]
if (is_clang) {
include_dirs += [
- "$tizen_clang_base_path/lib/clang/7.0.0/include",
+ "$tizen_clang_base_path/lib/clang/11.0.0/include",
"$deps_lib_path/gcc/armv7l-tizen-linux-gnueabi/$gcc_ver/include/c++",
"$deps_lib_path/gcc/armv7l-tizen-linux-gnueabi/$gcc_ver/include/c++/armv7l-tizen-linux-gnueabi",
]
/*===---- cuda_builtin_vars.h - CUDA built-in variables ---------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#define __DELETE
#endif
-// Make sure nobody can create instances of the special varible types. nvcc
+// Make sure nobody can create instances of the special variable types. nvcc
// also disallows taking address of special variables, so we disable address-of
// operator as well.
#define __CUDA_DISALLOW_BUILTINVAR_ACCESS(TypeName) \
/*===---- __clang_cuda_cmath.h - Device-side CUDA cmath support ------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#error "This file is for CUDA compilation only."
#endif
+#ifndef __OPENMP_NVPTX__
#include <limits>
+#endif
// CUDA lets us use various std math functions on the device side. This file
// works in concert with __clang_cuda_math_forward_declares.h to make this work.
// implementation. Declaring in the global namespace and pulling into namespace
// std covers all of the known knowns.
+#ifdef __OPENMP_NVPTX__
+#define __DEVICE__ static constexpr __attribute__((always_inline, nothrow))
+#else
#define __DEVICE__ static __device__ __inline__ __attribute__((always_inline))
+#endif
__DEVICE__ long long abs(long long __n) { return ::llabs(__n); }
__DEVICE__ long abs(long __n) { return ::labs(__n); }
}
// For inscrutable reasons, the CUDA headers define these functions for us on
-// Windows.
-#ifndef _MSC_VER
+// Windows. For OpenMP we omit these as some old system headers have
+// non-conforming `isinf(float)` and `isnan(float)` implementations that return
+// an `int`. The system versions of these functions should be fine anyway.
+#if !defined(_MSC_VER) && !defined(__OPENMP_NVPTX__)
__DEVICE__ bool isinf(float __x) { return ::__isinff(__x); }
__DEVICE__ bool isinf(double __x) { return ::__isinf(__x); }
__DEVICE__ bool isfinite(float __x) { return ::__finitef(__x); }
__DEVICE__ float log(float __x) { return ::logf(__x); }
__DEVICE__ float log10(float __x) { return ::log10f(__x); }
__DEVICE__ float modf(float __x, float *__iptr) { return ::modff(__x, __iptr); }
-__DEVICE__ float nexttoward(float __from, double __to) {
- return __builtin_nexttowardf(__from, __to);
-}
-__DEVICE__ double nexttoward(double __from, double __to) {
- return __builtin_nexttoward(__from, __to);
-}
-__DEVICE__ float nexttowardf(float __from, double __to) {
- return __builtin_nexttowardf(__from, __to);
-}
__DEVICE__ float pow(float __base, float __exp) {
return ::powf(__base, __exp);
}
__DEVICE__ float tan(float __x) { return ::tanf(__x); }
__DEVICE__ float tanh(float __x) { return ::tanhf(__x); }
+// Notably missing above is nexttoward. We omit it because
+// libdevice doesn't provide an implementation, and we don't want to be in the
+// business of implementing tricky libm functions in this header.
+
+#ifndef __OPENMP_NVPTX__
+
// Now we've defined everything we promised we'd define in
// __clang_cuda_math_forward_declares.h. We need to do two additional things to
// fix up our math functions.
return std::ldexp((double)__x, __exp);
}
-template <typename __T>
-__DEVICE__ typename __clang_cuda_enable_if<std::numeric_limits<__T>::is_integer,
- double>::type
-nexttoward(__T __from, double __to) {
- return std::nexttoward((double)__from, __to);
-}
-
template <typename __T1, typename __T2>
__DEVICE__ typename __clang_cuda_enable_if<
std::numeric_limits<__T1>::is_specialized &&
using ::lround;
using ::nearbyint;
using ::nextafter;
-using ::nexttoward;
using ::pow;
using ::remainder;
using ::remquo;
using ::modff;
using ::nearbyintf;
using ::nextafterf;
-using ::nexttowardf;
-using ::nexttowardf;
using ::powf;
using ::remainderf;
using ::remquof;
} // namespace std
#endif
+#endif // __OPENMP_NVPTX__
+
#undef __DEVICE__
#endif
--- /dev/null
+/*===-- __clang_cuda_complex_builtins - CUDA impls of runtime complex fns ---===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __CLANG_CUDA_COMPLEX_BUILTINS
+#define __CLANG_CUDA_COMPLEX_BUILTINS
+
+// This header defines __muldc3, __mulsc3, __divdc3, and __divsc3. These are
+// libgcc functions that clang assumes are available when compiling c99 complex
+// operations. (These implementations come from libc++, and have been modified
+// to work with CUDA and OpenMP target offloading [in C and C++ mode].)
+
+#pragma push_macro("__DEVICE__")
+#ifdef _OPENMP
+#pragma omp declare target
+#define __DEVICE__ __attribute__((noinline, nothrow, cold, weak))
+#else
+#define __DEVICE__ __device__ inline
+#endif
+
+// To make the algorithms available for C and C++ in CUDA and OpenMP we select
+// different but equivalent function versions. TODO: For OpenMP we currently
+// select the native builtins as the overload support for templates is lacking.
+#if !defined(_OPENMP)
+#define _ISNANd std::isnan
+#define _ISNANf std::isnan
+#define _ISINFd std::isinf
+#define _ISINFf std::isinf
+#define _ISFINITEd std::isfinite
+#define _ISFINITEf std::isfinite
+#define _COPYSIGNd std::copysign
+#define _COPYSIGNf std::copysign
+#define _SCALBNd std::scalbn
+#define _SCALBNf std::scalbn
+#define _ABSd std::abs
+#define _ABSf std::abs
+#define _LOGBd std::logb
+#define _LOGBf std::logb
+#else
+#define _ISNANd __nv_isnand
+#define _ISNANf __nv_isnanf
+#define _ISINFd __nv_isinfd
+#define _ISINFf __nv_isinff
+#define _ISFINITEd __nv_isfinited
+#define _ISFINITEf __nv_finitef
+#define _COPYSIGNd __nv_copysign
+#define _COPYSIGNf __nv_copysignf
+#define _SCALBNd __nv_scalbn
+#define _SCALBNf __nv_scalbnf
+#define _ABSd __nv_fabs
+#define _ABSf __nv_fabsf
+#define _LOGBd __nv_logb
+#define _LOGBf __nv_logbf
+#endif
+
+#if defined(__cplusplus)
+extern "C" {
+#endif
+
+__DEVICE__ double _Complex __muldc3(double __a, double __b, double __c,
+ double __d) {
+ double __ac = __a * __c;
+ double __bd = __b * __d;
+ double __ad = __a * __d;
+ double __bc = __b * __c;
+ double _Complex z;
+ __real__(z) = __ac - __bd;
+ __imag__(z) = __ad + __bc;
+ if (_ISNANd(__real__(z)) && _ISNANd(__imag__(z))) {
+ int __recalc = 0;
+ if (_ISINFd(__a) || _ISINFd(__b)) {
+ __a = _COPYSIGNd(_ISINFd(__a) ? 1 : 0, __a);
+ __b = _COPYSIGNd(_ISINFd(__b) ? 1 : 0, __b);
+ if (_ISNANd(__c))
+ __c = _COPYSIGNd(0, __c);
+ if (_ISNANd(__d))
+ __d = _COPYSIGNd(0, __d);
+ __recalc = 1;
+ }
+ if (_ISINFd(__c) || _ISINFd(__d)) {
+ __c = _COPYSIGNd(_ISINFd(__c) ? 1 : 0, __c);
+ __d = _COPYSIGNd(_ISINFd(__d) ? 1 : 0, __d);
+ if (_ISNANd(__a))
+ __a = _COPYSIGNd(0, __a);
+ if (_ISNANd(__b))
+ __b = _COPYSIGNd(0, __b);
+ __recalc = 1;
+ }
+ if (!__recalc &&
+ (_ISINFd(__ac) || _ISINFd(__bd) || _ISINFd(__ad) || _ISINFd(__bc))) {
+ if (_ISNANd(__a))
+ __a = _COPYSIGNd(0, __a);
+ if (_ISNANd(__b))
+ __b = _COPYSIGNd(0, __b);
+ if (_ISNANd(__c))
+ __c = _COPYSIGNd(0, __c);
+ if (_ISNANd(__d))
+ __d = _COPYSIGNd(0, __d);
+ __recalc = 1;
+ }
+ if (__recalc) {
+ // Can't use std::numeric_limits<double>::infinity() -- that doesn't have
+ // a device overload (and isn't constexpr before C++11, naturally).
+ __real__(z) = __builtin_huge_val() * (__a * __c - __b * __d);
+ __imag__(z) = __builtin_huge_val() * (__a * __d + __b * __c);
+ }
+ }
+ return z;
+}
+
+__DEVICE__ float _Complex __mulsc3(float __a, float __b, float __c, float __d) {
+ float __ac = __a * __c;
+ float __bd = __b * __d;
+ float __ad = __a * __d;
+ float __bc = __b * __c;
+ float _Complex z;
+ __real__(z) = __ac - __bd;
+ __imag__(z) = __ad + __bc;
+ if (_ISNANf(__real__(z)) && _ISNANf(__imag__(z))) {
+ int __recalc = 0;
+ if (_ISINFf(__a) || _ISINFf(__b)) {
+ __a = _COPYSIGNf(_ISINFf(__a) ? 1 : 0, __a);
+ __b = _COPYSIGNf(_ISINFf(__b) ? 1 : 0, __b);
+ if (_ISNANf(__c))
+ __c = _COPYSIGNf(0, __c);
+ if (_ISNANf(__d))
+ __d = _COPYSIGNf(0, __d);
+ __recalc = 1;
+ }
+ if (_ISINFf(__c) || _ISINFf(__d)) {
+ __c = _COPYSIGNf(_ISINFf(__c) ? 1 : 0, __c);
+ __d = _COPYSIGNf(_ISINFf(__d) ? 1 : 0, __d);
+ if (_ISNANf(__a))
+ __a = _COPYSIGNf(0, __a);
+ if (_ISNANf(__b))
+ __b = _COPYSIGNf(0, __b);
+ __recalc = 1;
+ }
+ if (!__recalc &&
+ (_ISINFf(__ac) || _ISINFf(__bd) || _ISINFf(__ad) || _ISINFf(__bc))) {
+ if (_ISNANf(__a))
+ __a = _COPYSIGNf(0, __a);
+ if (_ISNANf(__b))
+ __b = _COPYSIGNf(0, __b);
+ if (_ISNANf(__c))
+ __c = _COPYSIGNf(0, __c);
+ if (_ISNANf(__d))
+ __d = _COPYSIGNf(0, __d);
+ __recalc = 1;
+ }
+ if (__recalc) {
+ __real__(z) = __builtin_huge_valf() * (__a * __c - __b * __d);
+ __imag__(z) = __builtin_huge_valf() * (__a * __d + __b * __c);
+ }
+ }
+ return z;
+}
+
+__DEVICE__ double _Complex __divdc3(double __a, double __b, double __c,
+ double __d) {
+ int __ilogbw = 0;
+ // Can't use std::max, because that's defined in <algorithm>, and we don't
+ // want to pull that in for every compile. The CUDA headers define
+ // ::max(float, float) and ::max(double, double), which is sufficient for us.
+ double __logbw = _LOGBd(max(_ABSd(__c), _ABSd(__d)));
+ if (_ISFINITEd(__logbw)) {
+ __ilogbw = (int)__logbw;
+ __c = _SCALBNd(__c, -__ilogbw);
+ __d = _SCALBNd(__d, -__ilogbw);
+ }
+ double __denom = __c * __c + __d * __d;
+ double _Complex z;
+ __real__(z) = _SCALBNd((__a * __c + __b * __d) / __denom, -__ilogbw);
+ __imag__(z) = _SCALBNd((__b * __c - __a * __d) / __denom, -__ilogbw);
+ if (_ISNANd(__real__(z)) && _ISNANd(__imag__(z))) {
+ if ((__denom == 0.0) && (!_ISNANd(__a) || !_ISNANd(__b))) {
+ __real__(z) = _COPYSIGNd(__builtin_huge_val(), __c) * __a;
+ __imag__(z) = _COPYSIGNd(__builtin_huge_val(), __c) * __b;
+ } else if ((_ISINFd(__a) || _ISINFd(__b)) && _ISFINITEd(__c) &&
+ _ISFINITEd(__d)) {
+ __a = _COPYSIGNd(_ISINFd(__a) ? 1.0 : 0.0, __a);
+ __b = _COPYSIGNd(_ISINFd(__b) ? 1.0 : 0.0, __b);
+ __real__(z) = __builtin_huge_val() * (__a * __c + __b * __d);
+ __imag__(z) = __builtin_huge_val() * (__b * __c - __a * __d);
+ } else if (_ISINFd(__logbw) && __logbw > 0.0 && _ISFINITEd(__a) &&
+ _ISFINITEd(__b)) {
+ __c = _COPYSIGNd(_ISINFd(__c) ? 1.0 : 0.0, __c);
+ __d = _COPYSIGNd(_ISINFd(__d) ? 1.0 : 0.0, __d);
+ __real__(z) = 0.0 * (__a * __c + __b * __d);
+ __imag__(z) = 0.0 * (__b * __c - __a * __d);
+ }
+ }
+ return z;
+}
+
+__DEVICE__ float _Complex __divsc3(float __a, float __b, float __c, float __d) {
+ int __ilogbw = 0;
+ float __logbw = _LOGBf(max(_ABSf(__c), _ABSf(__d)));
+ if (_ISFINITEf(__logbw)) {
+ __ilogbw = (int)__logbw;
+ __c = _SCALBNf(__c, -__ilogbw);
+ __d = _SCALBNf(__d, -__ilogbw);
+ }
+ float __denom = __c * __c + __d * __d;
+ float _Complex z;
+ __real__(z) = _SCALBNf((__a * __c + __b * __d) / __denom, -__ilogbw);
+ __imag__(z) = _SCALBNf((__b * __c - __a * __d) / __denom, -__ilogbw);
+ if (_ISNANf(__real__(z)) && _ISNANf(__imag__(z))) {
+ if ((__denom == 0) && (!_ISNANf(__a) || !_ISNANf(__b))) {
+ __real__(z) = _COPYSIGNf(__builtin_huge_valf(), __c) * __a;
+ __imag__(z) = _COPYSIGNf(__builtin_huge_valf(), __c) * __b;
+ } else if ((_ISINFf(__a) || _ISINFf(__b)) && _ISFINITEf(__c) &&
+ _ISFINITEf(__d)) {
+ __a = _COPYSIGNf(_ISINFf(__a) ? 1 : 0, __a);
+ __b = _COPYSIGNf(_ISINFf(__b) ? 1 : 0, __b);
+ __real__(z) = __builtin_huge_valf() * (__a * __c + __b * __d);
+ __imag__(z) = __builtin_huge_valf() * (__b * __c - __a * __d);
+ } else if (_ISINFf(__logbw) && __logbw > 0 && _ISFINITEf(__a) &&
+ _ISFINITEf(__b)) {
+ __c = _COPYSIGNf(_ISINFf(__c) ? 1 : 0, __c);
+ __d = _COPYSIGNf(_ISINFf(__d) ? 1 : 0, __d);
+ __real__(z) = 0 * (__a * __c + __b * __d);
+ __imag__(z) = 0 * (__b * __c - __a * __d);
+ }
+ }
+ return z;
+}
+
+#if defined(__cplusplus)
+} // extern "C"
+#endif
+
+#undef _ISNANd
+#undef _ISNANf
+#undef _ISINFd
+#undef _ISINFf
+#undef _COPYSIGNd
+#undef _COPYSIGNf
+#undef _ISFINITEd
+#undef _ISFINITEf
+#undef _SCALBNd
+#undef _SCALBNf
+#undef _ABSd
+#undef _ABSf
+#undef _LOGBd
+#undef _LOGBf
+
+#ifdef _OPENMP
+#pragma omp end declare target
+#endif
+
+#pragma pop_macro("__DEVICE__")
+
+#endif // __CLANG_CUDA_COMPLEX_BUILTINS
--- /dev/null
+/*===---- __clang_cuda_device_functions.h - CUDA runtime support -----------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __CLANG_CUDA_DEVICE_FUNCTIONS_H__
+#define __CLANG_CUDA_DEVICE_FUNCTIONS_H__
+
+#ifndef __OPENMP_NVPTX__
+#if CUDA_VERSION < 9000
+#error This file is intended to be used with CUDA-9+ only.
+#endif
+#endif
+
+// __DEVICE__ is a helper macro with common set of attributes for the wrappers
+// we implement in this file. We need static in order to avoid emitting unused
+// functions and __forceinline__ helps inlining these wrappers at -O1.
+#pragma push_macro("__DEVICE__")
+#ifdef __OPENMP_NVPTX__
+#define __DEVICE__ static __attribute__((always_inline, nothrow))
+#else
+#define __DEVICE__ static __device__ __forceinline__
+#endif
+
+__DEVICE__ int __all(int __a) { return __nvvm_vote_all(__a); }
+__DEVICE__ int __any(int __a) { return __nvvm_vote_any(__a); }
+__DEVICE__ unsigned int __ballot(int __a) { return __nvvm_vote_ballot(__a); }
+__DEVICE__ unsigned int __brev(unsigned int __a) { return __nv_brev(__a); }
+__DEVICE__ unsigned long long __brevll(unsigned long long __a) {
+ return __nv_brevll(__a);
+}
+#if defined(__cplusplus)
+__DEVICE__ void __brkpt() { asm volatile("brkpt;"); }
+__DEVICE__ void __brkpt(int __a) { __brkpt(); }
+#else
+__DEVICE__ void __attribute__((overloadable)) __brkpt(void) { asm volatile("brkpt;"); }
+__DEVICE__ void __attribute__((overloadable)) __brkpt(int __a) { __brkpt(); }
+#endif
+__DEVICE__ unsigned int __byte_perm(unsigned int __a, unsigned int __b,
+ unsigned int __c) {
+ return __nv_byte_perm(__a, __b, __c);
+}
+__DEVICE__ int __clz(int __a) { return __nv_clz(__a); }
+__DEVICE__ int __clzll(long long __a) { return __nv_clzll(__a); }
+__DEVICE__ float __cosf(float __a) { return __nv_fast_cosf(__a); }
+__DEVICE__ double __dAtomicAdd(double *__p, double __v) {
+ return __nvvm_atom_add_gen_d(__p, __v);
+}
+__DEVICE__ double __dAtomicAdd_block(double *__p, double __v) {
+ return __nvvm_atom_cta_add_gen_d(__p, __v);
+}
+__DEVICE__ double __dAtomicAdd_system(double *__p, double __v) {
+ return __nvvm_atom_sys_add_gen_d(__p, __v);
+}
+__DEVICE__ double __dadd_rd(double __a, double __b) {
+ return __nv_dadd_rd(__a, __b);
+}
+__DEVICE__ double __dadd_rn(double __a, double __b) {
+ return __nv_dadd_rn(__a, __b);
+}
+__DEVICE__ double __dadd_ru(double __a, double __b) {
+ return __nv_dadd_ru(__a, __b);
+}
+__DEVICE__ double __dadd_rz(double __a, double __b) {
+ return __nv_dadd_rz(__a, __b);
+}
+__DEVICE__ double __ddiv_rd(double __a, double __b) {
+ return __nv_ddiv_rd(__a, __b);
+}
+__DEVICE__ double __ddiv_rn(double __a, double __b) {
+ return __nv_ddiv_rn(__a, __b);
+}
+__DEVICE__ double __ddiv_ru(double __a, double __b) {
+ return __nv_ddiv_ru(__a, __b);
+}
+__DEVICE__ double __ddiv_rz(double __a, double __b) {
+ return __nv_ddiv_rz(__a, __b);
+}
+__DEVICE__ double __dmul_rd(double __a, double __b) {
+ return __nv_dmul_rd(__a, __b);
+}
+__DEVICE__ double __dmul_rn(double __a, double __b) {
+ return __nv_dmul_rn(__a, __b);
+}
+__DEVICE__ double __dmul_ru(double __a, double __b) {
+ return __nv_dmul_ru(__a, __b);
+}
+__DEVICE__ double __dmul_rz(double __a, double __b) {
+ return __nv_dmul_rz(__a, __b);
+}
+__DEVICE__ float __double2float_rd(double __a) {
+ return __nv_double2float_rd(__a);
+}
+__DEVICE__ float __double2float_rn(double __a) {
+ return __nv_double2float_rn(__a);
+}
+__DEVICE__ float __double2float_ru(double __a) {
+ return __nv_double2float_ru(__a);
+}
+__DEVICE__ float __double2float_rz(double __a) {
+ return __nv_double2float_rz(__a);
+}
+__DEVICE__ int __double2hiint(double __a) { return __nv_double2hiint(__a); }
+__DEVICE__ int __double2int_rd(double __a) { return __nv_double2int_rd(__a); }
+__DEVICE__ int __double2int_rn(double __a) { return __nv_double2int_rn(__a); }
+__DEVICE__ int __double2int_ru(double __a) { return __nv_double2int_ru(__a); }
+__DEVICE__ int __double2int_rz(double __a) { return __nv_double2int_rz(__a); }
+__DEVICE__ long long __double2ll_rd(double __a) {
+ return __nv_double2ll_rd(__a);
+}
+__DEVICE__ long long __double2ll_rn(double __a) {
+ return __nv_double2ll_rn(__a);
+}
+__DEVICE__ long long __double2ll_ru(double __a) {
+ return __nv_double2ll_ru(__a);
+}
+__DEVICE__ long long __double2ll_rz(double __a) {
+ return __nv_double2ll_rz(__a);
+}
+__DEVICE__ int __double2loint(double __a) { return __nv_double2loint(__a); }
+__DEVICE__ unsigned int __double2uint_rd(double __a) {
+ return __nv_double2uint_rd(__a);
+}
+__DEVICE__ unsigned int __double2uint_rn(double __a) {
+ return __nv_double2uint_rn(__a);
+}
+__DEVICE__ unsigned int __double2uint_ru(double __a) {
+ return __nv_double2uint_ru(__a);
+}
+__DEVICE__ unsigned int __double2uint_rz(double __a) {
+ return __nv_double2uint_rz(__a);
+}
+__DEVICE__ unsigned long long __double2ull_rd(double __a) {
+ return __nv_double2ull_rd(__a);
+}
+__DEVICE__ unsigned long long __double2ull_rn(double __a) {
+ return __nv_double2ull_rn(__a);
+}
+__DEVICE__ unsigned long long __double2ull_ru(double __a) {
+ return __nv_double2ull_ru(__a);
+}
+__DEVICE__ unsigned long long __double2ull_rz(double __a) {
+ return __nv_double2ull_rz(__a);
+}
+__DEVICE__ long long __double_as_longlong(double __a) {
+ return __nv_double_as_longlong(__a);
+}
+__DEVICE__ double __drcp_rd(double __a) { return __nv_drcp_rd(__a); }
+__DEVICE__ double __drcp_rn(double __a) { return __nv_drcp_rn(__a); }
+__DEVICE__ double __drcp_ru(double __a) { return __nv_drcp_ru(__a); }
+__DEVICE__ double __drcp_rz(double __a) { return __nv_drcp_rz(__a); }
+__DEVICE__ double __dsqrt_rd(double __a) { return __nv_dsqrt_rd(__a); }
+__DEVICE__ double __dsqrt_rn(double __a) { return __nv_dsqrt_rn(__a); }
+__DEVICE__ double __dsqrt_ru(double __a) { return __nv_dsqrt_ru(__a); }
+__DEVICE__ double __dsqrt_rz(double __a) { return __nv_dsqrt_rz(__a); }
+__DEVICE__ double __dsub_rd(double __a, double __b) {
+ return __nv_dsub_rd(__a, __b);
+}
+__DEVICE__ double __dsub_rn(double __a, double __b) {
+ return __nv_dsub_rn(__a, __b);
+}
+__DEVICE__ double __dsub_ru(double __a, double __b) {
+ return __nv_dsub_ru(__a, __b);
+}
+__DEVICE__ double __dsub_rz(double __a, double __b) {
+ return __nv_dsub_rz(__a, __b);
+}
+__DEVICE__ float __exp10f(float __a) { return __nv_fast_exp10f(__a); }
+__DEVICE__ float __expf(float __a) { return __nv_fast_expf(__a); }
+__DEVICE__ float __fAtomicAdd(float *__p, float __v) {
+ return __nvvm_atom_add_gen_f(__p, __v);
+}
+__DEVICE__ float __fAtomicAdd_block(float *__p, float __v) {
+ return __nvvm_atom_cta_add_gen_f(__p, __v);
+}
+__DEVICE__ float __fAtomicAdd_system(float *__p, float __v) {
+ return __nvvm_atom_sys_add_gen_f(__p, __v);
+}
+__DEVICE__ float __fAtomicExch(float *__p, float __v) {
+ return __nv_int_as_float(
+ __nvvm_atom_xchg_gen_i((int *)__p, __nv_float_as_int(__v)));
+}
+__DEVICE__ float __fAtomicExch_block(float *__p, float __v) {
+ return __nv_int_as_float(
+ __nvvm_atom_cta_xchg_gen_i((int *)__p, __nv_float_as_int(__v)));
+}
+__DEVICE__ float __fAtomicExch_system(float *__p, float __v) {
+ return __nv_int_as_float(
+ __nvvm_atom_sys_xchg_gen_i((int *)__p, __nv_float_as_int(__v)));
+}
+__DEVICE__ float __fadd_rd(float __a, float __b) {
+ return __nv_fadd_rd(__a, __b);
+}
+__DEVICE__ float __fadd_rn(float __a, float __b) {
+ return __nv_fadd_rn(__a, __b);
+}
+__DEVICE__ float __fadd_ru(float __a, float __b) {
+ return __nv_fadd_ru(__a, __b);
+}
+__DEVICE__ float __fadd_rz(float __a, float __b) {
+ return __nv_fadd_rz(__a, __b);
+}
+__DEVICE__ float __fdiv_rd(float __a, float __b) {
+ return __nv_fdiv_rd(__a, __b);
+}
+__DEVICE__ float __fdiv_rn(float __a, float __b) {
+ return __nv_fdiv_rn(__a, __b);
+}
+__DEVICE__ float __fdiv_ru(float __a, float __b) {
+ return __nv_fdiv_ru(__a, __b);
+}
+__DEVICE__ float __fdiv_rz(float __a, float __b) {
+ return __nv_fdiv_rz(__a, __b);
+}
+__DEVICE__ float __fdividef(float __a, float __b) {
+ return __nv_fast_fdividef(__a, __b);
+}
+__DEVICE__ int __ffs(int __a) { return __nv_ffs(__a); }
+__DEVICE__ int __ffsll(long long __a) { return __nv_ffsll(__a); }
+__DEVICE__ int __finite(double __a) { return __nv_isfinited(__a); }
+__DEVICE__ int __finitef(float __a) { return __nv_finitef(__a); }
+#ifdef _MSC_VER
+__DEVICE__ int __finitel(long double __a);
+#endif
+__DEVICE__ int __float2int_rd(float __a) { return __nv_float2int_rd(__a); }
+__DEVICE__ int __float2int_rn(float __a) { return __nv_float2int_rn(__a); }
+__DEVICE__ int __float2int_ru(float __a) { return __nv_float2int_ru(__a); }
+__DEVICE__ int __float2int_rz(float __a) { return __nv_float2int_rz(__a); }
+__DEVICE__ long long __float2ll_rd(float __a) { return __nv_float2ll_rd(__a); }
+__DEVICE__ long long __float2ll_rn(float __a) { return __nv_float2ll_rn(__a); }
+__DEVICE__ long long __float2ll_ru(float __a) { return __nv_float2ll_ru(__a); }
+__DEVICE__ long long __float2ll_rz(float __a) { return __nv_float2ll_rz(__a); }
+__DEVICE__ unsigned int __float2uint_rd(float __a) {
+ return __nv_float2uint_rd(__a);
+}
+__DEVICE__ unsigned int __float2uint_rn(float __a) {
+ return __nv_float2uint_rn(__a);
+}
+__DEVICE__ unsigned int __float2uint_ru(float __a) {
+ return __nv_float2uint_ru(__a);
+}
+__DEVICE__ unsigned int __float2uint_rz(float __a) {
+ return __nv_float2uint_rz(__a);
+}
+__DEVICE__ unsigned long long __float2ull_rd(float __a) {
+ return __nv_float2ull_rd(__a);
+}
+__DEVICE__ unsigned long long __float2ull_rn(float __a) {
+ return __nv_float2ull_rn(__a);
+}
+__DEVICE__ unsigned long long __float2ull_ru(float __a) {
+ return __nv_float2ull_ru(__a);
+}
+__DEVICE__ unsigned long long __float2ull_rz(float __a) {
+ return __nv_float2ull_rz(__a);
+}
+__DEVICE__ int __float_as_int(float __a) { return __nv_float_as_int(__a); }
+__DEVICE__ unsigned int __float_as_uint(float __a) {
+ return __nv_float_as_uint(__a);
+}
+__DEVICE__ double __fma_rd(double __a, double __b, double __c) {
+ return __nv_fma_rd(__a, __b, __c);
+}
+__DEVICE__ double __fma_rn(double __a, double __b, double __c) {
+ return __nv_fma_rn(__a, __b, __c);
+}
+__DEVICE__ double __fma_ru(double __a, double __b, double __c) {
+ return __nv_fma_ru(__a, __b, __c);
+}
+__DEVICE__ double __fma_rz(double __a, double __b, double __c) {
+ return __nv_fma_rz(__a, __b, __c);
+}
+__DEVICE__ float __fmaf_ieee_rd(float __a, float __b, float __c) {
+ return __nv_fmaf_ieee_rd(__a, __b, __c);
+}
+__DEVICE__ float __fmaf_ieee_rn(float __a, float __b, float __c) {
+ return __nv_fmaf_ieee_rn(__a, __b, __c);
+}
+__DEVICE__ float __fmaf_ieee_ru(float __a, float __b, float __c) {
+ return __nv_fmaf_ieee_ru(__a, __b, __c);
+}
+__DEVICE__ float __fmaf_ieee_rz(float __a, float __b, float __c) {
+ return __nv_fmaf_ieee_rz(__a, __b, __c);
+}
+__DEVICE__ float __fmaf_rd(float __a, float __b, float __c) {
+ return __nv_fmaf_rd(__a, __b, __c);
+}
+__DEVICE__ float __fmaf_rn(float __a, float __b, float __c) {
+ return __nv_fmaf_rn(__a, __b, __c);
+}
+__DEVICE__ float __fmaf_ru(float __a, float __b, float __c) {
+ return __nv_fmaf_ru(__a, __b, __c);
+}
+__DEVICE__ float __fmaf_rz(float __a, float __b, float __c) {
+ return __nv_fmaf_rz(__a, __b, __c);
+}
+__DEVICE__ float __fmul_rd(float __a, float __b) {
+ return __nv_fmul_rd(__a, __b);
+}
+__DEVICE__ float __fmul_rn(float __a, float __b) {
+ return __nv_fmul_rn(__a, __b);
+}
+__DEVICE__ float __fmul_ru(float __a, float __b) {
+ return __nv_fmul_ru(__a, __b);
+}
+__DEVICE__ float __fmul_rz(float __a, float __b) {
+ return __nv_fmul_rz(__a, __b);
+}
+__DEVICE__ float __frcp_rd(float __a) { return __nv_frcp_rd(__a); }
+__DEVICE__ float __frcp_rn(float __a) { return __nv_frcp_rn(__a); }
+__DEVICE__ float __frcp_ru(float __a) { return __nv_frcp_ru(__a); }
+__DEVICE__ float __frcp_rz(float __a) { return __nv_frcp_rz(__a); }
+__DEVICE__ float __frsqrt_rn(float __a) { return __nv_frsqrt_rn(__a); }
+__DEVICE__ float __fsqrt_rd(float __a) { return __nv_fsqrt_rd(__a); }
+__DEVICE__ float __fsqrt_rn(float __a) { return __nv_fsqrt_rn(__a); }
+__DEVICE__ float __fsqrt_ru(float __a) { return __nv_fsqrt_ru(__a); }
+__DEVICE__ float __fsqrt_rz(float __a) { return __nv_fsqrt_rz(__a); }
+__DEVICE__ float __fsub_rd(float __a, float __b) {
+ return __nv_fsub_rd(__a, __b);
+}
+__DEVICE__ float __fsub_rn(float __a, float __b) {
+ return __nv_fsub_rn(__a, __b);
+}
+__DEVICE__ float __fsub_ru(float __a, float __b) {
+ return __nv_fsub_ru(__a, __b);
+}
+__DEVICE__ float __fsub_rz(float __a, float __b) {
+ return __nv_fsub_rz(__a, __b);
+}
+__DEVICE__ int __hadd(int __a, int __b) { return __nv_hadd(__a, __b); }
+__DEVICE__ double __hiloint2double(int __a, int __b) {
+ return __nv_hiloint2double(__a, __b);
+}
+__DEVICE__ int __iAtomicAdd(int *__p, int __v) {
+ return __nvvm_atom_add_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicAdd_block(int *__p, int __v) {
+ return __nvvm_atom_cta_add_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicAdd_system(int *__p, int __v) {
+ return __nvvm_atom_sys_add_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicAnd(int *__p, int __v) {
+ return __nvvm_atom_and_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicAnd_block(int *__p, int __v) {
+ return __nvvm_atom_cta_and_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicAnd_system(int *__p, int __v) {
+ return __nvvm_atom_sys_and_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicCAS(int *__p, int __cmp, int __v) {
+ return __nvvm_atom_cas_gen_i(__p, __cmp, __v);
+}
+__DEVICE__ int __iAtomicCAS_block(int *__p, int __cmp, int __v) {
+ return __nvvm_atom_cta_cas_gen_i(__p, __cmp, __v);
+}
+__DEVICE__ int __iAtomicCAS_system(int *__p, int __cmp, int __v) {
+ return __nvvm_atom_sys_cas_gen_i(__p, __cmp, __v);
+}
+__DEVICE__ int __iAtomicExch(int *__p, int __v) {
+ return __nvvm_atom_xchg_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicExch_block(int *__p, int __v) {
+ return __nvvm_atom_cta_xchg_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicExch_system(int *__p, int __v) {
+ return __nvvm_atom_sys_xchg_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicMax(int *__p, int __v) {
+ return __nvvm_atom_max_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicMax_block(int *__p, int __v) {
+ return __nvvm_atom_cta_max_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicMax_system(int *__p, int __v) {
+ return __nvvm_atom_sys_max_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicMin(int *__p, int __v) {
+ return __nvvm_atom_min_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicMin_block(int *__p, int __v) {
+ return __nvvm_atom_cta_min_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicMin_system(int *__p, int __v) {
+ return __nvvm_atom_sys_min_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicOr(int *__p, int __v) {
+ return __nvvm_atom_or_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicOr_block(int *__p, int __v) {
+ return __nvvm_atom_cta_or_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicOr_system(int *__p, int __v) {
+ return __nvvm_atom_sys_or_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicXor(int *__p, int __v) {
+ return __nvvm_atom_xor_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicXor_block(int *__p, int __v) {
+ return __nvvm_atom_cta_xor_gen_i(__p, __v);
+}
+__DEVICE__ int __iAtomicXor_system(int *__p, int __v) {
+ return __nvvm_atom_sys_xor_gen_i(__p, __v);
+}
+__DEVICE__ long long __illAtomicMax(long long *__p, long long __v) {
+ return __nvvm_atom_max_gen_ll(__p, __v);
+}
+__DEVICE__ long long __illAtomicMax_block(long long *__p, long long __v) {
+ return __nvvm_atom_cta_max_gen_ll(__p, __v);
+}
+__DEVICE__ long long __illAtomicMax_system(long long *__p, long long __v) {
+ return __nvvm_atom_sys_max_gen_ll(__p, __v);
+}
+__DEVICE__ long long __illAtomicMin(long long *__p, long long __v) {
+ return __nvvm_atom_min_gen_ll(__p, __v);
+}
+__DEVICE__ long long __illAtomicMin_block(long long *__p, long long __v) {
+ return __nvvm_atom_cta_min_gen_ll(__p, __v);
+}
+__DEVICE__ long long __illAtomicMin_system(long long *__p, long long __v) {
+ return __nvvm_atom_sys_min_gen_ll(__p, __v);
+}
+__DEVICE__ double __int2double_rn(int __a) { return __nv_int2double_rn(__a); }
+__DEVICE__ float __int2float_rd(int __a) { return __nv_int2float_rd(__a); }
+__DEVICE__ float __int2float_rn(int __a) { return __nv_int2float_rn(__a); }
+__DEVICE__ float __int2float_ru(int __a) { return __nv_int2float_ru(__a); }
+__DEVICE__ float __int2float_rz(int __a) { return __nv_int2float_rz(__a); }
+__DEVICE__ float __int_as_float(int __a) { return __nv_int_as_float(__a); }
+__DEVICE__ int __isfinited(double __a) { return __nv_isfinited(__a); }
+__DEVICE__ int __isinf(double __a) { return __nv_isinfd(__a); }
+__DEVICE__ int __isinff(float __a) { return __nv_isinff(__a); }
+#ifdef _MSC_VER
+__DEVICE__ int __isinfl(long double __a);
+#endif
+__DEVICE__ int __isnan(double __a) { return __nv_isnand(__a); }
+__DEVICE__ int __isnanf(float __a) { return __nv_isnanf(__a); }
+#ifdef _MSC_VER
+__DEVICE__ int __isnanl(long double __a);
+#endif
+__DEVICE__ double __ll2double_rd(long long __a) {
+ return __nv_ll2double_rd(__a);
+}
+__DEVICE__ double __ll2double_rn(long long __a) {
+ return __nv_ll2double_rn(__a);
+}
+__DEVICE__ double __ll2double_ru(long long __a) {
+ return __nv_ll2double_ru(__a);
+}
+__DEVICE__ double __ll2double_rz(long long __a) {
+ return __nv_ll2double_rz(__a);
+}
+__DEVICE__ float __ll2float_rd(long long __a) { return __nv_ll2float_rd(__a); }
+__DEVICE__ float __ll2float_rn(long long __a) { return __nv_ll2float_rn(__a); }
+__DEVICE__ float __ll2float_ru(long long __a) { return __nv_ll2float_ru(__a); }
+__DEVICE__ float __ll2float_rz(long long __a) { return __nv_ll2float_rz(__a); }
+__DEVICE__ long long __llAtomicAnd(long long *__p, long long __v) {
+ return __nvvm_atom_and_gen_ll(__p, __v);
+}
+__DEVICE__ long long __llAtomicAnd_block(long long *__p, long long __v) {
+ return __nvvm_atom_cta_and_gen_ll(__p, __v);
+}
+__DEVICE__ long long __llAtomicAnd_system(long long *__p, long long __v) {
+ return __nvvm_atom_sys_and_gen_ll(__p, __v);
+}
+__DEVICE__ long long __llAtomicOr(long long *__p, long long __v) {
+ return __nvvm_atom_or_gen_ll(__p, __v);
+}
+__DEVICE__ long long __llAtomicOr_block(long long *__p, long long __v) {
+ return __nvvm_atom_cta_or_gen_ll(__p, __v);
+}
+__DEVICE__ long long __llAtomicOr_system(long long *__p, long long __v) {
+ return __nvvm_atom_sys_or_gen_ll(__p, __v);
+}
+__DEVICE__ long long __llAtomicXor(long long *__p, long long __v) {
+ return __nvvm_atom_xor_gen_ll(__p, __v);
+}
+__DEVICE__ long long __llAtomicXor_block(long long *__p, long long __v) {
+ return __nvvm_atom_cta_xor_gen_ll(__p, __v);
+}
+__DEVICE__ long long __llAtomicXor_system(long long *__p, long long __v) {
+ return __nvvm_atom_sys_xor_gen_ll(__p, __v);
+}
+__DEVICE__ float __log10f(float __a) { return __nv_fast_log10f(__a); }
+__DEVICE__ float __log2f(float __a) { return __nv_fast_log2f(__a); }
+__DEVICE__ float __logf(float __a) { return __nv_fast_logf(__a); }
+__DEVICE__ double __longlong_as_double(long long __a) {
+ return __nv_longlong_as_double(__a);
+}
+__DEVICE__ int __mul24(int __a, int __b) { return __nv_mul24(__a, __b); }
+__DEVICE__ long long __mul64hi(long long __a, long long __b) {
+ return __nv_mul64hi(__a, __b);
+}
+__DEVICE__ int __mulhi(int __a, int __b) { return __nv_mulhi(__a, __b); }
+__DEVICE__ unsigned int __pm0(void) { return __nvvm_read_ptx_sreg_pm0(); }
+__DEVICE__ unsigned int __pm1(void) { return __nvvm_read_ptx_sreg_pm1(); }
+__DEVICE__ unsigned int __pm2(void) { return __nvvm_read_ptx_sreg_pm2(); }
+__DEVICE__ unsigned int __pm3(void) { return __nvvm_read_ptx_sreg_pm3(); }
+__DEVICE__ int __popc(int __a) { return __nv_popc(__a); }
+__DEVICE__ int __popcll(long long __a) { return __nv_popcll(__a); }
+__DEVICE__ float __powf(float __a, float __b) {
+ return __nv_fast_powf(__a, __b);
+}
+
+// Parameter must have a known integer value.
+#define __prof_trigger(__a) asm __volatile__("pmevent \t%0;" ::"i"(__a))
+__DEVICE__ int __rhadd(int __a, int __b) { return __nv_rhadd(__a, __b); }
+__DEVICE__ unsigned int __sad(int __a, int __b, unsigned int __c) {
+ return __nv_sad(__a, __b, __c);
+}
+__DEVICE__ float __saturatef(float __a) { return __nv_saturatef(__a); }
+__DEVICE__ int __signbitd(double __a) { return __nv_signbitd(__a); }
+__DEVICE__ int __signbitf(float __a) { return __nv_signbitf(__a); }
+__DEVICE__ void __sincosf(float __a, float *__s, float *__c) {
+ return __nv_fast_sincosf(__a, __s, __c);
+}
+__DEVICE__ float __sinf(float __a) { return __nv_fast_sinf(__a); }
+__DEVICE__ int __syncthreads_and(int __a) { return __nvvm_bar0_and(__a); }
+__DEVICE__ int __syncthreads_count(int __a) { return __nvvm_bar0_popc(__a); }
+__DEVICE__ int __syncthreads_or(int __a) { return __nvvm_bar0_or(__a); }
+__DEVICE__ float __tanf(float __a) { return __nv_fast_tanf(__a); }
+__DEVICE__ void __threadfence(void) { __nvvm_membar_gl(); }
+__DEVICE__ void __threadfence_block(void) { __nvvm_membar_cta(); };
+__DEVICE__ void __threadfence_system(void) { __nvvm_membar_sys(); };
+__DEVICE__ void __trap(void) { asm volatile("trap;"); }
+__DEVICE__ unsigned int __uAtomicAdd(unsigned int *__p, unsigned int __v) {
+ return __nvvm_atom_add_gen_i((int *)__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicAdd_block(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_cta_add_gen_i((int *)__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicAdd_system(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_sys_add_gen_i((int *)__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicAnd(unsigned int *__p, unsigned int __v) {
+ return __nvvm_atom_and_gen_i((int *)__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicAnd_block(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_cta_and_gen_i((int *)__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicAnd_system(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_sys_and_gen_i((int *)__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicCAS(unsigned int *__p, unsigned int __cmp,
+ unsigned int __v) {
+ return __nvvm_atom_cas_gen_i((int *)__p, __cmp, __v);
+}
+__DEVICE__ unsigned int
+__uAtomicCAS_block(unsigned int *__p, unsigned int __cmp, unsigned int __v) {
+ return __nvvm_atom_cta_cas_gen_i((int *)__p, __cmp, __v);
+}
+__DEVICE__ unsigned int
+__uAtomicCAS_system(unsigned int *__p, unsigned int __cmp, unsigned int __v) {
+ return __nvvm_atom_sys_cas_gen_i((int *)__p, __cmp, __v);
+}
+__DEVICE__ unsigned int __uAtomicDec(unsigned int *__p, unsigned int __v) {
+ return __nvvm_atom_dec_gen_ui(__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicDec_block(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_cta_dec_gen_ui(__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicDec_system(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_sys_dec_gen_ui(__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicExch(unsigned int *__p, unsigned int __v) {
+ return __nvvm_atom_xchg_gen_i((int *)__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicExch_block(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_cta_xchg_gen_i((int *)__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicExch_system(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_sys_xchg_gen_i((int *)__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicInc(unsigned int *__p, unsigned int __v) {
+ return __nvvm_atom_inc_gen_ui(__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicInc_block(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_cta_inc_gen_ui(__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicInc_system(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_sys_inc_gen_ui(__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicMax(unsigned int *__p, unsigned int __v) {
+ return __nvvm_atom_max_gen_ui(__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicMax_block(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_cta_max_gen_ui(__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicMax_system(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_sys_max_gen_ui(__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicMin(unsigned int *__p, unsigned int __v) {
+ return __nvvm_atom_min_gen_ui(__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicMin_block(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_cta_min_gen_ui(__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicMin_system(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_sys_min_gen_ui(__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicOr(unsigned int *__p, unsigned int __v) {
+ return __nvvm_atom_or_gen_i((int *)__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicOr_block(unsigned int *__p, unsigned int __v) {
+ return __nvvm_atom_cta_or_gen_i((int *)__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicOr_system(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_sys_or_gen_i((int *)__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicXor(unsigned int *__p, unsigned int __v) {
+ return __nvvm_atom_xor_gen_i((int *)__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicXor_block(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_cta_xor_gen_i((int *)__p, __v);
+}
+__DEVICE__ unsigned int __uAtomicXor_system(unsigned int *__p,
+ unsigned int __v) {
+ return __nvvm_atom_sys_xor_gen_i((int *)__p, __v);
+}
+__DEVICE__ unsigned int __uhadd(unsigned int __a, unsigned int __b) {
+ return __nv_uhadd(__a, __b);
+}
+__DEVICE__ double __uint2double_rn(unsigned int __a) {
+ return __nv_uint2double_rn(__a);
+}
+__DEVICE__ float __uint2float_rd(unsigned int __a) {
+ return __nv_uint2float_rd(__a);
+}
+__DEVICE__ float __uint2float_rn(unsigned int __a) {
+ return __nv_uint2float_rn(__a);
+}
+__DEVICE__ float __uint2float_ru(unsigned int __a) {
+ return __nv_uint2float_ru(__a);
+}
+__DEVICE__ float __uint2float_rz(unsigned int __a) {
+ return __nv_uint2float_rz(__a);
+}
+__DEVICE__ float __uint_as_float(unsigned int __a) {
+ return __nv_uint_as_float(__a);
+} //
+__DEVICE__ double __ull2double_rd(unsigned long long __a) {
+ return __nv_ull2double_rd(__a);
+}
+__DEVICE__ double __ull2double_rn(unsigned long long __a) {
+ return __nv_ull2double_rn(__a);
+}
+__DEVICE__ double __ull2double_ru(unsigned long long __a) {
+ return __nv_ull2double_ru(__a);
+}
+__DEVICE__ double __ull2double_rz(unsigned long long __a) {
+ return __nv_ull2double_rz(__a);
+}
+__DEVICE__ float __ull2float_rd(unsigned long long __a) {
+ return __nv_ull2float_rd(__a);
+}
+__DEVICE__ float __ull2float_rn(unsigned long long __a) {
+ return __nv_ull2float_rn(__a);
+}
+__DEVICE__ float __ull2float_ru(unsigned long long __a) {
+ return __nv_ull2float_ru(__a);
+}
+__DEVICE__ float __ull2float_rz(unsigned long long __a) {
+ return __nv_ull2float_rz(__a);
+}
+__DEVICE__ unsigned long long __ullAtomicAdd(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_add_gen_ll((long long *)__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicAdd_block(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_cta_add_gen_ll((long long *)__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicAdd_system(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_sys_add_gen_ll((long long *)__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicAnd(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_and_gen_ll((long long *)__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicAnd_block(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_cta_and_gen_ll((long long *)__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicAnd_system(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_sys_and_gen_ll((long long *)__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicCAS(unsigned long long *__p,
+ unsigned long long __cmp,
+ unsigned long long __v) {
+ return __nvvm_atom_cas_gen_ll((long long *)__p, __cmp, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicCAS_block(unsigned long long *__p,
+ unsigned long long __cmp,
+ unsigned long long __v) {
+ return __nvvm_atom_cta_cas_gen_ll((long long *)__p, __cmp, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicCAS_system(unsigned long long *__p,
+ unsigned long long __cmp,
+ unsigned long long __v) {
+ return __nvvm_atom_sys_cas_gen_ll((long long *)__p, __cmp, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicExch(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_xchg_gen_ll((long long *)__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicExch_block(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_cta_xchg_gen_ll((long long *)__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicExch_system(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_sys_xchg_gen_ll((long long *)__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicMax(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_max_gen_ull(__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicMax_block(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_cta_max_gen_ull(__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicMax_system(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_sys_max_gen_ull(__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicMin(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_min_gen_ull(__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicMin_block(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_cta_min_gen_ull(__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicMin_system(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_sys_min_gen_ull(__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicOr(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_or_gen_ll((long long *)__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicOr_block(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_cta_or_gen_ll((long long *)__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicOr_system(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_sys_or_gen_ll((long long *)__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicXor(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_xor_gen_ll((long long *)__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicXor_block(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_cta_xor_gen_ll((long long *)__p, __v);
+}
+__DEVICE__ unsigned long long __ullAtomicXor_system(unsigned long long *__p,
+ unsigned long long __v) {
+ return __nvvm_atom_sys_xor_gen_ll((long long *)__p, __v);
+}
+__DEVICE__ unsigned int __umul24(unsigned int __a, unsigned int __b) {
+ return __nv_umul24(__a, __b);
+}
+__DEVICE__ unsigned long long __umul64hi(unsigned long long __a,
+ unsigned long long __b) {
+ return __nv_umul64hi(__a, __b);
+}
+__DEVICE__ unsigned int __umulhi(unsigned int __a, unsigned int __b) {
+ return __nv_umulhi(__a, __b);
+}
+__DEVICE__ unsigned int __urhadd(unsigned int __a, unsigned int __b) {
+ return __nv_urhadd(__a, __b);
+}
+__DEVICE__ unsigned int __usad(unsigned int __a, unsigned int __b,
+ unsigned int __c) {
+ return __nv_usad(__a, __b, __c);
+}
+
+#if CUDA_VERSION >= 9000 && CUDA_VERSION < 9020
+__DEVICE__ unsigned int __vabs2(unsigned int __a) { return __nv_vabs2(__a); }
+__DEVICE__ unsigned int __vabs4(unsigned int __a) { return __nv_vabs4(__a); }
+__DEVICE__ unsigned int __vabsdiffs2(unsigned int __a, unsigned int __b) {
+ return __nv_vabsdiffs2(__a, __b);
+}
+__DEVICE__ unsigned int __vabsdiffs4(unsigned int __a, unsigned int __b) {
+ return __nv_vabsdiffs4(__a, __b);
+}
+__DEVICE__ unsigned int __vabsdiffu2(unsigned int __a, unsigned int __b) {
+ return __nv_vabsdiffu2(__a, __b);
+}
+__DEVICE__ unsigned int __vabsdiffu4(unsigned int __a, unsigned int __b) {
+ return __nv_vabsdiffu4(__a, __b);
+}
+__DEVICE__ unsigned int __vabsss2(unsigned int __a) {
+ return __nv_vabsss2(__a);
+}
+__DEVICE__ unsigned int __vabsss4(unsigned int __a) {
+ return __nv_vabsss4(__a);
+}
+__DEVICE__ unsigned int __vadd2(unsigned int __a, unsigned int __b) {
+ return __nv_vadd2(__a, __b);
+}
+__DEVICE__ unsigned int __vadd4(unsigned int __a, unsigned int __b) {
+ return __nv_vadd4(__a, __b);
+}
+__DEVICE__ unsigned int __vaddss2(unsigned int __a, unsigned int __b) {
+ return __nv_vaddss2(__a, __b);
+}
+__DEVICE__ unsigned int __vaddss4(unsigned int __a, unsigned int __b) {
+ return __nv_vaddss4(__a, __b);
+}
+__DEVICE__ unsigned int __vaddus2(unsigned int __a, unsigned int __b) {
+ return __nv_vaddus2(__a, __b);
+}
+__DEVICE__ unsigned int __vaddus4(unsigned int __a, unsigned int __b) {
+ return __nv_vaddus4(__a, __b);
+}
+__DEVICE__ unsigned int __vavgs2(unsigned int __a, unsigned int __b) {
+ return __nv_vavgs2(__a, __b);
+}
+__DEVICE__ unsigned int __vavgs4(unsigned int __a, unsigned int __b) {
+ return __nv_vavgs4(__a, __b);
+}
+__DEVICE__ unsigned int __vavgu2(unsigned int __a, unsigned int __b) {
+ return __nv_vavgu2(__a, __b);
+}
+__DEVICE__ unsigned int __vavgu4(unsigned int __a, unsigned int __b) {
+ return __nv_vavgu4(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpeq2(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpeq2(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpeq4(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpeq4(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpges2(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpges2(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpges4(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpges4(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpgeu2(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpgeu2(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpgeu4(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpgeu4(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpgts2(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpgts2(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpgts4(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpgts4(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpgtu2(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpgtu2(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpgtu4(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpgtu4(__a, __b);
+}
+__DEVICE__ unsigned int __vcmples2(unsigned int __a, unsigned int __b) {
+ return __nv_vcmples2(__a, __b);
+}
+__DEVICE__ unsigned int __vcmples4(unsigned int __a, unsigned int __b) {
+ return __nv_vcmples4(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpleu2(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpleu2(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpleu4(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpleu4(__a, __b);
+}
+__DEVICE__ unsigned int __vcmplts2(unsigned int __a, unsigned int __b) {
+ return __nv_vcmplts2(__a, __b);
+}
+__DEVICE__ unsigned int __vcmplts4(unsigned int __a, unsigned int __b) {
+ return __nv_vcmplts4(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpltu2(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpltu2(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpltu4(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpltu4(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpne2(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpne2(__a, __b);
+}
+__DEVICE__ unsigned int __vcmpne4(unsigned int __a, unsigned int __b) {
+ return __nv_vcmpne4(__a, __b);
+}
+__DEVICE__ unsigned int __vhaddu2(unsigned int __a, unsigned int __b) {
+ return __nv_vhaddu2(__a, __b);
+}
+__DEVICE__ unsigned int __vhaddu4(unsigned int __a, unsigned int __b) {
+ return __nv_vhaddu4(__a, __b);
+}
+__DEVICE__ unsigned int __vmaxs2(unsigned int __a, unsigned int __b) {
+ return __nv_vmaxs2(__a, __b);
+}
+__DEVICE__ unsigned int __vmaxs4(unsigned int __a, unsigned int __b) {
+ return __nv_vmaxs4(__a, __b);
+}
+__DEVICE__ unsigned int __vmaxu2(unsigned int __a, unsigned int __b) {
+ return __nv_vmaxu2(__a, __b);
+}
+__DEVICE__ unsigned int __vmaxu4(unsigned int __a, unsigned int __b) {
+ return __nv_vmaxu4(__a, __b);
+}
+__DEVICE__ unsigned int __vmins2(unsigned int __a, unsigned int __b) {
+ return __nv_vmins2(__a, __b);
+}
+__DEVICE__ unsigned int __vmins4(unsigned int __a, unsigned int __b) {
+ return __nv_vmins4(__a, __b);
+}
+__DEVICE__ unsigned int __vminu2(unsigned int __a, unsigned int __b) {
+ return __nv_vminu2(__a, __b);
+}
+__DEVICE__ unsigned int __vminu4(unsigned int __a, unsigned int __b) {
+ return __nv_vminu4(__a, __b);
+}
+__DEVICE__ unsigned int __vneg2(unsigned int __a) { return __nv_vneg2(__a); }
+__DEVICE__ unsigned int __vneg4(unsigned int __a) { return __nv_vneg4(__a); }
+__DEVICE__ unsigned int __vnegss2(unsigned int __a) {
+ return __nv_vnegss2(__a);
+}
+__DEVICE__ unsigned int __vnegss4(unsigned int __a) {
+ return __nv_vnegss4(__a);
+}
+__DEVICE__ unsigned int __vsads2(unsigned int __a, unsigned int __b) {
+ return __nv_vsads2(__a, __b);
+}
+__DEVICE__ unsigned int __vsads4(unsigned int __a, unsigned int __b) {
+ return __nv_vsads4(__a, __b);
+}
+__DEVICE__ unsigned int __vsadu2(unsigned int __a, unsigned int __b) {
+ return __nv_vsadu2(__a, __b);
+}
+__DEVICE__ unsigned int __vsadu4(unsigned int __a, unsigned int __b) {
+ return __nv_vsadu4(__a, __b);
+}
+__DEVICE__ unsigned int __vseteq2(unsigned int __a, unsigned int __b) {
+ return __nv_vseteq2(__a, __b);
+}
+__DEVICE__ unsigned int __vseteq4(unsigned int __a, unsigned int __b) {
+ return __nv_vseteq4(__a, __b);
+}
+__DEVICE__ unsigned int __vsetges2(unsigned int __a, unsigned int __b) {
+ return __nv_vsetges2(__a, __b);
+}
+__DEVICE__ unsigned int __vsetges4(unsigned int __a, unsigned int __b) {
+ return __nv_vsetges4(__a, __b);
+}
+__DEVICE__ unsigned int __vsetgeu2(unsigned int __a, unsigned int __b) {
+ return __nv_vsetgeu2(__a, __b);
+}
+__DEVICE__ unsigned int __vsetgeu4(unsigned int __a, unsigned int __b) {
+ return __nv_vsetgeu4(__a, __b);
+}
+__DEVICE__ unsigned int __vsetgts2(unsigned int __a, unsigned int __b) {
+ return __nv_vsetgts2(__a, __b);
+}
+__DEVICE__ unsigned int __vsetgts4(unsigned int __a, unsigned int __b) {
+ return __nv_vsetgts4(__a, __b);
+}
+__DEVICE__ unsigned int __vsetgtu2(unsigned int __a, unsigned int __b) {
+ return __nv_vsetgtu2(__a, __b);
+}
+__DEVICE__ unsigned int __vsetgtu4(unsigned int __a, unsigned int __b) {
+ return __nv_vsetgtu4(__a, __b);
+}
+__DEVICE__ unsigned int __vsetles2(unsigned int __a, unsigned int __b) {
+ return __nv_vsetles2(__a, __b);
+}
+__DEVICE__ unsigned int __vsetles4(unsigned int __a, unsigned int __b) {
+ return __nv_vsetles4(__a, __b);
+}
+__DEVICE__ unsigned int __vsetleu2(unsigned int __a, unsigned int __b) {
+ return __nv_vsetleu2(__a, __b);
+}
+__DEVICE__ unsigned int __vsetleu4(unsigned int __a, unsigned int __b) {
+ return __nv_vsetleu4(__a, __b);
+}
+__DEVICE__ unsigned int __vsetlts2(unsigned int __a, unsigned int __b) {
+ return __nv_vsetlts2(__a, __b);
+}
+__DEVICE__ unsigned int __vsetlts4(unsigned int __a, unsigned int __b) {
+ return __nv_vsetlts4(__a, __b);
+}
+__DEVICE__ unsigned int __vsetltu2(unsigned int __a, unsigned int __b) {
+ return __nv_vsetltu2(__a, __b);
+}
+__DEVICE__ unsigned int __vsetltu4(unsigned int __a, unsigned int __b) {
+ return __nv_vsetltu4(__a, __b);
+}
+__DEVICE__ unsigned int __vsetne2(unsigned int __a, unsigned int __b) {
+ return __nv_vsetne2(__a, __b);
+}
+__DEVICE__ unsigned int __vsetne4(unsigned int __a, unsigned int __b) {
+ return __nv_vsetne4(__a, __b);
+}
+__DEVICE__ unsigned int __vsub2(unsigned int __a, unsigned int __b) {
+ return __nv_vsub2(__a, __b);
+}
+__DEVICE__ unsigned int __vsub4(unsigned int __a, unsigned int __b) {
+ return __nv_vsub4(__a, __b);
+}
+__DEVICE__ unsigned int __vsubss2(unsigned int __a, unsigned int __b) {
+ return __nv_vsubss2(__a, __b);
+}
+__DEVICE__ unsigned int __vsubss4(unsigned int __a, unsigned int __b) {
+ return __nv_vsubss4(__a, __b);
+}
+__DEVICE__ unsigned int __vsubus2(unsigned int __a, unsigned int __b) {
+ return __nv_vsubus2(__a, __b);
+}
+__DEVICE__ unsigned int __vsubus4(unsigned int __a, unsigned int __b) {
+ return __nv_vsubus4(__a, __b);
+}
+#else // CUDA_VERSION >= 9020
+// CUDA no longer provides inline assembly (or bitcode) implementation of these
+// functions, so we have to reimplment them. The implementation is naive and is
+// not optimized for performance.
+
+// Helper function to convert N-bit boolean subfields into all-0 or all-1.
+// E.g. __bool2mask(0x01000100,8) -> 0xff00ff00
+// __bool2mask(0x00010000,16) -> 0xffff0000
+__DEVICE__ unsigned int __bool2mask(unsigned int __a, int shift) {
+ return (__a << shift) - __a;
+}
+__DEVICE__ unsigned int __vabs2(unsigned int __a) {
+ unsigned int r;
+ asm("vabsdiff2.s32.s32.s32 %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(0), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vabs4(unsigned int __a) {
+ unsigned int r;
+ asm("vabsdiff4.s32.s32.s32 %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(0), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vabsdiffs2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vabsdiff2.s32.s32.s32 %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+
+__DEVICE__ unsigned int __vabsdiffs4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vabsdiff4.s32.s32.s32 %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vabsdiffu2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vabsdiff2.u32.u32.u32 %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vabsdiffu4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vabsdiff4.u32.u32.u32 %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vabsss2(unsigned int __a) {
+ unsigned int r;
+ asm("vabsdiff2.s32.s32.s32.sat %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(0), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vabsss4(unsigned int __a) {
+ unsigned int r;
+ asm("vabsdiff4.s32.s32.s32.sat %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(0), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vadd2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vadd2.u32.u32.u32 %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vadd4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vadd4.u32.u32.u32 %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vaddss2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vadd2.s32.s32.s32.sat %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vaddss4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vadd4.s32.s32.s32.sat %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vaddus2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vadd2.u32.u32.u32.sat %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vaddus4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vadd4.u32.u32.u32.sat %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vavgs2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vavrg2.s32.s32.s32 %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vavgs4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vavrg4.s32.s32.s32 %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vavgu2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vavrg2.u32.u32.u32 %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vavgu4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vavrg4.u32.u32.u32 %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vseteq2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset2.u32.u32.eq %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpeq2(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vseteq2(__a, __b), 16);
+}
+__DEVICE__ unsigned int __vseteq4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset4.u32.u32.eq %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpeq4(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vseteq4(__a, __b), 8);
+}
+__DEVICE__ unsigned int __vsetges2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset2.s32.s32.ge %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpges2(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetges2(__a, __b), 16);
+}
+__DEVICE__ unsigned int __vsetges4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset4.s32.s32.ge %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpges4(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetges4(__a, __b), 8);
+}
+__DEVICE__ unsigned int __vsetgeu2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset2.u32.u32.ge %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpgeu2(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetgeu2(__a, __b), 16);
+}
+__DEVICE__ unsigned int __vsetgeu4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset4.u32.u32.ge %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpgeu4(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetgeu4(__a, __b), 8);
+}
+__DEVICE__ unsigned int __vsetgts2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset2.s32.s32.gt %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpgts2(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetgts2(__a, __b), 16);
+}
+__DEVICE__ unsigned int __vsetgts4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset4.s32.s32.gt %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpgts4(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetgts4(__a, __b), 8);
+}
+__DEVICE__ unsigned int __vsetgtu2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset2.u32.u32.gt %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpgtu2(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetgtu2(__a, __b), 16);
+}
+__DEVICE__ unsigned int __vsetgtu4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset4.u32.u32.gt %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpgtu4(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetgtu4(__a, __b), 8);
+}
+__DEVICE__ unsigned int __vsetles2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset2.s32.s32.le %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmples2(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetles2(__a, __b), 16);
+}
+__DEVICE__ unsigned int __vsetles4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset4.s32.s32.le %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmples4(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetles4(__a, __b), 8);
+}
+__DEVICE__ unsigned int __vsetleu2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset2.u32.u32.le %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpleu2(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetleu2(__a, __b), 16);
+}
+__DEVICE__ unsigned int __vsetleu4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset4.u32.u32.le %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpleu4(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetleu4(__a, __b), 8);
+}
+__DEVICE__ unsigned int __vsetlts2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset2.s32.s32.lt %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmplts2(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetlts2(__a, __b), 16);
+}
+__DEVICE__ unsigned int __vsetlts4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset4.s32.s32.lt %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmplts4(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetlts4(__a, __b), 8);
+}
+__DEVICE__ unsigned int __vsetltu2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset2.u32.u32.lt %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpltu2(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetltu2(__a, __b), 16);
+}
+__DEVICE__ unsigned int __vsetltu4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset4.u32.u32.lt %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpltu4(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetltu4(__a, __b), 8);
+}
+__DEVICE__ unsigned int __vsetne2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset2.u32.u32.ne %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpne2(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetne2(__a, __b), 16);
+}
+__DEVICE__ unsigned int __vsetne4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vset4.u32.u32.ne %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vcmpne4(unsigned int __a, unsigned int __b) {
+ return __bool2mask(__vsetne4(__a, __b), 8);
+}
+
+// Based on ITEM 23 in AIM-239: http://dspace.mit.edu/handle/1721.1/6086
+// (a & b) + (a | b) = a + b = (a ^ b) + 2 * (a & b) =>
+// (a + b) / 2 = ((a ^ b) >> 1) + (a & b)
+// To operate on multiple sub-elements we need to make sure to mask out bits
+// that crossed over into adjacent elements during the shift.
+__DEVICE__ unsigned int __vhaddu2(unsigned int __a, unsigned int __b) {
+ return (((__a ^ __b) >> 1) & ~0x80008000u) + (__a & __b);
+}
+__DEVICE__ unsigned int __vhaddu4(unsigned int __a, unsigned int __b) {
+ return (((__a ^ __b) >> 1) & ~0x80808080u) + (__a & __b);
+}
+
+__DEVICE__ unsigned int __vmaxs2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ if ((__a & 0x8000) && (__b & 0x8000)) {
+ // Work around a bug in ptxas which produces invalid result if low element
+ // is negative.
+ unsigned mask = __vcmpgts2(__a, __b);
+ r = (__a & mask) | (__b & ~mask);
+ } else {
+ asm("vmax2.s32.s32.s32 %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ }
+ return r;
+}
+__DEVICE__ unsigned int __vmaxs4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vmax4.s32.s32.s32 %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vmaxu2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vmax2.u32.u32.u32 %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vmaxu4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vmax4.u32.u32.u32 %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vmins2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vmin2.s32.s32.s32 %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vmins4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vmin4.s32.s32.s32 %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vminu2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vmin2.u32.u32.u32 %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vminu4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vmin4.u32.u32.u32 %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vsads2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vabsdiff2.s32.s32.s32.add %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vsads4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vabsdiff4.s32.s32.s32.add %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vsadu2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vabsdiff2.u32.u32.u32.add %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vsadu4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vabsdiff4.u32.u32.u32.add %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+
+__DEVICE__ unsigned int __vsub2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vsub2.u32.u32.u32 %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vneg2(unsigned int __a) { return __vsub2(0, __a); }
+
+__DEVICE__ unsigned int __vsub4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vsub4.u32.u32.u32 %0,%1,%2,%3;" : "=r"(r) : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vneg4(unsigned int __a) { return __vsub4(0, __a); }
+__DEVICE__ unsigned int __vsubss2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vsub2.s32.s32.s32.sat %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vnegss2(unsigned int __a) {
+ return __vsubss2(0, __a);
+}
+__DEVICE__ unsigned int __vsubss4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vsub4.s32.s32.s32.sat %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vnegss4(unsigned int __a) {
+ return __vsubss4(0, __a);
+}
+__DEVICE__ unsigned int __vsubus2(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vsub2.u32.u32.u32.sat %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+__DEVICE__ unsigned int __vsubus4(unsigned int __a, unsigned int __b) {
+ unsigned int r;
+ asm("vsub4.u32.u32.u32.sat %0,%1,%2,%3;"
+ : "=r"(r)
+ : "r"(__a), "r"(__b), "r"(0));
+ return r;
+}
+#endif // CUDA_VERSION >= 9020
+
+// For OpenMP we require the user to include <time.h> as we need to know what
+// clock_t is on the system.
+#ifndef __OPENMP_NVPTX__
+__DEVICE__ /* clock_t= */ int clock() { return __nvvm_read_ptx_sreg_clock(); }
+#endif
+__DEVICE__ long long clock64() { return __nvvm_read_ptx_sreg_clock64(); }
+
+// These functions shouldn't be declared when including this header
+// for math function resolution purposes.
+#ifndef __OPENMP_NVPTX__
+__DEVICE__ void *memcpy(void *__a, const void *__b, size_t __c) {
+ return __builtin_memcpy(__a, __b, __c);
+}
+__DEVICE__ void *memset(void *__a, int __b, size_t __c) {
+ return __builtin_memset(__a, __b, __c);
+}
+#endif
+
+#pragma pop_macro("__DEVICE__")
+#endif // __CLANG_CUDA_DEVICE_FUNCTIONS_H__
/*===--- __clang_cuda_intrinsics.h - Device-side CUDA intrinsic wrappers ---===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#if !defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 300
#pragma push_macro("__MAKE_SHUFFLES")
-#define __MAKE_SHUFFLES(__FnName, __IntIntrinsic, __FloatIntrinsic, __Mask) \
- inline __device__ int __FnName(int __val, int __offset, \
+#define __MAKE_SHUFFLES(__FnName, __IntIntrinsic, __FloatIntrinsic, __Mask, \
+ __Type) \
+ inline __device__ int __FnName(int __val, __Type __offset, \
int __width = warpSize) { \
return __IntIntrinsic(__val, __offset, \
((warpSize - __width) << 8) | (__Mask)); \
} \
- inline __device__ float __FnName(float __val, int __offset, \
+ inline __device__ float __FnName(float __val, __Type __offset, \
int __width = warpSize) { \
return __FloatIntrinsic(__val, __offset, \
((warpSize - __width) << 8) | (__Mask)); \
} \
- inline __device__ unsigned int __FnName(unsigned int __val, int __offset, \
+ inline __device__ unsigned int __FnName(unsigned int __val, __Type __offset, \
int __width = warpSize) { \
return static_cast<unsigned int>( \
::__FnName(static_cast<int>(__val), __offset, __width)); \
} \
- inline __device__ long long __FnName(long long __val, int __offset, \
+ inline __device__ long long __FnName(long long __val, __Type __offset, \
int __width = warpSize) { \
struct __Bits { \
int __a, __b; \
_Static_assert(sizeof(__val) == sizeof(__Bits)); \
_Static_assert(sizeof(__Bits) == 2 * sizeof(int)); \
__Bits __tmp; \
- memcpy(&__val, &__tmp, sizeof(__val)); \
+ memcpy(&__tmp, &__val, sizeof(__val)); \
__tmp.__a = ::__FnName(__tmp.__a, __offset, __width); \
__tmp.__b = ::__FnName(__tmp.__b, __offset, __width); \
long long __ret; \
memcpy(&__ret, &__tmp, sizeof(__tmp)); \
return __ret; \
} \
+ inline __device__ long __FnName(long __val, __Type __offset, \
+ int __width = warpSize) { \
+ _Static_assert(sizeof(long) == sizeof(long long) || \
+ sizeof(long) == sizeof(int)); \
+ if (sizeof(long) == sizeof(long long)) { \
+ return static_cast<long>( \
+ ::__FnName(static_cast<long long>(__val), __offset, __width)); \
+ } else if (sizeof(long) == sizeof(int)) { \
+ return static_cast<long>( \
+ ::__FnName(static_cast<int>(__val), __offset, __width)); \
+ } \
+ } \
+ inline __device__ unsigned long __FnName( \
+ unsigned long __val, __Type __offset, int __width = warpSize) { \
+ return static_cast<unsigned long>( \
+ ::__FnName(static_cast<long>(__val), __offset, __width)); \
+ } \
inline __device__ unsigned long long __FnName( \
- unsigned long long __val, int __offset, int __width = warpSize) { \
+ unsigned long long __val, __Type __offset, int __width = warpSize) { \
return static_cast<unsigned long long>(::__FnName( \
static_cast<unsigned long long>(__val), __offset, __width)); \
} \
- inline __device__ double __FnName(double __val, int __offset, \
+ inline __device__ double __FnName(double __val, __Type __offset, \
int __width = warpSize) { \
long long __tmp; \
_Static_assert(sizeof(__tmp) == sizeof(__val)); \
return __ret; \
}
-__MAKE_SHUFFLES(__shfl, __nvvm_shfl_idx_i32, __nvvm_shfl_idx_f32, 0x1f);
+__MAKE_SHUFFLES(__shfl, __nvvm_shfl_idx_i32, __nvvm_shfl_idx_f32, 0x1f, int);
// We use 0 rather than 31 as our mask, because shfl.up applies to lanes >=
// maxLane.
-__MAKE_SHUFFLES(__shfl_up, __nvvm_shfl_up_i32, __nvvm_shfl_up_f32, 0);
-__MAKE_SHUFFLES(__shfl_down, __nvvm_shfl_down_i32, __nvvm_shfl_down_f32, 0x1f);
-__MAKE_SHUFFLES(__shfl_xor, __nvvm_shfl_bfly_i32, __nvvm_shfl_bfly_f32, 0x1f);
-
+__MAKE_SHUFFLES(__shfl_up, __nvvm_shfl_up_i32, __nvvm_shfl_up_f32, 0,
+ unsigned int);
+__MAKE_SHUFFLES(__shfl_down, __nvvm_shfl_down_i32, __nvvm_shfl_down_f32, 0x1f,
+ unsigned int);
+__MAKE_SHUFFLES(__shfl_xor, __nvvm_shfl_bfly_i32, __nvvm_shfl_bfly_f32, 0x1f,
+ int);
#pragma pop_macro("__MAKE_SHUFFLES")
#endif // !defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 300
+#if CUDA_VERSION >= 9000
+#if (!defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 300)
+// __shfl_sync_* variants available in CUDA-9
+#pragma push_macro("__MAKE_SYNC_SHUFFLES")
+#define __MAKE_SYNC_SHUFFLES(__FnName, __IntIntrinsic, __FloatIntrinsic, \
+ __Mask, __Type) \
+ inline __device__ int __FnName(unsigned int __mask, int __val, \
+ __Type __offset, int __width = warpSize) { \
+ return __IntIntrinsic(__mask, __val, __offset, \
+ ((warpSize - __width) << 8) | (__Mask)); \
+ } \
+ inline __device__ float __FnName(unsigned int __mask, float __val, \
+ __Type __offset, int __width = warpSize) { \
+ return __FloatIntrinsic(__mask, __val, __offset, \
+ ((warpSize - __width) << 8) | (__Mask)); \
+ } \
+ inline __device__ unsigned int __FnName(unsigned int __mask, \
+ unsigned int __val, __Type __offset, \
+ int __width = warpSize) { \
+ return static_cast<unsigned int>( \
+ ::__FnName(__mask, static_cast<int>(__val), __offset, __width)); \
+ } \
+ inline __device__ long long __FnName(unsigned int __mask, long long __val, \
+ __Type __offset, \
+ int __width = warpSize) { \
+ struct __Bits { \
+ int __a, __b; \
+ }; \
+ _Static_assert(sizeof(__val) == sizeof(__Bits)); \
+ _Static_assert(sizeof(__Bits) == 2 * sizeof(int)); \
+ __Bits __tmp; \
+ memcpy(&__tmp, &__val, sizeof(__val)); \
+ __tmp.__a = ::__FnName(__mask, __tmp.__a, __offset, __width); \
+ __tmp.__b = ::__FnName(__mask, __tmp.__b, __offset, __width); \
+ long long __ret; \
+ memcpy(&__ret, &__tmp, sizeof(__tmp)); \
+ return __ret; \
+ } \
+ inline __device__ unsigned long long __FnName( \
+ unsigned int __mask, unsigned long long __val, __Type __offset, \
+ int __width = warpSize) { \
+ return static_cast<unsigned long long>(::__FnName( \
+ __mask, static_cast<unsigned long long>(__val), __offset, __width)); \
+ } \
+ inline __device__ long __FnName(unsigned int __mask, long __val, \
+ __Type __offset, int __width = warpSize) { \
+ _Static_assert(sizeof(long) == sizeof(long long) || \
+ sizeof(long) == sizeof(int)); \
+ if (sizeof(long) == sizeof(long long)) { \
+ return static_cast<long>(::__FnName( \
+ __mask, static_cast<long long>(__val), __offset, __width)); \
+ } else if (sizeof(long) == sizeof(int)) { \
+ return static_cast<long>( \
+ ::__FnName(__mask, static_cast<int>(__val), __offset, __width)); \
+ } \
+ } \
+ inline __device__ unsigned long __FnName( \
+ unsigned int __mask, unsigned long __val, __Type __offset, \
+ int __width = warpSize) { \
+ return static_cast<unsigned long>( \
+ ::__FnName(__mask, static_cast<long>(__val), __offset, __width)); \
+ } \
+ inline __device__ double __FnName(unsigned int __mask, double __val, \
+ __Type __offset, int __width = warpSize) { \
+ long long __tmp; \
+ _Static_assert(sizeof(__tmp) == sizeof(__val)); \
+ memcpy(&__tmp, &__val, sizeof(__val)); \
+ __tmp = ::__FnName(__mask, __tmp, __offset, __width); \
+ double __ret; \
+ memcpy(&__ret, &__tmp, sizeof(__ret)); \
+ return __ret; \
+ }
+__MAKE_SYNC_SHUFFLES(__shfl_sync, __nvvm_shfl_sync_idx_i32,
+ __nvvm_shfl_sync_idx_f32, 0x1f, int);
+// We use 0 rather than 31 as our mask, because shfl.up applies to lanes >=
+// maxLane.
+__MAKE_SYNC_SHUFFLES(__shfl_up_sync, __nvvm_shfl_sync_up_i32,
+ __nvvm_shfl_sync_up_f32, 0, unsigned int);
+__MAKE_SYNC_SHUFFLES(__shfl_down_sync, __nvvm_shfl_sync_down_i32,
+ __nvvm_shfl_sync_down_f32, 0x1f, unsigned int);
+__MAKE_SYNC_SHUFFLES(__shfl_xor_sync, __nvvm_shfl_sync_bfly_i32,
+ __nvvm_shfl_sync_bfly_f32, 0x1f, int);
+#pragma pop_macro("__MAKE_SYNC_SHUFFLES")
+
+inline __device__ void __syncwarp(unsigned int mask = 0xffffffff) {
+ return __nvvm_bar_warp_sync(mask);
+}
+
+inline __device__ void __barrier_sync(unsigned int id) {
+ __nvvm_barrier_sync(id);
+}
+
+inline __device__ void __barrier_sync_count(unsigned int id,
+ unsigned int count) {
+ __nvvm_barrier_sync_cnt(id, count);
+}
+
+inline __device__ int __all_sync(unsigned int mask, int pred) {
+ return __nvvm_vote_all_sync(mask, pred);
+}
+
+inline __device__ int __any_sync(unsigned int mask, int pred) {
+ return __nvvm_vote_any_sync(mask, pred);
+}
+
+inline __device__ int __uni_sync(unsigned int mask, int pred) {
+ return __nvvm_vote_uni_sync(mask, pred);
+}
+
+inline __device__ unsigned int __ballot_sync(unsigned int mask, int pred) {
+ return __nvvm_vote_ballot_sync(mask, pred);
+}
+
+inline __device__ unsigned int __activemask() {
+#if CUDA_VERSION < 9020
+ return __nvvm_vote_ballot(1);
+#else
+ unsigned int mask;
+ asm volatile("activemask.b32 %0;" : "=r"(mask));
+ return mask;
+#endif
+}
+
+inline __device__ unsigned int __fns(unsigned mask, unsigned base, int offset) {
+ return __nvvm_fns(mask, base, offset);
+}
+
+#endif // !defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 300
+
+// Define __match* builtins CUDA-9 headers expect to see.
+#if !defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 700
+inline __device__ unsigned int __match32_any_sync(unsigned int mask,
+ unsigned int value) {
+ return __nvvm_match_any_sync_i32(mask, value);
+}
+
+inline __device__ unsigned long long
+__match64_any_sync(unsigned int mask, unsigned long long value) {
+ return __nvvm_match_any_sync_i64(mask, value);
+}
+
+inline __device__ unsigned int
+__match32_all_sync(unsigned int mask, unsigned int value, int *pred) {
+ return __nvvm_match_all_sync_i32p(mask, value, pred);
+}
+
+inline __device__ unsigned long long
+__match64_all_sync(unsigned int mask, unsigned long long value, int *pred) {
+ return __nvvm_match_all_sync_i64p(mask, value, pred);
+}
+#include "crt/sm_70_rt.hpp"
+
+#endif // !defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 700
+#endif // __CUDA_VERSION >= 9000
+
// sm_32 intrinsics: __ldg and __funnelshift_{l,lc,r,rc}.
// Prevent the vanilla sm_32 intrinsics header from being included.
inline __device__ unsigned char __ldg(const unsigned char *ptr) {
return __nvvm_ldg_uc(ptr);
}
+inline __device__ signed char __ldg(const signed char *ptr) {
+ return __nvvm_ldg_uc((const unsigned char *)ptr);
+}
inline __device__ unsigned short __ldg(const unsigned short *ptr) {
return __nvvm_ldg_us(ptr);
}
--- /dev/null
+/*===-- __clang_cuda_libdevice_declares.h - decls for libdevice functions --===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __CLANG_CUDA_LIBDEVICE_DECLARES_H__
+#define __CLANG_CUDA_LIBDEVICE_DECLARES_H__
+
+#if defined(__cplusplus)
+extern "C" {
+#endif
+
+#if defined(__OPENMP_NVPTX__)
+#define __DEVICE__
+#elif defined(__CUDA__)
+#define __DEVICE__ __device__
+#endif
+
+__DEVICE__ int __nv_abs(int __a);
+__DEVICE__ double __nv_acos(double __a);
+__DEVICE__ float __nv_acosf(float __a);
+__DEVICE__ double __nv_acosh(double __a);
+__DEVICE__ float __nv_acoshf(float __a);
+__DEVICE__ double __nv_asin(double __a);
+__DEVICE__ float __nv_asinf(float __a);
+__DEVICE__ double __nv_asinh(double __a);
+__DEVICE__ float __nv_asinhf(float __a);
+__DEVICE__ double __nv_atan2(double __a, double __b);
+__DEVICE__ float __nv_atan2f(float __a, float __b);
+__DEVICE__ double __nv_atan(double __a);
+__DEVICE__ float __nv_atanf(float __a);
+__DEVICE__ double __nv_atanh(double __a);
+__DEVICE__ float __nv_atanhf(float __a);
+__DEVICE__ int __nv_brev(int __a);
+__DEVICE__ long long __nv_brevll(long long __a);
+__DEVICE__ int __nv_byte_perm(int __a, int __b, int __c);
+__DEVICE__ double __nv_cbrt(double __a);
+__DEVICE__ float __nv_cbrtf(float __a);
+__DEVICE__ double __nv_ceil(double __a);
+__DEVICE__ float __nv_ceilf(float __a);
+__DEVICE__ int __nv_clz(int __a);
+__DEVICE__ int __nv_clzll(long long __a);
+__DEVICE__ double __nv_copysign(double __a, double __b);
+__DEVICE__ float __nv_copysignf(float __a, float __b);
+__DEVICE__ double __nv_cos(double __a);
+__DEVICE__ float __nv_cosf(float __a);
+__DEVICE__ double __nv_cosh(double __a);
+__DEVICE__ float __nv_coshf(float __a);
+__DEVICE__ double __nv_cospi(double __a);
+__DEVICE__ float __nv_cospif(float __a);
+__DEVICE__ double __nv_cyl_bessel_i0(double __a);
+__DEVICE__ float __nv_cyl_bessel_i0f(float __a);
+__DEVICE__ double __nv_cyl_bessel_i1(double __a);
+__DEVICE__ float __nv_cyl_bessel_i1f(float __a);
+__DEVICE__ double __nv_dadd_rd(double __a, double __b);
+__DEVICE__ double __nv_dadd_rn(double __a, double __b);
+__DEVICE__ double __nv_dadd_ru(double __a, double __b);
+__DEVICE__ double __nv_dadd_rz(double __a, double __b);
+__DEVICE__ double __nv_ddiv_rd(double __a, double __b);
+__DEVICE__ double __nv_ddiv_rn(double __a, double __b);
+__DEVICE__ double __nv_ddiv_ru(double __a, double __b);
+__DEVICE__ double __nv_ddiv_rz(double __a, double __b);
+__DEVICE__ double __nv_dmul_rd(double __a, double __b);
+__DEVICE__ double __nv_dmul_rn(double __a, double __b);
+__DEVICE__ double __nv_dmul_ru(double __a, double __b);
+__DEVICE__ double __nv_dmul_rz(double __a, double __b);
+__DEVICE__ float __nv_double2float_rd(double __a);
+__DEVICE__ float __nv_double2float_rn(double __a);
+__DEVICE__ float __nv_double2float_ru(double __a);
+__DEVICE__ float __nv_double2float_rz(double __a);
+__DEVICE__ int __nv_double2hiint(double __a);
+__DEVICE__ int __nv_double2int_rd(double __a);
+__DEVICE__ int __nv_double2int_rn(double __a);
+__DEVICE__ int __nv_double2int_ru(double __a);
+__DEVICE__ int __nv_double2int_rz(double __a);
+__DEVICE__ long long __nv_double2ll_rd(double __a);
+__DEVICE__ long long __nv_double2ll_rn(double __a);
+__DEVICE__ long long __nv_double2ll_ru(double __a);
+__DEVICE__ long long __nv_double2ll_rz(double __a);
+__DEVICE__ int __nv_double2loint(double __a);
+__DEVICE__ unsigned int __nv_double2uint_rd(double __a);
+__DEVICE__ unsigned int __nv_double2uint_rn(double __a);
+__DEVICE__ unsigned int __nv_double2uint_ru(double __a);
+__DEVICE__ unsigned int __nv_double2uint_rz(double __a);
+__DEVICE__ unsigned long long __nv_double2ull_rd(double __a);
+__DEVICE__ unsigned long long __nv_double2ull_rn(double __a);
+__DEVICE__ unsigned long long __nv_double2ull_ru(double __a);
+__DEVICE__ unsigned long long __nv_double2ull_rz(double __a);
+__DEVICE__ unsigned long long __nv_double_as_longlong(double __a);
+__DEVICE__ double __nv_drcp_rd(double __a);
+__DEVICE__ double __nv_drcp_rn(double __a);
+__DEVICE__ double __nv_drcp_ru(double __a);
+__DEVICE__ double __nv_drcp_rz(double __a);
+__DEVICE__ double __nv_dsqrt_rd(double __a);
+__DEVICE__ double __nv_dsqrt_rn(double __a);
+__DEVICE__ double __nv_dsqrt_ru(double __a);
+__DEVICE__ double __nv_dsqrt_rz(double __a);
+__DEVICE__ double __nv_dsub_rd(double __a, double __b);
+__DEVICE__ double __nv_dsub_rn(double __a, double __b);
+__DEVICE__ double __nv_dsub_ru(double __a, double __b);
+__DEVICE__ double __nv_dsub_rz(double __a, double __b);
+__DEVICE__ double __nv_erfc(double __a);
+__DEVICE__ float __nv_erfcf(float __a);
+__DEVICE__ double __nv_erfcinv(double __a);
+__DEVICE__ float __nv_erfcinvf(float __a);
+__DEVICE__ double __nv_erfcx(double __a);
+__DEVICE__ float __nv_erfcxf(float __a);
+__DEVICE__ double __nv_erf(double __a);
+__DEVICE__ float __nv_erff(float __a);
+__DEVICE__ double __nv_erfinv(double __a);
+__DEVICE__ float __nv_erfinvf(float __a);
+__DEVICE__ double __nv_exp10(double __a);
+__DEVICE__ float __nv_exp10f(float __a);
+__DEVICE__ double __nv_exp2(double __a);
+__DEVICE__ float __nv_exp2f(float __a);
+__DEVICE__ double __nv_exp(double __a);
+__DEVICE__ float __nv_expf(float __a);
+__DEVICE__ double __nv_expm1(double __a);
+__DEVICE__ float __nv_expm1f(float __a);
+__DEVICE__ double __nv_fabs(double __a);
+__DEVICE__ float __nv_fabsf(float __a);
+__DEVICE__ float __nv_fadd_rd(float __a, float __b);
+__DEVICE__ float __nv_fadd_rn(float __a, float __b);
+__DEVICE__ float __nv_fadd_ru(float __a, float __b);
+__DEVICE__ float __nv_fadd_rz(float __a, float __b);
+__DEVICE__ float __nv_fast_cosf(float __a);
+__DEVICE__ float __nv_fast_exp10f(float __a);
+__DEVICE__ float __nv_fast_expf(float __a);
+__DEVICE__ float __nv_fast_fdividef(float __a, float __b);
+__DEVICE__ float __nv_fast_log10f(float __a);
+__DEVICE__ float __nv_fast_log2f(float __a);
+__DEVICE__ float __nv_fast_logf(float __a);
+__DEVICE__ float __nv_fast_powf(float __a, float __b);
+__DEVICE__ void __nv_fast_sincosf(float __a, float *__s, float *__c);
+__DEVICE__ float __nv_fast_sinf(float __a);
+__DEVICE__ float __nv_fast_tanf(float __a);
+__DEVICE__ double __nv_fdim(double __a, double __b);
+__DEVICE__ float __nv_fdimf(float __a, float __b);
+__DEVICE__ float __nv_fdiv_rd(float __a, float __b);
+__DEVICE__ float __nv_fdiv_rn(float __a, float __b);
+__DEVICE__ float __nv_fdiv_ru(float __a, float __b);
+__DEVICE__ float __nv_fdiv_rz(float __a, float __b);
+__DEVICE__ int __nv_ffs(int __a);
+__DEVICE__ int __nv_ffsll(long long __a);
+__DEVICE__ int __nv_finitef(float __a);
+__DEVICE__ unsigned short __nv_float2half_rn(float __a);
+__DEVICE__ int __nv_float2int_rd(float __a);
+__DEVICE__ int __nv_float2int_rn(float __a);
+__DEVICE__ int __nv_float2int_ru(float __a);
+__DEVICE__ int __nv_float2int_rz(float __a);
+__DEVICE__ long long __nv_float2ll_rd(float __a);
+__DEVICE__ long long __nv_float2ll_rn(float __a);
+__DEVICE__ long long __nv_float2ll_ru(float __a);
+__DEVICE__ long long __nv_float2ll_rz(float __a);
+__DEVICE__ unsigned int __nv_float2uint_rd(float __a);
+__DEVICE__ unsigned int __nv_float2uint_rn(float __a);
+__DEVICE__ unsigned int __nv_float2uint_ru(float __a);
+__DEVICE__ unsigned int __nv_float2uint_rz(float __a);
+__DEVICE__ unsigned long long __nv_float2ull_rd(float __a);
+__DEVICE__ unsigned long long __nv_float2ull_rn(float __a);
+__DEVICE__ unsigned long long __nv_float2ull_ru(float __a);
+__DEVICE__ unsigned long long __nv_float2ull_rz(float __a);
+__DEVICE__ int __nv_float_as_int(float __a);
+__DEVICE__ unsigned int __nv_float_as_uint(float __a);
+__DEVICE__ double __nv_floor(double __a);
+__DEVICE__ float __nv_floorf(float __a);
+__DEVICE__ double __nv_fma(double __a, double __b, double __c);
+__DEVICE__ float __nv_fmaf(float __a, float __b, float __c);
+__DEVICE__ float __nv_fmaf_ieee_rd(float __a, float __b, float __c);
+__DEVICE__ float __nv_fmaf_ieee_rn(float __a, float __b, float __c);
+__DEVICE__ float __nv_fmaf_ieee_ru(float __a, float __b, float __c);
+__DEVICE__ float __nv_fmaf_ieee_rz(float __a, float __b, float __c);
+__DEVICE__ float __nv_fmaf_rd(float __a, float __b, float __c);
+__DEVICE__ float __nv_fmaf_rn(float __a, float __b, float __c);
+__DEVICE__ float __nv_fmaf_ru(float __a, float __b, float __c);
+__DEVICE__ float __nv_fmaf_rz(float __a, float __b, float __c);
+__DEVICE__ double __nv_fma_rd(double __a, double __b, double __c);
+__DEVICE__ double __nv_fma_rn(double __a, double __b, double __c);
+__DEVICE__ double __nv_fma_ru(double __a, double __b, double __c);
+__DEVICE__ double __nv_fma_rz(double __a, double __b, double __c);
+__DEVICE__ double __nv_fmax(double __a, double __b);
+__DEVICE__ float __nv_fmaxf(float __a, float __b);
+__DEVICE__ double __nv_fmin(double __a, double __b);
+__DEVICE__ float __nv_fminf(float __a, float __b);
+__DEVICE__ double __nv_fmod(double __a, double __b);
+__DEVICE__ float __nv_fmodf(float __a, float __b);
+__DEVICE__ float __nv_fmul_rd(float __a, float __b);
+__DEVICE__ float __nv_fmul_rn(float __a, float __b);
+__DEVICE__ float __nv_fmul_ru(float __a, float __b);
+__DEVICE__ float __nv_fmul_rz(float __a, float __b);
+__DEVICE__ float __nv_frcp_rd(float __a);
+__DEVICE__ float __nv_frcp_rn(float __a);
+__DEVICE__ float __nv_frcp_ru(float __a);
+__DEVICE__ float __nv_frcp_rz(float __a);
+__DEVICE__ double __nv_frexp(double __a, int *__b);
+__DEVICE__ float __nv_frexpf(float __a, int *__b);
+__DEVICE__ float __nv_frsqrt_rn(float __a);
+__DEVICE__ float __nv_fsqrt_rd(float __a);
+__DEVICE__ float __nv_fsqrt_rn(float __a);
+__DEVICE__ float __nv_fsqrt_ru(float __a);
+__DEVICE__ float __nv_fsqrt_rz(float __a);
+__DEVICE__ float __nv_fsub_rd(float __a, float __b);
+__DEVICE__ float __nv_fsub_rn(float __a, float __b);
+__DEVICE__ float __nv_fsub_ru(float __a, float __b);
+__DEVICE__ float __nv_fsub_rz(float __a, float __b);
+__DEVICE__ int __nv_hadd(int __a, int __b);
+__DEVICE__ float __nv_half2float(unsigned short __h);
+__DEVICE__ double __nv_hiloint2double(int __a, int __b);
+__DEVICE__ double __nv_hypot(double __a, double __b);
+__DEVICE__ float __nv_hypotf(float __a, float __b);
+__DEVICE__ int __nv_ilogb(double __a);
+__DEVICE__ int __nv_ilogbf(float __a);
+__DEVICE__ double __nv_int2double_rn(int __a);
+__DEVICE__ float __nv_int2float_rd(int __a);
+__DEVICE__ float __nv_int2float_rn(int __a);
+__DEVICE__ float __nv_int2float_ru(int __a);
+__DEVICE__ float __nv_int2float_rz(int __a);
+__DEVICE__ float __nv_int_as_float(int __a);
+__DEVICE__ int __nv_isfinited(double __a);
+__DEVICE__ int __nv_isinfd(double __a);
+__DEVICE__ int __nv_isinff(float __a);
+__DEVICE__ int __nv_isnand(double __a);
+__DEVICE__ int __nv_isnanf(float __a);
+__DEVICE__ double __nv_j0(double __a);
+__DEVICE__ float __nv_j0f(float __a);
+__DEVICE__ double __nv_j1(double __a);
+__DEVICE__ float __nv_j1f(float __a);
+__DEVICE__ float __nv_jnf(int __a, float __b);
+__DEVICE__ double __nv_jn(int __a, double __b);
+__DEVICE__ double __nv_ldexp(double __a, int __b);
+__DEVICE__ float __nv_ldexpf(float __a, int __b);
+__DEVICE__ double __nv_lgamma(double __a);
+__DEVICE__ float __nv_lgammaf(float __a);
+__DEVICE__ double __nv_ll2double_rd(long long __a);
+__DEVICE__ double __nv_ll2double_rn(long long __a);
+__DEVICE__ double __nv_ll2double_ru(long long __a);
+__DEVICE__ double __nv_ll2double_rz(long long __a);
+__DEVICE__ float __nv_ll2float_rd(long long __a);
+__DEVICE__ float __nv_ll2float_rn(long long __a);
+__DEVICE__ float __nv_ll2float_ru(long long __a);
+__DEVICE__ float __nv_ll2float_rz(long long __a);
+__DEVICE__ long long __nv_llabs(long long __a);
+__DEVICE__ long long __nv_llmax(long long __a, long long __b);
+__DEVICE__ long long __nv_llmin(long long __a, long long __b);
+__DEVICE__ long long __nv_llrint(double __a);
+__DEVICE__ long long __nv_llrintf(float __a);
+__DEVICE__ long long __nv_llround(double __a);
+__DEVICE__ long long __nv_llroundf(float __a);
+__DEVICE__ double __nv_log10(double __a);
+__DEVICE__ float __nv_log10f(float __a);
+__DEVICE__ double __nv_log1p(double __a);
+__DEVICE__ float __nv_log1pf(float __a);
+__DEVICE__ double __nv_log2(double __a);
+__DEVICE__ float __nv_log2f(float __a);
+__DEVICE__ double __nv_logb(double __a);
+__DEVICE__ float __nv_logbf(float __a);
+__DEVICE__ double __nv_log(double __a);
+__DEVICE__ float __nv_logf(float __a);
+__DEVICE__ double __nv_longlong_as_double(long long __a);
+__DEVICE__ int __nv_max(int __a, int __b);
+__DEVICE__ int __nv_min(int __a, int __b);
+__DEVICE__ double __nv_modf(double __a, double *__b);
+__DEVICE__ float __nv_modff(float __a, float *__b);
+__DEVICE__ int __nv_mul24(int __a, int __b);
+__DEVICE__ long long __nv_mul64hi(long long __a, long long __b);
+__DEVICE__ int __nv_mulhi(int __a, int __b);
+__DEVICE__ double __nv_nan(const signed char *__a);
+__DEVICE__ float __nv_nanf(const signed char *__a);
+__DEVICE__ double __nv_nearbyint(double __a);
+__DEVICE__ float __nv_nearbyintf(float __a);
+__DEVICE__ double __nv_nextafter(double __a, double __b);
+__DEVICE__ float __nv_nextafterf(float __a, float __b);
+__DEVICE__ double __nv_norm3d(double __a, double __b, double __c);
+__DEVICE__ float __nv_norm3df(float __a, float __b, float __c);
+__DEVICE__ double __nv_norm4d(double __a, double __b, double __c, double __d);
+__DEVICE__ float __nv_norm4df(float __a, float __b, float __c, float __d);
+__DEVICE__ double __nv_normcdf(double __a);
+__DEVICE__ float __nv_normcdff(float __a);
+__DEVICE__ double __nv_normcdfinv(double __a);
+__DEVICE__ float __nv_normcdfinvf(float __a);
+__DEVICE__ float __nv_normf(int __a, const float *__b);
+__DEVICE__ double __nv_norm(int __a, const double *__b);
+__DEVICE__ int __nv_popc(int __a);
+__DEVICE__ int __nv_popcll(long long __a);
+__DEVICE__ double __nv_pow(double __a, double __b);
+__DEVICE__ float __nv_powf(float __a, float __b);
+__DEVICE__ double __nv_powi(double __a, int __b);
+__DEVICE__ float __nv_powif(float __a, int __b);
+__DEVICE__ double __nv_rcbrt(double __a);
+__DEVICE__ float __nv_rcbrtf(float __a);
+__DEVICE__ double __nv_rcp64h(double __a);
+__DEVICE__ double __nv_remainder(double __a, double __b);
+__DEVICE__ float __nv_remainderf(float __a, float __b);
+__DEVICE__ double __nv_remquo(double __a, double __b, int *__c);
+__DEVICE__ float __nv_remquof(float __a, float __b, int *__c);
+__DEVICE__ int __nv_rhadd(int __a, int __b);
+__DEVICE__ double __nv_rhypot(double __a, double __b);
+__DEVICE__ float __nv_rhypotf(float __a, float __b);
+__DEVICE__ double __nv_rint(double __a);
+__DEVICE__ float __nv_rintf(float __a);
+__DEVICE__ double __nv_rnorm3d(double __a, double __b, double __c);
+__DEVICE__ float __nv_rnorm3df(float __a, float __b, float __c);
+__DEVICE__ double __nv_rnorm4d(double __a, double __b, double __c, double __d);
+__DEVICE__ float __nv_rnorm4df(float __a, float __b, float __c, float __d);
+__DEVICE__ float __nv_rnormf(int __a, const float *__b);
+__DEVICE__ double __nv_rnorm(int __a, const double *__b);
+__DEVICE__ double __nv_round(double __a);
+__DEVICE__ float __nv_roundf(float __a);
+__DEVICE__ double __nv_rsqrt(double __a);
+__DEVICE__ float __nv_rsqrtf(float __a);
+__DEVICE__ int __nv_sad(int __a, int __b, int __c);
+__DEVICE__ float __nv_saturatef(float __a);
+__DEVICE__ double __nv_scalbn(double __a, int __b);
+__DEVICE__ float __nv_scalbnf(float __a, int __b);
+__DEVICE__ int __nv_signbitd(double __a);
+__DEVICE__ int __nv_signbitf(float __a);
+__DEVICE__ void __nv_sincos(double __a, double *__b, double *__c);
+__DEVICE__ void __nv_sincosf(float __a, float *__b, float *__c);
+__DEVICE__ void __nv_sincospi(double __a, double *__b, double *__c);
+__DEVICE__ void __nv_sincospif(float __a, float *__b, float *__c);
+__DEVICE__ double __nv_sin(double __a);
+__DEVICE__ float __nv_sinf(float __a);
+__DEVICE__ double __nv_sinh(double __a);
+__DEVICE__ float __nv_sinhf(float __a);
+__DEVICE__ double __nv_sinpi(double __a);
+__DEVICE__ float __nv_sinpif(float __a);
+__DEVICE__ double __nv_sqrt(double __a);
+__DEVICE__ float __nv_sqrtf(float __a);
+__DEVICE__ double __nv_tan(double __a);
+__DEVICE__ float __nv_tanf(float __a);
+__DEVICE__ double __nv_tanh(double __a);
+__DEVICE__ float __nv_tanhf(float __a);
+__DEVICE__ double __nv_tgamma(double __a);
+__DEVICE__ float __nv_tgammaf(float __a);
+__DEVICE__ double __nv_trunc(double __a);
+__DEVICE__ float __nv_truncf(float __a);
+__DEVICE__ int __nv_uhadd(unsigned int __a, unsigned int __b);
+__DEVICE__ double __nv_uint2double_rn(unsigned int __i);
+__DEVICE__ float __nv_uint2float_rd(unsigned int __a);
+__DEVICE__ float __nv_uint2float_rn(unsigned int __a);
+__DEVICE__ float __nv_uint2float_ru(unsigned int __a);
+__DEVICE__ float __nv_uint2float_rz(unsigned int __a);
+__DEVICE__ float __nv_uint_as_float(unsigned int __a);
+__DEVICE__ double __nv_ull2double_rd(unsigned long long __a);
+__DEVICE__ double __nv_ull2double_rn(unsigned long long __a);
+__DEVICE__ double __nv_ull2double_ru(unsigned long long __a);
+__DEVICE__ double __nv_ull2double_rz(unsigned long long __a);
+__DEVICE__ float __nv_ull2float_rd(unsigned long long __a);
+__DEVICE__ float __nv_ull2float_rn(unsigned long long __a);
+__DEVICE__ float __nv_ull2float_ru(unsigned long long __a);
+__DEVICE__ float __nv_ull2float_rz(unsigned long long __a);
+__DEVICE__ unsigned long long __nv_ullmax(unsigned long long __a,
+ unsigned long long __b);
+__DEVICE__ unsigned long long __nv_ullmin(unsigned long long __a,
+ unsigned long long __b);
+__DEVICE__ unsigned int __nv_umax(unsigned int __a, unsigned int __b);
+__DEVICE__ unsigned int __nv_umin(unsigned int __a, unsigned int __b);
+__DEVICE__ unsigned int __nv_umul24(unsigned int __a, unsigned int __b);
+__DEVICE__ unsigned long long __nv_umul64hi(unsigned long long __a,
+ unsigned long long __b);
+__DEVICE__ unsigned int __nv_umulhi(unsigned int __a, unsigned int __b);
+__DEVICE__ unsigned int __nv_urhadd(unsigned int __a, unsigned int __b);
+__DEVICE__ unsigned int __nv_usad(unsigned int __a, unsigned int __b,
+ unsigned int __c);
+#if CUDA_VERSION >= 9000 && CUDA_VERSION < 9020
+__DEVICE__ int __nv_vabs2(int __a);
+__DEVICE__ int __nv_vabs4(int __a);
+__DEVICE__ int __nv_vabsdiffs2(int __a, int __b);
+__DEVICE__ int __nv_vabsdiffs4(int __a, int __b);
+__DEVICE__ int __nv_vabsdiffu2(int __a, int __b);
+__DEVICE__ int __nv_vabsdiffu4(int __a, int __b);
+__DEVICE__ int __nv_vabsss2(int __a);
+__DEVICE__ int __nv_vabsss4(int __a);
+__DEVICE__ int __nv_vadd2(int __a, int __b);
+__DEVICE__ int __nv_vadd4(int __a, int __b);
+__DEVICE__ int __nv_vaddss2(int __a, int __b);
+__DEVICE__ int __nv_vaddss4(int __a, int __b);
+__DEVICE__ int __nv_vaddus2(int __a, int __b);
+__DEVICE__ int __nv_vaddus4(int __a, int __b);
+__DEVICE__ int __nv_vavgs2(int __a, int __b);
+__DEVICE__ int __nv_vavgs4(int __a, int __b);
+__DEVICE__ int __nv_vavgu2(int __a, int __b);
+__DEVICE__ int __nv_vavgu4(int __a, int __b);
+__DEVICE__ int __nv_vcmpeq2(int __a, int __b);
+__DEVICE__ int __nv_vcmpeq4(int __a, int __b);
+__DEVICE__ int __nv_vcmpges2(int __a, int __b);
+__DEVICE__ int __nv_vcmpges4(int __a, int __b);
+__DEVICE__ int __nv_vcmpgeu2(int __a, int __b);
+__DEVICE__ int __nv_vcmpgeu4(int __a, int __b);
+__DEVICE__ int __nv_vcmpgts2(int __a, int __b);
+__DEVICE__ int __nv_vcmpgts4(int __a, int __b);
+__DEVICE__ int __nv_vcmpgtu2(int __a, int __b);
+__DEVICE__ int __nv_vcmpgtu4(int __a, int __b);
+__DEVICE__ int __nv_vcmples2(int __a, int __b);
+__DEVICE__ int __nv_vcmples4(int __a, int __b);
+__DEVICE__ int __nv_vcmpleu2(int __a, int __b);
+__DEVICE__ int __nv_vcmpleu4(int __a, int __b);
+__DEVICE__ int __nv_vcmplts2(int __a, int __b);
+__DEVICE__ int __nv_vcmplts4(int __a, int __b);
+__DEVICE__ int __nv_vcmpltu2(int __a, int __b);
+__DEVICE__ int __nv_vcmpltu4(int __a, int __b);
+__DEVICE__ int __nv_vcmpne2(int __a, int __b);
+__DEVICE__ int __nv_vcmpne4(int __a, int __b);
+__DEVICE__ int __nv_vhaddu2(int __a, int __b);
+__DEVICE__ int __nv_vhaddu4(int __a, int __b);
+__DEVICE__ int __nv_vmaxs2(int __a, int __b);
+__DEVICE__ int __nv_vmaxs4(int __a, int __b);
+__DEVICE__ int __nv_vmaxu2(int __a, int __b);
+__DEVICE__ int __nv_vmaxu4(int __a, int __b);
+__DEVICE__ int __nv_vmins2(int __a, int __b);
+__DEVICE__ int __nv_vmins4(int __a, int __b);
+__DEVICE__ int __nv_vminu2(int __a, int __b);
+__DEVICE__ int __nv_vminu4(int __a, int __b);
+__DEVICE__ int __nv_vneg2(int __a);
+__DEVICE__ int __nv_vneg4(int __a);
+__DEVICE__ int __nv_vnegss2(int __a);
+__DEVICE__ int __nv_vnegss4(int __a);
+__DEVICE__ int __nv_vsads2(int __a, int __b);
+__DEVICE__ int __nv_vsads4(int __a, int __b);
+__DEVICE__ int __nv_vsadu2(int __a, int __b);
+__DEVICE__ int __nv_vsadu4(int __a, int __b);
+__DEVICE__ int __nv_vseteq2(int __a, int __b);
+__DEVICE__ int __nv_vseteq4(int __a, int __b);
+__DEVICE__ int __nv_vsetges2(int __a, int __b);
+__DEVICE__ int __nv_vsetges4(int __a, int __b);
+__DEVICE__ int __nv_vsetgeu2(int __a, int __b);
+__DEVICE__ int __nv_vsetgeu4(int __a, int __b);
+__DEVICE__ int __nv_vsetgts2(int __a, int __b);
+__DEVICE__ int __nv_vsetgts4(int __a, int __b);
+__DEVICE__ int __nv_vsetgtu2(int __a, int __b);
+__DEVICE__ int __nv_vsetgtu4(int __a, int __b);
+__DEVICE__ int __nv_vsetles2(int __a, int __b);
+__DEVICE__ int __nv_vsetles4(int __a, int __b);
+__DEVICE__ int __nv_vsetleu2(int __a, int __b);
+__DEVICE__ int __nv_vsetleu4(int __a, int __b);
+__DEVICE__ int __nv_vsetlts2(int __a, int __b);
+__DEVICE__ int __nv_vsetlts4(int __a, int __b);
+__DEVICE__ int __nv_vsetltu2(int __a, int __b);
+__DEVICE__ int __nv_vsetltu4(int __a, int __b);
+__DEVICE__ int __nv_vsetne2(int __a, int __b);
+__DEVICE__ int __nv_vsetne4(int __a, int __b);
+__DEVICE__ int __nv_vsub2(int __a, int __b);
+__DEVICE__ int __nv_vsub4(int __a, int __b);
+__DEVICE__ int __nv_vsubss2(int __a, int __b);
+__DEVICE__ int __nv_vsubss4(int __a, int __b);
+__DEVICE__ int __nv_vsubus2(int __a, int __b);
+__DEVICE__ int __nv_vsubus4(int __a, int __b);
+#endif // CUDA_VERSION
+__DEVICE__ double __nv_y0(double __a);
+__DEVICE__ float __nv_y0f(float __a);
+__DEVICE__ double __nv_y1(double __a);
+__DEVICE__ float __nv_y1f(float __a);
+__DEVICE__ float __nv_ynf(int __a, float __b);
+__DEVICE__ double __nv_yn(int __a, double __b);
+#if defined(__cplusplus)
+} // extern "C"
+#endif
+#endif // __CLANG_CUDA_LIBDEVICE_DECLARES_H__
--- /dev/null
+/*===---- __clang_cuda_math.h - Device-side CUDA math support --------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __CLANG_CUDA_MATH_H__
+#define __CLANG_CUDA_MATH_H__
+#ifndef __CUDA__
+#error "This file is for CUDA compilation only."
+#endif
+
+#ifndef __OPENMP_NVPTX__
+#if CUDA_VERSION < 9000
+#error This file is intended to be used with CUDA-9+ only.
+#endif
+#endif
+
+// __DEVICE__ is a helper macro with common set of attributes for the wrappers
+// we implement in this file. We need static in order to avoid emitting unused
+// functions and __forceinline__ helps inlining these wrappers at -O1.
+#pragma push_macro("__DEVICE__")
+#ifdef __OPENMP_NVPTX__
+#if defined(__cplusplus)
+#define __DEVICE__ static constexpr __attribute__((always_inline, nothrow))
+#else
+#define __DEVICE__ static __attribute__((always_inline, nothrow))
+#endif
+#else
+#define __DEVICE__ static __device__ __forceinline__
+#endif
+
+// Specialized version of __DEVICE__ for functions with void return type. Needed
+// because the OpenMP overlay requires constexpr functions here but prior to
+// c++14 void return functions could not be constexpr.
+#pragma push_macro("__DEVICE_VOID__")
+#ifdef __OPENMP_NVPTX__ && defined(__cplusplus) && __cplusplus < 201402L
+#define __DEVICE_VOID__ static __attribute__((always_inline, nothrow))
+#else
+#define __DEVICE_VOID__ __DEVICE__
+#endif
+
+// libdevice provides fast low precision and slow full-recision implementations
+// for some functions. Which one gets selected depends on
+// __CLANG_CUDA_APPROX_TRANSCENDENTALS__ which gets defined by clang if
+// -ffast-math or -fcuda-approx-transcendentals are in effect.
+#pragma push_macro("__FAST_OR_SLOW")
+#if defined(__CLANG_CUDA_APPROX_TRANSCENDENTALS__)
+#define __FAST_OR_SLOW(fast, slow) fast
+#else
+#define __FAST_OR_SLOW(fast, slow) slow
+#endif
+
+__DEVICE__ int abs(int __a) { return __nv_abs(__a); }
+__DEVICE__ double fabs(double __a) { return __nv_fabs(__a); }
+__DEVICE__ double acos(double __a) { return __nv_acos(__a); }
+__DEVICE__ float acosf(float __a) { return __nv_acosf(__a); }
+__DEVICE__ double acosh(double __a) { return __nv_acosh(__a); }
+__DEVICE__ float acoshf(float __a) { return __nv_acoshf(__a); }
+__DEVICE__ double asin(double __a) { return __nv_asin(__a); }
+__DEVICE__ float asinf(float __a) { return __nv_asinf(__a); }
+__DEVICE__ double asinh(double __a) { return __nv_asinh(__a); }
+__DEVICE__ float asinhf(float __a) { return __nv_asinhf(__a); }
+__DEVICE__ double atan(double __a) { return __nv_atan(__a); }
+__DEVICE__ double atan2(double __a, double __b) { return __nv_atan2(__a, __b); }
+__DEVICE__ float atan2f(float __a, float __b) { return __nv_atan2f(__a, __b); }
+__DEVICE__ float atanf(float __a) { return __nv_atanf(__a); }
+__DEVICE__ double atanh(double __a) { return __nv_atanh(__a); }
+__DEVICE__ float atanhf(float __a) { return __nv_atanhf(__a); }
+__DEVICE__ double cbrt(double __a) { return __nv_cbrt(__a); }
+__DEVICE__ float cbrtf(float __a) { return __nv_cbrtf(__a); }
+__DEVICE__ double ceil(double __a) { return __nv_ceil(__a); }
+__DEVICE__ float ceilf(float __a) { return __nv_ceilf(__a); }
+__DEVICE__ double copysign(double __a, double __b) {
+ return __nv_copysign(__a, __b);
+}
+__DEVICE__ float copysignf(float __a, float __b) {
+ return __nv_copysignf(__a, __b);
+}
+__DEVICE__ double cos(double __a) { return __nv_cos(__a); }
+__DEVICE__ float cosf(float __a) {
+ return __FAST_OR_SLOW(__nv_fast_cosf, __nv_cosf)(__a);
+}
+__DEVICE__ double cosh(double __a) { return __nv_cosh(__a); }
+__DEVICE__ float coshf(float __a) { return __nv_coshf(__a); }
+__DEVICE__ double cospi(double __a) { return __nv_cospi(__a); }
+__DEVICE__ float cospif(float __a) { return __nv_cospif(__a); }
+__DEVICE__ double cyl_bessel_i0(double __a) { return __nv_cyl_bessel_i0(__a); }
+__DEVICE__ float cyl_bessel_i0f(float __a) { return __nv_cyl_bessel_i0f(__a); }
+__DEVICE__ double cyl_bessel_i1(double __a) { return __nv_cyl_bessel_i1(__a); }
+__DEVICE__ float cyl_bessel_i1f(float __a) { return __nv_cyl_bessel_i1f(__a); }
+__DEVICE__ double erf(double __a) { return __nv_erf(__a); }
+__DEVICE__ double erfc(double __a) { return __nv_erfc(__a); }
+__DEVICE__ float erfcf(float __a) { return __nv_erfcf(__a); }
+__DEVICE__ double erfcinv(double __a) { return __nv_erfcinv(__a); }
+__DEVICE__ float erfcinvf(float __a) { return __nv_erfcinvf(__a); }
+__DEVICE__ double erfcx(double __a) { return __nv_erfcx(__a); }
+__DEVICE__ float erfcxf(float __a) { return __nv_erfcxf(__a); }
+__DEVICE__ float erff(float __a) { return __nv_erff(__a); }
+__DEVICE__ double erfinv(double __a) { return __nv_erfinv(__a); }
+__DEVICE__ float erfinvf(float __a) { return __nv_erfinvf(__a); }
+__DEVICE__ double exp(double __a) { return __nv_exp(__a); }
+__DEVICE__ double exp10(double __a) { return __nv_exp10(__a); }
+__DEVICE__ float exp10f(float __a) { return __nv_exp10f(__a); }
+__DEVICE__ double exp2(double __a) { return __nv_exp2(__a); }
+__DEVICE__ float exp2f(float __a) { return __nv_exp2f(__a); }
+__DEVICE__ float expf(float __a) { return __nv_expf(__a); }
+__DEVICE__ double expm1(double __a) { return __nv_expm1(__a); }
+__DEVICE__ float expm1f(float __a) { return __nv_expm1f(__a); }
+__DEVICE__ float fabsf(float __a) { return __nv_fabsf(__a); }
+__DEVICE__ double fdim(double __a, double __b) { return __nv_fdim(__a, __b); }
+__DEVICE__ float fdimf(float __a, float __b) { return __nv_fdimf(__a, __b); }
+__DEVICE__ double fdivide(double __a, double __b) { return __a / __b; }
+__DEVICE__ float fdividef(float __a, float __b) {
+#if __FAST_MATH__ && !__CUDA_PREC_DIV
+ return __nv_fast_fdividef(__a, __b);
+#else
+ return __a / __b;
+#endif
+}
+__DEVICE__ double floor(double __f) { return __nv_floor(__f); }
+__DEVICE__ float floorf(float __f) { return __nv_floorf(__f); }
+__DEVICE__ double fma(double __a, double __b, double __c) {
+ return __nv_fma(__a, __b, __c);
+}
+__DEVICE__ float fmaf(float __a, float __b, float __c) {
+ return __nv_fmaf(__a, __b, __c);
+}
+__DEVICE__ double fmax(double __a, double __b) { return __nv_fmax(__a, __b); }
+__DEVICE__ float fmaxf(float __a, float __b) { return __nv_fmaxf(__a, __b); }
+__DEVICE__ double fmin(double __a, double __b) { return __nv_fmin(__a, __b); }
+__DEVICE__ float fminf(float __a, float __b) { return __nv_fminf(__a, __b); }
+__DEVICE__ double fmod(double __a, double __b) { return __nv_fmod(__a, __b); }
+__DEVICE__ float fmodf(float __a, float __b) { return __nv_fmodf(__a, __b); }
+__DEVICE__ double frexp(double __a, int *__b) { return __nv_frexp(__a, __b); }
+__DEVICE__ float frexpf(float __a, int *__b) { return __nv_frexpf(__a, __b); }
+__DEVICE__ double hypot(double __a, double __b) { return __nv_hypot(__a, __b); }
+__DEVICE__ float hypotf(float __a, float __b) { return __nv_hypotf(__a, __b); }
+__DEVICE__ int ilogb(double __a) { return __nv_ilogb(__a); }
+__DEVICE__ int ilogbf(float __a) { return __nv_ilogbf(__a); }
+__DEVICE__ double j0(double __a) { return __nv_j0(__a); }
+__DEVICE__ float j0f(float __a) { return __nv_j0f(__a); }
+__DEVICE__ double j1(double __a) { return __nv_j1(__a); }
+__DEVICE__ float j1f(float __a) { return __nv_j1f(__a); }
+__DEVICE__ double jn(int __n, double __a) { return __nv_jn(__n, __a); }
+__DEVICE__ float jnf(int __n, float __a) { return __nv_jnf(__n, __a); }
+#if defined(__LP64__) || defined(_WIN64)
+__DEVICE__ long labs(long __a) { return __nv_llabs(__a); };
+#else
+__DEVICE__ long labs(long __a) { return __nv_abs(__a); };
+#endif
+__DEVICE__ double ldexp(double __a, int __b) { return __nv_ldexp(__a, __b); }
+__DEVICE__ float ldexpf(float __a, int __b) { return __nv_ldexpf(__a, __b); }
+__DEVICE__ double lgamma(double __a) { return __nv_lgamma(__a); }
+__DEVICE__ float lgammaf(float __a) { return __nv_lgammaf(__a); }
+__DEVICE__ long long llabs(long long __a) { return __nv_llabs(__a); }
+__DEVICE__ long long llmax(long long __a, long long __b) {
+ return __nv_llmax(__a, __b);
+}
+__DEVICE__ long long llmin(long long __a, long long __b) {
+ return __nv_llmin(__a, __b);
+}
+__DEVICE__ long long llrint(double __a) { return __nv_llrint(__a); }
+__DEVICE__ long long llrintf(float __a) { return __nv_llrintf(__a); }
+__DEVICE__ long long llround(double __a) { return __nv_llround(__a); }
+__DEVICE__ long long llroundf(float __a) { return __nv_llroundf(__a); }
+__DEVICE__ double log(double __a) { return __nv_log(__a); }
+__DEVICE__ double log10(double __a) { return __nv_log10(__a); }
+__DEVICE__ float log10f(float __a) { return __nv_log10f(__a); }
+__DEVICE__ double log1p(double __a) { return __nv_log1p(__a); }
+__DEVICE__ float log1pf(float __a) { return __nv_log1pf(__a); }
+__DEVICE__ double log2(double __a) { return __nv_log2(__a); }
+__DEVICE__ float log2f(float __a) {
+ return __FAST_OR_SLOW(__nv_fast_log2f, __nv_log2f)(__a);
+}
+__DEVICE__ double logb(double __a) { return __nv_logb(__a); }
+__DEVICE__ float logbf(float __a) { return __nv_logbf(__a); }
+__DEVICE__ float logf(float __a) {
+ return __FAST_OR_SLOW(__nv_fast_logf, __nv_logf)(__a);
+}
+#if defined(__LP64__) || defined(_WIN64)
+__DEVICE__ long lrint(double __a) { return llrint(__a); }
+__DEVICE__ long lrintf(float __a) { return __float2ll_rn(__a); }
+__DEVICE__ long lround(double __a) { return llround(__a); }
+__DEVICE__ long lroundf(float __a) { return llroundf(__a); }
+#else
+__DEVICE__ long lrint(double __a) { return (long)rint(__a); }
+__DEVICE__ long lrintf(float __a) { return __float2int_rn(__a); }
+__DEVICE__ long lround(double __a) { return round(__a); }
+__DEVICE__ long lroundf(float __a) { return roundf(__a); }
+#endif
+__DEVICE__ int max(int __a, int __b) { return __nv_max(__a, __b); }
+__DEVICE__ int min(int __a, int __b) { return __nv_min(__a, __b); }
+__DEVICE__ double modf(double __a, double *__b) { return __nv_modf(__a, __b); }
+__DEVICE__ float modff(float __a, float *__b) { return __nv_modff(__a, __b); }
+__DEVICE__ double nearbyint(double __a) { return __nv_nearbyint(__a); }
+__DEVICE__ float nearbyintf(float __a) { return __nv_nearbyintf(__a); }
+__DEVICE__ double nextafter(double __a, double __b) {
+ return __nv_nextafter(__a, __b);
+}
+__DEVICE__ float nextafterf(float __a, float __b) {
+ return __nv_nextafterf(__a, __b);
+}
+__DEVICE__ double norm(int __dim, const double *__t) {
+ return __nv_norm(__dim, __t);
+}
+__DEVICE__ double norm3d(double __a, double __b, double __c) {
+ return __nv_norm3d(__a, __b, __c);
+}
+__DEVICE__ float norm3df(float __a, float __b, float __c) {
+ return __nv_norm3df(__a, __b, __c);
+}
+__DEVICE__ double norm4d(double __a, double __b, double __c, double __d) {
+ return __nv_norm4d(__a, __b, __c, __d);
+}
+__DEVICE__ float norm4df(float __a, float __b, float __c, float __d) {
+ return __nv_norm4df(__a, __b, __c, __d);
+}
+__DEVICE__ double normcdf(double __a) { return __nv_normcdf(__a); }
+__DEVICE__ float normcdff(float __a) { return __nv_normcdff(__a); }
+__DEVICE__ double normcdfinv(double __a) { return __nv_normcdfinv(__a); }
+__DEVICE__ float normcdfinvf(float __a) { return __nv_normcdfinvf(__a); }
+__DEVICE__ float normf(int __dim, const float *__t) {
+ return __nv_normf(__dim, __t);
+}
+__DEVICE__ double pow(double __a, double __b) { return __nv_pow(__a, __b); }
+__DEVICE__ float powf(float __a, float __b) { return __nv_powf(__a, __b); }
+__DEVICE__ double powi(double __a, int __b) { return __nv_powi(__a, __b); }
+__DEVICE__ float powif(float __a, int __b) { return __nv_powif(__a, __b); }
+__DEVICE__ double rcbrt(double __a) { return __nv_rcbrt(__a); }
+__DEVICE__ float rcbrtf(float __a) { return __nv_rcbrtf(__a); }
+__DEVICE__ double remainder(double __a, double __b) {
+ return __nv_remainder(__a, __b);
+}
+__DEVICE__ float remainderf(float __a, float __b) {
+ return __nv_remainderf(__a, __b);
+}
+__DEVICE__ double remquo(double __a, double __b, int *__c) {
+ return __nv_remquo(__a, __b, __c);
+}
+__DEVICE__ float remquof(float __a, float __b, int *__c) {
+ return __nv_remquof(__a, __b, __c);
+}
+__DEVICE__ double rhypot(double __a, double __b) {
+ return __nv_rhypot(__a, __b);
+}
+__DEVICE__ float rhypotf(float __a, float __b) {
+ return __nv_rhypotf(__a, __b);
+}
+__DEVICE__ double rint(double __a) { return __nv_rint(__a); }
+__DEVICE__ float rintf(float __a) { return __nv_rintf(__a); }
+__DEVICE__ double rnorm(int __a, const double *__b) {
+ return __nv_rnorm(__a, __b);
+}
+__DEVICE__ double rnorm3d(double __a, double __b, double __c) {
+ return __nv_rnorm3d(__a, __b, __c);
+}
+__DEVICE__ float rnorm3df(float __a, float __b, float __c) {
+ return __nv_rnorm3df(__a, __b, __c);
+}
+__DEVICE__ double rnorm4d(double __a, double __b, double __c, double __d) {
+ return __nv_rnorm4d(__a, __b, __c, __d);
+}
+__DEVICE__ float rnorm4df(float __a, float __b, float __c, float __d) {
+ return __nv_rnorm4df(__a, __b, __c, __d);
+}
+__DEVICE__ float rnormf(int __dim, const float *__t) {
+ return __nv_rnormf(__dim, __t);
+}
+__DEVICE__ double round(double __a) { return __nv_round(__a); }
+__DEVICE__ float roundf(float __a) { return __nv_roundf(__a); }
+__DEVICE__ double rsqrt(double __a) { return __nv_rsqrt(__a); }
+__DEVICE__ float rsqrtf(float __a) { return __nv_rsqrtf(__a); }
+__DEVICE__ double scalbn(double __a, int __b) { return __nv_scalbn(__a, __b); }
+__DEVICE__ float scalbnf(float __a, int __b) { return __nv_scalbnf(__a, __b); }
+__DEVICE__ double scalbln(double __a, long __b) {
+ if (__b > INT_MAX)
+ return __a > 0 ? HUGE_VAL : -HUGE_VAL;
+ if (__b < INT_MIN)
+ return __a > 0 ? 0.0 : -0.0;
+ return scalbn(__a, (int)__b);
+}
+__DEVICE__ float scalblnf(float __a, long __b) {
+ if (__b > INT_MAX)
+ return __a > 0 ? HUGE_VALF : -HUGE_VALF;
+ if (__b < INT_MIN)
+ return __a > 0 ? 0.f : -0.f;
+ return scalbnf(__a, (int)__b);
+}
+__DEVICE__ double sin(double __a) { return __nv_sin(__a); }
+__DEVICE_VOID__ void sincos(double __a, double *__s, double *__c) {
+ return __nv_sincos(__a, __s, __c);
+}
+__DEVICE_VOID__ void sincosf(float __a, float *__s, float *__c) {
+ return __FAST_OR_SLOW(__nv_fast_sincosf, __nv_sincosf)(__a, __s, __c);
+}
+__DEVICE_VOID__ void sincospi(double __a, double *__s, double *__c) {
+ return __nv_sincospi(__a, __s, __c);
+}
+__DEVICE_VOID__ void sincospif(float __a, float *__s, float *__c) {
+ return __nv_sincospif(__a, __s, __c);
+}
+__DEVICE__ float sinf(float __a) {
+ return __FAST_OR_SLOW(__nv_fast_sinf, __nv_sinf)(__a);
+}
+__DEVICE__ double sinh(double __a) { return __nv_sinh(__a); }
+__DEVICE__ float sinhf(float __a) { return __nv_sinhf(__a); }
+__DEVICE__ double sinpi(double __a) { return __nv_sinpi(__a); }
+__DEVICE__ float sinpif(float __a) { return __nv_sinpif(__a); }
+__DEVICE__ double sqrt(double __a) { return __nv_sqrt(__a); }
+__DEVICE__ float sqrtf(float __a) { return __nv_sqrtf(__a); }
+__DEVICE__ double tan(double __a) { return __nv_tan(__a); }
+__DEVICE__ float tanf(float __a) { return __nv_tanf(__a); }
+__DEVICE__ double tanh(double __a) { return __nv_tanh(__a); }
+__DEVICE__ float tanhf(float __a) { return __nv_tanhf(__a); }
+__DEVICE__ double tgamma(double __a) { return __nv_tgamma(__a); }
+__DEVICE__ float tgammaf(float __a) { return __nv_tgammaf(__a); }
+__DEVICE__ double trunc(double __a) { return __nv_trunc(__a); }
+__DEVICE__ float truncf(float __a) { return __nv_truncf(__a); }
+__DEVICE__ unsigned long long ullmax(unsigned long long __a,
+ unsigned long long __b) {
+ return __nv_ullmax(__a, __b);
+}
+__DEVICE__ unsigned long long ullmin(unsigned long long __a,
+ unsigned long long __b) {
+ return __nv_ullmin(__a, __b);
+}
+__DEVICE__ unsigned int umax(unsigned int __a, unsigned int __b) {
+ return __nv_umax(__a, __b);
+}
+__DEVICE__ unsigned int umin(unsigned int __a, unsigned int __b) {
+ return __nv_umin(__a, __b);
+}
+__DEVICE__ double y0(double __a) { return __nv_y0(__a); }
+__DEVICE__ float y0f(float __a) { return __nv_y0f(__a); }
+__DEVICE__ double y1(double __a) { return __nv_y1(__a); }
+__DEVICE__ float y1f(float __a) { return __nv_y1f(__a); }
+__DEVICE__ double yn(int __a, double __b) { return __nv_yn(__a, __b); }
+__DEVICE__ float ynf(int __a, float __b) { return __nv_ynf(__a, __b); }
+
+#pragma pop_macro("__DEVICE__")
+#pragma pop_macro("__DEVICE_VOID__")
+#pragma pop_macro("__FAST_OR_SLOW")
+
+#endif // __CLANG_CUDA_DEVICE_FUNCTIONS_H__
/*===- __clang_math_forward_declares.h - Prototypes of __device__ math fns --===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#ifndef __CLANG__CUDA_MATH_FORWARD_DECLARES_H__
#define __CLANG__CUDA_MATH_FORWARD_DECLARES_H__
-#ifndef __CUDA__
-#error "This file is for CUDA compilation only."
+#if !defined(__CUDA__) && !__HIP__
+#error "This file is for CUDA/HIP compilation only."
#endif
// This file forward-declares of some math functions we (or the CUDA headers)
#define __DEVICE__ \
static __inline__ __attribute__((always_inline)) __attribute__((device))
+__DEVICE__ long abs(long);
+__DEVICE__ long long abs(long long);
__DEVICE__ double abs(double);
__DEVICE__ float abs(float);
__DEVICE__ int abs(int);
-__DEVICE__ long abs(long);
-__DEVICE__ long long abs(long long);
__DEVICE__ double acos(double);
__DEVICE__ float acos(float);
__DEVICE__ double acosh(double);
__DEVICE__ float hypot(float, float);
__DEVICE__ int ilogb(double);
__DEVICE__ int ilogb(float);
+#ifdef _MSC_VER
+__DEVICE__ bool isfinite(long double);
+#endif
__DEVICE__ bool isfinite(double);
__DEVICE__ bool isfinite(float);
__DEVICE__ bool isgreater(double, double);
__DEVICE__ bool isgreaterequal(double, double);
__DEVICE__ bool isgreaterequal(float, float);
__DEVICE__ bool isgreater(float, float);
+#ifdef _MSC_VER
+__DEVICE__ bool isinf(long double);
+#endif
__DEVICE__ bool isinf(double);
__DEVICE__ bool isinf(float);
__DEVICE__ bool isless(double, double);
__DEVICE__ bool isless(float, float);
__DEVICE__ bool islessgreater(double, double);
__DEVICE__ bool islessgreater(float, float);
+#ifdef _MSC_VER
+__DEVICE__ bool isnan(long double);
+#endif
__DEVICE__ bool isnan(double);
__DEVICE__ bool isnan(float);
__DEVICE__ bool isnormal(double);
__DEVICE__ float nearbyint(float);
__DEVICE__ double nextafter(double, double);
__DEVICE__ float nextafter(float, float);
-__DEVICE__ double nexttoward(double, double);
-__DEVICE__ float nexttoward(float, double);
-__DEVICE__ float nexttowardf(float, double);
__DEVICE__ double pow(double, double);
__DEVICE__ double pow(double, int);
__DEVICE__ float pow(float, float);
__DEVICE__ double trunc(double);
__DEVICE__ float trunc(float);
+// Notably missing above is nexttoward, which we don't define on
+// the device side because libdevice doesn't give us an implementation, and we
+// don't want to be in the business of writing one ourselves.
+
// We need to define these overloads in exactly the namespace our standard
// library uses (including the right inline namespace), otherwise they won't be
// picked up by other functions in the standard library (e.g. functions in
using ::nanf;
using ::nearbyint;
using ::nextafter;
-using ::nexttoward;
using ::pow;
using ::remainder;
using ::remquo;
/*===---- __clang_cuda_runtime_wrapper.h - CUDA runtime support -------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
// Include some forward declares that must come before cmath.
#include <__clang_cuda_math_forward_declares.h>
+// Define __CUDACC__ early as libstdc++ standard headers with GNU extensions
+// enabled depend on it to avoid using __float128, which is unsupported in
+// CUDA.
+#define __CUDACC__
+
// Include some standard headers to avoid CUDA headers including them
// while some required macros (like __THROW) are in a weird state.
#include <cmath>
#include <cstdlib>
#include <stdlib.h>
+#undef __CUDACC__
// Preserve common macros that will be changed below by us or by CUDA
// headers.
#include "cuda.h"
#if !defined(CUDA_VERSION)
#error "cuda.h did not define CUDA_VERSION"
-#elif CUDA_VERSION < 7000 || CUDA_VERSION > 8000
+#elif CUDA_VERSION < 7000
#error "Unsupported CUDA version!"
#endif
+#pragma push_macro("__CUDA_INCLUDE_COMPILER_INTERNAL_HEADERS__")
+#if CUDA_VERSION >= 10000
+#define __CUDA_INCLUDE_COMPILER_INTERNAL_HEADERS__
+#endif
+
// Make largest subset of device functions available during host
// compilation -- SM_35 for the time being.
#ifndef __CUDA_ARCH__
#define __DEVICE_FUNCTIONS_H__
#define __MATH_FUNCTIONS_H__
#define __COMMON_FUNCTIONS_H__
+// device_functions_decls is replaced by __clang_cuda_device_functions.h
+// included below.
+#define __DEVICE_FUNCTIONS_DECLS_H__
#undef __CUDACC__
+#if CUDA_VERSION < 9000
#define __CUDABE__
+#else
+#define __CUDACC__
+#define __CUDA_LIBDEVICE__
+#endif
// Disables definitions of device-side runtime support stubs in
// cuda_device_runtime_api.h
+#include "host_defines.h"
+#undef __CUDACC__
#include "driver_types.h"
#include "host_config.h"
-#include "host_defines.h"
+// Temporarily replace "nv_weak" with weak, so __attribute__((nv_weak)) in
+// cuda_device_runtime_api.h ends up being __attribute__((weak)) which is the
+// functional equivalent of what we need.
+#pragma push_macro("nv_weak")
+#define nv_weak weak
#undef __CUDABE__
+#undef __CUDA_LIBDEVICE__
#define __CUDACC__
#include "cuda_runtime.h"
+#pragma pop_macro("nv_weak")
#undef __CUDACC__
#define __CUDABE__
#define __nvvm_memcpy(s, d, n, a) __builtin_memcpy(s, d, n)
#define __nvvm_memset(d, c, n, a) __builtin_memset(d, c, n)
+#if CUDA_VERSION < 9000
#include "crt/device_runtime.h"
+#endif
#include "crt/host_runtime.h"
// device_runtime.h defines __cxa_* macros that will conflict with
// cxxabi.h.
}
#endif
-// We need decls for functions in CUDA's libdevice with __device__
-// attribute only. Alas they come either as __host__ __device__ or
-// with no attributes at all. To work around that, define __CUDA_RTC__
-// which produces HD variant and undef __host__ which gives us desided
-// decls with __device__ attribute.
-#pragma push_macro("__host__")
-#define __host__
-#define __CUDACC_RTC__
-#include "device_functions_decls.h"
-#undef __CUDACC_RTC__
+// CUDA 9.1 no longer provides declarations for libdevice functions, so we need
+// to provide our own.
+#include <__clang_cuda_libdevice_declares.h>
-// Temporarily poison __host__ macro to ensure it's not used by any of
-// the headers we're about to include.
-#define __host__ UNEXPECTED_HOST_ATTRIBUTE
+// Wrappers for many device-side standard library functions, incl. math
+// functions, became compiler builtins in CUDA-9 and have been removed from the
+// CUDA headers. Clang now provides its own implementation of the wrappers.
+#if CUDA_VERSION >= 9000
+#include <__clang_cuda_device_functions.h>
+#include <__clang_cuda_math.h>
+#endif
+
+// __THROW is redefined to be empty by device_functions_decls.h in CUDA. Clang's
+// counterpart does not do it, so we need to make it empty here to keep
+// following CUDA includes happy.
+#undef __THROW
+#define __THROW
// CUDA 8.0.41 relies on __USE_FAST_MATH__ and __CUDA_PREC_DIV's values.
// Previous versions used to check whether they are defined or not.
#endif
#endif
+// Temporarily poison __host__ macro to ensure it's not used by any of
+// the headers we're about to include.
+#pragma push_macro("__host__")
+#define __host__ UNEXPECTED_HOST_ATTRIBUTE
+
// device_functions.hpp and math_functions*.hpp use 'static
// __forceinline__' (with no __device__) for definitions of device
// functions. Temporarily redefine __forceinline__ to include
// __device__.
#pragma push_macro("__forceinline__")
#define __forceinline__ __device__ __inline__ __attribute__((always_inline))
+#if CUDA_VERSION < 9000
#include "device_functions.hpp"
+#endif
// math_function.hpp uses the __USE_FAST_MATH__ macro to determine whether we
// get the slow-but-accurate or fast-but-inaccurate versions of functions like
#if defined(__CLANG_CUDA_APPROX_TRANSCENDENTALS__)
#define __USE_FAST_MATH__ 1
#endif
+
+#if CUDA_VERSION >= 9000
+// CUDA-9.2 needs host-side memcpy for some host functions in
+// device_functions.hpp
+#if CUDA_VERSION >= 9020
+#include <string.h>
+#endif
+#include "crt/math_functions.hpp"
+#else
#include "math_functions.hpp"
+#endif
+
#pragma pop_macro("__USE_FAST_MATH__")
+#if CUDA_VERSION < 9000
#include "math_functions_dbl_ptx3.hpp"
+#endif
#pragma pop_macro("__forceinline__")
// Pull in host-only functions that are only available when neither
// __CUDACC__ nor __CUDABE__ are defined.
#undef __MATH_FUNCTIONS_HPP__
#undef __CUDABE__
+#if CUDA_VERSION < 9000
#include "math_functions.hpp"
+#endif
// Alas, additional overloads for these functions are hard to get to.
// Considering that we only need these overloads for a few functions,
// we can provide them here.
static inline float normcdf(float __a) { return normcdff(__a); }
static inline float erfcx(float __a) { return erfcxf(__a); }
+#if CUDA_VERSION < 9000
// For some reason single-argument variant is not always declared by
// CUDA headers. Alas, device_functions.hpp included below needs it.
static inline __device__ void __brkpt(int __c) { __brkpt(); }
+#endif
// Now include *.hpp with definitions of various GPU functions. Alas,
// a lot of thins get declared/defined with __host__ attribute which
// we don't want and we have to define it out. We also have to include
// {device,math}_functions.hpp again in order to extract the other
// branch of #if/else inside.
-
#define __host__
#undef __CUDABE__
#define __CUDACC__
+#if CUDA_VERSION >= 9000
+// Some atomic functions became compiler builtins in CUDA-9 , so we need their
+// declarations.
+#include "device_atomic_functions.h"
+#endif
#undef __DEVICE_FUNCTIONS_HPP__
#include "device_atomic_functions.hpp"
+#if CUDA_VERSION >= 9000
+#include "crt/device_functions.hpp"
+#include "crt/device_double_functions.hpp"
+#else
#include "device_functions.hpp"
+#define __CUDABE__
+#include "device_double_functions.h"
+#undef __CUDABE__
+#endif
#include "sm_20_atomic_functions.hpp"
#include "sm_20_intrinsics.hpp"
#include "sm_32_atomic_functions.hpp"
// reason about our code.
#if CUDA_VERSION >= 8000
+#pragma push_macro("__CUDA_ARCH__")
+#undef __CUDA_ARCH__
#include "sm_60_atomic_functions.hpp"
#include "sm_61_intrinsics.hpp"
+#pragma pop_macro("__CUDA_ARCH__")
#endif
#undef __MATH_FUNCTIONS_HPP__
#pragma push_macro("__GNUC__")
#undef __GNUC__
#define signbit __ignored_cuda_signbit
+
+// CUDA-9 omits device-side definitions of some math functions if it sees
+// include guard from math.h wrapper from libstdc++. We have to undo the header
+// guard temporarily to get the definitions we need.
+#pragma push_macro("_GLIBCXX_MATH_H")
+#pragma push_macro("_LIBCPP_VERSION")
+#if CUDA_VERSION >= 9000
+#undef _GLIBCXX_MATH_H
+// We also need to undo another guard that checks for libc++ 3.8+
+#ifdef _LIBCPP_VERSION
+#define _LIBCPP_VERSION 3700
+#endif
+#endif
+
+#if CUDA_VERSION >= 9000
+#include "crt/math_functions.hpp"
+#else
#include "math_functions.hpp"
+#endif
+#pragma pop_macro("_GLIBCXX_MATH_H")
+#pragma pop_macro("_LIBCPP_VERSION")
#pragma pop_macro("__GNUC__")
#pragma pop_macro("signbit")
#pragma pop_macro("dim3")
#pragma pop_macro("uint3")
#pragma pop_macro("__USE_FAST_MATH__")
+#pragma pop_macro("__CUDA_INCLUDE_COMPILER_INTERNAL_HEADERS__")
+
+// CUDA runtime uses this undocumented function to access kernel launch
+// configuration. The declaration is in crt/device_functions.h but that file
+// includes a lot of other stuff we don't want. Instead, we'll provide our own
+// declaration for it here.
+#if CUDA_VERSION >= 9020
+extern "C" unsigned __cudaPushCallConfiguration(dim3 gridDim, dim3 blockDim,
+ size_t sharedMem = 0,
+ void *stream = 0);
+#endif
#endif // __CUDA__
#endif // __CLANG_CUDA_RUNTIME_WRAPPER_H__
--- /dev/null
+/*===---- __clang_hip_libdevice_declares.h - HIP device library decls -------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __CLANG_HIP_LIBDEVICE_DECLARES_H__
+#define __CLANG_HIP_LIBDEVICE_DECLARES_H__
+
+extern "C" {
+
+// BEGIN FLOAT
+__device__ __attribute__((const)) float __ocml_acos_f32(float);
+__device__ __attribute__((pure)) float __ocml_acosh_f32(float);
+__device__ __attribute__((const)) float __ocml_asin_f32(float);
+__device__ __attribute__((pure)) float __ocml_asinh_f32(float);
+__device__ __attribute__((const)) float __ocml_atan2_f32(float, float);
+__device__ __attribute__((const)) float __ocml_atan_f32(float);
+__device__ __attribute__((pure)) float __ocml_atanh_f32(float);
+__device__ __attribute__((pure)) float __ocml_cbrt_f32(float);
+__device__ __attribute__((const)) float __ocml_ceil_f32(float);
+__device__ __attribute__((const)) __device__ float __ocml_copysign_f32(float,
+ float);
+__device__ float __ocml_cos_f32(float);
+__device__ float __ocml_native_cos_f32(float);
+__device__ __attribute__((pure)) __device__ float __ocml_cosh_f32(float);
+__device__ float __ocml_cospi_f32(float);
+__device__ float __ocml_i0_f32(float);
+__device__ float __ocml_i1_f32(float);
+__device__ __attribute__((pure)) float __ocml_erfc_f32(float);
+__device__ __attribute__((pure)) float __ocml_erfcinv_f32(float);
+__device__ __attribute__((pure)) float __ocml_erfcx_f32(float);
+__device__ __attribute__((pure)) float __ocml_erf_f32(float);
+__device__ __attribute__((pure)) float __ocml_erfinv_f32(float);
+__device__ __attribute__((pure)) float __ocml_exp10_f32(float);
+__device__ __attribute__((pure)) float __ocml_native_exp10_f32(float);
+__device__ __attribute__((pure)) float __ocml_exp2_f32(float);
+__device__ __attribute__((pure)) float __ocml_exp_f32(float);
+__device__ __attribute__((pure)) float __ocml_native_exp_f32(float);
+__device__ __attribute__((pure)) float __ocml_expm1_f32(float);
+__device__ __attribute__((const)) float __ocml_fabs_f32(float);
+__device__ __attribute__((const)) float __ocml_fdim_f32(float, float);
+__device__ __attribute__((const)) float __ocml_floor_f32(float);
+__device__ __attribute__((const)) float __ocml_fma_f32(float, float, float);
+__device__ __attribute__((const)) float __ocml_fmax_f32(float, float);
+__device__ __attribute__((const)) float __ocml_fmin_f32(float, float);
+__device__ __attribute__((const)) __device__ float __ocml_fmod_f32(float,
+ float);
+__device__ float __ocml_frexp_f32(float,
+ __attribute__((address_space(5))) int *);
+__device__ __attribute__((const)) float __ocml_hypot_f32(float, float);
+__device__ __attribute__((const)) int __ocml_ilogb_f32(float);
+__device__ __attribute__((const)) int __ocml_isfinite_f32(float);
+__device__ __attribute__((const)) int __ocml_isinf_f32(float);
+__device__ __attribute__((const)) int __ocml_isnan_f32(float);
+__device__ float __ocml_j0_f32(float);
+__device__ float __ocml_j1_f32(float);
+__device__ __attribute__((const)) float __ocml_ldexp_f32(float, int);
+__device__ float __ocml_lgamma_f32(float);
+__device__ __attribute__((pure)) float __ocml_log10_f32(float);
+__device__ __attribute__((pure)) float __ocml_native_log10_f32(float);
+__device__ __attribute__((pure)) float __ocml_log1p_f32(float);
+__device__ __attribute__((pure)) float __ocml_log2_f32(float);
+__device__ __attribute__((pure)) float __ocml_native_log2_f32(float);
+__device__ __attribute__((const)) float __ocml_logb_f32(float);
+__device__ __attribute__((pure)) float __ocml_log_f32(float);
+__device__ __attribute__((pure)) float __ocml_native_log_f32(float);
+__device__ float __ocml_modf_f32(float,
+ __attribute__((address_space(5))) float *);
+__device__ __attribute__((const)) float __ocml_nearbyint_f32(float);
+__device__ __attribute__((const)) float __ocml_nextafter_f32(float, float);
+__device__ __attribute__((const)) float __ocml_len3_f32(float, float, float);
+__device__ __attribute__((const)) float __ocml_len4_f32(float, float, float,
+ float);
+__device__ __attribute__((pure)) float __ocml_ncdf_f32(float);
+__device__ __attribute__((pure)) float __ocml_ncdfinv_f32(float);
+__device__ __attribute__((pure)) float __ocml_pow_f32(float, float);
+__device__ __attribute__((pure)) float __ocml_rcbrt_f32(float);
+__device__ __attribute__((const)) float __ocml_remainder_f32(float, float);
+__device__ float __ocml_remquo_f32(float, float,
+ __attribute__((address_space(5))) int *);
+__device__ __attribute__((const)) float __ocml_rhypot_f32(float, float);
+__device__ __attribute__((const)) float __ocml_rint_f32(float);
+__device__ __attribute__((const)) float __ocml_rlen3_f32(float, float, float);
+__device__ __attribute__((const)) float __ocml_rlen4_f32(float, float, float,
+ float);
+__device__ __attribute__((const)) float __ocml_round_f32(float);
+__device__ __attribute__((pure)) float __ocml_rsqrt_f32(float);
+__device__ __attribute__((const)) float __ocml_scalb_f32(float, float);
+__device__ __attribute__((const)) float __ocml_scalbn_f32(float, int);
+__device__ __attribute__((const)) int __ocml_signbit_f32(float);
+__device__ float __ocml_sincos_f32(float,
+ __attribute__((address_space(5))) float *);
+__device__ float __ocml_sincospi_f32(float,
+ __attribute__((address_space(5))) float *);
+__device__ float __ocml_sin_f32(float);
+__device__ float __ocml_native_sin_f32(float);
+__device__ __attribute__((pure)) float __ocml_sinh_f32(float);
+__device__ float __ocml_sinpi_f32(float);
+__device__ __attribute__((const)) float __ocml_sqrt_f32(float);
+__device__ __attribute__((const)) float __ocml_native_sqrt_f32(float);
+__device__ float __ocml_tan_f32(float);
+__device__ __attribute__((pure)) float __ocml_tanh_f32(float);
+__device__ float __ocml_tgamma_f32(float);
+__device__ __attribute__((const)) float __ocml_trunc_f32(float);
+__device__ float __ocml_y0_f32(float);
+__device__ float __ocml_y1_f32(float);
+
+// BEGIN INTRINSICS
+__device__ __attribute__((const)) float __ocml_add_rte_f32(float, float);
+__device__ __attribute__((const)) float __ocml_add_rtn_f32(float, float);
+__device__ __attribute__((const)) float __ocml_add_rtp_f32(float, float);
+__device__ __attribute__((const)) float __ocml_add_rtz_f32(float, float);
+__device__ __attribute__((const)) float __ocml_sub_rte_f32(float, float);
+__device__ __attribute__((const)) float __ocml_sub_rtn_f32(float, float);
+__device__ __attribute__((const)) float __ocml_sub_rtp_f32(float, float);
+__device__ __attribute__((const)) float __ocml_sub_rtz_f32(float, float);
+__device__ __attribute__((const)) float __ocml_mul_rte_f32(float, float);
+__device__ __attribute__((const)) float __ocml_mul_rtn_f32(float, float);
+__device__ __attribute__((const)) float __ocml_mul_rtp_f32(float, float);
+__device__ __attribute__((const)) float __ocml_mul_rtz_f32(float, float);
+__device__ __attribute__((const)) float __ocml_div_rte_f32(float, float);
+__device__ __attribute__((const)) float __ocml_div_rtn_f32(float, float);
+__device__ __attribute__((const)) float __ocml_div_rtp_f32(float, float);
+__device__ __attribute__((const)) float __ocml_div_rtz_f32(float, float);
+__device__ __attribute__((const)) float __ocml_sqrt_rte_f32(float, float);
+__device__ __attribute__((const)) float __ocml_sqrt_rtn_f32(float, float);
+__device__ __attribute__((const)) float __ocml_sqrt_rtp_f32(float, float);
+__device__ __attribute__((const)) float __ocml_sqrt_rtz_f32(float, float);
+__device__ __attribute__((const)) float __ocml_fma_rte_f32(float, float, float);
+__device__ __attribute__((const)) float __ocml_fma_rtn_f32(float, float, float);
+__device__ __attribute__((const)) float __ocml_fma_rtp_f32(float, float, float);
+__device__ __attribute__((const)) float __ocml_fma_rtz_f32(float, float, float);
+
+__device__ __attribute__((const)) float
+__llvm_amdgcn_cos_f32(float) __asm("llvm.amdgcn.cos.f32");
+__device__ __attribute__((const)) float
+__llvm_amdgcn_rcp_f32(float) __asm("llvm.amdgcn.rcp.f32");
+__device__ __attribute__((const)) float
+__llvm_amdgcn_rsq_f32(float) __asm("llvm.amdgcn.rsq.f32");
+__device__ __attribute__((const)) float
+__llvm_amdgcn_sin_f32(float) __asm("llvm.amdgcn.sin.f32");
+// END INTRINSICS
+// END FLOAT
+
+// BEGIN DOUBLE
+__device__ __attribute__((const)) double __ocml_acos_f64(double);
+__device__ __attribute__((pure)) double __ocml_acosh_f64(double);
+__device__ __attribute__((const)) double __ocml_asin_f64(double);
+__device__ __attribute__((pure)) double __ocml_asinh_f64(double);
+__device__ __attribute__((const)) double __ocml_atan2_f64(double, double);
+__device__ __attribute__((const)) double __ocml_atan_f64(double);
+__device__ __attribute__((pure)) double __ocml_atanh_f64(double);
+__device__ __attribute__((pure)) double __ocml_cbrt_f64(double);
+__device__ __attribute__((const)) double __ocml_ceil_f64(double);
+__device__ __attribute__((const)) double __ocml_copysign_f64(double, double);
+__device__ double __ocml_cos_f64(double);
+__device__ __attribute__((pure)) double __ocml_cosh_f64(double);
+__device__ double __ocml_cospi_f64(double);
+__device__ double __ocml_i0_f64(double);
+__device__ double __ocml_i1_f64(double);
+__device__ __attribute__((pure)) double __ocml_erfc_f64(double);
+__device__ __attribute__((pure)) double __ocml_erfcinv_f64(double);
+__device__ __attribute__((pure)) double __ocml_erfcx_f64(double);
+__device__ __attribute__((pure)) double __ocml_erf_f64(double);
+__device__ __attribute__((pure)) double __ocml_erfinv_f64(double);
+__device__ __attribute__((pure)) double __ocml_exp10_f64(double);
+__device__ __attribute__((pure)) double __ocml_exp2_f64(double);
+__device__ __attribute__((pure)) double __ocml_exp_f64(double);
+__device__ __attribute__((pure)) double __ocml_expm1_f64(double);
+__device__ __attribute__((const)) double __ocml_fabs_f64(double);
+__device__ __attribute__((const)) double __ocml_fdim_f64(double, double);
+__device__ __attribute__((const)) double __ocml_floor_f64(double);
+__device__ __attribute__((const)) double __ocml_fma_f64(double, double, double);
+__device__ __attribute__((const)) double __ocml_fmax_f64(double, double);
+__device__ __attribute__((const)) double __ocml_fmin_f64(double, double);
+__device__ __attribute__((const)) double __ocml_fmod_f64(double, double);
+__device__ double __ocml_frexp_f64(double,
+ __attribute__((address_space(5))) int *);
+__device__ __attribute__((const)) double __ocml_hypot_f64(double, double);
+__device__ __attribute__((const)) int __ocml_ilogb_f64(double);
+__device__ __attribute__((const)) int __ocml_isfinite_f64(double);
+__device__ __attribute__((const)) int __ocml_isinf_f64(double);
+__device__ __attribute__((const)) int __ocml_isnan_f64(double);
+__device__ double __ocml_j0_f64(double);
+__device__ double __ocml_j1_f64(double);
+__device__ __attribute__((const)) double __ocml_ldexp_f64(double, int);
+__device__ double __ocml_lgamma_f64(double);
+__device__ __attribute__((pure)) double __ocml_log10_f64(double);
+__device__ __attribute__((pure)) double __ocml_log1p_f64(double);
+__device__ __attribute__((pure)) double __ocml_log2_f64(double);
+__device__ __attribute__((const)) double __ocml_logb_f64(double);
+__device__ __attribute__((pure)) double __ocml_log_f64(double);
+__device__ double __ocml_modf_f64(double,
+ __attribute__((address_space(5))) double *);
+__device__ __attribute__((const)) double __ocml_nearbyint_f64(double);
+__device__ __attribute__((const)) double __ocml_nextafter_f64(double, double);
+__device__ __attribute__((const)) double __ocml_len3_f64(double, double,
+ double);
+__device__ __attribute__((const)) double __ocml_len4_f64(double, double, double,
+ double);
+__device__ __attribute__((pure)) double __ocml_ncdf_f64(double);
+__device__ __attribute__((pure)) double __ocml_ncdfinv_f64(double);
+__device__ __attribute__((pure)) double __ocml_pow_f64(double, double);
+__device__ __attribute__((pure)) double __ocml_rcbrt_f64(double);
+__device__ __attribute__((const)) double __ocml_remainder_f64(double, double);
+__device__ double __ocml_remquo_f64(double, double,
+ __attribute__((address_space(5))) int *);
+__device__ __attribute__((const)) double __ocml_rhypot_f64(double, double);
+__device__ __attribute__((const)) double __ocml_rint_f64(double);
+__device__ __attribute__((const)) double __ocml_rlen3_f64(double, double,
+ double);
+__device__ __attribute__((const)) double __ocml_rlen4_f64(double, double,
+ double, double);
+__device__ __attribute__((const)) double __ocml_round_f64(double);
+__device__ __attribute__((pure)) double __ocml_rsqrt_f64(double);
+__device__ __attribute__((const)) double __ocml_scalb_f64(double, double);
+__device__ __attribute__((const)) double __ocml_scalbn_f64(double, int);
+__device__ __attribute__((const)) int __ocml_signbit_f64(double);
+__device__ double __ocml_sincos_f64(double,
+ __attribute__((address_space(5))) double *);
+__device__ double
+__ocml_sincospi_f64(double, __attribute__((address_space(5))) double *);
+__device__ double __ocml_sin_f64(double);
+__device__ __attribute__((pure)) double __ocml_sinh_f64(double);
+__device__ double __ocml_sinpi_f64(double);
+__device__ __attribute__((const)) double __ocml_sqrt_f64(double);
+__device__ double __ocml_tan_f64(double);
+__device__ __attribute__((pure)) double __ocml_tanh_f64(double);
+__device__ double __ocml_tgamma_f64(double);
+__device__ __attribute__((const)) double __ocml_trunc_f64(double);
+__device__ double __ocml_y0_f64(double);
+__device__ double __ocml_y1_f64(double);
+
+// BEGIN INTRINSICS
+__device__ __attribute__((const)) double __ocml_add_rte_f64(double, double);
+__device__ __attribute__((const)) double __ocml_add_rtn_f64(double, double);
+__device__ __attribute__((const)) double __ocml_add_rtp_f64(double, double);
+__device__ __attribute__((const)) double __ocml_add_rtz_f64(double, double);
+__device__ __attribute__((const)) double __ocml_sub_rte_f64(double, double);
+__device__ __attribute__((const)) double __ocml_sub_rtn_f64(double, double);
+__device__ __attribute__((const)) double __ocml_sub_rtp_f64(double, double);
+__device__ __attribute__((const)) double __ocml_sub_rtz_f64(double, double);
+__device__ __attribute__((const)) double __ocml_mul_rte_f64(double, double);
+__device__ __attribute__((const)) double __ocml_mul_rtn_f64(double, double);
+__device__ __attribute__((const)) double __ocml_mul_rtp_f64(double, double);
+__device__ __attribute__((const)) double __ocml_mul_rtz_f64(double, double);
+__device__ __attribute__((const)) double __ocml_div_rte_f64(double, double);
+__device__ __attribute__((const)) double __ocml_div_rtn_f64(double, double);
+__device__ __attribute__((const)) double __ocml_div_rtp_f64(double, double);
+__device__ __attribute__((const)) double __ocml_div_rtz_f64(double, double);
+__device__ __attribute__((const)) double __ocml_sqrt_rte_f64(double, double);
+__device__ __attribute__((const)) double __ocml_sqrt_rtn_f64(double, double);
+__device__ __attribute__((const)) double __ocml_sqrt_rtp_f64(double, double);
+__device__ __attribute__((const)) double __ocml_sqrt_rtz_f64(double, double);
+__device__ __attribute__((const)) double __ocml_fma_rte_f64(double, double,
+ double);
+__device__ __attribute__((const)) double __ocml_fma_rtn_f64(double, double,
+ double);
+__device__ __attribute__((const)) double __ocml_fma_rtp_f64(double, double,
+ double);
+__device__ __attribute__((const)) double __ocml_fma_rtz_f64(double, double,
+ double);
+
+__device__ __attribute__((const)) double
+__llvm_amdgcn_rcp_f64(double) __asm("llvm.amdgcn.rcp.f64");
+__device__ __attribute__((const)) double
+__llvm_amdgcn_rsq_f64(double) __asm("llvm.amdgcn.rsq.f64");
+
+__device__ __attribute__((const)) _Float16 __ocml_ceil_f16(_Float16);
+__device__ _Float16 __ocml_cos_f16(_Float16);
+__device__ __attribute__((pure)) _Float16 __ocml_exp_f16(_Float16);
+__device__ __attribute__((pure)) _Float16 __ocml_exp10_f16(_Float16);
+__device__ __attribute__((pure)) _Float16 __ocml_exp2_f16(_Float16);
+__device__ __attribute__((const)) _Float16 __ocml_floor_f16(_Float16);
+__device__ __attribute__((const)) _Float16 __ocml_fma_f16(_Float16, _Float16,
+ _Float16);
+__device__ __attribute__((const)) _Float16 __ocml_fabs_f16(_Float16);
+__device__ __attribute__((const)) int __ocml_isinf_f16(_Float16);
+__device__ __attribute__((const)) int __ocml_isnan_f16(_Float16);
+__device__ __attribute__((pure)) _Float16 __ocml_log_f16(_Float16);
+__device__ __attribute__((pure)) _Float16 __ocml_log10_f16(_Float16);
+__device__ __attribute__((pure)) _Float16 __ocml_log2_f16(_Float16);
+__device__ __attribute__((const)) _Float16 __llvm_amdgcn_rcp_f16(_Float16);
+__device__ __attribute__((const)) _Float16 __ocml_rint_f16(_Float16);
+__device__ __attribute__((const)) _Float16 __ocml_rsqrt_f16(_Float16);
+__device__ _Float16 __ocml_sin_f16(_Float16);
+__device__ __attribute__((const)) _Float16 __ocml_sqrt_f16(_Float16);
+__device__ __attribute__((const)) _Float16 __ocml_trunc_f16(_Float16);
+
+typedef _Float16 __2f16 __attribute__((ext_vector_type(2)));
+typedef short __2i16 __attribute__((ext_vector_type(2)));
+
+__device__ __attribute__((const)) float __ockl_fdot2(__2f16 a, __2f16 b,
+ float c, bool s);
+__device__ __attribute__((const)) __2f16 __ocml_ceil_2f16(__2f16);
+__device__ __attribute__((const)) __2f16 __ocml_fabs_2f16(__2f16);
+__device__ __2f16 __ocml_cos_2f16(__2f16);
+__device__ __attribute__((pure)) __2f16 __ocml_exp_2f16(__2f16);
+__device__ __attribute__((pure)) __2f16 __ocml_exp10_2f16(__2f16);
+__device__ __attribute__((pure)) __2f16 __ocml_exp2_2f16(__2f16);
+__device__ __attribute__((const)) __2f16 __ocml_floor_2f16(__2f16);
+__device__ __attribute__((const))
+__2f16 __ocml_fma_2f16(__2f16, __2f16, __2f16);
+__device__ __attribute__((const)) __2i16 __ocml_isinf_2f16(__2f16);
+__device__ __attribute__((const)) __2i16 __ocml_isnan_2f16(__2f16);
+__device__ __attribute__((pure)) __2f16 __ocml_log_2f16(__2f16);
+__device__ __attribute__((pure)) __2f16 __ocml_log10_2f16(__2f16);
+__device__ __attribute__((pure)) __2f16 __ocml_log2_2f16(__2f16);
+__device__ inline __2f16
+__llvm_amdgcn_rcp_2f16(__2f16 __x) // Not currently exposed by ROCDL.
+{
+ return __2f16{__llvm_amdgcn_rcp_f16(__x.x), __llvm_amdgcn_rcp_f16(__x.y)};
+}
+__device__ __attribute__((const)) __2f16 __ocml_rint_2f16(__2f16);
+__device__ __attribute__((const)) __2f16 __ocml_rsqrt_2f16(__2f16);
+__device__ __2f16 __ocml_sin_2f16(__2f16);
+__device__ __attribute__((const)) __2f16 __ocml_sqrt_2f16(__2f16);
+__device__ __attribute__((const)) __2f16 __ocml_trunc_2f16(__2f16);
+
+} // extern "C"
+
+#endif // __CLANG_HIP_LIBDEVICE_DECLARES_H__
--- /dev/null
+/*===---- __clang_hip_math.h - HIP math decls -------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __CLANG_HIP_MATH_H__
+#define __CLANG_HIP_MATH_H__
+
+#include <algorithm>
+#include <limits.h>
+#include <limits>
+#include <stdint.h>
+
+#pragma push_macro("__DEVICE__")
+#pragma push_macro("__RETURN_TYPE")
+
+// to be consistent with __clang_cuda_math_forward_declares
+#define __DEVICE__ static __device__
+#define __RETURN_TYPE bool
+
+__DEVICE__
+inline uint64_t __make_mantissa_base8(const char *__tagp) {
+ uint64_t __r = 0;
+ while (__tagp) {
+ char __tmp = *__tagp;
+
+ if (__tmp >= '0' && __tmp <= '7')
+ __r = (__r * 8u) + __tmp - '0';
+ else
+ return 0;
+
+ ++__tagp;
+ }
+
+ return __r;
+}
+
+__DEVICE__
+inline uint64_t __make_mantissa_base10(const char *__tagp) {
+ uint64_t __r = 0;
+ while (__tagp) {
+ char __tmp = *__tagp;
+
+ if (__tmp >= '0' && __tmp <= '9')
+ __r = (__r * 10u) + __tmp - '0';
+ else
+ return 0;
+
+ ++__tagp;
+ }
+
+ return __r;
+}
+
+__DEVICE__
+inline uint64_t __make_mantissa_base16(const char *__tagp) {
+ uint64_t __r = 0;
+ while (__tagp) {
+ char __tmp = *__tagp;
+
+ if (__tmp >= '0' && __tmp <= '9')
+ __r = (__r * 16u) + __tmp - '0';
+ else if (__tmp >= 'a' && __tmp <= 'f')
+ __r = (__r * 16u) + __tmp - 'a' + 10;
+ else if (__tmp >= 'A' && __tmp <= 'F')
+ __r = (__r * 16u) + __tmp - 'A' + 10;
+ else
+ return 0;
+
+ ++__tagp;
+ }
+
+ return __r;
+}
+
+__DEVICE__
+inline uint64_t __make_mantissa(const char *__tagp) {
+ if (!__tagp)
+ return 0u;
+
+ if (*__tagp == '0') {
+ ++__tagp;
+
+ if (*__tagp == 'x' || *__tagp == 'X')
+ return __make_mantissa_base16(__tagp);
+ else
+ return __make_mantissa_base8(__tagp);
+ }
+
+ return __make_mantissa_base10(__tagp);
+}
+
+// BEGIN FLOAT
+__DEVICE__
+inline float abs(float __x) { return __ocml_fabs_f32(__x); }
+__DEVICE__
+inline float acosf(float __x) { return __ocml_acos_f32(__x); }
+__DEVICE__
+inline float acoshf(float __x) { return __ocml_acosh_f32(__x); }
+__DEVICE__
+inline float asinf(float __x) { return __ocml_asin_f32(__x); }
+__DEVICE__
+inline float asinhf(float __x) { return __ocml_asinh_f32(__x); }
+__DEVICE__
+inline float atan2f(float __x, float __y) { return __ocml_atan2_f32(__x, __y); }
+__DEVICE__
+inline float atanf(float __x) { return __ocml_atan_f32(__x); }
+__DEVICE__
+inline float atanhf(float __x) { return __ocml_atanh_f32(__x); }
+__DEVICE__
+inline float cbrtf(float __x) { return __ocml_cbrt_f32(__x); }
+__DEVICE__
+inline float ceilf(float __x) { return __ocml_ceil_f32(__x); }
+__DEVICE__
+inline float copysignf(float __x, float __y) {
+ return __ocml_copysign_f32(__x, __y);
+}
+__DEVICE__
+inline float cosf(float __x) { return __ocml_cos_f32(__x); }
+__DEVICE__
+inline float coshf(float __x) { return __ocml_cosh_f32(__x); }
+__DEVICE__
+inline float cospif(float __x) { return __ocml_cospi_f32(__x); }
+__DEVICE__
+inline float cyl_bessel_i0f(float __x) { return __ocml_i0_f32(__x); }
+__DEVICE__
+inline float cyl_bessel_i1f(float __x) { return __ocml_i1_f32(__x); }
+__DEVICE__
+inline float erfcf(float __x) { return __ocml_erfc_f32(__x); }
+__DEVICE__
+inline float erfcinvf(float __x) { return __ocml_erfcinv_f32(__x); }
+__DEVICE__
+inline float erfcxf(float __x) { return __ocml_erfcx_f32(__x); }
+__DEVICE__
+inline float erff(float __x) { return __ocml_erf_f32(__x); }
+__DEVICE__
+inline float erfinvf(float __x) { return __ocml_erfinv_f32(__x); }
+__DEVICE__
+inline float exp10f(float __x) { return __ocml_exp10_f32(__x); }
+__DEVICE__
+inline float exp2f(float __x) { return __ocml_exp2_f32(__x); }
+__DEVICE__
+inline float expf(float __x) { return __ocml_exp_f32(__x); }
+__DEVICE__
+inline float expm1f(float __x) { return __ocml_expm1_f32(__x); }
+__DEVICE__
+inline float fabsf(float __x) { return __ocml_fabs_f32(__x); }
+__DEVICE__
+inline float fdimf(float __x, float __y) { return __ocml_fdim_f32(__x, __y); }
+__DEVICE__
+inline float fdividef(float __x, float __y) { return __x / __y; }
+__DEVICE__
+inline float floorf(float __x) { return __ocml_floor_f32(__x); }
+__DEVICE__
+inline float fmaf(float __x, float __y, float __z) {
+ return __ocml_fma_f32(__x, __y, __z);
+}
+__DEVICE__
+inline float fmaxf(float __x, float __y) { return __ocml_fmax_f32(__x, __y); }
+__DEVICE__
+inline float fminf(float __x, float __y) { return __ocml_fmin_f32(__x, __y); }
+__DEVICE__
+inline float fmodf(float __x, float __y) { return __ocml_fmod_f32(__x, __y); }
+__DEVICE__
+inline float frexpf(float __x, int *__nptr) {
+ int __tmp;
+ float __r =
+ __ocml_frexp_f32(__x, (__attribute__((address_space(5))) int *)&__tmp);
+ *__nptr = __tmp;
+
+ return __r;
+}
+__DEVICE__
+inline float hypotf(float __x, float __y) { return __ocml_hypot_f32(__x, __y); }
+__DEVICE__
+inline int ilogbf(float __x) { return __ocml_ilogb_f32(__x); }
+__DEVICE__
+inline __RETURN_TYPE isfinite(float __x) { return __ocml_isfinite_f32(__x); }
+__DEVICE__
+inline __RETURN_TYPE isinf(float __x) { return __ocml_isinf_f32(__x); }
+__DEVICE__
+inline __RETURN_TYPE isnan(float __x) { return __ocml_isnan_f32(__x); }
+__DEVICE__
+inline float j0f(float __x) { return __ocml_j0_f32(__x); }
+__DEVICE__
+inline float j1f(float __x) { return __ocml_j1_f32(__x); }
+__DEVICE__
+inline float jnf(int __n,
+ float __x) { // TODO: we could use Ahmes multiplication
+ // and the Miller & Brown algorithm
+ // for linear recurrences to get O(log n) steps, but it's unclear if
+ // it'd be beneficial in this case.
+ if (__n == 0)
+ return j0f(__x);
+ if (__n == 1)
+ return j1f(__x);
+
+ float __x0 = j0f(__x);
+ float __x1 = j1f(__x);
+ for (int __i = 1; __i < __n; ++__i) {
+ float __x2 = (2 * __i) / __x * __x1 - __x0;
+ __x0 = __x1;
+ __x1 = __x2;
+ }
+
+ return __x1;
+}
+__DEVICE__
+inline float ldexpf(float __x, int __e) { return __ocml_ldexp_f32(__x, __e); }
+__DEVICE__
+inline float lgammaf(float __x) { return __ocml_lgamma_f32(__x); }
+__DEVICE__
+inline long long int llrintf(float __x) { return __ocml_rint_f32(__x); }
+__DEVICE__
+inline long long int llroundf(float __x) { return __ocml_round_f32(__x); }
+__DEVICE__
+inline float log10f(float __x) { return __ocml_log10_f32(__x); }
+__DEVICE__
+inline float log1pf(float __x) { return __ocml_log1p_f32(__x); }
+__DEVICE__
+inline float log2f(float __x) { return __ocml_log2_f32(__x); }
+__DEVICE__
+inline float logbf(float __x) { return __ocml_logb_f32(__x); }
+__DEVICE__
+inline float logf(float __x) { return __ocml_log_f32(__x); }
+__DEVICE__
+inline long int lrintf(float __x) { return __ocml_rint_f32(__x); }
+__DEVICE__
+inline long int lroundf(float __x) { return __ocml_round_f32(__x); }
+__DEVICE__
+inline float modff(float __x, float *__iptr) {
+ float __tmp;
+ float __r =
+ __ocml_modf_f32(__x, (__attribute__((address_space(5))) float *)&__tmp);
+ *__iptr = __tmp;
+
+ return __r;
+}
+__DEVICE__
+inline float nanf(const char *__tagp) {
+ union {
+ float val;
+ struct ieee_float {
+ uint32_t mantissa : 22;
+ uint32_t quiet : 1;
+ uint32_t exponent : 8;
+ uint32_t sign : 1;
+ } bits;
+
+ static_assert(sizeof(float) == sizeof(ieee_float), "");
+ } __tmp;
+
+ __tmp.bits.sign = 0u;
+ __tmp.bits.exponent = ~0u;
+ __tmp.bits.quiet = 1u;
+ __tmp.bits.mantissa = __make_mantissa(__tagp);
+
+ return __tmp.val;
+}
+__DEVICE__
+inline float nearbyintf(float __x) { return __ocml_nearbyint_f32(__x); }
+__DEVICE__
+inline float nextafterf(float __x, float __y) {
+ return __ocml_nextafter_f32(__x, __y);
+}
+__DEVICE__
+inline float norm3df(float __x, float __y, float __z) {
+ return __ocml_len3_f32(__x, __y, __z);
+}
+__DEVICE__
+inline float norm4df(float __x, float __y, float __z, float __w) {
+ return __ocml_len4_f32(__x, __y, __z, __w);
+}
+__DEVICE__
+inline float normcdff(float __x) { return __ocml_ncdf_f32(__x); }
+__DEVICE__
+inline float normcdfinvf(float __x) { return __ocml_ncdfinv_f32(__x); }
+__DEVICE__
+inline float
+normf(int __dim,
+ const float *__a) { // TODO: placeholder until OCML adds support.
+ float __r = 0;
+ while (__dim--) {
+ __r += __a[0] * __a[0];
+ ++__a;
+ }
+
+ return __ocml_sqrt_f32(__r);
+}
+__DEVICE__
+inline float powf(float __x, float __y) { return __ocml_pow_f32(__x, __y); }
+__DEVICE__
+inline float rcbrtf(float __x) { return __ocml_rcbrt_f32(__x); }
+__DEVICE__
+inline float remainderf(float __x, float __y) {
+ return __ocml_remainder_f32(__x, __y);
+}
+__DEVICE__
+inline float remquof(float __x, float __y, int *__quo) {
+ int __tmp;
+ float __r = __ocml_remquo_f32(
+ __x, __y, (__attribute__((address_space(5))) int *)&__tmp);
+ *__quo = __tmp;
+
+ return __r;
+}
+__DEVICE__
+inline float rhypotf(float __x, float __y) {
+ return __ocml_rhypot_f32(__x, __y);
+}
+__DEVICE__
+inline float rintf(float __x) { return __ocml_rint_f32(__x); }
+__DEVICE__
+inline float rnorm3df(float __x, float __y, float __z) {
+ return __ocml_rlen3_f32(__x, __y, __z);
+}
+
+__DEVICE__
+inline float rnorm4df(float __x, float __y, float __z, float __w) {
+ return __ocml_rlen4_f32(__x, __y, __z, __w);
+}
+__DEVICE__
+inline float
+rnormf(int __dim,
+ const float *__a) { // TODO: placeholder until OCML adds support.
+ float __r = 0;
+ while (__dim--) {
+ __r += __a[0] * __a[0];
+ ++__a;
+ }
+
+ return __ocml_rsqrt_f32(__r);
+}
+__DEVICE__
+inline float roundf(float __x) { return __ocml_round_f32(__x); }
+__DEVICE__
+inline float rsqrtf(float __x) { return __ocml_rsqrt_f32(__x); }
+__DEVICE__
+inline float scalblnf(float __x, long int __n) {
+ return (__n < INT_MAX) ? __ocml_scalbn_f32(__x, __n)
+ : __ocml_scalb_f32(__x, __n);
+}
+__DEVICE__
+inline float scalbnf(float __x, int __n) { return __ocml_scalbn_f32(__x, __n); }
+__DEVICE__
+inline __RETURN_TYPE signbit(float __x) { return __ocml_signbit_f32(__x); }
+__DEVICE__
+inline void sincosf(float __x, float *__sinptr, float *__cosptr) {
+ float __tmp;
+
+ *__sinptr =
+ __ocml_sincos_f32(__x, (__attribute__((address_space(5))) float *)&__tmp);
+ *__cosptr = __tmp;
+}
+__DEVICE__
+inline void sincospif(float __x, float *__sinptr, float *__cosptr) {
+ float __tmp;
+
+ *__sinptr = __ocml_sincospi_f32(
+ __x, (__attribute__((address_space(5))) float *)&__tmp);
+ *__cosptr = __tmp;
+}
+__DEVICE__
+inline float sinf(float __x) { return __ocml_sin_f32(__x); }
+__DEVICE__
+inline float sinhf(float __x) { return __ocml_sinh_f32(__x); }
+__DEVICE__
+inline float sinpif(float __x) { return __ocml_sinpi_f32(__x); }
+__DEVICE__
+inline float sqrtf(float __x) { return __ocml_sqrt_f32(__x); }
+__DEVICE__
+inline float tanf(float __x) { return __ocml_tan_f32(__x); }
+__DEVICE__
+inline float tanhf(float __x) { return __ocml_tanh_f32(__x); }
+__DEVICE__
+inline float tgammaf(float __x) { return __ocml_tgamma_f32(__x); }
+__DEVICE__
+inline float truncf(float __x) { return __ocml_trunc_f32(__x); }
+__DEVICE__
+inline float y0f(float __x) { return __ocml_y0_f32(__x); }
+__DEVICE__
+inline float y1f(float __x) { return __ocml_y1_f32(__x); }
+__DEVICE__
+inline float ynf(int __n,
+ float __x) { // TODO: we could use Ahmes multiplication
+ // and the Miller & Brown algorithm
+ // for linear recurrences to get O(log n) steps, but it's unclear if
+ // it'd be beneficial in this case. Placeholder until OCML adds
+ // support.
+ if (__n == 0)
+ return y0f(__x);
+ if (__n == 1)
+ return y1f(__x);
+
+ float __x0 = y0f(__x);
+ float __x1 = y1f(__x);
+ for (int __i = 1; __i < __n; ++__i) {
+ float __x2 = (2 * __i) / __x * __x1 - __x0;
+ __x0 = __x1;
+ __x1 = __x2;
+ }
+
+ return __x1;
+}
+
+// BEGIN INTRINSICS
+__DEVICE__
+inline float __cosf(float __x) { return __ocml_native_cos_f32(__x); }
+__DEVICE__
+inline float __exp10f(float __x) { return __ocml_native_exp10_f32(__x); }
+__DEVICE__
+inline float __expf(float __x) { return __ocml_native_exp_f32(__x); }
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline float __fadd_rd(float __x, float __y) {
+ return __ocml_add_rtn_f32(__x, __y);
+}
+#endif
+__DEVICE__
+inline float __fadd_rn(float __x, float __y) { return __x + __y; }
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline float __fadd_ru(float __x, float __y) {
+ return __ocml_add_rtp_f32(__x, __y);
+}
+__DEVICE__
+inline float __fadd_rz(float __x, float __y) {
+ return __ocml_add_rtz_f32(__x, __y);
+}
+__DEVICE__
+inline float __fdiv_rd(float __x, float __y) {
+ return __ocml_div_rtn_f32(__x, __y);
+}
+#endif
+__DEVICE__
+inline float __fdiv_rn(float __x, float __y) { return __x / __y; }
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline float __fdiv_ru(float __x, float __y) {
+ return __ocml_div_rtp_f32(__x, __y);
+}
+__DEVICE__
+inline float __fdiv_rz(float __x, float __y) {
+ return __ocml_div_rtz_f32(__x, __y);
+}
+#endif
+__DEVICE__
+inline float __fdividef(float __x, float __y) { return __x / __y; }
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline float __fmaf_rd(float __x, float __y, float __z) {
+ return __ocml_fma_rtn_f32(__x, __y, __z);
+}
+#endif
+__DEVICE__
+inline float __fmaf_rn(float __x, float __y, float __z) {
+ return __ocml_fma_f32(__x, __y, __z);
+}
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline float __fmaf_ru(float __x, float __y, float __z) {
+ return __ocml_fma_rtp_f32(__x, __y, __z);
+}
+__DEVICE__
+inline float __fmaf_rz(float __x, float __y, float __z) {
+ return __ocml_fma_rtz_f32(__x, __y, __z);
+}
+__DEVICE__
+inline float __fmul_rd(float __x, float __y) {
+ return __ocml_mul_rtn_f32(__x, __y);
+}
+#endif
+__DEVICE__
+inline float __fmul_rn(float __x, float __y) { return __x * __y; }
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline float __fmul_ru(float __x, float __y) {
+ return __ocml_mul_rtp_f32(__x, __y);
+}
+__DEVICE__
+inline float __fmul_rz(float __x, float __y) {
+ return __ocml_mul_rtz_f32(__x, __y);
+}
+__DEVICE__
+inline float __frcp_rd(float __x) { return __llvm_amdgcn_rcp_f32(__x); }
+#endif
+__DEVICE__
+inline float __frcp_rn(float __x) { return __llvm_amdgcn_rcp_f32(__x); }
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline float __frcp_ru(float __x) { return __llvm_amdgcn_rcp_f32(__x); }
+__DEVICE__
+inline float __frcp_rz(float __x) { return __llvm_amdgcn_rcp_f32(__x); }
+#endif
+__DEVICE__
+inline float __frsqrt_rn(float __x) { return __llvm_amdgcn_rsq_f32(__x); }
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline float __fsqrt_rd(float __x) { return __ocml_sqrt_rtn_f32(__x); }
+#endif
+__DEVICE__
+inline float __fsqrt_rn(float __x) { return __ocml_native_sqrt_f32(__x); }
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline float __fsqrt_ru(float __x) { return __ocml_sqrt_rtp_f32(__x); }
+__DEVICE__
+inline float __fsqrt_rz(float __x) { return __ocml_sqrt_rtz_f32(__x); }
+__DEVICE__
+inline float __fsub_rd(float __x, float __y) {
+ return __ocml_sub_rtn_f32(__x, __y);
+}
+#endif
+__DEVICE__
+inline float __fsub_rn(float __x, float __y) { return __x - __y; }
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline float __fsub_ru(float __x, float __y) {
+ return __ocml_sub_rtp_f32(__x, __y);
+}
+__DEVICE__
+inline float __fsub_rz(float __x, float __y) {
+ return __ocml_sub_rtz_f32(__x, __y);
+}
+#endif
+__DEVICE__
+inline float __log10f(float __x) { return __ocml_native_log10_f32(__x); }
+__DEVICE__
+inline float __log2f(float __x) { return __ocml_native_log2_f32(__x); }
+__DEVICE__
+inline float __logf(float __x) { return __ocml_native_log_f32(__x); }
+__DEVICE__
+inline float __powf(float __x, float __y) { return __ocml_pow_f32(__x, __y); }
+__DEVICE__
+inline float __saturatef(float __x) {
+ return (__x < 0) ? 0 : ((__x > 1) ? 1 : __x);
+}
+__DEVICE__
+inline void __sincosf(float __x, float *__sinptr, float *__cosptr) {
+ *__sinptr = __ocml_native_sin_f32(__x);
+ *__cosptr = __ocml_native_cos_f32(__x);
+}
+__DEVICE__
+inline float __sinf(float __x) { return __ocml_native_sin_f32(__x); }
+__DEVICE__
+inline float __tanf(float __x) { return __ocml_tan_f32(__x); }
+// END INTRINSICS
+// END FLOAT
+
+// BEGIN DOUBLE
+__DEVICE__
+inline double abs(double __x) { return __ocml_fabs_f64(__x); }
+__DEVICE__
+inline double acos(double __x) { return __ocml_acos_f64(__x); }
+__DEVICE__
+inline double acosh(double __x) { return __ocml_acosh_f64(__x); }
+__DEVICE__
+inline double asin(double __x) { return __ocml_asin_f64(__x); }
+__DEVICE__
+inline double asinh(double __x) { return __ocml_asinh_f64(__x); }
+__DEVICE__
+inline double atan(double __x) { return __ocml_atan_f64(__x); }
+__DEVICE__
+inline double atan2(double __x, double __y) {
+ return __ocml_atan2_f64(__x, __y);
+}
+__DEVICE__
+inline double atanh(double __x) { return __ocml_atanh_f64(__x); }
+__DEVICE__
+inline double cbrt(double __x) { return __ocml_cbrt_f64(__x); }
+__DEVICE__
+inline double ceil(double __x) { return __ocml_ceil_f64(__x); }
+__DEVICE__
+inline double copysign(double __x, double __y) {
+ return __ocml_copysign_f64(__x, __y);
+}
+__DEVICE__
+inline double cos(double __x) { return __ocml_cos_f64(__x); }
+__DEVICE__
+inline double cosh(double __x) { return __ocml_cosh_f64(__x); }
+__DEVICE__
+inline double cospi(double __x) { return __ocml_cospi_f64(__x); }
+__DEVICE__
+inline double cyl_bessel_i0(double __x) { return __ocml_i0_f64(__x); }
+__DEVICE__
+inline double cyl_bessel_i1(double __x) { return __ocml_i1_f64(__x); }
+__DEVICE__
+inline double erf(double __x) { return __ocml_erf_f64(__x); }
+__DEVICE__
+inline double erfc(double __x) { return __ocml_erfc_f64(__x); }
+__DEVICE__
+inline double erfcinv(double __x) { return __ocml_erfcinv_f64(__x); }
+__DEVICE__
+inline double erfcx(double __x) { return __ocml_erfcx_f64(__x); }
+__DEVICE__
+inline double erfinv(double __x) { return __ocml_erfinv_f64(__x); }
+__DEVICE__
+inline double exp(double __x) { return __ocml_exp_f64(__x); }
+__DEVICE__
+inline double exp10(double __x) { return __ocml_exp10_f64(__x); }
+__DEVICE__
+inline double exp2(double __x) { return __ocml_exp2_f64(__x); }
+__DEVICE__
+inline double expm1(double __x) { return __ocml_expm1_f64(__x); }
+__DEVICE__
+inline double fabs(double __x) { return __ocml_fabs_f64(__x); }
+__DEVICE__
+inline double fdim(double __x, double __y) { return __ocml_fdim_f64(__x, __y); }
+__DEVICE__
+inline double floor(double __x) { return __ocml_floor_f64(__x); }
+__DEVICE__
+inline double fma(double __x, double __y, double __z) {
+ return __ocml_fma_f64(__x, __y, __z);
+}
+__DEVICE__
+inline double fmax(double __x, double __y) { return __ocml_fmax_f64(__x, __y); }
+__DEVICE__
+inline double fmin(double __x, double __y) { return __ocml_fmin_f64(__x, __y); }
+__DEVICE__
+inline double fmod(double __x, double __y) { return __ocml_fmod_f64(__x, __y); }
+__DEVICE__
+inline double frexp(double __x, int *__nptr) {
+ int __tmp;
+ double __r =
+ __ocml_frexp_f64(__x, (__attribute__((address_space(5))) int *)&__tmp);
+ *__nptr = __tmp;
+
+ return __r;
+}
+__DEVICE__
+inline double hypot(double __x, double __y) {
+ return __ocml_hypot_f64(__x, __y);
+}
+__DEVICE__
+inline int ilogb(double __x) { return __ocml_ilogb_f64(__x); }
+__DEVICE__
+inline __RETURN_TYPE isfinite(double __x) { return __ocml_isfinite_f64(__x); }
+__DEVICE__
+inline __RETURN_TYPE isinf(double __x) { return __ocml_isinf_f64(__x); }
+__DEVICE__
+inline __RETURN_TYPE isnan(double __x) { return __ocml_isnan_f64(__x); }
+__DEVICE__
+inline double j0(double __x) { return __ocml_j0_f64(__x); }
+__DEVICE__
+inline double j1(double __x) { return __ocml_j1_f64(__x); }
+__DEVICE__
+inline double jn(int __n,
+ double __x) { // TODO: we could use Ahmes multiplication
+ // and the Miller & Brown algorithm
+ // for linear recurrences to get O(log n) steps, but it's unclear if
+ // it'd be beneficial in this case. Placeholder until OCML adds
+ // support.
+ if (__n == 0)
+ return j0f(__x);
+ if (__n == 1)
+ return j1f(__x);
+
+ double __x0 = j0f(__x);
+ double __x1 = j1f(__x);
+ for (int __i = 1; __i < __n; ++__i) {
+ double __x2 = (2 * __i) / __x * __x1 - __x0;
+ __x0 = __x1;
+ __x1 = __x2;
+ }
+
+ return __x1;
+}
+__DEVICE__
+inline double ldexp(double __x, int __e) { return __ocml_ldexp_f64(__x, __e); }
+__DEVICE__
+inline double lgamma(double __x) { return __ocml_lgamma_f64(__x); }
+__DEVICE__
+inline long long int llrint(double __x) { return __ocml_rint_f64(__x); }
+__DEVICE__
+inline long long int llround(double __x) { return __ocml_round_f64(__x); }
+__DEVICE__
+inline double log(double __x) { return __ocml_log_f64(__x); }
+__DEVICE__
+inline double log10(double __x) { return __ocml_log10_f64(__x); }
+__DEVICE__
+inline double log1p(double __x) { return __ocml_log1p_f64(__x); }
+__DEVICE__
+inline double log2(double __x) { return __ocml_log2_f64(__x); }
+__DEVICE__
+inline double logb(double __x) { return __ocml_logb_f64(__x); }
+__DEVICE__
+inline long int lrint(double __x) { return __ocml_rint_f64(__x); }
+__DEVICE__
+inline long int lround(double __x) { return __ocml_round_f64(__x); }
+__DEVICE__
+inline double modf(double __x, double *__iptr) {
+ double __tmp;
+ double __r =
+ __ocml_modf_f64(__x, (__attribute__((address_space(5))) double *)&__tmp);
+ *__iptr = __tmp;
+
+ return __r;
+}
+__DEVICE__
+inline double nan(const char *__tagp) {
+#if !_WIN32
+ union {
+ double val;
+ struct ieee_double {
+ uint64_t mantissa : 51;
+ uint32_t quiet : 1;
+ uint32_t exponent : 11;
+ uint32_t sign : 1;
+ } bits;
+ static_assert(sizeof(double) == sizeof(ieee_double), "");
+ } __tmp;
+
+ __tmp.bits.sign = 0u;
+ __tmp.bits.exponent = ~0u;
+ __tmp.bits.quiet = 1u;
+ __tmp.bits.mantissa = __make_mantissa(__tagp);
+
+ return __tmp.val;
+#else
+ static_assert(sizeof(uint64_t) == sizeof(double));
+ uint64_t val = __make_mantissa(__tagp);
+ val |= 0xFFF << 51;
+ return *reinterpret_cast<double *>(&val);
+#endif
+}
+__DEVICE__
+inline double nearbyint(double __x) { return __ocml_nearbyint_f64(__x); }
+__DEVICE__
+inline double nextafter(double __x, double __y) {
+ return __ocml_nextafter_f64(__x, __y);
+}
+__DEVICE__
+inline double
+norm(int __dim,
+ const double *__a) { // TODO: placeholder until OCML adds support.
+ double __r = 0;
+ while (__dim--) {
+ __r += __a[0] * __a[0];
+ ++__a;
+ }
+
+ return __ocml_sqrt_f64(__r);
+}
+__DEVICE__
+inline double norm3d(double __x, double __y, double __z) {
+ return __ocml_len3_f64(__x, __y, __z);
+}
+__DEVICE__
+inline double norm4d(double __x, double __y, double __z, double __w) {
+ return __ocml_len4_f64(__x, __y, __z, __w);
+}
+__DEVICE__
+inline double normcdf(double __x) { return __ocml_ncdf_f64(__x); }
+__DEVICE__
+inline double normcdfinv(double __x) { return __ocml_ncdfinv_f64(__x); }
+__DEVICE__
+inline double pow(double __x, double __y) { return __ocml_pow_f64(__x, __y); }
+__DEVICE__
+inline double rcbrt(double __x) { return __ocml_rcbrt_f64(__x); }
+__DEVICE__
+inline double remainder(double __x, double __y) {
+ return __ocml_remainder_f64(__x, __y);
+}
+__DEVICE__
+inline double remquo(double __x, double __y, int *__quo) {
+ int __tmp;
+ double __r = __ocml_remquo_f64(
+ __x, __y, (__attribute__((address_space(5))) int *)&__tmp);
+ *__quo = __tmp;
+
+ return __r;
+}
+__DEVICE__
+inline double rhypot(double __x, double __y) {
+ return __ocml_rhypot_f64(__x, __y);
+}
+__DEVICE__
+inline double rint(double __x) { return __ocml_rint_f64(__x); }
+__DEVICE__
+inline double
+rnorm(int __dim,
+ const double *__a) { // TODO: placeholder until OCML adds support.
+ double __r = 0;
+ while (__dim--) {
+ __r += __a[0] * __a[0];
+ ++__a;
+ }
+
+ return __ocml_rsqrt_f64(__r);
+}
+__DEVICE__
+inline double rnorm3d(double __x, double __y, double __z) {
+ return __ocml_rlen3_f64(__x, __y, __z);
+}
+__DEVICE__
+inline double rnorm4d(double __x, double __y, double __z, double __w) {
+ return __ocml_rlen4_f64(__x, __y, __z, __w);
+}
+__DEVICE__
+inline double round(double __x) { return __ocml_round_f64(__x); }
+__DEVICE__
+inline double rsqrt(double __x) { return __ocml_rsqrt_f64(__x); }
+__DEVICE__
+inline double scalbln(double __x, long int __n) {
+ return (__n < INT_MAX) ? __ocml_scalbn_f64(__x, __n)
+ : __ocml_scalb_f64(__x, __n);
+}
+__DEVICE__
+inline double scalbn(double __x, int __n) {
+ return __ocml_scalbn_f64(__x, __n);
+}
+__DEVICE__
+inline __RETURN_TYPE signbit(double __x) { return __ocml_signbit_f64(__x); }
+__DEVICE__
+inline double sin(double __x) { return __ocml_sin_f64(__x); }
+__DEVICE__
+inline void sincos(double __x, double *__sinptr, double *__cosptr) {
+ double __tmp;
+ *__sinptr = __ocml_sincos_f64(
+ __x, (__attribute__((address_space(5))) double *)&__tmp);
+ *__cosptr = __tmp;
+}
+__DEVICE__
+inline void sincospi(double __x, double *__sinptr, double *__cosptr) {
+ double __tmp;
+ *__sinptr = __ocml_sincospi_f64(
+ __x, (__attribute__((address_space(5))) double *)&__tmp);
+ *__cosptr = __tmp;
+}
+__DEVICE__
+inline double sinh(double __x) { return __ocml_sinh_f64(__x); }
+__DEVICE__
+inline double sinpi(double __x) { return __ocml_sinpi_f64(__x); }
+__DEVICE__
+inline double sqrt(double __x) { return __ocml_sqrt_f64(__x); }
+__DEVICE__
+inline double tan(double __x) { return __ocml_tan_f64(__x); }
+__DEVICE__
+inline double tanh(double __x) { return __ocml_tanh_f64(__x); }
+__DEVICE__
+inline double tgamma(double __x) { return __ocml_tgamma_f64(__x); }
+__DEVICE__
+inline double trunc(double __x) { return __ocml_trunc_f64(__x); }
+__DEVICE__
+inline double y0(double __x) { return __ocml_y0_f64(__x); }
+__DEVICE__
+inline double y1(double __x) { return __ocml_y1_f64(__x); }
+__DEVICE__
+inline double yn(int __n,
+ double __x) { // TODO: we could use Ahmes multiplication
+ // and the Miller & Brown algorithm
+ // for linear recurrences to get O(log n) steps, but it's unclear if
+ // it'd be beneficial in this case. Placeholder until OCML adds
+ // support.
+ if (__n == 0)
+ return j0f(__x);
+ if (__n == 1)
+ return j1f(__x);
+
+ double __x0 = j0f(__x);
+ double __x1 = j1f(__x);
+ for (int __i = 1; __i < __n; ++__i) {
+ double __x2 = (2 * __i) / __x * __x1 - __x0;
+ __x0 = __x1;
+ __x1 = __x2;
+ }
+
+ return __x1;
+}
+
+// BEGIN INTRINSICS
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline double __dadd_rd(double __x, double __y) {
+ return __ocml_add_rtn_f64(__x, __y);
+}
+#endif
+__DEVICE__
+inline double __dadd_rn(double __x, double __y) { return __x + __y; }
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline double __dadd_ru(double __x, double __y) {
+ return __ocml_add_rtp_f64(__x, __y);
+}
+__DEVICE__
+inline double __dadd_rz(double __x, double __y) {
+ return __ocml_add_rtz_f64(__x, __y);
+}
+__DEVICE__
+inline double __ddiv_rd(double __x, double __y) {
+ return __ocml_div_rtn_f64(__x, __y);
+}
+#endif
+__DEVICE__
+inline double __ddiv_rn(double __x, double __y) { return __x / __y; }
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline double __ddiv_ru(double __x, double __y) {
+ return __ocml_div_rtp_f64(__x, __y);
+}
+__DEVICE__
+inline double __ddiv_rz(double __x, double __y) {
+ return __ocml_div_rtz_f64(__x, __y);
+}
+__DEVICE__
+inline double __dmul_rd(double __x, double __y) {
+ return __ocml_mul_rtn_f64(__x, __y);
+}
+#endif
+__DEVICE__
+inline double __dmul_rn(double __x, double __y) { return __x * __y; }
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline double __dmul_ru(double __x, double __y) {
+ return __ocml_mul_rtp_f64(__x, __y);
+}
+__DEVICE__
+inline double __dmul_rz(double __x, double __y) {
+ return __ocml_mul_rtz_f64(__x, __y);
+}
+__DEVICE__
+inline double __drcp_rd(double __x) { return __llvm_amdgcn_rcp_f64(__x); }
+#endif
+__DEVICE__
+inline double __drcp_rn(double __x) { return __llvm_amdgcn_rcp_f64(__x); }
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline double __drcp_ru(double __x) { return __llvm_amdgcn_rcp_f64(__x); }
+__DEVICE__
+inline double __drcp_rz(double __x) { return __llvm_amdgcn_rcp_f64(__x); }
+__DEVICE__
+inline double __dsqrt_rd(double __x) { return __ocml_sqrt_rtn_f64(__x); }
+#endif
+__DEVICE__
+inline double __dsqrt_rn(double __x) { return __ocml_sqrt_f64(__x); }
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline double __dsqrt_ru(double __x) { return __ocml_sqrt_rtp_f64(__x); }
+__DEVICE__
+inline double __dsqrt_rz(double __x) { return __ocml_sqrt_rtz_f64(__x); }
+__DEVICE__
+inline double __dsub_rd(double __x, double __y) {
+ return __ocml_sub_rtn_f64(__x, __y);
+}
+#endif
+__DEVICE__
+inline double __dsub_rn(double __x, double __y) { return __x - __y; }
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline double __dsub_ru(double __x, double __y) {
+ return __ocml_sub_rtp_f64(__x, __y);
+}
+__DEVICE__
+inline double __dsub_rz(double __x, double __y) {
+ return __ocml_sub_rtz_f64(__x, __y);
+}
+__DEVICE__
+inline double __fma_rd(double __x, double __y, double __z) {
+ return __ocml_fma_rtn_f64(__x, __y, __z);
+}
+#endif
+__DEVICE__
+inline double __fma_rn(double __x, double __y, double __z) {
+ return __ocml_fma_f64(__x, __y, __z);
+}
+#if defined OCML_BASIC_ROUNDED_OPERATIONS
+__DEVICE__
+inline double __fma_ru(double __x, double __y, double __z) {
+ return __ocml_fma_rtp_f64(__x, __y, __z);
+}
+__DEVICE__
+inline double __fma_rz(double __x, double __y, double __z) {
+ return __ocml_fma_rtz_f64(__x, __y, __z);
+}
+#endif
+// END INTRINSICS
+// END DOUBLE
+
+// BEGIN INTEGER
+__DEVICE__
+inline int abs(int __x) {
+ int __sgn = __x >> (sizeof(int) * CHAR_BIT - 1);
+ return (__x ^ __sgn) - __sgn;
+}
+__DEVICE__
+inline long labs(long __x) {
+ long __sgn = __x >> (sizeof(long) * CHAR_BIT - 1);
+ return (__x ^ __sgn) - __sgn;
+}
+__DEVICE__
+inline long long llabs(long long __x) {
+ long long __sgn = __x >> (sizeof(long long) * CHAR_BIT - 1);
+ return (__x ^ __sgn) - __sgn;
+}
+
+#if defined(__cplusplus)
+__DEVICE__
+inline long abs(long __x) { return labs(__x); }
+__DEVICE__
+inline long long abs(long long __x) { return llabs(__x); }
+#endif
+// END INTEGER
+
+__DEVICE__
+inline _Float16 fma(_Float16 __x, _Float16 __y, _Float16 __z) {
+ return __ocml_fma_f16(__x, __y, __z);
+}
+
+__DEVICE__
+inline float fma(float __x, float __y, float __z) {
+ return fmaf(__x, __y, __z);
+}
+
+#pragma push_macro("__DEF_FUN1")
+#pragma push_macro("__DEF_FUN2")
+#pragma push_macro("__DEF_FUNI")
+#pragma push_macro("__DEF_FLOAT_FUN2I")
+#pragma push_macro("__HIP_OVERLOAD1")
+#pragma push_macro("__HIP_OVERLOAD2")
+
+// __hip_enable_if::type is a type function which returns __T if __B is true.
+template <bool __B, class __T = void> struct __hip_enable_if {};
+
+template <class __T> struct __hip_enable_if<true, __T> { typedef __T type; };
+
+// __HIP_OVERLOAD1 is used to resolve function calls with integer argument to
+// avoid compilation error due to ambibuity. e.g. floor(5) is resolved with
+// floor(double).
+#define __HIP_OVERLOAD1(__retty, __fn) \
+ template <typename __T> \
+ __DEVICE__ typename __hip_enable_if<std::numeric_limits<__T>::is_integer, \
+ __retty>::type \
+ __fn(__T __x) { \
+ return ::__fn((double)__x); \
+ }
+
+// __HIP_OVERLOAD2 is used to resolve function calls with mixed float/double
+// or integer argument to avoid compilation error due to ambibuity. e.g.
+// max(5.0f, 6.0) is resolved with max(double, double).
+#define __HIP_OVERLOAD2(__retty, __fn) \
+ template <typename __T1, typename __T2> \
+ __DEVICE__ \
+ typename __hip_enable_if<std::numeric_limits<__T1>::is_specialized && \
+ std::numeric_limits<__T2>::is_specialized, \
+ __retty>::type \
+ __fn(__T1 __x, __T2 __y) { \
+ return __fn((double)__x, (double)__y); \
+ }
+
+// Define cmath functions with float argument and returns float.
+#define __DEF_FUN1(__retty, __func) \
+ __DEVICE__ \
+ inline float __func(float __x) { return __func##f(__x); } \
+ __HIP_OVERLOAD1(__retty, __func)
+
+// Define cmath functions with float argument and returns __retty.
+#define __DEF_FUNI(__retty, __func) \
+ __DEVICE__ \
+ inline __retty __func(float __x) { return __func##f(__x); } \
+ __HIP_OVERLOAD1(__retty, __func)
+
+// define cmath functions with two float arguments.
+#define __DEF_FUN2(__retty, __func) \
+ __DEVICE__ \
+ inline float __func(float __x, float __y) { return __func##f(__x, __y); } \
+ __HIP_OVERLOAD2(__retty, __func)
+
+__DEF_FUN1(double, acos)
+__DEF_FUN1(double, acosh)
+__DEF_FUN1(double, asin)
+__DEF_FUN1(double, asinh)
+__DEF_FUN1(double, atan)
+__DEF_FUN2(double, atan2);
+__DEF_FUN1(double, atanh)
+__DEF_FUN1(double, cbrt)
+__DEF_FUN1(double, ceil)
+__DEF_FUN2(double, copysign);
+__DEF_FUN1(double, cos)
+__DEF_FUN1(double, cosh)
+__DEF_FUN1(double, erf)
+__DEF_FUN1(double, erfc)
+__DEF_FUN1(double, exp)
+__DEF_FUN1(double, exp2)
+__DEF_FUN1(double, expm1)
+__DEF_FUN1(double, fabs)
+__DEF_FUN2(double, fdim);
+__DEF_FUN1(double, floor)
+__DEF_FUN2(double, fmax);
+__DEF_FUN2(double, fmin);
+__DEF_FUN2(double, fmod);
+//__HIP_OVERLOAD1(int, fpclassify)
+__DEF_FUN2(double, hypot);
+__DEF_FUNI(int, ilogb)
+__HIP_OVERLOAD1(bool, isfinite)
+__HIP_OVERLOAD2(bool, isgreater);
+__HIP_OVERLOAD2(bool, isgreaterequal);
+__HIP_OVERLOAD1(bool, isinf);
+__HIP_OVERLOAD2(bool, isless);
+__HIP_OVERLOAD2(bool, islessequal);
+__HIP_OVERLOAD2(bool, islessgreater);
+__HIP_OVERLOAD1(bool, isnan);
+//__HIP_OVERLOAD1(bool, isnormal)
+__HIP_OVERLOAD2(bool, isunordered);
+__DEF_FUN1(double, lgamma)
+__DEF_FUN1(double, log)
+__DEF_FUN1(double, log10)
+__DEF_FUN1(double, log1p)
+__DEF_FUN1(double, log2)
+__DEF_FUN1(double, logb)
+__DEF_FUNI(long long, llrint)
+__DEF_FUNI(long long, llround)
+__DEF_FUNI(long, lrint)
+__DEF_FUNI(long, lround)
+__DEF_FUN1(double, nearbyint);
+__DEF_FUN2(double, nextafter);
+__DEF_FUN2(double, pow);
+__DEF_FUN2(double, remainder);
+__DEF_FUN1(double, rint);
+__DEF_FUN1(double, round);
+__HIP_OVERLOAD1(bool, signbit)
+__DEF_FUN1(double, sin)
+__DEF_FUN1(double, sinh)
+__DEF_FUN1(double, sqrt)
+__DEF_FUN1(double, tan)
+__DEF_FUN1(double, tanh)
+__DEF_FUN1(double, tgamma)
+__DEF_FUN1(double, trunc);
+
+// define cmath functions with a float and an integer argument.
+#define __DEF_FLOAT_FUN2I(__func) \
+ __DEVICE__ \
+ inline float __func(float __x, int __y) { return __func##f(__x, __y); }
+__DEF_FLOAT_FUN2I(scalbn)
+
+template <class T> __DEVICE__ inline T min(T __arg1, T __arg2) {
+ return (__arg1 < __arg2) ? __arg1 : __arg2;
+}
+
+template <class T> __DEVICE__ inline T max(T __arg1, T __arg2) {
+ return (__arg1 > __arg2) ? __arg1 : __arg2;
+}
+
+__DEVICE__ inline int min(int __arg1, int __arg2) {
+ return (__arg1 < __arg2) ? __arg1 : __arg2;
+}
+__DEVICE__ inline int max(int __arg1, int __arg2) {
+ return (__arg1 > __arg2) ? __arg1 : __arg2;
+}
+
+__DEVICE__
+inline float max(float __x, float __y) { return fmaxf(__x, __y); }
+
+__DEVICE__
+inline double max(double __x, double __y) { return fmax(__x, __y); }
+
+__DEVICE__
+inline float min(float __x, float __y) { return fminf(__x, __y); }
+
+__DEVICE__
+inline double min(double __x, double __y) { return fmin(__x, __y); }
+
+__HIP_OVERLOAD2(double, max)
+__HIP_OVERLOAD2(double, min)
+
+__host__ inline static int min(int __arg1, int __arg2) {
+ return std::min(__arg1, __arg2);
+}
+
+__host__ inline static int max(int __arg1, int __arg2) {
+ return std::max(__arg1, __arg2);
+}
+
+#pragma pop_macro("__DEF_FUN1")
+#pragma pop_macro("__DEF_FUN2")
+#pragma pop_macro("__DEF_FUNI")
+#pragma pop_macro("__DEF_FLOAT_FUN2I")
+#pragma pop_macro("__HIP_OVERLOAD1")
+#pragma pop_macro("__HIP_OVERLOAD2")
+#pragma pop_macro("__DEVICE__")
+#pragma pop_macro("__RETURN_TYPE")
+
+#endif // __CLANG_HIP_MATH_H__
--- /dev/null
+/*===---- __clang_hip_runtime_wrapper.h - HIP runtime support ---------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+/*
+ * WARNING: This header is intended to be directly -include'd by
+ * the compiler and is not supposed to be included by users.
+ *
+ */
+
+#ifndef __CLANG_HIP_RUNTIME_WRAPPER_H__
+#define __CLANG_HIP_RUNTIME_WRAPPER_H__
+
+#if __HIP__
+
+#include <cmath>
+#include <cstdlib>
+#include <stdlib.h>
+
+#define __host__ __attribute__((host))
+#define __device__ __attribute__((device))
+#define __global__ __attribute__((global))
+#define __shared__ __attribute__((shared))
+#define __constant__ __attribute__((constant))
+
+#if __HIP_ENABLE_DEVICE_MALLOC__
+extern "C" __device__ void *__hip_malloc(size_t __size);
+extern "C" __device__ void *__hip_free(void *__ptr);
+static inline __device__ void *malloc(size_t __size) {
+ return __hip_malloc(__size);
+}
+static inline __device__ void *free(void *__ptr) { return __hip_free(__ptr); }
+#else
+static inline __device__ void *malloc(size_t __size) {
+ __builtin_trap();
+ return nullptr;
+}
+static inline __device__ void *free(void *__ptr) {
+ __builtin_trap();
+ return nullptr;
+}
+#endif
+
+#include <__clang_hip_libdevice_declares.h>
+#include <__clang_hip_math.h>
+
+#if !_OPENMP || __HIP_ENABLE_CUDA_WRAPPER_FOR_OPENMP__
+#include <__clang_cuda_math_forward_declares.h>
+#include <__clang_cuda_complex_builtins.h>
+
+#include <algorithm>
+#include <complex>
+#include <new>
+#endif // !_OPENMP || __HIP_ENABLE_CUDA_WRAPPER_FOR_OPENMP__
+
+#define __CLANG_HIP_RUNTIME_WRAPPER_INCLUDED__ 1
+
+#endif // __HIP__
+#endif // __CLANG_HIP_RUNTIME_WRAPPER_H__
--- /dev/null
+/*===---- __stddef_max_align_t.h - Definition of max_align_t for modules ---===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __CLANG_MAX_ALIGN_T_DEFINED
+#define __CLANG_MAX_ALIGN_T_DEFINED
+
+#if defined(_MSC_VER)
+typedef double max_align_t;
+#elif defined(__APPLE__)
+typedef long double max_align_t;
+#else
+// Define 'max_align_t' to match the GCC definition.
+typedef struct {
+ long long __clang_max_align_nonce1
+ __attribute__((__aligned__(__alignof__(long long))));
+ long double __clang_max_align_nonce2
+ __attribute__((__aligned__(__alignof__(long double))));
+} max_align_t;
+#endif
+
+#endif
/*===---- __wmmintrin_aes.h - AES intrinsics -------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
-#ifndef _WMMINTRIN_AES_H
-#define _WMMINTRIN_AES_H
-#include <emmintrin.h>
+#ifndef __WMMINTRIN_H
+#error "Never use <__wmmintrin_aes.h> directly; include <wmmintrin.h> instead."
+#endif
+
+#ifndef __WMMINTRIN_AES_H
+#define __WMMINTRIN_AES_H
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("aes")))
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("aes"), __min_vector_width__(128)))
-/// \brief Performs a single round of AES encryption using the Equivalent
+/// Performs a single round of AES encryption using the Equivalent
/// Inverse Cipher, transforming the state value from the first source
/// operand using a 128-bit round key value contained in the second source
/// operand, and writes the result to the destination.
return (__m128i)__builtin_ia32_aesenc128((__v2di)__V, (__v2di)__R);
}
-/// \brief Performs the final round of AES encryption using the Equivalent
+/// Performs the final round of AES encryption using the Equivalent
/// Inverse Cipher, transforming the state value from the first source
/// operand using a 128-bit round key value contained in the second source
/// operand, and writes the result to the destination.
return (__m128i)__builtin_ia32_aesenclast128((__v2di)__V, (__v2di)__R);
}
-/// \brief Performs a single round of AES decryption using the Equivalent
+/// Performs a single round of AES decryption using the Equivalent
/// Inverse Cipher, transforming the state value from the first source
/// operand using a 128-bit round key value contained in the second source
/// operand, and writes the result to the destination.
return (__m128i)__builtin_ia32_aesdec128((__v2di)__V, (__v2di)__R);
}
-/// \brief Performs the final round of AES decryption using the Equivalent
+/// Performs the final round of AES decryption using the Equivalent
/// Inverse Cipher, transforming the state value from the first source
/// operand using a 128-bit round key value contained in the second source
/// operand, and writes the result to the destination.
return (__m128i)__builtin_ia32_aesdeclast128((__v2di)__V, (__v2di)__R);
}
-/// \brief Applies the AES InvMixColumns() transformation to an expanded key
+/// Applies the AES InvMixColumns() transformation to an expanded key
/// contained in the source operand, and writes the result to the
/// destination.
///
return (__m128i)__builtin_ia32_aesimc128((__v2di)__V);
}
-/// \brief Generates a round key for AES encyption, operating on 128-bit data
+/// Generates a round key for AES encryption, operating on 128-bit data
/// specified in the first source operand and using an 8-bit round constant
/// specified by the second source operand, and writes the result to the
/// destination.
#undef __DEFAULT_FN_ATTRS
-#endif /* _WMMINTRIN_AES_H */
+#endif /* __WMMINTRIN_AES_H */
--- /dev/null
+/*===---- __wmmintrin_pclmul.h - PCMUL intrinsics ---------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __WMMINTRIN_H
+#error "Never use <__wmmintrin_pclmul.h> directly; include <wmmintrin.h> instead."
+#endif
+
+#ifndef __WMMINTRIN_PCLMUL_H
+#define __WMMINTRIN_PCLMUL_H
+
+/// Multiplies two 64-bit integer values, which are selected from source
+/// operands using the immediate-value operand. The multiplication is a
+/// carry-less multiplication, and the 128-bit integer product is stored in
+/// the destination.
+///
+/// \headerfile <x86intrin.h>
+///
+/// \code
+/// __m128i _mm_clmulepi64_si128(__m128i __X, __m128i __Y, const int __I);
+/// \endcode
+///
+/// This intrinsic corresponds to the <c> VPCLMULQDQ </c> instruction.
+///
+/// \param __X
+/// A 128-bit vector of [2 x i64] containing one of the source operands.
+/// \param __Y
+/// A 128-bit vector of [2 x i64] containing one of the source operands.
+/// \param __I
+/// An immediate value specifying which 64-bit values to select from the
+/// operands. Bit 0 is used to select a value from operand \a __X, and bit
+/// 4 is used to select a value from operand \a __Y: \n
+/// Bit[0]=0 indicates that bits[63:0] of operand \a __X are used. \n
+/// Bit[0]=1 indicates that bits[127:64] of operand \a __X are used. \n
+/// Bit[4]=0 indicates that bits[63:0] of operand \a __Y are used. \n
+/// Bit[4]=1 indicates that bits[127:64] of operand \a __Y are used.
+/// \returns The 128-bit integer vector containing the result of the carry-less
+/// multiplication of the selected 64-bit values.
+#define _mm_clmulepi64_si128(X, Y, I) \
+ ((__m128i)__builtin_ia32_pclmulqdq128((__v2di)(__m128i)(X), \
+ (__v2di)(__m128i)(Y), (char)(I)))
+
+#endif /* __WMMINTRIN_PCLMUL_H */
/*===---- adxintrin.h - ADX intrinsics -------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
_addcarry_u32(unsigned char __cf, unsigned int __x, unsigned int __y,
unsigned int *__p)
{
- return __builtin_ia32_addcarry_u32(__cf, __x, __y, __p);
+ return __builtin_ia32_addcarryx_u32(__cf, __x, __y, __p);
}
#ifdef __x86_64__
_addcarry_u64(unsigned char __cf, unsigned long long __x,
unsigned long long __y, unsigned long long *__p)
{
- return __builtin_ia32_addcarry_u64(__cf, __x, __y, __p);
+ return __builtin_ia32_addcarryx_u64(__cf, __x, __y, __p);
}
#endif
/*===---- altivec.h - Standard header for type generic math ---------------===*\
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
\*===----------------------------------------------------------------------===*/
return (vector double)__builtin_vsx_lxvl(__a, (__b << 56));
}
-static __inline__ vector double __ATTRS_o_ai vec_xl_len_r(unsigned char *__a,
- size_t __b) {
+static __inline__ vector unsigned char __ATTRS_o_ai
+vec_xl_len_r(unsigned char *__a, size_t __b) {
vector unsigned char __res =
(vector unsigned char)__builtin_vsx_lxvll(__a, (__b << 56));
#ifdef __LITTLE_ENDIAN__
#ifdef __VSX__
#define vec_ctf(__a, __b) \
_Generic((__a), vector int \
- : (vector float)__builtin_altivec_vcfsx((__a), (__b)), \
+ : (vector float)__builtin_altivec_vcfsx((vector int)(__a), (__b)), \
vector unsigned int \
- : (vector float)__builtin_altivec_vcfux((vector int)(__a), (__b)), \
+ : (vector float)__builtin_altivec_vcfux((vector unsigned int)(__a), \
+ (__b)), \
vector unsigned long long \
: (__builtin_convertvector((vector unsigned long long)(__a), \
vector double) * \
#else
#define vec_ctf(__a, __b) \
_Generic((__a), vector int \
- : (vector float)__builtin_altivec_vcfsx((__a), (__b)), \
+ : (vector float)__builtin_altivec_vcfsx((vector int)(__a), (__b)), \
vector unsigned int \
- : (vector float)__builtin_altivec_vcfux((vector int)(__a), (__b)))
+ : (vector float)__builtin_altivec_vcfux((vector unsigned int)(__a), \
+ (__b)))
#endif
/* vec_vcfsx */
#ifdef __VSX__
#define vec_cts(__a, __b) \
_Generic((__a), vector float \
- : __builtin_altivec_vctsxs((__a), (__b)), vector double \
+ : __builtin_altivec_vctsxs((vector float)(__a), (__b)), \
+ vector double \
: __extension__({ \
vector double __ret = \
- (__a) * \
+ (vector double)(__a) * \
(vector double)(vector unsigned long long)((0x3ffULL + (__b)) \
<< 52); \
__builtin_convertvector(__ret, vector signed long long); \
#ifdef __VSX__
#define vec_ctu(__a, __b) \
_Generic((__a), vector float \
- : __builtin_altivec_vctuxs((__a), (__b)), vector double \
+ : __builtin_altivec_vctuxs((vector float)(__a), (__b)), \
+ vector double \
: __extension__({ \
vector double __ret = \
- (__a) * \
+ (vector double)(__a) * \
(vector double)(vector unsigned long long)((0x3ffULL + __b) \
<< 52); \
__builtin_convertvector(__ret, vector unsigned long long); \
/* vec_dss */
-static __inline__ void __attribute__((__always_inline__)) vec_dss(int __a) {
- __builtin_altivec_dss(__a);
-}
+#define vec_dss __builtin_altivec_dss
/* vec_dssall */
#ifdef __VSX__
static __inline__ vector double __ATTRS_o_ai vec_or(vector bool long long __a,
vector double __b) {
- return (vector unsigned long long)__a | (vector unsigned long long)__b;
+ return (vector double)((vector unsigned long long)__a |
+ (vector unsigned long long)__b);
}
static __inline__ vector double __ATTRS_o_ai vec_or(vector double __a,
vector bool long long __b) {
- return (vector unsigned long long)__a | (vector unsigned long long)__b;
+ return (vector double)((vector unsigned long long)__a |
+ (vector unsigned long long)__b);
}
static __inline__ vector double __ATTRS_o_ai vec_or(vector double __a,
vector double __b) {
- vector unsigned long long __res =
- (vector unsigned long long)__a | (vector unsigned long long)__b;
- return (vector double)__res;
+ return (vector double)((vector unsigned long long)__a |
+ (vector unsigned long long)__b);
}
static __inline__ vector signed long long __ATTRS_o_ai
/* vec_sr */
-static __inline__ vector signed char __ATTRS_o_ai
-vec_sr(vector signed char __a, vector unsigned char __b) {
- vector unsigned char __res = (vector unsigned char)__a >> __b;
- return (vector signed char)__res;
-}
-
+// vec_sr does modulo arithmetic on __b first, so __b is allowed to be more
+// than the length of __a.
static __inline__ vector unsigned char __ATTRS_o_ai
vec_sr(vector unsigned char __a, vector unsigned char __b) {
- return __a >> __b;
+ return __a >>
+ (__b % (vector unsigned char)(sizeof(unsigned char) * __CHAR_BIT__));
}
-static __inline__ vector signed short __ATTRS_o_ai
-vec_sr(vector signed short __a, vector unsigned short __b) {
- vector unsigned short __res = (vector unsigned short)__a >> __b;
- return (vector signed short)__res;
+static __inline__ vector signed char __ATTRS_o_ai
+vec_sr(vector signed char __a, vector unsigned char __b) {
+ return (vector signed char)vec_sr((vector unsigned char)__a, __b);
}
static __inline__ vector unsigned short __ATTRS_o_ai
vec_sr(vector unsigned short __a, vector unsigned short __b) {
- return __a >> __b;
+ return __a >>
+ (__b % (vector unsigned short)(sizeof(unsigned short) * __CHAR_BIT__));
}
-static __inline__ vector signed int __ATTRS_o_ai
-vec_sr(vector signed int __a, vector unsigned int __b) {
- vector unsigned int __res = (vector unsigned int)__a >> __b;
- return (vector signed int)__res;
+static __inline__ vector short __ATTRS_o_ai vec_sr(vector short __a,
+ vector unsigned short __b) {
+ return (vector short)vec_sr((vector unsigned short)__a, __b);
}
static __inline__ vector unsigned int __ATTRS_o_ai
vec_sr(vector unsigned int __a, vector unsigned int __b) {
- return __a >> __b;
+ return __a >>
+ (__b % (vector unsigned int)(sizeof(unsigned int) * __CHAR_BIT__));
}
-#ifdef __POWER8_VECTOR__
-static __inline__ vector signed long long __ATTRS_o_ai
-vec_sr(vector signed long long __a, vector unsigned long long __b) {
- vector unsigned long long __res = (vector unsigned long long)__a >> __b;
- return (vector signed long long)__res;
+static __inline__ vector int __ATTRS_o_ai vec_sr(vector int __a,
+ vector unsigned int __b) {
+ return (vector int)vec_sr((vector unsigned int)__a, __b);
}
+#ifdef __POWER8_VECTOR__
static __inline__ vector unsigned long long __ATTRS_o_ai
vec_sr(vector unsigned long long __a, vector unsigned long long __b) {
- return __a >> __b;
+ return __a >> (__b % (vector unsigned long long)(sizeof(unsigned long long) *
+ __CHAR_BIT__));
+}
+
+static __inline__ vector long long __ATTRS_o_ai
+vec_sr(vector long long __a, vector unsigned long long __b) {
+ return (vector long long)vec_sr((vector unsigned long long)__a, __b);
}
#endif
static __inline__ vector signed char __ATTRS_o_ai
vec_vsrb(vector signed char __a, vector unsigned char __b) {
- return __a >> (vector signed char)__b;
+ return vec_sr(__a, __b);
}
static __inline__ vector unsigned char __ATTRS_o_ai
vec_vsrb(vector unsigned char __a, vector unsigned char __b) {
- return __a >> __b;
+ return vec_sr(__a, __b);
}
/* vec_vsrh */
static __inline__ vector short __ATTRS_o_ai
vec_vsrh(vector short __a, vector unsigned short __b) {
- return __a >> (vector short)__b;
+ return vec_sr(__a, __b);
}
static __inline__ vector unsigned short __ATTRS_o_ai
vec_vsrh(vector unsigned short __a, vector unsigned short __b) {
- return __a >> __b;
+ return vec_sr(__a, __b);
}
/* vec_vsrw */
static __inline__ vector int __ATTRS_o_ai vec_vsrw(vector int __a,
vector unsigned int __b) {
- return __a >> (vector int)__b;
+ return vec_sr(__a, __b);
}
static __inline__ vector unsigned int __ATTRS_o_ai
vec_vsrw(vector unsigned int __a, vector unsigned int __b) {
- return __a >> __b;
+ return vec_sr(__a, __b);
}
/* vec_sra */
static __inline__ int __ATTRS_o_ai vec_all_ne(vector float __a,
vector float __b) {
#ifdef __VSX__
- return __builtin_vsx_xvcmpeqdp_p(__CR6_EQ, __a, __b);
+ return __builtin_vsx_xvcmpeqsp_p(__CR6_EQ, __a, __b);
#else
return __builtin_altivec_vcmpeqfp_p(__CR6_EQ, __a, __b);
#endif
/* vec_xl */
+typedef vector signed char unaligned_vec_schar __attribute__((aligned(1)));
+typedef vector unsigned char unaligned_vec_uchar __attribute__((aligned(1)));
+typedef vector signed short unaligned_vec_sshort __attribute__((aligned(1)));
+typedef vector unsigned short unaligned_vec_ushort __attribute__((aligned(1)));
+typedef vector signed int unaligned_vec_sint __attribute__((aligned(1)));
+typedef vector unsigned int unaligned_vec_uint __attribute__((aligned(1)));
+typedef vector float unaligned_vec_float __attribute__((aligned(1)));
+
static inline __ATTRS_o_ai vector signed char vec_xl(signed long long __offset,
signed char *__ptr) {
- return *(vector signed char *)(__ptr + __offset);
+ return *(unaligned_vec_schar *)(__ptr + __offset);
}
static inline __ATTRS_o_ai vector unsigned char
vec_xl(signed long long __offset, unsigned char *__ptr) {
- return *(vector unsigned char *)(__ptr + __offset);
+ return *(unaligned_vec_uchar*)(__ptr + __offset);
}
static inline __ATTRS_o_ai vector signed short vec_xl(signed long long __offset,
signed short *__ptr) {
- return *(vector signed short *)(__ptr + __offset);
+ signed char *__addr = (signed char *)__ptr + __offset;
+ return *(unaligned_vec_sshort *)__addr;
}
static inline __ATTRS_o_ai vector unsigned short
vec_xl(signed long long __offset, unsigned short *__ptr) {
- return *(vector unsigned short *)(__ptr + __offset);
+ signed char *__addr = (signed char *)__ptr + __offset;
+ return *(unaligned_vec_ushort *)__addr;
}
static inline __ATTRS_o_ai vector signed int vec_xl(signed long long __offset,
signed int *__ptr) {
- return *(vector signed int *)(__ptr + __offset);
+ signed char *__addr = (signed char *)__ptr + __offset;
+ return *(unaligned_vec_sint *)__addr;
}
static inline __ATTRS_o_ai vector unsigned int vec_xl(signed long long __offset,
unsigned int *__ptr) {
- return *(vector unsigned int *)(__ptr + __offset);
+ signed char *__addr = (signed char *)__ptr + __offset;
+ return *(unaligned_vec_uint *)__addr;
}
static inline __ATTRS_o_ai vector float vec_xl(signed long long __offset,
float *__ptr) {
- return *(vector float *)(__ptr + __offset);
+ signed char *__addr = (signed char *)__ptr + __offset;
+ return *(unaligned_vec_float *)__addr;
}
#ifdef __VSX__
+typedef vector signed long long unaligned_vec_sll __attribute__((aligned(1)));
+typedef vector unsigned long long unaligned_vec_ull __attribute__((aligned(1)));
+typedef vector double unaligned_vec_double __attribute__((aligned(1)));
+
static inline __ATTRS_o_ai vector signed long long
vec_xl(signed long long __offset, signed long long *__ptr) {
- return *(vector signed long long *)(__ptr + __offset);
+ signed char *__addr = (signed char *)__ptr + __offset;
+ return *(unaligned_vec_sll *)__addr;
}
static inline __ATTRS_o_ai vector unsigned long long
vec_xl(signed long long __offset, unsigned long long *__ptr) {
- return *(vector unsigned long long *)(__ptr + __offset);
+ signed char *__addr = (signed char *)__ptr + __offset;
+ return *(unaligned_vec_ull *)__addr;
}
static inline __ATTRS_o_ai vector double vec_xl(signed long long __offset,
double *__ptr) {
- return *(vector double *)(__ptr + __offset);
+ signed char *__addr = (signed char *)__ptr + __offset;
+ return *(unaligned_vec_double *)__addr;
}
#endif
#if defined(__POWER8_VECTOR__) && defined(__powerpc64__)
+typedef vector signed __int128 unaligned_vec_si128 __attribute__((aligned(1)));
+typedef vector unsigned __int128 unaligned_vec_ui128
+ __attribute__((aligned(1)));
static inline __ATTRS_o_ai vector signed __int128
vec_xl(signed long long __offset, signed __int128 *__ptr) {
- return *(vector signed __int128 *)(__ptr + __offset);
+ signed char *__addr = (signed char *)__ptr + __offset;
+ return *(unaligned_vec_si128 *)__addr;
}
static inline __ATTRS_o_ai vector unsigned __int128
vec_xl(signed long long __offset, unsigned __int128 *__ptr) {
- return *(vector unsigned __int128 *)(__ptr + __offset);
+ signed char *__addr = (signed char *)__ptr + __offset;
+ return *(unaligned_vec_ui128 *)__addr;
}
#endif
#ifdef __LITTLE_ENDIAN__
static __inline__ vector signed char __ATTRS_o_ai
vec_xl_be(signed long long __offset, signed char *__ptr) {
- vector signed char __vec = __builtin_vsx_lxvd2x_be(__offset, __ptr);
+ vector signed char __vec = (vector signed char)__builtin_vsx_lxvd2x_be(__offset, __ptr);
return __builtin_shufflevector(__vec, __vec, 7, 6, 5, 4, 3, 2, 1, 0, 15, 14,
13, 12, 11, 10, 9, 8);
}
static __inline__ vector unsigned char __ATTRS_o_ai
vec_xl_be(signed long long __offset, unsigned char *__ptr) {
- vector unsigned char __vec = __builtin_vsx_lxvd2x_be(__offset, __ptr);
+ vector unsigned char __vec = (vector unsigned char)__builtin_vsx_lxvd2x_be(__offset, __ptr);
return __builtin_shufflevector(__vec, __vec, 7, 6, 5, 4, 3, 2, 1, 0, 15, 14,
13, 12, 11, 10, 9, 8);
}
static __inline__ vector signed short __ATTRS_o_ai
vec_xl_be(signed long long __offset, signed short *__ptr) {
- vector signed short __vec = __builtin_vsx_lxvd2x_be(__offset, __ptr);
+ vector signed short __vec = (vector signed short)__builtin_vsx_lxvd2x_be(__offset, __ptr);
return __builtin_shufflevector(__vec, __vec, 3, 2, 1, 0, 7, 6, 5, 4);
}
static __inline__ vector unsigned short __ATTRS_o_ai
vec_xl_be(signed long long __offset, unsigned short *__ptr) {
- vector unsigned short __vec = __builtin_vsx_lxvd2x_be(__offset, __ptr);
+ vector unsigned short __vec = (vector unsigned short)__builtin_vsx_lxvd2x_be(__offset, __ptr);
return __builtin_shufflevector(__vec, __vec, 3, 2, 1, 0, 7, 6, 5, 4);
}
static inline __ATTRS_o_ai void vec_xst(vector signed char __vec,
signed long long __offset,
signed char *__ptr) {
- *(vector signed char *)(__ptr + __offset) = __vec;
+ *(unaligned_vec_schar *)(__ptr + __offset) = __vec;
}
static inline __ATTRS_o_ai void vec_xst(vector unsigned char __vec,
signed long long __offset,
unsigned char *__ptr) {
- *(vector unsigned char *)(__ptr + __offset) = __vec;
+ *(unaligned_vec_uchar *)(__ptr + __offset) = __vec;
}
static inline __ATTRS_o_ai void vec_xst(vector signed short __vec,
signed long long __offset,
signed short *__ptr) {
- *(vector signed short *)(__ptr + __offset) = __vec;
+ signed char *__addr = (signed char *)__ptr + __offset;
+ *(unaligned_vec_sshort *)__addr = __vec;
}
static inline __ATTRS_o_ai void vec_xst(vector unsigned short __vec,
signed long long __offset,
unsigned short *__ptr) {
- *(vector unsigned short *)(__ptr + __offset) = __vec;
+ signed char *__addr = (signed char *)__ptr + __offset;
+ *(unaligned_vec_ushort *)__addr = __vec;
}
static inline __ATTRS_o_ai void vec_xst(vector signed int __vec,
signed long long __offset,
signed int *__ptr) {
- *(vector signed int *)(__ptr + __offset) = __vec;
+ signed char *__addr = (signed char *)__ptr + __offset;
+ *(unaligned_vec_sint *)__addr = __vec;
}
static inline __ATTRS_o_ai void vec_xst(vector unsigned int __vec,
signed long long __offset,
unsigned int *__ptr) {
- *(vector unsigned int *)(__ptr + __offset) = __vec;
+ signed char *__addr = (signed char *)__ptr + __offset;
+ *(unaligned_vec_uint *)__addr = __vec;
}
static inline __ATTRS_o_ai void vec_xst(vector float __vec,
signed long long __offset,
float *__ptr) {
- *(vector float *)(__ptr + __offset) = __vec;
+ signed char *__addr = (signed char *)__ptr + __offset;
+ *(unaligned_vec_float *)__addr = __vec;
}
#ifdef __VSX__
static inline __ATTRS_o_ai void vec_xst(vector signed long long __vec,
signed long long __offset,
signed long long *__ptr) {
- *(vector signed long long *)(__ptr + __offset) = __vec;
+ signed char *__addr = (signed char *)__ptr + __offset;
+ *(unaligned_vec_sll *)__addr = __vec;
}
static inline __ATTRS_o_ai void vec_xst(vector unsigned long long __vec,
signed long long __offset,
unsigned long long *__ptr) {
- *(vector unsigned long long *)(__ptr + __offset) = __vec;
+ signed char *__addr = (signed char *)__ptr + __offset;
+ *(unaligned_vec_ull *)__addr = __vec;
}
static inline __ATTRS_o_ai void vec_xst(vector double __vec,
signed long long __offset,
double *__ptr) {
- *(vector double *)(__ptr + __offset) = __vec;
+ signed char *__addr = (signed char *)__ptr + __offset;
+ *(unaligned_vec_double *)__addr = __vec;
}
#endif
static inline __ATTRS_o_ai void vec_xst(vector signed __int128 __vec,
signed long long __offset,
signed __int128 *__ptr) {
- *(vector signed __int128 *)(__ptr + __offset) = __vec;
+ signed char *__addr = (signed char *)__ptr + __offset;
+ *(unaligned_vec_si128 *)__addr = __vec;
}
static inline __ATTRS_o_ai void vec_xst(vector unsigned __int128 __vec,
signed long long __offset,
unsigned __int128 *__ptr) {
- *(vector unsigned __int128 *)(__ptr + __offset) = __vec;
+ signed char *__addr = (signed char *)__ptr + __offset;
+ *(unaligned_vec_ui128 *)__addr = __vec;
}
#endif
vector signed char __tmp =
__builtin_shufflevector(__vec, __vec, 7, 6, 5, 4, 3, 2, 1, 0, 15, 14,
13, 12, 11, 10, 9, 8);
- __builtin_vsx_stxvd2x_be(__tmp, __offset, __ptr);
+ typedef __attribute__((vector_size(sizeof(__tmp)))) double __vector_double;
+ __builtin_vsx_stxvd2x_be((__vector_double)__tmp, __offset, __ptr);
}
static __inline__ void __ATTRS_o_ai vec_xst_be(vector unsigned char __vec,
vector unsigned char __tmp =
__builtin_shufflevector(__vec, __vec, 7, 6, 5, 4, 3, 2, 1, 0, 15, 14,
13, 12, 11, 10, 9, 8);
- __builtin_vsx_stxvd2x_be(__tmp, __offset, __ptr);
+ typedef __attribute__((vector_size(sizeof(__tmp)))) double __vector_double;
+ __builtin_vsx_stxvd2x_be((__vector_double)__tmp, __offset, __ptr);
}
static __inline__ void __ATTRS_o_ai vec_xst_be(vector signed short __vec,
signed short *__ptr) {
vector signed short __tmp =
__builtin_shufflevector(__vec, __vec, 3, 2, 1, 0, 7, 6, 5, 4);
- __builtin_vsx_stxvd2x_be(__tmp, __offset, __ptr);
+ typedef __attribute__((vector_size(sizeof(__tmp)))) double __vector_double;
+ __builtin_vsx_stxvd2x_be((__vector_double)__tmp, __offset, __ptr);
}
static __inline__ void __ATTRS_o_ai vec_xst_be(vector unsigned short __vec,
unsigned short *__ptr) {
vector unsigned short __tmp =
__builtin_shufflevector(__vec, __vec, 3, 2, 1, 0, 7, 6, 5, 4);
- __builtin_vsx_stxvd2x_be(__tmp, __offset, __ptr);
+ typedef __attribute__((vector_size(sizeof(__tmp)))) double __vector_double;
+ __builtin_vsx_stxvd2x_be((__vector_double)__tmp, __offset, __ptr);
}
static __inline__ void __ATTRS_o_ai vec_xst_be(vector signed int __vec,
static __inline__ void __ATTRS_o_ai vec_xst_be(vector unsigned int __vec,
signed long long __offset,
unsigned int *__ptr) {
- __builtin_vsx_stxvw4x_be(__vec, __offset, __ptr);
+ __builtin_vsx_stxvw4x_be((vector int)__vec, __offset, __ptr);
}
static __inline__ void __ATTRS_o_ai vec_xst_be(vector float __vec,
signed long long __offset,
float *__ptr) {
- __builtin_vsx_stxvw4x_be(__vec, __offset, __ptr);
+ __builtin_vsx_stxvw4x_be((vector int)__vec, __offset, __ptr);
}
#ifdef __VSX__
static __inline__ void __ATTRS_o_ai vec_xst_be(vector signed long long __vec,
signed long long __offset,
signed long long *__ptr) {
- __builtin_vsx_stxvd2x_be(__vec, __offset, __ptr);
+ __builtin_vsx_stxvd2x_be((vector double)__vec, __offset, __ptr);
}
static __inline__ void __ATTRS_o_ai vec_xst_be(vector unsigned long long __vec,
signed long long __offset,
unsigned long long *__ptr) {
- __builtin_vsx_stxvd2x_be(__vec, __offset, __ptr);
+ __builtin_vsx_stxvd2x_be((vector double)__vec, __offset, __ptr);
}
static __inline__ void __ATTRS_o_ai vec_xst_be(vector double __vec,
signed long long __offset,
double *__ptr) {
- __builtin_vsx_stxvd2x_be(__vec, __offset, __ptr);
+ __builtin_vsx_stxvd2x_be((vector double)__vec, __offset, __ptr);
}
#endif
#endif
#ifdef __POWER9_VECTOR__
-#define vec_test_data_class(__a, __b) \
- _Generic((__a), \
- vector float: \
- (vector bool int)__builtin_vsx_xvtstdcsp((__a), (__b)), \
- vector double: \
- (vector bool long long)__builtin_vsx_xvtstdcdp((__a), (__b)) \
- )
+#define vec_test_data_class(__a, __b) \
+ _Generic( \
+ (__a), vector float \
+ : (vector bool int)__builtin_vsx_xvtstdcsp((vector float)(__a), (__b)), \
+ vector double \
+ : (vector bool long long)__builtin_vsx_xvtstdcdp((vector double)(__a), \
+ (__b)))
#endif /* #ifdef __POWER9_VECTOR__ */
static vector signed char __ATTRS_o_ai vec_nabs(vector signed char __a) {
return __builtin_altivec_vminsb(__a, -__a);
}
+
+#ifdef __POWER10_VECTOR__
+/* vec_pdep */
+
+static __inline__ vector unsigned long long __ATTRS_o_ai
+vec_pdep(vector unsigned long long __a, vector unsigned long long __b) {
+ return __builtin_altivec_vpdepd(__a, __b);
+}
+
+/* vec_pext */
+
+static __inline__ vector unsigned long long __ATTRS_o_ai
+vec_pext(vector unsigned long long __a, vector unsigned long long __b) {
+ return __builtin_altivec_vpextd(__a, __b);
+}
+
+/* vec_cfuge */
+
+static __inline__ vector unsigned long long __ATTRS_o_ai
+vec_cfuge(vector unsigned long long __a, vector unsigned long long __b) {
+ return __builtin_altivec_vcfuged(__a, __b);
+}
+
+/* vec_gnb */
+
+#define vec_gnb(__a, __b) __builtin_altivec_vgnb(__a, __b)
+
+/* vec_ternarylogic */
+#ifdef __VSX__
+#define vec_ternarylogic(__a, __b, __c, __imm) \
+ _Generic((__a), vector unsigned char \
+ : __builtin_vsx_xxeval((vector unsigned long long)(__a), \
+ (vector unsigned long long)(__b), \
+ (vector unsigned long long)(__c), (__imm)), \
+ vector unsigned short \
+ : __builtin_vsx_xxeval((vector unsigned long long)(__a), \
+ (vector unsigned long long)(__b), \
+ (vector unsigned long long)(__c), (__imm)), \
+ vector unsigned int \
+ : __builtin_vsx_xxeval((vector unsigned long long)(__a), \
+ (vector unsigned long long)(__b), \
+ (vector unsigned long long)(__c), (__imm)), \
+ vector unsigned long long \
+ : __builtin_vsx_xxeval((vector unsigned long long)(__a), \
+ (vector unsigned long long)(__b), \
+ (vector unsigned long long)(__c), (__imm)), \
+ vector unsigned __int128 \
+ : __builtin_vsx_xxeval((vector unsigned long long)(__a), \
+ (vector unsigned long long)(__b), \
+ (vector unsigned long long)(__c), (__imm)))
+#endif /* __VSX__ */
+
+/* vec_genpcvm */
+
+#ifdef __VSX__
+#define vec_genpcvm(__a, __imm) \
+ _Generic((__a), vector unsigned char \
+ : __builtin_vsx_xxgenpcvbm((__a), (int)(__imm)), \
+ vector unsigned short \
+ : __builtin_vsx_xxgenpcvhm((__a), (int)(__imm)), \
+ vector unsigned int \
+ : __builtin_vsx_xxgenpcvwm((__a), (int)(__imm)), \
+ vector unsigned long long \
+ : __builtin_vsx_xxgenpcvdm((__a), (int)(__imm)))
+#endif /* __VSX__ */
+
+/* vec_clrl */
+
+static __inline__ vector signed char __ATTRS_o_ai
+vec_clrl(vector signed char __a, unsigned int __n) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vclrrb(__a, __n);
+#else
+ return __builtin_altivec_vclrlb( __a, __n);
+#endif
+}
+
+static __inline__ vector unsigned char __ATTRS_o_ai
+vec_clrl(vector unsigned char __a, unsigned int __n) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vclrrb((vector signed char)__a, __n);
+#else
+ return __builtin_altivec_vclrlb((vector signed char)__a, __n);
+#endif
+}
+
+/* vec_clrr */
+
+static __inline__ vector signed char __ATTRS_o_ai
+vec_clrr(vector signed char __a, unsigned int __n) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vclrlb(__a, __n);
+#else
+ return __builtin_altivec_vclrrb( __a, __n);
+#endif
+}
+
+static __inline__ vector unsigned char __ATTRS_o_ai
+vec_clrr(vector unsigned char __a, unsigned int __n) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vclrlb((vector signed char)__a, __n);
+#else
+ return __builtin_altivec_vclrrb((vector signed char)__a, __n);
+#endif
+}
+
+/* vec_cntlzm */
+
+static __inline__ vector unsigned long long __ATTRS_o_ai
+vec_cntlzm(vector unsigned long long __a, vector unsigned long long __b) {
+ return __builtin_altivec_vclzdm(__a, __b);
+}
+
+/* vec_cnttzm */
+
+static __inline__ vector unsigned long long __ATTRS_o_ai
+vec_cnttzm(vector unsigned long long __a, vector unsigned long long __b) {
+ return __builtin_altivec_vctzdm(__a, __b);
+}
+
+/* vec_sldbi */
+
+#define vec_sldb(__a, __b, __c) __builtin_altivec_vsldbi(__a, __b, (__c & 0x7))
+
+/* vec_srdbi */
+
+#define vec_srdb(__a, __b, __c) __builtin_altivec_vsrdbi(__a, __b, (__c & 0x7))
+
+/* vec_insertl */
+
+static __inline__ vector unsigned char __ATTRS_o_ai
+vec_insertl(unsigned char __a, vector unsigned char __b, unsigned int __c) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vinsbrx(__b, __c, __a);
+#else
+ return __builtin_altivec_vinsblx(__b, __c, __a);
+#endif
+}
+
+static __inline__ vector unsigned short __ATTRS_o_ai
+vec_insertl(unsigned short __a, vector unsigned short __b, unsigned int __c) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vinshrx(__b, __c, __a);
+#else
+ return __builtin_altivec_vinshlx(__b, __c, __a);
+#endif
+}
+
+static __inline__ vector unsigned int __ATTRS_o_ai
+vec_insertl(unsigned int __a, vector unsigned int __b, unsigned int __c) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vinswrx(__b, __c, __a);
+#else
+ return __builtin_altivec_vinswlx(__b, __c, __a);
+#endif
+}
+
+static __inline__ vector unsigned long long __ATTRS_o_ai
+vec_insertl(unsigned long long __a, vector unsigned long long __b,
+ unsigned int __c) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vinsdrx(__b, __c, __a);
+#else
+ return __builtin_altivec_vinsdlx(__b, __c, __a);
+#endif
+}
+
+static __inline__ vector unsigned char __ATTRS_o_ai
+vec_insertl(vector unsigned char __a, vector unsigned char __b,
+ unsigned int __c) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vinsbvrx(__b, __c, __a);
+#else
+ return __builtin_altivec_vinsbvlx(__b, __c, __a);
+#endif
+}
+
+static __inline__ vector unsigned short __ATTRS_o_ai
+vec_insertl(vector unsigned short __a, vector unsigned short __b,
+ unsigned int __c) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vinshvrx(__b, __c, __a);
+#else
+ return __builtin_altivec_vinshvlx(__b, __c, __a);
+#endif
+}
+
+static __inline__ vector unsigned int __ATTRS_o_ai
+vec_insertl(vector unsigned int __a, vector unsigned int __b,
+ unsigned int __c) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vinswvrx(__b, __c, __a);
+#else
+ return __builtin_altivec_vinswvlx(__b, __c, __a);
+#endif
+}
+
+/* vec_inserth */
+
+static __inline__ vector unsigned char __ATTRS_o_ai
+vec_inserth(unsigned char __a, vector unsigned char __b, unsigned int __c) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vinsblx(__b, __c, __a);
+#else
+ return __builtin_altivec_vinsbrx(__b, __c, __a);
+#endif
+}
+
+static __inline__ vector unsigned short __ATTRS_o_ai
+vec_inserth(unsigned short __a, vector unsigned short __b, unsigned int __c) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vinshlx(__b, __c, __a);
+#else
+ return __builtin_altivec_vinshrx(__b, __c, __a);
+#endif
+}
+
+static __inline__ vector unsigned int __ATTRS_o_ai
+vec_inserth(unsigned int __a, vector unsigned int __b, unsigned int __c) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vinswlx(__b, __c, __a);
+#else
+ return __builtin_altivec_vinswrx(__b, __c, __a);
+#endif
+}
+
+static __inline__ vector unsigned long long __ATTRS_o_ai
+vec_inserth(unsigned long long __a, vector unsigned long long __b,
+ unsigned int __c) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vinsdlx(__b, __c, __a);
+#else
+ return __builtin_altivec_vinsdrx(__b, __c, __a);
+#endif
+}
+
+static __inline__ vector unsigned char __ATTRS_o_ai
+vec_inserth(vector unsigned char __a, vector unsigned char __b,
+ unsigned int __c) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vinsbvlx(__b, __c, __a);
+#else
+ return __builtin_altivec_vinsbvrx(__b, __c, __a);
+#endif
+}
+
+static __inline__ vector unsigned short __ATTRS_o_ai
+vec_inserth(vector unsigned short __a, vector unsigned short __b,
+ unsigned int __c) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vinshvlx(__b, __c, __a);
+#else
+ return __builtin_altivec_vinshvrx(__b, __c, __a);
+#endif
+}
+
+static __inline__ vector unsigned int __ATTRS_o_ai
+vec_inserth(vector unsigned int __a, vector unsigned int __b,
+ unsigned int __c) {
+#ifdef __LITTLE_ENDIAN__
+ return __builtin_altivec_vinswvlx(__b, __c, __a);
+#else
+ return __builtin_altivec_vinswvrx(__b, __c, __a);
+#endif
+}
+
+#ifdef __VSX__
+
+/* vec_permx */
+
+#define vec_permx(__a, __b, __c, __d) \
+ __builtin_vsx_xxpermx((__a), (__b), (__c), (__d))
+
+/* vec_blendv */
+
+static __inline__ vector signed char __ATTRS_o_ai
+vec_blendv(vector signed char __a, vector signed char __b,
+ vector unsigned char __c) {
+ return __builtin_vsx_xxblendvb(__a, __b, __c);
+}
+
+static __inline__ vector unsigned char __ATTRS_o_ai
+vec_blendv(vector unsigned char __a, vector unsigned char __b,
+ vector unsigned char __c) {
+ return __builtin_vsx_xxblendvb(__a, __b, __c);
+}
+
+static __inline__ vector signed short __ATTRS_o_ai
+vec_blendv(vector signed short __a, vector signed short __b,
+ vector unsigned short __c) {
+ return __builtin_vsx_xxblendvh(__a, __b, __c);
+}
+
+static __inline__ vector unsigned short __ATTRS_o_ai
+vec_blendv(vector unsigned short __a, vector unsigned short __b,
+ vector unsigned short __c) {
+ return __builtin_vsx_xxblendvh(__a, __b, __c);
+}
+
+static __inline__ vector signed int __ATTRS_o_ai
+vec_blendv(vector signed int __a, vector signed int __b,
+ vector unsigned int __c) {
+ return __builtin_vsx_xxblendvw(__a, __b, __c);
+}
+
+static __inline__ vector unsigned int __ATTRS_o_ai
+vec_blendv(vector unsigned int __a, vector unsigned int __b,
+ vector unsigned int __c) {
+ return __builtin_vsx_xxblendvw(__a, __b, __c);
+}
+
+static __inline__ vector signed long long __ATTRS_o_ai
+vec_blendv(vector signed long long __a, vector signed long long __b,
+ vector unsigned long long __c) {
+ return __builtin_vsx_xxblendvd(__a, __b, __c);
+}
+
+static __inline__ vector unsigned long long __ATTRS_o_ai
+vec_blendv(vector unsigned long long __a, vector unsigned long long __b,
+ vector unsigned long long __c) {
+ return __builtin_vsx_xxblendvd(__a, __b, __c);
+}
+
+static __inline__ vector float __ATTRS_o_ai
+vec_blendv(vector float __a, vector float __b, vector unsigned int __c) {
+ return __builtin_vsx_xxblendvw(__a, __b, __c);
+}
+
+static __inline__ vector double __ATTRS_o_ai
+vec_blendv(vector double __a, vector double __b,
+ vector unsigned long long __c) {
+ return __builtin_vsx_xxblendvd(__a, __b, __c);
+}
+
+/* vec_splati */
+
+#define vec_splati(__a) \
+ _Generic((__a), signed int \
+ : ((vector signed int)__a), unsigned int \
+ : ((vector unsigned int)__a), float \
+ : ((vector float)__a))
+
+/* vec_spatid */
+
+static __inline__ vector double __ATTRS_o_ai vec_splatid(const float __a) {
+ return ((vector double)((double)__a));
+}
+
+/* vec_splati_ins */
+
+static __inline__ vector signed int __ATTRS_o_ai vec_splati_ins(
+ vector signed int __a, const unsigned int __b, const signed int __c) {
+#ifdef __LITTLE_ENDIAN__
+ __a[1 - __b] = __c;
+ __a[3 - __b] = __c;
+#else
+ __a[__b] = __c;
+ __a[2 + __b] = __c;
+#endif
+ return __a;
+}
+
+static __inline__ vector unsigned int __ATTRS_o_ai vec_splati_ins(
+ vector unsigned int __a, const unsigned int __b, const unsigned int __c) {
+#ifdef __LITTLE_ENDIAN__
+ __a[1 - __b] = __c;
+ __a[3 - __b] = __c;
+#else
+ __a[__b] = __c;
+ __a[2 + __b] = __c;
+#endif
+ return __a;
+}
+
+static __inline__ vector float __ATTRS_o_ai
+vec_splati_ins(vector float __a, const unsigned int __b, const float __c) {
+#ifdef __LITTLE_ENDIAN__
+ __a[1 - __b] = __c;
+ __a[3 - __b] = __c;
+#else
+ __a[__b] = __c;
+ __a[2 + __b] = __c;
+#endif
+ return __a;
+}
+
+/* vec_test_lsbb_all_ones */
+
+static __inline__ int __ATTRS_o_ai
+vec_test_lsbb_all_ones(vector unsigned char __a) {
+ return __builtin_vsx_xvtlsbb(__a, 1);
+}
+
+/* vec_test_lsbb_all_zeros */
+
+static __inline__ int __ATTRS_o_ai
+vec_test_lsbb_all_zeros(vector unsigned char __a) {
+ return __builtin_vsx_xvtlsbb(__a, 0);
+}
+#endif /* __VSX__ */
+#endif /* __POWER10_VECTOR__ */
+
#undef __ATTRS_o_ai
#endif /* __ALTIVEC_H */
/*===---- ammintrin.h - SSE4a intrinsics -----------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#include <pmmintrin.h>
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("sse4a")))
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("sse4a"), __min_vector_width__(128)))
-/// \brief Extracts the specified bits from the lower 64 bits of the 128-bit
+/// Extracts the specified bits from the lower 64 bits of the 128-bit
/// integer vector operand at the index \a idx and of the length \a len.
///
/// \headerfile <x86intrin.h>
((__m128i)__builtin_ia32_extrqi((__v2di)(__m128i)(x), \
(char)(len), (char)(idx)))
-/// \brief Extracts the specified bits from the lower 64 bits of the 128-bit
+/// Extracts the specified bits from the lower 64 bits of the 128-bit
/// integer vector operand at the index and of the length specified by
/// \a __y.
///
return (__m128i)__builtin_ia32_extrq((__v2di)__x, (__v16qi)__y);
}
-/// \brief Inserts bits of a specified length from the source integer vector
+/// Inserts bits of a specified length from the source integer vector
/// \a y into the lower 64 bits of the destination integer vector \a x at
/// the index \a idx and of the length \a len.
///
(__v2di)(__m128i)(y), \
(char)(len), (char)(idx)))
-/// \brief Inserts bits of a specified length from the source integer vector
+/// Inserts bits of a specified length from the source integer vector
/// \a __y into the lower 64 bits of the destination integer vector \a __x
/// at the index and of the length specified by \a __y.
///
return (__m128i)__builtin_ia32_insertq((__v2di)__x, (__v2di)__y);
}
-/// \brief Stores a 64-bit double-precision value in a 64-bit memory location.
+/// Stores a 64-bit double-precision value in a 64-bit memory location.
/// To minimize caching, the data is flagged as non-temporal (unlikely to be
/// used again soon).
///
__builtin_ia32_movntsd(__p, (__v2df)__a);
}
-/// \brief Stores a 32-bit single-precision floating-point value in a 32-bit
+/// Stores a 32-bit single-precision floating-point value in a 32-bit
/// memory location. To minimize caching, the data is flagged as
/// non-temporal (unlikely to be used again soon).
///
--- /dev/null
+/*===--------------- amxintrin.h - AMX intrinsics -*- C/C++ -*---------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===------------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <amxintrin.h> directly; include <immintrin.h> instead."
+#endif /* __IMMINTRIN_H */
+
+#ifndef __AMXINTRIN_H
+#define __AMXINTRIN_H
+#ifdef __x86_64__
+
+#define __DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("amx-tile")))
+
+/// Load tile configuration from a 64-byte memory location specified by
+/// "mem_addr". The tile configuration includes the tile type palette, the
+/// number of bytes per row, and the number of rows. If the specified
+/// palette_id is zero, that signifies the init state for both the tile
+/// config and the tile data, and the tiles are zeroed. Any invalid
+/// configurations will result in #GP fault.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> LDTILECFG </c> instruction.
+///
+/// \param __config
+/// A pointer to 512-bits configuration
+static __inline__ void __DEFAULT_FN_ATTRS
+_tile_loadconfig(const void *__config)
+{
+ __builtin_ia32_tile_loadconfig(__config);
+}
+
+/// Stores the current tile configuration to a 64-byte memory location
+/// specified by "mem_addr". The tile configuration includes the tile type
+/// palette, the number of bytes per row, and the number of rows. If tiles
+/// are not configured, all zeroes will be stored to memory.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> STTILECFG </c> instruction.
+///
+/// \param __config
+/// A pointer to 512-bits configuration
+static __inline__ void __DEFAULT_FN_ATTRS
+_tile_storeconfig(void *__config)
+{
+ __builtin_ia32_tile_storeconfig(__config);
+}
+
+/// Release the tile configuration to return to the init state, which
+/// releases all storage it currently holds.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> TILERELEASE </c> instruction.
+static __inline__ void __DEFAULT_FN_ATTRS
+_tile_release(void)
+{
+ __builtin_ia32_tilerelease();
+}
+
+/// Load tile rows from memory specifieid by "base" address and "stride" into
+/// destination tile "dst" using the tile configuration previously configured
+/// via "_tile_loadconfig".
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> TILELOADD </c> instruction.
+///
+/// \param dst
+/// A destination tile. Max size is 1024 Bytes.
+/// \param base
+/// A pointer to base address.
+/// \param stride
+/// The stride between the rows' data to be loaded in memory.
+#define _tile_loadd(dst, base, stride) \
+ __builtin_ia32_tileloadd64((dst), ((const void *)(base)), (__SIZE_TYPE__)(stride))
+
+/// Load tile rows from memory specifieid by "base" address and "stride" into
+/// destination tile "dst" using the tile configuration previously configured
+/// via "_tile_loadconfig". This intrinsic provides a hint to the implementation
+/// that the data will likely not be reused in the near future and the data
+/// caching can be optimized accordingly.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> TILELOADDT1 </c> instruction.
+///
+/// \param dst
+/// A destination tile. Max size is 1024 Bytes.
+/// \param base
+/// A pointer to base address.
+/// \param stride
+/// The stride between the rows' data to be loaded in memory.
+#define _tile_stream_loadd(dst, base, stride) \
+ __builtin_ia32_tileloaddt164((dst), ((const void *)(base)), (__SIZE_TYPE__)(stride))
+
+/// Store the tile specified by "src" to memory specifieid by "base" address and
+/// "stride" using the tile configuration previously configured via
+/// "_tile_loadconfig".
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> TILESTORED </c> instruction.
+///
+/// \param dst
+/// A destination tile. Max size is 1024 Bytes.
+/// \param base
+/// A pointer to base address.
+/// \param stride
+/// The stride between the rows' data to be stored in memory.
+#define _tile_stored(dst, base, stride) \
+ __builtin_ia32_tilestored64((dst), ((void *)(base)), (__SIZE_TYPE__)(stride))
+
+/// Zero the tile specified by "tdest".
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> TILEZERO </c> instruction.
+///
+/// \param tile
+/// The destination tile to be zero. Max size is 1024 Bytes.
+#define _tile_zero(tile) __builtin_ia32_tilezero((tile))
+
+/// Compute dot-product of bytes in tiles with a source/destination accumulator.
+/// Multiply groups of 4 adjacent pairs of signed 8-bit integers in src0 with
+/// corresponding signed 8-bit integers in src1, producing 4 intermediate 32-bit
+/// results. Sum these 4 results with the corresponding 32-bit integer in "dst",
+/// and store the 32-bit result back to tile "dst".
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> TDPBSSD </c> instruction.
+///
+/// \param dst
+/// The destination tile. Max size is 1024 Bytes.
+/// \param src0
+/// The 1st source tile. Max size is 1024 Bytes.
+/// \param src1
+/// The 2nd source tile. Max size is 1024 Bytes.
+#define _tile_dpbssd(dst, src0, src1) __builtin_ia32_tdpbssd((dst), (src0), (src1))
+
+/// Compute dot-product of bytes in tiles with a source/destination accumulator.
+/// Multiply groups of 4 adjacent pairs of signed 8-bit integers in src0 with
+/// corresponding unsigned 8-bit integers in src1, producing 4 intermediate
+/// 32-bit results. Sum these 4 results with the corresponding 32-bit integer
+/// in "dst", and store the 32-bit result back to tile "dst".
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> TDPBSUD </c> instruction.
+///
+/// \param dst
+/// The destination tile. Max size is 1024 Bytes.
+/// \param src0
+/// The 1st source tile. Max size is 1024 Bytes.
+/// \param src1
+/// The 2nd source tile. Max size is 1024 Bytes.
+#define _tile_dpbsud(dst, src0, src1) __builtin_ia32_tdpbsud((dst), (src0), (src1))
+
+/// Compute dot-product of bytes in tiles with a source/destination accumulator.
+/// Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in src0 with
+/// corresponding signed 8-bit integers in src1, producing 4 intermediate 32-bit
+/// results. Sum these 4 results with the corresponding 32-bit integer in "dst",
+/// and store the 32-bit result back to tile "dst".
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> TDPBUSD </c> instruction.
+///
+/// \param dst
+/// The destination tile. Max size is 1024 Bytes.
+/// \param src0
+/// The 1st source tile. Max size is 1024 Bytes.
+/// \param src1
+/// The 2nd source tile. Max size is 1024 Bytes.
+#define _tile_dpbusd(dst, src0, src1) __builtin_ia32_tdpbusd((dst), (src0), (src1))
+
+/// Compute dot-product of bytes in tiles with a source/destination accumulator.
+/// Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in src0 with
+/// corresponding unsigned 8-bit integers in src1, producing 4 intermediate
+/// 32-bit results. Sum these 4 results with the corresponding 32-bit integer in
+/// "dst", and store the 32-bit result back to tile "dst".
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> TDPBUUD </c> instruction.
+///
+/// \param dst
+/// The destination tile. Max size is 1024 Bytes.
+/// \param src0
+/// The 1st source tile. Max size is 1024 Bytes.
+/// \param src1
+/// The 2nd source tile. Max size is 1024 Bytes.
+#define _tile_dpbuud(dst, src0, src1) __builtin_ia32_tdpbuud((dst), (src0), (src1))
+
+/// Compute dot-product of BF16 (16-bit) floating-point pairs in tiles src0 and
+/// src1, accumulating the intermediate single-precision (32-bit) floating-point
+/// elements with elements in "dst", and store the 32-bit result back to tile
+/// "dst".
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> TDPBF16PS </c> instruction.
+///
+/// \param dst
+/// The destination tile. Max size is 1024 Bytes.
+/// \param src0
+/// The 1st source tile. Max size is 1024 Bytes.
+/// \param src1
+/// The 2nd source tile. Max size is 1024 Bytes.
+#define _tile_dpbf16ps(dst, src0, src1) \
+ __builtin_ia32_tdpbf16ps((dst), (src0), (src1))
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif /* __x86_64__ */
+#endif /* __AMXINTRIN_H */
--- /dev/null
+/*===---- arm64intr.h - ARM64 Windows intrinsics -------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+/* Only include this if we're compiling for the windows platform. */
+#ifndef _MSC_VER
+#include_next <arm64intr.h>
+#else
+
+#ifndef __ARM64INTR_H
+#define __ARM64INTR_H
+
+typedef enum
+{
+ _ARM64_BARRIER_SY = 0xF,
+ _ARM64_BARRIER_ST = 0xE,
+ _ARM64_BARRIER_LD = 0xD,
+ _ARM64_BARRIER_ISH = 0xB,
+ _ARM64_BARRIER_ISHST = 0xA,
+ _ARM64_BARRIER_ISHLD = 0x9,
+ _ARM64_BARRIER_NSH = 0x7,
+ _ARM64_BARRIER_NSHST = 0x6,
+ _ARM64_BARRIER_NSHLD = 0x5,
+ _ARM64_BARRIER_OSH = 0x3,
+ _ARM64_BARRIER_OSHST = 0x2,
+ _ARM64_BARRIER_OSHLD = 0x1
+} _ARM64INTR_BARRIER_TYPE;
+
+#endif /* __ARM64INTR_H */
+#endif /* _MSC_VER */
/*===---- arm_acle.h - ARM Non-Neon intrinsics -----------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
/* 8 SYNCHRONIZATION, BARRIER AND HINT INTRINSICS */
/* 8.3 Memory barriers */
-#if !defined(_MSC_VER)
+#if !__has_builtin(__dmb)
#define __dmb(i) __builtin_arm_dmb(i)
+#endif
+#if !__has_builtin(__dsb)
#define __dsb(i) __builtin_arm_dsb(i)
+#endif
+#if !__has_builtin(__isb)
#define __isb(i) __builtin_arm_isb(i)
#endif
/* 8.4 Hints */
-#if !defined(_MSC_VER)
+#if !__has_builtin(__wfi)
static __inline__ void __attribute__((__always_inline__, __nodebug__)) __wfi(void) {
__builtin_arm_wfi();
}
+#endif
+#if !__has_builtin(__wfe)
static __inline__ void __attribute__((__always_inline__, __nodebug__)) __wfe(void) {
__builtin_arm_wfe();
}
+#endif
+#if !__has_builtin(__sev)
static __inline__ void __attribute__((__always_inline__, __nodebug__)) __sev(void) {
__builtin_arm_sev();
}
+#endif
+#if !__has_builtin(__sevl)
static __inline__ void __attribute__((__always_inline__, __nodebug__)) __sevl(void) {
__builtin_arm_sevl();
}
+#endif
+#if !__has_builtin(__yield)
static __inline__ void __attribute__((__always_inline__, __nodebug__)) __yield(void) {
__builtin_arm_yield();
}
#endif
/* 8.7 NOP */
+#if !defined(_MSC_VER) || !defined(__aarch64__)
static __inline__ void __attribute__((__always_inline__, __nodebug__)) __nop(void) {
__builtin_arm_nop();
}
+#endif
/* 9 DATA-PROCESSING INTRINSICS */
/* 9.2 Miscellaneous data-processing intrinsics */
return __builtin_clzll(__t);
}
+/* CLS */
+static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__))
+__cls(uint32_t __t) {
+ return __builtin_arm_cls(__t);
+}
+
+static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__))
+__clsl(unsigned long __t) {
+#if __SIZEOF_LONG__ == 4
+ return __builtin_arm_cls(__t);
+#else
+ return __builtin_arm_cls64(__t);
+#endif
+}
+
+static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__))
+__clsll(uint64_t __t) {
+ return __builtin_arm_cls64(__t);
+}
+
/* REV */
static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__))
__rev(uint32_t __t) {
}
#endif
+/* Armv8.3-A Javascript conversion intrinsic */
+#if __ARM_64BIT_STATE && defined(__ARM_FEATURE_JCVT)
+static __inline__ int32_t __attribute__((__always_inline__, __nodebug__))
+__jcvt(double __a) {
+ return __builtin_arm_jcvt(__a);
+}
+#endif
+
/* 10.1 Special register intrinsics */
#define __arm_rsr(sysreg) __builtin_arm_rsr(sysreg)
#define __arm_rsr64(sysreg) __builtin_arm_rsr64(sysreg)
#define __arm_rsrp(sysreg) __builtin_arm_rsrp(sysreg)
+#define __arm_rsrf(sysreg) __builtin_bit_cast(float, __arm_rsr(sysreg))
+#define __arm_rsrf64(sysreg) __builtin_bit_cast(double, __arm_rsr64(sysreg))
#define __arm_wsr(sysreg, v) __builtin_arm_wsr(sysreg, v)
#define __arm_wsr64(sysreg, v) __builtin_arm_wsr64(sysreg, v)
#define __arm_wsrp(sysreg, v) __builtin_arm_wsrp(sysreg, v)
+#define __arm_wsrf(sysreg, v) __arm_wsr(sysreg, __builtin_bit_cast(uint32_t, v))
+#define __arm_wsrf64(sysreg, v) __arm_wsr64(sysreg, __builtin_bit_cast(uint64_t, v))
+
+/* Memory Tagging Extensions (MTE) Intrinsics */
+#if __ARM_FEATURE_MEMORY_TAGGING
+#define __arm_mte_create_random_tag(__ptr, __mask) __builtin_arm_irg(__ptr, __mask)
+#define __arm_mte_increment_tag(__ptr, __tag_offset) __builtin_arm_addg(__ptr, __tag_offset)
+#define __arm_mte_exclude_tag(__ptr, __excluded) __builtin_arm_gmi(__ptr, __excluded)
+#define __arm_mte_get_tag(__ptr) __builtin_arm_ldg(__ptr)
+#define __arm_mte_set_tag(__ptr) __builtin_arm_stg(__ptr)
+#define __arm_mte_ptrdiff(__ptra, __ptrb) __builtin_arm_subp(__ptra, __ptrb)
+#endif
+
+/* Transactional Memory Extension (TME) Intrinsics */
+#if __ARM_FEATURE_TME
+
+#define _TMFAILURE_REASON 0x00007fffu
+#define _TMFAILURE_RTRY 0x00008000u
+#define _TMFAILURE_CNCL 0x00010000u
+#define _TMFAILURE_MEM 0x00020000u
+#define _TMFAILURE_IMP 0x00040000u
+#define _TMFAILURE_ERR 0x00080000u
+#define _TMFAILURE_SIZE 0x00100000u
+#define _TMFAILURE_NEST 0x00200000u
+#define _TMFAILURE_DBG 0x00400000u
+#define _TMFAILURE_INT 0x00800000u
+#define _TMFAILURE_TRIVIAL 0x01000000u
+
+#define __tstart() __builtin_arm_tstart()
+#define __tcommit() __builtin_arm_tcommit()
+#define __tcancel(__arg) __builtin_arm_tcancel(__arg)
+#define __ttest() __builtin_arm_ttest()
+
+#endif /* __ARM_FEATURE_TME */
#if defined(__cplusplus)
}
--- /dev/null
+/*===---- arm_bf16.h - ARM BF16 intrinsics -----------------------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __ARM_BF16_H
+#define __ARM_BF16_H
+
+typedef __bf16 bfloat16_t;
+#define __ai static __inline__ __attribute__((__always_inline__, __nodebug__))
+
+
+#undef __ai
+
+#endif
--- /dev/null
+/*===---- arm_cde.h - ARM CDE intrinsics -----------------------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __ARM_CDE_H
+#define __ARM_CDE_H
+
+#if !__ARM_FEATURE_CDE
+#error "CDE support not enabled"
+#endif
+
+#include <stdint.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_cx1)))
+uint32_t __arm_cx1(int, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_cx1a)))
+uint32_t __arm_cx1a(int, uint32_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_cx1d)))
+uint64_t __arm_cx1d(int, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_cx1da)))
+uint64_t __arm_cx1da(int, uint64_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_cx2)))
+uint32_t __arm_cx2(int, uint32_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_cx2a)))
+uint32_t __arm_cx2a(int, uint32_t, uint32_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_cx2d)))
+uint64_t __arm_cx2d(int, uint32_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_cx2da)))
+uint64_t __arm_cx2da(int, uint64_t, uint32_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_cx3)))
+uint32_t __arm_cx3(int, uint32_t, uint32_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_cx3a)))
+uint32_t __arm_cx3a(int, uint32_t, uint32_t, uint32_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_cx3d)))
+uint64_t __arm_cx3d(int, uint32_t, uint32_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_cx3da)))
+uint64_t __arm_cx3da(int, uint64_t, uint32_t, uint32_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_vcx1_u32)))
+uint32_t __arm_vcx1_u32(int, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_vcx1a_u32)))
+uint32_t __arm_vcx1a_u32(int, uint32_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_vcx1d_u64)))
+uint64_t __arm_vcx1d_u64(int, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_vcx1da_u64)))
+uint64_t __arm_vcx1da_u64(int, uint64_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_vcx2_u32)))
+uint32_t __arm_vcx2_u32(int, uint32_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_vcx2a_u32)))
+uint32_t __arm_vcx2a_u32(int, uint32_t, uint32_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_vcx2d_u64)))
+uint64_t __arm_vcx2d_u64(int, uint64_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_vcx2da_u64)))
+uint64_t __arm_vcx2da_u64(int, uint64_t, uint64_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_vcx3_u32)))
+uint32_t __arm_vcx3_u32(int, uint32_t, uint32_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_vcx3a_u32)))
+uint32_t __arm_vcx3a_u32(int, uint32_t, uint32_t, uint32_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_vcx3d_u64)))
+uint64_t __arm_vcx3d_u64(int, uint64_t, uint64_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_vcx3da_u64)))
+uint64_t __arm_vcx3da_u64(int, uint64_t, uint64_t, uint64_t, uint32_t);
+
+#if __ARM_FEATURE_MVE
+
+typedef uint16_t mve_pred16_t;
+typedef __attribute__((__neon_vector_type__(8), __clang_arm_mve_strict_polymorphism)) int16_t int16x8_t;
+typedef __attribute__((__neon_vector_type__(4), __clang_arm_mve_strict_polymorphism)) int32_t int32x4_t;
+typedef __attribute__((__neon_vector_type__(2), __clang_arm_mve_strict_polymorphism)) int64_t int64x2_t;
+typedef __attribute__((__neon_vector_type__(16), __clang_arm_mve_strict_polymorphism)) int8_t int8x16_t;
+typedef __attribute__((__neon_vector_type__(8), __clang_arm_mve_strict_polymorphism)) uint16_t uint16x8_t;
+typedef __attribute__((__neon_vector_type__(4), __clang_arm_mve_strict_polymorphism)) uint32_t uint32x4_t;
+typedef __attribute__((__neon_vector_type__(2), __clang_arm_mve_strict_polymorphism)) uint64_t uint64x2_t;
+typedef __attribute__((__neon_vector_type__(16), __clang_arm_mve_strict_polymorphism)) uint8_t uint8x16_t;
+
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1q_m_s16)))
+int16x8_t __arm_vcx1q_m(int, int16x8_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1q_m_s32)))
+int32x4_t __arm_vcx1q_m(int, int32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1q_m_s64)))
+int64x2_t __arm_vcx1q_m(int, int64x2_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1q_m_s8)))
+int8x16_t __arm_vcx1q_m(int, int8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1q_m_u16)))
+uint16x8_t __arm_vcx1q_m(int, uint16x8_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1q_m_u32)))
+uint32x4_t __arm_vcx1q_m(int, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1q_m_u64)))
+uint64x2_t __arm_vcx1q_m(int, uint64x2_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1q_m_u8)))
+uint8x16_t __arm_vcx1q_m(int, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_cde_vcx1q_u8)))
+uint8x16_t __arm_vcx1q_u8(int, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_m_s16)))
+int16x8_t __arm_vcx1qa_m(int, int16x8_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_m_s32)))
+int32x4_t __arm_vcx1qa_m(int, int32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_m_s64)))
+int64x2_t __arm_vcx1qa_m(int, int64x2_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_m_s8)))
+int8x16_t __arm_vcx1qa_m(int, int8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_m_u16)))
+uint16x8_t __arm_vcx1qa_m(int, uint16x8_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_m_u32)))
+uint32x4_t __arm_vcx1qa_m(int, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_m_u64)))
+uint64x2_t __arm_vcx1qa_m(int, uint64x2_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_m_u8)))
+uint8x16_t __arm_vcx1qa_m(int, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_s16)))
+int16x8_t __arm_vcx1qa(int, int16x8_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_s32)))
+int32x4_t __arm_vcx1qa(int, int32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_s64)))
+int64x2_t __arm_vcx1qa(int, int64x2_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_s8)))
+int8x16_t __arm_vcx1qa(int, int8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_u16)))
+uint16x8_t __arm_vcx1qa(int, uint16x8_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_u32)))
+uint32x4_t __arm_vcx1qa(int, uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_u64)))
+uint64x2_t __arm_vcx1qa(int, uint64x2_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_u8)))
+uint8x16_t __arm_vcx1qa(int, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_m_impl_s16)))
+int16x8_t __arm_vcx2q_m_impl(int, int16x8_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_m_impl_s32)))
+int32x4_t __arm_vcx2q_m_impl(int, int32x4_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_m_impl_s64)))
+int64x2_t __arm_vcx2q_m_impl(int, int64x2_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_m_impl_s8)))
+int8x16_t __arm_vcx2q_m_impl(int, int8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_m_impl_u16)))
+uint16x8_t __arm_vcx2q_m_impl(int, uint16x8_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_m_impl_u32)))
+uint32x4_t __arm_vcx2q_m_impl(int, uint32x4_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_m_impl_u64)))
+uint64x2_t __arm_vcx2q_m_impl(int, uint64x2_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_m_impl_u8)))
+uint8x16_t __arm_vcx2q_m_impl(int, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_s16)))
+int16x8_t __arm_vcx2q(int, int16x8_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_s32)))
+int32x4_t __arm_vcx2q(int, int32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_s64)))
+int64x2_t __arm_vcx2q(int, int64x2_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_s8)))
+int8x16_t __arm_vcx2q(int, int8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_u16)))
+uint16x8_t __arm_vcx2q(int, uint16x8_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_u32)))
+uint32x4_t __arm_vcx2q(int, uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_u64)))
+uint64x2_t __arm_vcx2q(int, uint64x2_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_u8)))
+uint8x16_t __arm_vcx2q(int, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_u8_s16)))
+uint8x16_t __arm_vcx2q_u8(int, int16x8_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_u8_s32)))
+uint8x16_t __arm_vcx2q_u8(int, int32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_u8_s64)))
+uint8x16_t __arm_vcx2q_u8(int, int64x2_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_u8_s8)))
+uint8x16_t __arm_vcx2q_u8(int, int8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_u8_u16)))
+uint8x16_t __arm_vcx2q_u8(int, uint16x8_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_u8_u32)))
+uint8x16_t __arm_vcx2q_u8(int, uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_u8_u64)))
+uint8x16_t __arm_vcx2q_u8(int, uint64x2_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_u8_u8)))
+uint8x16_t __arm_vcx2q_u8(int, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_impl_s16)))
+int16x8_t __arm_vcx2qa_impl(int, int16x8_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_impl_s32)))
+int32x4_t __arm_vcx2qa_impl(int, int32x4_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_impl_s64)))
+int64x2_t __arm_vcx2qa_impl(int, int64x2_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_impl_s8)))
+int8x16_t __arm_vcx2qa_impl(int, int8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_impl_u16)))
+uint16x8_t __arm_vcx2qa_impl(int, uint16x8_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_impl_u32)))
+uint32x4_t __arm_vcx2qa_impl(int, uint32x4_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_impl_u64)))
+uint64x2_t __arm_vcx2qa_impl(int, uint64x2_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_impl_u8)))
+uint8x16_t __arm_vcx2qa_impl(int, uint8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_m_impl_s16)))
+int16x8_t __arm_vcx2qa_m_impl(int, int16x8_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_m_impl_s32)))
+int32x4_t __arm_vcx2qa_m_impl(int, int32x4_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_m_impl_s64)))
+int64x2_t __arm_vcx2qa_m_impl(int, int64x2_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_m_impl_s8)))
+int8x16_t __arm_vcx2qa_m_impl(int, int8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_m_impl_u16)))
+uint16x8_t __arm_vcx2qa_m_impl(int, uint16x8_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_m_impl_u32)))
+uint32x4_t __arm_vcx2qa_m_impl(int, uint32x4_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_m_impl_u64)))
+uint64x2_t __arm_vcx2qa_m_impl(int, uint64x2_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_m_impl_u8)))
+uint8x16_t __arm_vcx2qa_m_impl(int, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_impl_s16)))
+int16x8_t __arm_vcx3q_impl(int, int16x8_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_impl_s32)))
+int32x4_t __arm_vcx3q_impl(int, int32x4_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_impl_s64)))
+int64x2_t __arm_vcx3q_impl(int, int64x2_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_impl_s8)))
+int8x16_t __arm_vcx3q_impl(int, int8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_impl_u16)))
+uint16x8_t __arm_vcx3q_impl(int, uint16x8_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_impl_u32)))
+uint32x4_t __arm_vcx3q_impl(int, uint32x4_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_impl_u64)))
+uint64x2_t __arm_vcx3q_impl(int, uint64x2_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_impl_u8)))
+uint8x16_t __arm_vcx3q_impl(int, uint8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_m_impl_s16)))
+int16x8_t __arm_vcx3q_m_impl(int, int16x8_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_m_impl_s32)))
+int32x4_t __arm_vcx3q_m_impl(int, int32x4_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_m_impl_s64)))
+int64x2_t __arm_vcx3q_m_impl(int, int64x2_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_m_impl_s8)))
+int8x16_t __arm_vcx3q_m_impl(int, int8x16_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_m_impl_u16)))
+uint16x8_t __arm_vcx3q_m_impl(int, uint16x8_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_m_impl_u32)))
+uint32x4_t __arm_vcx3q_m_impl(int, uint32x4_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_m_impl_u64)))
+uint64x2_t __arm_vcx3q_m_impl(int, uint64x2_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_m_impl_u8)))
+uint8x16_t __arm_vcx3q_m_impl(int, uint8x16_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_u8_impl_s16)))
+uint8x16_t __arm_vcx3q_u8_impl(int, int16x8_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_u8_impl_s32)))
+uint8x16_t __arm_vcx3q_u8_impl(int, int32x4_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_u8_impl_s64)))
+uint8x16_t __arm_vcx3q_u8_impl(int, int64x2_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_u8_impl_s8)))
+uint8x16_t __arm_vcx3q_u8_impl(int, int8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_u8_impl_u16)))
+uint8x16_t __arm_vcx3q_u8_impl(int, uint16x8_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_u8_impl_u32)))
+uint8x16_t __arm_vcx3q_u8_impl(int, uint32x4_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_u8_impl_u64)))
+uint8x16_t __arm_vcx3q_u8_impl(int, uint64x2_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_u8_impl_u8)))
+uint8x16_t __arm_vcx3q_u8_impl(int, uint8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_impl_s16)))
+int16x8_t __arm_vcx3qa_impl(int, int16x8_t, uint8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_impl_s32)))
+int32x4_t __arm_vcx3qa_impl(int, int32x4_t, uint8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_impl_s64)))
+int64x2_t __arm_vcx3qa_impl(int, int64x2_t, uint8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_impl_s8)))
+int8x16_t __arm_vcx3qa_impl(int, int8x16_t, uint8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_impl_u16)))
+uint16x8_t __arm_vcx3qa_impl(int, uint16x8_t, uint8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_impl_u32)))
+uint32x4_t __arm_vcx3qa_impl(int, uint32x4_t, uint8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_impl_u64)))
+uint64x2_t __arm_vcx3qa_impl(int, uint64x2_t, uint8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_impl_u8)))
+uint8x16_t __arm_vcx3qa_impl(int, uint8x16_t, uint8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_m_impl_s16)))
+int16x8_t __arm_vcx3qa_m_impl(int, int16x8_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_m_impl_s32)))
+int32x4_t __arm_vcx3qa_m_impl(int, int32x4_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_m_impl_s64)))
+int64x2_t __arm_vcx3qa_m_impl(int, int64x2_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_m_impl_s8)))
+int8x16_t __arm_vcx3qa_m_impl(int, int8x16_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_m_impl_u16)))
+uint16x8_t __arm_vcx3qa_m_impl(int, uint16x8_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_m_impl_u32)))
+uint32x4_t __arm_vcx3qa_m_impl(int, uint32x4_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_m_impl_u64)))
+uint64x2_t __arm_vcx3qa_m_impl(int, uint64x2_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_m_impl_u8)))
+uint8x16_t __arm_vcx3qa_m_impl(int, uint8x16_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u8)))
+int16x8_t __arm_vreinterpretq_s16_u8(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u8)))
+int32x4_t __arm_vreinterpretq_s32_u8(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u8)))
+int64x2_t __arm_vreinterpretq_s64_u8(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u8)))
+int8x16_t __arm_vreinterpretq_s8_u8(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_u8)))
+uint16x8_t __arm_vreinterpretq_u16_u8(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_u8)))
+uint32x4_t __arm_vreinterpretq_u32_u8(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_u8)))
+uint64x2_t __arm_vreinterpretq_u64_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s16)))
+uint8x16_t __arm_vreinterpretq_u8(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s32)))
+uint8x16_t __arm_vreinterpretq_u8(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s64)))
+uint8x16_t __arm_vreinterpretq_u8(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s8)))
+uint8x16_t __arm_vreinterpretq_u8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_u16)))
+uint8x16_t __arm_vreinterpretq_u8(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_u32)))
+uint8x16_t __arm_vreinterpretq_u8(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_u64)))
+uint8x16_t __arm_vreinterpretq_u8(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vreinterpretq_u8_u8)))
+uint8x16_t __arm_vreinterpretq_u8(uint8x16_t);
+#define __arm_vcx2q_m(cp, inactive, n, imm, pred) __arm_vcx2q_m_impl((cp), (inactive), __arm_vreinterpretq_u8(n), (imm), (pred))
+#define __arm_vcx2qa(cp, acc, n, imm) __arm_vcx2qa_impl((cp), (acc), __arm_vreinterpretq_u8(n), (imm))
+#define __arm_vcx2qa_m(cp, acc, n, imm, pred) __arm_vcx2qa_m_impl((cp), (acc), __arm_vreinterpretq_u8(n), (imm), (pred))
+#define __arm_vcx3q(cp, n, m, imm) __arm_vcx3q_impl((cp), (n), __arm_vreinterpretq_u8(m), (imm))
+#define __arm_vcx3q_m(cp, inactive, n, m, imm, pred) __arm_vcx3q_m_impl((cp), (inactive), __arm_vreinterpretq_u8(n), __arm_vreinterpretq_u8(m), (imm), (pred))
+#define __arm_vcx3q_u8(cp, n, m, imm) __arm_vcx3q_u8_impl((cp), (n), __arm_vreinterpretq_u8(m), (imm))
+#define __arm_vcx3qa(cp, acc, n, m, imm) __arm_vcx3qa_impl((cp), (acc), __arm_vreinterpretq_u8(n), __arm_vreinterpretq_u8(m), (imm))
+#define __arm_vcx3qa_m(cp, acc, n, m, imm, pred) __arm_vcx3qa_m_impl((cp), (acc), __arm_vreinterpretq_u8(n), __arm_vreinterpretq_u8(m), (imm), (pred))
+
+#endif /* __ARM_FEATURE_MVE */
+
+#if __ARM_FEATURE_MVE & 2
+
+typedef __fp16 float16_t;
+typedef float float32_t;
+typedef __attribute__((__neon_vector_type__(8), __clang_arm_mve_strict_polymorphism)) float16_t float16x8_t;
+typedef __attribute__((__neon_vector_type__(4), __clang_arm_mve_strict_polymorphism)) float32_t float32x4_t;
+
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1q_m_f16)))
+float16x8_t __arm_vcx1q_m(int, float16x8_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1q_m_f32)))
+float32x4_t __arm_vcx1q_m(int, float32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_f16)))
+float16x8_t __arm_vcx1qa(int, float16x8_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_f32)))
+float32x4_t __arm_vcx1qa(int, float32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_m_f16)))
+float16x8_t __arm_vcx1qa_m(int, float16x8_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx1qa_m_f32)))
+float32x4_t __arm_vcx1qa_m(int, float32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_f16)))
+float16x8_t __arm_vcx2q(int, float16x8_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_f32)))
+float32x4_t __arm_vcx2q(int, float32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_m_impl_f16)))
+float16x8_t __arm_vcx2q_m_impl(int, float16x8_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_m_impl_f32)))
+float32x4_t __arm_vcx2q_m_impl(int, float32x4_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_u8_f16)))
+uint8x16_t __arm_vcx2q_u8(int, float16x8_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2q_u8_f32)))
+uint8x16_t __arm_vcx2q_u8(int, float32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_impl_f16)))
+float16x8_t __arm_vcx2qa_impl(int, float16x8_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_impl_f32)))
+float32x4_t __arm_vcx2qa_impl(int, float32x4_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_m_impl_f16)))
+float16x8_t __arm_vcx2qa_m_impl(int, float16x8_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx2qa_m_impl_f32)))
+float32x4_t __arm_vcx2qa_m_impl(int, float32x4_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_impl_f16)))
+float16x8_t __arm_vcx3q_impl(int, float16x8_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_impl_f32)))
+float32x4_t __arm_vcx3q_impl(int, float32x4_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_m_impl_f16)))
+float16x8_t __arm_vcx3q_m_impl(int, float16x8_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_m_impl_f32)))
+float32x4_t __arm_vcx3q_m_impl(int, float32x4_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_u8_impl_f16)))
+uint8x16_t __arm_vcx3q_u8_impl(int, float16x8_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3q_u8_impl_f32)))
+uint8x16_t __arm_vcx3q_u8_impl(int, float32x4_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_impl_f16)))
+float16x8_t __arm_vcx3qa_impl(int, float16x8_t, uint8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_impl_f32)))
+float32x4_t __arm_vcx3qa_impl(int, float32x4_t, uint8x16_t, uint8x16_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_m_impl_f16)))
+float16x8_t __arm_vcx3qa_m_impl(int, float16x8_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_cde_vcx3qa_m_impl_f32)))
+float32x4_t __arm_vcx3qa_m_impl(int, float32x4_t, uint8x16_t, uint8x16_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u8)))
+float16x8_t __arm_vreinterpretq_f16_u8(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u8)))
+float32x4_t __arm_vreinterpretq_f32_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_f16)))
+uint8x16_t __arm_vreinterpretq_u8(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_f32)))
+uint8x16_t __arm_vreinterpretq_u8(float32x4_t);
+
+#endif /* __ARM_FEATURE_MVE & 2 */
+
+#ifdef __cplusplus
+} /* extern "C" */
+#endif
+
+#endif /* __ARM_CDE_H */
--- /dev/null
+//===---- arm_cmse.h - Arm CMSE support -----------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef __ARM_CMSE_H
+#define __ARM_CMSE_H
+
+#if (__ARM_FEATURE_CMSE & 0x1)
+#include <stddef.h>
+#include <stdint.h>
+
+#define __ARM_CMSE_SECURE_MODE (__ARM_FEATURE_CMSE & 0x2)
+#define CMSE_MPU_READWRITE 1 /* checks if readwrite_ok field is set */
+#define CMSE_AU_NONSECURE 2 /* checks if permissions have secure field unset */
+#define CMSE_MPU_UNPRIV 4 /* sets T flag on TT insrtuction */
+#define CMSE_MPU_READ 8 /* checks if read_ok field is set */
+#define CMSE_MPU_NONSECURE 16 /* sets A flag, checks if secure field unset */
+#define CMSE_NONSECURE (CMSE_AU_NONSECURE | CMSE_MPU_NONSECURE)
+
+#define cmse_check_pointed_object(p, f) \
+ cmse_check_address_range((p), sizeof(*(p)), (f))
+
+#if defined(__cplusplus)
+extern "C" {
+#endif
+
+typedef union {
+ struct cmse_address_info {
+#ifdef __ARM_BIG_ENDIAN
+ /* __ARM_BIG_ENDIAN */
+#if (__ARM_CMSE_SECURE_MODE)
+ unsigned idau_region : 8;
+ unsigned idau_region_valid : 1;
+ unsigned secure : 1;
+ unsigned nonsecure_readwrite_ok : 1;
+ unsigned nonsecure_read_ok : 1;
+#else
+ unsigned : 12;
+#endif
+ unsigned readwrite_ok : 1;
+ unsigned read_ok : 1;
+#if (__ARM_CMSE_SECURE_MODE)
+ unsigned sau_region_valid : 1;
+#else
+ unsigned : 1;
+#endif
+ unsigned mpu_region_valid : 1;
+#if (__ARM_CMSE_SECURE_MODE)
+ unsigned sau_region : 8;
+#else
+ unsigned : 8;
+#endif
+ unsigned mpu_region : 8;
+
+#else /* __ARM_LITTLE_ENDIAN */
+ unsigned mpu_region : 8;
+#if (__ARM_CMSE_SECURE_MODE)
+ unsigned sau_region : 8;
+#else
+ unsigned : 8;
+#endif
+ unsigned mpu_region_valid : 1;
+#if (__ARM_CMSE_SECURE_MODE)
+ unsigned sau_region_valid : 1;
+#else
+ unsigned : 1;
+#endif
+ unsigned read_ok : 1;
+ unsigned readwrite_ok : 1;
+#if (__ARM_CMSE_SECURE_MODE)
+ unsigned nonsecure_read_ok : 1;
+ unsigned nonsecure_readwrite_ok : 1;
+ unsigned secure : 1;
+ unsigned idau_region_valid : 1;
+ unsigned idau_region : 8;
+#else
+ unsigned : 12;
+#endif
+#endif /*__ARM_LITTLE_ENDIAN */
+ } flags;
+ unsigned value;
+} cmse_address_info_t;
+
+static cmse_address_info_t __attribute__((__always_inline__, __nodebug__))
+cmse_TT(void *__p) {
+ cmse_address_info_t __u;
+ __u.value = __builtin_arm_cmse_TT(__p);
+ return __u;
+}
+static cmse_address_info_t __attribute__((__always_inline__, __nodebug__))
+cmse_TTT(void *__p) {
+ cmse_address_info_t __u;
+ __u.value = __builtin_arm_cmse_TTT(__p);
+ return __u;
+}
+
+#if __ARM_CMSE_SECURE_MODE
+static cmse_address_info_t __attribute__((__always_inline__, __nodebug__))
+cmse_TTA(void *__p) {
+ cmse_address_info_t __u;
+ __u.value = __builtin_arm_cmse_TTA(__p);
+ return __u;
+}
+static cmse_address_info_t __attribute__((__always_inline__, __nodebug__))
+cmse_TTAT(void *__p) {
+ cmse_address_info_t __u;
+ __u.value = __builtin_arm_cmse_TTAT(__p);
+ return __u;
+}
+#endif
+
+#define cmse_TT_fptr(p) cmse_TT(__builtin_bit_cast(void *, (p)))
+#define cmse_TTT_fptr(p) cmse_TTT(__builtin_bit_cast(void *, (p)))
+
+#if __ARM_CMSE_SECURE_MODE
+#define cmse_TTA_fptr(p) cmse_TTA(__builtin_bit_cast(void *, (p)))
+#define cmse_TTAT_fptr(p) cmse_TTAT(__builtin_bit_cast(void *, (p)))
+#endif
+
+static void *__attribute__((__always_inline__))
+cmse_check_address_range(void *__pb, size_t __s, int __flags) {
+ uintptr_t __begin = (uintptr_t)__pb;
+ uintptr_t __end = __begin + __s - 1;
+
+ if (__end < __begin)
+ return NULL; /* wrap around check */
+
+ /* Check whether the range crosses a 32-bytes aligned address */
+ const int __single_check = (__begin ^ __end) < 0x20u;
+
+ /* execute the right variant of the TT instructions */
+ void *__pe = (void *)__end;
+ cmse_address_info_t __permb, __perme;
+ switch (__flags & (CMSE_MPU_UNPRIV | CMSE_MPU_NONSECURE)) {
+ case 0:
+ __permb = cmse_TT(__pb);
+ __perme = __single_check ? __permb : cmse_TT(__pe);
+ break;
+ case CMSE_MPU_UNPRIV:
+ __permb = cmse_TTT(__pb);
+ __perme = __single_check ? __permb : cmse_TTT(__pe);
+ break;
+#if __ARM_CMSE_SECURE_MODE
+ case CMSE_MPU_NONSECURE:
+ __permb = cmse_TTA(__pb);
+ __perme = __single_check ? __permb : cmse_TTA(__pe);
+ break;
+ case CMSE_MPU_UNPRIV | CMSE_MPU_NONSECURE:
+ __permb = cmse_TTAT(__pb);
+ __perme = __single_check ? __permb : cmse_TTAT(__pe);
+ break;
+#endif
+ /* if CMSE_NONSECURE is specified w/o __ARM_CMSE_SECURE_MODE */
+ default:
+ return NULL;
+ }
+
+ /* check that the range does not cross MPU, SAU, or IDAU region boundaries */
+ if (__permb.value != __perme.value)
+ return NULL;
+#if !(__ARM_CMSE_SECURE_MODE)
+ /* CMSE_AU_NONSECURE is only supported when __ARM_FEATURE_CMSE & 0x2 */
+ if (__flags & CMSE_AU_NONSECURE)
+ return NULL;
+#endif
+
+ /* check the permission on the range */
+ switch (__flags & ~(CMSE_MPU_UNPRIV | CMSE_MPU_NONSECURE)) {
+#if (__ARM_CMSE_SECURE_MODE)
+ case CMSE_MPU_READ | CMSE_MPU_READWRITE | CMSE_AU_NONSECURE:
+ case CMSE_MPU_READWRITE | CMSE_AU_NONSECURE:
+ return __permb.flags.nonsecure_readwrite_ok ? __pb : NULL;
+
+ case CMSE_MPU_READ | CMSE_AU_NONSECURE:
+ return __permb.flags.nonsecure_read_ok ? __pb : NULL;
+
+ case CMSE_AU_NONSECURE:
+ return __permb.flags.secure ? NULL : __pb;
+#endif
+ case CMSE_MPU_READ | CMSE_MPU_READWRITE:
+ case CMSE_MPU_READWRITE:
+ return __permb.flags.readwrite_ok ? __pb : NULL;
+
+ case CMSE_MPU_READ:
+ return __permb.flags.read_ok ? __pb : NULL;
+
+ default:
+ return NULL;
+ }
+}
+
+#if __ARM_CMSE_SECURE_MODE
+static int __attribute__((__always_inline__, __nodebug__))
+cmse_nonsecure_caller(void) {
+ return !((uintptr_t)__builtin_return_address(0) & 1);
+}
+
+#define cmse_nsfptr_create(p) \
+ __builtin_bit_cast(__typeof__(p), \
+ (__builtin_bit_cast(uintptr_t, p) & ~(uintptr_t)1))
+
+#define cmse_is_nsfptr(p) ((__builtin_bit_cast(uintptr_t, p) & 1) == 0)
+
+#endif /* __ARM_CMSE_SECURE_MODE */
+
+void __attribute__((__noreturn__)) cmse_abort(void);
+#if defined(__cplusplus)
+}
+#endif
+
+#endif /* (__ARM_FEATURE_CMSE & 0x1) */
+
+#endif /* __ARM_CMSE_H */
--- /dev/null
+/*===---- arm_fp16.h - ARM FP16 intrinsics ---------------------------------===
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __ARM_FP16_H
+#define __ARM_FP16_H
+
+#include <stdint.h>
+
+typedef __fp16 float16_t;
+#define __ai static __inline__ __attribute__((__always_inline__, __nodebug__))
+
+#if defined(__ARM_FEATURE_FP16_SCALAR_ARITHMETIC) && defined(__aarch64__)
+#define vabdh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vabdh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vabsh_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vabsh_f16(__s0); \
+ __ret; \
+})
+#define vaddh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vaddh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vcageh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcageh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vcagth_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcagth_f16(__s0, __s1); \
+ __ret; \
+})
+#define vcaleh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcaleh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vcalth_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcalth_f16(__s0, __s1); \
+ __ret; \
+})
+#define vceqh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vceqh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vceqzh_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vceqzh_f16(__s0); \
+ __ret; \
+})
+#define vcgeh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcgeh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vcgezh_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcgezh_f16(__s0); \
+ __ret; \
+})
+#define vcgth_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcgth_f16(__s0, __s1); \
+ __ret; \
+})
+#define vcgtzh_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcgtzh_f16(__s0); \
+ __ret; \
+})
+#define vcleh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcleh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vclezh_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vclezh_f16(__s0); \
+ __ret; \
+})
+#define vclth_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vclth_f16(__s0, __s1); \
+ __ret; \
+})
+#define vcltzh_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcltzh_f16(__s0); \
+ __ret; \
+})
+#define vcvth_n_s16_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int16_t __ret; \
+ __ret = (int16_t) __builtin_neon_vcvth_n_s16_f16(__s0, __p1); \
+ __ret; \
+})
+#define vcvth_n_s32_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int32_t __ret; \
+ __ret = (int32_t) __builtin_neon_vcvth_n_s32_f16(__s0, __p1); \
+ __ret; \
+})
+#define vcvth_n_s64_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int64_t __ret; \
+ __ret = (int64_t) __builtin_neon_vcvth_n_s64_f16(__s0, __p1); \
+ __ret; \
+})
+#define vcvth_n_u16_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcvth_n_u16_f16(__s0, __p1); \
+ __ret; \
+})
+#define vcvth_n_u32_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint32_t __ret; \
+ __ret = (uint32_t) __builtin_neon_vcvth_n_u32_f16(__s0, __p1); \
+ __ret; \
+})
+#define vcvth_n_u64_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint64_t __ret; \
+ __ret = (uint64_t) __builtin_neon_vcvth_n_u64_f16(__s0, __p1); \
+ __ret; \
+})
+#define vcvth_s16_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int16_t __ret; \
+ __ret = (int16_t) __builtin_neon_vcvth_s16_f16(__s0); \
+ __ret; \
+})
+#define vcvth_s32_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int32_t __ret; \
+ __ret = (int32_t) __builtin_neon_vcvth_s32_f16(__s0); \
+ __ret; \
+})
+#define vcvth_s64_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int64_t __ret; \
+ __ret = (int64_t) __builtin_neon_vcvth_s64_f16(__s0); \
+ __ret; \
+})
+#define vcvth_u16_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcvth_u16_f16(__s0); \
+ __ret; \
+})
+#define vcvth_u32_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint32_t __ret; \
+ __ret = (uint32_t) __builtin_neon_vcvth_u32_f16(__s0); \
+ __ret; \
+})
+#define vcvth_u64_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint64_t __ret; \
+ __ret = (uint64_t) __builtin_neon_vcvth_u64_f16(__s0); \
+ __ret; \
+})
+#define vcvtah_s16_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int16_t __ret; \
+ __ret = (int16_t) __builtin_neon_vcvtah_s16_f16(__s0); \
+ __ret; \
+})
+#define vcvtah_s32_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int32_t __ret; \
+ __ret = (int32_t) __builtin_neon_vcvtah_s32_f16(__s0); \
+ __ret; \
+})
+#define vcvtah_s64_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int64_t __ret; \
+ __ret = (int64_t) __builtin_neon_vcvtah_s64_f16(__s0); \
+ __ret; \
+})
+#define vcvtah_u16_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcvtah_u16_f16(__s0); \
+ __ret; \
+})
+#define vcvtah_u32_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint32_t __ret; \
+ __ret = (uint32_t) __builtin_neon_vcvtah_u32_f16(__s0); \
+ __ret; \
+})
+#define vcvtah_u64_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint64_t __ret; \
+ __ret = (uint64_t) __builtin_neon_vcvtah_u64_f16(__s0); \
+ __ret; \
+})
+#define vcvth_f16_u16(__p0) __extension__ ({ \
+ uint16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vcvth_f16_u16(__s0); \
+ __ret; \
+})
+#define vcvth_f16_s16(__p0) __extension__ ({ \
+ int16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vcvth_f16_s16(__s0); \
+ __ret; \
+})
+#define vcvth_f16_u32(__p0) __extension__ ({ \
+ uint32_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vcvth_f16_u32(__s0); \
+ __ret; \
+})
+#define vcvth_f16_s32(__p0) __extension__ ({ \
+ int32_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vcvth_f16_s32(__s0); \
+ __ret; \
+})
+#define vcvth_f16_u64(__p0) __extension__ ({ \
+ uint64_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vcvth_f16_u64(__s0); \
+ __ret; \
+})
+#define vcvth_f16_s64(__p0) __extension__ ({ \
+ int64_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vcvth_f16_s64(__s0); \
+ __ret; \
+})
+#define vcvth_n_f16_u32(__p0, __p1) __extension__ ({ \
+ uint32_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vcvth_n_f16_u32(__s0, __p1); \
+ __ret; \
+})
+#define vcvth_n_f16_s32(__p0, __p1) __extension__ ({ \
+ int32_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vcvth_n_f16_s32(__s0, __p1); \
+ __ret; \
+})
+#define vcvth_n_f16_u64(__p0, __p1) __extension__ ({ \
+ uint64_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vcvth_n_f16_u64(__s0, __p1); \
+ __ret; \
+})
+#define vcvth_n_f16_s64(__p0, __p1) __extension__ ({ \
+ int64_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vcvth_n_f16_s64(__s0, __p1); \
+ __ret; \
+})
+#define vcvth_n_f16_u16(__p0, __p1) __extension__ ({ \
+ uint16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vcvth_n_f16_u16(__s0, __p1); \
+ __ret; \
+})
+#define vcvth_n_f16_s16(__p0, __p1) __extension__ ({ \
+ int16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vcvth_n_f16_s16(__s0, __p1); \
+ __ret; \
+})
+#define vcvtmh_s16_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int16_t __ret; \
+ __ret = (int16_t) __builtin_neon_vcvtmh_s16_f16(__s0); \
+ __ret; \
+})
+#define vcvtmh_s32_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int32_t __ret; \
+ __ret = (int32_t) __builtin_neon_vcvtmh_s32_f16(__s0); \
+ __ret; \
+})
+#define vcvtmh_s64_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int64_t __ret; \
+ __ret = (int64_t) __builtin_neon_vcvtmh_s64_f16(__s0); \
+ __ret; \
+})
+#define vcvtmh_u16_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcvtmh_u16_f16(__s0); \
+ __ret; \
+})
+#define vcvtmh_u32_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint32_t __ret; \
+ __ret = (uint32_t) __builtin_neon_vcvtmh_u32_f16(__s0); \
+ __ret; \
+})
+#define vcvtmh_u64_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint64_t __ret; \
+ __ret = (uint64_t) __builtin_neon_vcvtmh_u64_f16(__s0); \
+ __ret; \
+})
+#define vcvtnh_s16_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int16_t __ret; \
+ __ret = (int16_t) __builtin_neon_vcvtnh_s16_f16(__s0); \
+ __ret; \
+})
+#define vcvtnh_s32_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int32_t __ret; \
+ __ret = (int32_t) __builtin_neon_vcvtnh_s32_f16(__s0); \
+ __ret; \
+})
+#define vcvtnh_s64_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int64_t __ret; \
+ __ret = (int64_t) __builtin_neon_vcvtnh_s64_f16(__s0); \
+ __ret; \
+})
+#define vcvtnh_u16_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcvtnh_u16_f16(__s0); \
+ __ret; \
+})
+#define vcvtnh_u32_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint32_t __ret; \
+ __ret = (uint32_t) __builtin_neon_vcvtnh_u32_f16(__s0); \
+ __ret; \
+})
+#define vcvtnh_u64_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint64_t __ret; \
+ __ret = (uint64_t) __builtin_neon_vcvtnh_u64_f16(__s0); \
+ __ret; \
+})
+#define vcvtph_s16_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int16_t __ret; \
+ __ret = (int16_t) __builtin_neon_vcvtph_s16_f16(__s0); \
+ __ret; \
+})
+#define vcvtph_s32_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int32_t __ret; \
+ __ret = (int32_t) __builtin_neon_vcvtph_s32_f16(__s0); \
+ __ret; \
+})
+#define vcvtph_s64_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ int64_t __ret; \
+ __ret = (int64_t) __builtin_neon_vcvtph_s64_f16(__s0); \
+ __ret; \
+})
+#define vcvtph_u16_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint16_t __ret; \
+ __ret = (uint16_t) __builtin_neon_vcvtph_u16_f16(__s0); \
+ __ret; \
+})
+#define vcvtph_u32_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint32_t __ret; \
+ __ret = (uint32_t) __builtin_neon_vcvtph_u32_f16(__s0); \
+ __ret; \
+})
+#define vcvtph_u64_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ uint64_t __ret; \
+ __ret = (uint64_t) __builtin_neon_vcvtph_u64_f16(__s0); \
+ __ret; \
+})
+#define vdivh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vdivh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vfmah_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ float16_t __s2 = __p2; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vfmah_f16(__s0, __s1, __s2); \
+ __ret; \
+})
+#define vfmsh_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ float16_t __s2 = __p2; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vfmsh_f16(__s0, __s1, __s2); \
+ __ret; \
+})
+#define vmaxh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vmaxh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vmaxnmh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vmaxnmh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vminh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vminh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vminnmh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vminnmh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vmulh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vmulh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vmulxh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vmulxh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vnegh_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vnegh_f16(__s0); \
+ __ret; \
+})
+#define vrecpeh_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vrecpeh_f16(__s0); \
+ __ret; \
+})
+#define vrecpsh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vrecpsh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vrecpxh_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vrecpxh_f16(__s0); \
+ __ret; \
+})
+#define vrndh_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vrndh_f16(__s0); \
+ __ret; \
+})
+#define vrndah_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vrndah_f16(__s0); \
+ __ret; \
+})
+#define vrndih_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vrndih_f16(__s0); \
+ __ret; \
+})
+#define vrndmh_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vrndmh_f16(__s0); \
+ __ret; \
+})
+#define vrndnh_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vrndnh_f16(__s0); \
+ __ret; \
+})
+#define vrndph_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vrndph_f16(__s0); \
+ __ret; \
+})
+#define vrndxh_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vrndxh_f16(__s0); \
+ __ret; \
+})
+#define vrsqrteh_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vrsqrteh_f16(__s0); \
+ __ret; \
+})
+#define vrsqrtsh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vrsqrtsh_f16(__s0, __s1); \
+ __ret; \
+})
+#define vsqrth_f16(__p0) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vsqrth_f16(__s0); \
+ __ret; \
+})
+#define vsubh_f16(__p0, __p1) __extension__ ({ \
+ float16_t __s0 = __p0; \
+ float16_t __s1 = __p1; \
+ float16_t __ret; \
+ __ret = (float16_t) __builtin_neon_vsubh_f16(__s0, __s1); \
+ __ret; \
+})
+#endif
+
+#undef __ai
+
+#endif /* __ARM_FP16_H */
--- /dev/null
+/*===---- arm_mve.h - ARM MVE intrinsics -----------------------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __ARM_MVE_H
+#define __ARM_MVE_H
+
+#if !__ARM_FEATURE_MVE
+#error "MVE support not enabled"
+#endif
+
+#include <stdint.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+typedef uint16_t mve_pred16_t;
+typedef __attribute__((__neon_vector_type__(8), __clang_arm_mve_strict_polymorphism)) int16_t int16x8_t;
+typedef struct { int16x8_t val[2]; } int16x8x2_t;
+typedef struct { int16x8_t val[4]; } int16x8x4_t;
+typedef __attribute__((__neon_vector_type__(4), __clang_arm_mve_strict_polymorphism)) int32_t int32x4_t;
+typedef struct { int32x4_t val[2]; } int32x4x2_t;
+typedef struct { int32x4_t val[4]; } int32x4x4_t;
+typedef __attribute__((__neon_vector_type__(2), __clang_arm_mve_strict_polymorphism)) int64_t int64x2_t;
+typedef struct { int64x2_t val[2]; } int64x2x2_t;
+typedef struct { int64x2_t val[4]; } int64x2x4_t;
+typedef __attribute__((__neon_vector_type__(16), __clang_arm_mve_strict_polymorphism)) int8_t int8x16_t;
+typedef struct { int8x16_t val[2]; } int8x16x2_t;
+typedef struct { int8x16_t val[4]; } int8x16x4_t;
+typedef __attribute__((__neon_vector_type__(8), __clang_arm_mve_strict_polymorphism)) uint16_t uint16x8_t;
+typedef struct { uint16x8_t val[2]; } uint16x8x2_t;
+typedef struct { uint16x8_t val[4]; } uint16x8x4_t;
+typedef __attribute__((__neon_vector_type__(4), __clang_arm_mve_strict_polymorphism)) uint32_t uint32x4_t;
+typedef struct { uint32x4_t val[2]; } uint32x4x2_t;
+typedef struct { uint32x4_t val[4]; } uint32x4x4_t;
+typedef __attribute__((__neon_vector_type__(2), __clang_arm_mve_strict_polymorphism)) uint64_t uint64x2_t;
+typedef struct { uint64x2_t val[2]; } uint64x2x2_t;
+typedef struct { uint64x2_t val[4]; } uint64x2x4_t;
+typedef __attribute__((__neon_vector_type__(16), __clang_arm_mve_strict_polymorphism)) uint8_t uint8x16_t;
+typedef struct { uint8x16_t val[2]; } uint8x16x2_t;
+typedef struct { uint8x16_t val[4]; } uint8x16x4_t;
+
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_asrl)))
+int64_t __arm_asrl(int64_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_lsll)))
+uint64_t __arm_lsll(uint64_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_sqrshr)))
+int32_t __arm_sqrshr(int32_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_sqrshrl)))
+int64_t __arm_sqrshrl(int64_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_sqrshrl_sat48)))
+int64_t __arm_sqrshrl_sat48(int64_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_sqshl)))
+int32_t __arm_sqshl(int32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_sqshll)))
+int64_t __arm_sqshll(int64_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_srshr)))
+int32_t __arm_srshr(int32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_srshrl)))
+int64_t __arm_srshrl(int64_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_uqrshl)))
+uint32_t __arm_uqrshl(uint32_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_uqrshll)))
+uint64_t __arm_uqrshll(uint64_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_uqrshll_sat48)))
+uint64_t __arm_uqrshll_sat48(uint64_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_uqshl)))
+uint32_t __arm_uqshl(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_uqshll)))
+uint64_t __arm_uqshll(uint64_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_urshr)))
+uint32_t __arm_urshr(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_urshrl)))
+uint64_t __arm_urshrl(uint64_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_s16)))
+uint32_t __arm_vabavq_p_s16(uint32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_s16)))
+uint32_t __arm_vabavq_p(uint32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_s32)))
+uint32_t __arm_vabavq_p_s32(uint32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_s32)))
+uint32_t __arm_vabavq_p(uint32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_s8)))
+uint32_t __arm_vabavq_p_s8(uint32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_s8)))
+uint32_t __arm_vabavq_p(uint32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_u16)))
+uint32_t __arm_vabavq_p_u16(uint32_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_u16)))
+uint32_t __arm_vabavq_p(uint32_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_u32)))
+uint32_t __arm_vabavq_p_u32(uint32_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_u32)))
+uint32_t __arm_vabavq_p(uint32_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_u8)))
+uint32_t __arm_vabavq_p_u8(uint32_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_u8)))
+uint32_t __arm_vabavq_p(uint32_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_s16)))
+uint32_t __arm_vabavq_s16(uint32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_s16)))
+uint32_t __arm_vabavq(uint32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_s32)))
+uint32_t __arm_vabavq_s32(uint32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_s32)))
+uint32_t __arm_vabavq(uint32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_s8)))
+uint32_t __arm_vabavq_s8(uint32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_s8)))
+uint32_t __arm_vabavq(uint32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_u16)))
+uint32_t __arm_vabavq_u16(uint32_t, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_u16)))
+uint32_t __arm_vabavq(uint32_t, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_u32)))
+uint32_t __arm_vabavq_u32(uint32_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_u32)))
+uint32_t __arm_vabavq(uint32_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_u8)))
+uint32_t __arm_vabavq_u8(uint32_t, uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_u8)))
+uint32_t __arm_vabavq(uint32_t, uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_s16)))
+int16x8_t __arm_vabdq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_s16)))
+int16x8_t __arm_vabdq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_s32)))
+int32x4_t __arm_vabdq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_s32)))
+int32x4_t __arm_vabdq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_s8)))
+int8x16_t __arm_vabdq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_s8)))
+int8x16_t __arm_vabdq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_u16)))
+uint16x8_t __arm_vabdq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_u16)))
+uint16x8_t __arm_vabdq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_u32)))
+uint32x4_t __arm_vabdq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_u32)))
+uint32x4_t __arm_vabdq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_u8)))
+uint8x16_t __arm_vabdq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_u8)))
+uint8x16_t __arm_vabdq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_s16)))
+int16x8_t __arm_vabdq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_s16)))
+int16x8_t __arm_vabdq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_s32)))
+int32x4_t __arm_vabdq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_s32)))
+int32x4_t __arm_vabdq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_s8)))
+int8x16_t __arm_vabdq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_s8)))
+int8x16_t __arm_vabdq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_u16)))
+uint16x8_t __arm_vabdq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_u16)))
+uint16x8_t __arm_vabdq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_u32)))
+uint32x4_t __arm_vabdq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_u32)))
+uint32x4_t __arm_vabdq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_u8)))
+uint8x16_t __arm_vabdq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_u8)))
+uint8x16_t __arm_vabdq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_s16)))
+int16x8_t __arm_vabdq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_s16)))
+int16x8_t __arm_vabdq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_s32)))
+int32x4_t __arm_vabdq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_s32)))
+int32x4_t __arm_vabdq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_s8)))
+int8x16_t __arm_vabdq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_s8)))
+int8x16_t __arm_vabdq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_u16)))
+uint16x8_t __arm_vabdq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_u16)))
+uint16x8_t __arm_vabdq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_u32)))
+uint32x4_t __arm_vabdq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_u32)))
+uint32x4_t __arm_vabdq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_u8)))
+uint8x16_t __arm_vabdq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_u8)))
+uint8x16_t __arm_vabdq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_s16)))
+int16x8_t __arm_vabsq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_s16)))
+int16x8_t __arm_vabsq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_s32)))
+int32x4_t __arm_vabsq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_s32)))
+int32x4_t __arm_vabsq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_s8)))
+int8x16_t __arm_vabsq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_s8)))
+int8x16_t __arm_vabsq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_s16)))
+int16x8_t __arm_vabsq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_s16)))
+int16x8_t __arm_vabsq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_s32)))
+int32x4_t __arm_vabsq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_s32)))
+int32x4_t __arm_vabsq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_s8)))
+int8x16_t __arm_vabsq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_s8)))
+int8x16_t __arm_vabsq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_s16)))
+int16x8_t __arm_vabsq_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_s16)))
+int16x8_t __arm_vabsq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_s32)))
+int32x4_t __arm_vabsq_x_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_s32)))
+int32x4_t __arm_vabsq_x(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_s8)))
+int8x16_t __arm_vabsq_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_s8)))
+int8x16_t __arm_vabsq_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadciq_m_s32)))
+int32x4_t __arm_vadciq_m_s32(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadciq_m_s32)))
+int32x4_t __arm_vadciq_m(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadciq_m_u32)))
+uint32x4_t __arm_vadciq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadciq_m_u32)))
+uint32x4_t __arm_vadciq_m(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadciq_s32)))
+int32x4_t __arm_vadciq_s32(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadciq_s32)))
+int32x4_t __arm_vadciq(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadciq_u32)))
+uint32x4_t __arm_vadciq_u32(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadciq_u32)))
+uint32x4_t __arm_vadciq(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadcq_m_s32)))
+int32x4_t __arm_vadcq_m_s32(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadcq_m_s32)))
+int32x4_t __arm_vadcq_m(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadcq_m_u32)))
+uint32x4_t __arm_vadcq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadcq_m_u32)))
+uint32x4_t __arm_vadcq_m(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadcq_s32)))
+int32x4_t __arm_vadcq_s32(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadcq_s32)))
+int32x4_t __arm_vadcq(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadcq_u32)))
+uint32x4_t __arm_vadcq_u32(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadcq_u32)))
+uint32x4_t __arm_vadcq(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_p_s32)))
+int64_t __arm_vaddlvaq_p_s32(int64_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_p_s32)))
+int64_t __arm_vaddlvaq_p(int64_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_p_u32)))
+uint64_t __arm_vaddlvaq_p_u32(uint64_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_p_u32)))
+uint64_t __arm_vaddlvaq_p(uint64_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_s32)))
+int64_t __arm_vaddlvaq_s32(int64_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_s32)))
+int64_t __arm_vaddlvaq(int64_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_u32)))
+uint64_t __arm_vaddlvaq_u32(uint64_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_u32)))
+uint64_t __arm_vaddlvaq(uint64_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_p_s32)))
+int64_t __arm_vaddlvq_p_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_p_s32)))
+int64_t __arm_vaddlvq_p(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_p_u32)))
+uint64_t __arm_vaddlvq_p_u32(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_p_u32)))
+uint64_t __arm_vaddlvq_p(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_s32)))
+int64_t __arm_vaddlvq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_s32)))
+int64_t __arm_vaddlvq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_u32)))
+uint64_t __arm_vaddlvq_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_u32)))
+uint64_t __arm_vaddlvq(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_s16)))
+int16x8_t __arm_vaddq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_s16)))
+int16x8_t __arm_vaddq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_s32)))
+int32x4_t __arm_vaddq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_s32)))
+int32x4_t __arm_vaddq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_s8)))
+int8x16_t __arm_vaddq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_s8)))
+int8x16_t __arm_vaddq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_u16)))
+uint16x8_t __arm_vaddq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_u16)))
+uint16x8_t __arm_vaddq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_u32)))
+uint32x4_t __arm_vaddq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_u32)))
+uint32x4_t __arm_vaddq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_u8)))
+uint8x16_t __arm_vaddq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_u8)))
+uint8x16_t __arm_vaddq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_s16)))
+int16x8_t __arm_vaddq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_s16)))
+int16x8_t __arm_vaddq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_s32)))
+int32x4_t __arm_vaddq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_s32)))
+int32x4_t __arm_vaddq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_s8)))
+int8x16_t __arm_vaddq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_s8)))
+int8x16_t __arm_vaddq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_u16)))
+uint16x8_t __arm_vaddq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_u16)))
+uint16x8_t __arm_vaddq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_u32)))
+uint32x4_t __arm_vaddq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_u32)))
+uint32x4_t __arm_vaddq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_u8)))
+uint8x16_t __arm_vaddq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_u8)))
+uint8x16_t __arm_vaddq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_s16)))
+int16x8_t __arm_vaddq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_s16)))
+int16x8_t __arm_vaddq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_s32)))
+int32x4_t __arm_vaddq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_s32)))
+int32x4_t __arm_vaddq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_s8)))
+int8x16_t __arm_vaddq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_s8)))
+int8x16_t __arm_vaddq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_u16)))
+uint16x8_t __arm_vaddq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_u16)))
+uint16x8_t __arm_vaddq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_u32)))
+uint32x4_t __arm_vaddq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_u32)))
+uint32x4_t __arm_vaddq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_u8)))
+uint8x16_t __arm_vaddq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_u8)))
+uint8x16_t __arm_vaddq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_s16)))
+int16x8_t __arm_vaddq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_s16)))
+int16x8_t __arm_vaddq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_s32)))
+int32x4_t __arm_vaddq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_s32)))
+int32x4_t __arm_vaddq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_s8)))
+int8x16_t __arm_vaddq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_s8)))
+int8x16_t __arm_vaddq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_u16)))
+uint16x8_t __arm_vaddq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_u16)))
+uint16x8_t __arm_vaddq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_u32)))
+uint32x4_t __arm_vaddq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_u32)))
+uint32x4_t __arm_vaddq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_u8)))
+uint8x16_t __arm_vaddq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_u8)))
+uint8x16_t __arm_vaddq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_s16)))
+int16x8_t __arm_vaddq_x_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_s16)))
+int16x8_t __arm_vaddq_x(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_s32)))
+int32x4_t __arm_vaddq_x_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_s32)))
+int32x4_t __arm_vaddq_x(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_s8)))
+int8x16_t __arm_vaddq_x_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_s8)))
+int8x16_t __arm_vaddq_x(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_u16)))
+uint16x8_t __arm_vaddq_x_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_u16)))
+uint16x8_t __arm_vaddq_x(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_u32)))
+uint32x4_t __arm_vaddq_x_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_u32)))
+uint32x4_t __arm_vaddq_x(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_u8)))
+uint8x16_t __arm_vaddq_x_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_u8)))
+uint8x16_t __arm_vaddq_x(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_s16)))
+int16x8_t __arm_vaddq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_s16)))
+int16x8_t __arm_vaddq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_s32)))
+int32x4_t __arm_vaddq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_s32)))
+int32x4_t __arm_vaddq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_s8)))
+int8x16_t __arm_vaddq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_s8)))
+int8x16_t __arm_vaddq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_u16)))
+uint16x8_t __arm_vaddq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_u16)))
+uint16x8_t __arm_vaddq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_u32)))
+uint32x4_t __arm_vaddq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_u32)))
+uint32x4_t __arm_vaddq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_u8)))
+uint8x16_t __arm_vaddq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_u8)))
+uint8x16_t __arm_vaddq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_s16)))
+int32_t __arm_vaddvaq_p_s16(int32_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_s16)))
+int32_t __arm_vaddvaq_p(int32_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_s32)))
+int32_t __arm_vaddvaq_p_s32(int32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_s32)))
+int32_t __arm_vaddvaq_p(int32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_s8)))
+int32_t __arm_vaddvaq_p_s8(int32_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_s8)))
+int32_t __arm_vaddvaq_p(int32_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_u16)))
+uint32_t __arm_vaddvaq_p_u16(uint32_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_u16)))
+uint32_t __arm_vaddvaq_p(uint32_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_u32)))
+uint32_t __arm_vaddvaq_p_u32(uint32_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_u32)))
+uint32_t __arm_vaddvaq_p(uint32_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_u8)))
+uint32_t __arm_vaddvaq_p_u8(uint32_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_u8)))
+uint32_t __arm_vaddvaq_p(uint32_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_s16)))
+int32_t __arm_vaddvaq_s16(int32_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_s16)))
+int32_t __arm_vaddvaq(int32_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_s32)))
+int32_t __arm_vaddvaq_s32(int32_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_s32)))
+int32_t __arm_vaddvaq(int32_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_s8)))
+int32_t __arm_vaddvaq_s8(int32_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_s8)))
+int32_t __arm_vaddvaq(int32_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_u16)))
+uint32_t __arm_vaddvaq_u16(uint32_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_u16)))
+uint32_t __arm_vaddvaq(uint32_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_u32)))
+uint32_t __arm_vaddvaq_u32(uint32_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_u32)))
+uint32_t __arm_vaddvaq(uint32_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_u8)))
+uint32_t __arm_vaddvaq_u8(uint32_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_u8)))
+uint32_t __arm_vaddvaq(uint32_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_s16)))
+int32_t __arm_vaddvq_p_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_s16)))
+int32_t __arm_vaddvq_p(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_s32)))
+int32_t __arm_vaddvq_p_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_s32)))
+int32_t __arm_vaddvq_p(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_s8)))
+int32_t __arm_vaddvq_p_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_s8)))
+int32_t __arm_vaddvq_p(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_u16)))
+uint32_t __arm_vaddvq_p_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_u16)))
+uint32_t __arm_vaddvq_p(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_u32)))
+uint32_t __arm_vaddvq_p_u32(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_u32)))
+uint32_t __arm_vaddvq_p(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_u8)))
+uint32_t __arm_vaddvq_p_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_u8)))
+uint32_t __arm_vaddvq_p(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_s16)))
+int32_t __arm_vaddvq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_s16)))
+int32_t __arm_vaddvq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_s32)))
+int32_t __arm_vaddvq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_s32)))
+int32_t __arm_vaddvq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_s8)))
+int32_t __arm_vaddvq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_s8)))
+int32_t __arm_vaddvq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_u16)))
+uint32_t __arm_vaddvq_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_u16)))
+uint32_t __arm_vaddvq(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_u32)))
+uint32_t __arm_vaddvq_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_u32)))
+uint32_t __arm_vaddvq(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_u8)))
+uint32_t __arm_vaddvq_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_u8)))
+uint32_t __arm_vaddvq(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_s16)))
+int16x8_t __arm_vandq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_s16)))
+int16x8_t __arm_vandq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_s32)))
+int32x4_t __arm_vandq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_s32)))
+int32x4_t __arm_vandq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_s8)))
+int8x16_t __arm_vandq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_s8)))
+int8x16_t __arm_vandq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_u16)))
+uint16x8_t __arm_vandq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_u16)))
+uint16x8_t __arm_vandq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_u32)))
+uint32x4_t __arm_vandq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_u32)))
+uint32x4_t __arm_vandq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_u8)))
+uint8x16_t __arm_vandq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_u8)))
+uint8x16_t __arm_vandq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_s16)))
+int16x8_t __arm_vandq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_s16)))
+int16x8_t __arm_vandq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_s32)))
+int32x4_t __arm_vandq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_s32)))
+int32x4_t __arm_vandq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_s8)))
+int8x16_t __arm_vandq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_s8)))
+int8x16_t __arm_vandq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_u16)))
+uint16x8_t __arm_vandq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_u16)))
+uint16x8_t __arm_vandq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_u32)))
+uint32x4_t __arm_vandq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_u32)))
+uint32x4_t __arm_vandq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_u8)))
+uint8x16_t __arm_vandq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_u8)))
+uint8x16_t __arm_vandq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_s16)))
+int16x8_t __arm_vandq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_s16)))
+int16x8_t __arm_vandq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_s32)))
+int32x4_t __arm_vandq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_s32)))
+int32x4_t __arm_vandq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_s8)))
+int8x16_t __arm_vandq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_s8)))
+int8x16_t __arm_vandq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_u16)))
+uint16x8_t __arm_vandq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_u16)))
+uint16x8_t __arm_vandq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_u32)))
+uint32x4_t __arm_vandq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_u32)))
+uint32x4_t __arm_vandq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_u8)))
+uint8x16_t __arm_vandq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_u8)))
+uint8x16_t __arm_vandq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_s16)))
+int16x8_t __arm_vbicq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_s16)))
+int16x8_t __arm_vbicq_m_n(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_s32)))
+int32x4_t __arm_vbicq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_s32)))
+int32x4_t __arm_vbicq_m_n(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_u16)))
+uint16x8_t __arm_vbicq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_u16)))
+uint16x8_t __arm_vbicq_m_n(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_u32)))
+uint32x4_t __arm_vbicq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_u32)))
+uint32x4_t __arm_vbicq_m_n(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_s16)))
+int16x8_t __arm_vbicq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_s16)))
+int16x8_t __arm_vbicq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_s32)))
+int32x4_t __arm_vbicq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_s32)))
+int32x4_t __arm_vbicq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_s8)))
+int8x16_t __arm_vbicq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_s8)))
+int8x16_t __arm_vbicq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_u16)))
+uint16x8_t __arm_vbicq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_u16)))
+uint16x8_t __arm_vbicq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_u32)))
+uint32x4_t __arm_vbicq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_u32)))
+uint32x4_t __arm_vbicq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_u8)))
+uint8x16_t __arm_vbicq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_u8)))
+uint8x16_t __arm_vbicq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_s16)))
+int16x8_t __arm_vbicq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_s16)))
+int16x8_t __arm_vbicq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_s32)))
+int32x4_t __arm_vbicq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_s32)))
+int32x4_t __arm_vbicq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_u16)))
+uint16x8_t __arm_vbicq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_u16)))
+uint16x8_t __arm_vbicq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_u32)))
+uint32x4_t __arm_vbicq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_u32)))
+uint32x4_t __arm_vbicq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_s16)))
+int16x8_t __arm_vbicq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_s16)))
+int16x8_t __arm_vbicq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_s32)))
+int32x4_t __arm_vbicq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_s32)))
+int32x4_t __arm_vbicq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_s8)))
+int8x16_t __arm_vbicq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_s8)))
+int8x16_t __arm_vbicq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_u16)))
+uint16x8_t __arm_vbicq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_u16)))
+uint16x8_t __arm_vbicq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_u32)))
+uint32x4_t __arm_vbicq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_u32)))
+uint32x4_t __arm_vbicq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_u8)))
+uint8x16_t __arm_vbicq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_u8)))
+uint8x16_t __arm_vbicq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_s16)))
+int16x8_t __arm_vbicq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_s16)))
+int16x8_t __arm_vbicq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_s32)))
+int32x4_t __arm_vbicq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_s32)))
+int32x4_t __arm_vbicq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_s8)))
+int8x16_t __arm_vbicq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_s8)))
+int8x16_t __arm_vbicq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_u16)))
+uint16x8_t __arm_vbicq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_u16)))
+uint16x8_t __arm_vbicq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_u32)))
+uint32x4_t __arm_vbicq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_u32)))
+uint32x4_t __arm_vbicq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_u8)))
+uint8x16_t __arm_vbicq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_u8)))
+uint8x16_t __arm_vbicq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_s16)))
+int16x8_t __arm_vbrsrq_m_n_s16(int16x8_t, int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_s16)))
+int16x8_t __arm_vbrsrq_m(int16x8_t, int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_s32)))
+int32x4_t __arm_vbrsrq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_s32)))
+int32x4_t __arm_vbrsrq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_s8)))
+int8x16_t __arm_vbrsrq_m_n_s8(int8x16_t, int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_s8)))
+int8x16_t __arm_vbrsrq_m(int8x16_t, int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_u16)))
+uint16x8_t __arm_vbrsrq_m_n_u16(uint16x8_t, uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_u16)))
+uint16x8_t __arm_vbrsrq_m(uint16x8_t, uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_u32)))
+uint32x4_t __arm_vbrsrq_m_n_u32(uint32x4_t, uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_u32)))
+uint32x4_t __arm_vbrsrq_m(uint32x4_t, uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_u8)))
+uint8x16_t __arm_vbrsrq_m_n_u8(uint8x16_t, uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_u8)))
+uint8x16_t __arm_vbrsrq_m(uint8x16_t, uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_s16)))
+int16x8_t __arm_vbrsrq_n_s16(int16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_s16)))
+int16x8_t __arm_vbrsrq(int16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_s32)))
+int32x4_t __arm_vbrsrq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_s32)))
+int32x4_t __arm_vbrsrq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_s8)))
+int8x16_t __arm_vbrsrq_n_s8(int8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_s8)))
+int8x16_t __arm_vbrsrq(int8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_u16)))
+uint16x8_t __arm_vbrsrq_n_u16(uint16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_u16)))
+uint16x8_t __arm_vbrsrq(uint16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_u32)))
+uint32x4_t __arm_vbrsrq_n_u32(uint32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_u32)))
+uint32x4_t __arm_vbrsrq(uint32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_u8)))
+uint8x16_t __arm_vbrsrq_n_u8(uint8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_u8)))
+uint8x16_t __arm_vbrsrq(uint8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_s16)))
+int16x8_t __arm_vbrsrq_x_n_s16(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_s16)))
+int16x8_t __arm_vbrsrq_x(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_s32)))
+int32x4_t __arm_vbrsrq_x_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_s32)))
+int32x4_t __arm_vbrsrq_x(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_s8)))
+int8x16_t __arm_vbrsrq_x_n_s8(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_s8)))
+int8x16_t __arm_vbrsrq_x(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_u16)))
+uint16x8_t __arm_vbrsrq_x_n_u16(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_u16)))
+uint16x8_t __arm_vbrsrq_x(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_u32)))
+uint32x4_t __arm_vbrsrq_x_n_u32(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_u32)))
+uint32x4_t __arm_vbrsrq_x(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_u8)))
+uint8x16_t __arm_vbrsrq_x_n_u8(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_u8)))
+uint8x16_t __arm_vbrsrq_x(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_s16)))
+int16x8_t __arm_vcaddq_rot270_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_s16)))
+int16x8_t __arm_vcaddq_rot270_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_s32)))
+int32x4_t __arm_vcaddq_rot270_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_s32)))
+int32x4_t __arm_vcaddq_rot270_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_s8)))
+int8x16_t __arm_vcaddq_rot270_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_s8)))
+int8x16_t __arm_vcaddq_rot270_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_u16)))
+uint16x8_t __arm_vcaddq_rot270_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_u16)))
+uint16x8_t __arm_vcaddq_rot270_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_u32)))
+uint32x4_t __arm_vcaddq_rot270_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_u32)))
+uint32x4_t __arm_vcaddq_rot270_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_u8)))
+uint8x16_t __arm_vcaddq_rot270_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_u8)))
+uint8x16_t __arm_vcaddq_rot270_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_s16)))
+int16x8_t __arm_vcaddq_rot270_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_s16)))
+int16x8_t __arm_vcaddq_rot270(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_s32)))
+int32x4_t __arm_vcaddq_rot270_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_s32)))
+int32x4_t __arm_vcaddq_rot270(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_s8)))
+int8x16_t __arm_vcaddq_rot270_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_s8)))
+int8x16_t __arm_vcaddq_rot270(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_u16)))
+uint16x8_t __arm_vcaddq_rot270_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_u16)))
+uint16x8_t __arm_vcaddq_rot270(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_u32)))
+uint32x4_t __arm_vcaddq_rot270_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_u32)))
+uint32x4_t __arm_vcaddq_rot270(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_u8)))
+uint8x16_t __arm_vcaddq_rot270_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_u8)))
+uint8x16_t __arm_vcaddq_rot270(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_s16)))
+int16x8_t __arm_vcaddq_rot270_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_s16)))
+int16x8_t __arm_vcaddq_rot270_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_s32)))
+int32x4_t __arm_vcaddq_rot270_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_s32)))
+int32x4_t __arm_vcaddq_rot270_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_s8)))
+int8x16_t __arm_vcaddq_rot270_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_s8)))
+int8x16_t __arm_vcaddq_rot270_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_u16)))
+uint16x8_t __arm_vcaddq_rot270_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_u16)))
+uint16x8_t __arm_vcaddq_rot270_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_u32)))
+uint32x4_t __arm_vcaddq_rot270_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_u32)))
+uint32x4_t __arm_vcaddq_rot270_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_u8)))
+uint8x16_t __arm_vcaddq_rot270_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_u8)))
+uint8x16_t __arm_vcaddq_rot270_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_s16)))
+int16x8_t __arm_vcaddq_rot90_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_s16)))
+int16x8_t __arm_vcaddq_rot90_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_s32)))
+int32x4_t __arm_vcaddq_rot90_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_s32)))
+int32x4_t __arm_vcaddq_rot90_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_s8)))
+int8x16_t __arm_vcaddq_rot90_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_s8)))
+int8x16_t __arm_vcaddq_rot90_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_u16)))
+uint16x8_t __arm_vcaddq_rot90_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_u16)))
+uint16x8_t __arm_vcaddq_rot90_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_u32)))
+uint32x4_t __arm_vcaddq_rot90_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_u32)))
+uint32x4_t __arm_vcaddq_rot90_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_u8)))
+uint8x16_t __arm_vcaddq_rot90_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_u8)))
+uint8x16_t __arm_vcaddq_rot90_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_s16)))
+int16x8_t __arm_vcaddq_rot90_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_s16)))
+int16x8_t __arm_vcaddq_rot90(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_s32)))
+int32x4_t __arm_vcaddq_rot90_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_s32)))
+int32x4_t __arm_vcaddq_rot90(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_s8)))
+int8x16_t __arm_vcaddq_rot90_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_s8)))
+int8x16_t __arm_vcaddq_rot90(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_u16)))
+uint16x8_t __arm_vcaddq_rot90_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_u16)))
+uint16x8_t __arm_vcaddq_rot90(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_u32)))
+uint32x4_t __arm_vcaddq_rot90_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_u32)))
+uint32x4_t __arm_vcaddq_rot90(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_u8)))
+uint8x16_t __arm_vcaddq_rot90_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_u8)))
+uint8x16_t __arm_vcaddq_rot90(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_s16)))
+int16x8_t __arm_vcaddq_rot90_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_s16)))
+int16x8_t __arm_vcaddq_rot90_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_s32)))
+int32x4_t __arm_vcaddq_rot90_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_s32)))
+int32x4_t __arm_vcaddq_rot90_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_s8)))
+int8x16_t __arm_vcaddq_rot90_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_s8)))
+int8x16_t __arm_vcaddq_rot90_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_u16)))
+uint16x8_t __arm_vcaddq_rot90_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_u16)))
+uint16x8_t __arm_vcaddq_rot90_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_u32)))
+uint32x4_t __arm_vcaddq_rot90_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_u32)))
+uint32x4_t __arm_vcaddq_rot90_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_u8)))
+uint8x16_t __arm_vcaddq_rot90_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_u8)))
+uint8x16_t __arm_vcaddq_rot90_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_m_s16)))
+int16x8_t __arm_vclsq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_m_s16)))
+int16x8_t __arm_vclsq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_m_s32)))
+int32x4_t __arm_vclsq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_m_s32)))
+int32x4_t __arm_vclsq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_m_s8)))
+int8x16_t __arm_vclsq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_m_s8)))
+int8x16_t __arm_vclsq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_s16)))
+int16x8_t __arm_vclsq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_s16)))
+int16x8_t __arm_vclsq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_s32)))
+int32x4_t __arm_vclsq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_s32)))
+int32x4_t __arm_vclsq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_s8)))
+int8x16_t __arm_vclsq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_s8)))
+int8x16_t __arm_vclsq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_x_s16)))
+int16x8_t __arm_vclsq_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_x_s16)))
+int16x8_t __arm_vclsq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_x_s32)))
+int32x4_t __arm_vclsq_x_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_x_s32)))
+int32x4_t __arm_vclsq_x(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_x_s8)))
+int8x16_t __arm_vclsq_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_x_s8)))
+int8x16_t __arm_vclsq_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_s16)))
+int16x8_t __arm_vclzq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_s16)))
+int16x8_t __arm_vclzq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_s32)))
+int32x4_t __arm_vclzq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_s32)))
+int32x4_t __arm_vclzq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_s8)))
+int8x16_t __arm_vclzq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_s8)))
+int8x16_t __arm_vclzq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_u16)))
+uint16x8_t __arm_vclzq_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_u16)))
+uint16x8_t __arm_vclzq_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_u32)))
+uint32x4_t __arm_vclzq_m_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_u32)))
+uint32x4_t __arm_vclzq_m(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_u8)))
+uint8x16_t __arm_vclzq_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_u8)))
+uint8x16_t __arm_vclzq_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_s16)))
+int16x8_t __arm_vclzq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_s16)))
+int16x8_t __arm_vclzq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_s32)))
+int32x4_t __arm_vclzq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_s32)))
+int32x4_t __arm_vclzq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_s8)))
+int8x16_t __arm_vclzq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_s8)))
+int8x16_t __arm_vclzq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_u16)))
+uint16x8_t __arm_vclzq_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_u16)))
+uint16x8_t __arm_vclzq(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_u32)))
+uint32x4_t __arm_vclzq_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_u32)))
+uint32x4_t __arm_vclzq(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_u8)))
+uint8x16_t __arm_vclzq_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_u8)))
+uint8x16_t __arm_vclzq(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_s16)))
+int16x8_t __arm_vclzq_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_s16)))
+int16x8_t __arm_vclzq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_s32)))
+int32x4_t __arm_vclzq_x_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_s32)))
+int32x4_t __arm_vclzq_x(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_s8)))
+int8x16_t __arm_vclzq_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_s8)))
+int8x16_t __arm_vclzq_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_u16)))
+uint16x8_t __arm_vclzq_x_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_u16)))
+uint16x8_t __arm_vclzq_x(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_u32)))
+uint32x4_t __arm_vclzq_x_u32(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_u32)))
+uint32x4_t __arm_vclzq_x(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_u8)))
+uint8x16_t __arm_vclzq_x_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_u8)))
+uint8x16_t __arm_vclzq_x(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_n_u16)))
+mve_pred16_t __arm_vcmpcsq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_n_u16)))
+mve_pred16_t __arm_vcmpcsq_m(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_n_u32)))
+mve_pred16_t __arm_vcmpcsq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_n_u32)))
+mve_pred16_t __arm_vcmpcsq_m(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_n_u8)))
+mve_pred16_t __arm_vcmpcsq_m_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_n_u8)))
+mve_pred16_t __arm_vcmpcsq_m(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_u16)))
+mve_pred16_t __arm_vcmpcsq_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_u16)))
+mve_pred16_t __arm_vcmpcsq_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_u32)))
+mve_pred16_t __arm_vcmpcsq_m_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_u32)))
+mve_pred16_t __arm_vcmpcsq_m(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_u8)))
+mve_pred16_t __arm_vcmpcsq_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_u8)))
+mve_pred16_t __arm_vcmpcsq_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_n_u16)))
+mve_pred16_t __arm_vcmpcsq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_n_u16)))
+mve_pred16_t __arm_vcmpcsq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_n_u32)))
+mve_pred16_t __arm_vcmpcsq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_n_u32)))
+mve_pred16_t __arm_vcmpcsq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_n_u8)))
+mve_pred16_t __arm_vcmpcsq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_n_u8)))
+mve_pred16_t __arm_vcmpcsq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_u16)))
+mve_pred16_t __arm_vcmpcsq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_u16)))
+mve_pred16_t __arm_vcmpcsq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_u32)))
+mve_pred16_t __arm_vcmpcsq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_u32)))
+mve_pred16_t __arm_vcmpcsq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_u8)))
+mve_pred16_t __arm_vcmpcsq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_u8)))
+mve_pred16_t __arm_vcmpcsq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_s16)))
+mve_pred16_t __arm_vcmpeqq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_s16)))
+mve_pred16_t __arm_vcmpeqq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_s32)))
+mve_pred16_t __arm_vcmpeqq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_s32)))
+mve_pred16_t __arm_vcmpeqq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_s8)))
+mve_pred16_t __arm_vcmpeqq_m_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_s8)))
+mve_pred16_t __arm_vcmpeqq_m(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_u16)))
+mve_pred16_t __arm_vcmpeqq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_u16)))
+mve_pred16_t __arm_vcmpeqq_m(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_u32)))
+mve_pred16_t __arm_vcmpeqq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_u32)))
+mve_pred16_t __arm_vcmpeqq_m(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_u8)))
+mve_pred16_t __arm_vcmpeqq_m_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_u8)))
+mve_pred16_t __arm_vcmpeqq_m(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_s16)))
+mve_pred16_t __arm_vcmpeqq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_s16)))
+mve_pred16_t __arm_vcmpeqq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_s32)))
+mve_pred16_t __arm_vcmpeqq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_s32)))
+mve_pred16_t __arm_vcmpeqq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_s8)))
+mve_pred16_t __arm_vcmpeqq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_s8)))
+mve_pred16_t __arm_vcmpeqq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_u16)))
+mve_pred16_t __arm_vcmpeqq_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_u16)))
+mve_pred16_t __arm_vcmpeqq_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_u32)))
+mve_pred16_t __arm_vcmpeqq_m_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_u32)))
+mve_pred16_t __arm_vcmpeqq_m(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_u8)))
+mve_pred16_t __arm_vcmpeqq_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_u8)))
+mve_pred16_t __arm_vcmpeqq_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_s16)))
+mve_pred16_t __arm_vcmpeqq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_s16)))
+mve_pred16_t __arm_vcmpeqq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_s32)))
+mve_pred16_t __arm_vcmpeqq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_s32)))
+mve_pred16_t __arm_vcmpeqq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_s8)))
+mve_pred16_t __arm_vcmpeqq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_s8)))
+mve_pred16_t __arm_vcmpeqq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_u16)))
+mve_pred16_t __arm_vcmpeqq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_u16)))
+mve_pred16_t __arm_vcmpeqq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_u32)))
+mve_pred16_t __arm_vcmpeqq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_u32)))
+mve_pred16_t __arm_vcmpeqq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_u8)))
+mve_pred16_t __arm_vcmpeqq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_u8)))
+mve_pred16_t __arm_vcmpeqq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_s16)))
+mve_pred16_t __arm_vcmpeqq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_s16)))
+mve_pred16_t __arm_vcmpeqq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_s32)))
+mve_pred16_t __arm_vcmpeqq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_s32)))
+mve_pred16_t __arm_vcmpeqq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_s8)))
+mve_pred16_t __arm_vcmpeqq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_s8)))
+mve_pred16_t __arm_vcmpeqq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_u16)))
+mve_pred16_t __arm_vcmpeqq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_u16)))
+mve_pred16_t __arm_vcmpeqq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_u32)))
+mve_pred16_t __arm_vcmpeqq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_u32)))
+mve_pred16_t __arm_vcmpeqq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_u8)))
+mve_pred16_t __arm_vcmpeqq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_u8)))
+mve_pred16_t __arm_vcmpeqq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_s16)))
+mve_pred16_t __arm_vcmpgeq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_s16)))
+mve_pred16_t __arm_vcmpgeq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_s32)))
+mve_pred16_t __arm_vcmpgeq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_s32)))
+mve_pred16_t __arm_vcmpgeq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_s8)))
+mve_pred16_t __arm_vcmpgeq_m_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_s8)))
+mve_pred16_t __arm_vcmpgeq_m(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_s16)))
+mve_pred16_t __arm_vcmpgeq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_s16)))
+mve_pred16_t __arm_vcmpgeq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_s32)))
+mve_pred16_t __arm_vcmpgeq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_s32)))
+mve_pred16_t __arm_vcmpgeq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_s8)))
+mve_pred16_t __arm_vcmpgeq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_s8)))
+mve_pred16_t __arm_vcmpgeq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_s16)))
+mve_pred16_t __arm_vcmpgeq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_s16)))
+mve_pred16_t __arm_vcmpgeq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_s32)))
+mve_pred16_t __arm_vcmpgeq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_s32)))
+mve_pred16_t __arm_vcmpgeq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_s8)))
+mve_pred16_t __arm_vcmpgeq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_s8)))
+mve_pred16_t __arm_vcmpgeq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_s16)))
+mve_pred16_t __arm_vcmpgeq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_s16)))
+mve_pred16_t __arm_vcmpgeq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_s32)))
+mve_pred16_t __arm_vcmpgeq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_s32)))
+mve_pred16_t __arm_vcmpgeq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_s8)))
+mve_pred16_t __arm_vcmpgeq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_s8)))
+mve_pred16_t __arm_vcmpgeq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_s16)))
+mve_pred16_t __arm_vcmpgtq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_s16)))
+mve_pred16_t __arm_vcmpgtq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_s32)))
+mve_pred16_t __arm_vcmpgtq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_s32)))
+mve_pred16_t __arm_vcmpgtq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_s8)))
+mve_pred16_t __arm_vcmpgtq_m_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_s8)))
+mve_pred16_t __arm_vcmpgtq_m(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_s16)))
+mve_pred16_t __arm_vcmpgtq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_s16)))
+mve_pred16_t __arm_vcmpgtq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_s32)))
+mve_pred16_t __arm_vcmpgtq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_s32)))
+mve_pred16_t __arm_vcmpgtq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_s8)))
+mve_pred16_t __arm_vcmpgtq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_s8)))
+mve_pred16_t __arm_vcmpgtq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_s16)))
+mve_pred16_t __arm_vcmpgtq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_s16)))
+mve_pred16_t __arm_vcmpgtq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_s32)))
+mve_pred16_t __arm_vcmpgtq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_s32)))
+mve_pred16_t __arm_vcmpgtq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_s8)))
+mve_pred16_t __arm_vcmpgtq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_s8)))
+mve_pred16_t __arm_vcmpgtq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_s16)))
+mve_pred16_t __arm_vcmpgtq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_s16)))
+mve_pred16_t __arm_vcmpgtq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_s32)))
+mve_pred16_t __arm_vcmpgtq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_s32)))
+mve_pred16_t __arm_vcmpgtq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_s8)))
+mve_pred16_t __arm_vcmpgtq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_s8)))
+mve_pred16_t __arm_vcmpgtq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_n_u16)))
+mve_pred16_t __arm_vcmphiq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_n_u16)))
+mve_pred16_t __arm_vcmphiq_m(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_n_u32)))
+mve_pred16_t __arm_vcmphiq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_n_u32)))
+mve_pred16_t __arm_vcmphiq_m(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_n_u8)))
+mve_pred16_t __arm_vcmphiq_m_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_n_u8)))
+mve_pred16_t __arm_vcmphiq_m(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_u16)))
+mve_pred16_t __arm_vcmphiq_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_u16)))
+mve_pred16_t __arm_vcmphiq_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_u32)))
+mve_pred16_t __arm_vcmphiq_m_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_u32)))
+mve_pred16_t __arm_vcmphiq_m(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_u8)))
+mve_pred16_t __arm_vcmphiq_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_u8)))
+mve_pred16_t __arm_vcmphiq_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_n_u16)))
+mve_pred16_t __arm_vcmphiq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_n_u16)))
+mve_pred16_t __arm_vcmphiq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_n_u32)))
+mve_pred16_t __arm_vcmphiq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_n_u32)))
+mve_pred16_t __arm_vcmphiq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_n_u8)))
+mve_pred16_t __arm_vcmphiq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_n_u8)))
+mve_pred16_t __arm_vcmphiq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_u16)))
+mve_pred16_t __arm_vcmphiq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_u16)))
+mve_pred16_t __arm_vcmphiq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_u32)))
+mve_pred16_t __arm_vcmphiq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_u32)))
+mve_pred16_t __arm_vcmphiq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_u8)))
+mve_pred16_t __arm_vcmphiq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_u8)))
+mve_pred16_t __arm_vcmphiq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_s16)))
+mve_pred16_t __arm_vcmpleq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_s16)))
+mve_pred16_t __arm_vcmpleq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_s32)))
+mve_pred16_t __arm_vcmpleq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_s32)))
+mve_pred16_t __arm_vcmpleq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_s8)))
+mve_pred16_t __arm_vcmpleq_m_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_s8)))
+mve_pred16_t __arm_vcmpleq_m(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_s16)))
+mve_pred16_t __arm_vcmpleq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_s16)))
+mve_pred16_t __arm_vcmpleq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_s32)))
+mve_pred16_t __arm_vcmpleq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_s32)))
+mve_pred16_t __arm_vcmpleq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_s8)))
+mve_pred16_t __arm_vcmpleq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_s8)))
+mve_pred16_t __arm_vcmpleq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_s16)))
+mve_pred16_t __arm_vcmpleq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_s16)))
+mve_pred16_t __arm_vcmpleq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_s32)))
+mve_pred16_t __arm_vcmpleq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_s32)))
+mve_pred16_t __arm_vcmpleq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_s8)))
+mve_pred16_t __arm_vcmpleq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_s8)))
+mve_pred16_t __arm_vcmpleq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_s16)))
+mve_pred16_t __arm_vcmpleq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_s16)))
+mve_pred16_t __arm_vcmpleq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_s32)))
+mve_pred16_t __arm_vcmpleq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_s32)))
+mve_pred16_t __arm_vcmpleq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_s8)))
+mve_pred16_t __arm_vcmpleq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_s8)))
+mve_pred16_t __arm_vcmpleq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_s16)))
+mve_pred16_t __arm_vcmpltq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_s16)))
+mve_pred16_t __arm_vcmpltq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_s32)))
+mve_pred16_t __arm_vcmpltq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_s32)))
+mve_pred16_t __arm_vcmpltq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_s8)))
+mve_pred16_t __arm_vcmpltq_m_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_s8)))
+mve_pred16_t __arm_vcmpltq_m(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_s16)))
+mve_pred16_t __arm_vcmpltq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_s16)))
+mve_pred16_t __arm_vcmpltq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_s32)))
+mve_pred16_t __arm_vcmpltq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_s32)))
+mve_pred16_t __arm_vcmpltq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_s8)))
+mve_pred16_t __arm_vcmpltq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_s8)))
+mve_pred16_t __arm_vcmpltq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_s16)))
+mve_pred16_t __arm_vcmpltq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_s16)))
+mve_pred16_t __arm_vcmpltq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_s32)))
+mve_pred16_t __arm_vcmpltq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_s32)))
+mve_pred16_t __arm_vcmpltq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_s8)))
+mve_pred16_t __arm_vcmpltq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_s8)))
+mve_pred16_t __arm_vcmpltq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_s16)))
+mve_pred16_t __arm_vcmpltq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_s16)))
+mve_pred16_t __arm_vcmpltq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_s32)))
+mve_pred16_t __arm_vcmpltq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_s32)))
+mve_pred16_t __arm_vcmpltq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_s8)))
+mve_pred16_t __arm_vcmpltq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_s8)))
+mve_pred16_t __arm_vcmpltq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_s16)))
+mve_pred16_t __arm_vcmpneq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_s16)))
+mve_pred16_t __arm_vcmpneq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_s32)))
+mve_pred16_t __arm_vcmpneq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_s32)))
+mve_pred16_t __arm_vcmpneq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_s8)))
+mve_pred16_t __arm_vcmpneq_m_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_s8)))
+mve_pred16_t __arm_vcmpneq_m(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_u16)))
+mve_pred16_t __arm_vcmpneq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_u16)))
+mve_pred16_t __arm_vcmpneq_m(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_u32)))
+mve_pred16_t __arm_vcmpneq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_u32)))
+mve_pred16_t __arm_vcmpneq_m(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_u8)))
+mve_pred16_t __arm_vcmpneq_m_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_u8)))
+mve_pred16_t __arm_vcmpneq_m(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_s16)))
+mve_pred16_t __arm_vcmpneq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_s16)))
+mve_pred16_t __arm_vcmpneq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_s32)))
+mve_pred16_t __arm_vcmpneq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_s32)))
+mve_pred16_t __arm_vcmpneq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_s8)))
+mve_pred16_t __arm_vcmpneq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_s8)))
+mve_pred16_t __arm_vcmpneq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_u16)))
+mve_pred16_t __arm_vcmpneq_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_u16)))
+mve_pred16_t __arm_vcmpneq_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_u32)))
+mve_pred16_t __arm_vcmpneq_m_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_u32)))
+mve_pred16_t __arm_vcmpneq_m(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_u8)))
+mve_pred16_t __arm_vcmpneq_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_u8)))
+mve_pred16_t __arm_vcmpneq_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_s16)))
+mve_pred16_t __arm_vcmpneq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_s16)))
+mve_pred16_t __arm_vcmpneq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_s32)))
+mve_pred16_t __arm_vcmpneq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_s32)))
+mve_pred16_t __arm_vcmpneq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_s8)))
+mve_pred16_t __arm_vcmpneq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_s8)))
+mve_pred16_t __arm_vcmpneq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_u16)))
+mve_pred16_t __arm_vcmpneq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_u16)))
+mve_pred16_t __arm_vcmpneq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_u32)))
+mve_pred16_t __arm_vcmpneq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_u32)))
+mve_pred16_t __arm_vcmpneq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_u8)))
+mve_pred16_t __arm_vcmpneq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_u8)))
+mve_pred16_t __arm_vcmpneq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_s16)))
+mve_pred16_t __arm_vcmpneq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_s16)))
+mve_pred16_t __arm_vcmpneq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_s32)))
+mve_pred16_t __arm_vcmpneq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_s32)))
+mve_pred16_t __arm_vcmpneq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_s8)))
+mve_pred16_t __arm_vcmpneq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_s8)))
+mve_pred16_t __arm_vcmpneq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_u16)))
+mve_pred16_t __arm_vcmpneq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_u16)))
+mve_pred16_t __arm_vcmpneq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_u32)))
+mve_pred16_t __arm_vcmpneq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_u32)))
+mve_pred16_t __arm_vcmpneq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_u8)))
+mve_pred16_t __arm_vcmpneq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_u8)))
+mve_pred16_t __arm_vcmpneq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_s16)))
+int16x8_t __arm_vcreateq_s16(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_s32)))
+int32x4_t __arm_vcreateq_s32(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_s64)))
+int64x2_t __arm_vcreateq_s64(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_s8)))
+int8x16_t __arm_vcreateq_s8(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_u16)))
+uint16x8_t __arm_vcreateq_u16(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_u32)))
+uint32x4_t __arm_vcreateq_u32(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_u64)))
+uint64x2_t __arm_vcreateq_u64(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_u8)))
+uint8x16_t __arm_vcreateq_u8(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp16q)))
+mve_pred16_t __arm_vctp16q(uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp16q_m)))
+mve_pred16_t __arm_vctp16q_m(uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp32q)))
+mve_pred16_t __arm_vctp32q(uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp32q_m)))
+mve_pred16_t __arm_vctp32q_m(uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp64q)))
+mve_pred16_t __arm_vctp64q(uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp64q_m)))
+mve_pred16_t __arm_vctp64q_m(uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp8q)))
+mve_pred16_t __arm_vctp8q(uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp8q_m)))
+mve_pred16_t __arm_vctp8q_m(uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_n_u16)))
+uint16x8_t __arm_vddupq_m_n_u16(uint16x8_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_n_u16)))
+uint16x8_t __arm_vddupq_m(uint16x8_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_n_u32)))
+uint32x4_t __arm_vddupq_m_n_u32(uint32x4_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_n_u32)))
+uint32x4_t __arm_vddupq_m(uint32x4_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_n_u8)))
+uint8x16_t __arm_vddupq_m_n_u8(uint8x16_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_n_u8)))
+uint8x16_t __arm_vddupq_m(uint8x16_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_wb_u16)))
+uint16x8_t __arm_vddupq_m_wb_u16(uint16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_wb_u16)))
+uint16x8_t __arm_vddupq_m(uint16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_wb_u32)))
+uint32x4_t __arm_vddupq_m_wb_u32(uint32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_wb_u32)))
+uint32x4_t __arm_vddupq_m(uint32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_wb_u8)))
+uint8x16_t __arm_vddupq_m_wb_u8(uint8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_wb_u8)))
+uint8x16_t __arm_vddupq_m(uint8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_n_u16)))
+uint16x8_t __arm_vddupq_n_u16(uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_n_u16)))
+uint16x8_t __arm_vddupq_u16(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_n_u32)))
+uint32x4_t __arm_vddupq_n_u32(uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_n_u32)))
+uint32x4_t __arm_vddupq_u32(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_n_u8)))
+uint8x16_t __arm_vddupq_n_u8(uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_n_u8)))
+uint8x16_t __arm_vddupq_u8(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_wb_u16)))
+uint16x8_t __arm_vddupq_wb_u16(uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_wb_u16)))
+uint16x8_t __arm_vddupq_u16(uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_wb_u32)))
+uint32x4_t __arm_vddupq_wb_u32(uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_wb_u32)))
+uint32x4_t __arm_vddupq_u32(uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_wb_u8)))
+uint8x16_t __arm_vddupq_wb_u8(uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_wb_u8)))
+uint8x16_t __arm_vddupq_u8(uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_n_u16)))
+uint16x8_t __arm_vddupq_x_n_u16(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_n_u16)))
+uint16x8_t __arm_vddupq_x_u16(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_n_u32)))
+uint32x4_t __arm_vddupq_x_n_u32(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_n_u32)))
+uint32x4_t __arm_vddupq_x_u32(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_n_u8)))
+uint8x16_t __arm_vddupq_x_n_u8(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_n_u8)))
+uint8x16_t __arm_vddupq_x_u8(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_wb_u16)))
+uint16x8_t __arm_vddupq_x_wb_u16(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_wb_u16)))
+uint16x8_t __arm_vddupq_x_u16(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_wb_u32)))
+uint32x4_t __arm_vddupq_x_wb_u32(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_wb_u32)))
+uint32x4_t __arm_vddupq_x_u32(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_wb_u8)))
+uint8x16_t __arm_vddupq_x_wb_u8(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_wb_u8)))
+uint8x16_t __arm_vddupq_x_u8(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_s16)))
+int16x8_t __arm_vdupq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_s16)))
+int16x8_t __arm_vdupq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_s32)))
+int32x4_t __arm_vdupq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_s32)))
+int32x4_t __arm_vdupq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_s8)))
+int8x16_t __arm_vdupq_m_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_s8)))
+int8x16_t __arm_vdupq_m(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_u16)))
+uint16x8_t __arm_vdupq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_u16)))
+uint16x8_t __arm_vdupq_m(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_u32)))
+uint32x4_t __arm_vdupq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_u32)))
+uint32x4_t __arm_vdupq_m(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_u8)))
+uint8x16_t __arm_vdupq_m_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_u8)))
+uint8x16_t __arm_vdupq_m(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_s16)))
+int16x8_t __arm_vdupq_n_s16(int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_s32)))
+int32x4_t __arm_vdupq_n_s32(int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_s8)))
+int8x16_t __arm_vdupq_n_s8(int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_u16)))
+uint16x8_t __arm_vdupq_n_u16(uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_u32)))
+uint32x4_t __arm_vdupq_n_u32(uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_u8)))
+uint8x16_t __arm_vdupq_n_u8(uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_s16)))
+int16x8_t __arm_vdupq_x_n_s16(int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_s32)))
+int32x4_t __arm_vdupq_x_n_s32(int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_s8)))
+int8x16_t __arm_vdupq_x_n_s8(int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_u16)))
+uint16x8_t __arm_vdupq_x_n_u16(uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_u32)))
+uint32x4_t __arm_vdupq_x_n_u32(uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_u8)))
+uint8x16_t __arm_vdupq_x_n_u8(uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_n_u16)))
+uint16x8_t __arm_vdwdupq_m_n_u16(uint16x8_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_n_u16)))
+uint16x8_t __arm_vdwdupq_m(uint16x8_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_n_u32)))
+uint32x4_t __arm_vdwdupq_m_n_u32(uint32x4_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_n_u32)))
+uint32x4_t __arm_vdwdupq_m(uint32x4_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_n_u8)))
+uint8x16_t __arm_vdwdupq_m_n_u8(uint8x16_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_n_u8)))
+uint8x16_t __arm_vdwdupq_m(uint8x16_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_wb_u16)))
+uint16x8_t __arm_vdwdupq_m_wb_u16(uint16x8_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_wb_u16)))
+uint16x8_t __arm_vdwdupq_m(uint16x8_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_wb_u32)))
+uint32x4_t __arm_vdwdupq_m_wb_u32(uint32x4_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_wb_u32)))
+uint32x4_t __arm_vdwdupq_m(uint32x4_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_wb_u8)))
+uint8x16_t __arm_vdwdupq_m_wb_u8(uint8x16_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_wb_u8)))
+uint8x16_t __arm_vdwdupq_m(uint8x16_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_n_u16)))
+uint16x8_t __arm_vdwdupq_n_u16(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_n_u16)))
+uint16x8_t __arm_vdwdupq_u16(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_n_u32)))
+uint32x4_t __arm_vdwdupq_n_u32(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_n_u32)))
+uint32x4_t __arm_vdwdupq_u32(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_n_u8)))
+uint8x16_t __arm_vdwdupq_n_u8(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_n_u8)))
+uint8x16_t __arm_vdwdupq_u8(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_wb_u16)))
+uint16x8_t __arm_vdwdupq_wb_u16(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_wb_u16)))
+uint16x8_t __arm_vdwdupq_u16(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_wb_u32)))
+uint32x4_t __arm_vdwdupq_wb_u32(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_wb_u32)))
+uint32x4_t __arm_vdwdupq_u32(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_wb_u8)))
+uint8x16_t __arm_vdwdupq_wb_u8(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_wb_u8)))
+uint8x16_t __arm_vdwdupq_u8(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_n_u16)))
+uint16x8_t __arm_vdwdupq_x_n_u16(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_n_u16)))
+uint16x8_t __arm_vdwdupq_x_u16(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_n_u32)))
+uint32x4_t __arm_vdwdupq_x_n_u32(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_n_u32)))
+uint32x4_t __arm_vdwdupq_x_u32(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_n_u8)))
+uint8x16_t __arm_vdwdupq_x_n_u8(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_n_u8)))
+uint8x16_t __arm_vdwdupq_x_u8(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_wb_u16)))
+uint16x8_t __arm_vdwdupq_x_wb_u16(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_wb_u16)))
+uint16x8_t __arm_vdwdupq_x_u16(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_wb_u32)))
+uint32x4_t __arm_vdwdupq_x_wb_u32(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_wb_u32)))
+uint32x4_t __arm_vdwdupq_x_u32(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_wb_u8)))
+uint8x16_t __arm_vdwdupq_x_wb_u8(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_wb_u8)))
+uint8x16_t __arm_vdwdupq_x_u8(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_s16)))
+int16x8_t __arm_veorq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_s16)))
+int16x8_t __arm_veorq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_s32)))
+int32x4_t __arm_veorq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_s32)))
+int32x4_t __arm_veorq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_s8)))
+int8x16_t __arm_veorq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_s8)))
+int8x16_t __arm_veorq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_u16)))
+uint16x8_t __arm_veorq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_u16)))
+uint16x8_t __arm_veorq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_u32)))
+uint32x4_t __arm_veorq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_u32)))
+uint32x4_t __arm_veorq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_u8)))
+uint8x16_t __arm_veorq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_u8)))
+uint8x16_t __arm_veorq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_s16)))
+int16x8_t __arm_veorq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_s16)))
+int16x8_t __arm_veorq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_s32)))
+int32x4_t __arm_veorq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_s32)))
+int32x4_t __arm_veorq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_s8)))
+int8x16_t __arm_veorq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_s8)))
+int8x16_t __arm_veorq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_u16)))
+uint16x8_t __arm_veorq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_u16)))
+uint16x8_t __arm_veorq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_u32)))
+uint32x4_t __arm_veorq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_u32)))
+uint32x4_t __arm_veorq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_u8)))
+uint8x16_t __arm_veorq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_u8)))
+uint8x16_t __arm_veorq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_s16)))
+int16x8_t __arm_veorq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_s16)))
+int16x8_t __arm_veorq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_s32)))
+int32x4_t __arm_veorq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_s32)))
+int32x4_t __arm_veorq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_s8)))
+int8x16_t __arm_veorq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_s8)))
+int8x16_t __arm_veorq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_u16)))
+uint16x8_t __arm_veorq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_u16)))
+uint16x8_t __arm_veorq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_u32)))
+uint32x4_t __arm_veorq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_u32)))
+uint32x4_t __arm_veorq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_u8)))
+uint8x16_t __arm_veorq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_u8)))
+uint8x16_t __arm_veorq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s16)))
+int16_t __arm_vgetq_lane_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s16)))
+int16_t __arm_vgetq_lane(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s32)))
+int32_t __arm_vgetq_lane_s32(int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s32)))
+int32_t __arm_vgetq_lane(int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s64)))
+int64_t __arm_vgetq_lane_s64(int64x2_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s64)))
+int64_t __arm_vgetq_lane(int64x2_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s8)))
+int8_t __arm_vgetq_lane_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s8)))
+int8_t __arm_vgetq_lane(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u16)))
+uint16_t __arm_vgetq_lane_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u16)))
+uint16_t __arm_vgetq_lane(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u32)))
+uint32_t __arm_vgetq_lane_u32(uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u32)))
+uint32_t __arm_vgetq_lane(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u64)))
+uint64_t __arm_vgetq_lane_u64(uint64x2_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u64)))
+uint64_t __arm_vgetq_lane(uint64x2_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u8)))
+uint8_t __arm_vgetq_lane_u8(uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u8)))
+uint8_t __arm_vgetq_lane(uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_s16)))
+int16x8_t __arm_vhaddq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_s16)))
+int16x8_t __arm_vhaddq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_s32)))
+int32x4_t __arm_vhaddq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_s32)))
+int32x4_t __arm_vhaddq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_s8)))
+int8x16_t __arm_vhaddq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_s8)))
+int8x16_t __arm_vhaddq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_u16)))
+uint16x8_t __arm_vhaddq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_u16)))
+uint16x8_t __arm_vhaddq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_u32)))
+uint32x4_t __arm_vhaddq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_u32)))
+uint32x4_t __arm_vhaddq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_u8)))
+uint8x16_t __arm_vhaddq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_u8)))
+uint8x16_t __arm_vhaddq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_s16)))
+int16x8_t __arm_vhaddq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_s16)))
+int16x8_t __arm_vhaddq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_s32)))
+int32x4_t __arm_vhaddq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_s32)))
+int32x4_t __arm_vhaddq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_s8)))
+int8x16_t __arm_vhaddq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_s8)))
+int8x16_t __arm_vhaddq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_u16)))
+uint16x8_t __arm_vhaddq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_u16)))
+uint16x8_t __arm_vhaddq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_u32)))
+uint32x4_t __arm_vhaddq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_u32)))
+uint32x4_t __arm_vhaddq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_u8)))
+uint8x16_t __arm_vhaddq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_u8)))
+uint8x16_t __arm_vhaddq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_s16)))
+int16x8_t __arm_vhaddq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_s16)))
+int16x8_t __arm_vhaddq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_s32)))
+int32x4_t __arm_vhaddq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_s32)))
+int32x4_t __arm_vhaddq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_s8)))
+int8x16_t __arm_vhaddq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_s8)))
+int8x16_t __arm_vhaddq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_u16)))
+uint16x8_t __arm_vhaddq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_u16)))
+uint16x8_t __arm_vhaddq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_u32)))
+uint32x4_t __arm_vhaddq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_u32)))
+uint32x4_t __arm_vhaddq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_u8)))
+uint8x16_t __arm_vhaddq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_u8)))
+uint8x16_t __arm_vhaddq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_s16)))
+int16x8_t __arm_vhaddq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_s16)))
+int16x8_t __arm_vhaddq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_s32)))
+int32x4_t __arm_vhaddq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_s32)))
+int32x4_t __arm_vhaddq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_s8)))
+int8x16_t __arm_vhaddq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_s8)))
+int8x16_t __arm_vhaddq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_u16)))
+uint16x8_t __arm_vhaddq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_u16)))
+uint16x8_t __arm_vhaddq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_u32)))
+uint32x4_t __arm_vhaddq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_u32)))
+uint32x4_t __arm_vhaddq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_u8)))
+uint8x16_t __arm_vhaddq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_u8)))
+uint8x16_t __arm_vhaddq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_s16)))
+int16x8_t __arm_vhaddq_x_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_s16)))
+int16x8_t __arm_vhaddq_x(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_s32)))
+int32x4_t __arm_vhaddq_x_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_s32)))
+int32x4_t __arm_vhaddq_x(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_s8)))
+int8x16_t __arm_vhaddq_x_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_s8)))
+int8x16_t __arm_vhaddq_x(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_u16)))
+uint16x8_t __arm_vhaddq_x_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_u16)))
+uint16x8_t __arm_vhaddq_x(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_u32)))
+uint32x4_t __arm_vhaddq_x_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_u32)))
+uint32x4_t __arm_vhaddq_x(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_u8)))
+uint8x16_t __arm_vhaddq_x_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_u8)))
+uint8x16_t __arm_vhaddq_x(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_s16)))
+int16x8_t __arm_vhaddq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_s16)))
+int16x8_t __arm_vhaddq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_s32)))
+int32x4_t __arm_vhaddq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_s32)))
+int32x4_t __arm_vhaddq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_s8)))
+int8x16_t __arm_vhaddq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_s8)))
+int8x16_t __arm_vhaddq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_u16)))
+uint16x8_t __arm_vhaddq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_u16)))
+uint16x8_t __arm_vhaddq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_u32)))
+uint32x4_t __arm_vhaddq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_u32)))
+uint32x4_t __arm_vhaddq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_u8)))
+uint8x16_t __arm_vhaddq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_u8)))
+uint8x16_t __arm_vhaddq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_m_s16)))
+int16x8_t __arm_vhcaddq_rot270_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_m_s16)))
+int16x8_t __arm_vhcaddq_rot270_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_m_s32)))
+int32x4_t __arm_vhcaddq_rot270_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_m_s32)))
+int32x4_t __arm_vhcaddq_rot270_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_m_s8)))
+int8x16_t __arm_vhcaddq_rot270_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_m_s8)))
+int8x16_t __arm_vhcaddq_rot270_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_s16)))
+int16x8_t __arm_vhcaddq_rot270_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_s16)))
+int16x8_t __arm_vhcaddq_rot270(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_s32)))
+int32x4_t __arm_vhcaddq_rot270_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_s32)))
+int32x4_t __arm_vhcaddq_rot270(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_s8)))
+int8x16_t __arm_vhcaddq_rot270_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_s8)))
+int8x16_t __arm_vhcaddq_rot270(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_x_s16)))
+int16x8_t __arm_vhcaddq_rot270_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_x_s16)))
+int16x8_t __arm_vhcaddq_rot270_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_x_s32)))
+int32x4_t __arm_vhcaddq_rot270_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_x_s32)))
+int32x4_t __arm_vhcaddq_rot270_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_x_s8)))
+int8x16_t __arm_vhcaddq_rot270_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_x_s8)))
+int8x16_t __arm_vhcaddq_rot270_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_m_s16)))
+int16x8_t __arm_vhcaddq_rot90_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_m_s16)))
+int16x8_t __arm_vhcaddq_rot90_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_m_s32)))
+int32x4_t __arm_vhcaddq_rot90_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_m_s32)))
+int32x4_t __arm_vhcaddq_rot90_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_m_s8)))
+int8x16_t __arm_vhcaddq_rot90_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_m_s8)))
+int8x16_t __arm_vhcaddq_rot90_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_s16)))
+int16x8_t __arm_vhcaddq_rot90_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_s16)))
+int16x8_t __arm_vhcaddq_rot90(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_s32)))
+int32x4_t __arm_vhcaddq_rot90_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_s32)))
+int32x4_t __arm_vhcaddq_rot90(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_s8)))
+int8x16_t __arm_vhcaddq_rot90_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_s8)))
+int8x16_t __arm_vhcaddq_rot90(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_x_s16)))
+int16x8_t __arm_vhcaddq_rot90_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_x_s16)))
+int16x8_t __arm_vhcaddq_rot90_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_x_s32)))
+int32x4_t __arm_vhcaddq_rot90_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_x_s32)))
+int32x4_t __arm_vhcaddq_rot90_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_x_s8)))
+int8x16_t __arm_vhcaddq_rot90_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_x_s8)))
+int8x16_t __arm_vhcaddq_rot90_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_s16)))
+int16x8_t __arm_vhsubq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_s16)))
+int16x8_t __arm_vhsubq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_s32)))
+int32x4_t __arm_vhsubq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_s32)))
+int32x4_t __arm_vhsubq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_s8)))
+int8x16_t __arm_vhsubq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_s8)))
+int8x16_t __arm_vhsubq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_u16)))
+uint16x8_t __arm_vhsubq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_u16)))
+uint16x8_t __arm_vhsubq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_u32)))
+uint32x4_t __arm_vhsubq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_u32)))
+uint32x4_t __arm_vhsubq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_u8)))
+uint8x16_t __arm_vhsubq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_u8)))
+uint8x16_t __arm_vhsubq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_s16)))
+int16x8_t __arm_vhsubq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_s16)))
+int16x8_t __arm_vhsubq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_s32)))
+int32x4_t __arm_vhsubq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_s32)))
+int32x4_t __arm_vhsubq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_s8)))
+int8x16_t __arm_vhsubq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_s8)))
+int8x16_t __arm_vhsubq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_u16)))
+uint16x8_t __arm_vhsubq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_u16)))
+uint16x8_t __arm_vhsubq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_u32)))
+uint32x4_t __arm_vhsubq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_u32)))
+uint32x4_t __arm_vhsubq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_u8)))
+uint8x16_t __arm_vhsubq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_u8)))
+uint8x16_t __arm_vhsubq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_s16)))
+int16x8_t __arm_vhsubq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_s16)))
+int16x8_t __arm_vhsubq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_s32)))
+int32x4_t __arm_vhsubq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_s32)))
+int32x4_t __arm_vhsubq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_s8)))
+int8x16_t __arm_vhsubq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_s8)))
+int8x16_t __arm_vhsubq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_u16)))
+uint16x8_t __arm_vhsubq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_u16)))
+uint16x8_t __arm_vhsubq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_u32)))
+uint32x4_t __arm_vhsubq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_u32)))
+uint32x4_t __arm_vhsubq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_u8)))
+uint8x16_t __arm_vhsubq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_u8)))
+uint8x16_t __arm_vhsubq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_s16)))
+int16x8_t __arm_vhsubq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_s16)))
+int16x8_t __arm_vhsubq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_s32)))
+int32x4_t __arm_vhsubq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_s32)))
+int32x4_t __arm_vhsubq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_s8)))
+int8x16_t __arm_vhsubq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_s8)))
+int8x16_t __arm_vhsubq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_u16)))
+uint16x8_t __arm_vhsubq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_u16)))
+uint16x8_t __arm_vhsubq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_u32)))
+uint32x4_t __arm_vhsubq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_u32)))
+uint32x4_t __arm_vhsubq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_u8)))
+uint8x16_t __arm_vhsubq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_u8)))
+uint8x16_t __arm_vhsubq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_s16)))
+int16x8_t __arm_vhsubq_x_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_s16)))
+int16x8_t __arm_vhsubq_x(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_s32)))
+int32x4_t __arm_vhsubq_x_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_s32)))
+int32x4_t __arm_vhsubq_x(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_s8)))
+int8x16_t __arm_vhsubq_x_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_s8)))
+int8x16_t __arm_vhsubq_x(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_u16)))
+uint16x8_t __arm_vhsubq_x_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_u16)))
+uint16x8_t __arm_vhsubq_x(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_u32)))
+uint32x4_t __arm_vhsubq_x_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_u32)))
+uint32x4_t __arm_vhsubq_x(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_u8)))
+uint8x16_t __arm_vhsubq_x_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_u8)))
+uint8x16_t __arm_vhsubq_x(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_s16)))
+int16x8_t __arm_vhsubq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_s16)))
+int16x8_t __arm_vhsubq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_s32)))
+int32x4_t __arm_vhsubq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_s32)))
+int32x4_t __arm_vhsubq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_s8)))
+int8x16_t __arm_vhsubq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_s8)))
+int8x16_t __arm_vhsubq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_u16)))
+uint16x8_t __arm_vhsubq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_u16)))
+uint16x8_t __arm_vhsubq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_u32)))
+uint32x4_t __arm_vhsubq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_u32)))
+uint32x4_t __arm_vhsubq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_u8)))
+uint8x16_t __arm_vhsubq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_u8)))
+uint8x16_t __arm_vhsubq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_n_u16)))
+uint16x8_t __arm_vidupq_m_n_u16(uint16x8_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_n_u16)))
+uint16x8_t __arm_vidupq_m(uint16x8_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_n_u32)))
+uint32x4_t __arm_vidupq_m_n_u32(uint32x4_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_n_u32)))
+uint32x4_t __arm_vidupq_m(uint32x4_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_n_u8)))
+uint8x16_t __arm_vidupq_m_n_u8(uint8x16_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_n_u8)))
+uint8x16_t __arm_vidupq_m(uint8x16_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_wb_u16)))
+uint16x8_t __arm_vidupq_m_wb_u16(uint16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_wb_u16)))
+uint16x8_t __arm_vidupq_m(uint16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_wb_u32)))
+uint32x4_t __arm_vidupq_m_wb_u32(uint32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_wb_u32)))
+uint32x4_t __arm_vidupq_m(uint32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_wb_u8)))
+uint8x16_t __arm_vidupq_m_wb_u8(uint8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_wb_u8)))
+uint8x16_t __arm_vidupq_m(uint8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_n_u16)))
+uint16x8_t __arm_vidupq_n_u16(uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_n_u16)))
+uint16x8_t __arm_vidupq_u16(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_n_u32)))
+uint32x4_t __arm_vidupq_n_u32(uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_n_u32)))
+uint32x4_t __arm_vidupq_u32(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_n_u8)))
+uint8x16_t __arm_vidupq_n_u8(uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_n_u8)))
+uint8x16_t __arm_vidupq_u8(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_wb_u16)))
+uint16x8_t __arm_vidupq_wb_u16(uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_wb_u16)))
+uint16x8_t __arm_vidupq_u16(uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_wb_u32)))
+uint32x4_t __arm_vidupq_wb_u32(uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_wb_u32)))
+uint32x4_t __arm_vidupq_u32(uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_wb_u8)))
+uint8x16_t __arm_vidupq_wb_u8(uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_wb_u8)))
+uint8x16_t __arm_vidupq_u8(uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_n_u16)))
+uint16x8_t __arm_vidupq_x_n_u16(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_n_u16)))
+uint16x8_t __arm_vidupq_x_u16(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_n_u32)))
+uint32x4_t __arm_vidupq_x_n_u32(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_n_u32)))
+uint32x4_t __arm_vidupq_x_u32(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_n_u8)))
+uint8x16_t __arm_vidupq_x_n_u8(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_n_u8)))
+uint8x16_t __arm_vidupq_x_u8(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_wb_u16)))
+uint16x8_t __arm_vidupq_x_wb_u16(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_wb_u16)))
+uint16x8_t __arm_vidupq_x_u16(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_wb_u32)))
+uint32x4_t __arm_vidupq_x_wb_u32(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_wb_u32)))
+uint32x4_t __arm_vidupq_x_u32(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_wb_u8)))
+uint8x16_t __arm_vidupq_x_wb_u8(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_wb_u8)))
+uint8x16_t __arm_vidupq_x_u8(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_n_u16)))
+uint16x8_t __arm_viwdupq_m_n_u16(uint16x8_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_n_u16)))
+uint16x8_t __arm_viwdupq_m(uint16x8_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_n_u32)))
+uint32x4_t __arm_viwdupq_m_n_u32(uint32x4_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_n_u32)))
+uint32x4_t __arm_viwdupq_m(uint32x4_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_n_u8)))
+uint8x16_t __arm_viwdupq_m_n_u8(uint8x16_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_n_u8)))
+uint8x16_t __arm_viwdupq_m(uint8x16_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_wb_u16)))
+uint16x8_t __arm_viwdupq_m_wb_u16(uint16x8_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_wb_u16)))
+uint16x8_t __arm_viwdupq_m(uint16x8_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_wb_u32)))
+uint32x4_t __arm_viwdupq_m_wb_u32(uint32x4_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_wb_u32)))
+uint32x4_t __arm_viwdupq_m(uint32x4_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_wb_u8)))
+uint8x16_t __arm_viwdupq_m_wb_u8(uint8x16_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_wb_u8)))
+uint8x16_t __arm_viwdupq_m(uint8x16_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_n_u16)))
+uint16x8_t __arm_viwdupq_n_u16(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_n_u16)))
+uint16x8_t __arm_viwdupq_u16(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_n_u32)))
+uint32x4_t __arm_viwdupq_n_u32(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_n_u32)))
+uint32x4_t __arm_viwdupq_u32(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_n_u8)))
+uint8x16_t __arm_viwdupq_n_u8(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_n_u8)))
+uint8x16_t __arm_viwdupq_u8(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_wb_u16)))
+uint16x8_t __arm_viwdupq_wb_u16(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_wb_u16)))
+uint16x8_t __arm_viwdupq_u16(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_wb_u32)))
+uint32x4_t __arm_viwdupq_wb_u32(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_wb_u32)))
+uint32x4_t __arm_viwdupq_u32(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_wb_u8)))
+uint8x16_t __arm_viwdupq_wb_u8(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_wb_u8)))
+uint8x16_t __arm_viwdupq_u8(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_n_u16)))
+uint16x8_t __arm_viwdupq_x_n_u16(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_n_u16)))
+uint16x8_t __arm_viwdupq_x_u16(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_n_u32)))
+uint32x4_t __arm_viwdupq_x_n_u32(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_n_u32)))
+uint32x4_t __arm_viwdupq_x_u32(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_n_u8)))
+uint8x16_t __arm_viwdupq_x_n_u8(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_n_u8)))
+uint8x16_t __arm_viwdupq_x_u8(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_wb_u16)))
+uint16x8_t __arm_viwdupq_x_wb_u16(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_wb_u16)))
+uint16x8_t __arm_viwdupq_x_u16(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_wb_u32)))
+uint32x4_t __arm_viwdupq_x_wb_u32(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_wb_u32)))
+uint32x4_t __arm_viwdupq_x_u32(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_wb_u8)))
+uint8x16_t __arm_viwdupq_x_wb_u8(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_wb_u8)))
+uint8x16_t __arm_viwdupq_x_u8(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_s16)))
+int16x8_t __arm_vld1q_s16(const int16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_s16)))
+int16x8_t __arm_vld1q(const int16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_s32)))
+int32x4_t __arm_vld1q_s32(const int32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_s32)))
+int32x4_t __arm_vld1q(const int32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_s8)))
+int8x16_t __arm_vld1q_s8(const int8_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_s8)))
+int8x16_t __arm_vld1q(const int8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_u16)))
+uint16x8_t __arm_vld1q_u16(const uint16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_u16)))
+uint16x8_t __arm_vld1q(const uint16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_u32)))
+uint32x4_t __arm_vld1q_u32(const uint32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_u32)))
+uint32x4_t __arm_vld1q(const uint32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_u8)))
+uint8x16_t __arm_vld1q_u8(const uint8_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_u8)))
+uint8x16_t __arm_vld1q(const uint8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_s16)))
+int16x8_t __arm_vld1q_z_s16(const int16_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_s16)))
+int16x8_t __arm_vld1q_z(const int16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_s32)))
+int32x4_t __arm_vld1q_z_s32(const int32_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_s32)))
+int32x4_t __arm_vld1q_z(const int32_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_s8)))
+int8x16_t __arm_vld1q_z_s8(const int8_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_s8)))
+int8x16_t __arm_vld1q_z(const int8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_u16)))
+uint16x8_t __arm_vld1q_z_u16(const uint16_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_u16)))
+uint16x8_t __arm_vld1q_z(const uint16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_u32)))
+uint32x4_t __arm_vld1q_z_u32(const uint32_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_u32)))
+uint32x4_t __arm_vld1q_z(const uint32_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_u8)))
+uint8x16_t __arm_vld1q_z_u8(const uint8_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_u8)))
+uint8x16_t __arm_vld1q_z(const uint8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_s16)))
+int16x8x2_t __arm_vld2q_s16(const int16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_s16)))
+int16x8x2_t __arm_vld2q(const int16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_s32)))
+int32x4x2_t __arm_vld2q_s32(const int32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_s32)))
+int32x4x2_t __arm_vld2q(const int32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_s8)))
+int8x16x2_t __arm_vld2q_s8(const int8_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_s8)))
+int8x16x2_t __arm_vld2q(const int8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_u16)))
+uint16x8x2_t __arm_vld2q_u16(const uint16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_u16)))
+uint16x8x2_t __arm_vld2q(const uint16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_u32)))
+uint32x4x2_t __arm_vld2q_u32(const uint32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_u32)))
+uint32x4x2_t __arm_vld2q(const uint32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_u8)))
+uint8x16x2_t __arm_vld2q_u8(const uint8_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_u8)))
+uint8x16x2_t __arm_vld2q(const uint8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_s16)))
+int16x8x4_t __arm_vld4q_s16(const int16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_s16)))
+int16x8x4_t __arm_vld4q(const int16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_s32)))
+int32x4x4_t __arm_vld4q_s32(const int32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_s32)))
+int32x4x4_t __arm_vld4q(const int32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_s8)))
+int8x16x4_t __arm_vld4q_s8(const int8_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_s8)))
+int8x16x4_t __arm_vld4q(const int8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_u16)))
+uint16x8x4_t __arm_vld4q_u16(const uint16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_u16)))
+uint16x8x4_t __arm_vld4q(const uint16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_u32)))
+uint32x4x4_t __arm_vld4q_u32(const uint32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_u32)))
+uint32x4x4_t __arm_vld4q(const uint32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_u8)))
+uint8x16x4_t __arm_vld4q_u8(const uint8_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_u8)))
+uint8x16x4_t __arm_vld4q(const uint8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_s16)))
+int16x8_t __arm_vldrbq_gather_offset_s16(const int8_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_s16)))
+int16x8_t __arm_vldrbq_gather_offset(const int8_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_s32)))
+int32x4_t __arm_vldrbq_gather_offset_s32(const int8_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_s32)))
+int32x4_t __arm_vldrbq_gather_offset(const int8_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_s8)))
+int8x16_t __arm_vldrbq_gather_offset_s8(const int8_t *, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_s8)))
+int8x16_t __arm_vldrbq_gather_offset(const int8_t *, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_u16)))
+uint16x8_t __arm_vldrbq_gather_offset_u16(const uint8_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_u16)))
+uint16x8_t __arm_vldrbq_gather_offset(const uint8_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_u32)))
+uint32x4_t __arm_vldrbq_gather_offset_u32(const uint8_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_u32)))
+uint32x4_t __arm_vldrbq_gather_offset(const uint8_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_u8)))
+uint8x16_t __arm_vldrbq_gather_offset_u8(const uint8_t *, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_u8)))
+uint8x16_t __arm_vldrbq_gather_offset(const uint8_t *, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_s16)))
+int16x8_t __arm_vldrbq_gather_offset_z_s16(const int8_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_s16)))
+int16x8_t __arm_vldrbq_gather_offset_z(const int8_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_s32)))
+int32x4_t __arm_vldrbq_gather_offset_z_s32(const int8_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_s32)))
+int32x4_t __arm_vldrbq_gather_offset_z(const int8_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_s8)))
+int8x16_t __arm_vldrbq_gather_offset_z_s8(const int8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_s8)))
+int8x16_t __arm_vldrbq_gather_offset_z(const int8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_u16)))
+uint16x8_t __arm_vldrbq_gather_offset_z_u16(const uint8_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_u16)))
+uint16x8_t __arm_vldrbq_gather_offset_z(const uint8_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_u32)))
+uint32x4_t __arm_vldrbq_gather_offset_z_u32(const uint8_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_u32)))
+uint32x4_t __arm_vldrbq_gather_offset_z(const uint8_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_u8)))
+uint8x16_t __arm_vldrbq_gather_offset_z_u8(const uint8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_u8)))
+uint8x16_t __arm_vldrbq_gather_offset_z(const uint8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_s16)))
+int16x8_t __arm_vldrbq_s16(const int8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_s32)))
+int32x4_t __arm_vldrbq_s32(const int8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_s8)))
+int8x16_t __arm_vldrbq_s8(const int8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_u16)))
+uint16x8_t __arm_vldrbq_u16(const uint8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_u32)))
+uint32x4_t __arm_vldrbq_u32(const uint8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_u8)))
+uint8x16_t __arm_vldrbq_u8(const uint8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_z_s16)))
+int16x8_t __arm_vldrbq_z_s16(const int8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_z_s32)))
+int32x4_t __arm_vldrbq_z_s32(const int8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_z_s8)))
+int8x16_t __arm_vldrbq_z_s8(const int8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_z_u16)))
+uint16x8_t __arm_vldrbq_z_u16(const uint8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_z_u32)))
+uint32x4_t __arm_vldrbq_z_u32(const uint8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_z_u8)))
+uint8x16_t __arm_vldrbq_z_u8(const uint8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_s64)))
+int64x2_t __arm_vldrdq_gather_base_s64(uint64x2_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_u64)))
+uint64x2_t __arm_vldrdq_gather_base_u64(uint64x2_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_wb_s64)))
+int64x2_t __arm_vldrdq_gather_base_wb_s64(uint64x2_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_wb_u64)))
+uint64x2_t __arm_vldrdq_gather_base_wb_u64(uint64x2_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_wb_z_s64)))
+int64x2_t __arm_vldrdq_gather_base_wb_z_s64(uint64x2_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_wb_z_u64)))
+uint64x2_t __arm_vldrdq_gather_base_wb_z_u64(uint64x2_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_z_s64)))
+int64x2_t __arm_vldrdq_gather_base_z_s64(uint64x2_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_z_u64)))
+uint64x2_t __arm_vldrdq_gather_base_z_u64(uint64x2_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_s64)))
+int64x2_t __arm_vldrdq_gather_offset_s64(const int64_t *, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_s64)))
+int64x2_t __arm_vldrdq_gather_offset(const int64_t *, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_u64)))
+uint64x2_t __arm_vldrdq_gather_offset_u64(const uint64_t *, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_u64)))
+uint64x2_t __arm_vldrdq_gather_offset(const uint64_t *, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_z_s64)))
+int64x2_t __arm_vldrdq_gather_offset_z_s64(const int64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_z_s64)))
+int64x2_t __arm_vldrdq_gather_offset_z(const int64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_z_u64)))
+uint64x2_t __arm_vldrdq_gather_offset_z_u64(const uint64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_z_u64)))
+uint64x2_t __arm_vldrdq_gather_offset_z(const uint64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_s64)))
+int64x2_t __arm_vldrdq_gather_shifted_offset_s64(const int64_t *, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_s64)))
+int64x2_t __arm_vldrdq_gather_shifted_offset(const int64_t *, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_u64)))
+uint64x2_t __arm_vldrdq_gather_shifted_offset_u64(const uint64_t *, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_u64)))
+uint64x2_t __arm_vldrdq_gather_shifted_offset(const uint64_t *, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_z_s64)))
+int64x2_t __arm_vldrdq_gather_shifted_offset_z_s64(const int64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_z_s64)))
+int64x2_t __arm_vldrdq_gather_shifted_offset_z(const int64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_z_u64)))
+uint64x2_t __arm_vldrdq_gather_shifted_offset_z_u64(const uint64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_z_u64)))
+uint64x2_t __arm_vldrdq_gather_shifted_offset_z(const uint64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_s16)))
+int16x8_t __arm_vldrhq_gather_offset_s16(const int16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_s16)))
+int16x8_t __arm_vldrhq_gather_offset(const int16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_s32)))
+int32x4_t __arm_vldrhq_gather_offset_s32(const int16_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_s32)))
+int32x4_t __arm_vldrhq_gather_offset(const int16_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_u16)))
+uint16x8_t __arm_vldrhq_gather_offset_u16(const uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_u16)))
+uint16x8_t __arm_vldrhq_gather_offset(const uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_u32)))
+uint32x4_t __arm_vldrhq_gather_offset_u32(const uint16_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_u32)))
+uint32x4_t __arm_vldrhq_gather_offset(const uint16_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_s16)))
+int16x8_t __arm_vldrhq_gather_offset_z_s16(const int16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_s16)))
+int16x8_t __arm_vldrhq_gather_offset_z(const int16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_s32)))
+int32x4_t __arm_vldrhq_gather_offset_z_s32(const int16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_s32)))
+int32x4_t __arm_vldrhq_gather_offset_z(const int16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_u16)))
+uint16x8_t __arm_vldrhq_gather_offset_z_u16(const uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_u16)))
+uint16x8_t __arm_vldrhq_gather_offset_z(const uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_u32)))
+uint32x4_t __arm_vldrhq_gather_offset_z_u32(const uint16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_u32)))
+uint32x4_t __arm_vldrhq_gather_offset_z(const uint16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_s16)))
+int16x8_t __arm_vldrhq_gather_shifted_offset_s16(const int16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_s16)))
+int16x8_t __arm_vldrhq_gather_shifted_offset(const int16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_s32)))
+int32x4_t __arm_vldrhq_gather_shifted_offset_s32(const int16_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_s32)))
+int32x4_t __arm_vldrhq_gather_shifted_offset(const int16_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_u16)))
+uint16x8_t __arm_vldrhq_gather_shifted_offset_u16(const uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_u16)))
+uint16x8_t __arm_vldrhq_gather_shifted_offset(const uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_u32)))
+uint32x4_t __arm_vldrhq_gather_shifted_offset_u32(const uint16_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_u32)))
+uint32x4_t __arm_vldrhq_gather_shifted_offset(const uint16_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_s16)))
+int16x8_t __arm_vldrhq_gather_shifted_offset_z_s16(const int16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_s16)))
+int16x8_t __arm_vldrhq_gather_shifted_offset_z(const int16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_s32)))
+int32x4_t __arm_vldrhq_gather_shifted_offset_z_s32(const int16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_s32)))
+int32x4_t __arm_vldrhq_gather_shifted_offset_z(const int16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_u16)))
+uint16x8_t __arm_vldrhq_gather_shifted_offset_z_u16(const uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_u16)))
+uint16x8_t __arm_vldrhq_gather_shifted_offset_z(const uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_u32)))
+uint32x4_t __arm_vldrhq_gather_shifted_offset_z_u32(const uint16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_u32)))
+uint32x4_t __arm_vldrhq_gather_shifted_offset_z(const uint16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_s16)))
+int16x8_t __arm_vldrhq_s16(const int16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_s32)))
+int32x4_t __arm_vldrhq_s32(const int16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_u16)))
+uint16x8_t __arm_vldrhq_u16(const uint16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_u32)))
+uint32x4_t __arm_vldrhq_u32(const uint16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_z_s16)))
+int16x8_t __arm_vldrhq_z_s16(const int16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_z_s32)))
+int32x4_t __arm_vldrhq_z_s32(const int16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_z_u16)))
+uint16x8_t __arm_vldrhq_z_u16(const uint16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_z_u32)))
+uint32x4_t __arm_vldrhq_z_u32(const uint16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_s32)))
+int32x4_t __arm_vldrwq_gather_base_s32(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_u32)))
+uint32x4_t __arm_vldrwq_gather_base_u32(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_wb_s32)))
+int32x4_t __arm_vldrwq_gather_base_wb_s32(uint32x4_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_wb_u32)))
+uint32x4_t __arm_vldrwq_gather_base_wb_u32(uint32x4_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_wb_z_s32)))
+int32x4_t __arm_vldrwq_gather_base_wb_z_s32(uint32x4_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_wb_z_u32)))
+uint32x4_t __arm_vldrwq_gather_base_wb_z_u32(uint32x4_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_z_s32)))
+int32x4_t __arm_vldrwq_gather_base_z_s32(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_z_u32)))
+uint32x4_t __arm_vldrwq_gather_base_z_u32(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_s32)))
+int32x4_t __arm_vldrwq_gather_offset_s32(const int32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_s32)))
+int32x4_t __arm_vldrwq_gather_offset(const int32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_u32)))
+uint32x4_t __arm_vldrwq_gather_offset_u32(const uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_u32)))
+uint32x4_t __arm_vldrwq_gather_offset(const uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_z_s32)))
+int32x4_t __arm_vldrwq_gather_offset_z_s32(const int32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_z_s32)))
+int32x4_t __arm_vldrwq_gather_offset_z(const int32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_z_u32)))
+uint32x4_t __arm_vldrwq_gather_offset_z_u32(const uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_z_u32)))
+uint32x4_t __arm_vldrwq_gather_offset_z(const uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_s32)))
+int32x4_t __arm_vldrwq_gather_shifted_offset_s32(const int32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_s32)))
+int32x4_t __arm_vldrwq_gather_shifted_offset(const int32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_u32)))
+uint32x4_t __arm_vldrwq_gather_shifted_offset_u32(const uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_u32)))
+uint32x4_t __arm_vldrwq_gather_shifted_offset(const uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_z_s32)))
+int32x4_t __arm_vldrwq_gather_shifted_offset_z_s32(const int32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_z_s32)))
+int32x4_t __arm_vldrwq_gather_shifted_offset_z(const int32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_z_u32)))
+uint32x4_t __arm_vldrwq_gather_shifted_offset_z_u32(const uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_z_u32)))
+uint32x4_t __arm_vldrwq_gather_shifted_offset_z(const uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_s32)))
+int32x4_t __arm_vldrwq_s32(const int32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_u32)))
+uint32x4_t __arm_vldrwq_u32(const uint32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_z_s32)))
+int32x4_t __arm_vldrwq_z_s32(const int32_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_z_u32)))
+uint32x4_t __arm_vldrwq_z_u32(const uint32_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_m_s16)))
+uint16x8_t __arm_vmaxaq_m_s16(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_m_s16)))
+uint16x8_t __arm_vmaxaq_m(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_m_s32)))
+uint32x4_t __arm_vmaxaq_m_s32(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_m_s32)))
+uint32x4_t __arm_vmaxaq_m(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_m_s8)))
+uint8x16_t __arm_vmaxaq_m_s8(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_m_s8)))
+uint8x16_t __arm_vmaxaq_m(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_s16)))
+uint16x8_t __arm_vmaxaq_s16(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_s16)))
+uint16x8_t __arm_vmaxaq(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_s32)))
+uint32x4_t __arm_vmaxaq_s32(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_s32)))
+uint32x4_t __arm_vmaxaq(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_s8)))
+uint8x16_t __arm_vmaxaq_s8(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_s8)))
+uint8x16_t __arm_vmaxaq(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_p_s16)))
+uint16_t __arm_vmaxavq_p_s16(uint16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_p_s16)))
+uint16_t __arm_vmaxavq_p(uint16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_p_s32)))
+uint32_t __arm_vmaxavq_p_s32(uint32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_p_s32)))
+uint32_t __arm_vmaxavq_p(uint32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_p_s8)))
+uint8_t __arm_vmaxavq_p_s8(uint8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_p_s8)))
+uint8_t __arm_vmaxavq_p(uint8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_s16)))
+uint16_t __arm_vmaxavq_s16(uint16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_s16)))
+uint16_t __arm_vmaxavq(uint16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_s32)))
+uint32_t __arm_vmaxavq_s32(uint32_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_s32)))
+uint32_t __arm_vmaxavq(uint32_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_s8)))
+uint8_t __arm_vmaxavq_s8(uint8_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_s8)))
+uint8_t __arm_vmaxavq(uint8_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_s16)))
+int16x8_t __arm_vmaxq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_s16)))
+int16x8_t __arm_vmaxq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_s32)))
+int32x4_t __arm_vmaxq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_s32)))
+int32x4_t __arm_vmaxq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_s8)))
+int8x16_t __arm_vmaxq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_s8)))
+int8x16_t __arm_vmaxq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_u16)))
+uint16x8_t __arm_vmaxq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_u16)))
+uint16x8_t __arm_vmaxq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_u32)))
+uint32x4_t __arm_vmaxq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_u32)))
+uint32x4_t __arm_vmaxq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_u8)))
+uint8x16_t __arm_vmaxq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_u8)))
+uint8x16_t __arm_vmaxq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_s16)))
+int16x8_t __arm_vmaxq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_s16)))
+int16x8_t __arm_vmaxq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_s32)))
+int32x4_t __arm_vmaxq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_s32)))
+int32x4_t __arm_vmaxq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_s8)))
+int8x16_t __arm_vmaxq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_s8)))
+int8x16_t __arm_vmaxq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_u16)))
+uint16x8_t __arm_vmaxq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_u16)))
+uint16x8_t __arm_vmaxq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_u32)))
+uint32x4_t __arm_vmaxq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_u32)))
+uint32x4_t __arm_vmaxq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_u8)))
+uint8x16_t __arm_vmaxq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_u8)))
+uint8x16_t __arm_vmaxq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_s16)))
+int16x8_t __arm_vmaxq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_s16)))
+int16x8_t __arm_vmaxq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_s32)))
+int32x4_t __arm_vmaxq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_s32)))
+int32x4_t __arm_vmaxq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_s8)))
+int8x16_t __arm_vmaxq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_s8)))
+int8x16_t __arm_vmaxq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_u16)))
+uint16x8_t __arm_vmaxq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_u16)))
+uint16x8_t __arm_vmaxq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_u32)))
+uint32x4_t __arm_vmaxq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_u32)))
+uint32x4_t __arm_vmaxq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_u8)))
+uint8x16_t __arm_vmaxq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_u8)))
+uint8x16_t __arm_vmaxq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_s16)))
+int16_t __arm_vmaxvq_p_s16(int16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_s16)))
+int16_t __arm_vmaxvq_p(int16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_s32)))
+int32_t __arm_vmaxvq_p_s32(int32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_s32)))
+int32_t __arm_vmaxvq_p(int32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_s8)))
+int8_t __arm_vmaxvq_p_s8(int8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_s8)))
+int8_t __arm_vmaxvq_p(int8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_u16)))
+uint16_t __arm_vmaxvq_p_u16(uint16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_u16)))
+uint16_t __arm_vmaxvq_p(uint16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_u32)))
+uint32_t __arm_vmaxvq_p_u32(uint32_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_u32)))
+uint32_t __arm_vmaxvq_p(uint32_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_u8)))
+uint8_t __arm_vmaxvq_p_u8(uint8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_u8)))
+uint8_t __arm_vmaxvq_p(uint8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_s16)))
+int16_t __arm_vmaxvq_s16(int16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_s16)))
+int16_t __arm_vmaxvq(int16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_s32)))
+int32_t __arm_vmaxvq_s32(int32_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_s32)))
+int32_t __arm_vmaxvq(int32_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_s8)))
+int8_t __arm_vmaxvq_s8(int8_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_s8)))
+int8_t __arm_vmaxvq(int8_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_u16)))
+uint16_t __arm_vmaxvq_u16(uint16_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_u16)))
+uint16_t __arm_vmaxvq(uint16_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_u32)))
+uint32_t __arm_vmaxvq_u32(uint32_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_u32)))
+uint32_t __arm_vmaxvq(uint32_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_u8)))
+uint8_t __arm_vmaxvq_u8(uint8_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_u8)))
+uint8_t __arm_vmaxvq(uint8_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminaq_m_s16)))
+uint16x8_t __arm_vminaq_m_s16(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminaq_m_s16)))
+uint16x8_t __arm_vminaq_m(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminaq_m_s32)))
+uint32x4_t __arm_vminaq_m_s32(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminaq_m_s32)))
+uint32x4_t __arm_vminaq_m(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminaq_m_s8)))
+uint8x16_t __arm_vminaq_m_s8(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminaq_m_s8)))
+uint8x16_t __arm_vminaq_m(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminaq_s16)))
+uint16x8_t __arm_vminaq_s16(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminaq_s16)))
+uint16x8_t __arm_vminaq(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminaq_s32)))
+uint32x4_t __arm_vminaq_s32(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminaq_s32)))
+uint32x4_t __arm_vminaq(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminaq_s8)))
+uint8x16_t __arm_vminaq_s8(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminaq_s8)))
+uint8x16_t __arm_vminaq(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminavq_p_s16)))
+uint16_t __arm_vminavq_p_s16(uint16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminavq_p_s16)))
+uint16_t __arm_vminavq_p(uint16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminavq_p_s32)))
+uint32_t __arm_vminavq_p_s32(uint32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminavq_p_s32)))
+uint32_t __arm_vminavq_p(uint32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminavq_p_s8)))
+uint8_t __arm_vminavq_p_s8(uint8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminavq_p_s8)))
+uint8_t __arm_vminavq_p(uint8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminavq_s16)))
+uint16_t __arm_vminavq_s16(uint16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminavq_s16)))
+uint16_t __arm_vminavq(uint16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminavq_s32)))
+uint32_t __arm_vminavq_s32(uint32_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminavq_s32)))
+uint32_t __arm_vminavq(uint32_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminavq_s8)))
+uint8_t __arm_vminavq_s8(uint8_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminavq_s8)))
+uint8_t __arm_vminavq(uint8_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_s16)))
+int16x8_t __arm_vminq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_s16)))
+int16x8_t __arm_vminq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_s32)))
+int32x4_t __arm_vminq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_s32)))
+int32x4_t __arm_vminq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_s8)))
+int8x16_t __arm_vminq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_s8)))
+int8x16_t __arm_vminq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_u16)))
+uint16x8_t __arm_vminq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_u16)))
+uint16x8_t __arm_vminq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_u32)))
+uint32x4_t __arm_vminq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_u32)))
+uint32x4_t __arm_vminq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_u8)))
+uint8x16_t __arm_vminq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_u8)))
+uint8x16_t __arm_vminq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_s16)))
+int16x8_t __arm_vminq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_s16)))
+int16x8_t __arm_vminq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_s32)))
+int32x4_t __arm_vminq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_s32)))
+int32x4_t __arm_vminq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_s8)))
+int8x16_t __arm_vminq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_s8)))
+int8x16_t __arm_vminq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_u16)))
+uint16x8_t __arm_vminq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_u16)))
+uint16x8_t __arm_vminq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_u32)))
+uint32x4_t __arm_vminq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_u32)))
+uint32x4_t __arm_vminq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_u8)))
+uint8x16_t __arm_vminq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_u8)))
+uint8x16_t __arm_vminq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_s16)))
+int16x8_t __arm_vminq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_s16)))
+int16x8_t __arm_vminq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_s32)))
+int32x4_t __arm_vminq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_s32)))
+int32x4_t __arm_vminq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_s8)))
+int8x16_t __arm_vminq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_s8)))
+int8x16_t __arm_vminq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_u16)))
+uint16x8_t __arm_vminq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_u16)))
+uint16x8_t __arm_vminq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_u32)))
+uint32x4_t __arm_vminq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_u32)))
+uint32x4_t __arm_vminq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_u8)))
+uint8x16_t __arm_vminq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_u8)))
+uint8x16_t __arm_vminq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_s16)))
+int16_t __arm_vminvq_p_s16(int16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_s16)))
+int16_t __arm_vminvq_p(int16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_s32)))
+int32_t __arm_vminvq_p_s32(int32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_s32)))
+int32_t __arm_vminvq_p(int32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_s8)))
+int8_t __arm_vminvq_p_s8(int8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_s8)))
+int8_t __arm_vminvq_p(int8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_u16)))
+uint16_t __arm_vminvq_p_u16(uint16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_u16)))
+uint16_t __arm_vminvq_p(uint16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_u32)))
+uint32_t __arm_vminvq_p_u32(uint32_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_u32)))
+uint32_t __arm_vminvq_p(uint32_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_u8)))
+uint8_t __arm_vminvq_p_u8(uint8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_u8)))
+uint8_t __arm_vminvq_p(uint8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_s16)))
+int16_t __arm_vminvq_s16(int16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_s16)))
+int16_t __arm_vminvq(int16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_s32)))
+int32_t __arm_vminvq_s32(int32_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_s32)))
+int32_t __arm_vminvq(int32_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_s8)))
+int8_t __arm_vminvq_s8(int8_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_s8)))
+int8_t __arm_vminvq(int8_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_u16)))
+uint16_t __arm_vminvq_u16(uint16_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_u16)))
+uint16_t __arm_vminvq(uint16_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_u32)))
+uint32_t __arm_vminvq_u32(uint32_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_u32)))
+uint32_t __arm_vminvq(uint32_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_u8)))
+uint8_t __arm_vminvq_u8(uint8_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_u8)))
+uint8_t __arm_vminvq(uint8_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_s16)))
+int32_t __arm_vmladavaq_p_s16(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_s16)))
+int32_t __arm_vmladavaq_p(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_s32)))
+int32_t __arm_vmladavaq_p_s32(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_s32)))
+int32_t __arm_vmladavaq_p(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_s8)))
+int32_t __arm_vmladavaq_p_s8(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_s8)))
+int32_t __arm_vmladavaq_p(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_u16)))
+uint32_t __arm_vmladavaq_p_u16(uint32_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_u16)))
+uint32_t __arm_vmladavaq_p(uint32_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_u32)))
+uint32_t __arm_vmladavaq_p_u32(uint32_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_u32)))
+uint32_t __arm_vmladavaq_p(uint32_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_u8)))
+uint32_t __arm_vmladavaq_p_u8(uint32_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_u8)))
+uint32_t __arm_vmladavaq_p(uint32_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_s16)))
+int32_t __arm_vmladavaq_s16(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_s16)))
+int32_t __arm_vmladavaq(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_s32)))
+int32_t __arm_vmladavaq_s32(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_s32)))
+int32_t __arm_vmladavaq(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_s8)))
+int32_t __arm_vmladavaq_s8(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_s8)))
+int32_t __arm_vmladavaq(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_u16)))
+uint32_t __arm_vmladavaq_u16(uint32_t, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_u16)))
+uint32_t __arm_vmladavaq(uint32_t, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_u32)))
+uint32_t __arm_vmladavaq_u32(uint32_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_u32)))
+uint32_t __arm_vmladavaq(uint32_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_u8)))
+uint32_t __arm_vmladavaq_u8(uint32_t, uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_u8)))
+uint32_t __arm_vmladavaq(uint32_t, uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_p_s16)))
+int32_t __arm_vmladavaxq_p_s16(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_p_s16)))
+int32_t __arm_vmladavaxq_p(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_p_s32)))
+int32_t __arm_vmladavaxq_p_s32(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_p_s32)))
+int32_t __arm_vmladavaxq_p(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_p_s8)))
+int32_t __arm_vmladavaxq_p_s8(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_p_s8)))
+int32_t __arm_vmladavaxq_p(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_s16)))
+int32_t __arm_vmladavaxq_s16(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_s16)))
+int32_t __arm_vmladavaxq(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_s32)))
+int32_t __arm_vmladavaxq_s32(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_s32)))
+int32_t __arm_vmladavaxq(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_s8)))
+int32_t __arm_vmladavaxq_s8(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_s8)))
+int32_t __arm_vmladavaxq(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_s16)))
+int32_t __arm_vmladavq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_s16)))
+int32_t __arm_vmladavq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_s32)))
+int32_t __arm_vmladavq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_s32)))
+int32_t __arm_vmladavq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_s8)))
+int32_t __arm_vmladavq_p_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_s8)))
+int32_t __arm_vmladavq_p(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_u16)))
+uint32_t __arm_vmladavq_p_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_u16)))
+uint32_t __arm_vmladavq_p(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_u32)))
+uint32_t __arm_vmladavq_p_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_u32)))
+uint32_t __arm_vmladavq_p(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_u8)))
+uint32_t __arm_vmladavq_p_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_u8)))
+uint32_t __arm_vmladavq_p(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_s16)))
+int32_t __arm_vmladavq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_s16)))
+int32_t __arm_vmladavq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_s32)))
+int32_t __arm_vmladavq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_s32)))
+int32_t __arm_vmladavq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_s8)))
+int32_t __arm_vmladavq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_s8)))
+int32_t __arm_vmladavq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_u16)))
+uint32_t __arm_vmladavq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_u16)))
+uint32_t __arm_vmladavq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_u32)))
+uint32_t __arm_vmladavq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_u32)))
+uint32_t __arm_vmladavq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_u8)))
+uint32_t __arm_vmladavq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_u8)))
+uint32_t __arm_vmladavq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_p_s16)))
+int32_t __arm_vmladavxq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_p_s16)))
+int32_t __arm_vmladavxq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_p_s32)))
+int32_t __arm_vmladavxq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_p_s32)))
+int32_t __arm_vmladavxq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_p_s8)))
+int32_t __arm_vmladavxq_p_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_p_s8)))
+int32_t __arm_vmladavxq_p(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_s16)))
+int32_t __arm_vmladavxq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_s16)))
+int32_t __arm_vmladavxq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_s32)))
+int32_t __arm_vmladavxq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_s32)))
+int32_t __arm_vmladavxq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_s8)))
+int32_t __arm_vmladavxq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_s8)))
+int32_t __arm_vmladavxq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_s16)))
+int64_t __arm_vmlaldavaq_p_s16(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_s16)))
+int64_t __arm_vmlaldavaq_p(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_s32)))
+int64_t __arm_vmlaldavaq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_s32)))
+int64_t __arm_vmlaldavaq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_u16)))
+uint64_t __arm_vmlaldavaq_p_u16(uint64_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_u16)))
+uint64_t __arm_vmlaldavaq_p(uint64_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_u32)))
+uint64_t __arm_vmlaldavaq_p_u32(uint64_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_u32)))
+uint64_t __arm_vmlaldavaq_p(uint64_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_s16)))
+int64_t __arm_vmlaldavaq_s16(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_s16)))
+int64_t __arm_vmlaldavaq(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_s32)))
+int64_t __arm_vmlaldavaq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_s32)))
+int64_t __arm_vmlaldavaq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_u16)))
+uint64_t __arm_vmlaldavaq_u16(uint64_t, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_u16)))
+uint64_t __arm_vmlaldavaq(uint64_t, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_u32)))
+uint64_t __arm_vmlaldavaq_u32(uint64_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_u32)))
+uint64_t __arm_vmlaldavaq(uint64_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_p_s16)))
+int64_t __arm_vmlaldavaxq_p_s16(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_p_s16)))
+int64_t __arm_vmlaldavaxq_p(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_p_s32)))
+int64_t __arm_vmlaldavaxq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_p_s32)))
+int64_t __arm_vmlaldavaxq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_s16)))
+int64_t __arm_vmlaldavaxq_s16(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_s16)))
+int64_t __arm_vmlaldavaxq(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_s32)))
+int64_t __arm_vmlaldavaxq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_s32)))
+int64_t __arm_vmlaldavaxq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_s16)))
+int64_t __arm_vmlaldavq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_s16)))
+int64_t __arm_vmlaldavq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_s32)))
+int64_t __arm_vmlaldavq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_s32)))
+int64_t __arm_vmlaldavq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_u16)))
+uint64_t __arm_vmlaldavq_p_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_u16)))
+uint64_t __arm_vmlaldavq_p(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_u32)))
+uint64_t __arm_vmlaldavq_p_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_u32)))
+uint64_t __arm_vmlaldavq_p(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_s16)))
+int64_t __arm_vmlaldavq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_s16)))
+int64_t __arm_vmlaldavq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_s32)))
+int64_t __arm_vmlaldavq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_s32)))
+int64_t __arm_vmlaldavq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_u16)))
+uint64_t __arm_vmlaldavq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_u16)))
+uint64_t __arm_vmlaldavq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_u32)))
+uint64_t __arm_vmlaldavq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_u32)))
+uint64_t __arm_vmlaldavq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_p_s16)))
+int64_t __arm_vmlaldavxq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_p_s16)))
+int64_t __arm_vmlaldavxq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_p_s32)))
+int64_t __arm_vmlaldavxq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_p_s32)))
+int64_t __arm_vmlaldavxq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_s16)))
+int64_t __arm_vmlaldavxq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_s16)))
+int64_t __arm_vmlaldavxq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_s32)))
+int64_t __arm_vmlaldavxq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_s32)))
+int64_t __arm_vmlaldavxq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_s16)))
+int16x8_t __arm_vmlaq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_s16)))
+int16x8_t __arm_vmlaq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_s32)))
+int32x4_t __arm_vmlaq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_s32)))
+int32x4_t __arm_vmlaq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_s8)))
+int8x16_t __arm_vmlaq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_s8)))
+int8x16_t __arm_vmlaq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_u16)))
+uint16x8_t __arm_vmlaq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_u16)))
+uint16x8_t __arm_vmlaq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_u32)))
+uint32x4_t __arm_vmlaq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_u32)))
+uint32x4_t __arm_vmlaq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_u8)))
+uint8x16_t __arm_vmlaq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_u8)))
+uint8x16_t __arm_vmlaq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_s16)))
+int16x8_t __arm_vmlaq_n_s16(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_s16)))
+int16x8_t __arm_vmlaq(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_s32)))
+int32x4_t __arm_vmlaq_n_s32(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_s32)))
+int32x4_t __arm_vmlaq(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_s8)))
+int8x16_t __arm_vmlaq_n_s8(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_s8)))
+int8x16_t __arm_vmlaq(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_u16)))
+uint16x8_t __arm_vmlaq_n_u16(uint16x8_t, uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_u16)))
+uint16x8_t __arm_vmlaq(uint16x8_t, uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_u32)))
+uint32x4_t __arm_vmlaq_n_u32(uint32x4_t, uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_u32)))
+uint32x4_t __arm_vmlaq(uint32x4_t, uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_u8)))
+uint8x16_t __arm_vmlaq_n_u8(uint8x16_t, uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_u8)))
+uint8x16_t __arm_vmlaq(uint8x16_t, uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_s16)))
+int16x8_t __arm_vmlasq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_s16)))
+int16x8_t __arm_vmlasq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_s32)))
+int32x4_t __arm_vmlasq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_s32)))
+int32x4_t __arm_vmlasq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_s8)))
+int8x16_t __arm_vmlasq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_s8)))
+int8x16_t __arm_vmlasq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_u16)))
+uint16x8_t __arm_vmlasq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_u16)))
+uint16x8_t __arm_vmlasq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_u32)))
+uint32x4_t __arm_vmlasq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_u32)))
+uint32x4_t __arm_vmlasq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_u8)))
+uint8x16_t __arm_vmlasq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_u8)))
+uint8x16_t __arm_vmlasq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_s16)))
+int16x8_t __arm_vmlasq_n_s16(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_s16)))
+int16x8_t __arm_vmlasq(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_s32)))
+int32x4_t __arm_vmlasq_n_s32(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_s32)))
+int32x4_t __arm_vmlasq(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_s8)))
+int8x16_t __arm_vmlasq_n_s8(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_s8)))
+int8x16_t __arm_vmlasq(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_u16)))
+uint16x8_t __arm_vmlasq_n_u16(uint16x8_t, uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_u16)))
+uint16x8_t __arm_vmlasq(uint16x8_t, uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_u32)))
+uint32x4_t __arm_vmlasq_n_u32(uint32x4_t, uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_u32)))
+uint32x4_t __arm_vmlasq(uint32x4_t, uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_u8)))
+uint8x16_t __arm_vmlasq_n_u8(uint8x16_t, uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_u8)))
+uint8x16_t __arm_vmlasq(uint8x16_t, uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_p_s16)))
+int32_t __arm_vmlsdavaq_p_s16(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_p_s16)))
+int32_t __arm_vmlsdavaq_p(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_p_s32)))
+int32_t __arm_vmlsdavaq_p_s32(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_p_s32)))
+int32_t __arm_vmlsdavaq_p(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_p_s8)))
+int32_t __arm_vmlsdavaq_p_s8(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_p_s8)))
+int32_t __arm_vmlsdavaq_p(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_s16)))
+int32_t __arm_vmlsdavaq_s16(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_s16)))
+int32_t __arm_vmlsdavaq(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_s32)))
+int32_t __arm_vmlsdavaq_s32(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_s32)))
+int32_t __arm_vmlsdavaq(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_s8)))
+int32_t __arm_vmlsdavaq_s8(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_s8)))
+int32_t __arm_vmlsdavaq(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_p_s16)))
+int32_t __arm_vmlsdavaxq_p_s16(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_p_s16)))
+int32_t __arm_vmlsdavaxq_p(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_p_s32)))
+int32_t __arm_vmlsdavaxq_p_s32(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_p_s32)))
+int32_t __arm_vmlsdavaxq_p(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_p_s8)))
+int32_t __arm_vmlsdavaxq_p_s8(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_p_s8)))
+int32_t __arm_vmlsdavaxq_p(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_s16)))
+int32_t __arm_vmlsdavaxq_s16(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_s16)))
+int32_t __arm_vmlsdavaxq(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_s32)))
+int32_t __arm_vmlsdavaxq_s32(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_s32)))
+int32_t __arm_vmlsdavaxq(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_s8)))
+int32_t __arm_vmlsdavaxq_s8(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_s8)))
+int32_t __arm_vmlsdavaxq(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_p_s16)))
+int32_t __arm_vmlsdavq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_p_s16)))
+int32_t __arm_vmlsdavq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_p_s32)))
+int32_t __arm_vmlsdavq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_p_s32)))
+int32_t __arm_vmlsdavq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_p_s8)))
+int32_t __arm_vmlsdavq_p_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_p_s8)))
+int32_t __arm_vmlsdavq_p(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_s16)))
+int32_t __arm_vmlsdavq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_s16)))
+int32_t __arm_vmlsdavq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_s32)))
+int32_t __arm_vmlsdavq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_s32)))
+int32_t __arm_vmlsdavq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_s8)))
+int32_t __arm_vmlsdavq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_s8)))
+int32_t __arm_vmlsdavq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_p_s16)))
+int32_t __arm_vmlsdavxq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_p_s16)))
+int32_t __arm_vmlsdavxq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_p_s32)))
+int32_t __arm_vmlsdavxq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_p_s32)))
+int32_t __arm_vmlsdavxq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_p_s8)))
+int32_t __arm_vmlsdavxq_p_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_p_s8)))
+int32_t __arm_vmlsdavxq_p(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_s16)))
+int32_t __arm_vmlsdavxq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_s16)))
+int32_t __arm_vmlsdavxq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_s32)))
+int32_t __arm_vmlsdavxq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_s32)))
+int32_t __arm_vmlsdavxq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_s8)))
+int32_t __arm_vmlsdavxq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_s8)))
+int32_t __arm_vmlsdavxq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_p_s16)))
+int64_t __arm_vmlsldavaq_p_s16(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_p_s16)))
+int64_t __arm_vmlsldavaq_p(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_p_s32)))
+int64_t __arm_vmlsldavaq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_p_s32)))
+int64_t __arm_vmlsldavaq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_s16)))
+int64_t __arm_vmlsldavaq_s16(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_s16)))
+int64_t __arm_vmlsldavaq(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_s32)))
+int64_t __arm_vmlsldavaq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_s32)))
+int64_t __arm_vmlsldavaq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_p_s16)))
+int64_t __arm_vmlsldavaxq_p_s16(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_p_s16)))
+int64_t __arm_vmlsldavaxq_p(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_p_s32)))
+int64_t __arm_vmlsldavaxq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_p_s32)))
+int64_t __arm_vmlsldavaxq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_s16)))
+int64_t __arm_vmlsldavaxq_s16(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_s16)))
+int64_t __arm_vmlsldavaxq(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_s32)))
+int64_t __arm_vmlsldavaxq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_s32)))
+int64_t __arm_vmlsldavaxq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_p_s16)))
+int64_t __arm_vmlsldavq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_p_s16)))
+int64_t __arm_vmlsldavq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_p_s32)))
+int64_t __arm_vmlsldavq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_p_s32)))
+int64_t __arm_vmlsldavq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_s16)))
+int64_t __arm_vmlsldavq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_s16)))
+int64_t __arm_vmlsldavq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_s32)))
+int64_t __arm_vmlsldavq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_s32)))
+int64_t __arm_vmlsldavq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_p_s16)))
+int64_t __arm_vmlsldavxq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_p_s16)))
+int64_t __arm_vmlsldavxq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_p_s32)))
+int64_t __arm_vmlsldavxq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_p_s32)))
+int64_t __arm_vmlsldavxq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_s16)))
+int64_t __arm_vmlsldavxq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_s16)))
+int64_t __arm_vmlsldavxq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_s32)))
+int64_t __arm_vmlsldavxq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_s32)))
+int64_t __arm_vmlsldavxq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_s16)))
+int32x4_t __arm_vmovlbq_m_s16(int32x4_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_s16)))
+int32x4_t __arm_vmovlbq_m(int32x4_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_s8)))
+int16x8_t __arm_vmovlbq_m_s8(int16x8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_s8)))
+int16x8_t __arm_vmovlbq_m(int16x8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_u16)))
+uint32x4_t __arm_vmovlbq_m_u16(uint32x4_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_u16)))
+uint32x4_t __arm_vmovlbq_m(uint32x4_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_u8)))
+uint16x8_t __arm_vmovlbq_m_u8(uint16x8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_u8)))
+uint16x8_t __arm_vmovlbq_m(uint16x8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_s16)))
+int32x4_t __arm_vmovlbq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_s16)))
+int32x4_t __arm_vmovlbq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_s8)))
+int16x8_t __arm_vmovlbq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_s8)))
+int16x8_t __arm_vmovlbq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_u16)))
+uint32x4_t __arm_vmovlbq_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_u16)))
+uint32x4_t __arm_vmovlbq(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_u8)))
+uint16x8_t __arm_vmovlbq_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_u8)))
+uint16x8_t __arm_vmovlbq(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_s16)))
+int32x4_t __arm_vmovlbq_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_s16)))
+int32x4_t __arm_vmovlbq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_s8)))
+int16x8_t __arm_vmovlbq_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_s8)))
+int16x8_t __arm_vmovlbq_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_u16)))
+uint32x4_t __arm_vmovlbq_x_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_u16)))
+uint32x4_t __arm_vmovlbq_x(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_u8)))
+uint16x8_t __arm_vmovlbq_x_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_u8)))
+uint16x8_t __arm_vmovlbq_x(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_s16)))
+int32x4_t __arm_vmovltq_m_s16(int32x4_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_s16)))
+int32x4_t __arm_vmovltq_m(int32x4_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_s8)))
+int16x8_t __arm_vmovltq_m_s8(int16x8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_s8)))
+int16x8_t __arm_vmovltq_m(int16x8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_u16)))
+uint32x4_t __arm_vmovltq_m_u16(uint32x4_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_u16)))
+uint32x4_t __arm_vmovltq_m(uint32x4_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_u8)))
+uint16x8_t __arm_vmovltq_m_u8(uint16x8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_u8)))
+uint16x8_t __arm_vmovltq_m(uint16x8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_s16)))
+int32x4_t __arm_vmovltq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_s16)))
+int32x4_t __arm_vmovltq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_s8)))
+int16x8_t __arm_vmovltq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_s8)))
+int16x8_t __arm_vmovltq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_u16)))
+uint32x4_t __arm_vmovltq_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_u16)))
+uint32x4_t __arm_vmovltq(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_u8)))
+uint16x8_t __arm_vmovltq_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_u8)))
+uint16x8_t __arm_vmovltq(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_s16)))
+int32x4_t __arm_vmovltq_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_s16)))
+int32x4_t __arm_vmovltq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_s8)))
+int16x8_t __arm_vmovltq_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_s8)))
+int16x8_t __arm_vmovltq_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_u16)))
+uint32x4_t __arm_vmovltq_x_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_u16)))
+uint32x4_t __arm_vmovltq_x(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_u8)))
+uint16x8_t __arm_vmovltq_x_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_u8)))
+uint16x8_t __arm_vmovltq_x(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_s16)))
+int8x16_t __arm_vmovnbq_m_s16(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_s16)))
+int8x16_t __arm_vmovnbq_m(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_s32)))
+int16x8_t __arm_vmovnbq_m_s32(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_s32)))
+int16x8_t __arm_vmovnbq_m(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_u16)))
+uint8x16_t __arm_vmovnbq_m_u16(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_u16)))
+uint8x16_t __arm_vmovnbq_m(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_u32)))
+uint16x8_t __arm_vmovnbq_m_u32(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_u32)))
+uint16x8_t __arm_vmovnbq_m(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_s16)))
+int8x16_t __arm_vmovnbq_s16(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_s16)))
+int8x16_t __arm_vmovnbq(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_s32)))
+int16x8_t __arm_vmovnbq_s32(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_s32)))
+int16x8_t __arm_vmovnbq(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_u16)))
+uint8x16_t __arm_vmovnbq_u16(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_u16)))
+uint8x16_t __arm_vmovnbq(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_u32)))
+uint16x8_t __arm_vmovnbq_u32(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_u32)))
+uint16x8_t __arm_vmovnbq(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_s16)))
+int8x16_t __arm_vmovntq_m_s16(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_s16)))
+int8x16_t __arm_vmovntq_m(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_s32)))
+int16x8_t __arm_vmovntq_m_s32(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_s32)))
+int16x8_t __arm_vmovntq_m(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_u16)))
+uint8x16_t __arm_vmovntq_m_u16(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_u16)))
+uint8x16_t __arm_vmovntq_m(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_u32)))
+uint16x8_t __arm_vmovntq_m_u32(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_u32)))
+uint16x8_t __arm_vmovntq_m(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_s16)))
+int8x16_t __arm_vmovntq_s16(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_s16)))
+int8x16_t __arm_vmovntq(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_s32)))
+int16x8_t __arm_vmovntq_s32(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_s32)))
+int16x8_t __arm_vmovntq(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_u16)))
+uint8x16_t __arm_vmovntq_u16(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_u16)))
+uint8x16_t __arm_vmovntq(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_u32)))
+uint16x8_t __arm_vmovntq_u32(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_u32)))
+uint16x8_t __arm_vmovntq(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_s16)))
+int16x8_t __arm_vmulhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_s16)))
+int16x8_t __arm_vmulhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_s32)))
+int32x4_t __arm_vmulhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_s32)))
+int32x4_t __arm_vmulhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_s8)))
+int8x16_t __arm_vmulhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_s8)))
+int8x16_t __arm_vmulhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_u16)))
+uint16x8_t __arm_vmulhq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_u16)))
+uint16x8_t __arm_vmulhq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_u32)))
+uint32x4_t __arm_vmulhq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_u32)))
+uint32x4_t __arm_vmulhq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_u8)))
+uint8x16_t __arm_vmulhq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_u8)))
+uint8x16_t __arm_vmulhq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_s16)))
+int16x8_t __arm_vmulhq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_s16)))
+int16x8_t __arm_vmulhq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_s32)))
+int32x4_t __arm_vmulhq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_s32)))
+int32x4_t __arm_vmulhq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_s8)))
+int8x16_t __arm_vmulhq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_s8)))
+int8x16_t __arm_vmulhq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_u16)))
+uint16x8_t __arm_vmulhq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_u16)))
+uint16x8_t __arm_vmulhq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_u32)))
+uint32x4_t __arm_vmulhq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_u32)))
+uint32x4_t __arm_vmulhq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_u8)))
+uint8x16_t __arm_vmulhq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_u8)))
+uint8x16_t __arm_vmulhq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_s16)))
+int16x8_t __arm_vmulhq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_s16)))
+int16x8_t __arm_vmulhq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_s32)))
+int32x4_t __arm_vmulhq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_s32)))
+int32x4_t __arm_vmulhq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_s8)))
+int8x16_t __arm_vmulhq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_s8)))
+int8x16_t __arm_vmulhq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_u16)))
+uint16x8_t __arm_vmulhq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_u16)))
+uint16x8_t __arm_vmulhq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_u32)))
+uint32x4_t __arm_vmulhq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_u32)))
+uint32x4_t __arm_vmulhq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_u8)))
+uint8x16_t __arm_vmulhq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_u8)))
+uint8x16_t __arm_vmulhq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_s16)))
+int32x4_t __arm_vmullbq_int_m_s16(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_s16)))
+int32x4_t __arm_vmullbq_int_m(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_s32)))
+int64x2_t __arm_vmullbq_int_m_s32(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_s32)))
+int64x2_t __arm_vmullbq_int_m(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_s8)))
+int16x8_t __arm_vmullbq_int_m_s8(int16x8_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_s8)))
+int16x8_t __arm_vmullbq_int_m(int16x8_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_u16)))
+uint32x4_t __arm_vmullbq_int_m_u16(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_u16)))
+uint32x4_t __arm_vmullbq_int_m(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_u32)))
+uint64x2_t __arm_vmullbq_int_m_u32(uint64x2_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_u32)))
+uint64x2_t __arm_vmullbq_int_m(uint64x2_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_u8)))
+uint16x8_t __arm_vmullbq_int_m_u8(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_u8)))
+uint16x8_t __arm_vmullbq_int_m(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_s16)))
+int32x4_t __arm_vmullbq_int_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_s16)))
+int32x4_t __arm_vmullbq_int(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_s32)))
+int64x2_t __arm_vmullbq_int_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_s32)))
+int64x2_t __arm_vmullbq_int(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_s8)))
+int16x8_t __arm_vmullbq_int_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_s8)))
+int16x8_t __arm_vmullbq_int(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_u16)))
+uint32x4_t __arm_vmullbq_int_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_u16)))
+uint32x4_t __arm_vmullbq_int(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_u32)))
+uint64x2_t __arm_vmullbq_int_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_u32)))
+uint64x2_t __arm_vmullbq_int(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_u8)))
+uint16x8_t __arm_vmullbq_int_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_u8)))
+uint16x8_t __arm_vmullbq_int(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_s16)))
+int32x4_t __arm_vmullbq_int_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_s16)))
+int32x4_t __arm_vmullbq_int_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_s32)))
+int64x2_t __arm_vmullbq_int_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_s32)))
+int64x2_t __arm_vmullbq_int_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_s8)))
+int16x8_t __arm_vmullbq_int_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_s8)))
+int16x8_t __arm_vmullbq_int_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_u16)))
+uint32x4_t __arm_vmullbq_int_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_u16)))
+uint32x4_t __arm_vmullbq_int_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_u32)))
+uint64x2_t __arm_vmullbq_int_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_u32)))
+uint64x2_t __arm_vmullbq_int_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_u8)))
+uint16x8_t __arm_vmullbq_int_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_u8)))
+uint16x8_t __arm_vmullbq_int_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_m_p16)))
+uint32x4_t __arm_vmullbq_poly_m_p16(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_m_p16)))
+uint32x4_t __arm_vmullbq_poly_m(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_m_p8)))
+uint16x8_t __arm_vmullbq_poly_m_p8(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_m_p8)))
+uint16x8_t __arm_vmullbq_poly_m(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_p16)))
+uint32x4_t __arm_vmullbq_poly_p16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_p16)))
+uint32x4_t __arm_vmullbq_poly(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_p8)))
+uint16x8_t __arm_vmullbq_poly_p8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_p8)))
+uint16x8_t __arm_vmullbq_poly(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_x_p16)))
+uint32x4_t __arm_vmullbq_poly_x_p16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_x_p16)))
+uint32x4_t __arm_vmullbq_poly_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_x_p8)))
+uint16x8_t __arm_vmullbq_poly_x_p8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_x_p8)))
+uint16x8_t __arm_vmullbq_poly_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_s16)))
+int32x4_t __arm_vmulltq_int_m_s16(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_s16)))
+int32x4_t __arm_vmulltq_int_m(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_s32)))
+int64x2_t __arm_vmulltq_int_m_s32(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_s32)))
+int64x2_t __arm_vmulltq_int_m(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_s8)))
+int16x8_t __arm_vmulltq_int_m_s8(int16x8_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_s8)))
+int16x8_t __arm_vmulltq_int_m(int16x8_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_u16)))
+uint32x4_t __arm_vmulltq_int_m_u16(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_u16)))
+uint32x4_t __arm_vmulltq_int_m(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_u32)))
+uint64x2_t __arm_vmulltq_int_m_u32(uint64x2_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_u32)))
+uint64x2_t __arm_vmulltq_int_m(uint64x2_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_u8)))
+uint16x8_t __arm_vmulltq_int_m_u8(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_u8)))
+uint16x8_t __arm_vmulltq_int_m(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_s16)))
+int32x4_t __arm_vmulltq_int_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_s16)))
+int32x4_t __arm_vmulltq_int(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_s32)))
+int64x2_t __arm_vmulltq_int_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_s32)))
+int64x2_t __arm_vmulltq_int(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_s8)))
+int16x8_t __arm_vmulltq_int_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_s8)))
+int16x8_t __arm_vmulltq_int(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_u16)))
+uint32x4_t __arm_vmulltq_int_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_u16)))
+uint32x4_t __arm_vmulltq_int(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_u32)))
+uint64x2_t __arm_vmulltq_int_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_u32)))
+uint64x2_t __arm_vmulltq_int(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_u8)))
+uint16x8_t __arm_vmulltq_int_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_u8)))
+uint16x8_t __arm_vmulltq_int(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_s16)))
+int32x4_t __arm_vmulltq_int_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_s16)))
+int32x4_t __arm_vmulltq_int_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_s32)))
+int64x2_t __arm_vmulltq_int_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_s32)))
+int64x2_t __arm_vmulltq_int_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_s8)))
+int16x8_t __arm_vmulltq_int_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_s8)))
+int16x8_t __arm_vmulltq_int_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_u16)))
+uint32x4_t __arm_vmulltq_int_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_u16)))
+uint32x4_t __arm_vmulltq_int_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_u32)))
+uint64x2_t __arm_vmulltq_int_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_u32)))
+uint64x2_t __arm_vmulltq_int_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_u8)))
+uint16x8_t __arm_vmulltq_int_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_u8)))
+uint16x8_t __arm_vmulltq_int_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_m_p16)))
+uint32x4_t __arm_vmulltq_poly_m_p16(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_m_p16)))
+uint32x4_t __arm_vmulltq_poly_m(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_m_p8)))
+uint16x8_t __arm_vmulltq_poly_m_p8(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_m_p8)))
+uint16x8_t __arm_vmulltq_poly_m(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_p16)))
+uint32x4_t __arm_vmulltq_poly_p16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_p16)))
+uint32x4_t __arm_vmulltq_poly(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_p8)))
+uint16x8_t __arm_vmulltq_poly_p8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_p8)))
+uint16x8_t __arm_vmulltq_poly(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_x_p16)))
+uint32x4_t __arm_vmulltq_poly_x_p16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_x_p16)))
+uint32x4_t __arm_vmulltq_poly_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_x_p8)))
+uint16x8_t __arm_vmulltq_poly_x_p8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_x_p8)))
+uint16x8_t __arm_vmulltq_poly_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_s16)))
+int16x8_t __arm_vmulq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_s16)))
+int16x8_t __arm_vmulq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_s32)))
+int32x4_t __arm_vmulq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_s32)))
+int32x4_t __arm_vmulq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_s8)))
+int8x16_t __arm_vmulq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_s8)))
+int8x16_t __arm_vmulq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_u16)))
+uint16x8_t __arm_vmulq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_u16)))
+uint16x8_t __arm_vmulq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_u32)))
+uint32x4_t __arm_vmulq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_u32)))
+uint32x4_t __arm_vmulq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_u8)))
+uint8x16_t __arm_vmulq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_u8)))
+uint8x16_t __arm_vmulq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_s16)))
+int16x8_t __arm_vmulq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_s16)))
+int16x8_t __arm_vmulq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_s32)))
+int32x4_t __arm_vmulq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_s32)))
+int32x4_t __arm_vmulq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_s8)))
+int8x16_t __arm_vmulq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_s8)))
+int8x16_t __arm_vmulq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_u16)))
+uint16x8_t __arm_vmulq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_u16)))
+uint16x8_t __arm_vmulq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_u32)))
+uint32x4_t __arm_vmulq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_u32)))
+uint32x4_t __arm_vmulq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_u8)))
+uint8x16_t __arm_vmulq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_u8)))
+uint8x16_t __arm_vmulq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_s16)))
+int16x8_t __arm_vmulq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_s16)))
+int16x8_t __arm_vmulq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_s32)))
+int32x4_t __arm_vmulq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_s32)))
+int32x4_t __arm_vmulq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_s8)))
+int8x16_t __arm_vmulq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_s8)))
+int8x16_t __arm_vmulq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_u16)))
+uint16x8_t __arm_vmulq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_u16)))
+uint16x8_t __arm_vmulq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_u32)))
+uint32x4_t __arm_vmulq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_u32)))
+uint32x4_t __arm_vmulq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_u8)))
+uint8x16_t __arm_vmulq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_u8)))
+uint8x16_t __arm_vmulq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_s16)))
+int16x8_t __arm_vmulq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_s16)))
+int16x8_t __arm_vmulq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_s32)))
+int32x4_t __arm_vmulq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_s32)))
+int32x4_t __arm_vmulq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_s8)))
+int8x16_t __arm_vmulq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_s8)))
+int8x16_t __arm_vmulq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_u16)))
+uint16x8_t __arm_vmulq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_u16)))
+uint16x8_t __arm_vmulq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_u32)))
+uint32x4_t __arm_vmulq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_u32)))
+uint32x4_t __arm_vmulq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_u8)))
+uint8x16_t __arm_vmulq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_u8)))
+uint8x16_t __arm_vmulq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_s16)))
+int16x8_t __arm_vmulq_x_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_s16)))
+int16x8_t __arm_vmulq_x(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_s32)))
+int32x4_t __arm_vmulq_x_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_s32)))
+int32x4_t __arm_vmulq_x(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_s8)))
+int8x16_t __arm_vmulq_x_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_s8)))
+int8x16_t __arm_vmulq_x(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_u16)))
+uint16x8_t __arm_vmulq_x_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_u16)))
+uint16x8_t __arm_vmulq_x(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_u32)))
+uint32x4_t __arm_vmulq_x_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_u32)))
+uint32x4_t __arm_vmulq_x(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_u8)))
+uint8x16_t __arm_vmulq_x_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_u8)))
+uint8x16_t __arm_vmulq_x(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_s16)))
+int16x8_t __arm_vmulq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_s16)))
+int16x8_t __arm_vmulq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_s32)))
+int32x4_t __arm_vmulq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_s32)))
+int32x4_t __arm_vmulq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_s8)))
+int8x16_t __arm_vmulq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_s8)))
+int8x16_t __arm_vmulq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_u16)))
+uint16x8_t __arm_vmulq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_u16)))
+uint16x8_t __arm_vmulq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_u32)))
+uint32x4_t __arm_vmulq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_u32)))
+uint32x4_t __arm_vmulq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_u8)))
+uint8x16_t __arm_vmulq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_u8)))
+uint8x16_t __arm_vmulq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_s16)))
+int16x8_t __arm_vmvnq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_s16)))
+int16x8_t __arm_vmvnq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_s32)))
+int32x4_t __arm_vmvnq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_s32)))
+int32x4_t __arm_vmvnq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_u16)))
+uint16x8_t __arm_vmvnq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_u16)))
+uint16x8_t __arm_vmvnq_m(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_u32)))
+uint32x4_t __arm_vmvnq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_u32)))
+uint32x4_t __arm_vmvnq_m(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_s16)))
+int16x8_t __arm_vmvnq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_s16)))
+int16x8_t __arm_vmvnq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_s32)))
+int32x4_t __arm_vmvnq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_s32)))
+int32x4_t __arm_vmvnq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_s8)))
+int8x16_t __arm_vmvnq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_s8)))
+int8x16_t __arm_vmvnq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_u16)))
+uint16x8_t __arm_vmvnq_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_u16)))
+uint16x8_t __arm_vmvnq_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_u32)))
+uint32x4_t __arm_vmvnq_m_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_u32)))
+uint32x4_t __arm_vmvnq_m(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_u8)))
+uint8x16_t __arm_vmvnq_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_u8)))
+uint8x16_t __arm_vmvnq_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_n_s16)))
+int16x8_t __arm_vmvnq_n_s16(int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_n_s32)))
+int32x4_t __arm_vmvnq_n_s32(int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_n_u16)))
+uint16x8_t __arm_vmvnq_n_u16(uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_n_u32)))
+uint32x4_t __arm_vmvnq_n_u32(uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_s16)))
+int16x8_t __arm_vmvnq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_s16)))
+int16x8_t __arm_vmvnq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_s32)))
+int32x4_t __arm_vmvnq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_s32)))
+int32x4_t __arm_vmvnq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_s8)))
+int8x16_t __arm_vmvnq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_s8)))
+int8x16_t __arm_vmvnq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_u16)))
+uint16x8_t __arm_vmvnq_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_u16)))
+uint16x8_t __arm_vmvnq(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_u32)))
+uint32x4_t __arm_vmvnq_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_u32)))
+uint32x4_t __arm_vmvnq(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_u8)))
+uint8x16_t __arm_vmvnq_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_u8)))
+uint8x16_t __arm_vmvnq(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_n_s16)))
+int16x8_t __arm_vmvnq_x_n_s16(int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_n_s32)))
+int32x4_t __arm_vmvnq_x_n_s32(int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_n_u16)))
+uint16x8_t __arm_vmvnq_x_n_u16(uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_n_u32)))
+uint32x4_t __arm_vmvnq_x_n_u32(uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_s16)))
+int16x8_t __arm_vmvnq_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_s16)))
+int16x8_t __arm_vmvnq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_s32)))
+int32x4_t __arm_vmvnq_x_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_s32)))
+int32x4_t __arm_vmvnq_x(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_s8)))
+int8x16_t __arm_vmvnq_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_s8)))
+int8x16_t __arm_vmvnq_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_u16)))
+uint16x8_t __arm_vmvnq_x_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_u16)))
+uint16x8_t __arm_vmvnq_x(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_u32)))
+uint32x4_t __arm_vmvnq_x_u32(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_u32)))
+uint32x4_t __arm_vmvnq_x(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_u8)))
+uint8x16_t __arm_vmvnq_x_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_u8)))
+uint8x16_t __arm_vmvnq_x(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_s16)))
+int16x8_t __arm_vnegq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_s16)))
+int16x8_t __arm_vnegq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_s32)))
+int32x4_t __arm_vnegq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_s32)))
+int32x4_t __arm_vnegq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_s8)))
+int8x16_t __arm_vnegq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_s8)))
+int8x16_t __arm_vnegq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_s16)))
+int16x8_t __arm_vnegq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_s16)))
+int16x8_t __arm_vnegq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_s32)))
+int32x4_t __arm_vnegq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_s32)))
+int32x4_t __arm_vnegq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_s8)))
+int8x16_t __arm_vnegq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_s8)))
+int8x16_t __arm_vnegq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_s16)))
+int16x8_t __arm_vnegq_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_s16)))
+int16x8_t __arm_vnegq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_s32)))
+int32x4_t __arm_vnegq_x_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_s32)))
+int32x4_t __arm_vnegq_x(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_s8)))
+int8x16_t __arm_vnegq_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_s8)))
+int8x16_t __arm_vnegq_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_s16)))
+int16x8_t __arm_vornq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_s16)))
+int16x8_t __arm_vornq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_s32)))
+int32x4_t __arm_vornq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_s32)))
+int32x4_t __arm_vornq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_s8)))
+int8x16_t __arm_vornq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_s8)))
+int8x16_t __arm_vornq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_u16)))
+uint16x8_t __arm_vornq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_u16)))
+uint16x8_t __arm_vornq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_u32)))
+uint32x4_t __arm_vornq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_u32)))
+uint32x4_t __arm_vornq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_u8)))
+uint8x16_t __arm_vornq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_u8)))
+uint8x16_t __arm_vornq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_s16)))
+int16x8_t __arm_vornq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_s16)))
+int16x8_t __arm_vornq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_s32)))
+int32x4_t __arm_vornq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_s32)))
+int32x4_t __arm_vornq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_s8)))
+int8x16_t __arm_vornq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_s8)))
+int8x16_t __arm_vornq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_u16)))
+uint16x8_t __arm_vornq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_u16)))
+uint16x8_t __arm_vornq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_u32)))
+uint32x4_t __arm_vornq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_u32)))
+uint32x4_t __arm_vornq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_u8)))
+uint8x16_t __arm_vornq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_u8)))
+uint8x16_t __arm_vornq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_s16)))
+int16x8_t __arm_vornq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_s16)))
+int16x8_t __arm_vornq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_s32)))
+int32x4_t __arm_vornq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_s32)))
+int32x4_t __arm_vornq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_s8)))
+int8x16_t __arm_vornq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_s8)))
+int8x16_t __arm_vornq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_u16)))
+uint16x8_t __arm_vornq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_u16)))
+uint16x8_t __arm_vornq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_u32)))
+uint32x4_t __arm_vornq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_u32)))
+uint32x4_t __arm_vornq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_u8)))
+uint8x16_t __arm_vornq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_u8)))
+uint8x16_t __arm_vornq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_s16)))
+int16x8_t __arm_vorrq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_s16)))
+int16x8_t __arm_vorrq_m_n(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_s32)))
+int32x4_t __arm_vorrq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_s32)))
+int32x4_t __arm_vorrq_m_n(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_u16)))
+uint16x8_t __arm_vorrq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_u16)))
+uint16x8_t __arm_vorrq_m_n(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_u32)))
+uint32x4_t __arm_vorrq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_u32)))
+uint32x4_t __arm_vorrq_m_n(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_s16)))
+int16x8_t __arm_vorrq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_s16)))
+int16x8_t __arm_vorrq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_s32)))
+int32x4_t __arm_vorrq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_s32)))
+int32x4_t __arm_vorrq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_s8)))
+int8x16_t __arm_vorrq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_s8)))
+int8x16_t __arm_vorrq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_u16)))
+uint16x8_t __arm_vorrq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_u16)))
+uint16x8_t __arm_vorrq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_u32)))
+uint32x4_t __arm_vorrq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_u32)))
+uint32x4_t __arm_vorrq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_u8)))
+uint8x16_t __arm_vorrq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_u8)))
+uint8x16_t __arm_vorrq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_s16)))
+int16x8_t __arm_vorrq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_s16)))
+int16x8_t __arm_vorrq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_s32)))
+int32x4_t __arm_vorrq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_s32)))
+int32x4_t __arm_vorrq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_u16)))
+uint16x8_t __arm_vorrq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_u16)))
+uint16x8_t __arm_vorrq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_u32)))
+uint32x4_t __arm_vorrq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_u32)))
+uint32x4_t __arm_vorrq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_s16)))
+int16x8_t __arm_vorrq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_s16)))
+int16x8_t __arm_vorrq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_s32)))
+int32x4_t __arm_vorrq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_s32)))
+int32x4_t __arm_vorrq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_s8)))
+int8x16_t __arm_vorrq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_s8)))
+int8x16_t __arm_vorrq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_u16)))
+uint16x8_t __arm_vorrq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_u16)))
+uint16x8_t __arm_vorrq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_u32)))
+uint32x4_t __arm_vorrq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_u32)))
+uint32x4_t __arm_vorrq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_u8)))
+uint8x16_t __arm_vorrq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_u8)))
+uint8x16_t __arm_vorrq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_s16)))
+int16x8_t __arm_vorrq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_s16)))
+int16x8_t __arm_vorrq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_s32)))
+int32x4_t __arm_vorrq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_s32)))
+int32x4_t __arm_vorrq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_s8)))
+int8x16_t __arm_vorrq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_s8)))
+int8x16_t __arm_vorrq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_u16)))
+uint16x8_t __arm_vorrq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_u16)))
+uint16x8_t __arm_vorrq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_u32)))
+uint32x4_t __arm_vorrq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_u32)))
+uint32x4_t __arm_vorrq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_u8)))
+uint8x16_t __arm_vorrq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_u8)))
+uint8x16_t __arm_vorrq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpnot)))
+mve_pred16_t __arm_vpnot(mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s16)))
+int16x8_t __arm_vpselq_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s16)))
+int16x8_t __arm_vpselq(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s32)))
+int32x4_t __arm_vpselq_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s32)))
+int32x4_t __arm_vpselq(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s64)))
+int64x2_t __arm_vpselq_s64(int64x2_t, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s64)))
+int64x2_t __arm_vpselq(int64x2_t, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s8)))
+int8x16_t __arm_vpselq_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s8)))
+int8x16_t __arm_vpselq(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u16)))
+uint16x8_t __arm_vpselq_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u16)))
+uint16x8_t __arm_vpselq(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u32)))
+uint32x4_t __arm_vpselq_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u32)))
+uint32x4_t __arm_vpselq(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u64)))
+uint64x2_t __arm_vpselq_u64(uint64x2_t, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u64)))
+uint64x2_t __arm_vpselq(uint64x2_t, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u8)))
+uint8x16_t __arm_vpselq_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u8)))
+uint8x16_t __arm_vpselq(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_m_s16)))
+int16x8_t __arm_vqabsq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_m_s16)))
+int16x8_t __arm_vqabsq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_m_s32)))
+int32x4_t __arm_vqabsq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_m_s32)))
+int32x4_t __arm_vqabsq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_m_s8)))
+int8x16_t __arm_vqabsq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_m_s8)))
+int8x16_t __arm_vqabsq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_s16)))
+int16x8_t __arm_vqabsq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_s16)))
+int16x8_t __arm_vqabsq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_s32)))
+int32x4_t __arm_vqabsq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_s32)))
+int32x4_t __arm_vqabsq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_s8)))
+int8x16_t __arm_vqabsq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_s8)))
+int8x16_t __arm_vqabsq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_s16)))
+int16x8_t __arm_vqaddq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_s16)))
+int16x8_t __arm_vqaddq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_s32)))
+int32x4_t __arm_vqaddq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_s32)))
+int32x4_t __arm_vqaddq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_s8)))
+int8x16_t __arm_vqaddq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_s8)))
+int8x16_t __arm_vqaddq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_u16)))
+uint16x8_t __arm_vqaddq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_u16)))
+uint16x8_t __arm_vqaddq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_u32)))
+uint32x4_t __arm_vqaddq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_u32)))
+uint32x4_t __arm_vqaddq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_u8)))
+uint8x16_t __arm_vqaddq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_u8)))
+uint8x16_t __arm_vqaddq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_s16)))
+int16x8_t __arm_vqaddq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_s16)))
+int16x8_t __arm_vqaddq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_s32)))
+int32x4_t __arm_vqaddq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_s32)))
+int32x4_t __arm_vqaddq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_s8)))
+int8x16_t __arm_vqaddq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_s8)))
+int8x16_t __arm_vqaddq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_u16)))
+uint16x8_t __arm_vqaddq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_u16)))
+uint16x8_t __arm_vqaddq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_u32)))
+uint32x4_t __arm_vqaddq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_u32)))
+uint32x4_t __arm_vqaddq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_u8)))
+uint8x16_t __arm_vqaddq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_u8)))
+uint8x16_t __arm_vqaddq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_s16)))
+int16x8_t __arm_vqaddq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_s16)))
+int16x8_t __arm_vqaddq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_s32)))
+int32x4_t __arm_vqaddq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_s32)))
+int32x4_t __arm_vqaddq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_s8)))
+int8x16_t __arm_vqaddq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_s8)))
+int8x16_t __arm_vqaddq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_u16)))
+uint16x8_t __arm_vqaddq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_u16)))
+uint16x8_t __arm_vqaddq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_u32)))
+uint32x4_t __arm_vqaddq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_u32)))
+uint32x4_t __arm_vqaddq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_u8)))
+uint8x16_t __arm_vqaddq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_u8)))
+uint8x16_t __arm_vqaddq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_s16)))
+int16x8_t __arm_vqaddq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_s16)))
+int16x8_t __arm_vqaddq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_s32)))
+int32x4_t __arm_vqaddq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_s32)))
+int32x4_t __arm_vqaddq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_s8)))
+int8x16_t __arm_vqaddq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_s8)))
+int8x16_t __arm_vqaddq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_u16)))
+uint16x8_t __arm_vqaddq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_u16)))
+uint16x8_t __arm_vqaddq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_u32)))
+uint32x4_t __arm_vqaddq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_u32)))
+uint32x4_t __arm_vqaddq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_u8)))
+uint8x16_t __arm_vqaddq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_u8)))
+uint8x16_t __arm_vqaddq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_m_s16)))
+int16x8_t __arm_vqdmladhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_m_s16)))
+int16x8_t __arm_vqdmladhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_m_s32)))
+int32x4_t __arm_vqdmladhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_m_s32)))
+int32x4_t __arm_vqdmladhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_m_s8)))
+int8x16_t __arm_vqdmladhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_m_s8)))
+int8x16_t __arm_vqdmladhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_s16)))
+int16x8_t __arm_vqdmladhq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_s16)))
+int16x8_t __arm_vqdmladhq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_s32)))
+int32x4_t __arm_vqdmladhq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_s32)))
+int32x4_t __arm_vqdmladhq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_s8)))
+int8x16_t __arm_vqdmladhq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_s8)))
+int8x16_t __arm_vqdmladhq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_m_s16)))
+int16x8_t __arm_vqdmladhxq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_m_s16)))
+int16x8_t __arm_vqdmladhxq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_m_s32)))
+int32x4_t __arm_vqdmladhxq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_m_s32)))
+int32x4_t __arm_vqdmladhxq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_m_s8)))
+int8x16_t __arm_vqdmladhxq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_m_s8)))
+int8x16_t __arm_vqdmladhxq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_s16)))
+int16x8_t __arm_vqdmladhxq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_s16)))
+int16x8_t __arm_vqdmladhxq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_s32)))
+int32x4_t __arm_vqdmladhxq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_s32)))
+int32x4_t __arm_vqdmladhxq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_s8)))
+int8x16_t __arm_vqdmladhxq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_s8)))
+int8x16_t __arm_vqdmladhxq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_m_n_s16)))
+int16x8_t __arm_vqdmlahq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_m_n_s16)))
+int16x8_t __arm_vqdmlahq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_m_n_s32)))
+int32x4_t __arm_vqdmlahq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_m_n_s32)))
+int32x4_t __arm_vqdmlahq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_m_n_s8)))
+int8x16_t __arm_vqdmlahq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_m_n_s8)))
+int8x16_t __arm_vqdmlahq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_n_s16)))
+int16x8_t __arm_vqdmlahq_n_s16(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_n_s16)))
+int16x8_t __arm_vqdmlahq(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_n_s32)))
+int32x4_t __arm_vqdmlahq_n_s32(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_n_s32)))
+int32x4_t __arm_vqdmlahq(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_n_s8)))
+int8x16_t __arm_vqdmlahq_n_s8(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_n_s8)))
+int8x16_t __arm_vqdmlahq(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_m_n_s16)))
+int16x8_t __arm_vqdmlashq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_m_n_s16)))
+int16x8_t __arm_vqdmlashq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_m_n_s32)))
+int32x4_t __arm_vqdmlashq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_m_n_s32)))
+int32x4_t __arm_vqdmlashq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_m_n_s8)))
+int8x16_t __arm_vqdmlashq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_m_n_s8)))
+int8x16_t __arm_vqdmlashq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_n_s16)))
+int16x8_t __arm_vqdmlashq_n_s16(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_n_s16)))
+int16x8_t __arm_vqdmlashq(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_n_s32)))
+int32x4_t __arm_vqdmlashq_n_s32(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_n_s32)))
+int32x4_t __arm_vqdmlashq(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_n_s8)))
+int8x16_t __arm_vqdmlashq_n_s8(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_n_s8)))
+int8x16_t __arm_vqdmlashq(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_m_s16)))
+int16x8_t __arm_vqdmlsdhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_m_s16)))
+int16x8_t __arm_vqdmlsdhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_m_s32)))
+int32x4_t __arm_vqdmlsdhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_m_s32)))
+int32x4_t __arm_vqdmlsdhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_m_s8)))
+int8x16_t __arm_vqdmlsdhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_m_s8)))
+int8x16_t __arm_vqdmlsdhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_s16)))
+int16x8_t __arm_vqdmlsdhq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_s16)))
+int16x8_t __arm_vqdmlsdhq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_s32)))
+int32x4_t __arm_vqdmlsdhq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_s32)))
+int32x4_t __arm_vqdmlsdhq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_s8)))
+int8x16_t __arm_vqdmlsdhq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_s8)))
+int8x16_t __arm_vqdmlsdhq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_m_s16)))
+int16x8_t __arm_vqdmlsdhxq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_m_s16)))
+int16x8_t __arm_vqdmlsdhxq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_m_s32)))
+int32x4_t __arm_vqdmlsdhxq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_m_s32)))
+int32x4_t __arm_vqdmlsdhxq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_m_s8)))
+int8x16_t __arm_vqdmlsdhxq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_m_s8)))
+int8x16_t __arm_vqdmlsdhxq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_s16)))
+int16x8_t __arm_vqdmlsdhxq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_s16)))
+int16x8_t __arm_vqdmlsdhxq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_s32)))
+int32x4_t __arm_vqdmlsdhxq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_s32)))
+int32x4_t __arm_vqdmlsdhxq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_s8)))
+int8x16_t __arm_vqdmlsdhxq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_s8)))
+int8x16_t __arm_vqdmlsdhxq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_n_s16)))
+int16x8_t __arm_vqdmulhq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_n_s16)))
+int16x8_t __arm_vqdmulhq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_n_s32)))
+int32x4_t __arm_vqdmulhq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_n_s32)))
+int32x4_t __arm_vqdmulhq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_n_s8)))
+int8x16_t __arm_vqdmulhq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_n_s8)))
+int8x16_t __arm_vqdmulhq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_s16)))
+int16x8_t __arm_vqdmulhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_s16)))
+int16x8_t __arm_vqdmulhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_s32)))
+int32x4_t __arm_vqdmulhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_s32)))
+int32x4_t __arm_vqdmulhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_s8)))
+int8x16_t __arm_vqdmulhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_s8)))
+int8x16_t __arm_vqdmulhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_n_s16)))
+int16x8_t __arm_vqdmulhq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_n_s16)))
+int16x8_t __arm_vqdmulhq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_n_s32)))
+int32x4_t __arm_vqdmulhq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_n_s32)))
+int32x4_t __arm_vqdmulhq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_n_s8)))
+int8x16_t __arm_vqdmulhq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_n_s8)))
+int8x16_t __arm_vqdmulhq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_s16)))
+int16x8_t __arm_vqdmulhq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_s16)))
+int16x8_t __arm_vqdmulhq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_s32)))
+int32x4_t __arm_vqdmulhq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_s32)))
+int32x4_t __arm_vqdmulhq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_s8)))
+int8x16_t __arm_vqdmulhq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_s8)))
+int8x16_t __arm_vqdmulhq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_n_s16)))
+int32x4_t __arm_vqdmullbq_m_n_s16(int32x4_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_n_s16)))
+int32x4_t __arm_vqdmullbq_m(int32x4_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_n_s32)))
+int64x2_t __arm_vqdmullbq_m_n_s32(int64x2_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_n_s32)))
+int64x2_t __arm_vqdmullbq_m(int64x2_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_s16)))
+int32x4_t __arm_vqdmullbq_m_s16(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_s16)))
+int32x4_t __arm_vqdmullbq_m(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_s32)))
+int64x2_t __arm_vqdmullbq_m_s32(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_s32)))
+int64x2_t __arm_vqdmullbq_m(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_n_s16)))
+int32x4_t __arm_vqdmullbq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_n_s16)))
+int32x4_t __arm_vqdmullbq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_n_s32)))
+int64x2_t __arm_vqdmullbq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_n_s32)))
+int64x2_t __arm_vqdmullbq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_s16)))
+int32x4_t __arm_vqdmullbq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_s16)))
+int32x4_t __arm_vqdmullbq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_s32)))
+int64x2_t __arm_vqdmullbq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_s32)))
+int64x2_t __arm_vqdmullbq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_n_s16)))
+int32x4_t __arm_vqdmulltq_m_n_s16(int32x4_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_n_s16)))
+int32x4_t __arm_vqdmulltq_m(int32x4_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_n_s32)))
+int64x2_t __arm_vqdmulltq_m_n_s32(int64x2_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_n_s32)))
+int64x2_t __arm_vqdmulltq_m(int64x2_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_s16)))
+int32x4_t __arm_vqdmulltq_m_s16(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_s16)))
+int32x4_t __arm_vqdmulltq_m(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_s32)))
+int64x2_t __arm_vqdmulltq_m_s32(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_s32)))
+int64x2_t __arm_vqdmulltq_m(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_n_s16)))
+int32x4_t __arm_vqdmulltq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_n_s16)))
+int32x4_t __arm_vqdmulltq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_n_s32)))
+int64x2_t __arm_vqdmulltq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_n_s32)))
+int64x2_t __arm_vqdmulltq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_s16)))
+int32x4_t __arm_vqdmulltq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_s16)))
+int32x4_t __arm_vqdmulltq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_s32)))
+int64x2_t __arm_vqdmulltq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_s32)))
+int64x2_t __arm_vqdmulltq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_s16)))
+int8x16_t __arm_vqmovnbq_m_s16(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_s16)))
+int8x16_t __arm_vqmovnbq_m(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_s32)))
+int16x8_t __arm_vqmovnbq_m_s32(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_s32)))
+int16x8_t __arm_vqmovnbq_m(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_u16)))
+uint8x16_t __arm_vqmovnbq_m_u16(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_u16)))
+uint8x16_t __arm_vqmovnbq_m(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_u32)))
+uint16x8_t __arm_vqmovnbq_m_u32(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_u32)))
+uint16x8_t __arm_vqmovnbq_m(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_s16)))
+int8x16_t __arm_vqmovnbq_s16(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_s16)))
+int8x16_t __arm_vqmovnbq(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_s32)))
+int16x8_t __arm_vqmovnbq_s32(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_s32)))
+int16x8_t __arm_vqmovnbq(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_u16)))
+uint8x16_t __arm_vqmovnbq_u16(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_u16)))
+uint8x16_t __arm_vqmovnbq(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_u32)))
+uint16x8_t __arm_vqmovnbq_u32(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_u32)))
+uint16x8_t __arm_vqmovnbq(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_s16)))
+int8x16_t __arm_vqmovntq_m_s16(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_s16)))
+int8x16_t __arm_vqmovntq_m(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_s32)))
+int16x8_t __arm_vqmovntq_m_s32(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_s32)))
+int16x8_t __arm_vqmovntq_m(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_u16)))
+uint8x16_t __arm_vqmovntq_m_u16(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_u16)))
+uint8x16_t __arm_vqmovntq_m(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_u32)))
+uint16x8_t __arm_vqmovntq_m_u32(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_u32)))
+uint16x8_t __arm_vqmovntq_m(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_s16)))
+int8x16_t __arm_vqmovntq_s16(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_s16)))
+int8x16_t __arm_vqmovntq(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_s32)))
+int16x8_t __arm_vqmovntq_s32(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_s32)))
+int16x8_t __arm_vqmovntq(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_u16)))
+uint8x16_t __arm_vqmovntq_u16(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_u16)))
+uint8x16_t __arm_vqmovntq(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_u32)))
+uint16x8_t __arm_vqmovntq_u32(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_u32)))
+uint16x8_t __arm_vqmovntq(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_m_s16)))
+uint8x16_t __arm_vqmovunbq_m_s16(uint8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_m_s16)))
+uint8x16_t __arm_vqmovunbq_m(uint8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_m_s32)))
+uint16x8_t __arm_vqmovunbq_m_s32(uint16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_m_s32)))
+uint16x8_t __arm_vqmovunbq_m(uint16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_s16)))
+uint8x16_t __arm_vqmovunbq_s16(uint8x16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_s16)))
+uint8x16_t __arm_vqmovunbq(uint8x16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_s32)))
+uint16x8_t __arm_vqmovunbq_s32(uint16x8_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_s32)))
+uint16x8_t __arm_vqmovunbq(uint16x8_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_m_s16)))
+uint8x16_t __arm_vqmovuntq_m_s16(uint8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_m_s16)))
+uint8x16_t __arm_vqmovuntq_m(uint8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_m_s32)))
+uint16x8_t __arm_vqmovuntq_m_s32(uint16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_m_s32)))
+uint16x8_t __arm_vqmovuntq_m(uint16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_s16)))
+uint8x16_t __arm_vqmovuntq_s16(uint8x16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_s16)))
+uint8x16_t __arm_vqmovuntq(uint8x16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_s32)))
+uint16x8_t __arm_vqmovuntq_s32(uint16x8_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_s32)))
+uint16x8_t __arm_vqmovuntq(uint16x8_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_m_s16)))
+int16x8_t __arm_vqnegq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_m_s16)))
+int16x8_t __arm_vqnegq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_m_s32)))
+int32x4_t __arm_vqnegq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_m_s32)))
+int32x4_t __arm_vqnegq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_m_s8)))
+int8x16_t __arm_vqnegq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_m_s8)))
+int8x16_t __arm_vqnegq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_s16)))
+int16x8_t __arm_vqnegq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_s16)))
+int16x8_t __arm_vqnegq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_s32)))
+int32x4_t __arm_vqnegq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_s32)))
+int32x4_t __arm_vqnegq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_s8)))
+int8x16_t __arm_vqnegq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_s8)))
+int8x16_t __arm_vqnegq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_m_s16)))
+int16x8_t __arm_vqrdmladhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_m_s16)))
+int16x8_t __arm_vqrdmladhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_m_s32)))
+int32x4_t __arm_vqrdmladhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_m_s32)))
+int32x4_t __arm_vqrdmladhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_m_s8)))
+int8x16_t __arm_vqrdmladhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_m_s8)))
+int8x16_t __arm_vqrdmladhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_s16)))
+int16x8_t __arm_vqrdmladhq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_s16)))
+int16x8_t __arm_vqrdmladhq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_s32)))
+int32x4_t __arm_vqrdmladhq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_s32)))
+int32x4_t __arm_vqrdmladhq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_s8)))
+int8x16_t __arm_vqrdmladhq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_s8)))
+int8x16_t __arm_vqrdmladhq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_m_s16)))
+int16x8_t __arm_vqrdmladhxq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_m_s16)))
+int16x8_t __arm_vqrdmladhxq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_m_s32)))
+int32x4_t __arm_vqrdmladhxq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_m_s32)))
+int32x4_t __arm_vqrdmladhxq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_m_s8)))
+int8x16_t __arm_vqrdmladhxq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_m_s8)))
+int8x16_t __arm_vqrdmladhxq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_s16)))
+int16x8_t __arm_vqrdmladhxq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_s16)))
+int16x8_t __arm_vqrdmladhxq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_s32)))
+int32x4_t __arm_vqrdmladhxq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_s32)))
+int32x4_t __arm_vqrdmladhxq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_s8)))
+int8x16_t __arm_vqrdmladhxq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_s8)))
+int8x16_t __arm_vqrdmladhxq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_m_n_s16)))
+int16x8_t __arm_vqrdmlahq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_m_n_s16)))
+int16x8_t __arm_vqrdmlahq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_m_n_s32)))
+int32x4_t __arm_vqrdmlahq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_m_n_s32)))
+int32x4_t __arm_vqrdmlahq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_m_n_s8)))
+int8x16_t __arm_vqrdmlahq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_m_n_s8)))
+int8x16_t __arm_vqrdmlahq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_n_s16)))
+int16x8_t __arm_vqrdmlahq_n_s16(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_n_s16)))
+int16x8_t __arm_vqrdmlahq(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_n_s32)))
+int32x4_t __arm_vqrdmlahq_n_s32(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_n_s32)))
+int32x4_t __arm_vqrdmlahq(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_n_s8)))
+int8x16_t __arm_vqrdmlahq_n_s8(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_n_s8)))
+int8x16_t __arm_vqrdmlahq(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_m_n_s16)))
+int16x8_t __arm_vqrdmlashq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_m_n_s16)))
+int16x8_t __arm_vqrdmlashq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_m_n_s32)))
+int32x4_t __arm_vqrdmlashq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_m_n_s32)))
+int32x4_t __arm_vqrdmlashq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_m_n_s8)))
+int8x16_t __arm_vqrdmlashq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_m_n_s8)))
+int8x16_t __arm_vqrdmlashq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_n_s16)))
+int16x8_t __arm_vqrdmlashq_n_s16(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_n_s16)))
+int16x8_t __arm_vqrdmlashq(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_n_s32)))
+int32x4_t __arm_vqrdmlashq_n_s32(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_n_s32)))
+int32x4_t __arm_vqrdmlashq(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_n_s8)))
+int8x16_t __arm_vqrdmlashq_n_s8(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_n_s8)))
+int8x16_t __arm_vqrdmlashq(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_m_s16)))
+int16x8_t __arm_vqrdmlsdhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_m_s16)))
+int16x8_t __arm_vqrdmlsdhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_m_s32)))
+int32x4_t __arm_vqrdmlsdhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_m_s32)))
+int32x4_t __arm_vqrdmlsdhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_m_s8)))
+int8x16_t __arm_vqrdmlsdhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_m_s8)))
+int8x16_t __arm_vqrdmlsdhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_s16)))
+int16x8_t __arm_vqrdmlsdhq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_s16)))
+int16x8_t __arm_vqrdmlsdhq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_s32)))
+int32x4_t __arm_vqrdmlsdhq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_s32)))
+int32x4_t __arm_vqrdmlsdhq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_s8)))
+int8x16_t __arm_vqrdmlsdhq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_s8)))
+int8x16_t __arm_vqrdmlsdhq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_m_s16)))
+int16x8_t __arm_vqrdmlsdhxq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_m_s16)))
+int16x8_t __arm_vqrdmlsdhxq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_m_s32)))
+int32x4_t __arm_vqrdmlsdhxq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_m_s32)))
+int32x4_t __arm_vqrdmlsdhxq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_m_s8)))
+int8x16_t __arm_vqrdmlsdhxq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_m_s8)))
+int8x16_t __arm_vqrdmlsdhxq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_s16)))
+int16x8_t __arm_vqrdmlsdhxq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_s16)))
+int16x8_t __arm_vqrdmlsdhxq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_s32)))
+int32x4_t __arm_vqrdmlsdhxq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_s32)))
+int32x4_t __arm_vqrdmlsdhxq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_s8)))
+int8x16_t __arm_vqrdmlsdhxq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_s8)))
+int8x16_t __arm_vqrdmlsdhxq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_n_s16)))
+int16x8_t __arm_vqrdmulhq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_n_s16)))
+int16x8_t __arm_vqrdmulhq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_n_s32)))
+int32x4_t __arm_vqrdmulhq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_n_s32)))
+int32x4_t __arm_vqrdmulhq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_n_s8)))
+int8x16_t __arm_vqrdmulhq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_n_s8)))
+int8x16_t __arm_vqrdmulhq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_s16)))
+int16x8_t __arm_vqrdmulhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_s16)))
+int16x8_t __arm_vqrdmulhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_s32)))
+int32x4_t __arm_vqrdmulhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_s32)))
+int32x4_t __arm_vqrdmulhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_s8)))
+int8x16_t __arm_vqrdmulhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_s8)))
+int8x16_t __arm_vqrdmulhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_n_s16)))
+int16x8_t __arm_vqrdmulhq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_n_s16)))
+int16x8_t __arm_vqrdmulhq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_n_s32)))
+int32x4_t __arm_vqrdmulhq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_n_s32)))
+int32x4_t __arm_vqrdmulhq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_n_s8)))
+int8x16_t __arm_vqrdmulhq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_n_s8)))
+int8x16_t __arm_vqrdmulhq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_s16)))
+int16x8_t __arm_vqrdmulhq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_s16)))
+int16x8_t __arm_vqrdmulhq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_s32)))
+int32x4_t __arm_vqrdmulhq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_s32)))
+int32x4_t __arm_vqrdmulhq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_s8)))
+int8x16_t __arm_vqrdmulhq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_s8)))
+int8x16_t __arm_vqrdmulhq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_s16)))
+int16x8_t __arm_vqrshlq_m_n_s16(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_s16)))
+int16x8_t __arm_vqrshlq_m_n(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_s32)))
+int32x4_t __arm_vqrshlq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_s32)))
+int32x4_t __arm_vqrshlq_m_n(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_s8)))
+int8x16_t __arm_vqrshlq_m_n_s8(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_s8)))
+int8x16_t __arm_vqrshlq_m_n(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_u16)))
+uint16x8_t __arm_vqrshlq_m_n_u16(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_u16)))
+uint16x8_t __arm_vqrshlq_m_n(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_u32)))
+uint32x4_t __arm_vqrshlq_m_n_u32(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_u32)))
+uint32x4_t __arm_vqrshlq_m_n(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_u8)))
+uint8x16_t __arm_vqrshlq_m_n_u8(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_u8)))
+uint8x16_t __arm_vqrshlq_m_n(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_s16)))
+int16x8_t __arm_vqrshlq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_s16)))
+int16x8_t __arm_vqrshlq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_s32)))
+int32x4_t __arm_vqrshlq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_s32)))
+int32x4_t __arm_vqrshlq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_s8)))
+int8x16_t __arm_vqrshlq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_s8)))
+int8x16_t __arm_vqrshlq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_u16)))
+uint16x8_t __arm_vqrshlq_m_u16(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_u16)))
+uint16x8_t __arm_vqrshlq_m(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_u32)))
+uint32x4_t __arm_vqrshlq_m_u32(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_u32)))
+uint32x4_t __arm_vqrshlq_m(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_u8)))
+uint8x16_t __arm_vqrshlq_m_u8(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_u8)))
+uint8x16_t __arm_vqrshlq_m(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_s16)))
+int16x8_t __arm_vqrshlq_n_s16(int16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_s16)))
+int16x8_t __arm_vqrshlq(int16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_s32)))
+int32x4_t __arm_vqrshlq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_s32)))
+int32x4_t __arm_vqrshlq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_s8)))
+int8x16_t __arm_vqrshlq_n_s8(int8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_s8)))
+int8x16_t __arm_vqrshlq(int8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_u16)))
+uint16x8_t __arm_vqrshlq_n_u16(uint16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_u16)))
+uint16x8_t __arm_vqrshlq(uint16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_u32)))
+uint32x4_t __arm_vqrshlq_n_u32(uint32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_u32)))
+uint32x4_t __arm_vqrshlq(uint32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_u8)))
+uint8x16_t __arm_vqrshlq_n_u8(uint8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_u8)))
+uint8x16_t __arm_vqrshlq(uint8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_s16)))
+int16x8_t __arm_vqrshlq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_s16)))
+int16x8_t __arm_vqrshlq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_s32)))
+int32x4_t __arm_vqrshlq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_s32)))
+int32x4_t __arm_vqrshlq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_s8)))
+int8x16_t __arm_vqrshlq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_s8)))
+int8x16_t __arm_vqrshlq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_u16)))
+uint16x8_t __arm_vqrshlq_u16(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_u16)))
+uint16x8_t __arm_vqrshlq(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_u32)))
+uint32x4_t __arm_vqrshlq_u32(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_u32)))
+uint32x4_t __arm_vqrshlq(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_u8)))
+uint8x16_t __arm_vqrshlq_u8(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_u8)))
+uint8x16_t __arm_vqrshlq(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_s16)))
+int8x16_t __arm_vqrshrnbq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_s16)))
+int8x16_t __arm_vqrshrnbq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_s32)))
+int16x8_t __arm_vqrshrnbq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_s32)))
+int16x8_t __arm_vqrshrnbq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_u16)))
+uint8x16_t __arm_vqrshrnbq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_u16)))
+uint8x16_t __arm_vqrshrnbq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_u32)))
+uint16x8_t __arm_vqrshrnbq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_u32)))
+uint16x8_t __arm_vqrshrnbq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_s16)))
+int8x16_t __arm_vqrshrnbq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_s16)))
+int8x16_t __arm_vqrshrnbq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_s32)))
+int16x8_t __arm_vqrshrnbq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_s32)))
+int16x8_t __arm_vqrshrnbq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_u16)))
+uint8x16_t __arm_vqrshrnbq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_u16)))
+uint8x16_t __arm_vqrshrnbq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_u32)))
+uint16x8_t __arm_vqrshrnbq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_u32)))
+uint16x8_t __arm_vqrshrnbq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_s16)))
+int8x16_t __arm_vqrshrntq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_s16)))
+int8x16_t __arm_vqrshrntq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_s32)))
+int16x8_t __arm_vqrshrntq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_s32)))
+int16x8_t __arm_vqrshrntq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_u16)))
+uint8x16_t __arm_vqrshrntq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_u16)))
+uint8x16_t __arm_vqrshrntq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_u32)))
+uint16x8_t __arm_vqrshrntq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_u32)))
+uint16x8_t __arm_vqrshrntq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_s16)))
+int8x16_t __arm_vqrshrntq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_s16)))
+int8x16_t __arm_vqrshrntq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_s32)))
+int16x8_t __arm_vqrshrntq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_s32)))
+int16x8_t __arm_vqrshrntq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_u16)))
+uint8x16_t __arm_vqrshrntq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_u16)))
+uint8x16_t __arm_vqrshrntq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_u32)))
+uint16x8_t __arm_vqrshrntq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_u32)))
+uint16x8_t __arm_vqrshrntq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_m_n_s16)))
+uint8x16_t __arm_vqrshrunbq_m_n_s16(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_m_n_s16)))
+uint8x16_t __arm_vqrshrunbq_m(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_m_n_s32)))
+uint16x8_t __arm_vqrshrunbq_m_n_s32(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_m_n_s32)))
+uint16x8_t __arm_vqrshrunbq_m(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_n_s16)))
+uint8x16_t __arm_vqrshrunbq_n_s16(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_n_s16)))
+uint8x16_t __arm_vqrshrunbq(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_n_s32)))
+uint16x8_t __arm_vqrshrunbq_n_s32(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_n_s32)))
+uint16x8_t __arm_vqrshrunbq(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_m_n_s16)))
+uint8x16_t __arm_vqrshruntq_m_n_s16(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_m_n_s16)))
+uint8x16_t __arm_vqrshruntq_m(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_m_n_s32)))
+uint16x8_t __arm_vqrshruntq_m_n_s32(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_m_n_s32)))
+uint16x8_t __arm_vqrshruntq_m(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_n_s16)))
+uint8x16_t __arm_vqrshruntq_n_s16(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_n_s16)))
+uint8x16_t __arm_vqrshruntq(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_n_s32)))
+uint16x8_t __arm_vqrshruntq_n_s32(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_n_s32)))
+uint16x8_t __arm_vqrshruntq(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_s16)))
+int16x8_t __arm_vqshlq_m_n_s16(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_s16)))
+int16x8_t __arm_vqshlq_m_n(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_s32)))
+int32x4_t __arm_vqshlq_m_n_s32(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_s32)))
+int32x4_t __arm_vqshlq_m_n(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_s8)))
+int8x16_t __arm_vqshlq_m_n_s8(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_s8)))
+int8x16_t __arm_vqshlq_m_n(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_u16)))
+uint16x8_t __arm_vqshlq_m_n_u16(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_u16)))
+uint16x8_t __arm_vqshlq_m_n(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_u32)))
+uint32x4_t __arm_vqshlq_m_n_u32(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_u32)))
+uint32x4_t __arm_vqshlq_m_n(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_u8)))
+uint8x16_t __arm_vqshlq_m_n_u8(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_u8)))
+uint8x16_t __arm_vqshlq_m_n(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_s16)))
+int16x8_t __arm_vqshlq_m_r_s16(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_s16)))
+int16x8_t __arm_vqshlq_m_r(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_s32)))
+int32x4_t __arm_vqshlq_m_r_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_s32)))
+int32x4_t __arm_vqshlq_m_r(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_s8)))
+int8x16_t __arm_vqshlq_m_r_s8(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_s8)))
+int8x16_t __arm_vqshlq_m_r(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_u16)))
+uint16x8_t __arm_vqshlq_m_r_u16(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_u16)))
+uint16x8_t __arm_vqshlq_m_r(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_u32)))
+uint32x4_t __arm_vqshlq_m_r_u32(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_u32)))
+uint32x4_t __arm_vqshlq_m_r(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_u8)))
+uint8x16_t __arm_vqshlq_m_r_u8(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_u8)))
+uint8x16_t __arm_vqshlq_m_r(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_s16)))
+int16x8_t __arm_vqshlq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_s16)))
+int16x8_t __arm_vqshlq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_s32)))
+int32x4_t __arm_vqshlq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_s32)))
+int32x4_t __arm_vqshlq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_s8)))
+int8x16_t __arm_vqshlq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_s8)))
+int8x16_t __arm_vqshlq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_u16)))
+uint16x8_t __arm_vqshlq_m_u16(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_u16)))
+uint16x8_t __arm_vqshlq_m(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_u32)))
+uint32x4_t __arm_vqshlq_m_u32(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_u32)))
+uint32x4_t __arm_vqshlq_m(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_u8)))
+uint8x16_t __arm_vqshlq_m_u8(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_u8)))
+uint8x16_t __arm_vqshlq_m(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_s16)))
+int16x8_t __arm_vqshlq_n_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_s16)))
+int16x8_t __arm_vqshlq_n(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_s32)))
+int32x4_t __arm_vqshlq_n_s32(int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_s32)))
+int32x4_t __arm_vqshlq_n(int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_s8)))
+int8x16_t __arm_vqshlq_n_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_s8)))
+int8x16_t __arm_vqshlq_n(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_u16)))
+uint16x8_t __arm_vqshlq_n_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_u16)))
+uint16x8_t __arm_vqshlq_n(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_u32)))
+uint32x4_t __arm_vqshlq_n_u32(uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_u32)))
+uint32x4_t __arm_vqshlq_n(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_u8)))
+uint8x16_t __arm_vqshlq_n_u8(uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_u8)))
+uint8x16_t __arm_vqshlq_n(uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_s16)))
+int16x8_t __arm_vqshlq_r_s16(int16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_s16)))
+int16x8_t __arm_vqshlq_r(int16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_s32)))
+int32x4_t __arm_vqshlq_r_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_s32)))
+int32x4_t __arm_vqshlq_r(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_s8)))
+int8x16_t __arm_vqshlq_r_s8(int8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_s8)))
+int8x16_t __arm_vqshlq_r(int8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_u16)))
+uint16x8_t __arm_vqshlq_r_u16(uint16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_u16)))
+uint16x8_t __arm_vqshlq_r(uint16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_u32)))
+uint32x4_t __arm_vqshlq_r_u32(uint32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_u32)))
+uint32x4_t __arm_vqshlq_r(uint32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_u8)))
+uint8x16_t __arm_vqshlq_r_u8(uint8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_u8)))
+uint8x16_t __arm_vqshlq_r(uint8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_s16)))
+int16x8_t __arm_vqshlq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_s16)))
+int16x8_t __arm_vqshlq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_s32)))
+int32x4_t __arm_vqshlq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_s32)))
+int32x4_t __arm_vqshlq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_s8)))
+int8x16_t __arm_vqshlq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_s8)))
+int8x16_t __arm_vqshlq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_u16)))
+uint16x8_t __arm_vqshlq_u16(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_u16)))
+uint16x8_t __arm_vqshlq(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_u32)))
+uint32x4_t __arm_vqshlq_u32(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_u32)))
+uint32x4_t __arm_vqshlq(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_u8)))
+uint8x16_t __arm_vqshlq_u8(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_u8)))
+uint8x16_t __arm_vqshlq(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_m_n_s16)))
+uint16x8_t __arm_vqshluq_m_n_s16(uint16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_m_n_s16)))
+uint16x8_t __arm_vqshluq_m(uint16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_m_n_s32)))
+uint32x4_t __arm_vqshluq_m_n_s32(uint32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_m_n_s32)))
+uint32x4_t __arm_vqshluq_m(uint32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_m_n_s8)))
+uint8x16_t __arm_vqshluq_m_n_s8(uint8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_m_n_s8)))
+uint8x16_t __arm_vqshluq_m(uint8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_n_s16)))
+uint16x8_t __arm_vqshluq_n_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_n_s16)))
+uint16x8_t __arm_vqshluq(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_n_s32)))
+uint32x4_t __arm_vqshluq_n_s32(int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_n_s32)))
+uint32x4_t __arm_vqshluq(int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_n_s8)))
+uint8x16_t __arm_vqshluq_n_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_n_s8)))
+uint8x16_t __arm_vqshluq(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_s16)))
+int8x16_t __arm_vqshrnbq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_s16)))
+int8x16_t __arm_vqshrnbq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_s32)))
+int16x8_t __arm_vqshrnbq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_s32)))
+int16x8_t __arm_vqshrnbq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_u16)))
+uint8x16_t __arm_vqshrnbq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_u16)))
+uint8x16_t __arm_vqshrnbq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_u32)))
+uint16x8_t __arm_vqshrnbq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_u32)))
+uint16x8_t __arm_vqshrnbq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_s16)))
+int8x16_t __arm_vqshrnbq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_s16)))
+int8x16_t __arm_vqshrnbq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_s32)))
+int16x8_t __arm_vqshrnbq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_s32)))
+int16x8_t __arm_vqshrnbq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_u16)))
+uint8x16_t __arm_vqshrnbq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_u16)))
+uint8x16_t __arm_vqshrnbq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_u32)))
+uint16x8_t __arm_vqshrnbq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_u32)))
+uint16x8_t __arm_vqshrnbq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_s16)))
+int8x16_t __arm_vqshrntq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_s16)))
+int8x16_t __arm_vqshrntq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_s32)))
+int16x8_t __arm_vqshrntq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_s32)))
+int16x8_t __arm_vqshrntq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_u16)))
+uint8x16_t __arm_vqshrntq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_u16)))
+uint8x16_t __arm_vqshrntq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_u32)))
+uint16x8_t __arm_vqshrntq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_u32)))
+uint16x8_t __arm_vqshrntq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_s16)))
+int8x16_t __arm_vqshrntq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_s16)))
+int8x16_t __arm_vqshrntq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_s32)))
+int16x8_t __arm_vqshrntq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_s32)))
+int16x8_t __arm_vqshrntq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_u16)))
+uint8x16_t __arm_vqshrntq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_u16)))
+uint8x16_t __arm_vqshrntq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_u32)))
+uint16x8_t __arm_vqshrntq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_u32)))
+uint16x8_t __arm_vqshrntq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_m_n_s16)))
+uint8x16_t __arm_vqshrunbq_m_n_s16(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_m_n_s16)))
+uint8x16_t __arm_vqshrunbq_m(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_m_n_s32)))
+uint16x8_t __arm_vqshrunbq_m_n_s32(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_m_n_s32)))
+uint16x8_t __arm_vqshrunbq_m(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_n_s16)))
+uint8x16_t __arm_vqshrunbq_n_s16(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_n_s16)))
+uint8x16_t __arm_vqshrunbq(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_n_s32)))
+uint16x8_t __arm_vqshrunbq_n_s32(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_n_s32)))
+uint16x8_t __arm_vqshrunbq(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_m_n_s16)))
+uint8x16_t __arm_vqshruntq_m_n_s16(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_m_n_s16)))
+uint8x16_t __arm_vqshruntq_m(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_m_n_s32)))
+uint16x8_t __arm_vqshruntq_m_n_s32(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_m_n_s32)))
+uint16x8_t __arm_vqshruntq_m(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_n_s16)))
+uint8x16_t __arm_vqshruntq_n_s16(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_n_s16)))
+uint8x16_t __arm_vqshruntq(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_n_s32)))
+uint16x8_t __arm_vqshruntq_n_s32(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_n_s32)))
+uint16x8_t __arm_vqshruntq(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_s16)))
+int16x8_t __arm_vqsubq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_s16)))
+int16x8_t __arm_vqsubq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_s32)))
+int32x4_t __arm_vqsubq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_s32)))
+int32x4_t __arm_vqsubq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_s8)))
+int8x16_t __arm_vqsubq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_s8)))
+int8x16_t __arm_vqsubq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_u16)))
+uint16x8_t __arm_vqsubq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_u16)))
+uint16x8_t __arm_vqsubq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_u32)))
+uint32x4_t __arm_vqsubq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_u32)))
+uint32x4_t __arm_vqsubq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_u8)))
+uint8x16_t __arm_vqsubq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_u8)))
+uint8x16_t __arm_vqsubq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_s16)))
+int16x8_t __arm_vqsubq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_s16)))
+int16x8_t __arm_vqsubq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_s32)))
+int32x4_t __arm_vqsubq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_s32)))
+int32x4_t __arm_vqsubq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_s8)))
+int8x16_t __arm_vqsubq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_s8)))
+int8x16_t __arm_vqsubq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_u16)))
+uint16x8_t __arm_vqsubq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_u16)))
+uint16x8_t __arm_vqsubq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_u32)))
+uint32x4_t __arm_vqsubq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_u32)))
+uint32x4_t __arm_vqsubq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_u8)))
+uint8x16_t __arm_vqsubq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_u8)))
+uint8x16_t __arm_vqsubq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_s16)))
+int16x8_t __arm_vqsubq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_s16)))
+int16x8_t __arm_vqsubq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_s32)))
+int32x4_t __arm_vqsubq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_s32)))
+int32x4_t __arm_vqsubq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_s8)))
+int8x16_t __arm_vqsubq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_s8)))
+int8x16_t __arm_vqsubq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_u16)))
+uint16x8_t __arm_vqsubq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_u16)))
+uint16x8_t __arm_vqsubq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_u32)))
+uint32x4_t __arm_vqsubq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_u32)))
+uint32x4_t __arm_vqsubq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_u8)))
+uint8x16_t __arm_vqsubq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_u8)))
+uint8x16_t __arm_vqsubq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_s16)))
+int16x8_t __arm_vqsubq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_s16)))
+int16x8_t __arm_vqsubq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_s32)))
+int32x4_t __arm_vqsubq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_s32)))
+int32x4_t __arm_vqsubq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_s8)))
+int8x16_t __arm_vqsubq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_s8)))
+int8x16_t __arm_vqsubq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_u16)))
+uint16x8_t __arm_vqsubq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_u16)))
+uint16x8_t __arm_vqsubq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_u32)))
+uint32x4_t __arm_vqsubq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_u32)))
+uint32x4_t __arm_vqsubq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_u8)))
+uint8x16_t __arm_vqsubq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_u8)))
+uint8x16_t __arm_vqsubq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_s32)))
+int16x8_t __arm_vreinterpretq_s16_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_s32)))
+int16x8_t __arm_vreinterpretq_s16(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_s64)))
+int16x8_t __arm_vreinterpretq_s16_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_s64)))
+int16x8_t __arm_vreinterpretq_s16(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_s8)))
+int16x8_t __arm_vreinterpretq_s16_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_s8)))
+int16x8_t __arm_vreinterpretq_s16(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u16)))
+int16x8_t __arm_vreinterpretq_s16_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u16)))
+int16x8_t __arm_vreinterpretq_s16(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u32)))
+int16x8_t __arm_vreinterpretq_s16_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u32)))
+int16x8_t __arm_vreinterpretq_s16(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u64)))
+int16x8_t __arm_vreinterpretq_s16_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u64)))
+int16x8_t __arm_vreinterpretq_s16(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u8)))
+int16x8_t __arm_vreinterpretq_s16_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u8)))
+int16x8_t __arm_vreinterpretq_s16(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_s16)))
+int32x4_t __arm_vreinterpretq_s32_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_s16)))
+int32x4_t __arm_vreinterpretq_s32(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_s64)))
+int32x4_t __arm_vreinterpretq_s32_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_s64)))
+int32x4_t __arm_vreinterpretq_s32(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_s8)))
+int32x4_t __arm_vreinterpretq_s32_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_s8)))
+int32x4_t __arm_vreinterpretq_s32(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u16)))
+int32x4_t __arm_vreinterpretq_s32_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u16)))
+int32x4_t __arm_vreinterpretq_s32(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u32)))
+int32x4_t __arm_vreinterpretq_s32_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u32)))
+int32x4_t __arm_vreinterpretq_s32(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u64)))
+int32x4_t __arm_vreinterpretq_s32_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u64)))
+int32x4_t __arm_vreinterpretq_s32(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u8)))
+int32x4_t __arm_vreinterpretq_s32_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u8)))
+int32x4_t __arm_vreinterpretq_s32(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_s16)))
+int64x2_t __arm_vreinterpretq_s64_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_s16)))
+int64x2_t __arm_vreinterpretq_s64(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_s32)))
+int64x2_t __arm_vreinterpretq_s64_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_s32)))
+int64x2_t __arm_vreinterpretq_s64(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_s8)))
+int64x2_t __arm_vreinterpretq_s64_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_s8)))
+int64x2_t __arm_vreinterpretq_s64(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u16)))
+int64x2_t __arm_vreinterpretq_s64_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u16)))
+int64x2_t __arm_vreinterpretq_s64(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u32)))
+int64x2_t __arm_vreinterpretq_s64_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u32)))
+int64x2_t __arm_vreinterpretq_s64(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u64)))
+int64x2_t __arm_vreinterpretq_s64_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u64)))
+int64x2_t __arm_vreinterpretq_s64(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u8)))
+int64x2_t __arm_vreinterpretq_s64_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u8)))
+int64x2_t __arm_vreinterpretq_s64(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_s16)))
+int8x16_t __arm_vreinterpretq_s8_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_s16)))
+int8x16_t __arm_vreinterpretq_s8(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_s32)))
+int8x16_t __arm_vreinterpretq_s8_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_s32)))
+int8x16_t __arm_vreinterpretq_s8(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_s64)))
+int8x16_t __arm_vreinterpretq_s8_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_s64)))
+int8x16_t __arm_vreinterpretq_s8(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u16)))
+int8x16_t __arm_vreinterpretq_s8_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u16)))
+int8x16_t __arm_vreinterpretq_s8(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u32)))
+int8x16_t __arm_vreinterpretq_s8_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u32)))
+int8x16_t __arm_vreinterpretq_s8(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u64)))
+int8x16_t __arm_vreinterpretq_s8_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u64)))
+int8x16_t __arm_vreinterpretq_s8(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u8)))
+int8x16_t __arm_vreinterpretq_s8_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u8)))
+int8x16_t __arm_vreinterpretq_s8(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s16)))
+uint16x8_t __arm_vreinterpretq_u16_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s16)))
+uint16x8_t __arm_vreinterpretq_u16(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s32)))
+uint16x8_t __arm_vreinterpretq_u16_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s32)))
+uint16x8_t __arm_vreinterpretq_u16(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s64)))
+uint16x8_t __arm_vreinterpretq_u16_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s64)))
+uint16x8_t __arm_vreinterpretq_u16(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s8)))
+uint16x8_t __arm_vreinterpretq_u16_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s8)))
+uint16x8_t __arm_vreinterpretq_u16(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_u32)))
+uint16x8_t __arm_vreinterpretq_u16_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_u32)))
+uint16x8_t __arm_vreinterpretq_u16(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_u64)))
+uint16x8_t __arm_vreinterpretq_u16_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_u64)))
+uint16x8_t __arm_vreinterpretq_u16(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_u8)))
+uint16x8_t __arm_vreinterpretq_u16_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_u8)))
+uint16x8_t __arm_vreinterpretq_u16(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s16)))
+uint32x4_t __arm_vreinterpretq_u32_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s16)))
+uint32x4_t __arm_vreinterpretq_u32(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s32)))
+uint32x4_t __arm_vreinterpretq_u32_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s32)))
+uint32x4_t __arm_vreinterpretq_u32(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s64)))
+uint32x4_t __arm_vreinterpretq_u32_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s64)))
+uint32x4_t __arm_vreinterpretq_u32(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s8)))
+uint32x4_t __arm_vreinterpretq_u32_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s8)))
+uint32x4_t __arm_vreinterpretq_u32(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_u16)))
+uint32x4_t __arm_vreinterpretq_u32_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_u16)))
+uint32x4_t __arm_vreinterpretq_u32(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_u64)))
+uint32x4_t __arm_vreinterpretq_u32_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_u64)))
+uint32x4_t __arm_vreinterpretq_u32(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_u8)))
+uint32x4_t __arm_vreinterpretq_u32_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_u8)))
+uint32x4_t __arm_vreinterpretq_u32(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s16)))
+uint64x2_t __arm_vreinterpretq_u64_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s16)))
+uint64x2_t __arm_vreinterpretq_u64(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s32)))
+uint64x2_t __arm_vreinterpretq_u64_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s32)))
+uint64x2_t __arm_vreinterpretq_u64(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s64)))
+uint64x2_t __arm_vreinterpretq_u64_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s64)))
+uint64x2_t __arm_vreinterpretq_u64(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s8)))
+uint64x2_t __arm_vreinterpretq_u64_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s8)))
+uint64x2_t __arm_vreinterpretq_u64(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_u16)))
+uint64x2_t __arm_vreinterpretq_u64_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_u16)))
+uint64x2_t __arm_vreinterpretq_u64(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_u32)))
+uint64x2_t __arm_vreinterpretq_u64_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_u32)))
+uint64x2_t __arm_vreinterpretq_u64(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_u8)))
+uint64x2_t __arm_vreinterpretq_u64_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_u8)))
+uint64x2_t __arm_vreinterpretq_u64(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s16)))
+uint8x16_t __arm_vreinterpretq_u8_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s16)))
+uint8x16_t __arm_vreinterpretq_u8(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s32)))
+uint8x16_t __arm_vreinterpretq_u8_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s32)))
+uint8x16_t __arm_vreinterpretq_u8(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s64)))
+uint8x16_t __arm_vreinterpretq_u8_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s64)))
+uint8x16_t __arm_vreinterpretq_u8(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s8)))
+uint8x16_t __arm_vreinterpretq_u8_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s8)))
+uint8x16_t __arm_vreinterpretq_u8(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_u16)))
+uint8x16_t __arm_vreinterpretq_u8_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_u16)))
+uint8x16_t __arm_vreinterpretq_u8(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_u32)))
+uint8x16_t __arm_vreinterpretq_u8_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_u32)))
+uint8x16_t __arm_vreinterpretq_u8(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_u64)))
+uint8x16_t __arm_vreinterpretq_u8_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_u64)))
+uint8x16_t __arm_vreinterpretq_u8(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_m_s8)))
+int8x16_t __arm_vrev16q_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_m_s8)))
+int8x16_t __arm_vrev16q_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_m_u8)))
+uint8x16_t __arm_vrev16q_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_m_u8)))
+uint8x16_t __arm_vrev16q_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_s8)))
+int8x16_t __arm_vrev16q_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_s8)))
+int8x16_t __arm_vrev16q(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_u8)))
+uint8x16_t __arm_vrev16q_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_u8)))
+uint8x16_t __arm_vrev16q(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_x_s8)))
+int8x16_t __arm_vrev16q_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_x_s8)))
+int8x16_t __arm_vrev16q_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_x_u8)))
+uint8x16_t __arm_vrev16q_x_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_x_u8)))
+uint8x16_t __arm_vrev16q_x(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_s16)))
+int16x8_t __arm_vrev32q_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_s16)))
+int16x8_t __arm_vrev32q_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_s8)))
+int8x16_t __arm_vrev32q_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_s8)))
+int8x16_t __arm_vrev32q_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_u16)))
+uint16x8_t __arm_vrev32q_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_u16)))
+uint16x8_t __arm_vrev32q_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_u8)))
+uint8x16_t __arm_vrev32q_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_u8)))
+uint8x16_t __arm_vrev32q_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_s16)))
+int16x8_t __arm_vrev32q_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_s16)))
+int16x8_t __arm_vrev32q(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_s8)))
+int8x16_t __arm_vrev32q_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_s8)))
+int8x16_t __arm_vrev32q(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_u16)))
+uint16x8_t __arm_vrev32q_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_u16)))
+uint16x8_t __arm_vrev32q(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_u8)))
+uint8x16_t __arm_vrev32q_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_u8)))
+uint8x16_t __arm_vrev32q(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_s16)))
+int16x8_t __arm_vrev32q_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_s16)))
+int16x8_t __arm_vrev32q_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_s8)))
+int8x16_t __arm_vrev32q_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_s8)))
+int8x16_t __arm_vrev32q_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_u16)))
+uint16x8_t __arm_vrev32q_x_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_u16)))
+uint16x8_t __arm_vrev32q_x(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_u8)))
+uint8x16_t __arm_vrev32q_x_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_u8)))
+uint8x16_t __arm_vrev32q_x(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_s16)))
+int16x8_t __arm_vrev64q_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_s16)))
+int16x8_t __arm_vrev64q_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_s32)))
+int32x4_t __arm_vrev64q_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_s32)))
+int32x4_t __arm_vrev64q_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_s8)))
+int8x16_t __arm_vrev64q_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_s8)))
+int8x16_t __arm_vrev64q_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_u16)))
+uint16x8_t __arm_vrev64q_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_u16)))
+uint16x8_t __arm_vrev64q_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_u32)))
+uint32x4_t __arm_vrev64q_m_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_u32)))
+uint32x4_t __arm_vrev64q_m(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_u8)))
+uint8x16_t __arm_vrev64q_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_u8)))
+uint8x16_t __arm_vrev64q_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_s16)))
+int16x8_t __arm_vrev64q_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_s16)))
+int16x8_t __arm_vrev64q(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_s32)))
+int32x4_t __arm_vrev64q_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_s32)))
+int32x4_t __arm_vrev64q(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_s8)))
+int8x16_t __arm_vrev64q_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_s8)))
+int8x16_t __arm_vrev64q(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_u16)))
+uint16x8_t __arm_vrev64q_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_u16)))
+uint16x8_t __arm_vrev64q(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_u32)))
+uint32x4_t __arm_vrev64q_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_u32)))
+uint32x4_t __arm_vrev64q(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_u8)))
+uint8x16_t __arm_vrev64q_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_u8)))
+uint8x16_t __arm_vrev64q(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_s16)))
+int16x8_t __arm_vrev64q_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_s16)))
+int16x8_t __arm_vrev64q_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_s32)))
+int32x4_t __arm_vrev64q_x_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_s32)))
+int32x4_t __arm_vrev64q_x(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_s8)))
+int8x16_t __arm_vrev64q_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_s8)))
+int8x16_t __arm_vrev64q_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_u16)))
+uint16x8_t __arm_vrev64q_x_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_u16)))
+uint16x8_t __arm_vrev64q_x(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_u32)))
+uint32x4_t __arm_vrev64q_x_u32(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_u32)))
+uint32x4_t __arm_vrev64q_x(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_u8)))
+uint8x16_t __arm_vrev64q_x_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_u8)))
+uint8x16_t __arm_vrev64q_x(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_s16)))
+int16x8_t __arm_vrhaddq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_s16)))
+int16x8_t __arm_vrhaddq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_s32)))
+int32x4_t __arm_vrhaddq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_s32)))
+int32x4_t __arm_vrhaddq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_s8)))
+int8x16_t __arm_vrhaddq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_s8)))
+int8x16_t __arm_vrhaddq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_u16)))
+uint16x8_t __arm_vrhaddq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_u16)))
+uint16x8_t __arm_vrhaddq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_u32)))
+uint32x4_t __arm_vrhaddq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_u32)))
+uint32x4_t __arm_vrhaddq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_u8)))
+uint8x16_t __arm_vrhaddq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_u8)))
+uint8x16_t __arm_vrhaddq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_s16)))
+int16x8_t __arm_vrhaddq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_s16)))
+int16x8_t __arm_vrhaddq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_s32)))
+int32x4_t __arm_vrhaddq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_s32)))
+int32x4_t __arm_vrhaddq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_s8)))
+int8x16_t __arm_vrhaddq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_s8)))
+int8x16_t __arm_vrhaddq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_u16)))
+uint16x8_t __arm_vrhaddq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_u16)))
+uint16x8_t __arm_vrhaddq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_u32)))
+uint32x4_t __arm_vrhaddq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_u32)))
+uint32x4_t __arm_vrhaddq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_u8)))
+uint8x16_t __arm_vrhaddq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_u8)))
+uint8x16_t __arm_vrhaddq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_s16)))
+int16x8_t __arm_vrhaddq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_s16)))
+int16x8_t __arm_vrhaddq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_s32)))
+int32x4_t __arm_vrhaddq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_s32)))
+int32x4_t __arm_vrhaddq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_s8)))
+int8x16_t __arm_vrhaddq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_s8)))
+int8x16_t __arm_vrhaddq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_u16)))
+uint16x8_t __arm_vrhaddq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_u16)))
+uint16x8_t __arm_vrhaddq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_u32)))
+uint32x4_t __arm_vrhaddq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_u32)))
+uint32x4_t __arm_vrhaddq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_u8)))
+uint8x16_t __arm_vrhaddq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_u8)))
+uint8x16_t __arm_vrhaddq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_p_s32)))
+int64_t __arm_vrmlaldavhaq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_p_s32)))
+int64_t __arm_vrmlaldavhaq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_p_u32)))
+uint64_t __arm_vrmlaldavhaq_p_u32(uint64_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_p_u32)))
+uint64_t __arm_vrmlaldavhaq_p(uint64_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_s32)))
+int64_t __arm_vrmlaldavhaq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_s32)))
+int64_t __arm_vrmlaldavhaq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_u32)))
+uint64_t __arm_vrmlaldavhaq_u32(uint64_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_u32)))
+uint64_t __arm_vrmlaldavhaq(uint64_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaxq_p_s32)))
+int64_t __arm_vrmlaldavhaxq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaxq_p_s32)))
+int64_t __arm_vrmlaldavhaxq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaxq_s32)))
+int64_t __arm_vrmlaldavhaxq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaxq_s32)))
+int64_t __arm_vrmlaldavhaxq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_p_s32)))
+int64_t __arm_vrmlaldavhq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_p_s32)))
+int64_t __arm_vrmlaldavhq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_p_u32)))
+uint64_t __arm_vrmlaldavhq_p_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_p_u32)))
+uint64_t __arm_vrmlaldavhq_p(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_s32)))
+int64_t __arm_vrmlaldavhq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_s32)))
+int64_t __arm_vrmlaldavhq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_u32)))
+uint64_t __arm_vrmlaldavhq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_u32)))
+uint64_t __arm_vrmlaldavhq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhxq_p_s32)))
+int64_t __arm_vrmlaldavhxq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhxq_p_s32)))
+int64_t __arm_vrmlaldavhxq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhxq_s32)))
+int64_t __arm_vrmlaldavhxq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhxq_s32)))
+int64_t __arm_vrmlaldavhxq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaq_p_s32)))
+int64_t __arm_vrmlsldavhaq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaq_p_s32)))
+int64_t __arm_vrmlsldavhaq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaq_s32)))
+int64_t __arm_vrmlsldavhaq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaq_s32)))
+int64_t __arm_vrmlsldavhaq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaxq_p_s32)))
+int64_t __arm_vrmlsldavhaxq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaxq_p_s32)))
+int64_t __arm_vrmlsldavhaxq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaxq_s32)))
+int64_t __arm_vrmlsldavhaxq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaxq_s32)))
+int64_t __arm_vrmlsldavhaxq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhq_p_s32)))
+int64_t __arm_vrmlsldavhq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhq_p_s32)))
+int64_t __arm_vrmlsldavhq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhq_s32)))
+int64_t __arm_vrmlsldavhq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhq_s32)))
+int64_t __arm_vrmlsldavhq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhxq_p_s32)))
+int64_t __arm_vrmlsldavhxq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhxq_p_s32)))
+int64_t __arm_vrmlsldavhxq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhxq_s32)))
+int64_t __arm_vrmlsldavhxq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhxq_s32)))
+int64_t __arm_vrmlsldavhxq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_s16)))
+int16x8_t __arm_vrmulhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_s16)))
+int16x8_t __arm_vrmulhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_s32)))
+int32x4_t __arm_vrmulhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_s32)))
+int32x4_t __arm_vrmulhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_s8)))
+int8x16_t __arm_vrmulhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_s8)))
+int8x16_t __arm_vrmulhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_u16)))
+uint16x8_t __arm_vrmulhq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_u16)))
+uint16x8_t __arm_vrmulhq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_u32)))
+uint32x4_t __arm_vrmulhq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_u32)))
+uint32x4_t __arm_vrmulhq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_u8)))
+uint8x16_t __arm_vrmulhq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_u8)))
+uint8x16_t __arm_vrmulhq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_s16)))
+int16x8_t __arm_vrmulhq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_s16)))
+int16x8_t __arm_vrmulhq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_s32)))
+int32x4_t __arm_vrmulhq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_s32)))
+int32x4_t __arm_vrmulhq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_s8)))
+int8x16_t __arm_vrmulhq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_s8)))
+int8x16_t __arm_vrmulhq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_u16)))
+uint16x8_t __arm_vrmulhq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_u16)))
+uint16x8_t __arm_vrmulhq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_u32)))
+uint32x4_t __arm_vrmulhq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_u32)))
+uint32x4_t __arm_vrmulhq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_u8)))
+uint8x16_t __arm_vrmulhq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_u8)))
+uint8x16_t __arm_vrmulhq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_s16)))
+int16x8_t __arm_vrmulhq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_s16)))
+int16x8_t __arm_vrmulhq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_s32)))
+int32x4_t __arm_vrmulhq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_s32)))
+int32x4_t __arm_vrmulhq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_s8)))
+int8x16_t __arm_vrmulhq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_s8)))
+int8x16_t __arm_vrmulhq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_u16)))
+uint16x8_t __arm_vrmulhq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_u16)))
+uint16x8_t __arm_vrmulhq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_u32)))
+uint32x4_t __arm_vrmulhq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_u32)))
+uint32x4_t __arm_vrmulhq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_u8)))
+uint8x16_t __arm_vrmulhq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_u8)))
+uint8x16_t __arm_vrmulhq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_s16)))
+int16x8_t __arm_vrshlq_m_n_s16(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_s16)))
+int16x8_t __arm_vrshlq_m_n(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_s32)))
+int32x4_t __arm_vrshlq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_s32)))
+int32x4_t __arm_vrshlq_m_n(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_s8)))
+int8x16_t __arm_vrshlq_m_n_s8(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_s8)))
+int8x16_t __arm_vrshlq_m_n(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_u16)))
+uint16x8_t __arm_vrshlq_m_n_u16(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_u16)))
+uint16x8_t __arm_vrshlq_m_n(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_u32)))
+uint32x4_t __arm_vrshlq_m_n_u32(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_u32)))
+uint32x4_t __arm_vrshlq_m_n(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_u8)))
+uint8x16_t __arm_vrshlq_m_n_u8(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_u8)))
+uint8x16_t __arm_vrshlq_m_n(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_s16)))
+int16x8_t __arm_vrshlq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_s16)))
+int16x8_t __arm_vrshlq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_s32)))
+int32x4_t __arm_vrshlq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_s32)))
+int32x4_t __arm_vrshlq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_s8)))
+int8x16_t __arm_vrshlq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_s8)))
+int8x16_t __arm_vrshlq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_u16)))
+uint16x8_t __arm_vrshlq_m_u16(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_u16)))
+uint16x8_t __arm_vrshlq_m(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_u32)))
+uint32x4_t __arm_vrshlq_m_u32(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_u32)))
+uint32x4_t __arm_vrshlq_m(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_u8)))
+uint8x16_t __arm_vrshlq_m_u8(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_u8)))
+uint8x16_t __arm_vrshlq_m(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_s16)))
+int16x8_t __arm_vrshlq_n_s16(int16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_s16)))
+int16x8_t __arm_vrshlq(int16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_s32)))
+int32x4_t __arm_vrshlq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_s32)))
+int32x4_t __arm_vrshlq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_s8)))
+int8x16_t __arm_vrshlq_n_s8(int8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_s8)))
+int8x16_t __arm_vrshlq(int8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_u16)))
+uint16x8_t __arm_vrshlq_n_u16(uint16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_u16)))
+uint16x8_t __arm_vrshlq(uint16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_u32)))
+uint32x4_t __arm_vrshlq_n_u32(uint32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_u32)))
+uint32x4_t __arm_vrshlq(uint32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_u8)))
+uint8x16_t __arm_vrshlq_n_u8(uint8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_u8)))
+uint8x16_t __arm_vrshlq(uint8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_s16)))
+int16x8_t __arm_vrshlq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_s16)))
+int16x8_t __arm_vrshlq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_s32)))
+int32x4_t __arm_vrshlq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_s32)))
+int32x4_t __arm_vrshlq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_s8)))
+int8x16_t __arm_vrshlq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_s8)))
+int8x16_t __arm_vrshlq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_u16)))
+uint16x8_t __arm_vrshlq_u16(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_u16)))
+uint16x8_t __arm_vrshlq(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_u32)))
+uint32x4_t __arm_vrshlq_u32(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_u32)))
+uint32x4_t __arm_vrshlq(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_u8)))
+uint8x16_t __arm_vrshlq_u8(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_u8)))
+uint8x16_t __arm_vrshlq(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_s16)))
+int16x8_t __arm_vrshlq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_s16)))
+int16x8_t __arm_vrshlq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_s32)))
+int32x4_t __arm_vrshlq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_s32)))
+int32x4_t __arm_vrshlq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_s8)))
+int8x16_t __arm_vrshlq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_s8)))
+int8x16_t __arm_vrshlq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_u16)))
+uint16x8_t __arm_vrshlq_x_u16(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_u16)))
+uint16x8_t __arm_vrshlq_x(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_u32)))
+uint32x4_t __arm_vrshlq_x_u32(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_u32)))
+uint32x4_t __arm_vrshlq_x(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_u8)))
+uint8x16_t __arm_vrshlq_x_u8(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_u8)))
+uint8x16_t __arm_vrshlq_x(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_s16)))
+int8x16_t __arm_vrshrnbq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_s16)))
+int8x16_t __arm_vrshrnbq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_s32)))
+int16x8_t __arm_vrshrnbq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_s32)))
+int16x8_t __arm_vrshrnbq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_u16)))
+uint8x16_t __arm_vrshrnbq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_u16)))
+uint8x16_t __arm_vrshrnbq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_u32)))
+uint16x8_t __arm_vrshrnbq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_u32)))
+uint16x8_t __arm_vrshrnbq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_s16)))
+int8x16_t __arm_vrshrnbq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_s16)))
+int8x16_t __arm_vrshrnbq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_s32)))
+int16x8_t __arm_vrshrnbq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_s32)))
+int16x8_t __arm_vrshrnbq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_u16)))
+uint8x16_t __arm_vrshrnbq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_u16)))
+uint8x16_t __arm_vrshrnbq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_u32)))
+uint16x8_t __arm_vrshrnbq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_u32)))
+uint16x8_t __arm_vrshrnbq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_s16)))
+int8x16_t __arm_vrshrntq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_s16)))
+int8x16_t __arm_vrshrntq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_s32)))
+int16x8_t __arm_vrshrntq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_s32)))
+int16x8_t __arm_vrshrntq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_u16)))
+uint8x16_t __arm_vrshrntq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_u16)))
+uint8x16_t __arm_vrshrntq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_u32)))
+uint16x8_t __arm_vrshrntq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_u32)))
+uint16x8_t __arm_vrshrntq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_s16)))
+int8x16_t __arm_vrshrntq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_s16)))
+int8x16_t __arm_vrshrntq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_s32)))
+int16x8_t __arm_vrshrntq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_s32)))
+int16x8_t __arm_vrshrntq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_u16)))
+uint8x16_t __arm_vrshrntq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_u16)))
+uint8x16_t __arm_vrshrntq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_u32)))
+uint16x8_t __arm_vrshrntq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_u32)))
+uint16x8_t __arm_vrshrntq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_s16)))
+int16x8_t __arm_vrshrq_m_n_s16(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_s16)))
+int16x8_t __arm_vrshrq_m(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_s32)))
+int32x4_t __arm_vrshrq_m_n_s32(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_s32)))
+int32x4_t __arm_vrshrq_m(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_s8)))
+int8x16_t __arm_vrshrq_m_n_s8(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_s8)))
+int8x16_t __arm_vrshrq_m(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_u16)))
+uint16x8_t __arm_vrshrq_m_n_u16(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_u16)))
+uint16x8_t __arm_vrshrq_m(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_u32)))
+uint32x4_t __arm_vrshrq_m_n_u32(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_u32)))
+uint32x4_t __arm_vrshrq_m(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_u8)))
+uint8x16_t __arm_vrshrq_m_n_u8(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_u8)))
+uint8x16_t __arm_vrshrq_m(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_s16)))
+int16x8_t __arm_vrshrq_n_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_s16)))
+int16x8_t __arm_vrshrq(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_s32)))
+int32x4_t __arm_vrshrq_n_s32(int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_s32)))
+int32x4_t __arm_vrshrq(int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_s8)))
+int8x16_t __arm_vrshrq_n_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_s8)))
+int8x16_t __arm_vrshrq(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_u16)))
+uint16x8_t __arm_vrshrq_n_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_u16)))
+uint16x8_t __arm_vrshrq(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_u32)))
+uint32x4_t __arm_vrshrq_n_u32(uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_u32)))
+uint32x4_t __arm_vrshrq(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_u8)))
+uint8x16_t __arm_vrshrq_n_u8(uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_u8)))
+uint8x16_t __arm_vrshrq(uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_s16)))
+int16x8_t __arm_vrshrq_x_n_s16(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_s16)))
+int16x8_t __arm_vrshrq_x(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_s32)))
+int32x4_t __arm_vrshrq_x_n_s32(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_s32)))
+int32x4_t __arm_vrshrq_x(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_s8)))
+int8x16_t __arm_vrshrq_x_n_s8(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_s8)))
+int8x16_t __arm_vrshrq_x(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_u16)))
+uint16x8_t __arm_vrshrq_x_n_u16(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_u16)))
+uint16x8_t __arm_vrshrq_x(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_u32)))
+uint32x4_t __arm_vrshrq_x_n_u32(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_u32)))
+uint32x4_t __arm_vrshrq_x(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_u8)))
+uint8x16_t __arm_vrshrq_x_n_u8(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_u8)))
+uint8x16_t __arm_vrshrq_x(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_m_s32)))
+int32x4_t __arm_vsbciq_m_s32(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_m_s32)))
+int32x4_t __arm_vsbciq_m(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_m_u32)))
+uint32x4_t __arm_vsbciq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_m_u32)))
+uint32x4_t __arm_vsbciq_m(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_s32)))
+int32x4_t __arm_vsbciq_s32(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_s32)))
+int32x4_t __arm_vsbciq(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_u32)))
+uint32x4_t __arm_vsbciq_u32(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_u32)))
+uint32x4_t __arm_vsbciq(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_m_s32)))
+int32x4_t __arm_vsbcq_m_s32(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_m_s32)))
+int32x4_t __arm_vsbcq_m(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_m_u32)))
+uint32x4_t __arm_vsbcq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_m_u32)))
+uint32x4_t __arm_vsbcq_m(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_s32)))
+int32x4_t __arm_vsbcq_s32(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_s32)))
+int32x4_t __arm_vsbcq(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_u32)))
+uint32x4_t __arm_vsbcq_u32(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_u32)))
+uint32x4_t __arm_vsbcq(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s16)))
+int16x8_t __arm_vsetq_lane_s16(int16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s16)))
+int16x8_t __arm_vsetq_lane(int16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s32)))
+int32x4_t __arm_vsetq_lane_s32(int32_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s32)))
+int32x4_t __arm_vsetq_lane(int32_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s64)))
+int64x2_t __arm_vsetq_lane_s64(int64_t, int64x2_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s64)))
+int64x2_t __arm_vsetq_lane(int64_t, int64x2_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s8)))
+int8x16_t __arm_vsetq_lane_s8(int8_t, int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s8)))
+int8x16_t __arm_vsetq_lane(int8_t, int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u16)))
+uint16x8_t __arm_vsetq_lane_u16(uint16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u16)))
+uint16x8_t __arm_vsetq_lane(uint16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u32)))
+uint32x4_t __arm_vsetq_lane_u32(uint32_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u32)))
+uint32x4_t __arm_vsetq_lane(uint32_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u64)))
+uint64x2_t __arm_vsetq_lane_u64(uint64_t, uint64x2_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u64)))
+uint64x2_t __arm_vsetq_lane(uint64_t, uint64x2_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u8)))
+uint8x16_t __arm_vsetq_lane_u8(uint8_t, uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u8)))
+uint8x16_t __arm_vsetq_lane(uint8_t, uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_s16)))
+int16x8_t __arm_vshlcq_m_s16(int16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_s16)))
+int16x8_t __arm_vshlcq_m(int16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_s32)))
+int32x4_t __arm_vshlcq_m_s32(int32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_s32)))
+int32x4_t __arm_vshlcq_m(int32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_s8)))
+int8x16_t __arm_vshlcq_m_s8(int8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_s8)))
+int8x16_t __arm_vshlcq_m(int8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_u16)))
+uint16x8_t __arm_vshlcq_m_u16(uint16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_u16)))
+uint16x8_t __arm_vshlcq_m(uint16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_u32)))
+uint32x4_t __arm_vshlcq_m_u32(uint32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_u32)))
+uint32x4_t __arm_vshlcq_m(uint32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_u8)))
+uint8x16_t __arm_vshlcq_m_u8(uint8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_u8)))
+uint8x16_t __arm_vshlcq_m(uint8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_s16)))
+int16x8_t __arm_vshlcq_s16(int16x8_t, uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_s16)))
+int16x8_t __arm_vshlcq(int16x8_t, uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_s32)))
+int32x4_t __arm_vshlcq_s32(int32x4_t, uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_s32)))
+int32x4_t __arm_vshlcq(int32x4_t, uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_s8)))
+int8x16_t __arm_vshlcq_s8(int8x16_t, uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_s8)))
+int8x16_t __arm_vshlcq(int8x16_t, uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_u16)))
+uint16x8_t __arm_vshlcq_u16(uint16x8_t, uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_u16)))
+uint16x8_t __arm_vshlcq(uint16x8_t, uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_u32)))
+uint32x4_t __arm_vshlcq_u32(uint32x4_t, uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_u32)))
+uint32x4_t __arm_vshlcq(uint32x4_t, uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_u8)))
+uint8x16_t __arm_vshlcq_u8(uint8x16_t, uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_u8)))
+uint8x16_t __arm_vshlcq(uint8x16_t, uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_s16)))
+int32x4_t __arm_vshllbq_m_n_s16(int32x4_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_s16)))
+int32x4_t __arm_vshllbq_m(int32x4_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_s8)))
+int16x8_t __arm_vshllbq_m_n_s8(int16x8_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_s8)))
+int16x8_t __arm_vshllbq_m(int16x8_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_u16)))
+uint32x4_t __arm_vshllbq_m_n_u16(uint32x4_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_u16)))
+uint32x4_t __arm_vshllbq_m(uint32x4_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_u8)))
+uint16x8_t __arm_vshllbq_m_n_u8(uint16x8_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_u8)))
+uint16x8_t __arm_vshllbq_m(uint16x8_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_s16)))
+int32x4_t __arm_vshllbq_n_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_s16)))
+int32x4_t __arm_vshllbq(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_s8)))
+int16x8_t __arm_vshllbq_n_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_s8)))
+int16x8_t __arm_vshllbq(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_u16)))
+uint32x4_t __arm_vshllbq_n_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_u16)))
+uint32x4_t __arm_vshllbq(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_u8)))
+uint16x8_t __arm_vshllbq_n_u8(uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_u8)))
+uint16x8_t __arm_vshllbq(uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_s16)))
+int32x4_t __arm_vshllbq_x_n_s16(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_s16)))
+int32x4_t __arm_vshllbq_x(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_s8)))
+int16x8_t __arm_vshllbq_x_n_s8(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_s8)))
+int16x8_t __arm_vshllbq_x(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_u16)))
+uint32x4_t __arm_vshllbq_x_n_u16(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_u16)))
+uint32x4_t __arm_vshllbq_x(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_u8)))
+uint16x8_t __arm_vshllbq_x_n_u8(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_u8)))
+uint16x8_t __arm_vshllbq_x(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_s16)))
+int32x4_t __arm_vshlltq_m_n_s16(int32x4_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_s16)))
+int32x4_t __arm_vshlltq_m(int32x4_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_s8)))
+int16x8_t __arm_vshlltq_m_n_s8(int16x8_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_s8)))
+int16x8_t __arm_vshlltq_m(int16x8_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_u16)))
+uint32x4_t __arm_vshlltq_m_n_u16(uint32x4_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_u16)))
+uint32x4_t __arm_vshlltq_m(uint32x4_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_u8)))
+uint16x8_t __arm_vshlltq_m_n_u8(uint16x8_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_u8)))
+uint16x8_t __arm_vshlltq_m(uint16x8_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_s16)))
+int32x4_t __arm_vshlltq_n_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_s16)))
+int32x4_t __arm_vshlltq(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_s8)))
+int16x8_t __arm_vshlltq_n_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_s8)))
+int16x8_t __arm_vshlltq(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_u16)))
+uint32x4_t __arm_vshlltq_n_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_u16)))
+uint32x4_t __arm_vshlltq(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_u8)))
+uint16x8_t __arm_vshlltq_n_u8(uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_u8)))
+uint16x8_t __arm_vshlltq(uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_s16)))
+int32x4_t __arm_vshlltq_x_n_s16(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_s16)))
+int32x4_t __arm_vshlltq_x(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_s8)))
+int16x8_t __arm_vshlltq_x_n_s8(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_s8)))
+int16x8_t __arm_vshlltq_x(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_u16)))
+uint32x4_t __arm_vshlltq_x_n_u16(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_u16)))
+uint32x4_t __arm_vshlltq_x(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_u8)))
+uint16x8_t __arm_vshlltq_x_n_u8(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_u8)))
+uint16x8_t __arm_vshlltq_x(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_s16)))
+int16x8_t __arm_vshlq_m_n_s16(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_s16)))
+int16x8_t __arm_vshlq_m_n(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_s32)))
+int32x4_t __arm_vshlq_m_n_s32(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_s32)))
+int32x4_t __arm_vshlq_m_n(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_s8)))
+int8x16_t __arm_vshlq_m_n_s8(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_s8)))
+int8x16_t __arm_vshlq_m_n(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_u16)))
+uint16x8_t __arm_vshlq_m_n_u16(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_u16)))
+uint16x8_t __arm_vshlq_m_n(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_u32)))
+uint32x4_t __arm_vshlq_m_n_u32(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_u32)))
+uint32x4_t __arm_vshlq_m_n(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_u8)))
+uint8x16_t __arm_vshlq_m_n_u8(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_u8)))
+uint8x16_t __arm_vshlq_m_n(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_s16)))
+int16x8_t __arm_vshlq_m_r_s16(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_s16)))
+int16x8_t __arm_vshlq_m_r(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_s32)))
+int32x4_t __arm_vshlq_m_r_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_s32)))
+int32x4_t __arm_vshlq_m_r(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_s8)))
+int8x16_t __arm_vshlq_m_r_s8(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_s8)))
+int8x16_t __arm_vshlq_m_r(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_u16)))
+uint16x8_t __arm_vshlq_m_r_u16(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_u16)))
+uint16x8_t __arm_vshlq_m_r(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_u32)))
+uint32x4_t __arm_vshlq_m_r_u32(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_u32)))
+uint32x4_t __arm_vshlq_m_r(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_u8)))
+uint8x16_t __arm_vshlq_m_r_u8(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_u8)))
+uint8x16_t __arm_vshlq_m_r(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_s16)))
+int16x8_t __arm_vshlq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_s16)))
+int16x8_t __arm_vshlq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_s32)))
+int32x4_t __arm_vshlq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_s32)))
+int32x4_t __arm_vshlq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_s8)))
+int8x16_t __arm_vshlq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_s8)))
+int8x16_t __arm_vshlq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_u16)))
+uint16x8_t __arm_vshlq_m_u16(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_u16)))
+uint16x8_t __arm_vshlq_m(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_u32)))
+uint32x4_t __arm_vshlq_m_u32(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_u32)))
+uint32x4_t __arm_vshlq_m(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_u8)))
+uint8x16_t __arm_vshlq_m_u8(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_u8)))
+uint8x16_t __arm_vshlq_m(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_s16)))
+int16x8_t __arm_vshlq_n_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_s16)))
+int16x8_t __arm_vshlq_n(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_s32)))
+int32x4_t __arm_vshlq_n_s32(int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_s32)))
+int32x4_t __arm_vshlq_n(int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_s8)))
+int8x16_t __arm_vshlq_n_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_s8)))
+int8x16_t __arm_vshlq_n(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_u16)))
+uint16x8_t __arm_vshlq_n_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_u16)))
+uint16x8_t __arm_vshlq_n(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_u32)))
+uint32x4_t __arm_vshlq_n_u32(uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_u32)))
+uint32x4_t __arm_vshlq_n(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_u8)))
+uint8x16_t __arm_vshlq_n_u8(uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_u8)))
+uint8x16_t __arm_vshlq_n(uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_s16)))
+int16x8_t __arm_vshlq_r_s16(int16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_s16)))
+int16x8_t __arm_vshlq_r(int16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_s32)))
+int32x4_t __arm_vshlq_r_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_s32)))
+int32x4_t __arm_vshlq_r(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_s8)))
+int8x16_t __arm_vshlq_r_s8(int8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_s8)))
+int8x16_t __arm_vshlq_r(int8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_u16)))
+uint16x8_t __arm_vshlq_r_u16(uint16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_u16)))
+uint16x8_t __arm_vshlq_r(uint16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_u32)))
+uint32x4_t __arm_vshlq_r_u32(uint32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_u32)))
+uint32x4_t __arm_vshlq_r(uint32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_u8)))
+uint8x16_t __arm_vshlq_r_u8(uint8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_u8)))
+uint8x16_t __arm_vshlq_r(uint8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_s16)))
+int16x8_t __arm_vshlq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_s16)))
+int16x8_t __arm_vshlq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_s32)))
+int32x4_t __arm_vshlq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_s32)))
+int32x4_t __arm_vshlq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_s8)))
+int8x16_t __arm_vshlq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_s8)))
+int8x16_t __arm_vshlq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_u16)))
+uint16x8_t __arm_vshlq_u16(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_u16)))
+uint16x8_t __arm_vshlq(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_u32)))
+uint32x4_t __arm_vshlq_u32(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_u32)))
+uint32x4_t __arm_vshlq(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_u8)))
+uint8x16_t __arm_vshlq_u8(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_u8)))
+uint8x16_t __arm_vshlq(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_s16)))
+int16x8_t __arm_vshlq_x_n_s16(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_s16)))
+int16x8_t __arm_vshlq_x_n(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_s32)))
+int32x4_t __arm_vshlq_x_n_s32(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_s32)))
+int32x4_t __arm_vshlq_x_n(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_s8)))
+int8x16_t __arm_vshlq_x_n_s8(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_s8)))
+int8x16_t __arm_vshlq_x_n(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_u16)))
+uint16x8_t __arm_vshlq_x_n_u16(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_u16)))
+uint16x8_t __arm_vshlq_x_n(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_u32)))
+uint32x4_t __arm_vshlq_x_n_u32(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_u32)))
+uint32x4_t __arm_vshlq_x_n(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_u8)))
+uint8x16_t __arm_vshlq_x_n_u8(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_u8)))
+uint8x16_t __arm_vshlq_x_n(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_s16)))
+int16x8_t __arm_vshlq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_s16)))
+int16x8_t __arm_vshlq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_s32)))
+int32x4_t __arm_vshlq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_s32)))
+int32x4_t __arm_vshlq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_s8)))
+int8x16_t __arm_vshlq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_s8)))
+int8x16_t __arm_vshlq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_u16)))
+uint16x8_t __arm_vshlq_x_u16(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_u16)))
+uint16x8_t __arm_vshlq_x(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_u32)))
+uint32x4_t __arm_vshlq_x_u32(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_u32)))
+uint32x4_t __arm_vshlq_x(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_u8)))
+uint8x16_t __arm_vshlq_x_u8(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_u8)))
+uint8x16_t __arm_vshlq_x(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_s16)))
+int8x16_t __arm_vshrnbq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_s16)))
+int8x16_t __arm_vshrnbq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_s32)))
+int16x8_t __arm_vshrnbq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_s32)))
+int16x8_t __arm_vshrnbq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_u16)))
+uint8x16_t __arm_vshrnbq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_u16)))
+uint8x16_t __arm_vshrnbq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_u32)))
+uint16x8_t __arm_vshrnbq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_u32)))
+uint16x8_t __arm_vshrnbq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_s16)))
+int8x16_t __arm_vshrnbq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_s16)))
+int8x16_t __arm_vshrnbq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_s32)))
+int16x8_t __arm_vshrnbq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_s32)))
+int16x8_t __arm_vshrnbq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_u16)))
+uint8x16_t __arm_vshrnbq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_u16)))
+uint8x16_t __arm_vshrnbq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_u32)))
+uint16x8_t __arm_vshrnbq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_u32)))
+uint16x8_t __arm_vshrnbq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_s16)))
+int8x16_t __arm_vshrntq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_s16)))
+int8x16_t __arm_vshrntq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_s32)))
+int16x8_t __arm_vshrntq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_s32)))
+int16x8_t __arm_vshrntq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_u16)))
+uint8x16_t __arm_vshrntq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_u16)))
+uint8x16_t __arm_vshrntq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_u32)))
+uint16x8_t __arm_vshrntq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_u32)))
+uint16x8_t __arm_vshrntq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_s16)))
+int8x16_t __arm_vshrntq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_s16)))
+int8x16_t __arm_vshrntq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_s32)))
+int16x8_t __arm_vshrntq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_s32)))
+int16x8_t __arm_vshrntq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_u16)))
+uint8x16_t __arm_vshrntq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_u16)))
+uint8x16_t __arm_vshrntq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_u32)))
+uint16x8_t __arm_vshrntq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_u32)))
+uint16x8_t __arm_vshrntq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_s16)))
+int16x8_t __arm_vshrq_m_n_s16(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_s16)))
+int16x8_t __arm_vshrq_m(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_s32)))
+int32x4_t __arm_vshrq_m_n_s32(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_s32)))
+int32x4_t __arm_vshrq_m(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_s8)))
+int8x16_t __arm_vshrq_m_n_s8(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_s8)))
+int8x16_t __arm_vshrq_m(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_u16)))
+uint16x8_t __arm_vshrq_m_n_u16(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_u16)))
+uint16x8_t __arm_vshrq_m(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_u32)))
+uint32x4_t __arm_vshrq_m_n_u32(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_u32)))
+uint32x4_t __arm_vshrq_m(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_u8)))
+uint8x16_t __arm_vshrq_m_n_u8(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_u8)))
+uint8x16_t __arm_vshrq_m(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_s16)))
+int16x8_t __arm_vshrq_n_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_s16)))
+int16x8_t __arm_vshrq(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_s32)))
+int32x4_t __arm_vshrq_n_s32(int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_s32)))
+int32x4_t __arm_vshrq(int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_s8)))
+int8x16_t __arm_vshrq_n_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_s8)))
+int8x16_t __arm_vshrq(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_u16)))
+uint16x8_t __arm_vshrq_n_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_u16)))
+uint16x8_t __arm_vshrq(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_u32)))
+uint32x4_t __arm_vshrq_n_u32(uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_u32)))
+uint32x4_t __arm_vshrq(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_u8)))
+uint8x16_t __arm_vshrq_n_u8(uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_u8)))
+uint8x16_t __arm_vshrq(uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_s16)))
+int16x8_t __arm_vshrq_x_n_s16(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_s16)))
+int16x8_t __arm_vshrq_x(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_s32)))
+int32x4_t __arm_vshrq_x_n_s32(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_s32)))
+int32x4_t __arm_vshrq_x(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_s8)))
+int8x16_t __arm_vshrq_x_n_s8(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_s8)))
+int8x16_t __arm_vshrq_x(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_u16)))
+uint16x8_t __arm_vshrq_x_n_u16(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_u16)))
+uint16x8_t __arm_vshrq_x(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_u32)))
+uint32x4_t __arm_vshrq_x_n_u32(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_u32)))
+uint32x4_t __arm_vshrq_x(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_u8)))
+uint8x16_t __arm_vshrq_x_n_u8(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_u8)))
+uint8x16_t __arm_vshrq_x(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_s16)))
+int16x8_t __arm_vsliq_m_n_s16(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_s16)))
+int16x8_t __arm_vsliq_m(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_s32)))
+int32x4_t __arm_vsliq_m_n_s32(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_s32)))
+int32x4_t __arm_vsliq_m(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_s8)))
+int8x16_t __arm_vsliq_m_n_s8(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_s8)))
+int8x16_t __arm_vsliq_m(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_u16)))
+uint16x8_t __arm_vsliq_m_n_u16(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_u16)))
+uint16x8_t __arm_vsliq_m(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_u32)))
+uint32x4_t __arm_vsliq_m_n_u32(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_u32)))
+uint32x4_t __arm_vsliq_m(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_u8)))
+uint8x16_t __arm_vsliq_m_n_u8(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_u8)))
+uint8x16_t __arm_vsliq_m(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_s16)))
+int16x8_t __arm_vsliq_n_s16(int16x8_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_s16)))
+int16x8_t __arm_vsliq(int16x8_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_s32)))
+int32x4_t __arm_vsliq_n_s32(int32x4_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_s32)))
+int32x4_t __arm_vsliq(int32x4_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_s8)))
+int8x16_t __arm_vsliq_n_s8(int8x16_t, int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_s8)))
+int8x16_t __arm_vsliq(int8x16_t, int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_u16)))
+uint16x8_t __arm_vsliq_n_u16(uint16x8_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_u16)))
+uint16x8_t __arm_vsliq(uint16x8_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_u32)))
+uint32x4_t __arm_vsliq_n_u32(uint32x4_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_u32)))
+uint32x4_t __arm_vsliq(uint32x4_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_u8)))
+uint8x16_t __arm_vsliq_n_u8(uint8x16_t, uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_u8)))
+uint8x16_t __arm_vsliq(uint8x16_t, uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_s16)))
+int16x8_t __arm_vsriq_m_n_s16(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_s16)))
+int16x8_t __arm_vsriq_m(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_s32)))
+int32x4_t __arm_vsriq_m_n_s32(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_s32)))
+int32x4_t __arm_vsriq_m(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_s8)))
+int8x16_t __arm_vsriq_m_n_s8(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_s8)))
+int8x16_t __arm_vsriq_m(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_u16)))
+uint16x8_t __arm_vsriq_m_n_u16(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_u16)))
+uint16x8_t __arm_vsriq_m(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_u32)))
+uint32x4_t __arm_vsriq_m_n_u32(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_u32)))
+uint32x4_t __arm_vsriq_m(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_u8)))
+uint8x16_t __arm_vsriq_m_n_u8(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_u8)))
+uint8x16_t __arm_vsriq_m(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_s16)))
+int16x8_t __arm_vsriq_n_s16(int16x8_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_s16)))
+int16x8_t __arm_vsriq(int16x8_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_s32)))
+int32x4_t __arm_vsriq_n_s32(int32x4_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_s32)))
+int32x4_t __arm_vsriq(int32x4_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_s8)))
+int8x16_t __arm_vsriq_n_s8(int8x16_t, int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_s8)))
+int8x16_t __arm_vsriq(int8x16_t, int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_u16)))
+uint16x8_t __arm_vsriq_n_u16(uint16x8_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_u16)))
+uint16x8_t __arm_vsriq(uint16x8_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_u32)))
+uint32x4_t __arm_vsriq_n_u32(uint32x4_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_u32)))
+uint32x4_t __arm_vsriq(uint32x4_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_u8)))
+uint8x16_t __arm_vsriq_n_u8(uint8x16_t, uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_u8)))
+uint8x16_t __arm_vsriq(uint8x16_t, uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_s16)))
+void __arm_vst1q_p_s16(int16_t *, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_s16)))
+void __arm_vst1q_p(int16_t *, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_s32)))
+void __arm_vst1q_p_s32(int32_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_s32)))
+void __arm_vst1q_p(int32_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_s8)))
+void __arm_vst1q_p_s8(int8_t *, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_s8)))
+void __arm_vst1q_p(int8_t *, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_u16)))
+void __arm_vst1q_p_u16(uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_u16)))
+void __arm_vst1q_p(uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_u32)))
+void __arm_vst1q_p_u32(uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_u32)))
+void __arm_vst1q_p(uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_u8)))
+void __arm_vst1q_p_u8(uint8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_u8)))
+void __arm_vst1q_p(uint8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_s16)))
+void __arm_vst1q_s16(int16_t *, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_s16)))
+void __arm_vst1q(int16_t *, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_s32)))
+void __arm_vst1q_s32(int32_t *, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_s32)))
+void __arm_vst1q(int32_t *, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_s8)))
+void __arm_vst1q_s8(int8_t *, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_s8)))
+void __arm_vst1q(int8_t *, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_u16)))
+void __arm_vst1q_u16(uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_u16)))
+void __arm_vst1q(uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_u32)))
+void __arm_vst1q_u32(uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_u32)))
+void __arm_vst1q(uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_u8)))
+void __arm_vst1q_u8(uint8_t *, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_u8)))
+void __arm_vst1q(uint8_t *, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_s16)))
+void __arm_vst2q_s16(int16_t *, int16x8x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_s16)))
+void __arm_vst2q(int16_t *, int16x8x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_s32)))
+void __arm_vst2q_s32(int32_t *, int32x4x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_s32)))
+void __arm_vst2q(int32_t *, int32x4x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_s8)))
+void __arm_vst2q_s8(int8_t *, int8x16x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_s8)))
+void __arm_vst2q(int8_t *, int8x16x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_u16)))
+void __arm_vst2q_u16(uint16_t *, uint16x8x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_u16)))
+void __arm_vst2q(uint16_t *, uint16x8x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_u32)))
+void __arm_vst2q_u32(uint32_t *, uint32x4x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_u32)))
+void __arm_vst2q(uint32_t *, uint32x4x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_u8)))
+void __arm_vst2q_u8(uint8_t *, uint8x16x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_u8)))
+void __arm_vst2q(uint8_t *, uint8x16x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_s16)))
+void __arm_vst4q_s16(int16_t *, int16x8x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_s16)))
+void __arm_vst4q(int16_t *, int16x8x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_s32)))
+void __arm_vst4q_s32(int32_t *, int32x4x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_s32)))
+void __arm_vst4q(int32_t *, int32x4x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_s8)))
+void __arm_vst4q_s8(int8_t *, int8x16x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_s8)))
+void __arm_vst4q(int8_t *, int8x16x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_u16)))
+void __arm_vst4q_u16(uint16_t *, uint16x8x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_u16)))
+void __arm_vst4q(uint16_t *, uint16x8x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_u32)))
+void __arm_vst4q_u32(uint32_t *, uint32x4x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_u32)))
+void __arm_vst4q(uint32_t *, uint32x4x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_u8)))
+void __arm_vst4q_u8(uint8_t *, uint8x16x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_u8)))
+void __arm_vst4q(uint8_t *, uint8x16x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_s16)))
+void __arm_vstrbq_p_s16(int8_t *, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_s16)))
+void __arm_vstrbq_p(int8_t *, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_s32)))
+void __arm_vstrbq_p_s32(int8_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_s32)))
+void __arm_vstrbq_p(int8_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_s8)))
+void __arm_vstrbq_p_s8(int8_t *, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_s8)))
+void __arm_vstrbq_p(int8_t *, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_u16)))
+void __arm_vstrbq_p_u16(uint8_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_u16)))
+void __arm_vstrbq_p(uint8_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_u32)))
+void __arm_vstrbq_p_u32(uint8_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_u32)))
+void __arm_vstrbq_p(uint8_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_u8)))
+void __arm_vstrbq_p_u8(uint8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_u8)))
+void __arm_vstrbq_p(uint8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_s16)))
+void __arm_vstrbq_s16(int8_t *, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_s16)))
+void __arm_vstrbq(int8_t *, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_s32)))
+void __arm_vstrbq_s32(int8_t *, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_s32)))
+void __arm_vstrbq(int8_t *, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_s8)))
+void __arm_vstrbq_s8(int8_t *, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_s8)))
+void __arm_vstrbq(int8_t *, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_s16)))
+void __arm_vstrbq_scatter_offset_p_s16(int8_t *, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_s16)))
+void __arm_vstrbq_scatter_offset_p(int8_t *, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_s32)))
+void __arm_vstrbq_scatter_offset_p_s32(int8_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_s32)))
+void __arm_vstrbq_scatter_offset_p(int8_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_s8)))
+void __arm_vstrbq_scatter_offset_p_s8(int8_t *, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_s8)))
+void __arm_vstrbq_scatter_offset_p(int8_t *, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_u16)))
+void __arm_vstrbq_scatter_offset_p_u16(uint8_t *, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_u16)))
+void __arm_vstrbq_scatter_offset_p(uint8_t *, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_u32)))
+void __arm_vstrbq_scatter_offset_p_u32(uint8_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_u32)))
+void __arm_vstrbq_scatter_offset_p(uint8_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_u8)))
+void __arm_vstrbq_scatter_offset_p_u8(uint8_t *, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_u8)))
+void __arm_vstrbq_scatter_offset_p(uint8_t *, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_s16)))
+void __arm_vstrbq_scatter_offset_s16(int8_t *, uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_s16)))
+void __arm_vstrbq_scatter_offset(int8_t *, uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_s32)))
+void __arm_vstrbq_scatter_offset_s32(int8_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_s32)))
+void __arm_vstrbq_scatter_offset(int8_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_s8)))
+void __arm_vstrbq_scatter_offset_s8(int8_t *, uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_s8)))
+void __arm_vstrbq_scatter_offset(int8_t *, uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_u16)))
+void __arm_vstrbq_scatter_offset_u16(uint8_t *, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_u16)))
+void __arm_vstrbq_scatter_offset(uint8_t *, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_u32)))
+void __arm_vstrbq_scatter_offset_u32(uint8_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_u32)))
+void __arm_vstrbq_scatter_offset(uint8_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_u8)))
+void __arm_vstrbq_scatter_offset_u8(uint8_t *, uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_u8)))
+void __arm_vstrbq_scatter_offset(uint8_t *, uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_u16)))
+void __arm_vstrbq_u16(uint8_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_u16)))
+void __arm_vstrbq(uint8_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_u32)))
+void __arm_vstrbq_u32(uint8_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_u32)))
+void __arm_vstrbq(uint8_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_u8)))
+void __arm_vstrbq_u8(uint8_t *, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_u8)))
+void __arm_vstrbq(uint8_t *, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_p_s64)))
+void __arm_vstrdq_scatter_base_p_s64(uint64x2_t, int, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_p_s64)))
+void __arm_vstrdq_scatter_base_p(uint64x2_t, int, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_p_u64)))
+void __arm_vstrdq_scatter_base_p_u64(uint64x2_t, int, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_p_u64)))
+void __arm_vstrdq_scatter_base_p(uint64x2_t, int, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_s64)))
+void __arm_vstrdq_scatter_base_s64(uint64x2_t, int, int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_s64)))
+void __arm_vstrdq_scatter_base(uint64x2_t, int, int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_u64)))
+void __arm_vstrdq_scatter_base_u64(uint64x2_t, int, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_u64)))
+void __arm_vstrdq_scatter_base(uint64x2_t, int, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_p_s64)))
+void __arm_vstrdq_scatter_base_wb_p_s64(uint64x2_t *, int, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_p_s64)))
+void __arm_vstrdq_scatter_base_wb_p(uint64x2_t *, int, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_p_u64)))
+void __arm_vstrdq_scatter_base_wb_p_u64(uint64x2_t *, int, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_p_u64)))
+void __arm_vstrdq_scatter_base_wb_p(uint64x2_t *, int, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_s64)))
+void __arm_vstrdq_scatter_base_wb_s64(uint64x2_t *, int, int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_s64)))
+void __arm_vstrdq_scatter_base_wb(uint64x2_t *, int, int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_u64)))
+void __arm_vstrdq_scatter_base_wb_u64(uint64x2_t *, int, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_u64)))
+void __arm_vstrdq_scatter_base_wb(uint64x2_t *, int, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_p_s64)))
+void __arm_vstrdq_scatter_offset_p_s64(int64_t *, uint64x2_t, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_p_s64)))
+void __arm_vstrdq_scatter_offset_p(int64_t *, uint64x2_t, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_p_u64)))
+void __arm_vstrdq_scatter_offset_p_u64(uint64_t *, uint64x2_t, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_p_u64)))
+void __arm_vstrdq_scatter_offset_p(uint64_t *, uint64x2_t, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_s64)))
+void __arm_vstrdq_scatter_offset_s64(int64_t *, uint64x2_t, int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_s64)))
+void __arm_vstrdq_scatter_offset(int64_t *, uint64x2_t, int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_u64)))
+void __arm_vstrdq_scatter_offset_u64(uint64_t *, uint64x2_t, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_u64)))
+void __arm_vstrdq_scatter_offset(uint64_t *, uint64x2_t, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_p_s64)))
+void __arm_vstrdq_scatter_shifted_offset_p_s64(int64_t *, uint64x2_t, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_p_s64)))
+void __arm_vstrdq_scatter_shifted_offset_p(int64_t *, uint64x2_t, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_p_u64)))
+void __arm_vstrdq_scatter_shifted_offset_p_u64(uint64_t *, uint64x2_t, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_p_u64)))
+void __arm_vstrdq_scatter_shifted_offset_p(uint64_t *, uint64x2_t, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_s64)))
+void __arm_vstrdq_scatter_shifted_offset_s64(int64_t *, uint64x2_t, int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_s64)))
+void __arm_vstrdq_scatter_shifted_offset(int64_t *, uint64x2_t, int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_u64)))
+void __arm_vstrdq_scatter_shifted_offset_u64(uint64_t *, uint64x2_t, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_u64)))
+void __arm_vstrdq_scatter_shifted_offset(uint64_t *, uint64x2_t, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_s16)))
+void __arm_vstrhq_p_s16(int16_t *, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_s16)))
+void __arm_vstrhq_p(int16_t *, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_s32)))
+void __arm_vstrhq_p_s32(int16_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_s32)))
+void __arm_vstrhq_p(int16_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_u16)))
+void __arm_vstrhq_p_u16(uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_u16)))
+void __arm_vstrhq_p(uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_u32)))
+void __arm_vstrhq_p_u32(uint16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_u32)))
+void __arm_vstrhq_p(uint16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_s16)))
+void __arm_vstrhq_s16(int16_t *, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_s16)))
+void __arm_vstrhq(int16_t *, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_s32)))
+void __arm_vstrhq_s32(int16_t *, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_s32)))
+void __arm_vstrhq(int16_t *, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_s16)))
+void __arm_vstrhq_scatter_offset_p_s16(int16_t *, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_s16)))
+void __arm_vstrhq_scatter_offset_p(int16_t *, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_s32)))
+void __arm_vstrhq_scatter_offset_p_s32(int16_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_s32)))
+void __arm_vstrhq_scatter_offset_p(int16_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_u16)))
+void __arm_vstrhq_scatter_offset_p_u16(uint16_t *, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_u16)))
+void __arm_vstrhq_scatter_offset_p(uint16_t *, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_u32)))
+void __arm_vstrhq_scatter_offset_p_u32(uint16_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_u32)))
+void __arm_vstrhq_scatter_offset_p(uint16_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_s16)))
+void __arm_vstrhq_scatter_offset_s16(int16_t *, uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_s16)))
+void __arm_vstrhq_scatter_offset(int16_t *, uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_s32)))
+void __arm_vstrhq_scatter_offset_s32(int16_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_s32)))
+void __arm_vstrhq_scatter_offset(int16_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_u16)))
+void __arm_vstrhq_scatter_offset_u16(uint16_t *, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_u16)))
+void __arm_vstrhq_scatter_offset(uint16_t *, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_u32)))
+void __arm_vstrhq_scatter_offset_u32(uint16_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_u32)))
+void __arm_vstrhq_scatter_offset(uint16_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_s16)))
+void __arm_vstrhq_scatter_shifted_offset_p_s16(int16_t *, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_s16)))
+void __arm_vstrhq_scatter_shifted_offset_p(int16_t *, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_s32)))
+void __arm_vstrhq_scatter_shifted_offset_p_s32(int16_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_s32)))
+void __arm_vstrhq_scatter_shifted_offset_p(int16_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_u16)))
+void __arm_vstrhq_scatter_shifted_offset_p_u16(uint16_t *, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_u16)))
+void __arm_vstrhq_scatter_shifted_offset_p(uint16_t *, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_u32)))
+void __arm_vstrhq_scatter_shifted_offset_p_u32(uint16_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_u32)))
+void __arm_vstrhq_scatter_shifted_offset_p(uint16_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_s16)))
+void __arm_vstrhq_scatter_shifted_offset_s16(int16_t *, uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_s16)))
+void __arm_vstrhq_scatter_shifted_offset(int16_t *, uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_s32)))
+void __arm_vstrhq_scatter_shifted_offset_s32(int16_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_s32)))
+void __arm_vstrhq_scatter_shifted_offset(int16_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_u16)))
+void __arm_vstrhq_scatter_shifted_offset_u16(uint16_t *, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_u16)))
+void __arm_vstrhq_scatter_shifted_offset(uint16_t *, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_u32)))
+void __arm_vstrhq_scatter_shifted_offset_u32(uint16_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_u32)))
+void __arm_vstrhq_scatter_shifted_offset(uint16_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_u16)))
+void __arm_vstrhq_u16(uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_u16)))
+void __arm_vstrhq(uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_u32)))
+void __arm_vstrhq_u32(uint16_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_u32)))
+void __arm_vstrhq(uint16_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_p_s32)))
+void __arm_vstrwq_p_s32(int32_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_p_s32)))
+void __arm_vstrwq_p(int32_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_p_u32)))
+void __arm_vstrwq_p_u32(uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_p_u32)))
+void __arm_vstrwq_p(uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_s32)))
+void __arm_vstrwq_s32(int32_t *, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_s32)))
+void __arm_vstrwq(int32_t *, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_p_s32)))
+void __arm_vstrwq_scatter_base_p_s32(uint32x4_t, int, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_p_s32)))
+void __arm_vstrwq_scatter_base_p(uint32x4_t, int, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_p_u32)))
+void __arm_vstrwq_scatter_base_p_u32(uint32x4_t, int, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_p_u32)))
+void __arm_vstrwq_scatter_base_p(uint32x4_t, int, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_s32)))
+void __arm_vstrwq_scatter_base_s32(uint32x4_t, int, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_s32)))
+void __arm_vstrwq_scatter_base(uint32x4_t, int, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_u32)))
+void __arm_vstrwq_scatter_base_u32(uint32x4_t, int, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_u32)))
+void __arm_vstrwq_scatter_base(uint32x4_t, int, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_p_s32)))
+void __arm_vstrwq_scatter_base_wb_p_s32(uint32x4_t *, int, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_p_s32)))
+void __arm_vstrwq_scatter_base_wb_p(uint32x4_t *, int, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_p_u32)))
+void __arm_vstrwq_scatter_base_wb_p_u32(uint32x4_t *, int, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_p_u32)))
+void __arm_vstrwq_scatter_base_wb_p(uint32x4_t *, int, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_s32)))
+void __arm_vstrwq_scatter_base_wb_s32(uint32x4_t *, int, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_s32)))
+void __arm_vstrwq_scatter_base_wb(uint32x4_t *, int, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_u32)))
+void __arm_vstrwq_scatter_base_wb_u32(uint32x4_t *, int, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_u32)))
+void __arm_vstrwq_scatter_base_wb(uint32x4_t *, int, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_p_s32)))
+void __arm_vstrwq_scatter_offset_p_s32(int32_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_p_s32)))
+void __arm_vstrwq_scatter_offset_p(int32_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_p_u32)))
+void __arm_vstrwq_scatter_offset_p_u32(uint32_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_p_u32)))
+void __arm_vstrwq_scatter_offset_p(uint32_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_s32)))
+void __arm_vstrwq_scatter_offset_s32(int32_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_s32)))
+void __arm_vstrwq_scatter_offset(int32_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_u32)))
+void __arm_vstrwq_scatter_offset_u32(uint32_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_u32)))
+void __arm_vstrwq_scatter_offset(uint32_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_p_s32)))
+void __arm_vstrwq_scatter_shifted_offset_p_s32(int32_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_p_s32)))
+void __arm_vstrwq_scatter_shifted_offset_p(int32_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_p_u32)))
+void __arm_vstrwq_scatter_shifted_offset_p_u32(uint32_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_p_u32)))
+void __arm_vstrwq_scatter_shifted_offset_p(uint32_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_s32)))
+void __arm_vstrwq_scatter_shifted_offset_s32(int32_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_s32)))
+void __arm_vstrwq_scatter_shifted_offset(int32_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_u32)))
+void __arm_vstrwq_scatter_shifted_offset_u32(uint32_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_u32)))
+void __arm_vstrwq_scatter_shifted_offset(uint32_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_u32)))
+void __arm_vstrwq_u32(uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_u32)))
+void __arm_vstrwq(uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_s16)))
+int16x8_t __arm_vsubq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_s16)))
+int16x8_t __arm_vsubq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_s32)))
+int32x4_t __arm_vsubq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_s32)))
+int32x4_t __arm_vsubq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_s8)))
+int8x16_t __arm_vsubq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_s8)))
+int8x16_t __arm_vsubq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_u16)))
+uint16x8_t __arm_vsubq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_u16)))
+uint16x8_t __arm_vsubq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_u32)))
+uint32x4_t __arm_vsubq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_u32)))
+uint32x4_t __arm_vsubq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_u8)))
+uint8x16_t __arm_vsubq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_u8)))
+uint8x16_t __arm_vsubq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_s16)))
+int16x8_t __arm_vsubq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_s16)))
+int16x8_t __arm_vsubq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_s32)))
+int32x4_t __arm_vsubq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_s32)))
+int32x4_t __arm_vsubq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_s8)))
+int8x16_t __arm_vsubq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_s8)))
+int8x16_t __arm_vsubq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_u16)))
+uint16x8_t __arm_vsubq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_u16)))
+uint16x8_t __arm_vsubq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_u32)))
+uint32x4_t __arm_vsubq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_u32)))
+uint32x4_t __arm_vsubq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_u8)))
+uint8x16_t __arm_vsubq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_u8)))
+uint8x16_t __arm_vsubq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_s16)))
+int16x8_t __arm_vsubq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_s16)))
+int16x8_t __arm_vsubq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_s32)))
+int32x4_t __arm_vsubq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_s32)))
+int32x4_t __arm_vsubq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_s8)))
+int8x16_t __arm_vsubq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_s8)))
+int8x16_t __arm_vsubq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_u16)))
+uint16x8_t __arm_vsubq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_u16)))
+uint16x8_t __arm_vsubq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_u32)))
+uint32x4_t __arm_vsubq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_u32)))
+uint32x4_t __arm_vsubq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_u8)))
+uint8x16_t __arm_vsubq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_u8)))
+uint8x16_t __arm_vsubq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_s16)))
+int16x8_t __arm_vsubq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_s16)))
+int16x8_t __arm_vsubq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_s32)))
+int32x4_t __arm_vsubq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_s32)))
+int32x4_t __arm_vsubq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_s8)))
+int8x16_t __arm_vsubq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_s8)))
+int8x16_t __arm_vsubq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_u16)))
+uint16x8_t __arm_vsubq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_u16)))
+uint16x8_t __arm_vsubq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_u32)))
+uint32x4_t __arm_vsubq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_u32)))
+uint32x4_t __arm_vsubq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_u8)))
+uint8x16_t __arm_vsubq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_u8)))
+uint8x16_t __arm_vsubq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_s16)))
+int16x8_t __arm_vsubq_x_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_s16)))
+int16x8_t __arm_vsubq_x(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_s32)))
+int32x4_t __arm_vsubq_x_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_s32)))
+int32x4_t __arm_vsubq_x(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_s8)))
+int8x16_t __arm_vsubq_x_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_s8)))
+int8x16_t __arm_vsubq_x(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_u16)))
+uint16x8_t __arm_vsubq_x_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_u16)))
+uint16x8_t __arm_vsubq_x(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_u32)))
+uint32x4_t __arm_vsubq_x_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_u32)))
+uint32x4_t __arm_vsubq_x(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_u8)))
+uint8x16_t __arm_vsubq_x_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_u8)))
+uint8x16_t __arm_vsubq_x(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_s16)))
+int16x8_t __arm_vsubq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_s16)))
+int16x8_t __arm_vsubq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_s32)))
+int32x4_t __arm_vsubq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_s32)))
+int32x4_t __arm_vsubq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_s8)))
+int8x16_t __arm_vsubq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_s8)))
+int8x16_t __arm_vsubq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_u16)))
+uint16x8_t __arm_vsubq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_u16)))
+uint16x8_t __arm_vsubq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_u32)))
+uint32x4_t __arm_vsubq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_u32)))
+uint32x4_t __arm_vsubq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_u8)))
+uint8x16_t __arm_vsubq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_u8)))
+uint8x16_t __arm_vsubq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_s16)))
+int16x8_t __arm_vuninitializedq(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_s32)))
+int32x4_t __arm_vuninitializedq(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_s64)))
+int64x2_t __arm_vuninitializedq(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_s8)))
+int8x16_t __arm_vuninitializedq(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_u16)))
+uint16x8_t __arm_vuninitializedq(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_u32)))
+uint32x4_t __arm_vuninitializedq(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_u64)))
+uint64x2_t __arm_vuninitializedq(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_u8)))
+uint8x16_t __arm_vuninitializedq(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_s16)))
+int16x8_t __arm_vuninitializedq_s16();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_s32)))
+int32x4_t __arm_vuninitializedq_s32();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_s64)))
+int64x2_t __arm_vuninitializedq_s64();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_s8)))
+int8x16_t __arm_vuninitializedq_s8();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_u16)))
+uint16x8_t __arm_vuninitializedq_u16();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_u32)))
+uint32x4_t __arm_vuninitializedq_u32();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_u64)))
+uint64x2_t __arm_vuninitializedq_u64();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_u8)))
+uint8x16_t __arm_vuninitializedq_u8();
+
+#if (__ARM_FEATURE_MVE & 2)
+
+typedef __fp16 float16_t;
+typedef float float32_t;
+typedef __attribute__((__neon_vector_type__(8), __clang_arm_mve_strict_polymorphism)) float16_t float16x8_t;
+typedef struct { float16x8_t val[2]; } float16x8x2_t;
+typedef struct { float16x8_t val[4]; } float16x8x4_t;
+typedef __attribute__((__neon_vector_type__(4), __clang_arm_mve_strict_polymorphism)) float32_t float32x4_t;
+typedef struct { float32x4_t val[2]; } float32x4x2_t;
+typedef struct { float32x4_t val[4]; } float32x4x4_t;
+
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_f16)))
+float16x8_t __arm_vabdq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_f16)))
+float16x8_t __arm_vabdq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_f32)))
+float32x4_t __arm_vabdq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_f32)))
+float32x4_t __arm_vabdq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_f16)))
+float16x8_t __arm_vabdq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_f16)))
+float16x8_t __arm_vabdq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_f32)))
+float32x4_t __arm_vabdq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_f32)))
+float32x4_t __arm_vabdq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_f16)))
+float16x8_t __arm_vabdq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_f16)))
+float16x8_t __arm_vabdq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_f32)))
+float32x4_t __arm_vabdq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_f32)))
+float32x4_t __arm_vabdq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_f16)))
+float16x8_t __arm_vabsq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_f16)))
+float16x8_t __arm_vabsq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_f32)))
+float32x4_t __arm_vabsq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_f32)))
+float32x4_t __arm_vabsq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_f16)))
+float16x8_t __arm_vabsq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_f16)))
+float16x8_t __arm_vabsq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_f32)))
+float32x4_t __arm_vabsq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_f32)))
+float32x4_t __arm_vabsq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_f16)))
+float16x8_t __arm_vabsq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_f16)))
+float16x8_t __arm_vabsq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_f32)))
+float32x4_t __arm_vabsq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_f32)))
+float32x4_t __arm_vabsq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_f16)))
+float16x8_t __arm_vaddq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_f16)))
+float16x8_t __arm_vaddq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_f32)))
+float32x4_t __arm_vaddq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_f32)))
+float32x4_t __arm_vaddq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_f16)))
+float16x8_t __arm_vaddq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_f16)))
+float16x8_t __arm_vaddq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_f32)))
+float32x4_t __arm_vaddq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_f32)))
+float32x4_t __arm_vaddq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_f16)))
+float16x8_t __arm_vaddq_m_n_f16(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_f16)))
+float16x8_t __arm_vaddq_m(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_f32)))
+float32x4_t __arm_vaddq_m_n_f32(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_f32)))
+float32x4_t __arm_vaddq_m(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_f16)))
+float16x8_t __arm_vaddq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_f16)))
+float16x8_t __arm_vaddq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_f32)))
+float32x4_t __arm_vaddq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_f32)))
+float32x4_t __arm_vaddq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_f16)))
+float16x8_t __arm_vaddq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_f16)))
+float16x8_t __arm_vaddq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_f32)))
+float32x4_t __arm_vaddq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_f32)))
+float32x4_t __arm_vaddq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_f16)))
+float16x8_t __arm_vaddq_x_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_f16)))
+float16x8_t __arm_vaddq_x(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_f32)))
+float32x4_t __arm_vaddq_x_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_f32)))
+float32x4_t __arm_vaddq_x(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_f16)))
+float16x8_t __arm_vandq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_f16)))
+float16x8_t __arm_vandq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_f32)))
+float32x4_t __arm_vandq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_f32)))
+float32x4_t __arm_vandq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_f16)))
+float16x8_t __arm_vandq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_f16)))
+float16x8_t __arm_vandq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_f32)))
+float32x4_t __arm_vandq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_f32)))
+float32x4_t __arm_vandq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_f16)))
+float16x8_t __arm_vandq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_f16)))
+float16x8_t __arm_vandq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_f32)))
+float32x4_t __arm_vandq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_f32)))
+float32x4_t __arm_vandq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_f16)))
+float16x8_t __arm_vbicq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_f16)))
+float16x8_t __arm_vbicq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_f32)))
+float32x4_t __arm_vbicq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_f32)))
+float32x4_t __arm_vbicq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_f16)))
+float16x8_t __arm_vbicq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_f16)))
+float16x8_t __arm_vbicq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_f32)))
+float32x4_t __arm_vbicq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_f32)))
+float32x4_t __arm_vbicq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_f16)))
+float16x8_t __arm_vbicq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_f16)))
+float16x8_t __arm_vbicq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_f32)))
+float32x4_t __arm_vbicq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_f32)))
+float32x4_t __arm_vbicq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_f16)))
+float16x8_t __arm_vbrsrq_m_n_f16(float16x8_t, float16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_f16)))
+float16x8_t __arm_vbrsrq_m(float16x8_t, float16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_f32)))
+float32x4_t __arm_vbrsrq_m_n_f32(float32x4_t, float32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_f32)))
+float32x4_t __arm_vbrsrq_m(float32x4_t, float32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_f16)))
+float16x8_t __arm_vbrsrq_n_f16(float16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_f16)))
+float16x8_t __arm_vbrsrq(float16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_f32)))
+float32x4_t __arm_vbrsrq_n_f32(float32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_f32)))
+float32x4_t __arm_vbrsrq(float32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_f16)))
+float16x8_t __arm_vbrsrq_x_n_f16(float16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_f16)))
+float16x8_t __arm_vbrsrq_x(float16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_f32)))
+float32x4_t __arm_vbrsrq_x_n_f32(float32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_f32)))
+float32x4_t __arm_vbrsrq_x(float32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_f16)))
+float16x8_t __arm_vcaddq_rot270_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_f16)))
+float16x8_t __arm_vcaddq_rot270(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_f32)))
+float32x4_t __arm_vcaddq_rot270_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_f32)))
+float32x4_t __arm_vcaddq_rot270(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_f16)))
+float16x8_t __arm_vcaddq_rot270_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_f16)))
+float16x8_t __arm_vcaddq_rot270_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_f32)))
+float32x4_t __arm_vcaddq_rot270_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_f32)))
+float32x4_t __arm_vcaddq_rot270_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_f16)))
+float16x8_t __arm_vcaddq_rot270_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_f16)))
+float16x8_t __arm_vcaddq_rot270_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_f32)))
+float32x4_t __arm_vcaddq_rot270_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_f32)))
+float32x4_t __arm_vcaddq_rot270_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_f16)))
+float16x8_t __arm_vcaddq_rot90_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_f16)))
+float16x8_t __arm_vcaddq_rot90(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_f32)))
+float32x4_t __arm_vcaddq_rot90_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_f32)))
+float32x4_t __arm_vcaddq_rot90(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_f16)))
+float16x8_t __arm_vcaddq_rot90_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_f16)))
+float16x8_t __arm_vcaddq_rot90_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_f32)))
+float32x4_t __arm_vcaddq_rot90_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_f32)))
+float32x4_t __arm_vcaddq_rot90_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_f16)))
+float16x8_t __arm_vcaddq_rot90_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_f16)))
+float16x8_t __arm_vcaddq_rot90_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_f32)))
+float32x4_t __arm_vcaddq_rot90_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_f32)))
+float32x4_t __arm_vcaddq_rot90_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_f16)))
+float16x8_t __arm_vcmlaq_f16(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_f16)))
+float16x8_t __arm_vcmlaq(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_f32)))
+float32x4_t __arm_vcmlaq_f32(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_f32)))
+float32x4_t __arm_vcmlaq(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_m_f16)))
+float16x8_t __arm_vcmlaq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_m_f16)))
+float16x8_t __arm_vcmlaq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_m_f32)))
+float32x4_t __arm_vcmlaq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_m_f32)))
+float32x4_t __arm_vcmlaq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_f16)))
+float16x8_t __arm_vcmlaq_rot180_f16(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_f16)))
+float16x8_t __arm_vcmlaq_rot180(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_f32)))
+float32x4_t __arm_vcmlaq_rot180_f32(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_f32)))
+float32x4_t __arm_vcmlaq_rot180(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_m_f16)))
+float16x8_t __arm_vcmlaq_rot180_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_m_f16)))
+float16x8_t __arm_vcmlaq_rot180_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_m_f32)))
+float32x4_t __arm_vcmlaq_rot180_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_m_f32)))
+float32x4_t __arm_vcmlaq_rot180_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_f16)))
+float16x8_t __arm_vcmlaq_rot270_f16(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_f16)))
+float16x8_t __arm_vcmlaq_rot270(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_f32)))
+float32x4_t __arm_vcmlaq_rot270_f32(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_f32)))
+float32x4_t __arm_vcmlaq_rot270(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_m_f16)))
+float16x8_t __arm_vcmlaq_rot270_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_m_f16)))
+float16x8_t __arm_vcmlaq_rot270_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_m_f32)))
+float32x4_t __arm_vcmlaq_rot270_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_m_f32)))
+float32x4_t __arm_vcmlaq_rot270_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_f16)))
+float16x8_t __arm_vcmlaq_rot90_f16(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_f16)))
+float16x8_t __arm_vcmlaq_rot90(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_f32)))
+float32x4_t __arm_vcmlaq_rot90_f32(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_f32)))
+float32x4_t __arm_vcmlaq_rot90(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_m_f16)))
+float16x8_t __arm_vcmlaq_rot90_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_m_f16)))
+float16x8_t __arm_vcmlaq_rot90_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_m_f32)))
+float32x4_t __arm_vcmlaq_rot90_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_m_f32)))
+float32x4_t __arm_vcmlaq_rot90_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_f16)))
+mve_pred16_t __arm_vcmpeqq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_f16)))
+mve_pred16_t __arm_vcmpeqq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_f32)))
+mve_pred16_t __arm_vcmpeqq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_f32)))
+mve_pred16_t __arm_vcmpeqq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_f16)))
+mve_pred16_t __arm_vcmpeqq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_f16)))
+mve_pred16_t __arm_vcmpeqq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_f32)))
+mve_pred16_t __arm_vcmpeqq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_f32)))
+mve_pred16_t __arm_vcmpeqq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_f16)))
+mve_pred16_t __arm_vcmpeqq_m_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_f16)))
+mve_pred16_t __arm_vcmpeqq_m(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_f32)))
+mve_pred16_t __arm_vcmpeqq_m_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_f32)))
+mve_pred16_t __arm_vcmpeqq_m(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_f16)))
+mve_pred16_t __arm_vcmpeqq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_f16)))
+mve_pred16_t __arm_vcmpeqq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_f32)))
+mve_pred16_t __arm_vcmpeqq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_f32)))
+mve_pred16_t __arm_vcmpeqq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_f16)))
+mve_pred16_t __arm_vcmpgeq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_f16)))
+mve_pred16_t __arm_vcmpgeq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_f32)))
+mve_pred16_t __arm_vcmpgeq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_f32)))
+mve_pred16_t __arm_vcmpgeq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_f16)))
+mve_pred16_t __arm_vcmpgeq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_f16)))
+mve_pred16_t __arm_vcmpgeq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_f32)))
+mve_pred16_t __arm_vcmpgeq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_f32)))
+mve_pred16_t __arm_vcmpgeq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_f16)))
+mve_pred16_t __arm_vcmpgeq_m_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_f16)))
+mve_pred16_t __arm_vcmpgeq_m(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_f32)))
+mve_pred16_t __arm_vcmpgeq_m_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_f32)))
+mve_pred16_t __arm_vcmpgeq_m(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_f16)))
+mve_pred16_t __arm_vcmpgeq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_f16)))
+mve_pred16_t __arm_vcmpgeq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_f32)))
+mve_pred16_t __arm_vcmpgeq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_f32)))
+mve_pred16_t __arm_vcmpgeq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_f16)))
+mve_pred16_t __arm_vcmpgtq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_f16)))
+mve_pred16_t __arm_vcmpgtq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_f32)))
+mve_pred16_t __arm_vcmpgtq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_f32)))
+mve_pred16_t __arm_vcmpgtq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_f16)))
+mve_pred16_t __arm_vcmpgtq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_f16)))
+mve_pred16_t __arm_vcmpgtq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_f32)))
+mve_pred16_t __arm_vcmpgtq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_f32)))
+mve_pred16_t __arm_vcmpgtq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_f16)))
+mve_pred16_t __arm_vcmpgtq_m_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_f16)))
+mve_pred16_t __arm_vcmpgtq_m(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_f32)))
+mve_pred16_t __arm_vcmpgtq_m_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_f32)))
+mve_pred16_t __arm_vcmpgtq_m(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_f16)))
+mve_pred16_t __arm_vcmpgtq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_f16)))
+mve_pred16_t __arm_vcmpgtq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_f32)))
+mve_pred16_t __arm_vcmpgtq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_f32)))
+mve_pred16_t __arm_vcmpgtq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_f16)))
+mve_pred16_t __arm_vcmpleq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_f16)))
+mve_pred16_t __arm_vcmpleq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_f32)))
+mve_pred16_t __arm_vcmpleq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_f32)))
+mve_pred16_t __arm_vcmpleq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_f16)))
+mve_pred16_t __arm_vcmpleq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_f16)))
+mve_pred16_t __arm_vcmpleq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_f32)))
+mve_pred16_t __arm_vcmpleq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_f32)))
+mve_pred16_t __arm_vcmpleq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_f16)))
+mve_pred16_t __arm_vcmpleq_m_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_f16)))
+mve_pred16_t __arm_vcmpleq_m(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_f32)))
+mve_pred16_t __arm_vcmpleq_m_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_f32)))
+mve_pred16_t __arm_vcmpleq_m(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_f16)))
+mve_pred16_t __arm_vcmpleq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_f16)))
+mve_pred16_t __arm_vcmpleq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_f32)))
+mve_pred16_t __arm_vcmpleq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_f32)))
+mve_pred16_t __arm_vcmpleq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_f16)))
+mve_pred16_t __arm_vcmpltq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_f16)))
+mve_pred16_t __arm_vcmpltq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_f32)))
+mve_pred16_t __arm_vcmpltq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_f32)))
+mve_pred16_t __arm_vcmpltq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_f16)))
+mve_pred16_t __arm_vcmpltq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_f16)))
+mve_pred16_t __arm_vcmpltq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_f32)))
+mve_pred16_t __arm_vcmpltq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_f32)))
+mve_pred16_t __arm_vcmpltq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_f16)))
+mve_pred16_t __arm_vcmpltq_m_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_f16)))
+mve_pred16_t __arm_vcmpltq_m(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_f32)))
+mve_pred16_t __arm_vcmpltq_m_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_f32)))
+mve_pred16_t __arm_vcmpltq_m(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_f16)))
+mve_pred16_t __arm_vcmpltq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_f16)))
+mve_pred16_t __arm_vcmpltq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_f32)))
+mve_pred16_t __arm_vcmpltq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_f32)))
+mve_pred16_t __arm_vcmpltq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_f16)))
+mve_pred16_t __arm_vcmpneq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_f16)))
+mve_pred16_t __arm_vcmpneq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_f32)))
+mve_pred16_t __arm_vcmpneq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_f32)))
+mve_pred16_t __arm_vcmpneq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_f16)))
+mve_pred16_t __arm_vcmpneq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_f16)))
+mve_pred16_t __arm_vcmpneq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_f32)))
+mve_pred16_t __arm_vcmpneq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_f32)))
+mve_pred16_t __arm_vcmpneq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_f16)))
+mve_pred16_t __arm_vcmpneq_m_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_f16)))
+mve_pred16_t __arm_vcmpneq_m(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_f32)))
+mve_pred16_t __arm_vcmpneq_m_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_f32)))
+mve_pred16_t __arm_vcmpneq_m(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_f16)))
+mve_pred16_t __arm_vcmpneq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_f16)))
+mve_pred16_t __arm_vcmpneq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_f32)))
+mve_pred16_t __arm_vcmpneq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_f32)))
+mve_pred16_t __arm_vcmpneq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_f16)))
+float16x8_t __arm_vcmulq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_f16)))
+float16x8_t __arm_vcmulq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_f32)))
+float32x4_t __arm_vcmulq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_f32)))
+float32x4_t __arm_vcmulq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_m_f16)))
+float16x8_t __arm_vcmulq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_m_f16)))
+float16x8_t __arm_vcmulq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_m_f32)))
+float32x4_t __arm_vcmulq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_m_f32)))
+float32x4_t __arm_vcmulq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_f16)))
+float16x8_t __arm_vcmulq_rot180_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_f16)))
+float16x8_t __arm_vcmulq_rot180(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_f32)))
+float32x4_t __arm_vcmulq_rot180_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_f32)))
+float32x4_t __arm_vcmulq_rot180(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_m_f16)))
+float16x8_t __arm_vcmulq_rot180_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_m_f16)))
+float16x8_t __arm_vcmulq_rot180_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_m_f32)))
+float32x4_t __arm_vcmulq_rot180_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_m_f32)))
+float32x4_t __arm_vcmulq_rot180_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_x_f16)))
+float16x8_t __arm_vcmulq_rot180_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_x_f16)))
+float16x8_t __arm_vcmulq_rot180_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_x_f32)))
+float32x4_t __arm_vcmulq_rot180_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_x_f32)))
+float32x4_t __arm_vcmulq_rot180_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_f16)))
+float16x8_t __arm_vcmulq_rot270_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_f16)))
+float16x8_t __arm_vcmulq_rot270(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_f32)))
+float32x4_t __arm_vcmulq_rot270_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_f32)))
+float32x4_t __arm_vcmulq_rot270(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_m_f16)))
+float16x8_t __arm_vcmulq_rot270_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_m_f16)))
+float16x8_t __arm_vcmulq_rot270_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_m_f32)))
+float32x4_t __arm_vcmulq_rot270_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_m_f32)))
+float32x4_t __arm_vcmulq_rot270_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_x_f16)))
+float16x8_t __arm_vcmulq_rot270_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_x_f16)))
+float16x8_t __arm_vcmulq_rot270_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_x_f32)))
+float32x4_t __arm_vcmulq_rot270_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_x_f32)))
+float32x4_t __arm_vcmulq_rot270_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_f16)))
+float16x8_t __arm_vcmulq_rot90_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_f16)))
+float16x8_t __arm_vcmulq_rot90(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_f32)))
+float32x4_t __arm_vcmulq_rot90_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_f32)))
+float32x4_t __arm_vcmulq_rot90(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_m_f16)))
+float16x8_t __arm_vcmulq_rot90_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_m_f16)))
+float16x8_t __arm_vcmulq_rot90_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_m_f32)))
+float32x4_t __arm_vcmulq_rot90_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_m_f32)))
+float32x4_t __arm_vcmulq_rot90_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_x_f16)))
+float16x8_t __arm_vcmulq_rot90_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_x_f16)))
+float16x8_t __arm_vcmulq_rot90_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_x_f32)))
+float32x4_t __arm_vcmulq_rot90_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_x_f32)))
+float32x4_t __arm_vcmulq_rot90_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_x_f16)))
+float16x8_t __arm_vcmulq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_x_f16)))
+float16x8_t __arm_vcmulq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_x_f32)))
+float32x4_t __arm_vcmulq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_x_f32)))
+float32x4_t __arm_vcmulq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_f16)))
+float16x8_t __arm_vcreateq_f16(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_f32)))
+float32x4_t __arm_vcreateq_f32(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_s16_f16)))
+int16x8_t __arm_vcvtaq_m_s16_f16(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_s16_f16)))
+int16x8_t __arm_vcvtaq_m(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_s32_f32)))
+int32x4_t __arm_vcvtaq_m_s32_f32(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_s32_f32)))
+int32x4_t __arm_vcvtaq_m(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_u16_f16)))
+uint16x8_t __arm_vcvtaq_m_u16_f16(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_u16_f16)))
+uint16x8_t __arm_vcvtaq_m(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_u32_f32)))
+uint32x4_t __arm_vcvtaq_m_u32_f32(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_u32_f32)))
+uint32x4_t __arm_vcvtaq_m(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_s16_f16)))
+int16x8_t __arm_vcvtaq_s16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_s32_f32)))
+int32x4_t __arm_vcvtaq_s32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_u16_f16)))
+uint16x8_t __arm_vcvtaq_u16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_u32_f32)))
+uint32x4_t __arm_vcvtaq_u32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_x_s16_f16)))
+int16x8_t __arm_vcvtaq_x_s16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_x_s32_f32)))
+int32x4_t __arm_vcvtaq_x_s32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_x_u16_f16)))
+uint16x8_t __arm_vcvtaq_x_u16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_x_u32_f32)))
+uint32x4_t __arm_vcvtaq_x_u32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtbq_f16_f32)))
+float16x8_t __arm_vcvtbq_f16_f32(float16x8_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtbq_f32_f16)))
+float32x4_t __arm_vcvtbq_f32_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtbq_m_f16_f32)))
+float16x8_t __arm_vcvtbq_m_f16_f32(float16x8_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtbq_m_f32_f16)))
+float32x4_t __arm_vcvtbq_m_f32_f16(float32x4_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtbq_x_f32_f16)))
+float32x4_t __arm_vcvtbq_x_f32_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_s16_f16)))
+int16x8_t __arm_vcvtmq_m_s16_f16(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_s16_f16)))
+int16x8_t __arm_vcvtmq_m(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_s32_f32)))
+int32x4_t __arm_vcvtmq_m_s32_f32(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_s32_f32)))
+int32x4_t __arm_vcvtmq_m(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_u16_f16)))
+uint16x8_t __arm_vcvtmq_m_u16_f16(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_u16_f16)))
+uint16x8_t __arm_vcvtmq_m(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_u32_f32)))
+uint32x4_t __arm_vcvtmq_m_u32_f32(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_u32_f32)))
+uint32x4_t __arm_vcvtmq_m(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_s16_f16)))
+int16x8_t __arm_vcvtmq_s16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_s32_f32)))
+int32x4_t __arm_vcvtmq_s32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_u16_f16)))
+uint16x8_t __arm_vcvtmq_u16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_u32_f32)))
+uint32x4_t __arm_vcvtmq_u32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_x_s16_f16)))
+int16x8_t __arm_vcvtmq_x_s16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_x_s32_f32)))
+int32x4_t __arm_vcvtmq_x_s32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_x_u16_f16)))
+uint16x8_t __arm_vcvtmq_x_u16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_x_u32_f32)))
+uint32x4_t __arm_vcvtmq_x_u32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_s16_f16)))
+int16x8_t __arm_vcvtnq_m_s16_f16(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_s16_f16)))
+int16x8_t __arm_vcvtnq_m(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_s32_f32)))
+int32x4_t __arm_vcvtnq_m_s32_f32(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_s32_f32)))
+int32x4_t __arm_vcvtnq_m(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_u16_f16)))
+uint16x8_t __arm_vcvtnq_m_u16_f16(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_u16_f16)))
+uint16x8_t __arm_vcvtnq_m(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_u32_f32)))
+uint32x4_t __arm_vcvtnq_m_u32_f32(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_u32_f32)))
+uint32x4_t __arm_vcvtnq_m(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_s16_f16)))
+int16x8_t __arm_vcvtnq_s16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_s32_f32)))
+int32x4_t __arm_vcvtnq_s32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_u16_f16)))
+uint16x8_t __arm_vcvtnq_u16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_u32_f32)))
+uint32x4_t __arm_vcvtnq_u32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_x_s16_f16)))
+int16x8_t __arm_vcvtnq_x_s16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_x_s32_f32)))
+int32x4_t __arm_vcvtnq_x_s32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_x_u16_f16)))
+uint16x8_t __arm_vcvtnq_x_u16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_x_u32_f32)))
+uint32x4_t __arm_vcvtnq_x_u32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_s16_f16)))
+int16x8_t __arm_vcvtpq_m_s16_f16(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_s16_f16)))
+int16x8_t __arm_vcvtpq_m(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_s32_f32)))
+int32x4_t __arm_vcvtpq_m_s32_f32(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_s32_f32)))
+int32x4_t __arm_vcvtpq_m(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_u16_f16)))
+uint16x8_t __arm_vcvtpq_m_u16_f16(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_u16_f16)))
+uint16x8_t __arm_vcvtpq_m(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_u32_f32)))
+uint32x4_t __arm_vcvtpq_m_u32_f32(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_u32_f32)))
+uint32x4_t __arm_vcvtpq_m(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_s16_f16)))
+int16x8_t __arm_vcvtpq_s16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_s32_f32)))
+int32x4_t __arm_vcvtpq_s32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_u16_f16)))
+uint16x8_t __arm_vcvtpq_u16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_u32_f32)))
+uint32x4_t __arm_vcvtpq_u32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_x_s16_f16)))
+int16x8_t __arm_vcvtpq_x_s16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_x_s32_f32)))
+int32x4_t __arm_vcvtpq_x_s32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_x_u16_f16)))
+uint16x8_t __arm_vcvtpq_x_u16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_x_u32_f32)))
+uint32x4_t __arm_vcvtpq_x_u32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f16_s16)))
+float16x8_t __arm_vcvtq_f16_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f16_s16)))
+float16x8_t __arm_vcvtq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f16_u16)))
+float16x8_t __arm_vcvtq_f16_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f16_u16)))
+float16x8_t __arm_vcvtq(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f32_s32)))
+float32x4_t __arm_vcvtq_f32_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f32_s32)))
+float32x4_t __arm_vcvtq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f32_u32)))
+float32x4_t __arm_vcvtq_f32_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f32_u32)))
+float32x4_t __arm_vcvtq(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f16_s16)))
+float16x8_t __arm_vcvtq_m_f16_s16(float16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f16_s16)))
+float16x8_t __arm_vcvtq_m(float16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f16_u16)))
+float16x8_t __arm_vcvtq_m_f16_u16(float16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f16_u16)))
+float16x8_t __arm_vcvtq_m(float16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f32_s32)))
+float32x4_t __arm_vcvtq_m_f32_s32(float32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f32_s32)))
+float32x4_t __arm_vcvtq_m(float32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f32_u32)))
+float32x4_t __arm_vcvtq_m_f32_u32(float32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f32_u32)))
+float32x4_t __arm_vcvtq_m(float32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f16_s16)))
+float16x8_t __arm_vcvtq_m_n_f16_s16(float16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f16_s16)))
+float16x8_t __arm_vcvtq_m_n(float16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f16_u16)))
+float16x8_t __arm_vcvtq_m_n_f16_u16(float16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f16_u16)))
+float16x8_t __arm_vcvtq_m_n(float16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f32_s32)))
+float32x4_t __arm_vcvtq_m_n_f32_s32(float32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f32_s32)))
+float32x4_t __arm_vcvtq_m_n(float32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f32_u32)))
+float32x4_t __arm_vcvtq_m_n_f32_u32(float32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f32_u32)))
+float32x4_t __arm_vcvtq_m_n(float32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_s16_f16)))
+int16x8_t __arm_vcvtq_m_n_s16_f16(int16x8_t, float16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_s16_f16)))
+int16x8_t __arm_vcvtq_m_n(int16x8_t, float16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_s32_f32)))
+int32x4_t __arm_vcvtq_m_n_s32_f32(int32x4_t, float32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_s32_f32)))
+int32x4_t __arm_vcvtq_m_n(int32x4_t, float32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_u16_f16)))
+uint16x8_t __arm_vcvtq_m_n_u16_f16(uint16x8_t, float16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_u16_f16)))
+uint16x8_t __arm_vcvtq_m_n(uint16x8_t, float16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_u32_f32)))
+uint32x4_t __arm_vcvtq_m_n_u32_f32(uint32x4_t, float32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_u32_f32)))
+uint32x4_t __arm_vcvtq_m_n(uint32x4_t, float32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_s16_f16)))
+int16x8_t __arm_vcvtq_m_s16_f16(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_s16_f16)))
+int16x8_t __arm_vcvtq_m(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_s32_f32)))
+int32x4_t __arm_vcvtq_m_s32_f32(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_s32_f32)))
+int32x4_t __arm_vcvtq_m(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_u16_f16)))
+uint16x8_t __arm_vcvtq_m_u16_f16(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_u16_f16)))
+uint16x8_t __arm_vcvtq_m(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_u32_f32)))
+uint32x4_t __arm_vcvtq_m_u32_f32(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_u32_f32)))
+uint32x4_t __arm_vcvtq_m(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f16_s16)))
+float16x8_t __arm_vcvtq_n_f16_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f16_s16)))
+float16x8_t __arm_vcvtq_n(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f16_u16)))
+float16x8_t __arm_vcvtq_n_f16_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f16_u16)))
+float16x8_t __arm_vcvtq_n(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f32_s32)))
+float32x4_t __arm_vcvtq_n_f32_s32(int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f32_s32)))
+float32x4_t __arm_vcvtq_n(int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f32_u32)))
+float32x4_t __arm_vcvtq_n_f32_u32(uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f32_u32)))
+float32x4_t __arm_vcvtq_n(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_s16_f16)))
+int16x8_t __arm_vcvtq_n_s16_f16(float16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_s32_f32)))
+int32x4_t __arm_vcvtq_n_s32_f32(float32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_u16_f16)))
+uint16x8_t __arm_vcvtq_n_u16_f16(float16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_u32_f32)))
+uint32x4_t __arm_vcvtq_n_u32_f32(float32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_s16_f16)))
+int16x8_t __arm_vcvtq_s16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_s32_f32)))
+int32x4_t __arm_vcvtq_s32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_u16_f16)))
+uint16x8_t __arm_vcvtq_u16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_u32_f32)))
+uint32x4_t __arm_vcvtq_u32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f16_s16)))
+float16x8_t __arm_vcvtq_x_f16_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f16_s16)))
+float16x8_t __arm_vcvtq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f16_u16)))
+float16x8_t __arm_vcvtq_x_f16_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f16_u16)))
+float16x8_t __arm_vcvtq_x(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f32_s32)))
+float32x4_t __arm_vcvtq_x_f32_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f32_s32)))
+float32x4_t __arm_vcvtq_x(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f32_u32)))
+float32x4_t __arm_vcvtq_x_f32_u32(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f32_u32)))
+float32x4_t __arm_vcvtq_x(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f16_s16)))
+float16x8_t __arm_vcvtq_x_n_f16_s16(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f16_s16)))
+float16x8_t __arm_vcvtq_x_n(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f16_u16)))
+float16x8_t __arm_vcvtq_x_n_f16_u16(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f16_u16)))
+float16x8_t __arm_vcvtq_x_n(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f32_s32)))
+float32x4_t __arm_vcvtq_x_n_f32_s32(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f32_s32)))
+float32x4_t __arm_vcvtq_x_n(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f32_u32)))
+float32x4_t __arm_vcvtq_x_n_f32_u32(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f32_u32)))
+float32x4_t __arm_vcvtq_x_n(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_s16_f16)))
+int16x8_t __arm_vcvtq_x_n_s16_f16(float16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_s32_f32)))
+int32x4_t __arm_vcvtq_x_n_s32_f32(float32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_u16_f16)))
+uint16x8_t __arm_vcvtq_x_n_u16_f16(float16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_u32_f32)))
+uint32x4_t __arm_vcvtq_x_n_u32_f32(float32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_s16_f16)))
+int16x8_t __arm_vcvtq_x_s16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_s32_f32)))
+int32x4_t __arm_vcvtq_x_s32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_u16_f16)))
+uint16x8_t __arm_vcvtq_x_u16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_u32_f32)))
+uint32x4_t __arm_vcvtq_x_u32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvttq_f16_f32)))
+float16x8_t __arm_vcvttq_f16_f32(float16x8_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvttq_f32_f16)))
+float32x4_t __arm_vcvttq_f32_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvttq_m_f16_f32)))
+float16x8_t __arm_vcvttq_m_f16_f32(float16x8_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvttq_m_f32_f16)))
+float32x4_t __arm_vcvttq_m_f32_f16(float32x4_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvttq_x_f32_f16)))
+float32x4_t __arm_vcvttq_x_f32_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_f16)))
+float16x8_t __arm_vdupq_m_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_f16)))
+float16x8_t __arm_vdupq_m(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_f32)))
+float32x4_t __arm_vdupq_m_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_f32)))
+float32x4_t __arm_vdupq_m(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_f16)))
+float16x8_t __arm_vdupq_n_f16(float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_f32)))
+float32x4_t __arm_vdupq_n_f32(float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_f16)))
+float16x8_t __arm_vdupq_x_n_f16(float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_f32)))
+float32x4_t __arm_vdupq_x_n_f32(float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_f16)))
+float16x8_t __arm_veorq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_f16)))
+float16x8_t __arm_veorq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_f32)))
+float32x4_t __arm_veorq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_f32)))
+float32x4_t __arm_veorq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_f16)))
+float16x8_t __arm_veorq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_f16)))
+float16x8_t __arm_veorq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_f32)))
+float32x4_t __arm_veorq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_f32)))
+float32x4_t __arm_veorq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_f16)))
+float16x8_t __arm_veorq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_f16)))
+float16x8_t __arm_veorq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_f32)))
+float32x4_t __arm_veorq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_f32)))
+float32x4_t __arm_veorq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_f16)))
+float16x8_t __arm_vfmaq_f16(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_f16)))
+float16x8_t __arm_vfmaq(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_f32)))
+float32x4_t __arm_vfmaq_f32(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_f32)))
+float32x4_t __arm_vfmaq(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_f16)))
+float16x8_t __arm_vfmaq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_f16)))
+float16x8_t __arm_vfmaq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_f32)))
+float32x4_t __arm_vfmaq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_f32)))
+float32x4_t __arm_vfmaq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_n_f16)))
+float16x8_t __arm_vfmaq_m_n_f16(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_n_f16)))
+float16x8_t __arm_vfmaq_m(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_n_f32)))
+float32x4_t __arm_vfmaq_m_n_f32(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_n_f32)))
+float32x4_t __arm_vfmaq_m(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_n_f16)))
+float16x8_t __arm_vfmaq_n_f16(float16x8_t, float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_n_f16)))
+float16x8_t __arm_vfmaq(float16x8_t, float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_n_f32)))
+float32x4_t __arm_vfmaq_n_f32(float32x4_t, float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_n_f32)))
+float32x4_t __arm_vfmaq(float32x4_t, float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_m_n_f16)))
+float16x8_t __arm_vfmasq_m_n_f16(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_m_n_f16)))
+float16x8_t __arm_vfmasq_m(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_m_n_f32)))
+float32x4_t __arm_vfmasq_m_n_f32(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_m_n_f32)))
+float32x4_t __arm_vfmasq_m(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_n_f16)))
+float16x8_t __arm_vfmasq_n_f16(float16x8_t, float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_n_f16)))
+float16x8_t __arm_vfmasq(float16x8_t, float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_n_f32)))
+float32x4_t __arm_vfmasq_n_f32(float32x4_t, float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_n_f32)))
+float32x4_t __arm_vfmasq(float32x4_t, float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_f16)))
+float16x8_t __arm_vfmsq_f16(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_f16)))
+float16x8_t __arm_vfmsq(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_f32)))
+float32x4_t __arm_vfmsq_f32(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_f32)))
+float32x4_t __arm_vfmsq(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_m_f16)))
+float16x8_t __arm_vfmsq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_m_f16)))
+float16x8_t __arm_vfmsq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_m_f32)))
+float32x4_t __arm_vfmsq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_m_f32)))
+float32x4_t __arm_vfmsq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_f16)))
+float16_t __arm_vgetq_lane_f16(float16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_f16)))
+float16_t __arm_vgetq_lane(float16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_f32)))
+float32_t __arm_vgetq_lane_f32(float32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_f32)))
+float32_t __arm_vgetq_lane(float32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_f16)))
+float16x8_t __arm_vld1q_f16(const float16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_f16)))
+float16x8_t __arm_vld1q(const float16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_f32)))
+float32x4_t __arm_vld1q_f32(const float32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_f32)))
+float32x4_t __arm_vld1q(const float32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_f16)))
+float16x8_t __arm_vld1q_z_f16(const float16_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_f16)))
+float16x8_t __arm_vld1q_z(const float16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_f32)))
+float32x4_t __arm_vld1q_z_f32(const float32_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_f32)))
+float32x4_t __arm_vld1q_z(const float32_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_f16)))
+float16x8x2_t __arm_vld2q_f16(const float16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_f16)))
+float16x8x2_t __arm_vld2q(const float16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_f32)))
+float32x4x2_t __arm_vld2q_f32(const float32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_f32)))
+float32x4x2_t __arm_vld2q(const float32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_f16)))
+float16x8x4_t __arm_vld4q_f16(const float16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_f16)))
+float16x8x4_t __arm_vld4q(const float16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_f32)))
+float32x4x4_t __arm_vld4q_f32(const float32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_f32)))
+float32x4x4_t __arm_vld4q(const float32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_f16)))
+float16x8_t __arm_vldrhq_f16(const float16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_f16)))
+float16x8_t __arm_vldrhq_gather_offset_f16(const float16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_f16)))
+float16x8_t __arm_vldrhq_gather_offset(const float16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_f16)))
+float16x8_t __arm_vldrhq_gather_offset_z_f16(const float16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_f16)))
+float16x8_t __arm_vldrhq_gather_offset_z(const float16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_f16)))
+float16x8_t __arm_vldrhq_gather_shifted_offset_f16(const float16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_f16)))
+float16x8_t __arm_vldrhq_gather_shifted_offset(const float16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_f16)))
+float16x8_t __arm_vldrhq_gather_shifted_offset_z_f16(const float16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_f16)))
+float16x8_t __arm_vldrhq_gather_shifted_offset_z(const float16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_z_f16)))
+float16x8_t __arm_vldrhq_z_f16(const float16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_f32)))
+float32x4_t __arm_vldrwq_f32(const float32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_f32)))
+float32x4_t __arm_vldrwq_gather_base_f32(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_wb_f32)))
+float32x4_t __arm_vldrwq_gather_base_wb_f32(uint32x4_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_wb_z_f32)))
+float32x4_t __arm_vldrwq_gather_base_wb_z_f32(uint32x4_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_z_f32)))
+float32x4_t __arm_vldrwq_gather_base_z_f32(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_f32)))
+float32x4_t __arm_vldrwq_gather_offset_f32(const float32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_f32)))
+float32x4_t __arm_vldrwq_gather_offset(const float32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_z_f32)))
+float32x4_t __arm_vldrwq_gather_offset_z_f32(const float32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_z_f32)))
+float32x4_t __arm_vldrwq_gather_offset_z(const float32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_f32)))
+float32x4_t __arm_vldrwq_gather_shifted_offset_f32(const float32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_f32)))
+float32x4_t __arm_vldrwq_gather_shifted_offset(const float32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_z_f32)))
+float32x4_t __arm_vldrwq_gather_shifted_offset_z_f32(const float32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_z_f32)))
+float32x4_t __arm_vldrwq_gather_shifted_offset_z(const float32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_z_f32)))
+float32x4_t __arm_vldrwq_z_f32(const float32_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_f16)))
+float16x8_t __arm_vmaxnmaq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_f16)))
+float16x8_t __arm_vmaxnmaq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_f32)))
+float32x4_t __arm_vmaxnmaq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_f32)))
+float32x4_t __arm_vmaxnmaq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_m_f16)))
+float16x8_t __arm_vmaxnmaq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_m_f16)))
+float16x8_t __arm_vmaxnmaq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_m_f32)))
+float32x4_t __arm_vmaxnmaq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_m_f32)))
+float32x4_t __arm_vmaxnmaq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_f16)))
+float16_t __arm_vmaxnmavq_f16(float16_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_f16)))
+float16_t __arm_vmaxnmavq(float16_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_f32)))
+float32_t __arm_vmaxnmavq_f32(float32_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_f32)))
+float32_t __arm_vmaxnmavq(float32_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_p_f16)))
+float16_t __arm_vmaxnmavq_p_f16(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_p_f16)))
+float16_t __arm_vmaxnmavq_p(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_p_f32)))
+float32_t __arm_vmaxnmavq_p_f32(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_p_f32)))
+float32_t __arm_vmaxnmavq_p(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_f16)))
+float16x8_t __arm_vmaxnmq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_f16)))
+float16x8_t __arm_vmaxnmq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_f32)))
+float32x4_t __arm_vmaxnmq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_f32)))
+float32x4_t __arm_vmaxnmq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_m_f16)))
+float16x8_t __arm_vmaxnmq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_m_f16)))
+float16x8_t __arm_vmaxnmq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_m_f32)))
+float32x4_t __arm_vmaxnmq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_m_f32)))
+float32x4_t __arm_vmaxnmq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_x_f16)))
+float16x8_t __arm_vmaxnmq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_x_f16)))
+float16x8_t __arm_vmaxnmq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_x_f32)))
+float32x4_t __arm_vmaxnmq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_x_f32)))
+float32x4_t __arm_vmaxnmq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_f16)))
+float16_t __arm_vmaxnmvq_f16(float16_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_f16)))
+float16_t __arm_vmaxnmvq(float16_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_f32)))
+float32_t __arm_vmaxnmvq_f32(float32_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_f32)))
+float32_t __arm_vmaxnmvq(float32_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_p_f16)))
+float16_t __arm_vmaxnmvq_p_f16(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_p_f16)))
+float16_t __arm_vmaxnmvq_p(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_p_f32)))
+float32_t __arm_vmaxnmvq_p_f32(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_p_f32)))
+float32_t __arm_vmaxnmvq_p(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_f16)))
+float16x8_t __arm_vminnmaq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_f16)))
+float16x8_t __arm_vminnmaq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_f32)))
+float32x4_t __arm_vminnmaq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_f32)))
+float32x4_t __arm_vminnmaq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_m_f16)))
+float16x8_t __arm_vminnmaq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_m_f16)))
+float16x8_t __arm_vminnmaq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_m_f32)))
+float32x4_t __arm_vminnmaq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_m_f32)))
+float32x4_t __arm_vminnmaq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_f16)))
+float16_t __arm_vminnmavq_f16(float16_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_f16)))
+float16_t __arm_vminnmavq(float16_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_f32)))
+float32_t __arm_vminnmavq_f32(float32_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_f32)))
+float32_t __arm_vminnmavq(float32_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_p_f16)))
+float16_t __arm_vminnmavq_p_f16(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_p_f16)))
+float16_t __arm_vminnmavq_p(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_p_f32)))
+float32_t __arm_vminnmavq_p_f32(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_p_f32)))
+float32_t __arm_vminnmavq_p(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_f16)))
+float16x8_t __arm_vminnmq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_f16)))
+float16x8_t __arm_vminnmq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_f32)))
+float32x4_t __arm_vminnmq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_f32)))
+float32x4_t __arm_vminnmq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_m_f16)))
+float16x8_t __arm_vminnmq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_m_f16)))
+float16x8_t __arm_vminnmq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_m_f32)))
+float32x4_t __arm_vminnmq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_m_f32)))
+float32x4_t __arm_vminnmq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_x_f16)))
+float16x8_t __arm_vminnmq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_x_f16)))
+float16x8_t __arm_vminnmq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_x_f32)))
+float32x4_t __arm_vminnmq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_x_f32)))
+float32x4_t __arm_vminnmq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_f16)))
+float16_t __arm_vminnmvq_f16(float16_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_f16)))
+float16_t __arm_vminnmvq(float16_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_f32)))
+float32_t __arm_vminnmvq_f32(float32_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_f32)))
+float32_t __arm_vminnmvq(float32_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_p_f16)))
+float16_t __arm_vminnmvq_p_f16(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_p_f16)))
+float16_t __arm_vminnmvq_p(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_p_f32)))
+float32_t __arm_vminnmvq_p_f32(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_p_f32)))
+float32_t __arm_vminnmvq_p(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_f16)))
+float16x8_t __arm_vmulq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_f16)))
+float16x8_t __arm_vmulq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_f32)))
+float32x4_t __arm_vmulq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_f32)))
+float32x4_t __arm_vmulq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_f16)))
+float16x8_t __arm_vmulq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_f16)))
+float16x8_t __arm_vmulq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_f32)))
+float32x4_t __arm_vmulq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_f32)))
+float32x4_t __arm_vmulq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_f16)))
+float16x8_t __arm_vmulq_m_n_f16(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_f16)))
+float16x8_t __arm_vmulq_m(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_f32)))
+float32x4_t __arm_vmulq_m_n_f32(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_f32)))
+float32x4_t __arm_vmulq_m(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_f16)))
+float16x8_t __arm_vmulq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_f16)))
+float16x8_t __arm_vmulq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_f32)))
+float32x4_t __arm_vmulq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_f32)))
+float32x4_t __arm_vmulq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_f16)))
+float16x8_t __arm_vmulq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_f16)))
+float16x8_t __arm_vmulq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_f32)))
+float32x4_t __arm_vmulq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_f32)))
+float32x4_t __arm_vmulq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_f16)))
+float16x8_t __arm_vmulq_x_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_f16)))
+float16x8_t __arm_vmulq_x(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_f32)))
+float32x4_t __arm_vmulq_x_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_f32)))
+float32x4_t __arm_vmulq_x(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_f16)))
+float16x8_t __arm_vnegq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_f16)))
+float16x8_t __arm_vnegq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_f32)))
+float32x4_t __arm_vnegq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_f32)))
+float32x4_t __arm_vnegq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_f16)))
+float16x8_t __arm_vnegq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_f16)))
+float16x8_t __arm_vnegq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_f32)))
+float32x4_t __arm_vnegq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_f32)))
+float32x4_t __arm_vnegq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_f16)))
+float16x8_t __arm_vnegq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_f16)))
+float16x8_t __arm_vnegq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_f32)))
+float32x4_t __arm_vnegq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_f32)))
+float32x4_t __arm_vnegq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_f16)))
+float16x8_t __arm_vornq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_f16)))
+float16x8_t __arm_vornq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_f32)))
+float32x4_t __arm_vornq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_f32)))
+float32x4_t __arm_vornq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_f16)))
+float16x8_t __arm_vornq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_f16)))
+float16x8_t __arm_vornq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_f32)))
+float32x4_t __arm_vornq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_f32)))
+float32x4_t __arm_vornq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_f16)))
+float16x8_t __arm_vornq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_f16)))
+float16x8_t __arm_vornq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_f32)))
+float32x4_t __arm_vornq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_f32)))
+float32x4_t __arm_vornq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_f16)))
+float16x8_t __arm_vorrq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_f16)))
+float16x8_t __arm_vorrq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_f32)))
+float32x4_t __arm_vorrq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_f32)))
+float32x4_t __arm_vorrq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_f16)))
+float16x8_t __arm_vorrq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_f16)))
+float16x8_t __arm_vorrq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_f32)))
+float32x4_t __arm_vorrq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_f32)))
+float32x4_t __arm_vorrq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_f16)))
+float16x8_t __arm_vorrq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_f16)))
+float16x8_t __arm_vorrq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_f32)))
+float32x4_t __arm_vorrq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_f32)))
+float32x4_t __arm_vorrq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_f16)))
+float16x8_t __arm_vpselq_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_f16)))
+float16x8_t __arm_vpselq(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_f32)))
+float32x4_t __arm_vpselq_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_f32)))
+float32x4_t __arm_vpselq(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_f32)))
+float16x8_t __arm_vreinterpretq_f16_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_f32)))
+float16x8_t __arm_vreinterpretq_f16(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s16)))
+float16x8_t __arm_vreinterpretq_f16_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s16)))
+float16x8_t __arm_vreinterpretq_f16(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s32)))
+float16x8_t __arm_vreinterpretq_f16_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s32)))
+float16x8_t __arm_vreinterpretq_f16(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s64)))
+float16x8_t __arm_vreinterpretq_f16_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s64)))
+float16x8_t __arm_vreinterpretq_f16(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s8)))
+float16x8_t __arm_vreinterpretq_f16_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s8)))
+float16x8_t __arm_vreinterpretq_f16(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u16)))
+float16x8_t __arm_vreinterpretq_f16_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u16)))
+float16x8_t __arm_vreinterpretq_f16(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u32)))
+float16x8_t __arm_vreinterpretq_f16_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u32)))
+float16x8_t __arm_vreinterpretq_f16(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u64)))
+float16x8_t __arm_vreinterpretq_f16_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u64)))
+float16x8_t __arm_vreinterpretq_f16(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u8)))
+float16x8_t __arm_vreinterpretq_f16_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u8)))
+float16x8_t __arm_vreinterpretq_f16(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_f16)))
+float32x4_t __arm_vreinterpretq_f32_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_f16)))
+float32x4_t __arm_vreinterpretq_f32(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s16)))
+float32x4_t __arm_vreinterpretq_f32_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s16)))
+float32x4_t __arm_vreinterpretq_f32(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s32)))
+float32x4_t __arm_vreinterpretq_f32_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s32)))
+float32x4_t __arm_vreinterpretq_f32(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s64)))
+float32x4_t __arm_vreinterpretq_f32_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s64)))
+float32x4_t __arm_vreinterpretq_f32(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s8)))
+float32x4_t __arm_vreinterpretq_f32_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s8)))
+float32x4_t __arm_vreinterpretq_f32(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u16)))
+float32x4_t __arm_vreinterpretq_f32_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u16)))
+float32x4_t __arm_vreinterpretq_f32(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u32)))
+float32x4_t __arm_vreinterpretq_f32_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u32)))
+float32x4_t __arm_vreinterpretq_f32(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u64)))
+float32x4_t __arm_vreinterpretq_f32_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u64)))
+float32x4_t __arm_vreinterpretq_f32(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u8)))
+float32x4_t __arm_vreinterpretq_f32_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u8)))
+float32x4_t __arm_vreinterpretq_f32(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_f16)))
+int16x8_t __arm_vreinterpretq_s16_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_f16)))
+int16x8_t __arm_vreinterpretq_s16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_f32)))
+int16x8_t __arm_vreinterpretq_s16_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_f32)))
+int16x8_t __arm_vreinterpretq_s16(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_f16)))
+int32x4_t __arm_vreinterpretq_s32_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_f16)))
+int32x4_t __arm_vreinterpretq_s32(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_f32)))
+int32x4_t __arm_vreinterpretq_s32_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_f32)))
+int32x4_t __arm_vreinterpretq_s32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_f16)))
+int64x2_t __arm_vreinterpretq_s64_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_f16)))
+int64x2_t __arm_vreinterpretq_s64(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_f32)))
+int64x2_t __arm_vreinterpretq_s64_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_f32)))
+int64x2_t __arm_vreinterpretq_s64(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_f16)))
+int8x16_t __arm_vreinterpretq_s8_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_f16)))
+int8x16_t __arm_vreinterpretq_s8(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_f32)))
+int8x16_t __arm_vreinterpretq_s8_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_f32)))
+int8x16_t __arm_vreinterpretq_s8(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_f16)))
+uint16x8_t __arm_vreinterpretq_u16_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_f16)))
+uint16x8_t __arm_vreinterpretq_u16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_f32)))
+uint16x8_t __arm_vreinterpretq_u16_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_f32)))
+uint16x8_t __arm_vreinterpretq_u16(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_f16)))
+uint32x4_t __arm_vreinterpretq_u32_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_f16)))
+uint32x4_t __arm_vreinterpretq_u32(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_f32)))
+uint32x4_t __arm_vreinterpretq_u32_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_f32)))
+uint32x4_t __arm_vreinterpretq_u32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_f16)))
+uint64x2_t __arm_vreinterpretq_u64_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_f16)))
+uint64x2_t __arm_vreinterpretq_u64(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_f32)))
+uint64x2_t __arm_vreinterpretq_u64_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_f32)))
+uint64x2_t __arm_vreinterpretq_u64(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_f16)))
+uint8x16_t __arm_vreinterpretq_u8_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_f16)))
+uint8x16_t __arm_vreinterpretq_u8(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_f32)))
+uint8x16_t __arm_vreinterpretq_u8_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_f32)))
+uint8x16_t __arm_vreinterpretq_u8(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_f16)))
+float16x8_t __arm_vrev32q_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_f16)))
+float16x8_t __arm_vrev32q(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_f16)))
+float16x8_t __arm_vrev32q_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_f16)))
+float16x8_t __arm_vrev32q_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_f16)))
+float16x8_t __arm_vrev32q_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_f16)))
+float16x8_t __arm_vrev32q_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_f16)))
+float16x8_t __arm_vrev64q_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_f16)))
+float16x8_t __arm_vrev64q(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_f32)))
+float32x4_t __arm_vrev64q_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_f32)))
+float32x4_t __arm_vrev64q(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_f16)))
+float16x8_t __arm_vrev64q_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_f16)))
+float16x8_t __arm_vrev64q_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_f32)))
+float32x4_t __arm_vrev64q_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_f32)))
+float32x4_t __arm_vrev64q_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_f16)))
+float16x8_t __arm_vrev64q_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_f16)))
+float16x8_t __arm_vrev64q_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_f32)))
+float32x4_t __arm_vrev64q_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_f32)))
+float32x4_t __arm_vrev64q_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_f16)))
+float16x8_t __arm_vrndaq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_f16)))
+float16x8_t __arm_vrndaq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_f32)))
+float32x4_t __arm_vrndaq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_f32)))
+float32x4_t __arm_vrndaq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_m_f16)))
+float16x8_t __arm_vrndaq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_m_f16)))
+float16x8_t __arm_vrndaq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_m_f32)))
+float32x4_t __arm_vrndaq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_m_f32)))
+float32x4_t __arm_vrndaq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_x_f16)))
+float16x8_t __arm_vrndaq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_x_f16)))
+float16x8_t __arm_vrndaq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_x_f32)))
+float32x4_t __arm_vrndaq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_x_f32)))
+float32x4_t __arm_vrndaq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_f16)))
+float16x8_t __arm_vrndmq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_f16)))
+float16x8_t __arm_vrndmq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_f32)))
+float32x4_t __arm_vrndmq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_f32)))
+float32x4_t __arm_vrndmq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_m_f16)))
+float16x8_t __arm_vrndmq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_m_f16)))
+float16x8_t __arm_vrndmq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_m_f32)))
+float32x4_t __arm_vrndmq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_m_f32)))
+float32x4_t __arm_vrndmq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_x_f16)))
+float16x8_t __arm_vrndmq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_x_f16)))
+float16x8_t __arm_vrndmq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_x_f32)))
+float32x4_t __arm_vrndmq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_x_f32)))
+float32x4_t __arm_vrndmq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_f16)))
+float16x8_t __arm_vrndnq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_f16)))
+float16x8_t __arm_vrndnq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_f32)))
+float32x4_t __arm_vrndnq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_f32)))
+float32x4_t __arm_vrndnq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_m_f16)))
+float16x8_t __arm_vrndnq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_m_f16)))
+float16x8_t __arm_vrndnq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_m_f32)))
+float32x4_t __arm_vrndnq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_m_f32)))
+float32x4_t __arm_vrndnq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_x_f16)))
+float16x8_t __arm_vrndnq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_x_f16)))
+float16x8_t __arm_vrndnq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_x_f32)))
+float32x4_t __arm_vrndnq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_x_f32)))
+float32x4_t __arm_vrndnq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_f16)))
+float16x8_t __arm_vrndpq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_f16)))
+float16x8_t __arm_vrndpq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_f32)))
+float32x4_t __arm_vrndpq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_f32)))
+float32x4_t __arm_vrndpq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_m_f16)))
+float16x8_t __arm_vrndpq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_m_f16)))
+float16x8_t __arm_vrndpq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_m_f32)))
+float32x4_t __arm_vrndpq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_m_f32)))
+float32x4_t __arm_vrndpq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_x_f16)))
+float16x8_t __arm_vrndpq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_x_f16)))
+float16x8_t __arm_vrndpq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_x_f32)))
+float32x4_t __arm_vrndpq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_x_f32)))
+float32x4_t __arm_vrndpq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndq_f16)))
+float16x8_t __arm_vrndq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndq_f16)))
+float16x8_t __arm_vrndq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndq_f32)))
+float32x4_t __arm_vrndq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndq_f32)))
+float32x4_t __arm_vrndq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndq_m_f16)))
+float16x8_t __arm_vrndq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndq_m_f16)))
+float16x8_t __arm_vrndq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndq_m_f32)))
+float32x4_t __arm_vrndq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndq_m_f32)))
+float32x4_t __arm_vrndq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndq_x_f16)))
+float16x8_t __arm_vrndq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndq_x_f16)))
+float16x8_t __arm_vrndq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndq_x_f32)))
+float32x4_t __arm_vrndq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndq_x_f32)))
+float32x4_t __arm_vrndq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_f16)))
+float16x8_t __arm_vrndxq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_f16)))
+float16x8_t __arm_vrndxq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_f32)))
+float32x4_t __arm_vrndxq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_f32)))
+float32x4_t __arm_vrndxq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_m_f16)))
+float16x8_t __arm_vrndxq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_m_f16)))
+float16x8_t __arm_vrndxq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_m_f32)))
+float32x4_t __arm_vrndxq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_m_f32)))
+float32x4_t __arm_vrndxq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_x_f16)))
+float16x8_t __arm_vrndxq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_x_f16)))
+float16x8_t __arm_vrndxq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_x_f32)))
+float32x4_t __arm_vrndxq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_x_f32)))
+float32x4_t __arm_vrndxq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_f16)))
+float16x8_t __arm_vsetq_lane_f16(float16_t, float16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_f16)))
+float16x8_t __arm_vsetq_lane(float16_t, float16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_f32)))
+float32x4_t __arm_vsetq_lane_f32(float32_t, float32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_f32)))
+float32x4_t __arm_vsetq_lane(float32_t, float32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_f16)))
+void __arm_vst1q_f16(float16_t *, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_f16)))
+void __arm_vst1q(float16_t *, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_f32)))
+void __arm_vst1q_f32(float32_t *, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_f32)))
+void __arm_vst1q(float32_t *, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_f16)))
+void __arm_vst1q_p_f16(float16_t *, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_f16)))
+void __arm_vst1q_p(float16_t *, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_f32)))
+void __arm_vst1q_p_f32(float32_t *, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_f32)))
+void __arm_vst1q_p(float32_t *, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_f16)))
+void __arm_vst2q_f16(float16_t *, float16x8x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_f16)))
+void __arm_vst2q(float16_t *, float16x8x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_f32)))
+void __arm_vst2q_f32(float32_t *, float32x4x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_f32)))
+void __arm_vst2q(float32_t *, float32x4x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_f16)))
+void __arm_vst4q_f16(float16_t *, float16x8x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_f16)))
+void __arm_vst4q(float16_t *, float16x8x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_f32)))
+void __arm_vst4q_f32(float32_t *, float32x4x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_f32)))
+void __arm_vst4q(float32_t *, float32x4x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_f16)))
+void __arm_vstrhq_f16(float16_t *, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_f16)))
+void __arm_vstrhq(float16_t *, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_f16)))
+void __arm_vstrhq_p_f16(float16_t *, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_f16)))
+void __arm_vstrhq_p(float16_t *, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_f16)))
+void __arm_vstrhq_scatter_offset_f16(float16_t *, uint16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_f16)))
+void __arm_vstrhq_scatter_offset(float16_t *, uint16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_f16)))
+void __arm_vstrhq_scatter_offset_p_f16(float16_t *, uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_f16)))
+void __arm_vstrhq_scatter_offset_p(float16_t *, uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_f16)))
+void __arm_vstrhq_scatter_shifted_offset_f16(float16_t *, uint16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_f16)))
+void __arm_vstrhq_scatter_shifted_offset(float16_t *, uint16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_f16)))
+void __arm_vstrhq_scatter_shifted_offset_p_f16(float16_t *, uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_f16)))
+void __arm_vstrhq_scatter_shifted_offset_p(float16_t *, uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_f32)))
+void __arm_vstrwq_f32(float32_t *, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_f32)))
+void __arm_vstrwq(float32_t *, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_p_f32)))
+void __arm_vstrwq_p_f32(float32_t *, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_p_f32)))
+void __arm_vstrwq_p(float32_t *, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_f32)))
+void __arm_vstrwq_scatter_base_f32(uint32x4_t, int, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_f32)))
+void __arm_vstrwq_scatter_base(uint32x4_t, int, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_p_f32)))
+void __arm_vstrwq_scatter_base_p_f32(uint32x4_t, int, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_p_f32)))
+void __arm_vstrwq_scatter_base_p(uint32x4_t, int, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_f32)))
+void __arm_vstrwq_scatter_base_wb_f32(uint32x4_t *, int, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_f32)))
+void __arm_vstrwq_scatter_base_wb(uint32x4_t *, int, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_p_f32)))
+void __arm_vstrwq_scatter_base_wb_p_f32(uint32x4_t *, int, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_p_f32)))
+void __arm_vstrwq_scatter_base_wb_p(uint32x4_t *, int, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_f32)))
+void __arm_vstrwq_scatter_offset_f32(float32_t *, uint32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_f32)))
+void __arm_vstrwq_scatter_offset(float32_t *, uint32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_p_f32)))
+void __arm_vstrwq_scatter_offset_p_f32(float32_t *, uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_p_f32)))
+void __arm_vstrwq_scatter_offset_p(float32_t *, uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_f32)))
+void __arm_vstrwq_scatter_shifted_offset_f32(float32_t *, uint32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_f32)))
+void __arm_vstrwq_scatter_shifted_offset(float32_t *, uint32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_p_f32)))
+void __arm_vstrwq_scatter_shifted_offset_p_f32(float32_t *, uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_p_f32)))
+void __arm_vstrwq_scatter_shifted_offset_p(float32_t *, uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_f16)))
+float16x8_t __arm_vsubq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_f16)))
+float16x8_t __arm_vsubq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_f32)))
+float32x4_t __arm_vsubq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_f32)))
+float32x4_t __arm_vsubq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_f16)))
+float16x8_t __arm_vsubq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_f16)))
+float16x8_t __arm_vsubq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_f32)))
+float32x4_t __arm_vsubq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_f32)))
+float32x4_t __arm_vsubq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_f16)))
+float16x8_t __arm_vsubq_m_n_f16(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_f16)))
+float16x8_t __arm_vsubq_m(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_f32)))
+float32x4_t __arm_vsubq_m_n_f32(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_f32)))
+float32x4_t __arm_vsubq_m(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_f16)))
+float16x8_t __arm_vsubq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_f16)))
+float16x8_t __arm_vsubq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_f32)))
+float32x4_t __arm_vsubq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_f32)))
+float32x4_t __arm_vsubq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_f16)))
+float16x8_t __arm_vsubq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_f16)))
+float16x8_t __arm_vsubq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_f32)))
+float32x4_t __arm_vsubq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_f32)))
+float32x4_t __arm_vsubq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_f16)))
+float16x8_t __arm_vsubq_x_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_f16)))
+float16x8_t __arm_vsubq_x(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_f32)))
+float32x4_t __arm_vsubq_x_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_f32)))
+float32x4_t __arm_vsubq_x(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_f16)))
+float16x8_t __arm_vuninitializedq_f16();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_f32)))
+float32x4_t __arm_vuninitializedq_f32();
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_f16)))
+float16x8_t __arm_vuninitializedq(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_f32)))
+float32x4_t __arm_vuninitializedq(float32x4_t);
+
+#endif /* (__ARM_FEATURE_MVE & 2) */
+
+#if (!defined __ARM_MVE_PRESERVE_USER_NAMESPACE)
+
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_asrl)))
+int64_t asrl(int64_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_lsll)))
+uint64_t lsll(uint64_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_sqrshr)))
+int32_t sqrshr(int32_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_sqrshrl)))
+int64_t sqrshrl(int64_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_sqrshrl_sat48)))
+int64_t sqrshrl_sat48(int64_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_sqshl)))
+int32_t sqshl(int32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_sqshll)))
+int64_t sqshll(int64_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_srshr)))
+int32_t srshr(int32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_srshrl)))
+int64_t srshrl(int64_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_uqrshl)))
+uint32_t uqrshl(uint32_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_uqrshll)))
+uint64_t uqrshll(uint64_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_uqrshll_sat48)))
+uint64_t uqrshll_sat48(uint64_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_uqshl)))
+uint32_t uqshl(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_uqshll)))
+uint64_t uqshll(uint64_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_urshr)))
+uint32_t urshr(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_urshrl)))
+uint64_t urshrl(uint64_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_s16)))
+uint32_t vabavq_p_s16(uint32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_s16)))
+uint32_t vabavq_p(uint32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_s32)))
+uint32_t vabavq_p_s32(uint32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_s32)))
+uint32_t vabavq_p(uint32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_s8)))
+uint32_t vabavq_p_s8(uint32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_s8)))
+uint32_t vabavq_p(uint32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_u16)))
+uint32_t vabavq_p_u16(uint32_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_u16)))
+uint32_t vabavq_p(uint32_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_u32)))
+uint32_t vabavq_p_u32(uint32_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_u32)))
+uint32_t vabavq_p(uint32_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_u8)))
+uint32_t vabavq_p_u8(uint32_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_p_u8)))
+uint32_t vabavq_p(uint32_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_s16)))
+uint32_t vabavq_s16(uint32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_s16)))
+uint32_t vabavq(uint32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_s32)))
+uint32_t vabavq_s32(uint32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_s32)))
+uint32_t vabavq(uint32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_s8)))
+uint32_t vabavq_s8(uint32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_s8)))
+uint32_t vabavq(uint32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_u16)))
+uint32_t vabavq_u16(uint32_t, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_u16)))
+uint32_t vabavq(uint32_t, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_u32)))
+uint32_t vabavq_u32(uint32_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_u32)))
+uint32_t vabavq(uint32_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabavq_u8)))
+uint32_t vabavq_u8(uint32_t, uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabavq_u8)))
+uint32_t vabavq(uint32_t, uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_s16)))
+int16x8_t vabdq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_s16)))
+int16x8_t vabdq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_s32)))
+int32x4_t vabdq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_s32)))
+int32x4_t vabdq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_s8)))
+int8x16_t vabdq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_s8)))
+int8x16_t vabdq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_u16)))
+uint16x8_t vabdq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_u16)))
+uint16x8_t vabdq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_u32)))
+uint32x4_t vabdq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_u32)))
+uint32x4_t vabdq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_u8)))
+uint8x16_t vabdq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_u8)))
+uint8x16_t vabdq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_s16)))
+int16x8_t vabdq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_s16)))
+int16x8_t vabdq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_s32)))
+int32x4_t vabdq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_s32)))
+int32x4_t vabdq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_s8)))
+int8x16_t vabdq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_s8)))
+int8x16_t vabdq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_u16)))
+uint16x8_t vabdq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_u16)))
+uint16x8_t vabdq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_u32)))
+uint32x4_t vabdq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_u32)))
+uint32x4_t vabdq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_u8)))
+uint8x16_t vabdq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_u8)))
+uint8x16_t vabdq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_s16)))
+int16x8_t vabdq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_s16)))
+int16x8_t vabdq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_s32)))
+int32x4_t vabdq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_s32)))
+int32x4_t vabdq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_s8)))
+int8x16_t vabdq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_s8)))
+int8x16_t vabdq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_u16)))
+uint16x8_t vabdq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_u16)))
+uint16x8_t vabdq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_u32)))
+uint32x4_t vabdq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_u32)))
+uint32x4_t vabdq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_u8)))
+uint8x16_t vabdq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_u8)))
+uint8x16_t vabdq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_s16)))
+int16x8_t vabsq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_s16)))
+int16x8_t vabsq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_s32)))
+int32x4_t vabsq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_s32)))
+int32x4_t vabsq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_s8)))
+int8x16_t vabsq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_s8)))
+int8x16_t vabsq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_s16)))
+int16x8_t vabsq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_s16)))
+int16x8_t vabsq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_s32)))
+int32x4_t vabsq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_s32)))
+int32x4_t vabsq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_s8)))
+int8x16_t vabsq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_s8)))
+int8x16_t vabsq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_s16)))
+int16x8_t vabsq_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_s16)))
+int16x8_t vabsq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_s32)))
+int32x4_t vabsq_x_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_s32)))
+int32x4_t vabsq_x(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_s8)))
+int8x16_t vabsq_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_s8)))
+int8x16_t vabsq_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadciq_m_s32)))
+int32x4_t vadciq_m_s32(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadciq_m_s32)))
+int32x4_t vadciq_m(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadciq_m_u32)))
+uint32x4_t vadciq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadciq_m_u32)))
+uint32x4_t vadciq_m(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadciq_s32)))
+int32x4_t vadciq_s32(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadciq_s32)))
+int32x4_t vadciq(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadciq_u32)))
+uint32x4_t vadciq_u32(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadciq_u32)))
+uint32x4_t vadciq(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadcq_m_s32)))
+int32x4_t vadcq_m_s32(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadcq_m_s32)))
+int32x4_t vadcq_m(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadcq_m_u32)))
+uint32x4_t vadcq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadcq_m_u32)))
+uint32x4_t vadcq_m(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadcq_s32)))
+int32x4_t vadcq_s32(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadcq_s32)))
+int32x4_t vadcq(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vadcq_u32)))
+uint32x4_t vadcq_u32(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vadcq_u32)))
+uint32x4_t vadcq(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_p_s32)))
+int64_t vaddlvaq_p_s32(int64_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_p_s32)))
+int64_t vaddlvaq_p(int64_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_p_u32)))
+uint64_t vaddlvaq_p_u32(uint64_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_p_u32)))
+uint64_t vaddlvaq_p(uint64_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_s32)))
+int64_t vaddlvaq_s32(int64_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_s32)))
+int64_t vaddlvaq(int64_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_u32)))
+uint64_t vaddlvaq_u32(uint64_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvaq_u32)))
+uint64_t vaddlvaq(uint64_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_p_s32)))
+int64_t vaddlvq_p_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_p_s32)))
+int64_t vaddlvq_p(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_p_u32)))
+uint64_t vaddlvq_p_u32(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_p_u32)))
+uint64_t vaddlvq_p(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_s32)))
+int64_t vaddlvq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_s32)))
+int64_t vaddlvq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_u32)))
+uint64_t vaddlvq_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddlvq_u32)))
+uint64_t vaddlvq(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_s16)))
+int16x8_t vaddq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_s16)))
+int16x8_t vaddq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_s32)))
+int32x4_t vaddq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_s32)))
+int32x4_t vaddq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_s8)))
+int8x16_t vaddq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_s8)))
+int8x16_t vaddq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_u16)))
+uint16x8_t vaddq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_u16)))
+uint16x8_t vaddq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_u32)))
+uint32x4_t vaddq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_u32)))
+uint32x4_t vaddq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_u8)))
+uint8x16_t vaddq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_u8)))
+uint8x16_t vaddq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_s16)))
+int16x8_t vaddq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_s16)))
+int16x8_t vaddq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_s32)))
+int32x4_t vaddq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_s32)))
+int32x4_t vaddq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_s8)))
+int8x16_t vaddq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_s8)))
+int8x16_t vaddq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_u16)))
+uint16x8_t vaddq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_u16)))
+uint16x8_t vaddq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_u32)))
+uint32x4_t vaddq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_u32)))
+uint32x4_t vaddq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_u8)))
+uint8x16_t vaddq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_u8)))
+uint8x16_t vaddq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_s16)))
+int16x8_t vaddq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_s16)))
+int16x8_t vaddq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_s32)))
+int32x4_t vaddq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_s32)))
+int32x4_t vaddq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_s8)))
+int8x16_t vaddq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_s8)))
+int8x16_t vaddq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_u16)))
+uint16x8_t vaddq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_u16)))
+uint16x8_t vaddq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_u32)))
+uint32x4_t vaddq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_u32)))
+uint32x4_t vaddq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_u8)))
+uint8x16_t vaddq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_u8)))
+uint8x16_t vaddq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_s16)))
+int16x8_t vaddq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_s16)))
+int16x8_t vaddq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_s32)))
+int32x4_t vaddq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_s32)))
+int32x4_t vaddq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_s8)))
+int8x16_t vaddq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_s8)))
+int8x16_t vaddq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_u16)))
+uint16x8_t vaddq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_u16)))
+uint16x8_t vaddq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_u32)))
+uint32x4_t vaddq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_u32)))
+uint32x4_t vaddq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_u8)))
+uint8x16_t vaddq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_u8)))
+uint8x16_t vaddq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_s16)))
+int16x8_t vaddq_x_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_s16)))
+int16x8_t vaddq_x(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_s32)))
+int32x4_t vaddq_x_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_s32)))
+int32x4_t vaddq_x(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_s8)))
+int8x16_t vaddq_x_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_s8)))
+int8x16_t vaddq_x(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_u16)))
+uint16x8_t vaddq_x_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_u16)))
+uint16x8_t vaddq_x(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_u32)))
+uint32x4_t vaddq_x_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_u32)))
+uint32x4_t vaddq_x(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_u8)))
+uint8x16_t vaddq_x_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_u8)))
+uint8x16_t vaddq_x(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_s16)))
+int16x8_t vaddq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_s16)))
+int16x8_t vaddq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_s32)))
+int32x4_t vaddq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_s32)))
+int32x4_t vaddq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_s8)))
+int8x16_t vaddq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_s8)))
+int8x16_t vaddq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_u16)))
+uint16x8_t vaddq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_u16)))
+uint16x8_t vaddq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_u32)))
+uint32x4_t vaddq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_u32)))
+uint32x4_t vaddq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_u8)))
+uint8x16_t vaddq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_u8)))
+uint8x16_t vaddq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_s16)))
+int32_t vaddvaq_p_s16(int32_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_s16)))
+int32_t vaddvaq_p(int32_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_s32)))
+int32_t vaddvaq_p_s32(int32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_s32)))
+int32_t vaddvaq_p(int32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_s8)))
+int32_t vaddvaq_p_s8(int32_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_s8)))
+int32_t vaddvaq_p(int32_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_u16)))
+uint32_t vaddvaq_p_u16(uint32_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_u16)))
+uint32_t vaddvaq_p(uint32_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_u32)))
+uint32_t vaddvaq_p_u32(uint32_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_u32)))
+uint32_t vaddvaq_p(uint32_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_u8)))
+uint32_t vaddvaq_p_u8(uint32_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_p_u8)))
+uint32_t vaddvaq_p(uint32_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_s16)))
+int32_t vaddvaq_s16(int32_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_s16)))
+int32_t vaddvaq(int32_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_s32)))
+int32_t vaddvaq_s32(int32_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_s32)))
+int32_t vaddvaq(int32_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_s8)))
+int32_t vaddvaq_s8(int32_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_s8)))
+int32_t vaddvaq(int32_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_u16)))
+uint32_t vaddvaq_u16(uint32_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_u16)))
+uint32_t vaddvaq(uint32_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_u32)))
+uint32_t vaddvaq_u32(uint32_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_u32)))
+uint32_t vaddvaq(uint32_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_u8)))
+uint32_t vaddvaq_u8(uint32_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvaq_u8)))
+uint32_t vaddvaq(uint32_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_s16)))
+int32_t vaddvq_p_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_s16)))
+int32_t vaddvq_p(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_s32)))
+int32_t vaddvq_p_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_s32)))
+int32_t vaddvq_p(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_s8)))
+int32_t vaddvq_p_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_s8)))
+int32_t vaddvq_p(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_u16)))
+uint32_t vaddvq_p_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_u16)))
+uint32_t vaddvq_p(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_u32)))
+uint32_t vaddvq_p_u32(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_u32)))
+uint32_t vaddvq_p(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_u8)))
+uint32_t vaddvq_p_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_p_u8)))
+uint32_t vaddvq_p(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_s16)))
+int32_t vaddvq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_s16)))
+int32_t vaddvq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_s32)))
+int32_t vaddvq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_s32)))
+int32_t vaddvq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_s8)))
+int32_t vaddvq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_s8)))
+int32_t vaddvq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_u16)))
+uint32_t vaddvq_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_u16)))
+uint32_t vaddvq(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_u32)))
+uint32_t vaddvq_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_u32)))
+uint32_t vaddvq(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_u8)))
+uint32_t vaddvq_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddvq_u8)))
+uint32_t vaddvq(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_s16)))
+int16x8_t vandq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_s16)))
+int16x8_t vandq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_s32)))
+int32x4_t vandq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_s32)))
+int32x4_t vandq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_s8)))
+int8x16_t vandq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_s8)))
+int8x16_t vandq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_u16)))
+uint16x8_t vandq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_u16)))
+uint16x8_t vandq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_u32)))
+uint32x4_t vandq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_u32)))
+uint32x4_t vandq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_u8)))
+uint8x16_t vandq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_u8)))
+uint8x16_t vandq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_s16)))
+int16x8_t vandq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_s16)))
+int16x8_t vandq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_s32)))
+int32x4_t vandq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_s32)))
+int32x4_t vandq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_s8)))
+int8x16_t vandq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_s8)))
+int8x16_t vandq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_u16)))
+uint16x8_t vandq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_u16)))
+uint16x8_t vandq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_u32)))
+uint32x4_t vandq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_u32)))
+uint32x4_t vandq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_u8)))
+uint8x16_t vandq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_u8)))
+uint8x16_t vandq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_s16)))
+int16x8_t vandq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_s16)))
+int16x8_t vandq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_s32)))
+int32x4_t vandq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_s32)))
+int32x4_t vandq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_s8)))
+int8x16_t vandq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_s8)))
+int8x16_t vandq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_u16)))
+uint16x8_t vandq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_u16)))
+uint16x8_t vandq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_u32)))
+uint32x4_t vandq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_u32)))
+uint32x4_t vandq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_u8)))
+uint8x16_t vandq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_u8)))
+uint8x16_t vandq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_s16)))
+int16x8_t vbicq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_s16)))
+int16x8_t vbicq_m_n(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_s32)))
+int32x4_t vbicq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_s32)))
+int32x4_t vbicq_m_n(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_u16)))
+uint16x8_t vbicq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_u16)))
+uint16x8_t vbicq_m_n(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_u32)))
+uint32x4_t vbicq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_n_u32)))
+uint32x4_t vbicq_m_n(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_s16)))
+int16x8_t vbicq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_s16)))
+int16x8_t vbicq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_s32)))
+int32x4_t vbicq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_s32)))
+int32x4_t vbicq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_s8)))
+int8x16_t vbicq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_s8)))
+int8x16_t vbicq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_u16)))
+uint16x8_t vbicq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_u16)))
+uint16x8_t vbicq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_u32)))
+uint32x4_t vbicq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_u32)))
+uint32x4_t vbicq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_u8)))
+uint8x16_t vbicq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_u8)))
+uint8x16_t vbicq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_s16)))
+int16x8_t vbicq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_s16)))
+int16x8_t vbicq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_s32)))
+int32x4_t vbicq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_s32)))
+int32x4_t vbicq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_u16)))
+uint16x8_t vbicq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_u16)))
+uint16x8_t vbicq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_u32)))
+uint32x4_t vbicq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_n_u32)))
+uint32x4_t vbicq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_s16)))
+int16x8_t vbicq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_s16)))
+int16x8_t vbicq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_s32)))
+int32x4_t vbicq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_s32)))
+int32x4_t vbicq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_s8)))
+int8x16_t vbicq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_s8)))
+int8x16_t vbicq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_u16)))
+uint16x8_t vbicq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_u16)))
+uint16x8_t vbicq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_u32)))
+uint32x4_t vbicq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_u32)))
+uint32x4_t vbicq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_u8)))
+uint8x16_t vbicq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_u8)))
+uint8x16_t vbicq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_s16)))
+int16x8_t vbicq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_s16)))
+int16x8_t vbicq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_s32)))
+int32x4_t vbicq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_s32)))
+int32x4_t vbicq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_s8)))
+int8x16_t vbicq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_s8)))
+int8x16_t vbicq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_u16)))
+uint16x8_t vbicq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_u16)))
+uint16x8_t vbicq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_u32)))
+uint32x4_t vbicq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_u32)))
+uint32x4_t vbicq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_u8)))
+uint8x16_t vbicq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_u8)))
+uint8x16_t vbicq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_s16)))
+int16x8_t vbrsrq_m_n_s16(int16x8_t, int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_s16)))
+int16x8_t vbrsrq_m(int16x8_t, int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_s32)))
+int32x4_t vbrsrq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_s32)))
+int32x4_t vbrsrq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_s8)))
+int8x16_t vbrsrq_m_n_s8(int8x16_t, int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_s8)))
+int8x16_t vbrsrq_m(int8x16_t, int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_u16)))
+uint16x8_t vbrsrq_m_n_u16(uint16x8_t, uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_u16)))
+uint16x8_t vbrsrq_m(uint16x8_t, uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_u32)))
+uint32x4_t vbrsrq_m_n_u32(uint32x4_t, uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_u32)))
+uint32x4_t vbrsrq_m(uint32x4_t, uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_u8)))
+uint8x16_t vbrsrq_m_n_u8(uint8x16_t, uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_u8)))
+uint8x16_t vbrsrq_m(uint8x16_t, uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_s16)))
+int16x8_t vbrsrq_n_s16(int16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_s16)))
+int16x8_t vbrsrq(int16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_s32)))
+int32x4_t vbrsrq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_s32)))
+int32x4_t vbrsrq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_s8)))
+int8x16_t vbrsrq_n_s8(int8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_s8)))
+int8x16_t vbrsrq(int8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_u16)))
+uint16x8_t vbrsrq_n_u16(uint16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_u16)))
+uint16x8_t vbrsrq(uint16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_u32)))
+uint32x4_t vbrsrq_n_u32(uint32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_u32)))
+uint32x4_t vbrsrq(uint32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_u8)))
+uint8x16_t vbrsrq_n_u8(uint8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_u8)))
+uint8x16_t vbrsrq(uint8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_s16)))
+int16x8_t vbrsrq_x_n_s16(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_s16)))
+int16x8_t vbrsrq_x(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_s32)))
+int32x4_t vbrsrq_x_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_s32)))
+int32x4_t vbrsrq_x(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_s8)))
+int8x16_t vbrsrq_x_n_s8(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_s8)))
+int8x16_t vbrsrq_x(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_u16)))
+uint16x8_t vbrsrq_x_n_u16(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_u16)))
+uint16x8_t vbrsrq_x(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_u32)))
+uint32x4_t vbrsrq_x_n_u32(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_u32)))
+uint32x4_t vbrsrq_x(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_u8)))
+uint8x16_t vbrsrq_x_n_u8(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_u8)))
+uint8x16_t vbrsrq_x(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_s16)))
+int16x8_t vcaddq_rot270_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_s16)))
+int16x8_t vcaddq_rot270_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_s32)))
+int32x4_t vcaddq_rot270_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_s32)))
+int32x4_t vcaddq_rot270_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_s8)))
+int8x16_t vcaddq_rot270_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_s8)))
+int8x16_t vcaddq_rot270_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_u16)))
+uint16x8_t vcaddq_rot270_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_u16)))
+uint16x8_t vcaddq_rot270_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_u32)))
+uint32x4_t vcaddq_rot270_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_u32)))
+uint32x4_t vcaddq_rot270_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_u8)))
+uint8x16_t vcaddq_rot270_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_u8)))
+uint8x16_t vcaddq_rot270_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_s16)))
+int16x8_t vcaddq_rot270_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_s16)))
+int16x8_t vcaddq_rot270(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_s32)))
+int32x4_t vcaddq_rot270_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_s32)))
+int32x4_t vcaddq_rot270(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_s8)))
+int8x16_t vcaddq_rot270_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_s8)))
+int8x16_t vcaddq_rot270(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_u16)))
+uint16x8_t vcaddq_rot270_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_u16)))
+uint16x8_t vcaddq_rot270(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_u32)))
+uint32x4_t vcaddq_rot270_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_u32)))
+uint32x4_t vcaddq_rot270(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_u8)))
+uint8x16_t vcaddq_rot270_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_u8)))
+uint8x16_t vcaddq_rot270(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_s16)))
+int16x8_t vcaddq_rot270_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_s16)))
+int16x8_t vcaddq_rot270_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_s32)))
+int32x4_t vcaddq_rot270_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_s32)))
+int32x4_t vcaddq_rot270_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_s8)))
+int8x16_t vcaddq_rot270_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_s8)))
+int8x16_t vcaddq_rot270_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_u16)))
+uint16x8_t vcaddq_rot270_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_u16)))
+uint16x8_t vcaddq_rot270_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_u32)))
+uint32x4_t vcaddq_rot270_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_u32)))
+uint32x4_t vcaddq_rot270_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_u8)))
+uint8x16_t vcaddq_rot270_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_u8)))
+uint8x16_t vcaddq_rot270_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_s16)))
+int16x8_t vcaddq_rot90_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_s16)))
+int16x8_t vcaddq_rot90_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_s32)))
+int32x4_t vcaddq_rot90_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_s32)))
+int32x4_t vcaddq_rot90_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_s8)))
+int8x16_t vcaddq_rot90_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_s8)))
+int8x16_t vcaddq_rot90_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_u16)))
+uint16x8_t vcaddq_rot90_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_u16)))
+uint16x8_t vcaddq_rot90_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_u32)))
+uint32x4_t vcaddq_rot90_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_u32)))
+uint32x4_t vcaddq_rot90_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_u8)))
+uint8x16_t vcaddq_rot90_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_u8)))
+uint8x16_t vcaddq_rot90_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_s16)))
+int16x8_t vcaddq_rot90_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_s16)))
+int16x8_t vcaddq_rot90(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_s32)))
+int32x4_t vcaddq_rot90_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_s32)))
+int32x4_t vcaddq_rot90(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_s8)))
+int8x16_t vcaddq_rot90_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_s8)))
+int8x16_t vcaddq_rot90(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_u16)))
+uint16x8_t vcaddq_rot90_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_u16)))
+uint16x8_t vcaddq_rot90(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_u32)))
+uint32x4_t vcaddq_rot90_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_u32)))
+uint32x4_t vcaddq_rot90(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_u8)))
+uint8x16_t vcaddq_rot90_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_u8)))
+uint8x16_t vcaddq_rot90(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_s16)))
+int16x8_t vcaddq_rot90_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_s16)))
+int16x8_t vcaddq_rot90_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_s32)))
+int32x4_t vcaddq_rot90_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_s32)))
+int32x4_t vcaddq_rot90_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_s8)))
+int8x16_t vcaddq_rot90_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_s8)))
+int8x16_t vcaddq_rot90_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_u16)))
+uint16x8_t vcaddq_rot90_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_u16)))
+uint16x8_t vcaddq_rot90_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_u32)))
+uint32x4_t vcaddq_rot90_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_u32)))
+uint32x4_t vcaddq_rot90_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_u8)))
+uint8x16_t vcaddq_rot90_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_u8)))
+uint8x16_t vcaddq_rot90_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_m_s16)))
+int16x8_t vclsq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_m_s16)))
+int16x8_t vclsq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_m_s32)))
+int32x4_t vclsq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_m_s32)))
+int32x4_t vclsq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_m_s8)))
+int8x16_t vclsq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_m_s8)))
+int8x16_t vclsq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_s16)))
+int16x8_t vclsq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_s16)))
+int16x8_t vclsq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_s32)))
+int32x4_t vclsq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_s32)))
+int32x4_t vclsq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_s8)))
+int8x16_t vclsq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_s8)))
+int8x16_t vclsq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_x_s16)))
+int16x8_t vclsq_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_x_s16)))
+int16x8_t vclsq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_x_s32)))
+int32x4_t vclsq_x_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_x_s32)))
+int32x4_t vclsq_x(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclsq_x_s8)))
+int8x16_t vclsq_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclsq_x_s8)))
+int8x16_t vclsq_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_s16)))
+int16x8_t vclzq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_s16)))
+int16x8_t vclzq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_s32)))
+int32x4_t vclzq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_s32)))
+int32x4_t vclzq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_s8)))
+int8x16_t vclzq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_s8)))
+int8x16_t vclzq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_u16)))
+uint16x8_t vclzq_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_u16)))
+uint16x8_t vclzq_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_u32)))
+uint32x4_t vclzq_m_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_u32)))
+uint32x4_t vclzq_m(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_u8)))
+uint8x16_t vclzq_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_m_u8)))
+uint8x16_t vclzq_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_s16)))
+int16x8_t vclzq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_s16)))
+int16x8_t vclzq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_s32)))
+int32x4_t vclzq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_s32)))
+int32x4_t vclzq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_s8)))
+int8x16_t vclzq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_s8)))
+int8x16_t vclzq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_u16)))
+uint16x8_t vclzq_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_u16)))
+uint16x8_t vclzq(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_u32)))
+uint32x4_t vclzq_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_u32)))
+uint32x4_t vclzq(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_u8)))
+uint8x16_t vclzq_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_u8)))
+uint8x16_t vclzq(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_s16)))
+int16x8_t vclzq_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_s16)))
+int16x8_t vclzq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_s32)))
+int32x4_t vclzq_x_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_s32)))
+int32x4_t vclzq_x(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_s8)))
+int8x16_t vclzq_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_s8)))
+int8x16_t vclzq_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_u16)))
+uint16x8_t vclzq_x_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_u16)))
+uint16x8_t vclzq_x(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_u32)))
+uint32x4_t vclzq_x_u32(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_u32)))
+uint32x4_t vclzq_x(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_u8)))
+uint8x16_t vclzq_x_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vclzq_x_u8)))
+uint8x16_t vclzq_x(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_n_u16)))
+mve_pred16_t vcmpcsq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_n_u16)))
+mve_pred16_t vcmpcsq_m(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_n_u32)))
+mve_pred16_t vcmpcsq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_n_u32)))
+mve_pred16_t vcmpcsq_m(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_n_u8)))
+mve_pred16_t vcmpcsq_m_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_n_u8)))
+mve_pred16_t vcmpcsq_m(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_u16)))
+mve_pred16_t vcmpcsq_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_u16)))
+mve_pred16_t vcmpcsq_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_u32)))
+mve_pred16_t vcmpcsq_m_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_u32)))
+mve_pred16_t vcmpcsq_m(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_u8)))
+mve_pred16_t vcmpcsq_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_m_u8)))
+mve_pred16_t vcmpcsq_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_n_u16)))
+mve_pred16_t vcmpcsq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_n_u16)))
+mve_pred16_t vcmpcsq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_n_u32)))
+mve_pred16_t vcmpcsq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_n_u32)))
+mve_pred16_t vcmpcsq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_n_u8)))
+mve_pred16_t vcmpcsq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_n_u8)))
+mve_pred16_t vcmpcsq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_u16)))
+mve_pred16_t vcmpcsq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_u16)))
+mve_pred16_t vcmpcsq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_u32)))
+mve_pred16_t vcmpcsq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_u32)))
+mve_pred16_t vcmpcsq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_u8)))
+mve_pred16_t vcmpcsq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpcsq_u8)))
+mve_pred16_t vcmpcsq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_s16)))
+mve_pred16_t vcmpeqq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_s16)))
+mve_pred16_t vcmpeqq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_s32)))
+mve_pred16_t vcmpeqq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_s32)))
+mve_pred16_t vcmpeqq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_s8)))
+mve_pred16_t vcmpeqq_m_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_s8)))
+mve_pred16_t vcmpeqq_m(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_u16)))
+mve_pred16_t vcmpeqq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_u16)))
+mve_pred16_t vcmpeqq_m(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_u32)))
+mve_pred16_t vcmpeqq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_u32)))
+mve_pred16_t vcmpeqq_m(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_u8)))
+mve_pred16_t vcmpeqq_m_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_u8)))
+mve_pred16_t vcmpeqq_m(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_s16)))
+mve_pred16_t vcmpeqq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_s16)))
+mve_pred16_t vcmpeqq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_s32)))
+mve_pred16_t vcmpeqq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_s32)))
+mve_pred16_t vcmpeqq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_s8)))
+mve_pred16_t vcmpeqq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_s8)))
+mve_pred16_t vcmpeqq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_u16)))
+mve_pred16_t vcmpeqq_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_u16)))
+mve_pred16_t vcmpeqq_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_u32)))
+mve_pred16_t vcmpeqq_m_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_u32)))
+mve_pred16_t vcmpeqq_m(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_u8)))
+mve_pred16_t vcmpeqq_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_u8)))
+mve_pred16_t vcmpeqq_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_s16)))
+mve_pred16_t vcmpeqq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_s16)))
+mve_pred16_t vcmpeqq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_s32)))
+mve_pred16_t vcmpeqq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_s32)))
+mve_pred16_t vcmpeqq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_s8)))
+mve_pred16_t vcmpeqq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_s8)))
+mve_pred16_t vcmpeqq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_u16)))
+mve_pred16_t vcmpeqq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_u16)))
+mve_pred16_t vcmpeqq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_u32)))
+mve_pred16_t vcmpeqq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_u32)))
+mve_pred16_t vcmpeqq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_u8)))
+mve_pred16_t vcmpeqq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_u8)))
+mve_pred16_t vcmpeqq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_s16)))
+mve_pred16_t vcmpeqq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_s16)))
+mve_pred16_t vcmpeqq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_s32)))
+mve_pred16_t vcmpeqq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_s32)))
+mve_pred16_t vcmpeqq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_s8)))
+mve_pred16_t vcmpeqq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_s8)))
+mve_pred16_t vcmpeqq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_u16)))
+mve_pred16_t vcmpeqq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_u16)))
+mve_pred16_t vcmpeqq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_u32)))
+mve_pred16_t vcmpeqq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_u32)))
+mve_pred16_t vcmpeqq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_u8)))
+mve_pred16_t vcmpeqq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_u8)))
+mve_pred16_t vcmpeqq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_s16)))
+mve_pred16_t vcmpgeq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_s16)))
+mve_pred16_t vcmpgeq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_s32)))
+mve_pred16_t vcmpgeq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_s32)))
+mve_pred16_t vcmpgeq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_s8)))
+mve_pred16_t vcmpgeq_m_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_s8)))
+mve_pred16_t vcmpgeq_m(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_s16)))
+mve_pred16_t vcmpgeq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_s16)))
+mve_pred16_t vcmpgeq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_s32)))
+mve_pred16_t vcmpgeq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_s32)))
+mve_pred16_t vcmpgeq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_s8)))
+mve_pred16_t vcmpgeq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_s8)))
+mve_pred16_t vcmpgeq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_s16)))
+mve_pred16_t vcmpgeq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_s16)))
+mve_pred16_t vcmpgeq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_s32)))
+mve_pred16_t vcmpgeq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_s32)))
+mve_pred16_t vcmpgeq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_s8)))
+mve_pred16_t vcmpgeq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_s8)))
+mve_pred16_t vcmpgeq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_s16)))
+mve_pred16_t vcmpgeq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_s16)))
+mve_pred16_t vcmpgeq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_s32)))
+mve_pred16_t vcmpgeq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_s32)))
+mve_pred16_t vcmpgeq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_s8)))
+mve_pred16_t vcmpgeq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_s8)))
+mve_pred16_t vcmpgeq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_s16)))
+mve_pred16_t vcmpgtq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_s16)))
+mve_pred16_t vcmpgtq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_s32)))
+mve_pred16_t vcmpgtq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_s32)))
+mve_pred16_t vcmpgtq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_s8)))
+mve_pred16_t vcmpgtq_m_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_s8)))
+mve_pred16_t vcmpgtq_m(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_s16)))
+mve_pred16_t vcmpgtq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_s16)))
+mve_pred16_t vcmpgtq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_s32)))
+mve_pred16_t vcmpgtq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_s32)))
+mve_pred16_t vcmpgtq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_s8)))
+mve_pred16_t vcmpgtq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_s8)))
+mve_pred16_t vcmpgtq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_s16)))
+mve_pred16_t vcmpgtq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_s16)))
+mve_pred16_t vcmpgtq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_s32)))
+mve_pred16_t vcmpgtq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_s32)))
+mve_pred16_t vcmpgtq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_s8)))
+mve_pred16_t vcmpgtq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_s8)))
+mve_pred16_t vcmpgtq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_s16)))
+mve_pred16_t vcmpgtq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_s16)))
+mve_pred16_t vcmpgtq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_s32)))
+mve_pred16_t vcmpgtq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_s32)))
+mve_pred16_t vcmpgtq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_s8)))
+mve_pred16_t vcmpgtq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_s8)))
+mve_pred16_t vcmpgtq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_n_u16)))
+mve_pred16_t vcmphiq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_n_u16)))
+mve_pred16_t vcmphiq_m(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_n_u32)))
+mve_pred16_t vcmphiq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_n_u32)))
+mve_pred16_t vcmphiq_m(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_n_u8)))
+mve_pred16_t vcmphiq_m_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_n_u8)))
+mve_pred16_t vcmphiq_m(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_u16)))
+mve_pred16_t vcmphiq_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_u16)))
+mve_pred16_t vcmphiq_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_u32)))
+mve_pred16_t vcmphiq_m_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_u32)))
+mve_pred16_t vcmphiq_m(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_u8)))
+mve_pred16_t vcmphiq_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_m_u8)))
+mve_pred16_t vcmphiq_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_n_u16)))
+mve_pred16_t vcmphiq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_n_u16)))
+mve_pred16_t vcmphiq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_n_u32)))
+mve_pred16_t vcmphiq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_n_u32)))
+mve_pred16_t vcmphiq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_n_u8)))
+mve_pred16_t vcmphiq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_n_u8)))
+mve_pred16_t vcmphiq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_u16)))
+mve_pred16_t vcmphiq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_u16)))
+mve_pred16_t vcmphiq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_u32)))
+mve_pred16_t vcmphiq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_u32)))
+mve_pred16_t vcmphiq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_u8)))
+mve_pred16_t vcmphiq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmphiq_u8)))
+mve_pred16_t vcmphiq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_s16)))
+mve_pred16_t vcmpleq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_s16)))
+mve_pred16_t vcmpleq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_s32)))
+mve_pred16_t vcmpleq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_s32)))
+mve_pred16_t vcmpleq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_s8)))
+mve_pred16_t vcmpleq_m_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_s8)))
+mve_pred16_t vcmpleq_m(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_s16)))
+mve_pred16_t vcmpleq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_s16)))
+mve_pred16_t vcmpleq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_s32)))
+mve_pred16_t vcmpleq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_s32)))
+mve_pred16_t vcmpleq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_s8)))
+mve_pred16_t vcmpleq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_s8)))
+mve_pred16_t vcmpleq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_s16)))
+mve_pred16_t vcmpleq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_s16)))
+mve_pred16_t vcmpleq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_s32)))
+mve_pred16_t vcmpleq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_s32)))
+mve_pred16_t vcmpleq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_s8)))
+mve_pred16_t vcmpleq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_s8)))
+mve_pred16_t vcmpleq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_s16)))
+mve_pred16_t vcmpleq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_s16)))
+mve_pred16_t vcmpleq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_s32)))
+mve_pred16_t vcmpleq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_s32)))
+mve_pred16_t vcmpleq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_s8)))
+mve_pred16_t vcmpleq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_s8)))
+mve_pred16_t vcmpleq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_s16)))
+mve_pred16_t vcmpltq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_s16)))
+mve_pred16_t vcmpltq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_s32)))
+mve_pred16_t vcmpltq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_s32)))
+mve_pred16_t vcmpltq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_s8)))
+mve_pred16_t vcmpltq_m_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_s8)))
+mve_pred16_t vcmpltq_m(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_s16)))
+mve_pred16_t vcmpltq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_s16)))
+mve_pred16_t vcmpltq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_s32)))
+mve_pred16_t vcmpltq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_s32)))
+mve_pred16_t vcmpltq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_s8)))
+mve_pred16_t vcmpltq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_s8)))
+mve_pred16_t vcmpltq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_s16)))
+mve_pred16_t vcmpltq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_s16)))
+mve_pred16_t vcmpltq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_s32)))
+mve_pred16_t vcmpltq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_s32)))
+mve_pred16_t vcmpltq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_s8)))
+mve_pred16_t vcmpltq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_s8)))
+mve_pred16_t vcmpltq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_s16)))
+mve_pred16_t vcmpltq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_s16)))
+mve_pred16_t vcmpltq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_s32)))
+mve_pred16_t vcmpltq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_s32)))
+mve_pred16_t vcmpltq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_s8)))
+mve_pred16_t vcmpltq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_s8)))
+mve_pred16_t vcmpltq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_s16)))
+mve_pred16_t vcmpneq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_s16)))
+mve_pred16_t vcmpneq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_s32)))
+mve_pred16_t vcmpneq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_s32)))
+mve_pred16_t vcmpneq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_s8)))
+mve_pred16_t vcmpneq_m_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_s8)))
+mve_pred16_t vcmpneq_m(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_u16)))
+mve_pred16_t vcmpneq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_u16)))
+mve_pred16_t vcmpneq_m(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_u32)))
+mve_pred16_t vcmpneq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_u32)))
+mve_pred16_t vcmpneq_m(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_u8)))
+mve_pred16_t vcmpneq_m_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_u8)))
+mve_pred16_t vcmpneq_m(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_s16)))
+mve_pred16_t vcmpneq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_s16)))
+mve_pred16_t vcmpneq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_s32)))
+mve_pred16_t vcmpneq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_s32)))
+mve_pred16_t vcmpneq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_s8)))
+mve_pred16_t vcmpneq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_s8)))
+mve_pred16_t vcmpneq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_u16)))
+mve_pred16_t vcmpneq_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_u16)))
+mve_pred16_t vcmpneq_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_u32)))
+mve_pred16_t vcmpneq_m_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_u32)))
+mve_pred16_t vcmpneq_m(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_u8)))
+mve_pred16_t vcmpneq_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_u8)))
+mve_pred16_t vcmpneq_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_s16)))
+mve_pred16_t vcmpneq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_s16)))
+mve_pred16_t vcmpneq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_s32)))
+mve_pred16_t vcmpneq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_s32)))
+mve_pred16_t vcmpneq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_s8)))
+mve_pred16_t vcmpneq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_s8)))
+mve_pred16_t vcmpneq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_u16)))
+mve_pred16_t vcmpneq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_u16)))
+mve_pred16_t vcmpneq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_u32)))
+mve_pred16_t vcmpneq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_u32)))
+mve_pred16_t vcmpneq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_u8)))
+mve_pred16_t vcmpneq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_u8)))
+mve_pred16_t vcmpneq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_s16)))
+mve_pred16_t vcmpneq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_s16)))
+mve_pred16_t vcmpneq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_s32)))
+mve_pred16_t vcmpneq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_s32)))
+mve_pred16_t vcmpneq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_s8)))
+mve_pred16_t vcmpneq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_s8)))
+mve_pred16_t vcmpneq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_u16)))
+mve_pred16_t vcmpneq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_u16)))
+mve_pred16_t vcmpneq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_u32)))
+mve_pred16_t vcmpneq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_u32)))
+mve_pred16_t vcmpneq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_u8)))
+mve_pred16_t vcmpneq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_u8)))
+mve_pred16_t vcmpneq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_s16)))
+int16x8_t vcreateq_s16(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_s32)))
+int32x4_t vcreateq_s32(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_s64)))
+int64x2_t vcreateq_s64(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_s8)))
+int8x16_t vcreateq_s8(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_u16)))
+uint16x8_t vcreateq_u16(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_u32)))
+uint32x4_t vcreateq_u32(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_u64)))
+uint64x2_t vcreateq_u64(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_u8)))
+uint8x16_t vcreateq_u8(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp16q)))
+mve_pred16_t vctp16q(uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp16q_m)))
+mve_pred16_t vctp16q_m(uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp32q)))
+mve_pred16_t vctp32q(uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp32q_m)))
+mve_pred16_t vctp32q_m(uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp64q)))
+mve_pred16_t vctp64q(uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp64q_m)))
+mve_pred16_t vctp64q_m(uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp8q)))
+mve_pred16_t vctp8q(uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vctp8q_m)))
+mve_pred16_t vctp8q_m(uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_n_u16)))
+uint16x8_t vddupq_m_n_u16(uint16x8_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_n_u16)))
+uint16x8_t vddupq_m(uint16x8_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_n_u32)))
+uint32x4_t vddupq_m_n_u32(uint32x4_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_n_u32)))
+uint32x4_t vddupq_m(uint32x4_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_n_u8)))
+uint8x16_t vddupq_m_n_u8(uint8x16_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_n_u8)))
+uint8x16_t vddupq_m(uint8x16_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_wb_u16)))
+uint16x8_t vddupq_m_wb_u16(uint16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_wb_u16)))
+uint16x8_t vddupq_m(uint16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_wb_u32)))
+uint32x4_t vddupq_m_wb_u32(uint32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_wb_u32)))
+uint32x4_t vddupq_m(uint32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_wb_u8)))
+uint8x16_t vddupq_m_wb_u8(uint8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_m_wb_u8)))
+uint8x16_t vddupq_m(uint8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_n_u16)))
+uint16x8_t vddupq_n_u16(uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_n_u16)))
+uint16x8_t vddupq_u16(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_n_u32)))
+uint32x4_t vddupq_n_u32(uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_n_u32)))
+uint32x4_t vddupq_u32(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_n_u8)))
+uint8x16_t vddupq_n_u8(uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_n_u8)))
+uint8x16_t vddupq_u8(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_wb_u16)))
+uint16x8_t vddupq_wb_u16(uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_wb_u16)))
+uint16x8_t vddupq_u16(uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_wb_u32)))
+uint32x4_t vddupq_wb_u32(uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_wb_u32)))
+uint32x4_t vddupq_u32(uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_wb_u8)))
+uint8x16_t vddupq_wb_u8(uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_wb_u8)))
+uint8x16_t vddupq_u8(uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_n_u16)))
+uint16x8_t vddupq_x_n_u16(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_n_u16)))
+uint16x8_t vddupq_x_u16(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_n_u32)))
+uint32x4_t vddupq_x_n_u32(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_n_u32)))
+uint32x4_t vddupq_x_u32(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_n_u8)))
+uint8x16_t vddupq_x_n_u8(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_n_u8)))
+uint8x16_t vddupq_x_u8(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_wb_u16)))
+uint16x8_t vddupq_x_wb_u16(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_wb_u16)))
+uint16x8_t vddupq_x_u16(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_wb_u32)))
+uint32x4_t vddupq_x_wb_u32(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_wb_u32)))
+uint32x4_t vddupq_x_u32(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_wb_u8)))
+uint8x16_t vddupq_x_wb_u8(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vddupq_x_wb_u8)))
+uint8x16_t vddupq_x_u8(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_s16)))
+int16x8_t vdupq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_s16)))
+int16x8_t vdupq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_s32)))
+int32x4_t vdupq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_s32)))
+int32x4_t vdupq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_s8)))
+int8x16_t vdupq_m_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_s8)))
+int8x16_t vdupq_m(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_u16)))
+uint16x8_t vdupq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_u16)))
+uint16x8_t vdupq_m(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_u32)))
+uint32x4_t vdupq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_u32)))
+uint32x4_t vdupq_m(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_u8)))
+uint8x16_t vdupq_m_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_u8)))
+uint8x16_t vdupq_m(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_s16)))
+int16x8_t vdupq_n_s16(int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_s32)))
+int32x4_t vdupq_n_s32(int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_s8)))
+int8x16_t vdupq_n_s8(int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_u16)))
+uint16x8_t vdupq_n_u16(uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_u32)))
+uint32x4_t vdupq_n_u32(uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_u8)))
+uint8x16_t vdupq_n_u8(uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_s16)))
+int16x8_t vdupq_x_n_s16(int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_s32)))
+int32x4_t vdupq_x_n_s32(int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_s8)))
+int8x16_t vdupq_x_n_s8(int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_u16)))
+uint16x8_t vdupq_x_n_u16(uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_u32)))
+uint32x4_t vdupq_x_n_u32(uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_u8)))
+uint8x16_t vdupq_x_n_u8(uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_n_u16)))
+uint16x8_t vdwdupq_m_n_u16(uint16x8_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_n_u16)))
+uint16x8_t vdwdupq_m(uint16x8_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_n_u32)))
+uint32x4_t vdwdupq_m_n_u32(uint32x4_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_n_u32)))
+uint32x4_t vdwdupq_m(uint32x4_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_n_u8)))
+uint8x16_t vdwdupq_m_n_u8(uint8x16_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_n_u8)))
+uint8x16_t vdwdupq_m(uint8x16_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_wb_u16)))
+uint16x8_t vdwdupq_m_wb_u16(uint16x8_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_wb_u16)))
+uint16x8_t vdwdupq_m(uint16x8_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_wb_u32)))
+uint32x4_t vdwdupq_m_wb_u32(uint32x4_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_wb_u32)))
+uint32x4_t vdwdupq_m(uint32x4_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_wb_u8)))
+uint8x16_t vdwdupq_m_wb_u8(uint8x16_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_m_wb_u8)))
+uint8x16_t vdwdupq_m(uint8x16_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_n_u16)))
+uint16x8_t vdwdupq_n_u16(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_n_u16)))
+uint16x8_t vdwdupq_u16(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_n_u32)))
+uint32x4_t vdwdupq_n_u32(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_n_u32)))
+uint32x4_t vdwdupq_u32(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_n_u8)))
+uint8x16_t vdwdupq_n_u8(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_n_u8)))
+uint8x16_t vdwdupq_u8(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_wb_u16)))
+uint16x8_t vdwdupq_wb_u16(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_wb_u16)))
+uint16x8_t vdwdupq_u16(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_wb_u32)))
+uint32x4_t vdwdupq_wb_u32(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_wb_u32)))
+uint32x4_t vdwdupq_u32(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_wb_u8)))
+uint8x16_t vdwdupq_wb_u8(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_wb_u8)))
+uint8x16_t vdwdupq_u8(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_n_u16)))
+uint16x8_t vdwdupq_x_n_u16(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_n_u16)))
+uint16x8_t vdwdupq_x_u16(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_n_u32)))
+uint32x4_t vdwdupq_x_n_u32(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_n_u32)))
+uint32x4_t vdwdupq_x_u32(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_n_u8)))
+uint8x16_t vdwdupq_x_n_u8(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_n_u8)))
+uint8x16_t vdwdupq_x_u8(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_wb_u16)))
+uint16x8_t vdwdupq_x_wb_u16(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_wb_u16)))
+uint16x8_t vdwdupq_x_u16(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_wb_u32)))
+uint32x4_t vdwdupq_x_wb_u32(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_wb_u32)))
+uint32x4_t vdwdupq_x_u32(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_wb_u8)))
+uint8x16_t vdwdupq_x_wb_u8(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdwdupq_x_wb_u8)))
+uint8x16_t vdwdupq_x_u8(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_s16)))
+int16x8_t veorq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_s16)))
+int16x8_t veorq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_s32)))
+int32x4_t veorq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_s32)))
+int32x4_t veorq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_s8)))
+int8x16_t veorq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_s8)))
+int8x16_t veorq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_u16)))
+uint16x8_t veorq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_u16)))
+uint16x8_t veorq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_u32)))
+uint32x4_t veorq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_u32)))
+uint32x4_t veorq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_u8)))
+uint8x16_t veorq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_u8)))
+uint8x16_t veorq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_s16)))
+int16x8_t veorq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_s16)))
+int16x8_t veorq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_s32)))
+int32x4_t veorq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_s32)))
+int32x4_t veorq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_s8)))
+int8x16_t veorq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_s8)))
+int8x16_t veorq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_u16)))
+uint16x8_t veorq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_u16)))
+uint16x8_t veorq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_u32)))
+uint32x4_t veorq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_u32)))
+uint32x4_t veorq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_u8)))
+uint8x16_t veorq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_u8)))
+uint8x16_t veorq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_s16)))
+int16x8_t veorq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_s16)))
+int16x8_t veorq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_s32)))
+int32x4_t veorq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_s32)))
+int32x4_t veorq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_s8)))
+int8x16_t veorq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_s8)))
+int8x16_t veorq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_u16)))
+uint16x8_t veorq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_u16)))
+uint16x8_t veorq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_u32)))
+uint32x4_t veorq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_u32)))
+uint32x4_t veorq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_u8)))
+uint8x16_t veorq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_u8)))
+uint8x16_t veorq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s16)))
+int16_t vgetq_lane_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s16)))
+int16_t vgetq_lane(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s32)))
+int32_t vgetq_lane_s32(int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s32)))
+int32_t vgetq_lane(int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s64)))
+int64_t vgetq_lane_s64(int64x2_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s64)))
+int64_t vgetq_lane(int64x2_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s8)))
+int8_t vgetq_lane_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_s8)))
+int8_t vgetq_lane(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u16)))
+uint16_t vgetq_lane_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u16)))
+uint16_t vgetq_lane(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u32)))
+uint32_t vgetq_lane_u32(uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u32)))
+uint32_t vgetq_lane(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u64)))
+uint64_t vgetq_lane_u64(uint64x2_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u64)))
+uint64_t vgetq_lane(uint64x2_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u8)))
+uint8_t vgetq_lane_u8(uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_u8)))
+uint8_t vgetq_lane(uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_s16)))
+int16x8_t vhaddq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_s16)))
+int16x8_t vhaddq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_s32)))
+int32x4_t vhaddq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_s32)))
+int32x4_t vhaddq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_s8)))
+int8x16_t vhaddq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_s8)))
+int8x16_t vhaddq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_u16)))
+uint16x8_t vhaddq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_u16)))
+uint16x8_t vhaddq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_u32)))
+uint32x4_t vhaddq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_u32)))
+uint32x4_t vhaddq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_u8)))
+uint8x16_t vhaddq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_n_u8)))
+uint8x16_t vhaddq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_s16)))
+int16x8_t vhaddq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_s16)))
+int16x8_t vhaddq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_s32)))
+int32x4_t vhaddq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_s32)))
+int32x4_t vhaddq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_s8)))
+int8x16_t vhaddq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_s8)))
+int8x16_t vhaddq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_u16)))
+uint16x8_t vhaddq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_u16)))
+uint16x8_t vhaddq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_u32)))
+uint32x4_t vhaddq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_u32)))
+uint32x4_t vhaddq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_u8)))
+uint8x16_t vhaddq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_m_u8)))
+uint8x16_t vhaddq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_s16)))
+int16x8_t vhaddq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_s16)))
+int16x8_t vhaddq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_s32)))
+int32x4_t vhaddq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_s32)))
+int32x4_t vhaddq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_s8)))
+int8x16_t vhaddq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_s8)))
+int8x16_t vhaddq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_u16)))
+uint16x8_t vhaddq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_u16)))
+uint16x8_t vhaddq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_u32)))
+uint32x4_t vhaddq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_u32)))
+uint32x4_t vhaddq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_u8)))
+uint8x16_t vhaddq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_n_u8)))
+uint8x16_t vhaddq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_s16)))
+int16x8_t vhaddq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_s16)))
+int16x8_t vhaddq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_s32)))
+int32x4_t vhaddq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_s32)))
+int32x4_t vhaddq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_s8)))
+int8x16_t vhaddq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_s8)))
+int8x16_t vhaddq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_u16)))
+uint16x8_t vhaddq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_u16)))
+uint16x8_t vhaddq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_u32)))
+uint32x4_t vhaddq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_u32)))
+uint32x4_t vhaddq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_u8)))
+uint8x16_t vhaddq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_u8)))
+uint8x16_t vhaddq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_s16)))
+int16x8_t vhaddq_x_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_s16)))
+int16x8_t vhaddq_x(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_s32)))
+int32x4_t vhaddq_x_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_s32)))
+int32x4_t vhaddq_x(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_s8)))
+int8x16_t vhaddq_x_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_s8)))
+int8x16_t vhaddq_x(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_u16)))
+uint16x8_t vhaddq_x_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_u16)))
+uint16x8_t vhaddq_x(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_u32)))
+uint32x4_t vhaddq_x_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_u32)))
+uint32x4_t vhaddq_x(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_u8)))
+uint8x16_t vhaddq_x_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_n_u8)))
+uint8x16_t vhaddq_x(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_s16)))
+int16x8_t vhaddq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_s16)))
+int16x8_t vhaddq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_s32)))
+int32x4_t vhaddq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_s32)))
+int32x4_t vhaddq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_s8)))
+int8x16_t vhaddq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_s8)))
+int8x16_t vhaddq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_u16)))
+uint16x8_t vhaddq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_u16)))
+uint16x8_t vhaddq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_u32)))
+uint32x4_t vhaddq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_u32)))
+uint32x4_t vhaddq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_u8)))
+uint8x16_t vhaddq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhaddq_x_u8)))
+uint8x16_t vhaddq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_m_s16)))
+int16x8_t vhcaddq_rot270_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_m_s16)))
+int16x8_t vhcaddq_rot270_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_m_s32)))
+int32x4_t vhcaddq_rot270_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_m_s32)))
+int32x4_t vhcaddq_rot270_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_m_s8)))
+int8x16_t vhcaddq_rot270_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_m_s8)))
+int8x16_t vhcaddq_rot270_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_s16)))
+int16x8_t vhcaddq_rot270_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_s16)))
+int16x8_t vhcaddq_rot270(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_s32)))
+int32x4_t vhcaddq_rot270_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_s32)))
+int32x4_t vhcaddq_rot270(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_s8)))
+int8x16_t vhcaddq_rot270_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_s8)))
+int8x16_t vhcaddq_rot270(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_x_s16)))
+int16x8_t vhcaddq_rot270_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_x_s16)))
+int16x8_t vhcaddq_rot270_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_x_s32)))
+int32x4_t vhcaddq_rot270_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_x_s32)))
+int32x4_t vhcaddq_rot270_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_x_s8)))
+int8x16_t vhcaddq_rot270_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot270_x_s8)))
+int8x16_t vhcaddq_rot270_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_m_s16)))
+int16x8_t vhcaddq_rot90_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_m_s16)))
+int16x8_t vhcaddq_rot90_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_m_s32)))
+int32x4_t vhcaddq_rot90_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_m_s32)))
+int32x4_t vhcaddq_rot90_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_m_s8)))
+int8x16_t vhcaddq_rot90_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_m_s8)))
+int8x16_t vhcaddq_rot90_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_s16)))
+int16x8_t vhcaddq_rot90_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_s16)))
+int16x8_t vhcaddq_rot90(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_s32)))
+int32x4_t vhcaddq_rot90_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_s32)))
+int32x4_t vhcaddq_rot90(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_s8)))
+int8x16_t vhcaddq_rot90_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_s8)))
+int8x16_t vhcaddq_rot90(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_x_s16)))
+int16x8_t vhcaddq_rot90_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_x_s16)))
+int16x8_t vhcaddq_rot90_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_x_s32)))
+int32x4_t vhcaddq_rot90_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_x_s32)))
+int32x4_t vhcaddq_rot90_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_x_s8)))
+int8x16_t vhcaddq_rot90_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhcaddq_rot90_x_s8)))
+int8x16_t vhcaddq_rot90_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_s16)))
+int16x8_t vhsubq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_s16)))
+int16x8_t vhsubq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_s32)))
+int32x4_t vhsubq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_s32)))
+int32x4_t vhsubq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_s8)))
+int8x16_t vhsubq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_s8)))
+int8x16_t vhsubq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_u16)))
+uint16x8_t vhsubq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_u16)))
+uint16x8_t vhsubq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_u32)))
+uint32x4_t vhsubq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_u32)))
+uint32x4_t vhsubq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_u8)))
+uint8x16_t vhsubq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_n_u8)))
+uint8x16_t vhsubq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_s16)))
+int16x8_t vhsubq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_s16)))
+int16x8_t vhsubq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_s32)))
+int32x4_t vhsubq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_s32)))
+int32x4_t vhsubq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_s8)))
+int8x16_t vhsubq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_s8)))
+int8x16_t vhsubq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_u16)))
+uint16x8_t vhsubq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_u16)))
+uint16x8_t vhsubq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_u32)))
+uint32x4_t vhsubq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_u32)))
+uint32x4_t vhsubq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_u8)))
+uint8x16_t vhsubq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_m_u8)))
+uint8x16_t vhsubq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_s16)))
+int16x8_t vhsubq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_s16)))
+int16x8_t vhsubq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_s32)))
+int32x4_t vhsubq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_s32)))
+int32x4_t vhsubq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_s8)))
+int8x16_t vhsubq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_s8)))
+int8x16_t vhsubq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_u16)))
+uint16x8_t vhsubq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_u16)))
+uint16x8_t vhsubq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_u32)))
+uint32x4_t vhsubq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_u32)))
+uint32x4_t vhsubq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_u8)))
+uint8x16_t vhsubq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_n_u8)))
+uint8x16_t vhsubq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_s16)))
+int16x8_t vhsubq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_s16)))
+int16x8_t vhsubq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_s32)))
+int32x4_t vhsubq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_s32)))
+int32x4_t vhsubq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_s8)))
+int8x16_t vhsubq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_s8)))
+int8x16_t vhsubq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_u16)))
+uint16x8_t vhsubq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_u16)))
+uint16x8_t vhsubq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_u32)))
+uint32x4_t vhsubq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_u32)))
+uint32x4_t vhsubq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_u8)))
+uint8x16_t vhsubq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_u8)))
+uint8x16_t vhsubq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_s16)))
+int16x8_t vhsubq_x_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_s16)))
+int16x8_t vhsubq_x(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_s32)))
+int32x4_t vhsubq_x_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_s32)))
+int32x4_t vhsubq_x(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_s8)))
+int8x16_t vhsubq_x_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_s8)))
+int8x16_t vhsubq_x(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_u16)))
+uint16x8_t vhsubq_x_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_u16)))
+uint16x8_t vhsubq_x(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_u32)))
+uint32x4_t vhsubq_x_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_u32)))
+uint32x4_t vhsubq_x(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_u8)))
+uint8x16_t vhsubq_x_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_n_u8)))
+uint8x16_t vhsubq_x(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_s16)))
+int16x8_t vhsubq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_s16)))
+int16x8_t vhsubq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_s32)))
+int32x4_t vhsubq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_s32)))
+int32x4_t vhsubq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_s8)))
+int8x16_t vhsubq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_s8)))
+int8x16_t vhsubq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_u16)))
+uint16x8_t vhsubq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_u16)))
+uint16x8_t vhsubq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_u32)))
+uint32x4_t vhsubq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_u32)))
+uint32x4_t vhsubq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_u8)))
+uint8x16_t vhsubq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vhsubq_x_u8)))
+uint8x16_t vhsubq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_n_u16)))
+uint16x8_t vidupq_m_n_u16(uint16x8_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_n_u16)))
+uint16x8_t vidupq_m(uint16x8_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_n_u32)))
+uint32x4_t vidupq_m_n_u32(uint32x4_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_n_u32)))
+uint32x4_t vidupq_m(uint32x4_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_n_u8)))
+uint8x16_t vidupq_m_n_u8(uint8x16_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_n_u8)))
+uint8x16_t vidupq_m(uint8x16_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_wb_u16)))
+uint16x8_t vidupq_m_wb_u16(uint16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_wb_u16)))
+uint16x8_t vidupq_m(uint16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_wb_u32)))
+uint32x4_t vidupq_m_wb_u32(uint32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_wb_u32)))
+uint32x4_t vidupq_m(uint32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_wb_u8)))
+uint8x16_t vidupq_m_wb_u8(uint8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_m_wb_u8)))
+uint8x16_t vidupq_m(uint8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_n_u16)))
+uint16x8_t vidupq_n_u16(uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_n_u16)))
+uint16x8_t vidupq_u16(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_n_u32)))
+uint32x4_t vidupq_n_u32(uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_n_u32)))
+uint32x4_t vidupq_u32(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_n_u8)))
+uint8x16_t vidupq_n_u8(uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_n_u8)))
+uint8x16_t vidupq_u8(uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_wb_u16)))
+uint16x8_t vidupq_wb_u16(uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_wb_u16)))
+uint16x8_t vidupq_u16(uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_wb_u32)))
+uint32x4_t vidupq_wb_u32(uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_wb_u32)))
+uint32x4_t vidupq_u32(uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_wb_u8)))
+uint8x16_t vidupq_wb_u8(uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_wb_u8)))
+uint8x16_t vidupq_u8(uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_n_u16)))
+uint16x8_t vidupq_x_n_u16(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_n_u16)))
+uint16x8_t vidupq_x_u16(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_n_u32)))
+uint32x4_t vidupq_x_n_u32(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_n_u32)))
+uint32x4_t vidupq_x_u32(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_n_u8)))
+uint8x16_t vidupq_x_n_u8(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_n_u8)))
+uint8x16_t vidupq_x_u8(uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_wb_u16)))
+uint16x8_t vidupq_x_wb_u16(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_wb_u16)))
+uint16x8_t vidupq_x_u16(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_wb_u32)))
+uint32x4_t vidupq_x_wb_u32(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_wb_u32)))
+uint32x4_t vidupq_x_u32(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_wb_u8)))
+uint8x16_t vidupq_x_wb_u8(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vidupq_x_wb_u8)))
+uint8x16_t vidupq_x_u8(uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_n_u16)))
+uint16x8_t viwdupq_m_n_u16(uint16x8_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_n_u16)))
+uint16x8_t viwdupq_m(uint16x8_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_n_u32)))
+uint32x4_t viwdupq_m_n_u32(uint32x4_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_n_u32)))
+uint32x4_t viwdupq_m(uint32x4_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_n_u8)))
+uint8x16_t viwdupq_m_n_u8(uint8x16_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_n_u8)))
+uint8x16_t viwdupq_m(uint8x16_t, uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_wb_u16)))
+uint16x8_t viwdupq_m_wb_u16(uint16x8_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_wb_u16)))
+uint16x8_t viwdupq_m(uint16x8_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_wb_u32)))
+uint32x4_t viwdupq_m_wb_u32(uint32x4_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_wb_u32)))
+uint32x4_t viwdupq_m(uint32x4_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_wb_u8)))
+uint8x16_t viwdupq_m_wb_u8(uint8x16_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_m_wb_u8)))
+uint8x16_t viwdupq_m(uint8x16_t, uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_n_u16)))
+uint16x8_t viwdupq_n_u16(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_n_u16)))
+uint16x8_t viwdupq_u16(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_n_u32)))
+uint32x4_t viwdupq_n_u32(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_n_u32)))
+uint32x4_t viwdupq_u32(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_n_u8)))
+uint8x16_t viwdupq_n_u8(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_n_u8)))
+uint8x16_t viwdupq_u8(uint32_t, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_wb_u16)))
+uint16x8_t viwdupq_wb_u16(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_wb_u16)))
+uint16x8_t viwdupq_u16(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_wb_u32)))
+uint32x4_t viwdupq_wb_u32(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_wb_u32)))
+uint32x4_t viwdupq_u32(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_wb_u8)))
+uint8x16_t viwdupq_wb_u8(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_wb_u8)))
+uint8x16_t viwdupq_u8(uint32_t *, uint32_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_n_u16)))
+uint16x8_t viwdupq_x_n_u16(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_n_u16)))
+uint16x8_t viwdupq_x_u16(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_n_u32)))
+uint32x4_t viwdupq_x_n_u32(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_n_u32)))
+uint32x4_t viwdupq_x_u32(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_n_u8)))
+uint8x16_t viwdupq_x_n_u8(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_n_u8)))
+uint8x16_t viwdupq_x_u8(uint32_t, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_wb_u16)))
+uint16x8_t viwdupq_x_wb_u16(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_wb_u16)))
+uint16x8_t viwdupq_x_u16(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_wb_u32)))
+uint32x4_t viwdupq_x_wb_u32(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_wb_u32)))
+uint32x4_t viwdupq_x_u32(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_wb_u8)))
+uint8x16_t viwdupq_x_wb_u8(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_viwdupq_x_wb_u8)))
+uint8x16_t viwdupq_x_u8(uint32_t *, uint32_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_s16)))
+int16x8_t vld1q_s16(const int16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_s16)))
+int16x8_t vld1q(const int16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_s32)))
+int32x4_t vld1q_s32(const int32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_s32)))
+int32x4_t vld1q(const int32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_s8)))
+int8x16_t vld1q_s8(const int8_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_s8)))
+int8x16_t vld1q(const int8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_u16)))
+uint16x8_t vld1q_u16(const uint16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_u16)))
+uint16x8_t vld1q(const uint16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_u32)))
+uint32x4_t vld1q_u32(const uint32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_u32)))
+uint32x4_t vld1q(const uint32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_u8)))
+uint8x16_t vld1q_u8(const uint8_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_u8)))
+uint8x16_t vld1q(const uint8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_s16)))
+int16x8_t vld1q_z_s16(const int16_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_s16)))
+int16x8_t vld1q_z(const int16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_s32)))
+int32x4_t vld1q_z_s32(const int32_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_s32)))
+int32x4_t vld1q_z(const int32_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_s8)))
+int8x16_t vld1q_z_s8(const int8_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_s8)))
+int8x16_t vld1q_z(const int8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_u16)))
+uint16x8_t vld1q_z_u16(const uint16_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_u16)))
+uint16x8_t vld1q_z(const uint16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_u32)))
+uint32x4_t vld1q_z_u32(const uint32_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_u32)))
+uint32x4_t vld1q_z(const uint32_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_u8)))
+uint8x16_t vld1q_z_u8(const uint8_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_u8)))
+uint8x16_t vld1q_z(const uint8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_s16)))
+int16x8x2_t vld2q_s16(const int16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_s16)))
+int16x8x2_t vld2q(const int16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_s32)))
+int32x4x2_t vld2q_s32(const int32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_s32)))
+int32x4x2_t vld2q(const int32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_s8)))
+int8x16x2_t vld2q_s8(const int8_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_s8)))
+int8x16x2_t vld2q(const int8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_u16)))
+uint16x8x2_t vld2q_u16(const uint16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_u16)))
+uint16x8x2_t vld2q(const uint16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_u32)))
+uint32x4x2_t vld2q_u32(const uint32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_u32)))
+uint32x4x2_t vld2q(const uint32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_u8)))
+uint8x16x2_t vld2q_u8(const uint8_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_u8)))
+uint8x16x2_t vld2q(const uint8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_s16)))
+int16x8x4_t vld4q_s16(const int16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_s16)))
+int16x8x4_t vld4q(const int16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_s32)))
+int32x4x4_t vld4q_s32(const int32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_s32)))
+int32x4x4_t vld4q(const int32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_s8)))
+int8x16x4_t vld4q_s8(const int8_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_s8)))
+int8x16x4_t vld4q(const int8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_u16)))
+uint16x8x4_t vld4q_u16(const uint16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_u16)))
+uint16x8x4_t vld4q(const uint16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_u32)))
+uint32x4x4_t vld4q_u32(const uint32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_u32)))
+uint32x4x4_t vld4q(const uint32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_u8)))
+uint8x16x4_t vld4q_u8(const uint8_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_u8)))
+uint8x16x4_t vld4q(const uint8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_s16)))
+int16x8_t vldrbq_gather_offset_s16(const int8_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_s16)))
+int16x8_t vldrbq_gather_offset(const int8_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_s32)))
+int32x4_t vldrbq_gather_offset_s32(const int8_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_s32)))
+int32x4_t vldrbq_gather_offset(const int8_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_s8)))
+int8x16_t vldrbq_gather_offset_s8(const int8_t *, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_s8)))
+int8x16_t vldrbq_gather_offset(const int8_t *, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_u16)))
+uint16x8_t vldrbq_gather_offset_u16(const uint8_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_u16)))
+uint16x8_t vldrbq_gather_offset(const uint8_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_u32)))
+uint32x4_t vldrbq_gather_offset_u32(const uint8_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_u32)))
+uint32x4_t vldrbq_gather_offset(const uint8_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_u8)))
+uint8x16_t vldrbq_gather_offset_u8(const uint8_t *, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_u8)))
+uint8x16_t vldrbq_gather_offset(const uint8_t *, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_s16)))
+int16x8_t vldrbq_gather_offset_z_s16(const int8_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_s16)))
+int16x8_t vldrbq_gather_offset_z(const int8_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_s32)))
+int32x4_t vldrbq_gather_offset_z_s32(const int8_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_s32)))
+int32x4_t vldrbq_gather_offset_z(const int8_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_s8)))
+int8x16_t vldrbq_gather_offset_z_s8(const int8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_s8)))
+int8x16_t vldrbq_gather_offset_z(const int8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_u16)))
+uint16x8_t vldrbq_gather_offset_z_u16(const uint8_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_u16)))
+uint16x8_t vldrbq_gather_offset_z(const uint8_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_u32)))
+uint32x4_t vldrbq_gather_offset_z_u32(const uint8_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_u32)))
+uint32x4_t vldrbq_gather_offset_z(const uint8_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_u8)))
+uint8x16_t vldrbq_gather_offset_z_u8(const uint8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_gather_offset_z_u8)))
+uint8x16_t vldrbq_gather_offset_z(const uint8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_s16)))
+int16x8_t vldrbq_s16(const int8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_s32)))
+int32x4_t vldrbq_s32(const int8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_s8)))
+int8x16_t vldrbq_s8(const int8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_u16)))
+uint16x8_t vldrbq_u16(const uint8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_u32)))
+uint32x4_t vldrbq_u32(const uint8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_u8)))
+uint8x16_t vldrbq_u8(const uint8_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_z_s16)))
+int16x8_t vldrbq_z_s16(const int8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_z_s32)))
+int32x4_t vldrbq_z_s32(const int8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_z_s8)))
+int8x16_t vldrbq_z_s8(const int8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_z_u16)))
+uint16x8_t vldrbq_z_u16(const uint8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_z_u32)))
+uint32x4_t vldrbq_z_u32(const uint8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrbq_z_u8)))
+uint8x16_t vldrbq_z_u8(const uint8_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_s64)))
+int64x2_t vldrdq_gather_base_s64(uint64x2_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_u64)))
+uint64x2_t vldrdq_gather_base_u64(uint64x2_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_wb_s64)))
+int64x2_t vldrdq_gather_base_wb_s64(uint64x2_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_wb_u64)))
+uint64x2_t vldrdq_gather_base_wb_u64(uint64x2_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_wb_z_s64)))
+int64x2_t vldrdq_gather_base_wb_z_s64(uint64x2_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_wb_z_u64)))
+uint64x2_t vldrdq_gather_base_wb_z_u64(uint64x2_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_z_s64)))
+int64x2_t vldrdq_gather_base_z_s64(uint64x2_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_base_z_u64)))
+uint64x2_t vldrdq_gather_base_z_u64(uint64x2_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_s64)))
+int64x2_t vldrdq_gather_offset_s64(const int64_t *, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_s64)))
+int64x2_t vldrdq_gather_offset(const int64_t *, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_u64)))
+uint64x2_t vldrdq_gather_offset_u64(const uint64_t *, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_u64)))
+uint64x2_t vldrdq_gather_offset(const uint64_t *, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_z_s64)))
+int64x2_t vldrdq_gather_offset_z_s64(const int64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_z_s64)))
+int64x2_t vldrdq_gather_offset_z(const int64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_z_u64)))
+uint64x2_t vldrdq_gather_offset_z_u64(const uint64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_offset_z_u64)))
+uint64x2_t vldrdq_gather_offset_z(const uint64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_s64)))
+int64x2_t vldrdq_gather_shifted_offset_s64(const int64_t *, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_s64)))
+int64x2_t vldrdq_gather_shifted_offset(const int64_t *, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_u64)))
+uint64x2_t vldrdq_gather_shifted_offset_u64(const uint64_t *, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_u64)))
+uint64x2_t vldrdq_gather_shifted_offset(const uint64_t *, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_z_s64)))
+int64x2_t vldrdq_gather_shifted_offset_z_s64(const int64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_z_s64)))
+int64x2_t vldrdq_gather_shifted_offset_z(const int64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_z_u64)))
+uint64x2_t vldrdq_gather_shifted_offset_z_u64(const uint64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrdq_gather_shifted_offset_z_u64)))
+uint64x2_t vldrdq_gather_shifted_offset_z(const uint64_t *, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_s16)))
+int16x8_t vldrhq_gather_offset_s16(const int16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_s16)))
+int16x8_t vldrhq_gather_offset(const int16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_s32)))
+int32x4_t vldrhq_gather_offset_s32(const int16_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_s32)))
+int32x4_t vldrhq_gather_offset(const int16_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_u16)))
+uint16x8_t vldrhq_gather_offset_u16(const uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_u16)))
+uint16x8_t vldrhq_gather_offset(const uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_u32)))
+uint32x4_t vldrhq_gather_offset_u32(const uint16_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_u32)))
+uint32x4_t vldrhq_gather_offset(const uint16_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_s16)))
+int16x8_t vldrhq_gather_offset_z_s16(const int16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_s16)))
+int16x8_t vldrhq_gather_offset_z(const int16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_s32)))
+int32x4_t vldrhq_gather_offset_z_s32(const int16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_s32)))
+int32x4_t vldrhq_gather_offset_z(const int16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_u16)))
+uint16x8_t vldrhq_gather_offset_z_u16(const uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_u16)))
+uint16x8_t vldrhq_gather_offset_z(const uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_u32)))
+uint32x4_t vldrhq_gather_offset_z_u32(const uint16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_u32)))
+uint32x4_t vldrhq_gather_offset_z(const uint16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_s16)))
+int16x8_t vldrhq_gather_shifted_offset_s16(const int16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_s16)))
+int16x8_t vldrhq_gather_shifted_offset(const int16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_s32)))
+int32x4_t vldrhq_gather_shifted_offset_s32(const int16_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_s32)))
+int32x4_t vldrhq_gather_shifted_offset(const int16_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_u16)))
+uint16x8_t vldrhq_gather_shifted_offset_u16(const uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_u16)))
+uint16x8_t vldrhq_gather_shifted_offset(const uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_u32)))
+uint32x4_t vldrhq_gather_shifted_offset_u32(const uint16_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_u32)))
+uint32x4_t vldrhq_gather_shifted_offset(const uint16_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_s16)))
+int16x8_t vldrhq_gather_shifted_offset_z_s16(const int16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_s16)))
+int16x8_t vldrhq_gather_shifted_offset_z(const int16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_s32)))
+int32x4_t vldrhq_gather_shifted_offset_z_s32(const int16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_s32)))
+int32x4_t vldrhq_gather_shifted_offset_z(const int16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_u16)))
+uint16x8_t vldrhq_gather_shifted_offset_z_u16(const uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_u16)))
+uint16x8_t vldrhq_gather_shifted_offset_z(const uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_u32)))
+uint32x4_t vldrhq_gather_shifted_offset_z_u32(const uint16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_u32)))
+uint32x4_t vldrhq_gather_shifted_offset_z(const uint16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_s16)))
+int16x8_t vldrhq_s16(const int16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_s32)))
+int32x4_t vldrhq_s32(const int16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_u16)))
+uint16x8_t vldrhq_u16(const uint16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_u32)))
+uint32x4_t vldrhq_u32(const uint16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_z_s16)))
+int16x8_t vldrhq_z_s16(const int16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_z_s32)))
+int32x4_t vldrhq_z_s32(const int16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_z_u16)))
+uint16x8_t vldrhq_z_u16(const uint16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_z_u32)))
+uint32x4_t vldrhq_z_u32(const uint16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_s32)))
+int32x4_t vldrwq_gather_base_s32(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_u32)))
+uint32x4_t vldrwq_gather_base_u32(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_wb_s32)))
+int32x4_t vldrwq_gather_base_wb_s32(uint32x4_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_wb_u32)))
+uint32x4_t vldrwq_gather_base_wb_u32(uint32x4_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_wb_z_s32)))
+int32x4_t vldrwq_gather_base_wb_z_s32(uint32x4_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_wb_z_u32)))
+uint32x4_t vldrwq_gather_base_wb_z_u32(uint32x4_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_z_s32)))
+int32x4_t vldrwq_gather_base_z_s32(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_z_u32)))
+uint32x4_t vldrwq_gather_base_z_u32(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_s32)))
+int32x4_t vldrwq_gather_offset_s32(const int32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_s32)))
+int32x4_t vldrwq_gather_offset(const int32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_u32)))
+uint32x4_t vldrwq_gather_offset_u32(const uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_u32)))
+uint32x4_t vldrwq_gather_offset(const uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_z_s32)))
+int32x4_t vldrwq_gather_offset_z_s32(const int32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_z_s32)))
+int32x4_t vldrwq_gather_offset_z(const int32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_z_u32)))
+uint32x4_t vldrwq_gather_offset_z_u32(const uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_z_u32)))
+uint32x4_t vldrwq_gather_offset_z(const uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_s32)))
+int32x4_t vldrwq_gather_shifted_offset_s32(const int32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_s32)))
+int32x4_t vldrwq_gather_shifted_offset(const int32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_u32)))
+uint32x4_t vldrwq_gather_shifted_offset_u32(const uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_u32)))
+uint32x4_t vldrwq_gather_shifted_offset(const uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_z_s32)))
+int32x4_t vldrwq_gather_shifted_offset_z_s32(const int32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_z_s32)))
+int32x4_t vldrwq_gather_shifted_offset_z(const int32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_z_u32)))
+uint32x4_t vldrwq_gather_shifted_offset_z_u32(const uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_z_u32)))
+uint32x4_t vldrwq_gather_shifted_offset_z(const uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_s32)))
+int32x4_t vldrwq_s32(const int32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_u32)))
+uint32x4_t vldrwq_u32(const uint32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_z_s32)))
+int32x4_t vldrwq_z_s32(const int32_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_z_u32)))
+uint32x4_t vldrwq_z_u32(const uint32_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_m_s16)))
+uint16x8_t vmaxaq_m_s16(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_m_s16)))
+uint16x8_t vmaxaq_m(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_m_s32)))
+uint32x4_t vmaxaq_m_s32(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_m_s32)))
+uint32x4_t vmaxaq_m(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_m_s8)))
+uint8x16_t vmaxaq_m_s8(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_m_s8)))
+uint8x16_t vmaxaq_m(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_s16)))
+uint16x8_t vmaxaq_s16(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_s16)))
+uint16x8_t vmaxaq(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_s32)))
+uint32x4_t vmaxaq_s32(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_s32)))
+uint32x4_t vmaxaq(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_s8)))
+uint8x16_t vmaxaq_s8(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxaq_s8)))
+uint8x16_t vmaxaq(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_p_s16)))
+uint16_t vmaxavq_p_s16(uint16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_p_s16)))
+uint16_t vmaxavq_p(uint16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_p_s32)))
+uint32_t vmaxavq_p_s32(uint32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_p_s32)))
+uint32_t vmaxavq_p(uint32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_p_s8)))
+uint8_t vmaxavq_p_s8(uint8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_p_s8)))
+uint8_t vmaxavq_p(uint8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_s16)))
+uint16_t vmaxavq_s16(uint16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_s16)))
+uint16_t vmaxavq(uint16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_s32)))
+uint32_t vmaxavq_s32(uint32_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_s32)))
+uint32_t vmaxavq(uint32_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_s8)))
+uint8_t vmaxavq_s8(uint8_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxavq_s8)))
+uint8_t vmaxavq(uint8_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_s16)))
+int16x8_t vmaxq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_s16)))
+int16x8_t vmaxq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_s32)))
+int32x4_t vmaxq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_s32)))
+int32x4_t vmaxq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_s8)))
+int8x16_t vmaxq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_s8)))
+int8x16_t vmaxq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_u16)))
+uint16x8_t vmaxq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_u16)))
+uint16x8_t vmaxq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_u32)))
+uint32x4_t vmaxq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_u32)))
+uint32x4_t vmaxq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_u8)))
+uint8x16_t vmaxq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_m_u8)))
+uint8x16_t vmaxq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_s16)))
+int16x8_t vmaxq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_s16)))
+int16x8_t vmaxq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_s32)))
+int32x4_t vmaxq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_s32)))
+int32x4_t vmaxq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_s8)))
+int8x16_t vmaxq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_s8)))
+int8x16_t vmaxq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_u16)))
+uint16x8_t vmaxq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_u16)))
+uint16x8_t vmaxq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_u32)))
+uint32x4_t vmaxq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_u32)))
+uint32x4_t vmaxq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_u8)))
+uint8x16_t vmaxq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_u8)))
+uint8x16_t vmaxq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_s16)))
+int16x8_t vmaxq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_s16)))
+int16x8_t vmaxq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_s32)))
+int32x4_t vmaxq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_s32)))
+int32x4_t vmaxq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_s8)))
+int8x16_t vmaxq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_s8)))
+int8x16_t vmaxq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_u16)))
+uint16x8_t vmaxq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_u16)))
+uint16x8_t vmaxq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_u32)))
+uint32x4_t vmaxq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_u32)))
+uint32x4_t vmaxq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_u8)))
+uint8x16_t vmaxq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxq_x_u8)))
+uint8x16_t vmaxq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_s16)))
+int16_t vmaxvq_p_s16(int16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_s16)))
+int16_t vmaxvq_p(int16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_s32)))
+int32_t vmaxvq_p_s32(int32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_s32)))
+int32_t vmaxvq_p(int32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_s8)))
+int8_t vmaxvq_p_s8(int8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_s8)))
+int8_t vmaxvq_p(int8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_u16)))
+uint16_t vmaxvq_p_u16(uint16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_u16)))
+uint16_t vmaxvq_p(uint16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_u32)))
+uint32_t vmaxvq_p_u32(uint32_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_u32)))
+uint32_t vmaxvq_p(uint32_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_u8)))
+uint8_t vmaxvq_p_u8(uint8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_p_u8)))
+uint8_t vmaxvq_p(uint8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_s16)))
+int16_t vmaxvq_s16(int16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_s16)))
+int16_t vmaxvq(int16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_s32)))
+int32_t vmaxvq_s32(int32_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_s32)))
+int32_t vmaxvq(int32_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_s8)))
+int8_t vmaxvq_s8(int8_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_s8)))
+int8_t vmaxvq(int8_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_u16)))
+uint16_t vmaxvq_u16(uint16_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_u16)))
+uint16_t vmaxvq(uint16_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_u32)))
+uint32_t vmaxvq_u32(uint32_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_u32)))
+uint32_t vmaxvq(uint32_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_u8)))
+uint8_t vmaxvq_u8(uint8_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxvq_u8)))
+uint8_t vmaxvq(uint8_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminaq_m_s16)))
+uint16x8_t vminaq_m_s16(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminaq_m_s16)))
+uint16x8_t vminaq_m(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminaq_m_s32)))
+uint32x4_t vminaq_m_s32(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminaq_m_s32)))
+uint32x4_t vminaq_m(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminaq_m_s8)))
+uint8x16_t vminaq_m_s8(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminaq_m_s8)))
+uint8x16_t vminaq_m(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminaq_s16)))
+uint16x8_t vminaq_s16(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminaq_s16)))
+uint16x8_t vminaq(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminaq_s32)))
+uint32x4_t vminaq_s32(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminaq_s32)))
+uint32x4_t vminaq(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminaq_s8)))
+uint8x16_t vminaq_s8(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminaq_s8)))
+uint8x16_t vminaq(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminavq_p_s16)))
+uint16_t vminavq_p_s16(uint16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminavq_p_s16)))
+uint16_t vminavq_p(uint16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminavq_p_s32)))
+uint32_t vminavq_p_s32(uint32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminavq_p_s32)))
+uint32_t vminavq_p(uint32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminavq_p_s8)))
+uint8_t vminavq_p_s8(uint8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminavq_p_s8)))
+uint8_t vminavq_p(uint8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminavq_s16)))
+uint16_t vminavq_s16(uint16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminavq_s16)))
+uint16_t vminavq(uint16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminavq_s32)))
+uint32_t vminavq_s32(uint32_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminavq_s32)))
+uint32_t vminavq(uint32_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminavq_s8)))
+uint8_t vminavq_s8(uint8_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminavq_s8)))
+uint8_t vminavq(uint8_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_s16)))
+int16x8_t vminq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_s16)))
+int16x8_t vminq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_s32)))
+int32x4_t vminq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_s32)))
+int32x4_t vminq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_s8)))
+int8x16_t vminq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_s8)))
+int8x16_t vminq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_u16)))
+uint16x8_t vminq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_u16)))
+uint16x8_t vminq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_u32)))
+uint32x4_t vminq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_u32)))
+uint32x4_t vminq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_u8)))
+uint8x16_t vminq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_m_u8)))
+uint8x16_t vminq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_s16)))
+int16x8_t vminq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_s16)))
+int16x8_t vminq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_s32)))
+int32x4_t vminq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_s32)))
+int32x4_t vminq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_s8)))
+int8x16_t vminq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_s8)))
+int8x16_t vminq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_u16)))
+uint16x8_t vminq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_u16)))
+uint16x8_t vminq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_u32)))
+uint32x4_t vminq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_u32)))
+uint32x4_t vminq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_u8)))
+uint8x16_t vminq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_u8)))
+uint8x16_t vminq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_s16)))
+int16x8_t vminq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_s16)))
+int16x8_t vminq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_s32)))
+int32x4_t vminq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_s32)))
+int32x4_t vminq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_s8)))
+int8x16_t vminq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_s8)))
+int8x16_t vminq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_u16)))
+uint16x8_t vminq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_u16)))
+uint16x8_t vminq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_u32)))
+uint32x4_t vminq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_u32)))
+uint32x4_t vminq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_u8)))
+uint8x16_t vminq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminq_x_u8)))
+uint8x16_t vminq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_s16)))
+int16_t vminvq_p_s16(int16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_s16)))
+int16_t vminvq_p(int16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_s32)))
+int32_t vminvq_p_s32(int32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_s32)))
+int32_t vminvq_p(int32_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_s8)))
+int8_t vminvq_p_s8(int8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_s8)))
+int8_t vminvq_p(int8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_u16)))
+uint16_t vminvq_p_u16(uint16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_u16)))
+uint16_t vminvq_p(uint16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_u32)))
+uint32_t vminvq_p_u32(uint32_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_u32)))
+uint32_t vminvq_p(uint32_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_u8)))
+uint8_t vminvq_p_u8(uint8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_p_u8)))
+uint8_t vminvq_p(uint8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_s16)))
+int16_t vminvq_s16(int16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_s16)))
+int16_t vminvq(int16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_s32)))
+int32_t vminvq_s32(int32_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_s32)))
+int32_t vminvq(int32_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_s8)))
+int8_t vminvq_s8(int8_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_s8)))
+int8_t vminvq(int8_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_u16)))
+uint16_t vminvq_u16(uint16_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_u16)))
+uint16_t vminvq(uint16_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_u32)))
+uint32_t vminvq_u32(uint32_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_u32)))
+uint32_t vminvq(uint32_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminvq_u8)))
+uint8_t vminvq_u8(uint8_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminvq_u8)))
+uint8_t vminvq(uint8_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_s16)))
+int32_t vmladavaq_p_s16(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_s16)))
+int32_t vmladavaq_p(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_s32)))
+int32_t vmladavaq_p_s32(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_s32)))
+int32_t vmladavaq_p(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_s8)))
+int32_t vmladavaq_p_s8(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_s8)))
+int32_t vmladavaq_p(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_u16)))
+uint32_t vmladavaq_p_u16(uint32_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_u16)))
+uint32_t vmladavaq_p(uint32_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_u32)))
+uint32_t vmladavaq_p_u32(uint32_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_u32)))
+uint32_t vmladavaq_p(uint32_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_u8)))
+uint32_t vmladavaq_p_u8(uint32_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_p_u8)))
+uint32_t vmladavaq_p(uint32_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_s16)))
+int32_t vmladavaq_s16(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_s16)))
+int32_t vmladavaq(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_s32)))
+int32_t vmladavaq_s32(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_s32)))
+int32_t vmladavaq(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_s8)))
+int32_t vmladavaq_s8(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_s8)))
+int32_t vmladavaq(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_u16)))
+uint32_t vmladavaq_u16(uint32_t, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_u16)))
+uint32_t vmladavaq(uint32_t, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_u32)))
+uint32_t vmladavaq_u32(uint32_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_u32)))
+uint32_t vmladavaq(uint32_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_u8)))
+uint32_t vmladavaq_u8(uint32_t, uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaq_u8)))
+uint32_t vmladavaq(uint32_t, uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_p_s16)))
+int32_t vmladavaxq_p_s16(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_p_s16)))
+int32_t vmladavaxq_p(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_p_s32)))
+int32_t vmladavaxq_p_s32(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_p_s32)))
+int32_t vmladavaxq_p(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_p_s8)))
+int32_t vmladavaxq_p_s8(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_p_s8)))
+int32_t vmladavaxq_p(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_s16)))
+int32_t vmladavaxq_s16(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_s16)))
+int32_t vmladavaxq(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_s32)))
+int32_t vmladavaxq_s32(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_s32)))
+int32_t vmladavaxq(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_s8)))
+int32_t vmladavaxq_s8(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavaxq_s8)))
+int32_t vmladavaxq(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_s16)))
+int32_t vmladavq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_s16)))
+int32_t vmladavq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_s32)))
+int32_t vmladavq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_s32)))
+int32_t vmladavq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_s8)))
+int32_t vmladavq_p_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_s8)))
+int32_t vmladavq_p(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_u16)))
+uint32_t vmladavq_p_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_u16)))
+uint32_t vmladavq_p(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_u32)))
+uint32_t vmladavq_p_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_u32)))
+uint32_t vmladavq_p(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_u8)))
+uint32_t vmladavq_p_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_p_u8)))
+uint32_t vmladavq_p(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_s16)))
+int32_t vmladavq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_s16)))
+int32_t vmladavq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_s32)))
+int32_t vmladavq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_s32)))
+int32_t vmladavq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_s8)))
+int32_t vmladavq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_s8)))
+int32_t vmladavq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_u16)))
+uint32_t vmladavq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_u16)))
+uint32_t vmladavq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_u32)))
+uint32_t vmladavq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_u32)))
+uint32_t vmladavq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_u8)))
+uint32_t vmladavq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavq_u8)))
+uint32_t vmladavq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_p_s16)))
+int32_t vmladavxq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_p_s16)))
+int32_t vmladavxq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_p_s32)))
+int32_t vmladavxq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_p_s32)))
+int32_t vmladavxq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_p_s8)))
+int32_t vmladavxq_p_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_p_s8)))
+int32_t vmladavxq_p(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_s16)))
+int32_t vmladavxq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_s16)))
+int32_t vmladavxq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_s32)))
+int32_t vmladavxq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_s32)))
+int32_t vmladavxq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_s8)))
+int32_t vmladavxq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmladavxq_s8)))
+int32_t vmladavxq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_s16)))
+int64_t vmlaldavaq_p_s16(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_s16)))
+int64_t vmlaldavaq_p(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_s32)))
+int64_t vmlaldavaq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_s32)))
+int64_t vmlaldavaq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_u16)))
+uint64_t vmlaldavaq_p_u16(uint64_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_u16)))
+uint64_t vmlaldavaq_p(uint64_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_u32)))
+uint64_t vmlaldavaq_p_u32(uint64_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_p_u32)))
+uint64_t vmlaldavaq_p(uint64_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_s16)))
+int64_t vmlaldavaq_s16(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_s16)))
+int64_t vmlaldavaq(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_s32)))
+int64_t vmlaldavaq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_s32)))
+int64_t vmlaldavaq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_u16)))
+uint64_t vmlaldavaq_u16(uint64_t, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_u16)))
+uint64_t vmlaldavaq(uint64_t, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_u32)))
+uint64_t vmlaldavaq_u32(uint64_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaq_u32)))
+uint64_t vmlaldavaq(uint64_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_p_s16)))
+int64_t vmlaldavaxq_p_s16(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_p_s16)))
+int64_t vmlaldavaxq_p(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_p_s32)))
+int64_t vmlaldavaxq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_p_s32)))
+int64_t vmlaldavaxq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_s16)))
+int64_t vmlaldavaxq_s16(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_s16)))
+int64_t vmlaldavaxq(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_s32)))
+int64_t vmlaldavaxq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavaxq_s32)))
+int64_t vmlaldavaxq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_s16)))
+int64_t vmlaldavq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_s16)))
+int64_t vmlaldavq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_s32)))
+int64_t vmlaldavq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_s32)))
+int64_t vmlaldavq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_u16)))
+uint64_t vmlaldavq_p_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_u16)))
+uint64_t vmlaldavq_p(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_u32)))
+uint64_t vmlaldavq_p_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_p_u32)))
+uint64_t vmlaldavq_p(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_s16)))
+int64_t vmlaldavq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_s16)))
+int64_t vmlaldavq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_s32)))
+int64_t vmlaldavq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_s32)))
+int64_t vmlaldavq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_u16)))
+uint64_t vmlaldavq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_u16)))
+uint64_t vmlaldavq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_u32)))
+uint64_t vmlaldavq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavq_u32)))
+uint64_t vmlaldavq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_p_s16)))
+int64_t vmlaldavxq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_p_s16)))
+int64_t vmlaldavxq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_p_s32)))
+int64_t vmlaldavxq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_p_s32)))
+int64_t vmlaldavxq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_s16)))
+int64_t vmlaldavxq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_s16)))
+int64_t vmlaldavxq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_s32)))
+int64_t vmlaldavxq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaldavxq_s32)))
+int64_t vmlaldavxq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_s16)))
+int16x8_t vmlaq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_s16)))
+int16x8_t vmlaq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_s32)))
+int32x4_t vmlaq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_s32)))
+int32x4_t vmlaq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_s8)))
+int8x16_t vmlaq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_s8)))
+int8x16_t vmlaq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_u16)))
+uint16x8_t vmlaq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_u16)))
+uint16x8_t vmlaq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_u32)))
+uint32x4_t vmlaq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_u32)))
+uint32x4_t vmlaq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_u8)))
+uint8x16_t vmlaq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_m_n_u8)))
+uint8x16_t vmlaq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_s16)))
+int16x8_t vmlaq_n_s16(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_s16)))
+int16x8_t vmlaq(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_s32)))
+int32x4_t vmlaq_n_s32(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_s32)))
+int32x4_t vmlaq(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_s8)))
+int8x16_t vmlaq_n_s8(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_s8)))
+int8x16_t vmlaq(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_u16)))
+uint16x8_t vmlaq_n_u16(uint16x8_t, uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_u16)))
+uint16x8_t vmlaq(uint16x8_t, uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_u32)))
+uint32x4_t vmlaq_n_u32(uint32x4_t, uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_u32)))
+uint32x4_t vmlaq(uint32x4_t, uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_u8)))
+uint8x16_t vmlaq_n_u8(uint8x16_t, uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlaq_n_u8)))
+uint8x16_t vmlaq(uint8x16_t, uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_s16)))
+int16x8_t vmlasq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_s16)))
+int16x8_t vmlasq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_s32)))
+int32x4_t vmlasq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_s32)))
+int32x4_t vmlasq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_s8)))
+int8x16_t vmlasq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_s8)))
+int8x16_t vmlasq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_u16)))
+uint16x8_t vmlasq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_u16)))
+uint16x8_t vmlasq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_u32)))
+uint32x4_t vmlasq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_u32)))
+uint32x4_t vmlasq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_u8)))
+uint8x16_t vmlasq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_m_n_u8)))
+uint8x16_t vmlasq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_s16)))
+int16x8_t vmlasq_n_s16(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_s16)))
+int16x8_t vmlasq(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_s32)))
+int32x4_t vmlasq_n_s32(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_s32)))
+int32x4_t vmlasq(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_s8)))
+int8x16_t vmlasq_n_s8(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_s8)))
+int8x16_t vmlasq(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_u16)))
+uint16x8_t vmlasq_n_u16(uint16x8_t, uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_u16)))
+uint16x8_t vmlasq(uint16x8_t, uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_u32)))
+uint32x4_t vmlasq_n_u32(uint32x4_t, uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_u32)))
+uint32x4_t vmlasq(uint32x4_t, uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_u8)))
+uint8x16_t vmlasq_n_u8(uint8x16_t, uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlasq_n_u8)))
+uint8x16_t vmlasq(uint8x16_t, uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_p_s16)))
+int32_t vmlsdavaq_p_s16(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_p_s16)))
+int32_t vmlsdavaq_p(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_p_s32)))
+int32_t vmlsdavaq_p_s32(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_p_s32)))
+int32_t vmlsdavaq_p(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_p_s8)))
+int32_t vmlsdavaq_p_s8(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_p_s8)))
+int32_t vmlsdavaq_p(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_s16)))
+int32_t vmlsdavaq_s16(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_s16)))
+int32_t vmlsdavaq(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_s32)))
+int32_t vmlsdavaq_s32(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_s32)))
+int32_t vmlsdavaq(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_s8)))
+int32_t vmlsdavaq_s8(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaq_s8)))
+int32_t vmlsdavaq(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_p_s16)))
+int32_t vmlsdavaxq_p_s16(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_p_s16)))
+int32_t vmlsdavaxq_p(int32_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_p_s32)))
+int32_t vmlsdavaxq_p_s32(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_p_s32)))
+int32_t vmlsdavaxq_p(int32_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_p_s8)))
+int32_t vmlsdavaxq_p_s8(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_p_s8)))
+int32_t vmlsdavaxq_p(int32_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_s16)))
+int32_t vmlsdavaxq_s16(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_s16)))
+int32_t vmlsdavaxq(int32_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_s32)))
+int32_t vmlsdavaxq_s32(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_s32)))
+int32_t vmlsdavaxq(int32_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_s8)))
+int32_t vmlsdavaxq_s8(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavaxq_s8)))
+int32_t vmlsdavaxq(int32_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_p_s16)))
+int32_t vmlsdavq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_p_s16)))
+int32_t vmlsdavq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_p_s32)))
+int32_t vmlsdavq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_p_s32)))
+int32_t vmlsdavq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_p_s8)))
+int32_t vmlsdavq_p_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_p_s8)))
+int32_t vmlsdavq_p(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_s16)))
+int32_t vmlsdavq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_s16)))
+int32_t vmlsdavq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_s32)))
+int32_t vmlsdavq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_s32)))
+int32_t vmlsdavq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_s8)))
+int32_t vmlsdavq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavq_s8)))
+int32_t vmlsdavq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_p_s16)))
+int32_t vmlsdavxq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_p_s16)))
+int32_t vmlsdavxq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_p_s32)))
+int32_t vmlsdavxq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_p_s32)))
+int32_t vmlsdavxq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_p_s8)))
+int32_t vmlsdavxq_p_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_p_s8)))
+int32_t vmlsdavxq_p(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_s16)))
+int32_t vmlsdavxq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_s16)))
+int32_t vmlsdavxq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_s32)))
+int32_t vmlsdavxq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_s32)))
+int32_t vmlsdavxq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_s8)))
+int32_t vmlsdavxq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsdavxq_s8)))
+int32_t vmlsdavxq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_p_s16)))
+int64_t vmlsldavaq_p_s16(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_p_s16)))
+int64_t vmlsldavaq_p(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_p_s32)))
+int64_t vmlsldavaq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_p_s32)))
+int64_t vmlsldavaq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_s16)))
+int64_t vmlsldavaq_s16(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_s16)))
+int64_t vmlsldavaq(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_s32)))
+int64_t vmlsldavaq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaq_s32)))
+int64_t vmlsldavaq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_p_s16)))
+int64_t vmlsldavaxq_p_s16(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_p_s16)))
+int64_t vmlsldavaxq_p(int64_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_p_s32)))
+int64_t vmlsldavaxq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_p_s32)))
+int64_t vmlsldavaxq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_s16)))
+int64_t vmlsldavaxq_s16(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_s16)))
+int64_t vmlsldavaxq(int64_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_s32)))
+int64_t vmlsldavaxq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavaxq_s32)))
+int64_t vmlsldavaxq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_p_s16)))
+int64_t vmlsldavq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_p_s16)))
+int64_t vmlsldavq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_p_s32)))
+int64_t vmlsldavq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_p_s32)))
+int64_t vmlsldavq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_s16)))
+int64_t vmlsldavq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_s16)))
+int64_t vmlsldavq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_s32)))
+int64_t vmlsldavq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavq_s32)))
+int64_t vmlsldavq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_p_s16)))
+int64_t vmlsldavxq_p_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_p_s16)))
+int64_t vmlsldavxq_p(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_p_s32)))
+int64_t vmlsldavxq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_p_s32)))
+int64_t vmlsldavxq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_s16)))
+int64_t vmlsldavxq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_s16)))
+int64_t vmlsldavxq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_s32)))
+int64_t vmlsldavxq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmlsldavxq_s32)))
+int64_t vmlsldavxq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_s16)))
+int32x4_t vmovlbq_m_s16(int32x4_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_s16)))
+int32x4_t vmovlbq_m(int32x4_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_s8)))
+int16x8_t vmovlbq_m_s8(int16x8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_s8)))
+int16x8_t vmovlbq_m(int16x8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_u16)))
+uint32x4_t vmovlbq_m_u16(uint32x4_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_u16)))
+uint32x4_t vmovlbq_m(uint32x4_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_u8)))
+uint16x8_t vmovlbq_m_u8(uint16x8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_m_u8)))
+uint16x8_t vmovlbq_m(uint16x8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_s16)))
+int32x4_t vmovlbq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_s16)))
+int32x4_t vmovlbq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_s8)))
+int16x8_t vmovlbq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_s8)))
+int16x8_t vmovlbq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_u16)))
+uint32x4_t vmovlbq_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_u16)))
+uint32x4_t vmovlbq(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_u8)))
+uint16x8_t vmovlbq_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_u8)))
+uint16x8_t vmovlbq(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_s16)))
+int32x4_t vmovlbq_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_s16)))
+int32x4_t vmovlbq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_s8)))
+int16x8_t vmovlbq_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_s8)))
+int16x8_t vmovlbq_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_u16)))
+uint32x4_t vmovlbq_x_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_u16)))
+uint32x4_t vmovlbq_x(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_u8)))
+uint16x8_t vmovlbq_x_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovlbq_x_u8)))
+uint16x8_t vmovlbq_x(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_s16)))
+int32x4_t vmovltq_m_s16(int32x4_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_s16)))
+int32x4_t vmovltq_m(int32x4_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_s8)))
+int16x8_t vmovltq_m_s8(int16x8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_s8)))
+int16x8_t vmovltq_m(int16x8_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_u16)))
+uint32x4_t vmovltq_m_u16(uint32x4_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_u16)))
+uint32x4_t vmovltq_m(uint32x4_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_u8)))
+uint16x8_t vmovltq_m_u8(uint16x8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_m_u8)))
+uint16x8_t vmovltq_m(uint16x8_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_s16)))
+int32x4_t vmovltq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_s16)))
+int32x4_t vmovltq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_s8)))
+int16x8_t vmovltq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_s8)))
+int16x8_t vmovltq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_u16)))
+uint32x4_t vmovltq_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_u16)))
+uint32x4_t vmovltq(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_u8)))
+uint16x8_t vmovltq_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_u8)))
+uint16x8_t vmovltq(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_s16)))
+int32x4_t vmovltq_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_s16)))
+int32x4_t vmovltq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_s8)))
+int16x8_t vmovltq_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_s8)))
+int16x8_t vmovltq_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_u16)))
+uint32x4_t vmovltq_x_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_u16)))
+uint32x4_t vmovltq_x(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_u8)))
+uint16x8_t vmovltq_x_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovltq_x_u8)))
+uint16x8_t vmovltq_x(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_s16)))
+int8x16_t vmovnbq_m_s16(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_s16)))
+int8x16_t vmovnbq_m(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_s32)))
+int16x8_t vmovnbq_m_s32(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_s32)))
+int16x8_t vmovnbq_m(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_u16)))
+uint8x16_t vmovnbq_m_u16(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_u16)))
+uint8x16_t vmovnbq_m(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_u32)))
+uint16x8_t vmovnbq_m_u32(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_m_u32)))
+uint16x8_t vmovnbq_m(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_s16)))
+int8x16_t vmovnbq_s16(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_s16)))
+int8x16_t vmovnbq(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_s32)))
+int16x8_t vmovnbq_s32(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_s32)))
+int16x8_t vmovnbq(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_u16)))
+uint8x16_t vmovnbq_u16(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_u16)))
+uint8x16_t vmovnbq(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_u32)))
+uint16x8_t vmovnbq_u32(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovnbq_u32)))
+uint16x8_t vmovnbq(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_s16)))
+int8x16_t vmovntq_m_s16(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_s16)))
+int8x16_t vmovntq_m(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_s32)))
+int16x8_t vmovntq_m_s32(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_s32)))
+int16x8_t vmovntq_m(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_u16)))
+uint8x16_t vmovntq_m_u16(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_u16)))
+uint8x16_t vmovntq_m(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_u32)))
+uint16x8_t vmovntq_m_u32(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_m_u32)))
+uint16x8_t vmovntq_m(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_s16)))
+int8x16_t vmovntq_s16(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_s16)))
+int8x16_t vmovntq(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_s32)))
+int16x8_t vmovntq_s32(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_s32)))
+int16x8_t vmovntq(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_u16)))
+uint8x16_t vmovntq_u16(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_u16)))
+uint8x16_t vmovntq(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_u32)))
+uint16x8_t vmovntq_u32(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmovntq_u32)))
+uint16x8_t vmovntq(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_s16)))
+int16x8_t vmulhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_s16)))
+int16x8_t vmulhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_s32)))
+int32x4_t vmulhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_s32)))
+int32x4_t vmulhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_s8)))
+int8x16_t vmulhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_s8)))
+int8x16_t vmulhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_u16)))
+uint16x8_t vmulhq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_u16)))
+uint16x8_t vmulhq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_u32)))
+uint32x4_t vmulhq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_u32)))
+uint32x4_t vmulhq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_u8)))
+uint8x16_t vmulhq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_m_u8)))
+uint8x16_t vmulhq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_s16)))
+int16x8_t vmulhq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_s16)))
+int16x8_t vmulhq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_s32)))
+int32x4_t vmulhq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_s32)))
+int32x4_t vmulhq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_s8)))
+int8x16_t vmulhq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_s8)))
+int8x16_t vmulhq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_u16)))
+uint16x8_t vmulhq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_u16)))
+uint16x8_t vmulhq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_u32)))
+uint32x4_t vmulhq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_u32)))
+uint32x4_t vmulhq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_u8)))
+uint8x16_t vmulhq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_u8)))
+uint8x16_t vmulhq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_s16)))
+int16x8_t vmulhq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_s16)))
+int16x8_t vmulhq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_s32)))
+int32x4_t vmulhq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_s32)))
+int32x4_t vmulhq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_s8)))
+int8x16_t vmulhq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_s8)))
+int8x16_t vmulhq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_u16)))
+uint16x8_t vmulhq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_u16)))
+uint16x8_t vmulhq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_u32)))
+uint32x4_t vmulhq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_u32)))
+uint32x4_t vmulhq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_u8)))
+uint8x16_t vmulhq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulhq_x_u8)))
+uint8x16_t vmulhq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_s16)))
+int32x4_t vmullbq_int_m_s16(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_s16)))
+int32x4_t vmullbq_int_m(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_s32)))
+int64x2_t vmullbq_int_m_s32(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_s32)))
+int64x2_t vmullbq_int_m(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_s8)))
+int16x8_t vmullbq_int_m_s8(int16x8_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_s8)))
+int16x8_t vmullbq_int_m(int16x8_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_u16)))
+uint32x4_t vmullbq_int_m_u16(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_u16)))
+uint32x4_t vmullbq_int_m(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_u32)))
+uint64x2_t vmullbq_int_m_u32(uint64x2_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_u32)))
+uint64x2_t vmullbq_int_m(uint64x2_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_u8)))
+uint16x8_t vmullbq_int_m_u8(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_m_u8)))
+uint16x8_t vmullbq_int_m(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_s16)))
+int32x4_t vmullbq_int_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_s16)))
+int32x4_t vmullbq_int(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_s32)))
+int64x2_t vmullbq_int_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_s32)))
+int64x2_t vmullbq_int(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_s8)))
+int16x8_t vmullbq_int_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_s8)))
+int16x8_t vmullbq_int(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_u16)))
+uint32x4_t vmullbq_int_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_u16)))
+uint32x4_t vmullbq_int(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_u32)))
+uint64x2_t vmullbq_int_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_u32)))
+uint64x2_t vmullbq_int(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_u8)))
+uint16x8_t vmullbq_int_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_u8)))
+uint16x8_t vmullbq_int(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_s16)))
+int32x4_t vmullbq_int_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_s16)))
+int32x4_t vmullbq_int_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_s32)))
+int64x2_t vmullbq_int_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_s32)))
+int64x2_t vmullbq_int_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_s8)))
+int16x8_t vmullbq_int_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_s8)))
+int16x8_t vmullbq_int_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_u16)))
+uint32x4_t vmullbq_int_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_u16)))
+uint32x4_t vmullbq_int_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_u32)))
+uint64x2_t vmullbq_int_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_u32)))
+uint64x2_t vmullbq_int_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_u8)))
+uint16x8_t vmullbq_int_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_int_x_u8)))
+uint16x8_t vmullbq_int_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_m_p16)))
+uint32x4_t vmullbq_poly_m_p16(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_m_p16)))
+uint32x4_t vmullbq_poly_m(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_m_p8)))
+uint16x8_t vmullbq_poly_m_p8(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_m_p8)))
+uint16x8_t vmullbq_poly_m(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_p16)))
+uint32x4_t vmullbq_poly_p16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_p16)))
+uint32x4_t vmullbq_poly(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_p8)))
+uint16x8_t vmullbq_poly_p8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_p8)))
+uint16x8_t vmullbq_poly(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_x_p16)))
+uint32x4_t vmullbq_poly_x_p16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_x_p16)))
+uint32x4_t vmullbq_poly_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_x_p8)))
+uint16x8_t vmullbq_poly_x_p8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmullbq_poly_x_p8)))
+uint16x8_t vmullbq_poly_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_s16)))
+int32x4_t vmulltq_int_m_s16(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_s16)))
+int32x4_t vmulltq_int_m(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_s32)))
+int64x2_t vmulltq_int_m_s32(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_s32)))
+int64x2_t vmulltq_int_m(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_s8)))
+int16x8_t vmulltq_int_m_s8(int16x8_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_s8)))
+int16x8_t vmulltq_int_m(int16x8_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_u16)))
+uint32x4_t vmulltq_int_m_u16(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_u16)))
+uint32x4_t vmulltq_int_m(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_u32)))
+uint64x2_t vmulltq_int_m_u32(uint64x2_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_u32)))
+uint64x2_t vmulltq_int_m(uint64x2_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_u8)))
+uint16x8_t vmulltq_int_m_u8(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_m_u8)))
+uint16x8_t vmulltq_int_m(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_s16)))
+int32x4_t vmulltq_int_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_s16)))
+int32x4_t vmulltq_int(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_s32)))
+int64x2_t vmulltq_int_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_s32)))
+int64x2_t vmulltq_int(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_s8)))
+int16x8_t vmulltq_int_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_s8)))
+int16x8_t vmulltq_int(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_u16)))
+uint32x4_t vmulltq_int_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_u16)))
+uint32x4_t vmulltq_int(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_u32)))
+uint64x2_t vmulltq_int_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_u32)))
+uint64x2_t vmulltq_int(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_u8)))
+uint16x8_t vmulltq_int_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_u8)))
+uint16x8_t vmulltq_int(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_s16)))
+int32x4_t vmulltq_int_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_s16)))
+int32x4_t vmulltq_int_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_s32)))
+int64x2_t vmulltq_int_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_s32)))
+int64x2_t vmulltq_int_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_s8)))
+int16x8_t vmulltq_int_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_s8)))
+int16x8_t vmulltq_int_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_u16)))
+uint32x4_t vmulltq_int_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_u16)))
+uint32x4_t vmulltq_int_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_u32)))
+uint64x2_t vmulltq_int_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_u32)))
+uint64x2_t vmulltq_int_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_u8)))
+uint16x8_t vmulltq_int_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_int_x_u8)))
+uint16x8_t vmulltq_int_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_m_p16)))
+uint32x4_t vmulltq_poly_m_p16(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_m_p16)))
+uint32x4_t vmulltq_poly_m(uint32x4_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_m_p8)))
+uint16x8_t vmulltq_poly_m_p8(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_m_p8)))
+uint16x8_t vmulltq_poly_m(uint16x8_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_p16)))
+uint32x4_t vmulltq_poly_p16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_p16)))
+uint32x4_t vmulltq_poly(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_p8)))
+uint16x8_t vmulltq_poly_p8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_p8)))
+uint16x8_t vmulltq_poly(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_x_p16)))
+uint32x4_t vmulltq_poly_x_p16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_x_p16)))
+uint32x4_t vmulltq_poly_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_x_p8)))
+uint16x8_t vmulltq_poly_x_p8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulltq_poly_x_p8)))
+uint16x8_t vmulltq_poly_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_s16)))
+int16x8_t vmulq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_s16)))
+int16x8_t vmulq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_s32)))
+int32x4_t vmulq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_s32)))
+int32x4_t vmulq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_s8)))
+int8x16_t vmulq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_s8)))
+int8x16_t vmulq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_u16)))
+uint16x8_t vmulq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_u16)))
+uint16x8_t vmulq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_u32)))
+uint32x4_t vmulq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_u32)))
+uint32x4_t vmulq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_u8)))
+uint8x16_t vmulq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_u8)))
+uint8x16_t vmulq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_s16)))
+int16x8_t vmulq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_s16)))
+int16x8_t vmulq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_s32)))
+int32x4_t vmulq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_s32)))
+int32x4_t vmulq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_s8)))
+int8x16_t vmulq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_s8)))
+int8x16_t vmulq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_u16)))
+uint16x8_t vmulq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_u16)))
+uint16x8_t vmulq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_u32)))
+uint32x4_t vmulq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_u32)))
+uint32x4_t vmulq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_u8)))
+uint8x16_t vmulq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_u8)))
+uint8x16_t vmulq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_s16)))
+int16x8_t vmulq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_s16)))
+int16x8_t vmulq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_s32)))
+int32x4_t vmulq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_s32)))
+int32x4_t vmulq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_s8)))
+int8x16_t vmulq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_s8)))
+int8x16_t vmulq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_u16)))
+uint16x8_t vmulq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_u16)))
+uint16x8_t vmulq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_u32)))
+uint32x4_t vmulq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_u32)))
+uint32x4_t vmulq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_u8)))
+uint8x16_t vmulq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_u8)))
+uint8x16_t vmulq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_s16)))
+int16x8_t vmulq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_s16)))
+int16x8_t vmulq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_s32)))
+int32x4_t vmulq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_s32)))
+int32x4_t vmulq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_s8)))
+int8x16_t vmulq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_s8)))
+int8x16_t vmulq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_u16)))
+uint16x8_t vmulq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_u16)))
+uint16x8_t vmulq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_u32)))
+uint32x4_t vmulq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_u32)))
+uint32x4_t vmulq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_u8)))
+uint8x16_t vmulq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_u8)))
+uint8x16_t vmulq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_s16)))
+int16x8_t vmulq_x_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_s16)))
+int16x8_t vmulq_x(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_s32)))
+int32x4_t vmulq_x_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_s32)))
+int32x4_t vmulq_x(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_s8)))
+int8x16_t vmulq_x_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_s8)))
+int8x16_t vmulq_x(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_u16)))
+uint16x8_t vmulq_x_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_u16)))
+uint16x8_t vmulq_x(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_u32)))
+uint32x4_t vmulq_x_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_u32)))
+uint32x4_t vmulq_x(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_u8)))
+uint8x16_t vmulq_x_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_u8)))
+uint8x16_t vmulq_x(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_s16)))
+int16x8_t vmulq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_s16)))
+int16x8_t vmulq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_s32)))
+int32x4_t vmulq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_s32)))
+int32x4_t vmulq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_s8)))
+int8x16_t vmulq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_s8)))
+int8x16_t vmulq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_u16)))
+uint16x8_t vmulq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_u16)))
+uint16x8_t vmulq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_u32)))
+uint32x4_t vmulq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_u32)))
+uint32x4_t vmulq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_u8)))
+uint8x16_t vmulq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_u8)))
+uint8x16_t vmulq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_s16)))
+int16x8_t vmvnq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_s16)))
+int16x8_t vmvnq_m(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_s32)))
+int32x4_t vmvnq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_s32)))
+int32x4_t vmvnq_m(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_u16)))
+uint16x8_t vmvnq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_u16)))
+uint16x8_t vmvnq_m(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_u32)))
+uint32x4_t vmvnq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_n_u32)))
+uint32x4_t vmvnq_m(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_s16)))
+int16x8_t vmvnq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_s16)))
+int16x8_t vmvnq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_s32)))
+int32x4_t vmvnq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_s32)))
+int32x4_t vmvnq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_s8)))
+int8x16_t vmvnq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_s8)))
+int8x16_t vmvnq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_u16)))
+uint16x8_t vmvnq_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_u16)))
+uint16x8_t vmvnq_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_u32)))
+uint32x4_t vmvnq_m_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_u32)))
+uint32x4_t vmvnq_m(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_u8)))
+uint8x16_t vmvnq_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_m_u8)))
+uint8x16_t vmvnq_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_n_s16)))
+int16x8_t vmvnq_n_s16(int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_n_s32)))
+int32x4_t vmvnq_n_s32(int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_n_u16)))
+uint16x8_t vmvnq_n_u16(uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_n_u32)))
+uint32x4_t vmvnq_n_u32(uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_s16)))
+int16x8_t vmvnq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_s16)))
+int16x8_t vmvnq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_s32)))
+int32x4_t vmvnq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_s32)))
+int32x4_t vmvnq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_s8)))
+int8x16_t vmvnq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_s8)))
+int8x16_t vmvnq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_u16)))
+uint16x8_t vmvnq_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_u16)))
+uint16x8_t vmvnq(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_u32)))
+uint32x4_t vmvnq_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_u32)))
+uint32x4_t vmvnq(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_u8)))
+uint8x16_t vmvnq_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_u8)))
+uint8x16_t vmvnq(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_n_s16)))
+int16x8_t vmvnq_x_n_s16(int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_n_s32)))
+int32x4_t vmvnq_x_n_s32(int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_n_u16)))
+uint16x8_t vmvnq_x_n_u16(uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_n_u32)))
+uint32x4_t vmvnq_x_n_u32(uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_s16)))
+int16x8_t vmvnq_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_s16)))
+int16x8_t vmvnq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_s32)))
+int32x4_t vmvnq_x_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_s32)))
+int32x4_t vmvnq_x(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_s8)))
+int8x16_t vmvnq_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_s8)))
+int8x16_t vmvnq_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_u16)))
+uint16x8_t vmvnq_x_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_u16)))
+uint16x8_t vmvnq_x(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_u32)))
+uint32x4_t vmvnq_x_u32(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_u32)))
+uint32x4_t vmvnq_x(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_u8)))
+uint8x16_t vmvnq_x_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmvnq_x_u8)))
+uint8x16_t vmvnq_x(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_s16)))
+int16x8_t vnegq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_s16)))
+int16x8_t vnegq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_s32)))
+int32x4_t vnegq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_s32)))
+int32x4_t vnegq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_s8)))
+int8x16_t vnegq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_s8)))
+int8x16_t vnegq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_s16)))
+int16x8_t vnegq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_s16)))
+int16x8_t vnegq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_s32)))
+int32x4_t vnegq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_s32)))
+int32x4_t vnegq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_s8)))
+int8x16_t vnegq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_s8)))
+int8x16_t vnegq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_s16)))
+int16x8_t vnegq_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_s16)))
+int16x8_t vnegq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_s32)))
+int32x4_t vnegq_x_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_s32)))
+int32x4_t vnegq_x(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_s8)))
+int8x16_t vnegq_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_s8)))
+int8x16_t vnegq_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_s16)))
+int16x8_t vornq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_s16)))
+int16x8_t vornq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_s32)))
+int32x4_t vornq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_s32)))
+int32x4_t vornq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_s8)))
+int8x16_t vornq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_s8)))
+int8x16_t vornq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_u16)))
+uint16x8_t vornq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_u16)))
+uint16x8_t vornq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_u32)))
+uint32x4_t vornq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_u32)))
+uint32x4_t vornq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_u8)))
+uint8x16_t vornq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_u8)))
+uint8x16_t vornq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_s16)))
+int16x8_t vornq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_s16)))
+int16x8_t vornq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_s32)))
+int32x4_t vornq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_s32)))
+int32x4_t vornq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_s8)))
+int8x16_t vornq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_s8)))
+int8x16_t vornq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_u16)))
+uint16x8_t vornq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_u16)))
+uint16x8_t vornq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_u32)))
+uint32x4_t vornq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_u32)))
+uint32x4_t vornq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_u8)))
+uint8x16_t vornq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_u8)))
+uint8x16_t vornq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_s16)))
+int16x8_t vornq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_s16)))
+int16x8_t vornq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_s32)))
+int32x4_t vornq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_s32)))
+int32x4_t vornq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_s8)))
+int8x16_t vornq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_s8)))
+int8x16_t vornq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_u16)))
+uint16x8_t vornq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_u16)))
+uint16x8_t vornq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_u32)))
+uint32x4_t vornq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_u32)))
+uint32x4_t vornq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_u8)))
+uint8x16_t vornq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_u8)))
+uint8x16_t vornq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_s16)))
+int16x8_t vorrq_m_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_s16)))
+int16x8_t vorrq_m_n(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_s32)))
+int32x4_t vorrq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_s32)))
+int32x4_t vorrq_m_n(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_u16)))
+uint16x8_t vorrq_m_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_u16)))
+uint16x8_t vorrq_m_n(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_u32)))
+uint32x4_t vorrq_m_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_n_u32)))
+uint32x4_t vorrq_m_n(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_s16)))
+int16x8_t vorrq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_s16)))
+int16x8_t vorrq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_s32)))
+int32x4_t vorrq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_s32)))
+int32x4_t vorrq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_s8)))
+int8x16_t vorrq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_s8)))
+int8x16_t vorrq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_u16)))
+uint16x8_t vorrq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_u16)))
+uint16x8_t vorrq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_u32)))
+uint32x4_t vorrq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_u32)))
+uint32x4_t vorrq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_u8)))
+uint8x16_t vorrq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_u8)))
+uint8x16_t vorrq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_s16)))
+int16x8_t vorrq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_s16)))
+int16x8_t vorrq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_s32)))
+int32x4_t vorrq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_s32)))
+int32x4_t vorrq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_u16)))
+uint16x8_t vorrq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_u16)))
+uint16x8_t vorrq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_u32)))
+uint32x4_t vorrq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_n_u32)))
+uint32x4_t vorrq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_s16)))
+int16x8_t vorrq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_s16)))
+int16x8_t vorrq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_s32)))
+int32x4_t vorrq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_s32)))
+int32x4_t vorrq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_s8)))
+int8x16_t vorrq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_s8)))
+int8x16_t vorrq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_u16)))
+uint16x8_t vorrq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_u16)))
+uint16x8_t vorrq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_u32)))
+uint32x4_t vorrq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_u32)))
+uint32x4_t vorrq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_u8)))
+uint8x16_t vorrq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_u8)))
+uint8x16_t vorrq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_s16)))
+int16x8_t vorrq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_s16)))
+int16x8_t vorrq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_s32)))
+int32x4_t vorrq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_s32)))
+int32x4_t vorrq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_s8)))
+int8x16_t vorrq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_s8)))
+int8x16_t vorrq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_u16)))
+uint16x8_t vorrq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_u16)))
+uint16x8_t vorrq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_u32)))
+uint32x4_t vorrq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_u32)))
+uint32x4_t vorrq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_u8)))
+uint8x16_t vorrq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_u8)))
+uint8x16_t vorrq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpnot)))
+mve_pred16_t vpnot(mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s16)))
+int16x8_t vpselq_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s16)))
+int16x8_t vpselq(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s32)))
+int32x4_t vpselq_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s32)))
+int32x4_t vpselq(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s64)))
+int64x2_t vpselq_s64(int64x2_t, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s64)))
+int64x2_t vpselq(int64x2_t, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s8)))
+int8x16_t vpselq_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_s8)))
+int8x16_t vpselq(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u16)))
+uint16x8_t vpselq_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u16)))
+uint16x8_t vpselq(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u32)))
+uint32x4_t vpselq_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u32)))
+uint32x4_t vpselq(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u64)))
+uint64x2_t vpselq_u64(uint64x2_t, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u64)))
+uint64x2_t vpselq(uint64x2_t, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u8)))
+uint8x16_t vpselq_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_u8)))
+uint8x16_t vpselq(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_m_s16)))
+int16x8_t vqabsq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_m_s16)))
+int16x8_t vqabsq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_m_s32)))
+int32x4_t vqabsq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_m_s32)))
+int32x4_t vqabsq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_m_s8)))
+int8x16_t vqabsq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_m_s8)))
+int8x16_t vqabsq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_s16)))
+int16x8_t vqabsq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_s16)))
+int16x8_t vqabsq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_s32)))
+int32x4_t vqabsq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_s32)))
+int32x4_t vqabsq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_s8)))
+int8x16_t vqabsq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqabsq_s8)))
+int8x16_t vqabsq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_s16)))
+int16x8_t vqaddq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_s16)))
+int16x8_t vqaddq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_s32)))
+int32x4_t vqaddq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_s32)))
+int32x4_t vqaddq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_s8)))
+int8x16_t vqaddq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_s8)))
+int8x16_t vqaddq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_u16)))
+uint16x8_t vqaddq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_u16)))
+uint16x8_t vqaddq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_u32)))
+uint32x4_t vqaddq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_u32)))
+uint32x4_t vqaddq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_u8)))
+uint8x16_t vqaddq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_n_u8)))
+uint8x16_t vqaddq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_s16)))
+int16x8_t vqaddq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_s16)))
+int16x8_t vqaddq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_s32)))
+int32x4_t vqaddq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_s32)))
+int32x4_t vqaddq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_s8)))
+int8x16_t vqaddq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_s8)))
+int8x16_t vqaddq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_u16)))
+uint16x8_t vqaddq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_u16)))
+uint16x8_t vqaddq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_u32)))
+uint32x4_t vqaddq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_u32)))
+uint32x4_t vqaddq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_u8)))
+uint8x16_t vqaddq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_m_u8)))
+uint8x16_t vqaddq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_s16)))
+int16x8_t vqaddq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_s16)))
+int16x8_t vqaddq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_s32)))
+int32x4_t vqaddq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_s32)))
+int32x4_t vqaddq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_s8)))
+int8x16_t vqaddq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_s8)))
+int8x16_t vqaddq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_u16)))
+uint16x8_t vqaddq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_u16)))
+uint16x8_t vqaddq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_u32)))
+uint32x4_t vqaddq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_u32)))
+uint32x4_t vqaddq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_u8)))
+uint8x16_t vqaddq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_n_u8)))
+uint8x16_t vqaddq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_s16)))
+int16x8_t vqaddq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_s16)))
+int16x8_t vqaddq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_s32)))
+int32x4_t vqaddq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_s32)))
+int32x4_t vqaddq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_s8)))
+int8x16_t vqaddq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_s8)))
+int8x16_t vqaddq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_u16)))
+uint16x8_t vqaddq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_u16)))
+uint16x8_t vqaddq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_u32)))
+uint32x4_t vqaddq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_u32)))
+uint32x4_t vqaddq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_u8)))
+uint8x16_t vqaddq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqaddq_u8)))
+uint8x16_t vqaddq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_m_s16)))
+int16x8_t vqdmladhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_m_s16)))
+int16x8_t vqdmladhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_m_s32)))
+int32x4_t vqdmladhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_m_s32)))
+int32x4_t vqdmladhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_m_s8)))
+int8x16_t vqdmladhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_m_s8)))
+int8x16_t vqdmladhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_s16)))
+int16x8_t vqdmladhq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_s16)))
+int16x8_t vqdmladhq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_s32)))
+int32x4_t vqdmladhq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_s32)))
+int32x4_t vqdmladhq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_s8)))
+int8x16_t vqdmladhq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhq_s8)))
+int8x16_t vqdmladhq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_m_s16)))
+int16x8_t vqdmladhxq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_m_s16)))
+int16x8_t vqdmladhxq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_m_s32)))
+int32x4_t vqdmladhxq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_m_s32)))
+int32x4_t vqdmladhxq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_m_s8)))
+int8x16_t vqdmladhxq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_m_s8)))
+int8x16_t vqdmladhxq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_s16)))
+int16x8_t vqdmladhxq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_s16)))
+int16x8_t vqdmladhxq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_s32)))
+int32x4_t vqdmladhxq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_s32)))
+int32x4_t vqdmladhxq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_s8)))
+int8x16_t vqdmladhxq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmladhxq_s8)))
+int8x16_t vqdmladhxq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_m_n_s16)))
+int16x8_t vqdmlahq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_m_n_s16)))
+int16x8_t vqdmlahq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_m_n_s32)))
+int32x4_t vqdmlahq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_m_n_s32)))
+int32x4_t vqdmlahq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_m_n_s8)))
+int8x16_t vqdmlahq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_m_n_s8)))
+int8x16_t vqdmlahq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_n_s16)))
+int16x8_t vqdmlahq_n_s16(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_n_s16)))
+int16x8_t vqdmlahq(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_n_s32)))
+int32x4_t vqdmlahq_n_s32(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_n_s32)))
+int32x4_t vqdmlahq(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_n_s8)))
+int8x16_t vqdmlahq_n_s8(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlahq_n_s8)))
+int8x16_t vqdmlahq(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_m_n_s16)))
+int16x8_t vqdmlashq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_m_n_s16)))
+int16x8_t vqdmlashq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_m_n_s32)))
+int32x4_t vqdmlashq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_m_n_s32)))
+int32x4_t vqdmlashq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_m_n_s8)))
+int8x16_t vqdmlashq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_m_n_s8)))
+int8x16_t vqdmlashq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_n_s16)))
+int16x8_t vqdmlashq_n_s16(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_n_s16)))
+int16x8_t vqdmlashq(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_n_s32)))
+int32x4_t vqdmlashq_n_s32(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_n_s32)))
+int32x4_t vqdmlashq(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_n_s8)))
+int8x16_t vqdmlashq_n_s8(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlashq_n_s8)))
+int8x16_t vqdmlashq(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_m_s16)))
+int16x8_t vqdmlsdhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_m_s16)))
+int16x8_t vqdmlsdhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_m_s32)))
+int32x4_t vqdmlsdhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_m_s32)))
+int32x4_t vqdmlsdhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_m_s8)))
+int8x16_t vqdmlsdhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_m_s8)))
+int8x16_t vqdmlsdhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_s16)))
+int16x8_t vqdmlsdhq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_s16)))
+int16x8_t vqdmlsdhq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_s32)))
+int32x4_t vqdmlsdhq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_s32)))
+int32x4_t vqdmlsdhq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_s8)))
+int8x16_t vqdmlsdhq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhq_s8)))
+int8x16_t vqdmlsdhq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_m_s16)))
+int16x8_t vqdmlsdhxq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_m_s16)))
+int16x8_t vqdmlsdhxq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_m_s32)))
+int32x4_t vqdmlsdhxq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_m_s32)))
+int32x4_t vqdmlsdhxq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_m_s8)))
+int8x16_t vqdmlsdhxq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_m_s8)))
+int8x16_t vqdmlsdhxq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_s16)))
+int16x8_t vqdmlsdhxq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_s16)))
+int16x8_t vqdmlsdhxq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_s32)))
+int32x4_t vqdmlsdhxq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_s32)))
+int32x4_t vqdmlsdhxq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_s8)))
+int8x16_t vqdmlsdhxq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmlsdhxq_s8)))
+int8x16_t vqdmlsdhxq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_n_s16)))
+int16x8_t vqdmulhq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_n_s16)))
+int16x8_t vqdmulhq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_n_s32)))
+int32x4_t vqdmulhq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_n_s32)))
+int32x4_t vqdmulhq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_n_s8)))
+int8x16_t vqdmulhq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_n_s8)))
+int8x16_t vqdmulhq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_s16)))
+int16x8_t vqdmulhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_s16)))
+int16x8_t vqdmulhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_s32)))
+int32x4_t vqdmulhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_s32)))
+int32x4_t vqdmulhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_s8)))
+int8x16_t vqdmulhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_m_s8)))
+int8x16_t vqdmulhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_n_s16)))
+int16x8_t vqdmulhq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_n_s16)))
+int16x8_t vqdmulhq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_n_s32)))
+int32x4_t vqdmulhq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_n_s32)))
+int32x4_t vqdmulhq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_n_s8)))
+int8x16_t vqdmulhq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_n_s8)))
+int8x16_t vqdmulhq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_s16)))
+int16x8_t vqdmulhq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_s16)))
+int16x8_t vqdmulhq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_s32)))
+int32x4_t vqdmulhq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_s32)))
+int32x4_t vqdmulhq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_s8)))
+int8x16_t vqdmulhq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulhq_s8)))
+int8x16_t vqdmulhq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_n_s16)))
+int32x4_t vqdmullbq_m_n_s16(int32x4_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_n_s16)))
+int32x4_t vqdmullbq_m(int32x4_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_n_s32)))
+int64x2_t vqdmullbq_m_n_s32(int64x2_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_n_s32)))
+int64x2_t vqdmullbq_m(int64x2_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_s16)))
+int32x4_t vqdmullbq_m_s16(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_s16)))
+int32x4_t vqdmullbq_m(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_s32)))
+int64x2_t vqdmullbq_m_s32(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_m_s32)))
+int64x2_t vqdmullbq_m(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_n_s16)))
+int32x4_t vqdmullbq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_n_s16)))
+int32x4_t vqdmullbq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_n_s32)))
+int64x2_t vqdmullbq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_n_s32)))
+int64x2_t vqdmullbq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_s16)))
+int32x4_t vqdmullbq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_s16)))
+int32x4_t vqdmullbq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_s32)))
+int64x2_t vqdmullbq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmullbq_s32)))
+int64x2_t vqdmullbq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_n_s16)))
+int32x4_t vqdmulltq_m_n_s16(int32x4_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_n_s16)))
+int32x4_t vqdmulltq_m(int32x4_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_n_s32)))
+int64x2_t vqdmulltq_m_n_s32(int64x2_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_n_s32)))
+int64x2_t vqdmulltq_m(int64x2_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_s16)))
+int32x4_t vqdmulltq_m_s16(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_s16)))
+int32x4_t vqdmulltq_m(int32x4_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_s32)))
+int64x2_t vqdmulltq_m_s32(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_m_s32)))
+int64x2_t vqdmulltq_m(int64x2_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_n_s16)))
+int32x4_t vqdmulltq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_n_s16)))
+int32x4_t vqdmulltq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_n_s32)))
+int64x2_t vqdmulltq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_n_s32)))
+int64x2_t vqdmulltq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_s16)))
+int32x4_t vqdmulltq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_s16)))
+int32x4_t vqdmulltq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_s32)))
+int64x2_t vqdmulltq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqdmulltq_s32)))
+int64x2_t vqdmulltq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_s16)))
+int8x16_t vqmovnbq_m_s16(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_s16)))
+int8x16_t vqmovnbq_m(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_s32)))
+int16x8_t vqmovnbq_m_s32(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_s32)))
+int16x8_t vqmovnbq_m(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_u16)))
+uint8x16_t vqmovnbq_m_u16(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_u16)))
+uint8x16_t vqmovnbq_m(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_u32)))
+uint16x8_t vqmovnbq_m_u32(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_m_u32)))
+uint16x8_t vqmovnbq_m(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_s16)))
+int8x16_t vqmovnbq_s16(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_s16)))
+int8x16_t vqmovnbq(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_s32)))
+int16x8_t vqmovnbq_s32(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_s32)))
+int16x8_t vqmovnbq(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_u16)))
+uint8x16_t vqmovnbq_u16(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_u16)))
+uint8x16_t vqmovnbq(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_u32)))
+uint16x8_t vqmovnbq_u32(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovnbq_u32)))
+uint16x8_t vqmovnbq(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_s16)))
+int8x16_t vqmovntq_m_s16(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_s16)))
+int8x16_t vqmovntq_m(int8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_s32)))
+int16x8_t vqmovntq_m_s32(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_s32)))
+int16x8_t vqmovntq_m(int16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_u16)))
+uint8x16_t vqmovntq_m_u16(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_u16)))
+uint8x16_t vqmovntq_m(uint8x16_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_u32)))
+uint16x8_t vqmovntq_m_u32(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_m_u32)))
+uint16x8_t vqmovntq_m(uint16x8_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_s16)))
+int8x16_t vqmovntq_s16(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_s16)))
+int8x16_t vqmovntq(int8x16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_s32)))
+int16x8_t vqmovntq_s32(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_s32)))
+int16x8_t vqmovntq(int16x8_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_u16)))
+uint8x16_t vqmovntq_u16(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_u16)))
+uint8x16_t vqmovntq(uint8x16_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_u32)))
+uint16x8_t vqmovntq_u32(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovntq_u32)))
+uint16x8_t vqmovntq(uint16x8_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_m_s16)))
+uint8x16_t vqmovunbq_m_s16(uint8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_m_s16)))
+uint8x16_t vqmovunbq_m(uint8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_m_s32)))
+uint16x8_t vqmovunbq_m_s32(uint16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_m_s32)))
+uint16x8_t vqmovunbq_m(uint16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_s16)))
+uint8x16_t vqmovunbq_s16(uint8x16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_s16)))
+uint8x16_t vqmovunbq(uint8x16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_s32)))
+uint16x8_t vqmovunbq_s32(uint16x8_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovunbq_s32)))
+uint16x8_t vqmovunbq(uint16x8_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_m_s16)))
+uint8x16_t vqmovuntq_m_s16(uint8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_m_s16)))
+uint8x16_t vqmovuntq_m(uint8x16_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_m_s32)))
+uint16x8_t vqmovuntq_m_s32(uint16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_m_s32)))
+uint16x8_t vqmovuntq_m(uint16x8_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_s16)))
+uint8x16_t vqmovuntq_s16(uint8x16_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_s16)))
+uint8x16_t vqmovuntq(uint8x16_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_s32)))
+uint16x8_t vqmovuntq_s32(uint16x8_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqmovuntq_s32)))
+uint16x8_t vqmovuntq(uint16x8_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_m_s16)))
+int16x8_t vqnegq_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_m_s16)))
+int16x8_t vqnegq_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_m_s32)))
+int32x4_t vqnegq_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_m_s32)))
+int32x4_t vqnegq_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_m_s8)))
+int8x16_t vqnegq_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_m_s8)))
+int8x16_t vqnegq_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_s16)))
+int16x8_t vqnegq_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_s16)))
+int16x8_t vqnegq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_s32)))
+int32x4_t vqnegq_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_s32)))
+int32x4_t vqnegq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_s8)))
+int8x16_t vqnegq_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqnegq_s8)))
+int8x16_t vqnegq(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_m_s16)))
+int16x8_t vqrdmladhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_m_s16)))
+int16x8_t vqrdmladhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_m_s32)))
+int32x4_t vqrdmladhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_m_s32)))
+int32x4_t vqrdmladhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_m_s8)))
+int8x16_t vqrdmladhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_m_s8)))
+int8x16_t vqrdmladhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_s16)))
+int16x8_t vqrdmladhq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_s16)))
+int16x8_t vqrdmladhq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_s32)))
+int32x4_t vqrdmladhq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_s32)))
+int32x4_t vqrdmladhq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_s8)))
+int8x16_t vqrdmladhq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhq_s8)))
+int8x16_t vqrdmladhq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_m_s16)))
+int16x8_t vqrdmladhxq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_m_s16)))
+int16x8_t vqrdmladhxq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_m_s32)))
+int32x4_t vqrdmladhxq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_m_s32)))
+int32x4_t vqrdmladhxq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_m_s8)))
+int8x16_t vqrdmladhxq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_m_s8)))
+int8x16_t vqrdmladhxq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_s16)))
+int16x8_t vqrdmladhxq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_s16)))
+int16x8_t vqrdmladhxq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_s32)))
+int32x4_t vqrdmladhxq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_s32)))
+int32x4_t vqrdmladhxq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_s8)))
+int8x16_t vqrdmladhxq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmladhxq_s8)))
+int8x16_t vqrdmladhxq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_m_n_s16)))
+int16x8_t vqrdmlahq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_m_n_s16)))
+int16x8_t vqrdmlahq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_m_n_s32)))
+int32x4_t vqrdmlahq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_m_n_s32)))
+int32x4_t vqrdmlahq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_m_n_s8)))
+int8x16_t vqrdmlahq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_m_n_s8)))
+int8x16_t vqrdmlahq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_n_s16)))
+int16x8_t vqrdmlahq_n_s16(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_n_s16)))
+int16x8_t vqrdmlahq(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_n_s32)))
+int32x4_t vqrdmlahq_n_s32(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_n_s32)))
+int32x4_t vqrdmlahq(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_n_s8)))
+int8x16_t vqrdmlahq_n_s8(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlahq_n_s8)))
+int8x16_t vqrdmlahq(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_m_n_s16)))
+int16x8_t vqrdmlashq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_m_n_s16)))
+int16x8_t vqrdmlashq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_m_n_s32)))
+int32x4_t vqrdmlashq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_m_n_s32)))
+int32x4_t vqrdmlashq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_m_n_s8)))
+int8x16_t vqrdmlashq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_m_n_s8)))
+int8x16_t vqrdmlashq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_n_s16)))
+int16x8_t vqrdmlashq_n_s16(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_n_s16)))
+int16x8_t vqrdmlashq(int16x8_t, int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_n_s32)))
+int32x4_t vqrdmlashq_n_s32(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_n_s32)))
+int32x4_t vqrdmlashq(int32x4_t, int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_n_s8)))
+int8x16_t vqrdmlashq_n_s8(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlashq_n_s8)))
+int8x16_t vqrdmlashq(int8x16_t, int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_m_s16)))
+int16x8_t vqrdmlsdhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_m_s16)))
+int16x8_t vqrdmlsdhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_m_s32)))
+int32x4_t vqrdmlsdhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_m_s32)))
+int32x4_t vqrdmlsdhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_m_s8)))
+int8x16_t vqrdmlsdhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_m_s8)))
+int8x16_t vqrdmlsdhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_s16)))
+int16x8_t vqrdmlsdhq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_s16)))
+int16x8_t vqrdmlsdhq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_s32)))
+int32x4_t vqrdmlsdhq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_s32)))
+int32x4_t vqrdmlsdhq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_s8)))
+int8x16_t vqrdmlsdhq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhq_s8)))
+int8x16_t vqrdmlsdhq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_m_s16)))
+int16x8_t vqrdmlsdhxq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_m_s16)))
+int16x8_t vqrdmlsdhxq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_m_s32)))
+int32x4_t vqrdmlsdhxq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_m_s32)))
+int32x4_t vqrdmlsdhxq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_m_s8)))
+int8x16_t vqrdmlsdhxq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_m_s8)))
+int8x16_t vqrdmlsdhxq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_s16)))
+int16x8_t vqrdmlsdhxq_s16(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_s16)))
+int16x8_t vqrdmlsdhxq(int16x8_t, int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_s32)))
+int32x4_t vqrdmlsdhxq_s32(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_s32)))
+int32x4_t vqrdmlsdhxq(int32x4_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_s8)))
+int8x16_t vqrdmlsdhxq_s8(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmlsdhxq_s8)))
+int8x16_t vqrdmlsdhxq(int8x16_t, int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_n_s16)))
+int16x8_t vqrdmulhq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_n_s16)))
+int16x8_t vqrdmulhq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_n_s32)))
+int32x4_t vqrdmulhq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_n_s32)))
+int32x4_t vqrdmulhq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_n_s8)))
+int8x16_t vqrdmulhq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_n_s8)))
+int8x16_t vqrdmulhq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_s16)))
+int16x8_t vqrdmulhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_s16)))
+int16x8_t vqrdmulhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_s32)))
+int32x4_t vqrdmulhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_s32)))
+int32x4_t vqrdmulhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_s8)))
+int8x16_t vqrdmulhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_m_s8)))
+int8x16_t vqrdmulhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_n_s16)))
+int16x8_t vqrdmulhq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_n_s16)))
+int16x8_t vqrdmulhq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_n_s32)))
+int32x4_t vqrdmulhq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_n_s32)))
+int32x4_t vqrdmulhq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_n_s8)))
+int8x16_t vqrdmulhq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_n_s8)))
+int8x16_t vqrdmulhq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_s16)))
+int16x8_t vqrdmulhq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_s16)))
+int16x8_t vqrdmulhq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_s32)))
+int32x4_t vqrdmulhq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_s32)))
+int32x4_t vqrdmulhq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_s8)))
+int8x16_t vqrdmulhq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrdmulhq_s8)))
+int8x16_t vqrdmulhq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_s16)))
+int16x8_t vqrshlq_m_n_s16(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_s16)))
+int16x8_t vqrshlq_m_n(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_s32)))
+int32x4_t vqrshlq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_s32)))
+int32x4_t vqrshlq_m_n(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_s8)))
+int8x16_t vqrshlq_m_n_s8(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_s8)))
+int8x16_t vqrshlq_m_n(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_u16)))
+uint16x8_t vqrshlq_m_n_u16(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_u16)))
+uint16x8_t vqrshlq_m_n(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_u32)))
+uint32x4_t vqrshlq_m_n_u32(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_u32)))
+uint32x4_t vqrshlq_m_n(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_u8)))
+uint8x16_t vqrshlq_m_n_u8(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_n_u8)))
+uint8x16_t vqrshlq_m_n(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_s16)))
+int16x8_t vqrshlq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_s16)))
+int16x8_t vqrshlq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_s32)))
+int32x4_t vqrshlq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_s32)))
+int32x4_t vqrshlq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_s8)))
+int8x16_t vqrshlq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_s8)))
+int8x16_t vqrshlq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_u16)))
+uint16x8_t vqrshlq_m_u16(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_u16)))
+uint16x8_t vqrshlq_m(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_u32)))
+uint32x4_t vqrshlq_m_u32(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_u32)))
+uint32x4_t vqrshlq_m(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_u8)))
+uint8x16_t vqrshlq_m_u8(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_m_u8)))
+uint8x16_t vqrshlq_m(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_s16)))
+int16x8_t vqrshlq_n_s16(int16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_s16)))
+int16x8_t vqrshlq(int16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_s32)))
+int32x4_t vqrshlq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_s32)))
+int32x4_t vqrshlq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_s8)))
+int8x16_t vqrshlq_n_s8(int8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_s8)))
+int8x16_t vqrshlq(int8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_u16)))
+uint16x8_t vqrshlq_n_u16(uint16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_u16)))
+uint16x8_t vqrshlq(uint16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_u32)))
+uint32x4_t vqrshlq_n_u32(uint32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_u32)))
+uint32x4_t vqrshlq(uint32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_u8)))
+uint8x16_t vqrshlq_n_u8(uint8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_n_u8)))
+uint8x16_t vqrshlq(uint8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_s16)))
+int16x8_t vqrshlq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_s16)))
+int16x8_t vqrshlq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_s32)))
+int32x4_t vqrshlq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_s32)))
+int32x4_t vqrshlq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_s8)))
+int8x16_t vqrshlq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_s8)))
+int8x16_t vqrshlq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_u16)))
+uint16x8_t vqrshlq_u16(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_u16)))
+uint16x8_t vqrshlq(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_u32)))
+uint32x4_t vqrshlq_u32(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_u32)))
+uint32x4_t vqrshlq(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_u8)))
+uint8x16_t vqrshlq_u8(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshlq_u8)))
+uint8x16_t vqrshlq(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_s16)))
+int8x16_t vqrshrnbq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_s16)))
+int8x16_t vqrshrnbq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_s32)))
+int16x8_t vqrshrnbq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_s32)))
+int16x8_t vqrshrnbq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_u16)))
+uint8x16_t vqrshrnbq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_u16)))
+uint8x16_t vqrshrnbq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_u32)))
+uint16x8_t vqrshrnbq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_m_n_u32)))
+uint16x8_t vqrshrnbq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_s16)))
+int8x16_t vqrshrnbq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_s16)))
+int8x16_t vqrshrnbq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_s32)))
+int16x8_t vqrshrnbq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_s32)))
+int16x8_t vqrshrnbq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_u16)))
+uint8x16_t vqrshrnbq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_u16)))
+uint8x16_t vqrshrnbq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_u32)))
+uint16x8_t vqrshrnbq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrnbq_n_u32)))
+uint16x8_t vqrshrnbq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_s16)))
+int8x16_t vqrshrntq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_s16)))
+int8x16_t vqrshrntq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_s32)))
+int16x8_t vqrshrntq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_s32)))
+int16x8_t vqrshrntq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_u16)))
+uint8x16_t vqrshrntq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_u16)))
+uint8x16_t vqrshrntq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_u32)))
+uint16x8_t vqrshrntq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_m_n_u32)))
+uint16x8_t vqrshrntq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_s16)))
+int8x16_t vqrshrntq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_s16)))
+int8x16_t vqrshrntq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_s32)))
+int16x8_t vqrshrntq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_s32)))
+int16x8_t vqrshrntq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_u16)))
+uint8x16_t vqrshrntq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_u16)))
+uint8x16_t vqrshrntq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_u32)))
+uint16x8_t vqrshrntq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrntq_n_u32)))
+uint16x8_t vqrshrntq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_m_n_s16)))
+uint8x16_t vqrshrunbq_m_n_s16(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_m_n_s16)))
+uint8x16_t vqrshrunbq_m(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_m_n_s32)))
+uint16x8_t vqrshrunbq_m_n_s32(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_m_n_s32)))
+uint16x8_t vqrshrunbq_m(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_n_s16)))
+uint8x16_t vqrshrunbq_n_s16(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_n_s16)))
+uint8x16_t vqrshrunbq(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_n_s32)))
+uint16x8_t vqrshrunbq_n_s32(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshrunbq_n_s32)))
+uint16x8_t vqrshrunbq(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_m_n_s16)))
+uint8x16_t vqrshruntq_m_n_s16(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_m_n_s16)))
+uint8x16_t vqrshruntq_m(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_m_n_s32)))
+uint16x8_t vqrshruntq_m_n_s32(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_m_n_s32)))
+uint16x8_t vqrshruntq_m(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_n_s16)))
+uint8x16_t vqrshruntq_n_s16(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_n_s16)))
+uint8x16_t vqrshruntq(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_n_s32)))
+uint16x8_t vqrshruntq_n_s32(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqrshruntq_n_s32)))
+uint16x8_t vqrshruntq(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_s16)))
+int16x8_t vqshlq_m_n_s16(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_s16)))
+int16x8_t vqshlq_m_n(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_s32)))
+int32x4_t vqshlq_m_n_s32(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_s32)))
+int32x4_t vqshlq_m_n(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_s8)))
+int8x16_t vqshlq_m_n_s8(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_s8)))
+int8x16_t vqshlq_m_n(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_u16)))
+uint16x8_t vqshlq_m_n_u16(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_u16)))
+uint16x8_t vqshlq_m_n(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_u32)))
+uint32x4_t vqshlq_m_n_u32(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_u32)))
+uint32x4_t vqshlq_m_n(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_u8)))
+uint8x16_t vqshlq_m_n_u8(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_n_u8)))
+uint8x16_t vqshlq_m_n(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_s16)))
+int16x8_t vqshlq_m_r_s16(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_s16)))
+int16x8_t vqshlq_m_r(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_s32)))
+int32x4_t vqshlq_m_r_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_s32)))
+int32x4_t vqshlq_m_r(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_s8)))
+int8x16_t vqshlq_m_r_s8(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_s8)))
+int8x16_t vqshlq_m_r(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_u16)))
+uint16x8_t vqshlq_m_r_u16(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_u16)))
+uint16x8_t vqshlq_m_r(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_u32)))
+uint32x4_t vqshlq_m_r_u32(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_u32)))
+uint32x4_t vqshlq_m_r(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_u8)))
+uint8x16_t vqshlq_m_r_u8(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_r_u8)))
+uint8x16_t vqshlq_m_r(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_s16)))
+int16x8_t vqshlq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_s16)))
+int16x8_t vqshlq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_s32)))
+int32x4_t vqshlq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_s32)))
+int32x4_t vqshlq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_s8)))
+int8x16_t vqshlq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_s8)))
+int8x16_t vqshlq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_u16)))
+uint16x8_t vqshlq_m_u16(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_u16)))
+uint16x8_t vqshlq_m(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_u32)))
+uint32x4_t vqshlq_m_u32(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_u32)))
+uint32x4_t vqshlq_m(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_u8)))
+uint8x16_t vqshlq_m_u8(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_m_u8)))
+uint8x16_t vqshlq_m(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_s16)))
+int16x8_t vqshlq_n_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_s16)))
+int16x8_t vqshlq_n(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_s32)))
+int32x4_t vqshlq_n_s32(int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_s32)))
+int32x4_t vqshlq_n(int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_s8)))
+int8x16_t vqshlq_n_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_s8)))
+int8x16_t vqshlq_n(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_u16)))
+uint16x8_t vqshlq_n_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_u16)))
+uint16x8_t vqshlq_n(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_u32)))
+uint32x4_t vqshlq_n_u32(uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_u32)))
+uint32x4_t vqshlq_n(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_u8)))
+uint8x16_t vqshlq_n_u8(uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_n_u8)))
+uint8x16_t vqshlq_n(uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_s16)))
+int16x8_t vqshlq_r_s16(int16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_s16)))
+int16x8_t vqshlq_r(int16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_s32)))
+int32x4_t vqshlq_r_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_s32)))
+int32x4_t vqshlq_r(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_s8)))
+int8x16_t vqshlq_r_s8(int8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_s8)))
+int8x16_t vqshlq_r(int8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_u16)))
+uint16x8_t vqshlq_r_u16(uint16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_u16)))
+uint16x8_t vqshlq_r(uint16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_u32)))
+uint32x4_t vqshlq_r_u32(uint32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_u32)))
+uint32x4_t vqshlq_r(uint32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_u8)))
+uint8x16_t vqshlq_r_u8(uint8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_r_u8)))
+uint8x16_t vqshlq_r(uint8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_s16)))
+int16x8_t vqshlq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_s16)))
+int16x8_t vqshlq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_s32)))
+int32x4_t vqshlq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_s32)))
+int32x4_t vqshlq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_s8)))
+int8x16_t vqshlq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_s8)))
+int8x16_t vqshlq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_u16)))
+uint16x8_t vqshlq_u16(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_u16)))
+uint16x8_t vqshlq(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_u32)))
+uint32x4_t vqshlq_u32(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_u32)))
+uint32x4_t vqshlq(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_u8)))
+uint8x16_t vqshlq_u8(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshlq_u8)))
+uint8x16_t vqshlq(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_m_n_s16)))
+uint16x8_t vqshluq_m_n_s16(uint16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_m_n_s16)))
+uint16x8_t vqshluq_m(uint16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_m_n_s32)))
+uint32x4_t vqshluq_m_n_s32(uint32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_m_n_s32)))
+uint32x4_t vqshluq_m(uint32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_m_n_s8)))
+uint8x16_t vqshluq_m_n_s8(uint8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_m_n_s8)))
+uint8x16_t vqshluq_m(uint8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_n_s16)))
+uint16x8_t vqshluq_n_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_n_s16)))
+uint16x8_t vqshluq(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_n_s32)))
+uint32x4_t vqshluq_n_s32(int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_n_s32)))
+uint32x4_t vqshluq(int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_n_s8)))
+uint8x16_t vqshluq_n_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshluq_n_s8)))
+uint8x16_t vqshluq(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_s16)))
+int8x16_t vqshrnbq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_s16)))
+int8x16_t vqshrnbq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_s32)))
+int16x8_t vqshrnbq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_s32)))
+int16x8_t vqshrnbq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_u16)))
+uint8x16_t vqshrnbq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_u16)))
+uint8x16_t vqshrnbq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_u32)))
+uint16x8_t vqshrnbq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_m_n_u32)))
+uint16x8_t vqshrnbq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_s16)))
+int8x16_t vqshrnbq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_s16)))
+int8x16_t vqshrnbq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_s32)))
+int16x8_t vqshrnbq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_s32)))
+int16x8_t vqshrnbq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_u16)))
+uint8x16_t vqshrnbq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_u16)))
+uint8x16_t vqshrnbq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_u32)))
+uint16x8_t vqshrnbq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrnbq_n_u32)))
+uint16x8_t vqshrnbq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_s16)))
+int8x16_t vqshrntq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_s16)))
+int8x16_t vqshrntq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_s32)))
+int16x8_t vqshrntq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_s32)))
+int16x8_t vqshrntq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_u16)))
+uint8x16_t vqshrntq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_u16)))
+uint8x16_t vqshrntq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_u32)))
+uint16x8_t vqshrntq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_m_n_u32)))
+uint16x8_t vqshrntq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_s16)))
+int8x16_t vqshrntq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_s16)))
+int8x16_t vqshrntq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_s32)))
+int16x8_t vqshrntq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_s32)))
+int16x8_t vqshrntq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_u16)))
+uint8x16_t vqshrntq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_u16)))
+uint8x16_t vqshrntq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_u32)))
+uint16x8_t vqshrntq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrntq_n_u32)))
+uint16x8_t vqshrntq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_m_n_s16)))
+uint8x16_t vqshrunbq_m_n_s16(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_m_n_s16)))
+uint8x16_t vqshrunbq_m(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_m_n_s32)))
+uint16x8_t vqshrunbq_m_n_s32(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_m_n_s32)))
+uint16x8_t vqshrunbq_m(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_n_s16)))
+uint8x16_t vqshrunbq_n_s16(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_n_s16)))
+uint8x16_t vqshrunbq(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_n_s32)))
+uint16x8_t vqshrunbq_n_s32(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshrunbq_n_s32)))
+uint16x8_t vqshrunbq(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_m_n_s16)))
+uint8x16_t vqshruntq_m_n_s16(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_m_n_s16)))
+uint8x16_t vqshruntq_m(uint8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_m_n_s32)))
+uint16x8_t vqshruntq_m_n_s32(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_m_n_s32)))
+uint16x8_t vqshruntq_m(uint16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_n_s16)))
+uint8x16_t vqshruntq_n_s16(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_n_s16)))
+uint8x16_t vqshruntq(uint8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_n_s32)))
+uint16x8_t vqshruntq_n_s32(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqshruntq_n_s32)))
+uint16x8_t vqshruntq(uint16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_s16)))
+int16x8_t vqsubq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_s16)))
+int16x8_t vqsubq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_s32)))
+int32x4_t vqsubq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_s32)))
+int32x4_t vqsubq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_s8)))
+int8x16_t vqsubq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_s8)))
+int8x16_t vqsubq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_u16)))
+uint16x8_t vqsubq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_u16)))
+uint16x8_t vqsubq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_u32)))
+uint32x4_t vqsubq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_u32)))
+uint32x4_t vqsubq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_u8)))
+uint8x16_t vqsubq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_n_u8)))
+uint8x16_t vqsubq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_s16)))
+int16x8_t vqsubq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_s16)))
+int16x8_t vqsubq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_s32)))
+int32x4_t vqsubq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_s32)))
+int32x4_t vqsubq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_s8)))
+int8x16_t vqsubq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_s8)))
+int8x16_t vqsubq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_u16)))
+uint16x8_t vqsubq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_u16)))
+uint16x8_t vqsubq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_u32)))
+uint32x4_t vqsubq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_u32)))
+uint32x4_t vqsubq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_u8)))
+uint8x16_t vqsubq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_m_u8)))
+uint8x16_t vqsubq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_s16)))
+int16x8_t vqsubq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_s16)))
+int16x8_t vqsubq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_s32)))
+int32x4_t vqsubq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_s32)))
+int32x4_t vqsubq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_s8)))
+int8x16_t vqsubq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_s8)))
+int8x16_t vqsubq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_u16)))
+uint16x8_t vqsubq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_u16)))
+uint16x8_t vqsubq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_u32)))
+uint32x4_t vqsubq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_u32)))
+uint32x4_t vqsubq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_u8)))
+uint8x16_t vqsubq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_n_u8)))
+uint8x16_t vqsubq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_s16)))
+int16x8_t vqsubq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_s16)))
+int16x8_t vqsubq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_s32)))
+int32x4_t vqsubq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_s32)))
+int32x4_t vqsubq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_s8)))
+int8x16_t vqsubq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_s8)))
+int8x16_t vqsubq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_u16)))
+uint16x8_t vqsubq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_u16)))
+uint16x8_t vqsubq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_u32)))
+uint32x4_t vqsubq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_u32)))
+uint32x4_t vqsubq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_u8)))
+uint8x16_t vqsubq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vqsubq_u8)))
+uint8x16_t vqsubq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_s32)))
+int16x8_t vreinterpretq_s16_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_s32)))
+int16x8_t vreinterpretq_s16(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_s64)))
+int16x8_t vreinterpretq_s16_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_s64)))
+int16x8_t vreinterpretq_s16(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_s8)))
+int16x8_t vreinterpretq_s16_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_s8)))
+int16x8_t vreinterpretq_s16(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u16)))
+int16x8_t vreinterpretq_s16_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u16)))
+int16x8_t vreinterpretq_s16(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u32)))
+int16x8_t vreinterpretq_s16_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u32)))
+int16x8_t vreinterpretq_s16(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u64)))
+int16x8_t vreinterpretq_s16_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u64)))
+int16x8_t vreinterpretq_s16(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u8)))
+int16x8_t vreinterpretq_s16_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_u8)))
+int16x8_t vreinterpretq_s16(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_s16)))
+int32x4_t vreinterpretq_s32_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_s16)))
+int32x4_t vreinterpretq_s32(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_s64)))
+int32x4_t vreinterpretq_s32_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_s64)))
+int32x4_t vreinterpretq_s32(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_s8)))
+int32x4_t vreinterpretq_s32_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_s8)))
+int32x4_t vreinterpretq_s32(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u16)))
+int32x4_t vreinterpretq_s32_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u16)))
+int32x4_t vreinterpretq_s32(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u32)))
+int32x4_t vreinterpretq_s32_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u32)))
+int32x4_t vreinterpretq_s32(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u64)))
+int32x4_t vreinterpretq_s32_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u64)))
+int32x4_t vreinterpretq_s32(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u8)))
+int32x4_t vreinterpretq_s32_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_u8)))
+int32x4_t vreinterpretq_s32(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_s16)))
+int64x2_t vreinterpretq_s64_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_s16)))
+int64x2_t vreinterpretq_s64(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_s32)))
+int64x2_t vreinterpretq_s64_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_s32)))
+int64x2_t vreinterpretq_s64(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_s8)))
+int64x2_t vreinterpretq_s64_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_s8)))
+int64x2_t vreinterpretq_s64(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u16)))
+int64x2_t vreinterpretq_s64_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u16)))
+int64x2_t vreinterpretq_s64(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u32)))
+int64x2_t vreinterpretq_s64_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u32)))
+int64x2_t vreinterpretq_s64(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u64)))
+int64x2_t vreinterpretq_s64_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u64)))
+int64x2_t vreinterpretq_s64(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u8)))
+int64x2_t vreinterpretq_s64_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_u8)))
+int64x2_t vreinterpretq_s64(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_s16)))
+int8x16_t vreinterpretq_s8_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_s16)))
+int8x16_t vreinterpretq_s8(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_s32)))
+int8x16_t vreinterpretq_s8_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_s32)))
+int8x16_t vreinterpretq_s8(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_s64)))
+int8x16_t vreinterpretq_s8_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_s64)))
+int8x16_t vreinterpretq_s8(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u16)))
+int8x16_t vreinterpretq_s8_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u16)))
+int8x16_t vreinterpretq_s8(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u32)))
+int8x16_t vreinterpretq_s8_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u32)))
+int8x16_t vreinterpretq_s8(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u64)))
+int8x16_t vreinterpretq_s8_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u64)))
+int8x16_t vreinterpretq_s8(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u8)))
+int8x16_t vreinterpretq_s8_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_u8)))
+int8x16_t vreinterpretq_s8(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s16)))
+uint16x8_t vreinterpretq_u16_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s16)))
+uint16x8_t vreinterpretq_u16(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s32)))
+uint16x8_t vreinterpretq_u16_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s32)))
+uint16x8_t vreinterpretq_u16(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s64)))
+uint16x8_t vreinterpretq_u16_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s64)))
+uint16x8_t vreinterpretq_u16(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s8)))
+uint16x8_t vreinterpretq_u16_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_s8)))
+uint16x8_t vreinterpretq_u16(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_u32)))
+uint16x8_t vreinterpretq_u16_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_u32)))
+uint16x8_t vreinterpretq_u16(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_u64)))
+uint16x8_t vreinterpretq_u16_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_u64)))
+uint16x8_t vreinterpretq_u16(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_u8)))
+uint16x8_t vreinterpretq_u16_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_u8)))
+uint16x8_t vreinterpretq_u16(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s16)))
+uint32x4_t vreinterpretq_u32_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s16)))
+uint32x4_t vreinterpretq_u32(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s32)))
+uint32x4_t vreinterpretq_u32_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s32)))
+uint32x4_t vreinterpretq_u32(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s64)))
+uint32x4_t vreinterpretq_u32_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s64)))
+uint32x4_t vreinterpretq_u32(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s8)))
+uint32x4_t vreinterpretq_u32_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_s8)))
+uint32x4_t vreinterpretq_u32(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_u16)))
+uint32x4_t vreinterpretq_u32_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_u16)))
+uint32x4_t vreinterpretq_u32(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_u64)))
+uint32x4_t vreinterpretq_u32_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_u64)))
+uint32x4_t vreinterpretq_u32(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_u8)))
+uint32x4_t vreinterpretq_u32_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_u8)))
+uint32x4_t vreinterpretq_u32(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s16)))
+uint64x2_t vreinterpretq_u64_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s16)))
+uint64x2_t vreinterpretq_u64(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s32)))
+uint64x2_t vreinterpretq_u64_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s32)))
+uint64x2_t vreinterpretq_u64(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s64)))
+uint64x2_t vreinterpretq_u64_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s64)))
+uint64x2_t vreinterpretq_u64(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s8)))
+uint64x2_t vreinterpretq_u64_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_s8)))
+uint64x2_t vreinterpretq_u64(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_u16)))
+uint64x2_t vreinterpretq_u64_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_u16)))
+uint64x2_t vreinterpretq_u64(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_u32)))
+uint64x2_t vreinterpretq_u64_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_u32)))
+uint64x2_t vreinterpretq_u64(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_u8)))
+uint64x2_t vreinterpretq_u64_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_u8)))
+uint64x2_t vreinterpretq_u64(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s16)))
+uint8x16_t vreinterpretq_u8_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s16)))
+uint8x16_t vreinterpretq_u8(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s32)))
+uint8x16_t vreinterpretq_u8_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s32)))
+uint8x16_t vreinterpretq_u8(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s64)))
+uint8x16_t vreinterpretq_u8_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s64)))
+uint8x16_t vreinterpretq_u8(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s8)))
+uint8x16_t vreinterpretq_u8_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_s8)))
+uint8x16_t vreinterpretq_u8(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_u16)))
+uint8x16_t vreinterpretq_u8_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_u16)))
+uint8x16_t vreinterpretq_u8(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_u32)))
+uint8x16_t vreinterpretq_u8_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_u32)))
+uint8x16_t vreinterpretq_u8(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_u64)))
+uint8x16_t vreinterpretq_u8_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_u64)))
+uint8x16_t vreinterpretq_u8(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_m_s8)))
+int8x16_t vrev16q_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_m_s8)))
+int8x16_t vrev16q_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_m_u8)))
+uint8x16_t vrev16q_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_m_u8)))
+uint8x16_t vrev16q_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_s8)))
+int8x16_t vrev16q_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_s8)))
+int8x16_t vrev16q(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_u8)))
+uint8x16_t vrev16q_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_u8)))
+uint8x16_t vrev16q(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_x_s8)))
+int8x16_t vrev16q_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_x_s8)))
+int8x16_t vrev16q_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_x_u8)))
+uint8x16_t vrev16q_x_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev16q_x_u8)))
+uint8x16_t vrev16q_x(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_s16)))
+int16x8_t vrev32q_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_s16)))
+int16x8_t vrev32q_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_s8)))
+int8x16_t vrev32q_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_s8)))
+int8x16_t vrev32q_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_u16)))
+uint16x8_t vrev32q_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_u16)))
+uint16x8_t vrev32q_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_u8)))
+uint8x16_t vrev32q_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_u8)))
+uint8x16_t vrev32q_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_s16)))
+int16x8_t vrev32q_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_s16)))
+int16x8_t vrev32q(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_s8)))
+int8x16_t vrev32q_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_s8)))
+int8x16_t vrev32q(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_u16)))
+uint16x8_t vrev32q_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_u16)))
+uint16x8_t vrev32q(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_u8)))
+uint8x16_t vrev32q_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_u8)))
+uint8x16_t vrev32q(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_s16)))
+int16x8_t vrev32q_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_s16)))
+int16x8_t vrev32q_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_s8)))
+int8x16_t vrev32q_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_s8)))
+int8x16_t vrev32q_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_u16)))
+uint16x8_t vrev32q_x_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_u16)))
+uint16x8_t vrev32q_x(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_u8)))
+uint8x16_t vrev32q_x_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_u8)))
+uint8x16_t vrev32q_x(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_s16)))
+int16x8_t vrev64q_m_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_s16)))
+int16x8_t vrev64q_m(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_s32)))
+int32x4_t vrev64q_m_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_s32)))
+int32x4_t vrev64q_m(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_s8)))
+int8x16_t vrev64q_m_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_s8)))
+int8x16_t vrev64q_m(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_u16)))
+uint16x8_t vrev64q_m_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_u16)))
+uint16x8_t vrev64q_m(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_u32)))
+uint32x4_t vrev64q_m_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_u32)))
+uint32x4_t vrev64q_m(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_u8)))
+uint8x16_t vrev64q_m_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_u8)))
+uint8x16_t vrev64q_m(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_s16)))
+int16x8_t vrev64q_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_s16)))
+int16x8_t vrev64q(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_s32)))
+int32x4_t vrev64q_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_s32)))
+int32x4_t vrev64q(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_s8)))
+int8x16_t vrev64q_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_s8)))
+int8x16_t vrev64q(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_u16)))
+uint16x8_t vrev64q_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_u16)))
+uint16x8_t vrev64q(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_u32)))
+uint32x4_t vrev64q_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_u32)))
+uint32x4_t vrev64q(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_u8)))
+uint8x16_t vrev64q_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_u8)))
+uint8x16_t vrev64q(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_s16)))
+int16x8_t vrev64q_x_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_s16)))
+int16x8_t vrev64q_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_s32)))
+int32x4_t vrev64q_x_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_s32)))
+int32x4_t vrev64q_x(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_s8)))
+int8x16_t vrev64q_x_s8(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_s8)))
+int8x16_t vrev64q_x(int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_u16)))
+uint16x8_t vrev64q_x_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_u16)))
+uint16x8_t vrev64q_x(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_u32)))
+uint32x4_t vrev64q_x_u32(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_u32)))
+uint32x4_t vrev64q_x(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_u8)))
+uint8x16_t vrev64q_x_u8(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_u8)))
+uint8x16_t vrev64q_x(uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_s16)))
+int16x8_t vrhaddq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_s16)))
+int16x8_t vrhaddq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_s32)))
+int32x4_t vrhaddq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_s32)))
+int32x4_t vrhaddq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_s8)))
+int8x16_t vrhaddq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_s8)))
+int8x16_t vrhaddq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_u16)))
+uint16x8_t vrhaddq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_u16)))
+uint16x8_t vrhaddq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_u32)))
+uint32x4_t vrhaddq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_u32)))
+uint32x4_t vrhaddq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_u8)))
+uint8x16_t vrhaddq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_m_u8)))
+uint8x16_t vrhaddq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_s16)))
+int16x8_t vrhaddq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_s16)))
+int16x8_t vrhaddq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_s32)))
+int32x4_t vrhaddq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_s32)))
+int32x4_t vrhaddq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_s8)))
+int8x16_t vrhaddq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_s8)))
+int8x16_t vrhaddq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_u16)))
+uint16x8_t vrhaddq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_u16)))
+uint16x8_t vrhaddq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_u32)))
+uint32x4_t vrhaddq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_u32)))
+uint32x4_t vrhaddq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_u8)))
+uint8x16_t vrhaddq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_u8)))
+uint8x16_t vrhaddq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_s16)))
+int16x8_t vrhaddq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_s16)))
+int16x8_t vrhaddq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_s32)))
+int32x4_t vrhaddq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_s32)))
+int32x4_t vrhaddq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_s8)))
+int8x16_t vrhaddq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_s8)))
+int8x16_t vrhaddq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_u16)))
+uint16x8_t vrhaddq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_u16)))
+uint16x8_t vrhaddq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_u32)))
+uint32x4_t vrhaddq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_u32)))
+uint32x4_t vrhaddq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_u8)))
+uint8x16_t vrhaddq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrhaddq_x_u8)))
+uint8x16_t vrhaddq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_p_s32)))
+int64_t vrmlaldavhaq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_p_s32)))
+int64_t vrmlaldavhaq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_p_u32)))
+uint64_t vrmlaldavhaq_p_u32(uint64_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_p_u32)))
+uint64_t vrmlaldavhaq_p(uint64_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_s32)))
+int64_t vrmlaldavhaq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_s32)))
+int64_t vrmlaldavhaq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_u32)))
+uint64_t vrmlaldavhaq_u32(uint64_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaq_u32)))
+uint64_t vrmlaldavhaq(uint64_t, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaxq_p_s32)))
+int64_t vrmlaldavhaxq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaxq_p_s32)))
+int64_t vrmlaldavhaxq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaxq_s32)))
+int64_t vrmlaldavhaxq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhaxq_s32)))
+int64_t vrmlaldavhaxq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_p_s32)))
+int64_t vrmlaldavhq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_p_s32)))
+int64_t vrmlaldavhq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_p_u32)))
+uint64_t vrmlaldavhq_p_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_p_u32)))
+uint64_t vrmlaldavhq_p(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_s32)))
+int64_t vrmlaldavhq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_s32)))
+int64_t vrmlaldavhq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_u32)))
+uint64_t vrmlaldavhq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhq_u32)))
+uint64_t vrmlaldavhq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhxq_p_s32)))
+int64_t vrmlaldavhxq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhxq_p_s32)))
+int64_t vrmlaldavhxq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhxq_s32)))
+int64_t vrmlaldavhxq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlaldavhxq_s32)))
+int64_t vrmlaldavhxq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaq_p_s32)))
+int64_t vrmlsldavhaq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaq_p_s32)))
+int64_t vrmlsldavhaq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaq_s32)))
+int64_t vrmlsldavhaq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaq_s32)))
+int64_t vrmlsldavhaq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaxq_p_s32)))
+int64_t vrmlsldavhaxq_p_s32(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaxq_p_s32)))
+int64_t vrmlsldavhaxq_p(int64_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaxq_s32)))
+int64_t vrmlsldavhaxq_s32(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhaxq_s32)))
+int64_t vrmlsldavhaxq(int64_t, int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhq_p_s32)))
+int64_t vrmlsldavhq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhq_p_s32)))
+int64_t vrmlsldavhq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhq_s32)))
+int64_t vrmlsldavhq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhq_s32)))
+int64_t vrmlsldavhq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhxq_p_s32)))
+int64_t vrmlsldavhxq_p_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhxq_p_s32)))
+int64_t vrmlsldavhxq_p(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhxq_s32)))
+int64_t vrmlsldavhxq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmlsldavhxq_s32)))
+int64_t vrmlsldavhxq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_s16)))
+int16x8_t vrmulhq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_s16)))
+int16x8_t vrmulhq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_s32)))
+int32x4_t vrmulhq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_s32)))
+int32x4_t vrmulhq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_s8)))
+int8x16_t vrmulhq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_s8)))
+int8x16_t vrmulhq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_u16)))
+uint16x8_t vrmulhq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_u16)))
+uint16x8_t vrmulhq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_u32)))
+uint32x4_t vrmulhq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_u32)))
+uint32x4_t vrmulhq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_u8)))
+uint8x16_t vrmulhq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_m_u8)))
+uint8x16_t vrmulhq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_s16)))
+int16x8_t vrmulhq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_s16)))
+int16x8_t vrmulhq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_s32)))
+int32x4_t vrmulhq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_s32)))
+int32x4_t vrmulhq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_s8)))
+int8x16_t vrmulhq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_s8)))
+int8x16_t vrmulhq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_u16)))
+uint16x8_t vrmulhq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_u16)))
+uint16x8_t vrmulhq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_u32)))
+uint32x4_t vrmulhq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_u32)))
+uint32x4_t vrmulhq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_u8)))
+uint8x16_t vrmulhq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_u8)))
+uint8x16_t vrmulhq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_s16)))
+int16x8_t vrmulhq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_s16)))
+int16x8_t vrmulhq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_s32)))
+int32x4_t vrmulhq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_s32)))
+int32x4_t vrmulhq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_s8)))
+int8x16_t vrmulhq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_s8)))
+int8x16_t vrmulhq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_u16)))
+uint16x8_t vrmulhq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_u16)))
+uint16x8_t vrmulhq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_u32)))
+uint32x4_t vrmulhq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_u32)))
+uint32x4_t vrmulhq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_u8)))
+uint8x16_t vrmulhq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrmulhq_x_u8)))
+uint8x16_t vrmulhq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_s16)))
+int16x8_t vrshlq_m_n_s16(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_s16)))
+int16x8_t vrshlq_m_n(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_s32)))
+int32x4_t vrshlq_m_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_s32)))
+int32x4_t vrshlq_m_n(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_s8)))
+int8x16_t vrshlq_m_n_s8(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_s8)))
+int8x16_t vrshlq_m_n(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_u16)))
+uint16x8_t vrshlq_m_n_u16(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_u16)))
+uint16x8_t vrshlq_m_n(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_u32)))
+uint32x4_t vrshlq_m_n_u32(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_u32)))
+uint32x4_t vrshlq_m_n(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_u8)))
+uint8x16_t vrshlq_m_n_u8(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_n_u8)))
+uint8x16_t vrshlq_m_n(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_s16)))
+int16x8_t vrshlq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_s16)))
+int16x8_t vrshlq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_s32)))
+int32x4_t vrshlq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_s32)))
+int32x4_t vrshlq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_s8)))
+int8x16_t vrshlq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_s8)))
+int8x16_t vrshlq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_u16)))
+uint16x8_t vrshlq_m_u16(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_u16)))
+uint16x8_t vrshlq_m(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_u32)))
+uint32x4_t vrshlq_m_u32(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_u32)))
+uint32x4_t vrshlq_m(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_u8)))
+uint8x16_t vrshlq_m_u8(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_m_u8)))
+uint8x16_t vrshlq_m(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_s16)))
+int16x8_t vrshlq_n_s16(int16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_s16)))
+int16x8_t vrshlq(int16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_s32)))
+int32x4_t vrshlq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_s32)))
+int32x4_t vrshlq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_s8)))
+int8x16_t vrshlq_n_s8(int8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_s8)))
+int8x16_t vrshlq(int8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_u16)))
+uint16x8_t vrshlq_n_u16(uint16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_u16)))
+uint16x8_t vrshlq(uint16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_u32)))
+uint32x4_t vrshlq_n_u32(uint32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_u32)))
+uint32x4_t vrshlq(uint32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_u8)))
+uint8x16_t vrshlq_n_u8(uint8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_n_u8)))
+uint8x16_t vrshlq(uint8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_s16)))
+int16x8_t vrshlq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_s16)))
+int16x8_t vrshlq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_s32)))
+int32x4_t vrshlq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_s32)))
+int32x4_t vrshlq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_s8)))
+int8x16_t vrshlq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_s8)))
+int8x16_t vrshlq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_u16)))
+uint16x8_t vrshlq_u16(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_u16)))
+uint16x8_t vrshlq(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_u32)))
+uint32x4_t vrshlq_u32(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_u32)))
+uint32x4_t vrshlq(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_u8)))
+uint8x16_t vrshlq_u8(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_u8)))
+uint8x16_t vrshlq(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_s16)))
+int16x8_t vrshlq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_s16)))
+int16x8_t vrshlq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_s32)))
+int32x4_t vrshlq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_s32)))
+int32x4_t vrshlq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_s8)))
+int8x16_t vrshlq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_s8)))
+int8x16_t vrshlq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_u16)))
+uint16x8_t vrshlq_x_u16(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_u16)))
+uint16x8_t vrshlq_x(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_u32)))
+uint32x4_t vrshlq_x_u32(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_u32)))
+uint32x4_t vrshlq_x(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_u8)))
+uint8x16_t vrshlq_x_u8(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshlq_x_u8)))
+uint8x16_t vrshlq_x(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_s16)))
+int8x16_t vrshrnbq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_s16)))
+int8x16_t vrshrnbq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_s32)))
+int16x8_t vrshrnbq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_s32)))
+int16x8_t vrshrnbq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_u16)))
+uint8x16_t vrshrnbq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_u16)))
+uint8x16_t vrshrnbq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_u32)))
+uint16x8_t vrshrnbq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_m_n_u32)))
+uint16x8_t vrshrnbq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_s16)))
+int8x16_t vrshrnbq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_s16)))
+int8x16_t vrshrnbq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_s32)))
+int16x8_t vrshrnbq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_s32)))
+int16x8_t vrshrnbq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_u16)))
+uint8x16_t vrshrnbq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_u16)))
+uint8x16_t vrshrnbq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_u32)))
+uint16x8_t vrshrnbq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrnbq_n_u32)))
+uint16x8_t vrshrnbq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_s16)))
+int8x16_t vrshrntq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_s16)))
+int8x16_t vrshrntq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_s32)))
+int16x8_t vrshrntq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_s32)))
+int16x8_t vrshrntq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_u16)))
+uint8x16_t vrshrntq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_u16)))
+uint8x16_t vrshrntq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_u32)))
+uint16x8_t vrshrntq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_m_n_u32)))
+uint16x8_t vrshrntq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_s16)))
+int8x16_t vrshrntq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_s16)))
+int8x16_t vrshrntq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_s32)))
+int16x8_t vrshrntq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_s32)))
+int16x8_t vrshrntq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_u16)))
+uint8x16_t vrshrntq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_u16)))
+uint8x16_t vrshrntq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_u32)))
+uint16x8_t vrshrntq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrntq_n_u32)))
+uint16x8_t vrshrntq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_s16)))
+int16x8_t vrshrq_m_n_s16(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_s16)))
+int16x8_t vrshrq_m(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_s32)))
+int32x4_t vrshrq_m_n_s32(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_s32)))
+int32x4_t vrshrq_m(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_s8)))
+int8x16_t vrshrq_m_n_s8(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_s8)))
+int8x16_t vrshrq_m(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_u16)))
+uint16x8_t vrshrq_m_n_u16(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_u16)))
+uint16x8_t vrshrq_m(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_u32)))
+uint32x4_t vrshrq_m_n_u32(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_u32)))
+uint32x4_t vrshrq_m(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_u8)))
+uint8x16_t vrshrq_m_n_u8(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_m_n_u8)))
+uint8x16_t vrshrq_m(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_s16)))
+int16x8_t vrshrq_n_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_s16)))
+int16x8_t vrshrq(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_s32)))
+int32x4_t vrshrq_n_s32(int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_s32)))
+int32x4_t vrshrq(int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_s8)))
+int8x16_t vrshrq_n_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_s8)))
+int8x16_t vrshrq(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_u16)))
+uint16x8_t vrshrq_n_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_u16)))
+uint16x8_t vrshrq(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_u32)))
+uint32x4_t vrshrq_n_u32(uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_u32)))
+uint32x4_t vrshrq(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_u8)))
+uint8x16_t vrshrq_n_u8(uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_n_u8)))
+uint8x16_t vrshrq(uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_s16)))
+int16x8_t vrshrq_x_n_s16(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_s16)))
+int16x8_t vrshrq_x(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_s32)))
+int32x4_t vrshrq_x_n_s32(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_s32)))
+int32x4_t vrshrq_x(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_s8)))
+int8x16_t vrshrq_x_n_s8(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_s8)))
+int8x16_t vrshrq_x(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_u16)))
+uint16x8_t vrshrq_x_n_u16(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_u16)))
+uint16x8_t vrshrq_x(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_u32)))
+uint32x4_t vrshrq_x_n_u32(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_u32)))
+uint32x4_t vrshrq_x(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_u8)))
+uint8x16_t vrshrq_x_n_u8(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrshrq_x_n_u8)))
+uint8x16_t vrshrq_x(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_m_s32)))
+int32x4_t vsbciq_m_s32(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_m_s32)))
+int32x4_t vsbciq_m(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_m_u32)))
+uint32x4_t vsbciq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_m_u32)))
+uint32x4_t vsbciq_m(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_s32)))
+int32x4_t vsbciq_s32(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_s32)))
+int32x4_t vsbciq(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_u32)))
+uint32x4_t vsbciq_u32(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbciq_u32)))
+uint32x4_t vsbciq(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_m_s32)))
+int32x4_t vsbcq_m_s32(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_m_s32)))
+int32x4_t vsbcq_m(int32x4_t, int32x4_t, int32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_m_u32)))
+uint32x4_t vsbcq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_m_u32)))
+uint32x4_t vsbcq_m(uint32x4_t, uint32x4_t, uint32x4_t, unsigned *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_s32)))
+int32x4_t vsbcq_s32(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_s32)))
+int32x4_t vsbcq(int32x4_t, int32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_u32)))
+uint32x4_t vsbcq_u32(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsbcq_u32)))
+uint32x4_t vsbcq(uint32x4_t, uint32x4_t, unsigned *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s16)))
+int16x8_t vsetq_lane_s16(int16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s16)))
+int16x8_t vsetq_lane(int16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s32)))
+int32x4_t vsetq_lane_s32(int32_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s32)))
+int32x4_t vsetq_lane(int32_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s64)))
+int64x2_t vsetq_lane_s64(int64_t, int64x2_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s64)))
+int64x2_t vsetq_lane(int64_t, int64x2_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s8)))
+int8x16_t vsetq_lane_s8(int8_t, int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_s8)))
+int8x16_t vsetq_lane(int8_t, int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u16)))
+uint16x8_t vsetq_lane_u16(uint16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u16)))
+uint16x8_t vsetq_lane(uint16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u32)))
+uint32x4_t vsetq_lane_u32(uint32_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u32)))
+uint32x4_t vsetq_lane(uint32_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u64)))
+uint64x2_t vsetq_lane_u64(uint64_t, uint64x2_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u64)))
+uint64x2_t vsetq_lane(uint64_t, uint64x2_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u8)))
+uint8x16_t vsetq_lane_u8(uint8_t, uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_u8)))
+uint8x16_t vsetq_lane(uint8_t, uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_s16)))
+int16x8_t vshlcq_m_s16(int16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_s16)))
+int16x8_t vshlcq_m(int16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_s32)))
+int32x4_t vshlcq_m_s32(int32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_s32)))
+int32x4_t vshlcq_m(int32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_s8)))
+int8x16_t vshlcq_m_s8(int8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_s8)))
+int8x16_t vshlcq_m(int8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_u16)))
+uint16x8_t vshlcq_m_u16(uint16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_u16)))
+uint16x8_t vshlcq_m(uint16x8_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_u32)))
+uint32x4_t vshlcq_m_u32(uint32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_u32)))
+uint32x4_t vshlcq_m(uint32x4_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_u8)))
+uint8x16_t vshlcq_m_u8(uint8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_m_u8)))
+uint8x16_t vshlcq_m(uint8x16_t, uint32_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_s16)))
+int16x8_t vshlcq_s16(int16x8_t, uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_s16)))
+int16x8_t vshlcq(int16x8_t, uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_s32)))
+int32x4_t vshlcq_s32(int32x4_t, uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_s32)))
+int32x4_t vshlcq(int32x4_t, uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_s8)))
+int8x16_t vshlcq_s8(int8x16_t, uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_s8)))
+int8x16_t vshlcq(int8x16_t, uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_u16)))
+uint16x8_t vshlcq_u16(uint16x8_t, uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_u16)))
+uint16x8_t vshlcq(uint16x8_t, uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_u32)))
+uint32x4_t vshlcq_u32(uint32x4_t, uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_u32)))
+uint32x4_t vshlcq(uint32x4_t, uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_u8)))
+uint8x16_t vshlcq_u8(uint8x16_t, uint32_t *, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlcq_u8)))
+uint8x16_t vshlcq(uint8x16_t, uint32_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_s16)))
+int32x4_t vshllbq_m_n_s16(int32x4_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_s16)))
+int32x4_t vshllbq_m(int32x4_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_s8)))
+int16x8_t vshllbq_m_n_s8(int16x8_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_s8)))
+int16x8_t vshllbq_m(int16x8_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_u16)))
+uint32x4_t vshllbq_m_n_u16(uint32x4_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_u16)))
+uint32x4_t vshllbq_m(uint32x4_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_u8)))
+uint16x8_t vshllbq_m_n_u8(uint16x8_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_m_n_u8)))
+uint16x8_t vshllbq_m(uint16x8_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_s16)))
+int32x4_t vshllbq_n_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_s16)))
+int32x4_t vshllbq(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_s8)))
+int16x8_t vshllbq_n_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_s8)))
+int16x8_t vshllbq(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_u16)))
+uint32x4_t vshllbq_n_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_u16)))
+uint32x4_t vshllbq(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_u8)))
+uint16x8_t vshllbq_n_u8(uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_n_u8)))
+uint16x8_t vshllbq(uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_s16)))
+int32x4_t vshllbq_x_n_s16(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_s16)))
+int32x4_t vshllbq_x(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_s8)))
+int16x8_t vshllbq_x_n_s8(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_s8)))
+int16x8_t vshllbq_x(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_u16)))
+uint32x4_t vshllbq_x_n_u16(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_u16)))
+uint32x4_t vshllbq_x(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_u8)))
+uint16x8_t vshllbq_x_n_u8(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshllbq_x_n_u8)))
+uint16x8_t vshllbq_x(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_s16)))
+int32x4_t vshlltq_m_n_s16(int32x4_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_s16)))
+int32x4_t vshlltq_m(int32x4_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_s8)))
+int16x8_t vshlltq_m_n_s8(int16x8_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_s8)))
+int16x8_t vshlltq_m(int16x8_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_u16)))
+uint32x4_t vshlltq_m_n_u16(uint32x4_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_u16)))
+uint32x4_t vshlltq_m(uint32x4_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_u8)))
+uint16x8_t vshlltq_m_n_u8(uint16x8_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_m_n_u8)))
+uint16x8_t vshlltq_m(uint16x8_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_s16)))
+int32x4_t vshlltq_n_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_s16)))
+int32x4_t vshlltq(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_s8)))
+int16x8_t vshlltq_n_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_s8)))
+int16x8_t vshlltq(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_u16)))
+uint32x4_t vshlltq_n_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_u16)))
+uint32x4_t vshlltq(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_u8)))
+uint16x8_t vshlltq_n_u8(uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_n_u8)))
+uint16x8_t vshlltq(uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_s16)))
+int32x4_t vshlltq_x_n_s16(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_s16)))
+int32x4_t vshlltq_x(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_s8)))
+int16x8_t vshlltq_x_n_s8(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_s8)))
+int16x8_t vshlltq_x(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_u16)))
+uint32x4_t vshlltq_x_n_u16(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_u16)))
+uint32x4_t vshlltq_x(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_u8)))
+uint16x8_t vshlltq_x_n_u8(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlltq_x_n_u8)))
+uint16x8_t vshlltq_x(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_s16)))
+int16x8_t vshlq_m_n_s16(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_s16)))
+int16x8_t vshlq_m_n(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_s32)))
+int32x4_t vshlq_m_n_s32(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_s32)))
+int32x4_t vshlq_m_n(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_s8)))
+int8x16_t vshlq_m_n_s8(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_s8)))
+int8x16_t vshlq_m_n(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_u16)))
+uint16x8_t vshlq_m_n_u16(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_u16)))
+uint16x8_t vshlq_m_n(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_u32)))
+uint32x4_t vshlq_m_n_u32(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_u32)))
+uint32x4_t vshlq_m_n(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_u8)))
+uint8x16_t vshlq_m_n_u8(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_n_u8)))
+uint8x16_t vshlq_m_n(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_s16)))
+int16x8_t vshlq_m_r_s16(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_s16)))
+int16x8_t vshlq_m_r(int16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_s32)))
+int32x4_t vshlq_m_r_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_s32)))
+int32x4_t vshlq_m_r(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_s8)))
+int8x16_t vshlq_m_r_s8(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_s8)))
+int8x16_t vshlq_m_r(int8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_u16)))
+uint16x8_t vshlq_m_r_u16(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_u16)))
+uint16x8_t vshlq_m_r(uint16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_u32)))
+uint32x4_t vshlq_m_r_u32(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_u32)))
+uint32x4_t vshlq_m_r(uint32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_u8)))
+uint8x16_t vshlq_m_r_u8(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_r_u8)))
+uint8x16_t vshlq_m_r(uint8x16_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_s16)))
+int16x8_t vshlq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_s16)))
+int16x8_t vshlq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_s32)))
+int32x4_t vshlq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_s32)))
+int32x4_t vshlq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_s8)))
+int8x16_t vshlq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_s8)))
+int8x16_t vshlq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_u16)))
+uint16x8_t vshlq_m_u16(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_u16)))
+uint16x8_t vshlq_m(uint16x8_t, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_u32)))
+uint32x4_t vshlq_m_u32(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_u32)))
+uint32x4_t vshlq_m(uint32x4_t, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_u8)))
+uint8x16_t vshlq_m_u8(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_m_u8)))
+uint8x16_t vshlq_m(uint8x16_t, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_s16)))
+int16x8_t vshlq_n_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_s16)))
+int16x8_t vshlq_n(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_s32)))
+int32x4_t vshlq_n_s32(int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_s32)))
+int32x4_t vshlq_n(int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_s8)))
+int8x16_t vshlq_n_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_s8)))
+int8x16_t vshlq_n(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_u16)))
+uint16x8_t vshlq_n_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_u16)))
+uint16x8_t vshlq_n(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_u32)))
+uint32x4_t vshlq_n_u32(uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_u32)))
+uint32x4_t vshlq_n(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_u8)))
+uint8x16_t vshlq_n_u8(uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_n_u8)))
+uint8x16_t vshlq_n(uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_s16)))
+int16x8_t vshlq_r_s16(int16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_s16)))
+int16x8_t vshlq_r(int16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_s32)))
+int32x4_t vshlq_r_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_s32)))
+int32x4_t vshlq_r(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_s8)))
+int8x16_t vshlq_r_s8(int8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_s8)))
+int8x16_t vshlq_r(int8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_u16)))
+uint16x8_t vshlq_r_u16(uint16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_u16)))
+uint16x8_t vshlq_r(uint16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_u32)))
+uint32x4_t vshlq_r_u32(uint32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_u32)))
+uint32x4_t vshlq_r(uint32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_u8)))
+uint8x16_t vshlq_r_u8(uint8x16_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_r_u8)))
+uint8x16_t vshlq_r(uint8x16_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_s16)))
+int16x8_t vshlq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_s16)))
+int16x8_t vshlq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_s32)))
+int32x4_t vshlq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_s32)))
+int32x4_t vshlq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_s8)))
+int8x16_t vshlq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_s8)))
+int8x16_t vshlq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_u16)))
+uint16x8_t vshlq_u16(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_u16)))
+uint16x8_t vshlq(uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_u32)))
+uint32x4_t vshlq_u32(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_u32)))
+uint32x4_t vshlq(uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_u8)))
+uint8x16_t vshlq_u8(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_u8)))
+uint8x16_t vshlq(uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_s16)))
+int16x8_t vshlq_x_n_s16(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_s16)))
+int16x8_t vshlq_x_n(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_s32)))
+int32x4_t vshlq_x_n_s32(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_s32)))
+int32x4_t vshlq_x_n(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_s8)))
+int8x16_t vshlq_x_n_s8(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_s8)))
+int8x16_t vshlq_x_n(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_u16)))
+uint16x8_t vshlq_x_n_u16(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_u16)))
+uint16x8_t vshlq_x_n(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_u32)))
+uint32x4_t vshlq_x_n_u32(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_u32)))
+uint32x4_t vshlq_x_n(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_u8)))
+uint8x16_t vshlq_x_n_u8(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_n_u8)))
+uint8x16_t vshlq_x_n(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_s16)))
+int16x8_t vshlq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_s16)))
+int16x8_t vshlq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_s32)))
+int32x4_t vshlq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_s32)))
+int32x4_t vshlq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_s8)))
+int8x16_t vshlq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_s8)))
+int8x16_t vshlq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_u16)))
+uint16x8_t vshlq_x_u16(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_u16)))
+uint16x8_t vshlq_x(uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_u32)))
+uint32x4_t vshlq_x_u32(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_u32)))
+uint32x4_t vshlq_x(uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_u8)))
+uint8x16_t vshlq_x_u8(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshlq_x_u8)))
+uint8x16_t vshlq_x(uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_s16)))
+int8x16_t vshrnbq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_s16)))
+int8x16_t vshrnbq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_s32)))
+int16x8_t vshrnbq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_s32)))
+int16x8_t vshrnbq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_u16)))
+uint8x16_t vshrnbq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_u16)))
+uint8x16_t vshrnbq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_u32)))
+uint16x8_t vshrnbq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_m_n_u32)))
+uint16x8_t vshrnbq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_s16)))
+int8x16_t vshrnbq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_s16)))
+int8x16_t vshrnbq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_s32)))
+int16x8_t vshrnbq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_s32)))
+int16x8_t vshrnbq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_u16)))
+uint8x16_t vshrnbq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_u16)))
+uint8x16_t vshrnbq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_u32)))
+uint16x8_t vshrnbq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrnbq_n_u32)))
+uint16x8_t vshrnbq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_s16)))
+int8x16_t vshrntq_m_n_s16(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_s16)))
+int8x16_t vshrntq_m(int8x16_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_s32)))
+int16x8_t vshrntq_m_n_s32(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_s32)))
+int16x8_t vshrntq_m(int16x8_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_u16)))
+uint8x16_t vshrntq_m_n_u16(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_u16)))
+uint8x16_t vshrntq_m(uint8x16_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_u32)))
+uint16x8_t vshrntq_m_n_u32(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_m_n_u32)))
+uint16x8_t vshrntq_m(uint16x8_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_s16)))
+int8x16_t vshrntq_n_s16(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_s16)))
+int8x16_t vshrntq(int8x16_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_s32)))
+int16x8_t vshrntq_n_s32(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_s32)))
+int16x8_t vshrntq(int16x8_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_u16)))
+uint8x16_t vshrntq_n_u16(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_u16)))
+uint8x16_t vshrntq(uint8x16_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_u32)))
+uint16x8_t vshrntq_n_u32(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrntq_n_u32)))
+uint16x8_t vshrntq(uint16x8_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_s16)))
+int16x8_t vshrq_m_n_s16(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_s16)))
+int16x8_t vshrq_m(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_s32)))
+int32x4_t vshrq_m_n_s32(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_s32)))
+int32x4_t vshrq_m(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_s8)))
+int8x16_t vshrq_m_n_s8(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_s8)))
+int8x16_t vshrq_m(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_u16)))
+uint16x8_t vshrq_m_n_u16(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_u16)))
+uint16x8_t vshrq_m(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_u32)))
+uint32x4_t vshrq_m_n_u32(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_u32)))
+uint32x4_t vshrq_m(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_u8)))
+uint8x16_t vshrq_m_n_u8(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_m_n_u8)))
+uint8x16_t vshrq_m(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_s16)))
+int16x8_t vshrq_n_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_s16)))
+int16x8_t vshrq(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_s32)))
+int32x4_t vshrq_n_s32(int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_s32)))
+int32x4_t vshrq(int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_s8)))
+int8x16_t vshrq_n_s8(int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_s8)))
+int8x16_t vshrq(int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_u16)))
+uint16x8_t vshrq_n_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_u16)))
+uint16x8_t vshrq(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_u32)))
+uint32x4_t vshrq_n_u32(uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_u32)))
+uint32x4_t vshrq(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_u8)))
+uint8x16_t vshrq_n_u8(uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_n_u8)))
+uint8x16_t vshrq(uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_s16)))
+int16x8_t vshrq_x_n_s16(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_s16)))
+int16x8_t vshrq_x(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_s32)))
+int32x4_t vshrq_x_n_s32(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_s32)))
+int32x4_t vshrq_x(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_s8)))
+int8x16_t vshrq_x_n_s8(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_s8)))
+int8x16_t vshrq_x(int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_u16)))
+uint16x8_t vshrq_x_n_u16(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_u16)))
+uint16x8_t vshrq_x(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_u32)))
+uint32x4_t vshrq_x_n_u32(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_u32)))
+uint32x4_t vshrq_x(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_u8)))
+uint8x16_t vshrq_x_n_u8(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vshrq_x_n_u8)))
+uint8x16_t vshrq_x(uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_s16)))
+int16x8_t vsliq_m_n_s16(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_s16)))
+int16x8_t vsliq_m(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_s32)))
+int32x4_t vsliq_m_n_s32(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_s32)))
+int32x4_t vsliq_m(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_s8)))
+int8x16_t vsliq_m_n_s8(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_s8)))
+int8x16_t vsliq_m(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_u16)))
+uint16x8_t vsliq_m_n_u16(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_u16)))
+uint16x8_t vsliq_m(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_u32)))
+uint32x4_t vsliq_m_n_u32(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_u32)))
+uint32x4_t vsliq_m(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_u8)))
+uint8x16_t vsliq_m_n_u8(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_m_n_u8)))
+uint8x16_t vsliq_m(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_s16)))
+int16x8_t vsliq_n_s16(int16x8_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_s16)))
+int16x8_t vsliq(int16x8_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_s32)))
+int32x4_t vsliq_n_s32(int32x4_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_s32)))
+int32x4_t vsliq(int32x4_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_s8)))
+int8x16_t vsliq_n_s8(int8x16_t, int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_s8)))
+int8x16_t vsliq(int8x16_t, int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_u16)))
+uint16x8_t vsliq_n_u16(uint16x8_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_u16)))
+uint16x8_t vsliq(uint16x8_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_u32)))
+uint32x4_t vsliq_n_u32(uint32x4_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_u32)))
+uint32x4_t vsliq(uint32x4_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_u8)))
+uint8x16_t vsliq_n_u8(uint8x16_t, uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsliq_n_u8)))
+uint8x16_t vsliq(uint8x16_t, uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_s16)))
+int16x8_t vsriq_m_n_s16(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_s16)))
+int16x8_t vsriq_m(int16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_s32)))
+int32x4_t vsriq_m_n_s32(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_s32)))
+int32x4_t vsriq_m(int32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_s8)))
+int8x16_t vsriq_m_n_s8(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_s8)))
+int8x16_t vsriq_m(int8x16_t, int8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_u16)))
+uint16x8_t vsriq_m_n_u16(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_u16)))
+uint16x8_t vsriq_m(uint16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_u32)))
+uint32x4_t vsriq_m_n_u32(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_u32)))
+uint32x4_t vsriq_m(uint32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_u8)))
+uint8x16_t vsriq_m_n_u8(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_m_n_u8)))
+uint8x16_t vsriq_m(uint8x16_t, uint8x16_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_s16)))
+int16x8_t vsriq_n_s16(int16x8_t, int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_s16)))
+int16x8_t vsriq(int16x8_t, int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_s32)))
+int32x4_t vsriq_n_s32(int32x4_t, int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_s32)))
+int32x4_t vsriq(int32x4_t, int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_s8)))
+int8x16_t vsriq_n_s8(int8x16_t, int8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_s8)))
+int8x16_t vsriq(int8x16_t, int8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_u16)))
+uint16x8_t vsriq_n_u16(uint16x8_t, uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_u16)))
+uint16x8_t vsriq(uint16x8_t, uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_u32)))
+uint32x4_t vsriq_n_u32(uint32x4_t, uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_u32)))
+uint32x4_t vsriq(uint32x4_t, uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_u8)))
+uint8x16_t vsriq_n_u8(uint8x16_t, uint8x16_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsriq_n_u8)))
+uint8x16_t vsriq(uint8x16_t, uint8x16_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_s16)))
+void vst1q_p_s16(int16_t *, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_s16)))
+void vst1q_p(int16_t *, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_s32)))
+void vst1q_p_s32(int32_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_s32)))
+void vst1q_p(int32_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_s8)))
+void vst1q_p_s8(int8_t *, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_s8)))
+void vst1q_p(int8_t *, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_u16)))
+void vst1q_p_u16(uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_u16)))
+void vst1q_p(uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_u32)))
+void vst1q_p_u32(uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_u32)))
+void vst1q_p(uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_u8)))
+void vst1q_p_u8(uint8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_u8)))
+void vst1q_p(uint8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_s16)))
+void vst1q_s16(int16_t *, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_s16)))
+void vst1q(int16_t *, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_s32)))
+void vst1q_s32(int32_t *, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_s32)))
+void vst1q(int32_t *, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_s8)))
+void vst1q_s8(int8_t *, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_s8)))
+void vst1q(int8_t *, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_u16)))
+void vst1q_u16(uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_u16)))
+void vst1q(uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_u32)))
+void vst1q_u32(uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_u32)))
+void vst1q(uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_u8)))
+void vst1q_u8(uint8_t *, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_u8)))
+void vst1q(uint8_t *, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_s16)))
+void vst2q_s16(int16_t *, int16x8x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_s16)))
+void vst2q(int16_t *, int16x8x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_s32)))
+void vst2q_s32(int32_t *, int32x4x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_s32)))
+void vst2q(int32_t *, int32x4x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_s8)))
+void vst2q_s8(int8_t *, int8x16x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_s8)))
+void vst2q(int8_t *, int8x16x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_u16)))
+void vst2q_u16(uint16_t *, uint16x8x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_u16)))
+void vst2q(uint16_t *, uint16x8x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_u32)))
+void vst2q_u32(uint32_t *, uint32x4x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_u32)))
+void vst2q(uint32_t *, uint32x4x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_u8)))
+void vst2q_u8(uint8_t *, uint8x16x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_u8)))
+void vst2q(uint8_t *, uint8x16x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_s16)))
+void vst4q_s16(int16_t *, int16x8x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_s16)))
+void vst4q(int16_t *, int16x8x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_s32)))
+void vst4q_s32(int32_t *, int32x4x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_s32)))
+void vst4q(int32_t *, int32x4x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_s8)))
+void vst4q_s8(int8_t *, int8x16x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_s8)))
+void vst4q(int8_t *, int8x16x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_u16)))
+void vst4q_u16(uint16_t *, uint16x8x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_u16)))
+void vst4q(uint16_t *, uint16x8x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_u32)))
+void vst4q_u32(uint32_t *, uint32x4x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_u32)))
+void vst4q(uint32_t *, uint32x4x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_u8)))
+void vst4q_u8(uint8_t *, uint8x16x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_u8)))
+void vst4q(uint8_t *, uint8x16x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_s16)))
+void vstrbq_p_s16(int8_t *, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_s16)))
+void vstrbq_p(int8_t *, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_s32)))
+void vstrbq_p_s32(int8_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_s32)))
+void vstrbq_p(int8_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_s8)))
+void vstrbq_p_s8(int8_t *, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_s8)))
+void vstrbq_p(int8_t *, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_u16)))
+void vstrbq_p_u16(uint8_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_u16)))
+void vstrbq_p(uint8_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_u32)))
+void vstrbq_p_u32(uint8_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_u32)))
+void vstrbq_p(uint8_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_u8)))
+void vstrbq_p_u8(uint8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_p_u8)))
+void vstrbq_p(uint8_t *, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_s16)))
+void vstrbq_s16(int8_t *, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_s16)))
+void vstrbq(int8_t *, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_s32)))
+void vstrbq_s32(int8_t *, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_s32)))
+void vstrbq(int8_t *, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_s8)))
+void vstrbq_s8(int8_t *, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_s8)))
+void vstrbq(int8_t *, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_s16)))
+void vstrbq_scatter_offset_p_s16(int8_t *, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_s16)))
+void vstrbq_scatter_offset_p(int8_t *, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_s32)))
+void vstrbq_scatter_offset_p_s32(int8_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_s32)))
+void vstrbq_scatter_offset_p(int8_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_s8)))
+void vstrbq_scatter_offset_p_s8(int8_t *, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_s8)))
+void vstrbq_scatter_offset_p(int8_t *, uint8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_u16)))
+void vstrbq_scatter_offset_p_u16(uint8_t *, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_u16)))
+void vstrbq_scatter_offset_p(uint8_t *, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_u32)))
+void vstrbq_scatter_offset_p_u32(uint8_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_u32)))
+void vstrbq_scatter_offset_p(uint8_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_u8)))
+void vstrbq_scatter_offset_p_u8(uint8_t *, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_p_u8)))
+void vstrbq_scatter_offset_p(uint8_t *, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_s16)))
+void vstrbq_scatter_offset_s16(int8_t *, uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_s16)))
+void vstrbq_scatter_offset(int8_t *, uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_s32)))
+void vstrbq_scatter_offset_s32(int8_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_s32)))
+void vstrbq_scatter_offset(int8_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_s8)))
+void vstrbq_scatter_offset_s8(int8_t *, uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_s8)))
+void vstrbq_scatter_offset(int8_t *, uint8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_u16)))
+void vstrbq_scatter_offset_u16(uint8_t *, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_u16)))
+void vstrbq_scatter_offset(uint8_t *, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_u32)))
+void vstrbq_scatter_offset_u32(uint8_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_u32)))
+void vstrbq_scatter_offset(uint8_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_u8)))
+void vstrbq_scatter_offset_u8(uint8_t *, uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_scatter_offset_u8)))
+void vstrbq_scatter_offset(uint8_t *, uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_u16)))
+void vstrbq_u16(uint8_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_u16)))
+void vstrbq(uint8_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_u32)))
+void vstrbq_u32(uint8_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_u32)))
+void vstrbq(uint8_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_u8)))
+void vstrbq_u8(uint8_t *, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrbq_u8)))
+void vstrbq(uint8_t *, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_p_s64)))
+void vstrdq_scatter_base_p_s64(uint64x2_t, int, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_p_s64)))
+void vstrdq_scatter_base_p(uint64x2_t, int, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_p_u64)))
+void vstrdq_scatter_base_p_u64(uint64x2_t, int, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_p_u64)))
+void vstrdq_scatter_base_p(uint64x2_t, int, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_s64)))
+void vstrdq_scatter_base_s64(uint64x2_t, int, int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_s64)))
+void vstrdq_scatter_base(uint64x2_t, int, int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_u64)))
+void vstrdq_scatter_base_u64(uint64x2_t, int, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_u64)))
+void vstrdq_scatter_base(uint64x2_t, int, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_p_s64)))
+void vstrdq_scatter_base_wb_p_s64(uint64x2_t *, int, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_p_s64)))
+void vstrdq_scatter_base_wb_p(uint64x2_t *, int, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_p_u64)))
+void vstrdq_scatter_base_wb_p_u64(uint64x2_t *, int, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_p_u64)))
+void vstrdq_scatter_base_wb_p(uint64x2_t *, int, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_s64)))
+void vstrdq_scatter_base_wb_s64(uint64x2_t *, int, int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_s64)))
+void vstrdq_scatter_base_wb(uint64x2_t *, int, int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_u64)))
+void vstrdq_scatter_base_wb_u64(uint64x2_t *, int, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_base_wb_u64)))
+void vstrdq_scatter_base_wb(uint64x2_t *, int, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_p_s64)))
+void vstrdq_scatter_offset_p_s64(int64_t *, uint64x2_t, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_p_s64)))
+void vstrdq_scatter_offset_p(int64_t *, uint64x2_t, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_p_u64)))
+void vstrdq_scatter_offset_p_u64(uint64_t *, uint64x2_t, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_p_u64)))
+void vstrdq_scatter_offset_p(uint64_t *, uint64x2_t, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_s64)))
+void vstrdq_scatter_offset_s64(int64_t *, uint64x2_t, int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_s64)))
+void vstrdq_scatter_offset(int64_t *, uint64x2_t, int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_u64)))
+void vstrdq_scatter_offset_u64(uint64_t *, uint64x2_t, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_offset_u64)))
+void vstrdq_scatter_offset(uint64_t *, uint64x2_t, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_p_s64)))
+void vstrdq_scatter_shifted_offset_p_s64(int64_t *, uint64x2_t, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_p_s64)))
+void vstrdq_scatter_shifted_offset_p(int64_t *, uint64x2_t, int64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_p_u64)))
+void vstrdq_scatter_shifted_offset_p_u64(uint64_t *, uint64x2_t, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_p_u64)))
+void vstrdq_scatter_shifted_offset_p(uint64_t *, uint64x2_t, uint64x2_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_s64)))
+void vstrdq_scatter_shifted_offset_s64(int64_t *, uint64x2_t, int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_s64)))
+void vstrdq_scatter_shifted_offset(int64_t *, uint64x2_t, int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_u64)))
+void vstrdq_scatter_shifted_offset_u64(uint64_t *, uint64x2_t, uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrdq_scatter_shifted_offset_u64)))
+void vstrdq_scatter_shifted_offset(uint64_t *, uint64x2_t, uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_s16)))
+void vstrhq_p_s16(int16_t *, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_s16)))
+void vstrhq_p(int16_t *, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_s32)))
+void vstrhq_p_s32(int16_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_s32)))
+void vstrhq_p(int16_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_u16)))
+void vstrhq_p_u16(uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_u16)))
+void vstrhq_p(uint16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_u32)))
+void vstrhq_p_u32(uint16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_u32)))
+void vstrhq_p(uint16_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_s16)))
+void vstrhq_s16(int16_t *, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_s16)))
+void vstrhq(int16_t *, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_s32)))
+void vstrhq_s32(int16_t *, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_s32)))
+void vstrhq(int16_t *, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_s16)))
+void vstrhq_scatter_offset_p_s16(int16_t *, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_s16)))
+void vstrhq_scatter_offset_p(int16_t *, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_s32)))
+void vstrhq_scatter_offset_p_s32(int16_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_s32)))
+void vstrhq_scatter_offset_p(int16_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_u16)))
+void vstrhq_scatter_offset_p_u16(uint16_t *, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_u16)))
+void vstrhq_scatter_offset_p(uint16_t *, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_u32)))
+void vstrhq_scatter_offset_p_u32(uint16_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_u32)))
+void vstrhq_scatter_offset_p(uint16_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_s16)))
+void vstrhq_scatter_offset_s16(int16_t *, uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_s16)))
+void vstrhq_scatter_offset(int16_t *, uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_s32)))
+void vstrhq_scatter_offset_s32(int16_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_s32)))
+void vstrhq_scatter_offset(int16_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_u16)))
+void vstrhq_scatter_offset_u16(uint16_t *, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_u16)))
+void vstrhq_scatter_offset(uint16_t *, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_u32)))
+void vstrhq_scatter_offset_u32(uint16_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_u32)))
+void vstrhq_scatter_offset(uint16_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_s16)))
+void vstrhq_scatter_shifted_offset_p_s16(int16_t *, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_s16)))
+void vstrhq_scatter_shifted_offset_p(int16_t *, uint16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_s32)))
+void vstrhq_scatter_shifted_offset_p_s32(int16_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_s32)))
+void vstrhq_scatter_shifted_offset_p(int16_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_u16)))
+void vstrhq_scatter_shifted_offset_p_u16(uint16_t *, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_u16)))
+void vstrhq_scatter_shifted_offset_p(uint16_t *, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_u32)))
+void vstrhq_scatter_shifted_offset_p_u32(uint16_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_u32)))
+void vstrhq_scatter_shifted_offset_p(uint16_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_s16)))
+void vstrhq_scatter_shifted_offset_s16(int16_t *, uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_s16)))
+void vstrhq_scatter_shifted_offset(int16_t *, uint16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_s32)))
+void vstrhq_scatter_shifted_offset_s32(int16_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_s32)))
+void vstrhq_scatter_shifted_offset(int16_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_u16)))
+void vstrhq_scatter_shifted_offset_u16(uint16_t *, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_u16)))
+void vstrhq_scatter_shifted_offset(uint16_t *, uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_u32)))
+void vstrhq_scatter_shifted_offset_u32(uint16_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_u32)))
+void vstrhq_scatter_shifted_offset(uint16_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_u16)))
+void vstrhq_u16(uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_u16)))
+void vstrhq(uint16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_u32)))
+void vstrhq_u32(uint16_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_u32)))
+void vstrhq(uint16_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_p_s32)))
+void vstrwq_p_s32(int32_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_p_s32)))
+void vstrwq_p(int32_t *, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_p_u32)))
+void vstrwq_p_u32(uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_p_u32)))
+void vstrwq_p(uint32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_s32)))
+void vstrwq_s32(int32_t *, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_s32)))
+void vstrwq(int32_t *, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_p_s32)))
+void vstrwq_scatter_base_p_s32(uint32x4_t, int, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_p_s32)))
+void vstrwq_scatter_base_p(uint32x4_t, int, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_p_u32)))
+void vstrwq_scatter_base_p_u32(uint32x4_t, int, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_p_u32)))
+void vstrwq_scatter_base_p(uint32x4_t, int, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_s32)))
+void vstrwq_scatter_base_s32(uint32x4_t, int, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_s32)))
+void vstrwq_scatter_base(uint32x4_t, int, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_u32)))
+void vstrwq_scatter_base_u32(uint32x4_t, int, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_u32)))
+void vstrwq_scatter_base(uint32x4_t, int, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_p_s32)))
+void vstrwq_scatter_base_wb_p_s32(uint32x4_t *, int, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_p_s32)))
+void vstrwq_scatter_base_wb_p(uint32x4_t *, int, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_p_u32)))
+void vstrwq_scatter_base_wb_p_u32(uint32x4_t *, int, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_p_u32)))
+void vstrwq_scatter_base_wb_p(uint32x4_t *, int, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_s32)))
+void vstrwq_scatter_base_wb_s32(uint32x4_t *, int, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_s32)))
+void vstrwq_scatter_base_wb(uint32x4_t *, int, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_u32)))
+void vstrwq_scatter_base_wb_u32(uint32x4_t *, int, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_u32)))
+void vstrwq_scatter_base_wb(uint32x4_t *, int, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_p_s32)))
+void vstrwq_scatter_offset_p_s32(int32_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_p_s32)))
+void vstrwq_scatter_offset_p(int32_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_p_u32)))
+void vstrwq_scatter_offset_p_u32(uint32_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_p_u32)))
+void vstrwq_scatter_offset_p(uint32_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_s32)))
+void vstrwq_scatter_offset_s32(int32_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_s32)))
+void vstrwq_scatter_offset(int32_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_u32)))
+void vstrwq_scatter_offset_u32(uint32_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_u32)))
+void vstrwq_scatter_offset(uint32_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_p_s32)))
+void vstrwq_scatter_shifted_offset_p_s32(int32_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_p_s32)))
+void vstrwq_scatter_shifted_offset_p(int32_t *, uint32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_p_u32)))
+void vstrwq_scatter_shifted_offset_p_u32(uint32_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_p_u32)))
+void vstrwq_scatter_shifted_offset_p(uint32_t *, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_s32)))
+void vstrwq_scatter_shifted_offset_s32(int32_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_s32)))
+void vstrwq_scatter_shifted_offset(int32_t *, uint32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_u32)))
+void vstrwq_scatter_shifted_offset_u32(uint32_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_u32)))
+void vstrwq_scatter_shifted_offset(uint32_t *, uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_u32)))
+void vstrwq_u32(uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_u32)))
+void vstrwq(uint32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_s16)))
+int16x8_t vsubq_m_n_s16(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_s16)))
+int16x8_t vsubq_m(int16x8_t, int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_s32)))
+int32x4_t vsubq_m_n_s32(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_s32)))
+int32x4_t vsubq_m(int32x4_t, int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_s8)))
+int8x16_t vsubq_m_n_s8(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_s8)))
+int8x16_t vsubq_m(int8x16_t, int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_u16)))
+uint16x8_t vsubq_m_n_u16(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_u16)))
+uint16x8_t vsubq_m(uint16x8_t, uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_u32)))
+uint32x4_t vsubq_m_n_u32(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_u32)))
+uint32x4_t vsubq_m(uint32x4_t, uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_u8)))
+uint8x16_t vsubq_m_n_u8(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_u8)))
+uint8x16_t vsubq_m(uint8x16_t, uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_s16)))
+int16x8_t vsubq_m_s16(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_s16)))
+int16x8_t vsubq_m(int16x8_t, int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_s32)))
+int32x4_t vsubq_m_s32(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_s32)))
+int32x4_t vsubq_m(int32x4_t, int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_s8)))
+int8x16_t vsubq_m_s8(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_s8)))
+int8x16_t vsubq_m(int8x16_t, int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_u16)))
+uint16x8_t vsubq_m_u16(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_u16)))
+uint16x8_t vsubq_m(uint16x8_t, uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_u32)))
+uint32x4_t vsubq_m_u32(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_u32)))
+uint32x4_t vsubq_m(uint32x4_t, uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_u8)))
+uint8x16_t vsubq_m_u8(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_u8)))
+uint8x16_t vsubq_m(uint8x16_t, uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_s16)))
+int16x8_t vsubq_n_s16(int16x8_t, int16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_s16)))
+int16x8_t vsubq(int16x8_t, int16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_s32)))
+int32x4_t vsubq_n_s32(int32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_s32)))
+int32x4_t vsubq(int32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_s8)))
+int8x16_t vsubq_n_s8(int8x16_t, int8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_s8)))
+int8x16_t vsubq(int8x16_t, int8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_u16)))
+uint16x8_t vsubq_n_u16(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_u16)))
+uint16x8_t vsubq(uint16x8_t, uint16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_u32)))
+uint32x4_t vsubq_n_u32(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_u32)))
+uint32x4_t vsubq(uint32x4_t, uint32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_u8)))
+uint8x16_t vsubq_n_u8(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_u8)))
+uint8x16_t vsubq(uint8x16_t, uint8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_s16)))
+int16x8_t vsubq_s16(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_s16)))
+int16x8_t vsubq(int16x8_t, int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_s32)))
+int32x4_t vsubq_s32(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_s32)))
+int32x4_t vsubq(int32x4_t, int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_s8)))
+int8x16_t vsubq_s8(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_s8)))
+int8x16_t vsubq(int8x16_t, int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_u16)))
+uint16x8_t vsubq_u16(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_u16)))
+uint16x8_t vsubq(uint16x8_t, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_u32)))
+uint32x4_t vsubq_u32(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_u32)))
+uint32x4_t vsubq(uint32x4_t, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_u8)))
+uint8x16_t vsubq_u8(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_u8)))
+uint8x16_t vsubq(uint8x16_t, uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_s16)))
+int16x8_t vsubq_x_n_s16(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_s16)))
+int16x8_t vsubq_x(int16x8_t, int16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_s32)))
+int32x4_t vsubq_x_n_s32(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_s32)))
+int32x4_t vsubq_x(int32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_s8)))
+int8x16_t vsubq_x_n_s8(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_s8)))
+int8x16_t vsubq_x(int8x16_t, int8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_u16)))
+uint16x8_t vsubq_x_n_u16(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_u16)))
+uint16x8_t vsubq_x(uint16x8_t, uint16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_u32)))
+uint32x4_t vsubq_x_n_u32(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_u32)))
+uint32x4_t vsubq_x(uint32x4_t, uint32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_u8)))
+uint8x16_t vsubq_x_n_u8(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_u8)))
+uint8x16_t vsubq_x(uint8x16_t, uint8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_s16)))
+int16x8_t vsubq_x_s16(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_s16)))
+int16x8_t vsubq_x(int16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_s32)))
+int32x4_t vsubq_x_s32(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_s32)))
+int32x4_t vsubq_x(int32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_s8)))
+int8x16_t vsubq_x_s8(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_s8)))
+int8x16_t vsubq_x(int8x16_t, int8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_u16)))
+uint16x8_t vsubq_x_u16(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_u16)))
+uint16x8_t vsubq_x(uint16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_u32)))
+uint32x4_t vsubq_x_u32(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_u32)))
+uint32x4_t vsubq_x(uint32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_u8)))
+uint8x16_t vsubq_x_u8(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_u8)))
+uint8x16_t vsubq_x(uint8x16_t, uint8x16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_s16)))
+int16x8_t vuninitializedq(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_s32)))
+int32x4_t vuninitializedq(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_s64)))
+int64x2_t vuninitializedq(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_s8)))
+int8x16_t vuninitializedq(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_u16)))
+uint16x8_t vuninitializedq(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_u32)))
+uint32x4_t vuninitializedq(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_u64)))
+uint64x2_t vuninitializedq(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_u8)))
+uint8x16_t vuninitializedq(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_s16)))
+int16x8_t vuninitializedq_s16();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_s32)))
+int32x4_t vuninitializedq_s32();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_s64)))
+int64x2_t vuninitializedq_s64();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_s8)))
+int8x16_t vuninitializedq_s8();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_u16)))
+uint16x8_t vuninitializedq_u16();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_u32)))
+uint32x4_t vuninitializedq_u32();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_u64)))
+uint64x2_t vuninitializedq_u64();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_u8)))
+uint8x16_t vuninitializedq_u8();
+
+#endif /* (!defined __ARM_MVE_PRESERVE_USER_NAMESPACE) */
+
+#if (__ARM_FEATURE_MVE & 2) && (!defined __ARM_MVE_PRESERVE_USER_NAMESPACE)
+
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_f16)))
+float16x8_t vabdq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_f16)))
+float16x8_t vabdq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_f32)))
+float32x4_t vabdq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_f32)))
+float32x4_t vabdq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_f16)))
+float16x8_t vabdq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_f16)))
+float16x8_t vabdq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_f32)))
+float32x4_t vabdq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_m_f32)))
+float32x4_t vabdq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_f16)))
+float16x8_t vabdq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_f16)))
+float16x8_t vabdq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_f32)))
+float32x4_t vabdq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabdq_x_f32)))
+float32x4_t vabdq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_f16)))
+float16x8_t vabsq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_f16)))
+float16x8_t vabsq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_f32)))
+float32x4_t vabsq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_f32)))
+float32x4_t vabsq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_f16)))
+float16x8_t vabsq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_f16)))
+float16x8_t vabsq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_f32)))
+float32x4_t vabsq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_m_f32)))
+float32x4_t vabsq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_f16)))
+float16x8_t vabsq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_f16)))
+float16x8_t vabsq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_f32)))
+float32x4_t vabsq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vabsq_x_f32)))
+float32x4_t vabsq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_f16)))
+float16x8_t vaddq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_f16)))
+float16x8_t vaddq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_f32)))
+float32x4_t vaddq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_f32)))
+float32x4_t vaddq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_f16)))
+float16x8_t vaddq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_f16)))
+float16x8_t vaddq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_f32)))
+float32x4_t vaddq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_f32)))
+float32x4_t vaddq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_f16)))
+float16x8_t vaddq_m_n_f16(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_f16)))
+float16x8_t vaddq_m(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_f32)))
+float32x4_t vaddq_m_n_f32(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_m_n_f32)))
+float32x4_t vaddq_m(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_f16)))
+float16x8_t vaddq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_f16)))
+float16x8_t vaddq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_f32)))
+float32x4_t vaddq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_n_f32)))
+float32x4_t vaddq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_f16)))
+float16x8_t vaddq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_f16)))
+float16x8_t vaddq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_f32)))
+float32x4_t vaddq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_f32)))
+float32x4_t vaddq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_f16)))
+float16x8_t vaddq_x_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_f16)))
+float16x8_t vaddq_x(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_f32)))
+float32x4_t vaddq_x_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vaddq_x_n_f32)))
+float32x4_t vaddq_x(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_f16)))
+float16x8_t vandq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_f16)))
+float16x8_t vandq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_f32)))
+float32x4_t vandq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_f32)))
+float32x4_t vandq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_f16)))
+float16x8_t vandq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_f16)))
+float16x8_t vandq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_f32)))
+float32x4_t vandq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_m_f32)))
+float32x4_t vandq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_f16)))
+float16x8_t vandq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_f16)))
+float16x8_t vandq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_f32)))
+float32x4_t vandq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vandq_x_f32)))
+float32x4_t vandq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_f16)))
+float16x8_t vbicq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_f16)))
+float16x8_t vbicq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_f32)))
+float32x4_t vbicq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_f32)))
+float32x4_t vbicq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_f16)))
+float16x8_t vbicq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_f16)))
+float16x8_t vbicq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_f32)))
+float32x4_t vbicq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_m_f32)))
+float32x4_t vbicq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_f16)))
+float16x8_t vbicq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_f16)))
+float16x8_t vbicq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_f32)))
+float32x4_t vbicq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbicq_x_f32)))
+float32x4_t vbicq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_f16)))
+float16x8_t vbrsrq_m_n_f16(float16x8_t, float16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_f16)))
+float16x8_t vbrsrq_m(float16x8_t, float16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_f32)))
+float32x4_t vbrsrq_m_n_f32(float32x4_t, float32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_m_n_f32)))
+float32x4_t vbrsrq_m(float32x4_t, float32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_f16)))
+float16x8_t vbrsrq_n_f16(float16x8_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_f16)))
+float16x8_t vbrsrq(float16x8_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_f32)))
+float32x4_t vbrsrq_n_f32(float32x4_t, int32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_n_f32)))
+float32x4_t vbrsrq(float32x4_t, int32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_f16)))
+float16x8_t vbrsrq_x_n_f16(float16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_f16)))
+float16x8_t vbrsrq_x(float16x8_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_f32)))
+float32x4_t vbrsrq_x_n_f32(float32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vbrsrq_x_n_f32)))
+float32x4_t vbrsrq_x(float32x4_t, int32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_f16)))
+float16x8_t vcaddq_rot270_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_f16)))
+float16x8_t vcaddq_rot270(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_f32)))
+float32x4_t vcaddq_rot270_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_f32)))
+float32x4_t vcaddq_rot270(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_f16)))
+float16x8_t vcaddq_rot270_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_f16)))
+float16x8_t vcaddq_rot270_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_f32)))
+float32x4_t vcaddq_rot270_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_m_f32)))
+float32x4_t vcaddq_rot270_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_f16)))
+float16x8_t vcaddq_rot270_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_f16)))
+float16x8_t vcaddq_rot270_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_f32)))
+float32x4_t vcaddq_rot270_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot270_x_f32)))
+float32x4_t vcaddq_rot270_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_f16)))
+float16x8_t vcaddq_rot90_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_f16)))
+float16x8_t vcaddq_rot90(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_f32)))
+float32x4_t vcaddq_rot90_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_f32)))
+float32x4_t vcaddq_rot90(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_f16)))
+float16x8_t vcaddq_rot90_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_f16)))
+float16x8_t vcaddq_rot90_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_f32)))
+float32x4_t vcaddq_rot90_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_m_f32)))
+float32x4_t vcaddq_rot90_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_f16)))
+float16x8_t vcaddq_rot90_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_f16)))
+float16x8_t vcaddq_rot90_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_f32)))
+float32x4_t vcaddq_rot90_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcaddq_rot90_x_f32)))
+float32x4_t vcaddq_rot90_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_f16)))
+float16x8_t vcmlaq_f16(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_f16)))
+float16x8_t vcmlaq(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_f32)))
+float32x4_t vcmlaq_f32(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_f32)))
+float32x4_t vcmlaq(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_m_f16)))
+float16x8_t vcmlaq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_m_f16)))
+float16x8_t vcmlaq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_m_f32)))
+float32x4_t vcmlaq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_m_f32)))
+float32x4_t vcmlaq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_f16)))
+float16x8_t vcmlaq_rot180_f16(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_f16)))
+float16x8_t vcmlaq_rot180(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_f32)))
+float32x4_t vcmlaq_rot180_f32(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_f32)))
+float32x4_t vcmlaq_rot180(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_m_f16)))
+float16x8_t vcmlaq_rot180_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_m_f16)))
+float16x8_t vcmlaq_rot180_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_m_f32)))
+float32x4_t vcmlaq_rot180_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot180_m_f32)))
+float32x4_t vcmlaq_rot180_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_f16)))
+float16x8_t vcmlaq_rot270_f16(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_f16)))
+float16x8_t vcmlaq_rot270(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_f32)))
+float32x4_t vcmlaq_rot270_f32(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_f32)))
+float32x4_t vcmlaq_rot270(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_m_f16)))
+float16x8_t vcmlaq_rot270_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_m_f16)))
+float16x8_t vcmlaq_rot270_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_m_f32)))
+float32x4_t vcmlaq_rot270_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot270_m_f32)))
+float32x4_t vcmlaq_rot270_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_f16)))
+float16x8_t vcmlaq_rot90_f16(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_f16)))
+float16x8_t vcmlaq_rot90(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_f32)))
+float32x4_t vcmlaq_rot90_f32(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_f32)))
+float32x4_t vcmlaq_rot90(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_m_f16)))
+float16x8_t vcmlaq_rot90_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_m_f16)))
+float16x8_t vcmlaq_rot90_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_m_f32)))
+float32x4_t vcmlaq_rot90_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmlaq_rot90_m_f32)))
+float32x4_t vcmlaq_rot90_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_f16)))
+mve_pred16_t vcmpeqq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_f16)))
+mve_pred16_t vcmpeqq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_f32)))
+mve_pred16_t vcmpeqq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_f32)))
+mve_pred16_t vcmpeqq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_f16)))
+mve_pred16_t vcmpeqq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_f16)))
+mve_pred16_t vcmpeqq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_f32)))
+mve_pred16_t vcmpeqq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_f32)))
+mve_pred16_t vcmpeqq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_f16)))
+mve_pred16_t vcmpeqq_m_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_f16)))
+mve_pred16_t vcmpeqq_m(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_f32)))
+mve_pred16_t vcmpeqq_m_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_m_n_f32)))
+mve_pred16_t vcmpeqq_m(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_f16)))
+mve_pred16_t vcmpeqq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_f16)))
+mve_pred16_t vcmpeqq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_f32)))
+mve_pred16_t vcmpeqq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpeqq_n_f32)))
+mve_pred16_t vcmpeqq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_f16)))
+mve_pred16_t vcmpgeq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_f16)))
+mve_pred16_t vcmpgeq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_f32)))
+mve_pred16_t vcmpgeq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_f32)))
+mve_pred16_t vcmpgeq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_f16)))
+mve_pred16_t vcmpgeq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_f16)))
+mve_pred16_t vcmpgeq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_f32)))
+mve_pred16_t vcmpgeq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_f32)))
+mve_pred16_t vcmpgeq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_f16)))
+mve_pred16_t vcmpgeq_m_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_f16)))
+mve_pred16_t vcmpgeq_m(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_f32)))
+mve_pred16_t vcmpgeq_m_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_m_n_f32)))
+mve_pred16_t vcmpgeq_m(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_f16)))
+mve_pred16_t vcmpgeq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_f16)))
+mve_pred16_t vcmpgeq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_f32)))
+mve_pred16_t vcmpgeq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgeq_n_f32)))
+mve_pred16_t vcmpgeq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_f16)))
+mve_pred16_t vcmpgtq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_f16)))
+mve_pred16_t vcmpgtq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_f32)))
+mve_pred16_t vcmpgtq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_f32)))
+mve_pred16_t vcmpgtq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_f16)))
+mve_pred16_t vcmpgtq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_f16)))
+mve_pred16_t vcmpgtq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_f32)))
+mve_pred16_t vcmpgtq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_f32)))
+mve_pred16_t vcmpgtq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_f16)))
+mve_pred16_t vcmpgtq_m_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_f16)))
+mve_pred16_t vcmpgtq_m(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_f32)))
+mve_pred16_t vcmpgtq_m_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_m_n_f32)))
+mve_pred16_t vcmpgtq_m(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_f16)))
+mve_pred16_t vcmpgtq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_f16)))
+mve_pred16_t vcmpgtq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_f32)))
+mve_pred16_t vcmpgtq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpgtq_n_f32)))
+mve_pred16_t vcmpgtq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_f16)))
+mve_pred16_t vcmpleq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_f16)))
+mve_pred16_t vcmpleq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_f32)))
+mve_pred16_t vcmpleq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_f32)))
+mve_pred16_t vcmpleq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_f16)))
+mve_pred16_t vcmpleq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_f16)))
+mve_pred16_t vcmpleq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_f32)))
+mve_pred16_t vcmpleq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_f32)))
+mve_pred16_t vcmpleq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_f16)))
+mve_pred16_t vcmpleq_m_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_f16)))
+mve_pred16_t vcmpleq_m(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_f32)))
+mve_pred16_t vcmpleq_m_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_m_n_f32)))
+mve_pred16_t vcmpleq_m(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_f16)))
+mve_pred16_t vcmpleq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_f16)))
+mve_pred16_t vcmpleq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_f32)))
+mve_pred16_t vcmpleq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpleq_n_f32)))
+mve_pred16_t vcmpleq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_f16)))
+mve_pred16_t vcmpltq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_f16)))
+mve_pred16_t vcmpltq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_f32)))
+mve_pred16_t vcmpltq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_f32)))
+mve_pred16_t vcmpltq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_f16)))
+mve_pred16_t vcmpltq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_f16)))
+mve_pred16_t vcmpltq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_f32)))
+mve_pred16_t vcmpltq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_f32)))
+mve_pred16_t vcmpltq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_f16)))
+mve_pred16_t vcmpltq_m_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_f16)))
+mve_pred16_t vcmpltq_m(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_f32)))
+mve_pred16_t vcmpltq_m_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_m_n_f32)))
+mve_pred16_t vcmpltq_m(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_f16)))
+mve_pred16_t vcmpltq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_f16)))
+mve_pred16_t vcmpltq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_f32)))
+mve_pred16_t vcmpltq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpltq_n_f32)))
+mve_pred16_t vcmpltq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_f16)))
+mve_pred16_t vcmpneq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_f16)))
+mve_pred16_t vcmpneq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_f32)))
+mve_pred16_t vcmpneq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_f32)))
+mve_pred16_t vcmpneq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_f16)))
+mve_pred16_t vcmpneq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_f16)))
+mve_pred16_t vcmpneq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_f32)))
+mve_pred16_t vcmpneq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_f32)))
+mve_pred16_t vcmpneq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_f16)))
+mve_pred16_t vcmpneq_m_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_f16)))
+mve_pred16_t vcmpneq_m(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_f32)))
+mve_pred16_t vcmpneq_m_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_m_n_f32)))
+mve_pred16_t vcmpneq_m(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_f16)))
+mve_pred16_t vcmpneq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_f16)))
+mve_pred16_t vcmpneq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_f32)))
+mve_pred16_t vcmpneq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmpneq_n_f32)))
+mve_pred16_t vcmpneq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_f16)))
+float16x8_t vcmulq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_f16)))
+float16x8_t vcmulq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_f32)))
+float32x4_t vcmulq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_f32)))
+float32x4_t vcmulq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_m_f16)))
+float16x8_t vcmulq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_m_f16)))
+float16x8_t vcmulq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_m_f32)))
+float32x4_t vcmulq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_m_f32)))
+float32x4_t vcmulq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_f16)))
+float16x8_t vcmulq_rot180_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_f16)))
+float16x8_t vcmulq_rot180(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_f32)))
+float32x4_t vcmulq_rot180_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_f32)))
+float32x4_t vcmulq_rot180(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_m_f16)))
+float16x8_t vcmulq_rot180_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_m_f16)))
+float16x8_t vcmulq_rot180_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_m_f32)))
+float32x4_t vcmulq_rot180_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_m_f32)))
+float32x4_t vcmulq_rot180_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_x_f16)))
+float16x8_t vcmulq_rot180_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_x_f16)))
+float16x8_t vcmulq_rot180_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_x_f32)))
+float32x4_t vcmulq_rot180_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot180_x_f32)))
+float32x4_t vcmulq_rot180_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_f16)))
+float16x8_t vcmulq_rot270_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_f16)))
+float16x8_t vcmulq_rot270(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_f32)))
+float32x4_t vcmulq_rot270_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_f32)))
+float32x4_t vcmulq_rot270(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_m_f16)))
+float16x8_t vcmulq_rot270_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_m_f16)))
+float16x8_t vcmulq_rot270_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_m_f32)))
+float32x4_t vcmulq_rot270_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_m_f32)))
+float32x4_t vcmulq_rot270_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_x_f16)))
+float16x8_t vcmulq_rot270_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_x_f16)))
+float16x8_t vcmulq_rot270_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_x_f32)))
+float32x4_t vcmulq_rot270_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot270_x_f32)))
+float32x4_t vcmulq_rot270_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_f16)))
+float16x8_t vcmulq_rot90_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_f16)))
+float16x8_t vcmulq_rot90(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_f32)))
+float32x4_t vcmulq_rot90_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_f32)))
+float32x4_t vcmulq_rot90(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_m_f16)))
+float16x8_t vcmulq_rot90_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_m_f16)))
+float16x8_t vcmulq_rot90_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_m_f32)))
+float32x4_t vcmulq_rot90_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_m_f32)))
+float32x4_t vcmulq_rot90_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_x_f16)))
+float16x8_t vcmulq_rot90_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_x_f16)))
+float16x8_t vcmulq_rot90_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_x_f32)))
+float32x4_t vcmulq_rot90_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_rot90_x_f32)))
+float32x4_t vcmulq_rot90_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_x_f16)))
+float16x8_t vcmulq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_x_f16)))
+float16x8_t vcmulq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_x_f32)))
+float32x4_t vcmulq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcmulq_x_f32)))
+float32x4_t vcmulq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_f16)))
+float16x8_t vcreateq_f16(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcreateq_f32)))
+float32x4_t vcreateq_f32(uint64_t, uint64_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_s16_f16)))
+int16x8_t vcvtaq_m_s16_f16(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_s16_f16)))
+int16x8_t vcvtaq_m(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_s32_f32)))
+int32x4_t vcvtaq_m_s32_f32(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_s32_f32)))
+int32x4_t vcvtaq_m(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_u16_f16)))
+uint16x8_t vcvtaq_m_u16_f16(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_u16_f16)))
+uint16x8_t vcvtaq_m(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_u32_f32)))
+uint32x4_t vcvtaq_m_u32_f32(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_m_u32_f32)))
+uint32x4_t vcvtaq_m(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_s16_f16)))
+int16x8_t vcvtaq_s16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_s32_f32)))
+int32x4_t vcvtaq_s32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_u16_f16)))
+uint16x8_t vcvtaq_u16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_u32_f32)))
+uint32x4_t vcvtaq_u32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_x_s16_f16)))
+int16x8_t vcvtaq_x_s16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_x_s32_f32)))
+int32x4_t vcvtaq_x_s32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_x_u16_f16)))
+uint16x8_t vcvtaq_x_u16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtaq_x_u32_f32)))
+uint32x4_t vcvtaq_x_u32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtbq_f16_f32)))
+float16x8_t vcvtbq_f16_f32(float16x8_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtbq_f32_f16)))
+float32x4_t vcvtbq_f32_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtbq_m_f16_f32)))
+float16x8_t vcvtbq_m_f16_f32(float16x8_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtbq_m_f32_f16)))
+float32x4_t vcvtbq_m_f32_f16(float32x4_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtbq_x_f32_f16)))
+float32x4_t vcvtbq_x_f32_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_s16_f16)))
+int16x8_t vcvtmq_m_s16_f16(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_s16_f16)))
+int16x8_t vcvtmq_m(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_s32_f32)))
+int32x4_t vcvtmq_m_s32_f32(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_s32_f32)))
+int32x4_t vcvtmq_m(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_u16_f16)))
+uint16x8_t vcvtmq_m_u16_f16(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_u16_f16)))
+uint16x8_t vcvtmq_m(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_u32_f32)))
+uint32x4_t vcvtmq_m_u32_f32(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_m_u32_f32)))
+uint32x4_t vcvtmq_m(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_s16_f16)))
+int16x8_t vcvtmq_s16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_s32_f32)))
+int32x4_t vcvtmq_s32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_u16_f16)))
+uint16x8_t vcvtmq_u16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_u32_f32)))
+uint32x4_t vcvtmq_u32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_x_s16_f16)))
+int16x8_t vcvtmq_x_s16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_x_s32_f32)))
+int32x4_t vcvtmq_x_s32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_x_u16_f16)))
+uint16x8_t vcvtmq_x_u16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtmq_x_u32_f32)))
+uint32x4_t vcvtmq_x_u32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_s16_f16)))
+int16x8_t vcvtnq_m_s16_f16(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_s16_f16)))
+int16x8_t vcvtnq_m(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_s32_f32)))
+int32x4_t vcvtnq_m_s32_f32(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_s32_f32)))
+int32x4_t vcvtnq_m(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_u16_f16)))
+uint16x8_t vcvtnq_m_u16_f16(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_u16_f16)))
+uint16x8_t vcvtnq_m(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_u32_f32)))
+uint32x4_t vcvtnq_m_u32_f32(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_m_u32_f32)))
+uint32x4_t vcvtnq_m(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_s16_f16)))
+int16x8_t vcvtnq_s16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_s32_f32)))
+int32x4_t vcvtnq_s32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_u16_f16)))
+uint16x8_t vcvtnq_u16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_u32_f32)))
+uint32x4_t vcvtnq_u32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_x_s16_f16)))
+int16x8_t vcvtnq_x_s16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_x_s32_f32)))
+int32x4_t vcvtnq_x_s32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_x_u16_f16)))
+uint16x8_t vcvtnq_x_u16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtnq_x_u32_f32)))
+uint32x4_t vcvtnq_x_u32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_s16_f16)))
+int16x8_t vcvtpq_m_s16_f16(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_s16_f16)))
+int16x8_t vcvtpq_m(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_s32_f32)))
+int32x4_t vcvtpq_m_s32_f32(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_s32_f32)))
+int32x4_t vcvtpq_m(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_u16_f16)))
+uint16x8_t vcvtpq_m_u16_f16(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_u16_f16)))
+uint16x8_t vcvtpq_m(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_u32_f32)))
+uint32x4_t vcvtpq_m_u32_f32(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_m_u32_f32)))
+uint32x4_t vcvtpq_m(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_s16_f16)))
+int16x8_t vcvtpq_s16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_s32_f32)))
+int32x4_t vcvtpq_s32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_u16_f16)))
+uint16x8_t vcvtpq_u16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_u32_f32)))
+uint32x4_t vcvtpq_u32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_x_s16_f16)))
+int16x8_t vcvtpq_x_s16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_x_s32_f32)))
+int32x4_t vcvtpq_x_s32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_x_u16_f16)))
+uint16x8_t vcvtpq_x_u16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtpq_x_u32_f32)))
+uint32x4_t vcvtpq_x_u32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f16_s16)))
+float16x8_t vcvtq_f16_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f16_s16)))
+float16x8_t vcvtq(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f16_u16)))
+float16x8_t vcvtq_f16_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f16_u16)))
+float16x8_t vcvtq(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f32_s32)))
+float32x4_t vcvtq_f32_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f32_s32)))
+float32x4_t vcvtq(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f32_u32)))
+float32x4_t vcvtq_f32_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_f32_u32)))
+float32x4_t vcvtq(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f16_s16)))
+float16x8_t vcvtq_m_f16_s16(float16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f16_s16)))
+float16x8_t vcvtq_m(float16x8_t, int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f16_u16)))
+float16x8_t vcvtq_m_f16_u16(float16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f16_u16)))
+float16x8_t vcvtq_m(float16x8_t, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f32_s32)))
+float32x4_t vcvtq_m_f32_s32(float32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f32_s32)))
+float32x4_t vcvtq_m(float32x4_t, int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f32_u32)))
+float32x4_t vcvtq_m_f32_u32(float32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_f32_u32)))
+float32x4_t vcvtq_m(float32x4_t, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f16_s16)))
+float16x8_t vcvtq_m_n_f16_s16(float16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f16_s16)))
+float16x8_t vcvtq_m_n(float16x8_t, int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f16_u16)))
+float16x8_t vcvtq_m_n_f16_u16(float16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f16_u16)))
+float16x8_t vcvtq_m_n(float16x8_t, uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f32_s32)))
+float32x4_t vcvtq_m_n_f32_s32(float32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f32_s32)))
+float32x4_t vcvtq_m_n(float32x4_t, int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f32_u32)))
+float32x4_t vcvtq_m_n_f32_u32(float32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_f32_u32)))
+float32x4_t vcvtq_m_n(float32x4_t, uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_s16_f16)))
+int16x8_t vcvtq_m_n_s16_f16(int16x8_t, float16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_s16_f16)))
+int16x8_t vcvtq_m_n(int16x8_t, float16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_s32_f32)))
+int32x4_t vcvtq_m_n_s32_f32(int32x4_t, float32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_s32_f32)))
+int32x4_t vcvtq_m_n(int32x4_t, float32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_u16_f16)))
+uint16x8_t vcvtq_m_n_u16_f16(uint16x8_t, float16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_u16_f16)))
+uint16x8_t vcvtq_m_n(uint16x8_t, float16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_u32_f32)))
+uint32x4_t vcvtq_m_n_u32_f32(uint32x4_t, float32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_n_u32_f32)))
+uint32x4_t vcvtq_m_n(uint32x4_t, float32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_s16_f16)))
+int16x8_t vcvtq_m_s16_f16(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_s16_f16)))
+int16x8_t vcvtq_m(int16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_s32_f32)))
+int32x4_t vcvtq_m_s32_f32(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_s32_f32)))
+int32x4_t vcvtq_m(int32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_u16_f16)))
+uint16x8_t vcvtq_m_u16_f16(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_u16_f16)))
+uint16x8_t vcvtq_m(uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_u32_f32)))
+uint32x4_t vcvtq_m_u32_f32(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_m_u32_f32)))
+uint32x4_t vcvtq_m(uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f16_s16)))
+float16x8_t vcvtq_n_f16_s16(int16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f16_s16)))
+float16x8_t vcvtq_n(int16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f16_u16)))
+float16x8_t vcvtq_n_f16_u16(uint16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f16_u16)))
+float16x8_t vcvtq_n(uint16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f32_s32)))
+float32x4_t vcvtq_n_f32_s32(int32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f32_s32)))
+float32x4_t vcvtq_n(int32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f32_u32)))
+float32x4_t vcvtq_n_f32_u32(uint32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_f32_u32)))
+float32x4_t vcvtq_n(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_s16_f16)))
+int16x8_t vcvtq_n_s16_f16(float16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_s32_f32)))
+int32x4_t vcvtq_n_s32_f32(float32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_u16_f16)))
+uint16x8_t vcvtq_n_u16_f16(float16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_n_u32_f32)))
+uint32x4_t vcvtq_n_u32_f32(float32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_s16_f16)))
+int16x8_t vcvtq_s16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_s32_f32)))
+int32x4_t vcvtq_s32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_u16_f16)))
+uint16x8_t vcvtq_u16_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_u32_f32)))
+uint32x4_t vcvtq_u32_f32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f16_s16)))
+float16x8_t vcvtq_x_f16_s16(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f16_s16)))
+float16x8_t vcvtq_x(int16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f16_u16)))
+float16x8_t vcvtq_x_f16_u16(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f16_u16)))
+float16x8_t vcvtq_x(uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f32_s32)))
+float32x4_t vcvtq_x_f32_s32(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f32_s32)))
+float32x4_t vcvtq_x(int32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f32_u32)))
+float32x4_t vcvtq_x_f32_u32(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_f32_u32)))
+float32x4_t vcvtq_x(uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f16_s16)))
+float16x8_t vcvtq_x_n_f16_s16(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f16_s16)))
+float16x8_t vcvtq_x_n(int16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f16_u16)))
+float16x8_t vcvtq_x_n_f16_u16(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f16_u16)))
+float16x8_t vcvtq_x_n(uint16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f32_s32)))
+float32x4_t vcvtq_x_n_f32_s32(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f32_s32)))
+float32x4_t vcvtq_x_n(int32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f32_u32)))
+float32x4_t vcvtq_x_n_f32_u32(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_f32_u32)))
+float32x4_t vcvtq_x_n(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_s16_f16)))
+int16x8_t vcvtq_x_n_s16_f16(float16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_s32_f32)))
+int32x4_t vcvtq_x_n_s32_f32(float32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_u16_f16)))
+uint16x8_t vcvtq_x_n_u16_f16(float16x8_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_n_u32_f32)))
+uint32x4_t vcvtq_x_n_u32_f32(float32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_s16_f16)))
+int16x8_t vcvtq_x_s16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_s32_f32)))
+int32x4_t vcvtq_x_s32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_u16_f16)))
+uint16x8_t vcvtq_x_u16_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvtq_x_u32_f32)))
+uint32x4_t vcvtq_x_u32_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvttq_f16_f32)))
+float16x8_t vcvttq_f16_f32(float16x8_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvttq_f32_f16)))
+float32x4_t vcvttq_f32_f16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvttq_m_f16_f32)))
+float16x8_t vcvttq_m_f16_f32(float16x8_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvttq_m_f32_f16)))
+float32x4_t vcvttq_m_f32_f16(float32x4_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vcvttq_x_f32_f16)))
+float32x4_t vcvttq_x_f32_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_f16)))
+float16x8_t vdupq_m_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_f16)))
+float16x8_t vdupq_m(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_f32)))
+float32x4_t vdupq_m_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vdupq_m_n_f32)))
+float32x4_t vdupq_m(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_f16)))
+float16x8_t vdupq_n_f16(float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_n_f32)))
+float32x4_t vdupq_n_f32(float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_f16)))
+float16x8_t vdupq_x_n_f16(float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vdupq_x_n_f32)))
+float32x4_t vdupq_x_n_f32(float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_f16)))
+float16x8_t veorq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_f16)))
+float16x8_t veorq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_f32)))
+float32x4_t veorq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_f32)))
+float32x4_t veorq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_f16)))
+float16x8_t veorq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_f16)))
+float16x8_t veorq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_f32)))
+float32x4_t veorq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_m_f32)))
+float32x4_t veorq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_f16)))
+float16x8_t veorq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_f16)))
+float16x8_t veorq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_f32)))
+float32x4_t veorq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_veorq_x_f32)))
+float32x4_t veorq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_f16)))
+float16x8_t vfmaq_f16(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_f16)))
+float16x8_t vfmaq(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_f32)))
+float32x4_t vfmaq_f32(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_f32)))
+float32x4_t vfmaq(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_f16)))
+float16x8_t vfmaq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_f16)))
+float16x8_t vfmaq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_f32)))
+float32x4_t vfmaq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_f32)))
+float32x4_t vfmaq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_n_f16)))
+float16x8_t vfmaq_m_n_f16(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_n_f16)))
+float16x8_t vfmaq_m(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_n_f32)))
+float32x4_t vfmaq_m_n_f32(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_m_n_f32)))
+float32x4_t vfmaq_m(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_n_f16)))
+float16x8_t vfmaq_n_f16(float16x8_t, float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_n_f16)))
+float16x8_t vfmaq(float16x8_t, float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_n_f32)))
+float32x4_t vfmaq_n_f32(float32x4_t, float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmaq_n_f32)))
+float32x4_t vfmaq(float32x4_t, float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_m_n_f16)))
+float16x8_t vfmasq_m_n_f16(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_m_n_f16)))
+float16x8_t vfmasq_m(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_m_n_f32)))
+float32x4_t vfmasq_m_n_f32(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_m_n_f32)))
+float32x4_t vfmasq_m(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_n_f16)))
+float16x8_t vfmasq_n_f16(float16x8_t, float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_n_f16)))
+float16x8_t vfmasq(float16x8_t, float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_n_f32)))
+float32x4_t vfmasq_n_f32(float32x4_t, float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmasq_n_f32)))
+float32x4_t vfmasq(float32x4_t, float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_f16)))
+float16x8_t vfmsq_f16(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_f16)))
+float16x8_t vfmsq(float16x8_t, float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_f32)))
+float32x4_t vfmsq_f32(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_f32)))
+float32x4_t vfmsq(float32x4_t, float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_m_f16)))
+float16x8_t vfmsq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_m_f16)))
+float16x8_t vfmsq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_m_f32)))
+float32x4_t vfmsq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vfmsq_m_f32)))
+float32x4_t vfmsq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_f16)))
+float16_t vgetq_lane_f16(float16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_f16)))
+float16_t vgetq_lane(float16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_f32)))
+float32_t vgetq_lane_f32(float32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vgetq_lane_f32)))
+float32_t vgetq_lane(float32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_f16)))
+float16x8_t vld1q_f16(const float16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_f16)))
+float16x8_t vld1q(const float16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_f32)))
+float32x4_t vld1q_f32(const float32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_f32)))
+float32x4_t vld1q(const float32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_f16)))
+float16x8_t vld1q_z_f16(const float16_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_f16)))
+float16x8_t vld1q_z(const float16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_f32)))
+float32x4_t vld1q_z_f32(const float32_t *, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld1q_z_f32)))
+float32x4_t vld1q_z(const float32_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_f16)))
+float16x8x2_t vld2q_f16(const float16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_f16)))
+float16x8x2_t vld2q(const float16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld2q_f32)))
+float32x4x2_t vld2q_f32(const float32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld2q_f32)))
+float32x4x2_t vld2q(const float32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_f16)))
+float16x8x4_t vld4q_f16(const float16_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_f16)))
+float16x8x4_t vld4q(const float16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vld4q_f32)))
+float32x4x4_t vld4q_f32(const float32_t *);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vld4q_f32)))
+float32x4x4_t vld4q(const float32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_f16)))
+float16x8_t vldrhq_f16(const float16_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_f16)))
+float16x8_t vldrhq_gather_offset_f16(const float16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_f16)))
+float16x8_t vldrhq_gather_offset(const float16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_f16)))
+float16x8_t vldrhq_gather_offset_z_f16(const float16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_offset_z_f16)))
+float16x8_t vldrhq_gather_offset_z(const float16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_f16)))
+float16x8_t vldrhq_gather_shifted_offset_f16(const float16_t *, uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_f16)))
+float16x8_t vldrhq_gather_shifted_offset(const float16_t *, uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_f16)))
+float16x8_t vldrhq_gather_shifted_offset_z_f16(const float16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_gather_shifted_offset_z_f16)))
+float16x8_t vldrhq_gather_shifted_offset_z(const float16_t *, uint16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrhq_z_f16)))
+float16x8_t vldrhq_z_f16(const float16_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_f32)))
+float32x4_t vldrwq_f32(const float32_t *);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_f32)))
+float32x4_t vldrwq_gather_base_f32(uint32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_wb_f32)))
+float32x4_t vldrwq_gather_base_wb_f32(uint32x4_t *, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_wb_z_f32)))
+float32x4_t vldrwq_gather_base_wb_z_f32(uint32x4_t *, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_base_z_f32)))
+float32x4_t vldrwq_gather_base_z_f32(uint32x4_t, int, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_f32)))
+float32x4_t vldrwq_gather_offset_f32(const float32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_f32)))
+float32x4_t vldrwq_gather_offset(const float32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_z_f32)))
+float32x4_t vldrwq_gather_offset_z_f32(const float32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_offset_z_f32)))
+float32x4_t vldrwq_gather_offset_z(const float32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_f32)))
+float32x4_t vldrwq_gather_shifted_offset_f32(const float32_t *, uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_f32)))
+float32x4_t vldrwq_gather_shifted_offset(const float32_t *, uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_z_f32)))
+float32x4_t vldrwq_gather_shifted_offset_z_f32(const float32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_gather_shifted_offset_z_f32)))
+float32x4_t vldrwq_gather_shifted_offset_z(const float32_t *, uint32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vldrwq_z_f32)))
+float32x4_t vldrwq_z_f32(const float32_t *, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_f16)))
+float16x8_t vmaxnmaq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_f16)))
+float16x8_t vmaxnmaq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_f32)))
+float32x4_t vmaxnmaq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_f32)))
+float32x4_t vmaxnmaq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_m_f16)))
+float16x8_t vmaxnmaq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_m_f16)))
+float16x8_t vmaxnmaq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_m_f32)))
+float32x4_t vmaxnmaq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmaq_m_f32)))
+float32x4_t vmaxnmaq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_f16)))
+float16_t vmaxnmavq_f16(float16_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_f16)))
+float16_t vmaxnmavq(float16_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_f32)))
+float32_t vmaxnmavq_f32(float32_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_f32)))
+float32_t vmaxnmavq(float32_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_p_f16)))
+float16_t vmaxnmavq_p_f16(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_p_f16)))
+float16_t vmaxnmavq_p(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_p_f32)))
+float32_t vmaxnmavq_p_f32(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmavq_p_f32)))
+float32_t vmaxnmavq_p(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_f16)))
+float16x8_t vmaxnmq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_f16)))
+float16x8_t vmaxnmq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_f32)))
+float32x4_t vmaxnmq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_f32)))
+float32x4_t vmaxnmq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_m_f16)))
+float16x8_t vmaxnmq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_m_f16)))
+float16x8_t vmaxnmq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_m_f32)))
+float32x4_t vmaxnmq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_m_f32)))
+float32x4_t vmaxnmq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_x_f16)))
+float16x8_t vmaxnmq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_x_f16)))
+float16x8_t vmaxnmq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_x_f32)))
+float32x4_t vmaxnmq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmq_x_f32)))
+float32x4_t vmaxnmq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_f16)))
+float16_t vmaxnmvq_f16(float16_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_f16)))
+float16_t vmaxnmvq(float16_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_f32)))
+float32_t vmaxnmvq_f32(float32_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_f32)))
+float32_t vmaxnmvq(float32_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_p_f16)))
+float16_t vmaxnmvq_p_f16(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_p_f16)))
+float16_t vmaxnmvq_p(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_p_f32)))
+float32_t vmaxnmvq_p_f32(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmaxnmvq_p_f32)))
+float32_t vmaxnmvq_p(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_f16)))
+float16x8_t vminnmaq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_f16)))
+float16x8_t vminnmaq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_f32)))
+float32x4_t vminnmaq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_f32)))
+float32x4_t vminnmaq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_m_f16)))
+float16x8_t vminnmaq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_m_f16)))
+float16x8_t vminnmaq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_m_f32)))
+float32x4_t vminnmaq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmaq_m_f32)))
+float32x4_t vminnmaq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_f16)))
+float16_t vminnmavq_f16(float16_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_f16)))
+float16_t vminnmavq(float16_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_f32)))
+float32_t vminnmavq_f32(float32_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_f32)))
+float32_t vminnmavq(float32_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_p_f16)))
+float16_t vminnmavq_p_f16(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_p_f16)))
+float16_t vminnmavq_p(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_p_f32)))
+float32_t vminnmavq_p_f32(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmavq_p_f32)))
+float32_t vminnmavq_p(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_f16)))
+float16x8_t vminnmq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_f16)))
+float16x8_t vminnmq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_f32)))
+float32x4_t vminnmq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_f32)))
+float32x4_t vminnmq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_m_f16)))
+float16x8_t vminnmq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_m_f16)))
+float16x8_t vminnmq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_m_f32)))
+float32x4_t vminnmq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_m_f32)))
+float32x4_t vminnmq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_x_f16)))
+float16x8_t vminnmq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_x_f16)))
+float16x8_t vminnmq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_x_f32)))
+float32x4_t vminnmq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmq_x_f32)))
+float32x4_t vminnmq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_f16)))
+float16_t vminnmvq_f16(float16_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_f16)))
+float16_t vminnmvq(float16_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_f32)))
+float32_t vminnmvq_f32(float32_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_f32)))
+float32_t vminnmvq(float32_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_p_f16)))
+float16_t vminnmvq_p_f16(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_p_f16)))
+float16_t vminnmvq_p(float16_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_p_f32)))
+float32_t vminnmvq_p_f32(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vminnmvq_p_f32)))
+float32_t vminnmvq_p(float32_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_f16)))
+float16x8_t vmulq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_f16)))
+float16x8_t vmulq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_f32)))
+float32x4_t vmulq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_f32)))
+float32x4_t vmulq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_f16)))
+float16x8_t vmulq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_f16)))
+float16x8_t vmulq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_f32)))
+float32x4_t vmulq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_f32)))
+float32x4_t vmulq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_f16)))
+float16x8_t vmulq_m_n_f16(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_f16)))
+float16x8_t vmulq_m(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_f32)))
+float32x4_t vmulq_m_n_f32(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_m_n_f32)))
+float32x4_t vmulq_m(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_f16)))
+float16x8_t vmulq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_f16)))
+float16x8_t vmulq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_f32)))
+float32x4_t vmulq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_n_f32)))
+float32x4_t vmulq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_f16)))
+float16x8_t vmulq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_f16)))
+float16x8_t vmulq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_f32)))
+float32x4_t vmulq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_f32)))
+float32x4_t vmulq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_f16)))
+float16x8_t vmulq_x_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_f16)))
+float16x8_t vmulq_x(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_f32)))
+float32x4_t vmulq_x_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vmulq_x_n_f32)))
+float32x4_t vmulq_x(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_f16)))
+float16x8_t vnegq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_f16)))
+float16x8_t vnegq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_f32)))
+float32x4_t vnegq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_f32)))
+float32x4_t vnegq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_f16)))
+float16x8_t vnegq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_f16)))
+float16x8_t vnegq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_f32)))
+float32x4_t vnegq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_m_f32)))
+float32x4_t vnegq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_f16)))
+float16x8_t vnegq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_f16)))
+float16x8_t vnegq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_f32)))
+float32x4_t vnegq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vnegq_x_f32)))
+float32x4_t vnegq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_f16)))
+float16x8_t vornq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_f16)))
+float16x8_t vornq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_f32)))
+float32x4_t vornq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_f32)))
+float32x4_t vornq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_f16)))
+float16x8_t vornq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_f16)))
+float16x8_t vornq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_f32)))
+float32x4_t vornq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_m_f32)))
+float32x4_t vornq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_f16)))
+float16x8_t vornq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_f16)))
+float16x8_t vornq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_f32)))
+float32x4_t vornq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vornq_x_f32)))
+float32x4_t vornq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_f16)))
+float16x8_t vorrq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_f16)))
+float16x8_t vorrq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_f32)))
+float32x4_t vorrq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_f32)))
+float32x4_t vorrq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_f16)))
+float16x8_t vorrq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_f16)))
+float16x8_t vorrq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_f32)))
+float32x4_t vorrq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_m_f32)))
+float32x4_t vorrq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_f16)))
+float16x8_t vorrq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_f16)))
+float16x8_t vorrq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_f32)))
+float32x4_t vorrq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vorrq_x_f32)))
+float32x4_t vorrq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_f16)))
+float16x8_t vpselq_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_f16)))
+float16x8_t vpselq(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vpselq_f32)))
+float32x4_t vpselq_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vpselq_f32)))
+float32x4_t vpselq(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_f32)))
+float16x8_t vreinterpretq_f16_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_f32)))
+float16x8_t vreinterpretq_f16(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s16)))
+float16x8_t vreinterpretq_f16_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s16)))
+float16x8_t vreinterpretq_f16(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s32)))
+float16x8_t vreinterpretq_f16_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s32)))
+float16x8_t vreinterpretq_f16(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s64)))
+float16x8_t vreinterpretq_f16_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s64)))
+float16x8_t vreinterpretq_f16(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s8)))
+float16x8_t vreinterpretq_f16_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_s8)))
+float16x8_t vreinterpretq_f16(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u16)))
+float16x8_t vreinterpretq_f16_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u16)))
+float16x8_t vreinterpretq_f16(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u32)))
+float16x8_t vreinterpretq_f16_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u32)))
+float16x8_t vreinterpretq_f16(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u64)))
+float16x8_t vreinterpretq_f16_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u64)))
+float16x8_t vreinterpretq_f16(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u8)))
+float16x8_t vreinterpretq_f16_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f16_u8)))
+float16x8_t vreinterpretq_f16(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_f16)))
+float32x4_t vreinterpretq_f32_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_f16)))
+float32x4_t vreinterpretq_f32(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s16)))
+float32x4_t vreinterpretq_f32_s16(int16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s16)))
+float32x4_t vreinterpretq_f32(int16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s32)))
+float32x4_t vreinterpretq_f32_s32(int32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s32)))
+float32x4_t vreinterpretq_f32(int32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s64)))
+float32x4_t vreinterpretq_f32_s64(int64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s64)))
+float32x4_t vreinterpretq_f32(int64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s8)))
+float32x4_t vreinterpretq_f32_s8(int8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_s8)))
+float32x4_t vreinterpretq_f32(int8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u16)))
+float32x4_t vreinterpretq_f32_u16(uint16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u16)))
+float32x4_t vreinterpretq_f32(uint16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u32)))
+float32x4_t vreinterpretq_f32_u32(uint32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u32)))
+float32x4_t vreinterpretq_f32(uint32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u64)))
+float32x4_t vreinterpretq_f32_u64(uint64x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u64)))
+float32x4_t vreinterpretq_f32(uint64x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u8)))
+float32x4_t vreinterpretq_f32_u8(uint8x16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_f32_u8)))
+float32x4_t vreinterpretq_f32(uint8x16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_f16)))
+int16x8_t vreinterpretq_s16_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_f16)))
+int16x8_t vreinterpretq_s16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_f32)))
+int16x8_t vreinterpretq_s16_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s16_f32)))
+int16x8_t vreinterpretq_s16(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_f16)))
+int32x4_t vreinterpretq_s32_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_f16)))
+int32x4_t vreinterpretq_s32(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_f32)))
+int32x4_t vreinterpretq_s32_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s32_f32)))
+int32x4_t vreinterpretq_s32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_f16)))
+int64x2_t vreinterpretq_s64_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_f16)))
+int64x2_t vreinterpretq_s64(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_f32)))
+int64x2_t vreinterpretq_s64_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s64_f32)))
+int64x2_t vreinterpretq_s64(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_f16)))
+int8x16_t vreinterpretq_s8_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_f16)))
+int8x16_t vreinterpretq_s8(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_f32)))
+int8x16_t vreinterpretq_s8_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_s8_f32)))
+int8x16_t vreinterpretq_s8(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_f16)))
+uint16x8_t vreinterpretq_u16_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_f16)))
+uint16x8_t vreinterpretq_u16(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_f32)))
+uint16x8_t vreinterpretq_u16_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u16_f32)))
+uint16x8_t vreinterpretq_u16(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_f16)))
+uint32x4_t vreinterpretq_u32_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_f16)))
+uint32x4_t vreinterpretq_u32(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_f32)))
+uint32x4_t vreinterpretq_u32_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u32_f32)))
+uint32x4_t vreinterpretq_u32(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_f16)))
+uint64x2_t vreinterpretq_u64_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_f16)))
+uint64x2_t vreinterpretq_u64(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_f32)))
+uint64x2_t vreinterpretq_u64_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u64_f32)))
+uint64x2_t vreinterpretq_u64(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_f16)))
+uint8x16_t vreinterpretq_u8_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_f16)))
+uint8x16_t vreinterpretq_u8(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_f32)))
+uint8x16_t vreinterpretq_u8_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vreinterpretq_u8_f32)))
+uint8x16_t vreinterpretq_u8(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_f16)))
+float16x8_t vrev32q_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_f16)))
+float16x8_t vrev32q(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_f16)))
+float16x8_t vrev32q_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_m_f16)))
+float16x8_t vrev32q_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_f16)))
+float16x8_t vrev32q_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev32q_x_f16)))
+float16x8_t vrev32q_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_f16)))
+float16x8_t vrev64q_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_f16)))
+float16x8_t vrev64q(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_f32)))
+float32x4_t vrev64q_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_f32)))
+float32x4_t vrev64q(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_f16)))
+float16x8_t vrev64q_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_f16)))
+float16x8_t vrev64q_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_f32)))
+float32x4_t vrev64q_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_m_f32)))
+float32x4_t vrev64q_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_f16)))
+float16x8_t vrev64q_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_f16)))
+float16x8_t vrev64q_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_f32)))
+float32x4_t vrev64q_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrev64q_x_f32)))
+float32x4_t vrev64q_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_f16)))
+float16x8_t vrndaq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_f16)))
+float16x8_t vrndaq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_f32)))
+float32x4_t vrndaq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_f32)))
+float32x4_t vrndaq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_m_f16)))
+float16x8_t vrndaq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_m_f16)))
+float16x8_t vrndaq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_m_f32)))
+float32x4_t vrndaq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_m_f32)))
+float32x4_t vrndaq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_x_f16)))
+float16x8_t vrndaq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_x_f16)))
+float16x8_t vrndaq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_x_f32)))
+float32x4_t vrndaq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndaq_x_f32)))
+float32x4_t vrndaq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_f16)))
+float16x8_t vrndmq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_f16)))
+float16x8_t vrndmq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_f32)))
+float32x4_t vrndmq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_f32)))
+float32x4_t vrndmq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_m_f16)))
+float16x8_t vrndmq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_m_f16)))
+float16x8_t vrndmq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_m_f32)))
+float32x4_t vrndmq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_m_f32)))
+float32x4_t vrndmq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_x_f16)))
+float16x8_t vrndmq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_x_f16)))
+float16x8_t vrndmq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_x_f32)))
+float32x4_t vrndmq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndmq_x_f32)))
+float32x4_t vrndmq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_f16)))
+float16x8_t vrndnq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_f16)))
+float16x8_t vrndnq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_f32)))
+float32x4_t vrndnq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_f32)))
+float32x4_t vrndnq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_m_f16)))
+float16x8_t vrndnq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_m_f16)))
+float16x8_t vrndnq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_m_f32)))
+float32x4_t vrndnq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_m_f32)))
+float32x4_t vrndnq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_x_f16)))
+float16x8_t vrndnq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_x_f16)))
+float16x8_t vrndnq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_x_f32)))
+float32x4_t vrndnq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndnq_x_f32)))
+float32x4_t vrndnq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_f16)))
+float16x8_t vrndpq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_f16)))
+float16x8_t vrndpq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_f32)))
+float32x4_t vrndpq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_f32)))
+float32x4_t vrndpq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_m_f16)))
+float16x8_t vrndpq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_m_f16)))
+float16x8_t vrndpq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_m_f32)))
+float32x4_t vrndpq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_m_f32)))
+float32x4_t vrndpq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_x_f16)))
+float16x8_t vrndpq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_x_f16)))
+float16x8_t vrndpq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_x_f32)))
+float32x4_t vrndpq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndpq_x_f32)))
+float32x4_t vrndpq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndq_f16)))
+float16x8_t vrndq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndq_f16)))
+float16x8_t vrndq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndq_f32)))
+float32x4_t vrndq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndq_f32)))
+float32x4_t vrndq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndq_m_f16)))
+float16x8_t vrndq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndq_m_f16)))
+float16x8_t vrndq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndq_m_f32)))
+float32x4_t vrndq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndq_m_f32)))
+float32x4_t vrndq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndq_x_f16)))
+float16x8_t vrndq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndq_x_f16)))
+float16x8_t vrndq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndq_x_f32)))
+float32x4_t vrndq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndq_x_f32)))
+float32x4_t vrndq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_f16)))
+float16x8_t vrndxq_f16(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_f16)))
+float16x8_t vrndxq(float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_f32)))
+float32x4_t vrndxq_f32(float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_f32)))
+float32x4_t vrndxq(float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_m_f16)))
+float16x8_t vrndxq_m_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_m_f16)))
+float16x8_t vrndxq_m(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_m_f32)))
+float32x4_t vrndxq_m_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_m_f32)))
+float32x4_t vrndxq_m(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_x_f16)))
+float16x8_t vrndxq_x_f16(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_x_f16)))
+float16x8_t vrndxq_x(float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_x_f32)))
+float32x4_t vrndxq_x_f32(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vrndxq_x_f32)))
+float32x4_t vrndxq_x(float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_f16)))
+float16x8_t vsetq_lane_f16(float16_t, float16x8_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_f16)))
+float16x8_t vsetq_lane(float16_t, float16x8_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_f32)))
+float32x4_t vsetq_lane_f32(float32_t, float32x4_t, int);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsetq_lane_f32)))
+float32x4_t vsetq_lane(float32_t, float32x4_t, int);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_f16)))
+void vst1q_f16(float16_t *, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_f16)))
+void vst1q(float16_t *, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_f32)))
+void vst1q_f32(float32_t *, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_f32)))
+void vst1q(float32_t *, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_f16)))
+void vst1q_p_f16(float16_t *, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_f16)))
+void vst1q_p(float16_t *, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_f32)))
+void vst1q_p_f32(float32_t *, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst1q_p_f32)))
+void vst1q_p(float32_t *, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_f16)))
+void vst2q_f16(float16_t *, float16x8x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_f16)))
+void vst2q(float16_t *, float16x8x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst2q_f32)))
+void vst2q_f32(float32_t *, float32x4x2_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst2q_f32)))
+void vst2q(float32_t *, float32x4x2_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_f16)))
+void vst4q_f16(float16_t *, float16x8x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_f16)))
+void vst4q(float16_t *, float16x8x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vst4q_f32)))
+void vst4q_f32(float32_t *, float32x4x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vst4q_f32)))
+void vst4q(float32_t *, float32x4x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_f16)))
+void vstrhq_f16(float16_t *, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_f16)))
+void vstrhq(float16_t *, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_f16)))
+void vstrhq_p_f16(float16_t *, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_p_f16)))
+void vstrhq_p(float16_t *, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_f16)))
+void vstrhq_scatter_offset_f16(float16_t *, uint16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_f16)))
+void vstrhq_scatter_offset(float16_t *, uint16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_f16)))
+void vstrhq_scatter_offset_p_f16(float16_t *, uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_offset_p_f16)))
+void vstrhq_scatter_offset_p(float16_t *, uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_f16)))
+void vstrhq_scatter_shifted_offset_f16(float16_t *, uint16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_f16)))
+void vstrhq_scatter_shifted_offset(float16_t *, uint16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_f16)))
+void vstrhq_scatter_shifted_offset_p_f16(float16_t *, uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrhq_scatter_shifted_offset_p_f16)))
+void vstrhq_scatter_shifted_offset_p(float16_t *, uint16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_f32)))
+void vstrwq_f32(float32_t *, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_f32)))
+void vstrwq(float32_t *, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_p_f32)))
+void vstrwq_p_f32(float32_t *, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_p_f32)))
+void vstrwq_p(float32_t *, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_f32)))
+void vstrwq_scatter_base_f32(uint32x4_t, int, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_f32)))
+void vstrwq_scatter_base(uint32x4_t, int, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_p_f32)))
+void vstrwq_scatter_base_p_f32(uint32x4_t, int, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_p_f32)))
+void vstrwq_scatter_base_p(uint32x4_t, int, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_f32)))
+void vstrwq_scatter_base_wb_f32(uint32x4_t *, int, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_f32)))
+void vstrwq_scatter_base_wb(uint32x4_t *, int, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_p_f32)))
+void vstrwq_scatter_base_wb_p_f32(uint32x4_t *, int, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_base_wb_p_f32)))
+void vstrwq_scatter_base_wb_p(uint32x4_t *, int, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_f32)))
+void vstrwq_scatter_offset_f32(float32_t *, uint32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_f32)))
+void vstrwq_scatter_offset(float32_t *, uint32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_p_f32)))
+void vstrwq_scatter_offset_p_f32(float32_t *, uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_offset_p_f32)))
+void vstrwq_scatter_offset_p(float32_t *, uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_f32)))
+void vstrwq_scatter_shifted_offset_f32(float32_t *, uint32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_f32)))
+void vstrwq_scatter_shifted_offset(float32_t *, uint32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_p_f32)))
+void vstrwq_scatter_shifted_offset_p_f32(float32_t *, uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vstrwq_scatter_shifted_offset_p_f32)))
+void vstrwq_scatter_shifted_offset_p(float32_t *, uint32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_f16)))
+float16x8_t vsubq_f16(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_f16)))
+float16x8_t vsubq(float16x8_t, float16x8_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_f32)))
+float32x4_t vsubq_f32(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_f32)))
+float32x4_t vsubq(float32x4_t, float32x4_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_f16)))
+float16x8_t vsubq_m_f16(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_f16)))
+float16x8_t vsubq_m(float16x8_t, float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_f32)))
+float32x4_t vsubq_m_f32(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_f32)))
+float32x4_t vsubq_m(float32x4_t, float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_f16)))
+float16x8_t vsubq_m_n_f16(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_f16)))
+float16x8_t vsubq_m(float16x8_t, float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_f32)))
+float32x4_t vsubq_m_n_f32(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_m_n_f32)))
+float32x4_t vsubq_m(float32x4_t, float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_f16)))
+float16x8_t vsubq_n_f16(float16x8_t, float16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_f16)))
+float16x8_t vsubq(float16x8_t, float16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_f32)))
+float32x4_t vsubq_n_f32(float32x4_t, float32_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_n_f32)))
+float32x4_t vsubq(float32x4_t, float32_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_f16)))
+float16x8_t vsubq_x_f16(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_f16)))
+float16x8_t vsubq_x(float16x8_t, float16x8_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_f32)))
+float32x4_t vsubq_x_f32(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_f32)))
+float32x4_t vsubq_x(float32x4_t, float32x4_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_f16)))
+float16x8_t vsubq_x_n_f16(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_f16)))
+float16x8_t vsubq_x(float16x8_t, float16_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_f32)))
+float32x4_t vsubq_x_n_f32(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vsubq_x_n_f32)))
+float32x4_t vsubq_x(float32x4_t, float32_t, mve_pred16_t);
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_f16)))
+float16x8_t vuninitializedq_f16();
+static __inline__ __attribute__((__clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_f32)))
+float32x4_t vuninitializedq_f32();
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_f16)))
+float16x8_t vuninitializedq(float16x8_t);
+static __inline__ __attribute__((__overloadable__, __clang_arm_builtin_alias(__builtin_arm_mve_vuninitializedq_polymorphic_f32)))
+float32x4_t vuninitializedq(float32x4_t);
+
+#endif /* (__ARM_FEATURE_MVE & 2) && (!defined __ARM_MVE_PRESERVE_USER_NAMESPACE) */
+
+#ifdef __cplusplus
+} /* extern "C" */
+#endif
+
+#endif /* __ARM_MVE_H */
#ifndef __ARM_NEON_H
#define __ARM_NEON_H
+#ifndef __ARM_FP
+#error "NEON intrinsics not available with the soft-float ABI. Please use -mfloat-abi=softfp or -mfloat-abi=hard"
+#else
+
#if !defined(__ARM_NEON)
#error "NEON support not enabled"
-#endif
+#else
#include <stdint.h>
+#ifdef __ARM_FEATURE_BF16
+#include <arm_bf16.h>
+typedef __bf16 bfloat16_t;
+#endif
+
typedef float float32_t;
typedef __fp16 float16_t;
#ifdef __aarch64__
#else
typedef int8_t poly8_t;
typedef int16_t poly16_t;
+typedef int64_t poly64_t;
#endif
typedef __attribute__((neon_vector_type(8))) int8_t int8x8_t;
typedef __attribute__((neon_vector_type(16))) int8_t int8x16_t;
typedef __attribute__((neon_polyvector_type(16))) poly8_t poly8x16_t;
typedef __attribute__((neon_polyvector_type(4))) poly16_t poly16x4_t;
typedef __attribute__((neon_polyvector_type(8))) poly16_t poly16x8_t;
-#ifdef __aarch64__
typedef __attribute__((neon_polyvector_type(1))) poly64_t poly64x1_t;
typedef __attribute__((neon_polyvector_type(2))) poly64_t poly64x2_t;
-#endif
typedef struct int8x8x2_t {
int8x8_t val[2];
poly16x8_t val[2];
} poly16x8x2_t;
-#ifdef __aarch64__
typedef struct poly64x1x2_t {
poly64x1_t val[2];
} poly64x1x2_t;
poly64x2_t val[2];
} poly64x2x2_t;
-#endif
typedef struct int8x8x3_t {
int8x8_t val[3];
} int8x8x3_t;
poly16x8_t val[3];
} poly16x8x3_t;
-#ifdef __aarch64__
typedef struct poly64x1x3_t {
poly64x1_t val[3];
} poly64x1x3_t;
poly64x2_t val[3];
} poly64x2x3_t;
-#endif
typedef struct int8x8x4_t {
int8x8_t val[4];
} int8x8x4_t;
poly16x8_t val[4];
} poly16x8x4_t;
-#ifdef __aarch64__
typedef struct poly64x1x4_t {
poly64x1_t val[4];
} poly64x1x4_t;
poly64x2_t val[4];
} poly64x2x4_t;
+#ifdef __ARM_FEATURE_BF16
+typedef __attribute__((neon_vector_type(4))) bfloat16_t bfloat16x4_t;
+typedef __attribute__((neon_vector_type(8))) bfloat16_t bfloat16x8_t;
+
+typedef struct bfloat16x4x2_t {
+ bfloat16x4_t val[2];
+} bfloat16x4x2_t;
+
+typedef struct bfloat16x8x2_t {
+ bfloat16x8_t val[2];
+} bfloat16x8x2_t;
+
+typedef struct bfloat16x4x3_t {
+ bfloat16x4_t val[3];
+} bfloat16x4x3_t;
+
+typedef struct bfloat16x8x3_t {
+ bfloat16x8_t val[3];
+} bfloat16x8x3_t;
+
+typedef struct bfloat16x4x4_t {
+ bfloat16x4_t val[4];
+} bfloat16x4x4_t;
+
+typedef struct bfloat16x8x4_t {
+ bfloat16x8_t val[4];
+} bfloat16x8x4_t;
+
#endif
-#define __ai static inline __attribute__((__always_inline__, __nodebug__))
+#define __ai static __inline__ __attribute__((__always_inline__, __nodebug__))
#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vabdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
- uint8x16_t __ret;
- __ret = (uint8x16_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
- return __ret;
-}
+#define splat_lane_p8(__p0, __p1) __extension__ ({ \
+ poly8x8_t __s0 = __p0; \
+ poly8x8_t __ret; \
+ __ret = (poly8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 4); \
+ __ret; \
+})
#else
-__ai uint8x16_t vabdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
- uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- uint8x16_t __ret;
- __ret = (uint8x16_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
- __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret;
-}
-__ai uint8x16_t __noswap_vabdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
- uint8x16_t __ret;
- __ret = (uint8x16_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
- return __ret;
-}
+#define splat_lane_p8(__p0, __p1) __extension__ ({ \
+ poly8x8_t __s0 = __p0; \
+ poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x8_t __ret; \
+ __ret = (poly8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 4); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_lane_p8(__p0, __p1) __extension__ ({ \
+ poly8x8_t __s0 = __p0; \
+ poly8x8_t __ret; \
+ __ret = (poly8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 4); \
+ __ret; \
+})
#endif
+#define splat_lane_p64(__p0, __p1) __extension__ ({ \
+ poly64x1_t __s0 = __p0; \
+ poly64x1_t __ret; \
+ __ret = (poly64x1_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 6); \
+ __ret; \
+})
#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vabdq_u32(uint32x4_t __p0, uint32x4_t __p1) {
- uint32x4_t __ret;
- __ret = (uint32x4_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
- return __ret;
-}
+#define splat_lane_p16(__p0, __p1) __extension__ ({ \
+ poly16x4_t __s0 = __p0; \
+ poly16x4_t __ret; \
+ __ret = (poly16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 5); \
+ __ret; \
+})
#else
-__ai uint32x4_t vabdq_u32(uint32x4_t __p0, uint32x4_t __p1) {
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
- uint32x4_t __ret;
- __ret = (uint32x4_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
-__ai uint32x4_t __noswap_vabdq_u32(uint32x4_t __p0, uint32x4_t __p1) {
- uint32x4_t __ret;
- __ret = (uint32x4_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
- return __ret;
-}
+#define splat_lane_p16(__p0, __p1) __extension__ ({ \
+ poly16x4_t __s0 = __p0; \
+ poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ poly16x4_t __ret; \
+ __ret = (poly16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 5); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_lane_p16(__p0, __p1) __extension__ ({ \
+ poly16x4_t __s0 = __p0; \
+ poly16x4_t __ret; \
+ __ret = (poly16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 5); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vabdq_u16(uint16x8_t __p0, uint16x8_t __p1) {
- uint16x8_t __ret;
- __ret = (uint16x8_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
- return __ret;
-}
+#define splatq_lane_p8(__p0, __p1) __extension__ ({ \
+ poly8x8_t __s0 = __p0; \
+ poly8x16_t __ret; \
+ __ret = (poly8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 4); \
+ __ret; \
+})
#else
-__ai uint16x8_t vabdq_u16(uint16x8_t __p0, uint16x8_t __p1) {
- uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
- uint16x8_t __ret;
- __ret = (uint16x8_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret;
-}
-__ai uint16x8_t __noswap_vabdq_u16(uint16x8_t __p0, uint16x8_t __p1) {
- uint16x8_t __ret;
- __ret = (uint16x8_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
- return __ret;
-}
+#define splatq_lane_p8(__p0, __p1) __extension__ ({ \
+ poly8x8_t __s0 = __p0; \
+ poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x16_t __ret; \
+ __ret = (poly8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 4); \
+ __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_lane_p8(__p0, __p1) __extension__ ({ \
+ poly8x8_t __s0 = __p0; \
+ poly8x16_t __ret; \
+ __ret = (poly8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 4); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vabdq_s8(int8x16_t __p0, int8x16_t __p1) {
- int8x16_t __ret;
- __ret = (int8x16_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
- return __ret;
-}
+#define splatq_lane_p64(__p0, __p1) __extension__ ({ \
+ poly64x1_t __s0 = __p0; \
+ poly64x2_t __ret; \
+ __ret = (poly64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 6); \
+ __ret; \
+})
#else
-__ai int8x16_t vabdq_s8(int8x16_t __p0, int8x16_t __p1) {
- int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x16_t __ret;
- __ret = (int8x16_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
- __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret;
-}
-__ai int8x16_t __noswap_vabdq_s8(int8x16_t __p0, int8x16_t __p1) {
- int8x16_t __ret;
- __ret = (int8x16_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
- return __ret;
-}
+#define splatq_lane_p64(__p0, __p1) __extension__ ({ \
+ poly64x1_t __s0 = __p0; \
+ poly64x2_t __ret; \
+ __ret = (poly64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 6); \
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_lane_p64(__p0, __p1) __extension__ ({ \
+ poly64x1_t __s0 = __p0; \
+ poly64x2_t __ret; \
+ __ret = (poly64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 6); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vabdq_f32(float32x4_t __p0, float32x4_t __p1) {
- float32x4_t __ret;
- __ret = (float32x4_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
- return __ret;
-}
+#define splatq_lane_p16(__p0, __p1) __extension__ ({ \
+ poly16x4_t __s0 = __p0; \
+ poly16x8_t __ret; \
+ __ret = (poly16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 5); \
+ __ret; \
+})
#else
-__ai float32x4_t vabdq_f32(float32x4_t __p0, float32x4_t __p1) {
- float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
- float32x4_t __ret;
- __ret = (float32x4_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
+#define splatq_lane_p16(__p0, __p1) __extension__ ({ \
+ poly16x4_t __s0 = __p0; \
+ poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ poly16x8_t __ret; \
+ __ret = (poly16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 5); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_lane_p16(__p0, __p1) __extension__ ({ \
+ poly16x4_t __s0 = __p0; \
+ poly16x8_t __ret; \
+ __ret = (poly16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 5); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vabdq_s32(int32x4_t __p0, int32x4_t __p1) {
- int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
- return __ret;
-}
+#define splatq_lane_u8(__p0, __p1) __extension__ ({ \
+ uint8x8_t __s0 = __p0; \
+ uint8x16_t __ret; \
+ __ret = (uint8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 16); \
+ __ret; \
+})
#else
-__ai int32x4_t vabdq_s32(int32x4_t __p0, int32x4_t __p1) {
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
- int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
-__ai int32x4_t __noswap_vabdq_s32(int32x4_t __p0, int32x4_t __p1) {
- int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
- return __ret;
-}
+#define splatq_lane_u8(__p0, __p1) __extension__ ({ \
+ uint8x8_t __s0 = __p0; \
+ uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __ret; \
+ __ret = (uint8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 16); \
+ __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_lane_u8(__p0, __p1) __extension__ ({ \
+ uint8x8_t __s0 = __p0; \
+ uint8x16_t __ret; \
+ __ret = (uint8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 16); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vabdq_s16(int16x8_t __p0, int16x8_t __p1) {
- int16x8_t __ret;
- __ret = (int16x8_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
- return __ret;
-}
+#define splatq_lane_u32(__p0, __p1) __extension__ ({ \
+ uint32x2_t __s0 = __p0; \
+ uint32x4_t __ret; \
+ __ret = (uint32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 18); \
+ __ret; \
+})
#else
-__ai int16x8_t vabdq_s16(int16x8_t __p0, int16x8_t __p1) {
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
- int16x8_t __ret;
- __ret = (int16x8_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret;
-}
-__ai int16x8_t __noswap_vabdq_s16(int16x8_t __p0, int16x8_t __p1) {
- int16x8_t __ret;
- __ret = (int16x8_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
- return __ret;
-}
+#define splatq_lane_u32(__p0, __p1) __extension__ ({ \
+ uint32x2_t __s0 = __p0; \
+ uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ uint32x4_t __ret; \
+ __ret = (uint32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 18); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_lane_u32(__p0, __p1) __extension__ ({ \
+ uint32x2_t __s0 = __p0; \
+ uint32x4_t __ret; \
+ __ret = (uint32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 18); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vabd_u8(uint8x8_t __p0, uint8x8_t __p1) {
- uint8x8_t __ret;
- __ret = (uint8x8_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
- return __ret;
-}
+#define splatq_lane_u64(__p0, __p1) __extension__ ({ \
+ uint64x1_t __s0 = __p0; \
+ uint64x2_t __ret; \
+ __ret = (uint64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 19); \
+ __ret; \
+})
#else
-__ai uint8x8_t vabd_u8(uint8x8_t __p0, uint8x8_t __p1) {
- uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
- uint8x8_t __ret;
- __ret = (uint8x8_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret;
-}
-__ai uint8x8_t __noswap_vabd_u8(uint8x8_t __p0, uint8x8_t __p1) {
- uint8x8_t __ret;
- __ret = (uint8x8_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
- return __ret;
-}
+#define splatq_lane_u64(__p0, __p1) __extension__ ({ \
+ uint64x1_t __s0 = __p0; \
+ uint64x2_t __ret; \
+ __ret = (uint64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 19); \
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_lane_u64(__p0, __p1) __extension__ ({ \
+ uint64x1_t __s0 = __p0; \
+ uint64x2_t __ret; \
+ __ret = (uint64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 19); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vabd_u32(uint32x2_t __p0, uint32x2_t __p1) {
- uint32x2_t __ret;
- __ret = (uint32x2_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
- return __ret;
-}
+#define splatq_lane_u16(__p0, __p1) __extension__ ({ \
+ uint16x4_t __s0 = __p0; \
+ uint16x8_t __ret; \
+ __ret = (uint16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 17); \
+ __ret; \
+})
#else
-__ai uint32x2_t vabd_u32(uint32x2_t __p0, uint32x2_t __p1) {
- uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
- uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
- uint32x2_t __ret;
- __ret = (uint32x2_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
- return __ret;
-}
-__ai uint32x2_t __noswap_vabd_u32(uint32x2_t __p0, uint32x2_t __p1) {
- uint32x2_t __ret;
- __ret = (uint32x2_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
- return __ret;
-}
+#define splatq_lane_u16(__p0, __p1) __extension__ ({ \
+ uint16x4_t __s0 = __p0; \
+ uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ uint16x8_t __ret; \
+ __ret = (uint16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 17); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_lane_u16(__p0, __p1) __extension__ ({ \
+ uint16x4_t __s0 = __p0; \
+ uint16x8_t __ret; \
+ __ret = (uint16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 17); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vabd_u16(uint16x4_t __p0, uint16x4_t __p1) {
- uint16x4_t __ret;
- __ret = (uint16x4_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
- return __ret;
-}
+#define splatq_lane_s8(__p0, __p1) __extension__ ({ \
+ int8x8_t __s0 = __p0; \
+ int8x16_t __ret; \
+ __ret = (int8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 0); \
+ __ret; \
+})
#else
-__ai uint16x4_t vabd_u16(uint16x4_t __p0, uint16x4_t __p1) {
- uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
- uint16x4_t __ret;
- __ret = (uint16x4_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
-__ai uint16x4_t __noswap_vabd_u16(uint16x4_t __p0, uint16x4_t __p1) {
- uint16x4_t __ret;
- __ret = (uint16x4_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
- return __ret;
-}
+#define splatq_lane_s8(__p0, __p1) __extension__ ({ \
+ int8x8_t __s0 = __p0; \
+ int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __ret; \
+ __ret = (int8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 0); \
+ __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_lane_s8(__p0, __p1) __extension__ ({ \
+ int8x8_t __s0 = __p0; \
+ int8x16_t __ret; \
+ __ret = (int8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vabd_s8(int8x8_t __p0, int8x8_t __p1) {
- int8x8_t __ret;
- __ret = (int8x8_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
- return __ret;
-}
+#define splatq_lane_f64(__p0, __p1) __extension__ ({ \
+ float64x1_t __s0 = __p0; \
+ float64x2_t __ret; \
+ __ret = (float64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 10); \
+ __ret; \
+})
#else
-__ai int8x8_t vabd_s8(int8x8_t __p0, int8x8_t __p1) {
- int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x8_t __ret;
- __ret = (int8x8_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret;
-}
-__ai int8x8_t __noswap_vabd_s8(int8x8_t __p0, int8x8_t __p1) {
- int8x8_t __ret;
- __ret = (int8x8_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
- return __ret;
-}
+#define splatq_lane_f64(__p0, __p1) __extension__ ({ \
+ float64x1_t __s0 = __p0; \
+ float64x2_t __ret; \
+ __ret = (float64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 10); \
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_lane_f64(__p0, __p1) __extension__ ({ \
+ float64x1_t __s0 = __p0; \
+ float64x2_t __ret; \
+ __ret = (float64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 10); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vabd_f32(float32x2_t __p0, float32x2_t __p1) {
- float32x2_t __ret;
- __ret = (float32x2_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
- return __ret;
-}
+#define splatq_lane_f32(__p0, __p1) __extension__ ({ \
+ float32x2_t __s0 = __p0; \
+ float32x4_t __ret; \
+ __ret = (float32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 9); \
+ __ret; \
+})
#else
-__ai float32x2_t vabd_f32(float32x2_t __p0, float32x2_t __p1) {
- float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
- float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
- float32x2_t __ret;
- __ret = (float32x2_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
- return __ret;
-}
+#define splatq_lane_f32(__p0, __p1) __extension__ ({ \
+ float32x2_t __s0 = __p0; \
+ float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ float32x4_t __ret; \
+ __ret = (float32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 9); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_lane_f32(__p0, __p1) __extension__ ({ \
+ float32x2_t __s0 = __p0; \
+ float32x4_t __ret; \
+ __ret = (float32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 9); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vabd_s32(int32x2_t __p0, int32x2_t __p1) {
- int32x2_t __ret;
- __ret = (int32x2_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
- return __ret;
-}
+#define splatq_lane_f16(__p0, __p1) __extension__ ({ \
+ float16x4_t __s0 = __p0; \
+ float16x8_t __ret; \
+ __ret = (float16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 8); \
+ __ret; \
+})
#else
-__ai int32x2_t vabd_s32(int32x2_t __p0, int32x2_t __p1) {
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
- int32x2_t __ret;
- __ret = (int32x2_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
- return __ret;
-}
-__ai int32x2_t __noswap_vabd_s32(int32x2_t __p0, int32x2_t __p1) {
- int32x2_t __ret;
- __ret = (int32x2_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
- return __ret;
-}
+#define splatq_lane_f16(__p0, __p1) __extension__ ({ \
+ float16x4_t __s0 = __p0; \
+ float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ float16x8_t __ret; \
+ __ret = (float16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 8); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_lane_f16(__p0, __p1) __extension__ ({ \
+ float16x4_t __s0 = __p0; \
+ float16x8_t __ret; \
+ __ret = (float16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 8); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vabd_s16(int16x4_t __p0, int16x4_t __p1) {
- int16x4_t __ret;
- __ret = (int16x4_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
- return __ret;
-}
+#define splatq_lane_s32(__p0, __p1) __extension__ ({ \
+ int32x2_t __s0 = __p0; \
+ int32x4_t __ret; \
+ __ret = (int32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 2); \
+ __ret; \
+})
#else
-__ai int16x4_t vabd_s16(int16x4_t __p0, int16x4_t __p1) {
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
- int16x4_t __ret;
- __ret = (int16x4_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
-__ai int16x4_t __noswap_vabd_s16(int16x4_t __p0, int16x4_t __p1) {
- int16x4_t __ret;
- __ret = (int16x4_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
- return __ret;
-}
+#define splatq_lane_s32(__p0, __p1) __extension__ ({ \
+ int32x2_t __s0 = __p0; \
+ int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ int32x4_t __ret; \
+ __ret = (int32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 2); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_lane_s32(__p0, __p1) __extension__ ({ \
+ int32x2_t __s0 = __p0; \
+ int32x4_t __ret; \
+ __ret = (int32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 2); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vabsq_s8(int8x16_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t) __builtin_neon_vabsq_v((int8x16_t)__p0, 32);
- return __ret;
-}
+#define splatq_lane_s64(__p0, __p1) __extension__ ({ \
+ int64x1_t __s0 = __p0; \
+ int64x2_t __ret; \
+ __ret = (int64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 3); \
+ __ret; \
+})
#else
-__ai int8x16_t vabsq_s8(int8x16_t __p0) {
- int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x16_t __ret;
- __ret = (int8x16_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 32);
- __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret;
-}
+#define splatq_lane_s64(__p0, __p1) __extension__ ({ \
+ int64x1_t __s0 = __p0; \
+ int64x2_t __ret; \
+ __ret = (int64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 3); \
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_lane_s64(__p0, __p1) __extension__ ({ \
+ int64x1_t __s0 = __p0; \
+ int64x2_t __ret; \
+ __ret = (int64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 3); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vabsq_f32(float32x4_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t) __builtin_neon_vabsq_v((int8x16_t)__p0, 41);
- return __ret;
-}
+#define splatq_lane_s16(__p0, __p1) __extension__ ({ \
+ int16x4_t __s0 = __p0; \
+ int16x8_t __ret; \
+ __ret = (int16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 1); \
+ __ret; \
+})
#else
-__ai float32x4_t vabsq_f32(float32x4_t __p0) {
- float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- float32x4_t __ret;
- __ret = (float32x4_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 41);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
+#define splatq_lane_s16(__p0, __p1) __extension__ ({ \
+ int16x4_t __s0 = __p0; \
+ int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ int16x8_t __ret; \
+ __ret = (int16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 1); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_lane_s16(__p0, __p1) __extension__ ({ \
+ int16x4_t __s0 = __p0; \
+ int16x8_t __ret; \
+ __ret = (int16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 1); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vabsq_s32(int32x4_t __p0) {
- int32x4_t __ret;
+#define splat_lane_u8(__p0, __p1) __extension__ ({ \
+ uint8x8_t __s0 = __p0; \
+ uint8x8_t __ret; \
+ __ret = (uint8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 16); \
+ __ret; \
+})
+#else
+#define splat_lane_u8(__p0, __p1) __extension__ ({ \
+ uint8x8_t __s0 = __p0; \
+ uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x8_t __ret; \
+ __ret = (uint8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 16); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_lane_u8(__p0, __p1) __extension__ ({ \
+ uint8x8_t __s0 = __p0; \
+ uint8x8_t __ret; \
+ __ret = (uint8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 16); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_lane_u32(__p0, __p1) __extension__ ({ \
+ uint32x2_t __s0 = __p0; \
+ uint32x2_t __ret; \
+ __ret = (uint32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 18); \
+ __ret; \
+})
+#else
+#define splat_lane_u32(__p0, __p1) __extension__ ({ \
+ uint32x2_t __s0 = __p0; \
+ uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ uint32x2_t __ret; \
+ __ret = (uint32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 18); \
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_lane_u32(__p0, __p1) __extension__ ({ \
+ uint32x2_t __s0 = __p0; \
+ uint32x2_t __ret; \
+ __ret = (uint32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 18); \
+ __ret; \
+})
+#endif
+
+#define splat_lane_u64(__p0, __p1) __extension__ ({ \
+ uint64x1_t __s0 = __p0; \
+ uint64x1_t __ret; \
+ __ret = (uint64x1_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 19); \
+ __ret; \
+})
+#ifdef __LITTLE_ENDIAN__
+#define splat_lane_u16(__p0, __p1) __extension__ ({ \
+ uint16x4_t __s0 = __p0; \
+ uint16x4_t __ret; \
+ __ret = (uint16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 17); \
+ __ret; \
+})
+#else
+#define splat_lane_u16(__p0, __p1) __extension__ ({ \
+ uint16x4_t __s0 = __p0; \
+ uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ uint16x4_t __ret; \
+ __ret = (uint16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 17); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_lane_u16(__p0, __p1) __extension__ ({ \
+ uint16x4_t __s0 = __p0; \
+ uint16x4_t __ret; \
+ __ret = (uint16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 17); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_lane_s8(__p0, __p1) __extension__ ({ \
+ int8x8_t __s0 = __p0; \
+ int8x8_t __ret; \
+ __ret = (int8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 0); \
+ __ret; \
+})
+#else
+#define splat_lane_s8(__p0, __p1) __extension__ ({ \
+ int8x8_t __s0 = __p0; \
+ int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x8_t __ret; \
+ __ret = (int8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 0); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_lane_s8(__p0, __p1) __extension__ ({ \
+ int8x8_t __s0 = __p0; \
+ int8x8_t __ret; \
+ __ret = (int8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 0); \
+ __ret; \
+})
+#endif
+
+#define splat_lane_f64(__p0, __p1) __extension__ ({ \
+ float64x1_t __s0 = __p0; \
+ float64x1_t __ret; \
+ __ret = (float64x1_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 10); \
+ __ret; \
+})
+#ifdef __LITTLE_ENDIAN__
+#define splat_lane_f32(__p0, __p1) __extension__ ({ \
+ float32x2_t __s0 = __p0; \
+ float32x2_t __ret; \
+ __ret = (float32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 9); \
+ __ret; \
+})
+#else
+#define splat_lane_f32(__p0, __p1) __extension__ ({ \
+ float32x2_t __s0 = __p0; \
+ float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ float32x2_t __ret; \
+ __ret = (float32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 9); \
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_lane_f32(__p0, __p1) __extension__ ({ \
+ float32x2_t __s0 = __p0; \
+ float32x2_t __ret; \
+ __ret = (float32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 9); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_lane_f16(__p0, __p1) __extension__ ({ \
+ float16x4_t __s0 = __p0; \
+ float16x4_t __ret; \
+ __ret = (float16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 8); \
+ __ret; \
+})
+#else
+#define splat_lane_f16(__p0, __p1) __extension__ ({ \
+ float16x4_t __s0 = __p0; \
+ float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ float16x4_t __ret; \
+ __ret = (float16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 8); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_lane_f16(__p0, __p1) __extension__ ({ \
+ float16x4_t __s0 = __p0; \
+ float16x4_t __ret; \
+ __ret = (float16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 8); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_lane_s32(__p0, __p1) __extension__ ({ \
+ int32x2_t __s0 = __p0; \
+ int32x2_t __ret; \
+ __ret = (int32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 2); \
+ __ret; \
+})
+#else
+#define splat_lane_s32(__p0, __p1) __extension__ ({ \
+ int32x2_t __s0 = __p0; \
+ int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ int32x2_t __ret; \
+ __ret = (int32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 2); \
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_lane_s32(__p0, __p1) __extension__ ({ \
+ int32x2_t __s0 = __p0; \
+ int32x2_t __ret; \
+ __ret = (int32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 2); \
+ __ret; \
+})
+#endif
+
+#define splat_lane_s64(__p0, __p1) __extension__ ({ \
+ int64x1_t __s0 = __p0; \
+ int64x1_t __ret; \
+ __ret = (int64x1_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 3); \
+ __ret; \
+})
+#ifdef __LITTLE_ENDIAN__
+#define splat_lane_s16(__p0, __p1) __extension__ ({ \
+ int16x4_t __s0 = __p0; \
+ int16x4_t __ret; \
+ __ret = (int16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 1); \
+ __ret; \
+})
+#else
+#define splat_lane_s16(__p0, __p1) __extension__ ({ \
+ int16x4_t __s0 = __p0; \
+ int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ int16x4_t __ret; \
+ __ret = (int16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 1); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_lane_s16(__p0, __p1) __extension__ ({ \
+ int16x4_t __s0 = __p0; \
+ int16x4_t __ret; \
+ __ret = (int16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 1); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_laneq_p8(__p0, __p1) __extension__ ({ \
+ poly8x16_t __s0 = __p0; \
+ poly8x8_t __ret; \
+ __ret = (poly8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 36); \
+ __ret; \
+})
+#else
+#define splat_laneq_p8(__p0, __p1) __extension__ ({ \
+ poly8x16_t __s0 = __p0; \
+ poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x8_t __ret; \
+ __ret = (poly8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 36); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_laneq_p8(__p0, __p1) __extension__ ({ \
+ poly8x16_t __s0 = __p0; \
+ poly8x8_t __ret; \
+ __ret = (poly8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 36); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_laneq_p64(__p0, __p1) __extension__ ({ \
+ poly64x2_t __s0 = __p0; \
+ poly64x1_t __ret; \
+ __ret = (poly64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 38); \
+ __ret; \
+})
+#else
+#define splat_laneq_p64(__p0, __p1) __extension__ ({ \
+ poly64x2_t __s0 = __p0; \
+ poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ poly64x1_t __ret; \
+ __ret = (poly64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 38); \
+ __ret; \
+})
+#define __noswap_splat_laneq_p64(__p0, __p1) __extension__ ({ \
+ poly64x2_t __s0 = __p0; \
+ poly64x1_t __ret; \
+ __ret = (poly64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 38); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_laneq_p16(__p0, __p1) __extension__ ({ \
+ poly16x8_t __s0 = __p0; \
+ poly16x4_t __ret; \
+ __ret = (poly16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 37); \
+ __ret; \
+})
+#else
+#define splat_laneq_p16(__p0, __p1) __extension__ ({ \
+ poly16x8_t __s0 = __p0; \
+ poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly16x4_t __ret; \
+ __ret = (poly16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 37); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_laneq_p16(__p0, __p1) __extension__ ({ \
+ poly16x8_t __s0 = __p0; \
+ poly16x4_t __ret; \
+ __ret = (poly16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 37); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splatq_laneq_p8(__p0, __p1) __extension__ ({ \
+ poly8x16_t __s0 = __p0; \
+ poly8x16_t __ret; \
+ __ret = (poly8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 36); \
+ __ret; \
+})
+#else
+#define splatq_laneq_p8(__p0, __p1) __extension__ ({ \
+ poly8x16_t __s0 = __p0; \
+ poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x16_t __ret; \
+ __ret = (poly8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 36); \
+ __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_laneq_p8(__p0, __p1) __extension__ ({ \
+ poly8x16_t __s0 = __p0; \
+ poly8x16_t __ret; \
+ __ret = (poly8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 36); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splatq_laneq_p64(__p0, __p1) __extension__ ({ \
+ poly64x2_t __s0 = __p0; \
+ poly64x2_t __ret; \
+ __ret = (poly64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 38); \
+ __ret; \
+})
+#else
+#define splatq_laneq_p64(__p0, __p1) __extension__ ({ \
+ poly64x2_t __s0 = __p0; \
+ poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ poly64x2_t __ret; \
+ __ret = (poly64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 38); \
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_laneq_p64(__p0, __p1) __extension__ ({ \
+ poly64x2_t __s0 = __p0; \
+ poly64x2_t __ret; \
+ __ret = (poly64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 38); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splatq_laneq_p16(__p0, __p1) __extension__ ({ \
+ poly16x8_t __s0 = __p0; \
+ poly16x8_t __ret; \
+ __ret = (poly16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 37); \
+ __ret; \
+})
+#else
+#define splatq_laneq_p16(__p0, __p1) __extension__ ({ \
+ poly16x8_t __s0 = __p0; \
+ poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly16x8_t __ret; \
+ __ret = (poly16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 37); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_laneq_p16(__p0, __p1) __extension__ ({ \
+ poly16x8_t __s0 = __p0; \
+ poly16x8_t __ret; \
+ __ret = (poly16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 37); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splatq_laneq_u8(__p0, __p1) __extension__ ({ \
+ uint8x16_t __s0 = __p0; \
+ uint8x16_t __ret; \
+ __ret = (uint8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 48); \
+ __ret; \
+})
+#else
+#define splatq_laneq_u8(__p0, __p1) __extension__ ({ \
+ uint8x16_t __s0 = __p0; \
+ uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __ret; \
+ __ret = (uint8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 48); \
+ __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_laneq_u8(__p0, __p1) __extension__ ({ \
+ uint8x16_t __s0 = __p0; \
+ uint8x16_t __ret; \
+ __ret = (uint8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 48); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splatq_laneq_u32(__p0, __p1) __extension__ ({ \
+ uint32x4_t __s0 = __p0; \
+ uint32x4_t __ret; \
+ __ret = (uint32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 50); \
+ __ret; \
+})
+#else
+#define splatq_laneq_u32(__p0, __p1) __extension__ ({ \
+ uint32x4_t __s0 = __p0; \
+ uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ uint32x4_t __ret; \
+ __ret = (uint32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 50); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_laneq_u32(__p0, __p1) __extension__ ({ \
+ uint32x4_t __s0 = __p0; \
+ uint32x4_t __ret; \
+ __ret = (uint32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 50); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splatq_laneq_u64(__p0, __p1) __extension__ ({ \
+ uint64x2_t __s0 = __p0; \
+ uint64x2_t __ret; \
+ __ret = (uint64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 51); \
+ __ret; \
+})
+#else
+#define splatq_laneq_u64(__p0, __p1) __extension__ ({ \
+ uint64x2_t __s0 = __p0; \
+ uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ uint64x2_t __ret; \
+ __ret = (uint64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 51); \
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_laneq_u64(__p0, __p1) __extension__ ({ \
+ uint64x2_t __s0 = __p0; \
+ uint64x2_t __ret; \
+ __ret = (uint64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 51); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splatq_laneq_u16(__p0, __p1) __extension__ ({ \
+ uint16x8_t __s0 = __p0; \
+ uint16x8_t __ret; \
+ __ret = (uint16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 49); \
+ __ret; \
+})
+#else
+#define splatq_laneq_u16(__p0, __p1) __extension__ ({ \
+ uint16x8_t __s0 = __p0; \
+ uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __ret; \
+ __ret = (uint16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 49); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_laneq_u16(__p0, __p1) __extension__ ({ \
+ uint16x8_t __s0 = __p0; \
+ uint16x8_t __ret; \
+ __ret = (uint16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 49); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splatq_laneq_s8(__p0, __p1) __extension__ ({ \
+ int8x16_t __s0 = __p0; \
+ int8x16_t __ret; \
+ __ret = (int8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 32); \
+ __ret; \
+})
+#else
+#define splatq_laneq_s8(__p0, __p1) __extension__ ({ \
+ int8x16_t __s0 = __p0; \
+ int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __ret; \
+ __ret = (int8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 32); \
+ __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_laneq_s8(__p0, __p1) __extension__ ({ \
+ int8x16_t __s0 = __p0; \
+ int8x16_t __ret; \
+ __ret = (int8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 32); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splatq_laneq_f64(__p0, __p1) __extension__ ({ \
+ float64x2_t __s0 = __p0; \
+ float64x2_t __ret; \
+ __ret = (float64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 42); \
+ __ret; \
+})
+#else
+#define splatq_laneq_f64(__p0, __p1) __extension__ ({ \
+ float64x2_t __s0 = __p0; \
+ float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ float64x2_t __ret; \
+ __ret = (float64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 42); \
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_laneq_f64(__p0, __p1) __extension__ ({ \
+ float64x2_t __s0 = __p0; \
+ float64x2_t __ret; \
+ __ret = (float64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 42); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splatq_laneq_f32(__p0, __p1) __extension__ ({ \
+ float32x4_t __s0 = __p0; \
+ float32x4_t __ret; \
+ __ret = (float32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 41); \
+ __ret; \
+})
+#else
+#define splatq_laneq_f32(__p0, __p1) __extension__ ({ \
+ float32x4_t __s0 = __p0; \
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ float32x4_t __ret; \
+ __ret = (float32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 41); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_laneq_f32(__p0, __p1) __extension__ ({ \
+ float32x4_t __s0 = __p0; \
+ float32x4_t __ret; \
+ __ret = (float32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 41); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splatq_laneq_f16(__p0, __p1) __extension__ ({ \
+ float16x8_t __s0 = __p0; \
+ float16x8_t __ret; \
+ __ret = (float16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 40); \
+ __ret; \
+})
+#else
+#define splatq_laneq_f16(__p0, __p1) __extension__ ({ \
+ float16x8_t __s0 = __p0; \
+ float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __ret; \
+ __ret = (float16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 40); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_laneq_f16(__p0, __p1) __extension__ ({ \
+ float16x8_t __s0 = __p0; \
+ float16x8_t __ret; \
+ __ret = (float16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 40); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splatq_laneq_s32(__p0, __p1) __extension__ ({ \
+ int32x4_t __s0 = __p0; \
+ int32x4_t __ret; \
+ __ret = (int32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 34); \
+ __ret; \
+})
+#else
+#define splatq_laneq_s32(__p0, __p1) __extension__ ({ \
+ int32x4_t __s0 = __p0; \
+ int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ int32x4_t __ret; \
+ __ret = (int32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 34); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_laneq_s32(__p0, __p1) __extension__ ({ \
+ int32x4_t __s0 = __p0; \
+ int32x4_t __ret; \
+ __ret = (int32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 34); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splatq_laneq_s64(__p0, __p1) __extension__ ({ \
+ int64x2_t __s0 = __p0; \
+ int64x2_t __ret; \
+ __ret = (int64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 35); \
+ __ret; \
+})
+#else
+#define splatq_laneq_s64(__p0, __p1) __extension__ ({ \
+ int64x2_t __s0 = __p0; \
+ int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ int64x2_t __ret; \
+ __ret = (int64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 35); \
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_laneq_s64(__p0, __p1) __extension__ ({ \
+ int64x2_t __s0 = __p0; \
+ int64x2_t __ret; \
+ __ret = (int64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 35); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splatq_laneq_s16(__p0, __p1) __extension__ ({ \
+ int16x8_t __s0 = __p0; \
+ int16x8_t __ret; \
+ __ret = (int16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 33); \
+ __ret; \
+})
+#else
+#define splatq_laneq_s16(__p0, __p1) __extension__ ({ \
+ int16x8_t __s0 = __p0; \
+ int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __ret; \
+ __ret = (int16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 33); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_laneq_s16(__p0, __p1) __extension__ ({ \
+ int16x8_t __s0 = __p0; \
+ int16x8_t __ret; \
+ __ret = (int16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 33); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_laneq_u8(__p0, __p1) __extension__ ({ \
+ uint8x16_t __s0 = __p0; \
+ uint8x8_t __ret; \
+ __ret = (uint8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 48); \
+ __ret; \
+})
+#else
+#define splat_laneq_u8(__p0, __p1) __extension__ ({ \
+ uint8x16_t __s0 = __p0; \
+ uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x8_t __ret; \
+ __ret = (uint8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 48); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_laneq_u8(__p0, __p1) __extension__ ({ \
+ uint8x16_t __s0 = __p0; \
+ uint8x8_t __ret; \
+ __ret = (uint8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 48); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_laneq_u32(__p0, __p1) __extension__ ({ \
+ uint32x4_t __s0 = __p0; \
+ uint32x2_t __ret; \
+ __ret = (uint32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 50); \
+ __ret; \
+})
+#else
+#define splat_laneq_u32(__p0, __p1) __extension__ ({ \
+ uint32x4_t __s0 = __p0; \
+ uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ uint32x2_t __ret; \
+ __ret = (uint32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 50); \
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_laneq_u32(__p0, __p1) __extension__ ({ \
+ uint32x4_t __s0 = __p0; \
+ uint32x2_t __ret; \
+ __ret = (uint32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 50); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_laneq_u64(__p0, __p1) __extension__ ({ \
+ uint64x2_t __s0 = __p0; \
+ uint64x1_t __ret; \
+ __ret = (uint64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 51); \
+ __ret; \
+})
+#else
+#define splat_laneq_u64(__p0, __p1) __extension__ ({ \
+ uint64x2_t __s0 = __p0; \
+ uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ uint64x1_t __ret; \
+ __ret = (uint64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 51); \
+ __ret; \
+})
+#define __noswap_splat_laneq_u64(__p0, __p1) __extension__ ({ \
+ uint64x2_t __s0 = __p0; \
+ uint64x1_t __ret; \
+ __ret = (uint64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 51); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_laneq_u16(__p0, __p1) __extension__ ({ \
+ uint16x8_t __s0 = __p0; \
+ uint16x4_t __ret; \
+ __ret = (uint16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 49); \
+ __ret; \
+})
+#else
+#define splat_laneq_u16(__p0, __p1) __extension__ ({ \
+ uint16x8_t __s0 = __p0; \
+ uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x4_t __ret; \
+ __ret = (uint16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 49); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_laneq_u16(__p0, __p1) __extension__ ({ \
+ uint16x8_t __s0 = __p0; \
+ uint16x4_t __ret; \
+ __ret = (uint16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 49); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_laneq_s8(__p0, __p1) __extension__ ({ \
+ int8x16_t __s0 = __p0; \
+ int8x8_t __ret; \
+ __ret = (int8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 32); \
+ __ret; \
+})
+#else
+#define splat_laneq_s8(__p0, __p1) __extension__ ({ \
+ int8x16_t __s0 = __p0; \
+ int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x8_t __ret; \
+ __ret = (int8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 32); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_laneq_s8(__p0, __p1) __extension__ ({ \
+ int8x16_t __s0 = __p0; \
+ int8x8_t __ret; \
+ __ret = (int8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 32); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_laneq_f64(__p0, __p1) __extension__ ({ \
+ float64x2_t __s0 = __p0; \
+ float64x1_t __ret; \
+ __ret = (float64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 42); \
+ __ret; \
+})
+#else
+#define splat_laneq_f64(__p0, __p1) __extension__ ({ \
+ float64x2_t __s0 = __p0; \
+ float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ float64x1_t __ret; \
+ __ret = (float64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 42); \
+ __ret; \
+})
+#define __noswap_splat_laneq_f64(__p0, __p1) __extension__ ({ \
+ float64x2_t __s0 = __p0; \
+ float64x1_t __ret; \
+ __ret = (float64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 42); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_laneq_f32(__p0, __p1) __extension__ ({ \
+ float32x4_t __s0 = __p0; \
+ float32x2_t __ret; \
+ __ret = (float32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 41); \
+ __ret; \
+})
+#else
+#define splat_laneq_f32(__p0, __p1) __extension__ ({ \
+ float32x4_t __s0 = __p0; \
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ float32x2_t __ret; \
+ __ret = (float32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 41); \
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_laneq_f32(__p0, __p1) __extension__ ({ \
+ float32x4_t __s0 = __p0; \
+ float32x2_t __ret; \
+ __ret = (float32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 41); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_laneq_f16(__p0, __p1) __extension__ ({ \
+ float16x8_t __s0 = __p0; \
+ float16x4_t __ret; \
+ __ret = (float16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 40); \
+ __ret; \
+})
+#else
+#define splat_laneq_f16(__p0, __p1) __extension__ ({ \
+ float16x8_t __s0 = __p0; \
+ float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x4_t __ret; \
+ __ret = (float16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 40); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_laneq_f16(__p0, __p1) __extension__ ({ \
+ float16x8_t __s0 = __p0; \
+ float16x4_t __ret; \
+ __ret = (float16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 40); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_laneq_s32(__p0, __p1) __extension__ ({ \
+ int32x4_t __s0 = __p0; \
+ int32x2_t __ret; \
+ __ret = (int32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 34); \
+ __ret; \
+})
+#else
+#define splat_laneq_s32(__p0, __p1) __extension__ ({ \
+ int32x4_t __s0 = __p0; \
+ int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ int32x2_t __ret; \
+ __ret = (int32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 34); \
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_laneq_s32(__p0, __p1) __extension__ ({ \
+ int32x4_t __s0 = __p0; \
+ int32x2_t __ret; \
+ __ret = (int32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 34); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_laneq_s64(__p0, __p1) __extension__ ({ \
+ int64x2_t __s0 = __p0; \
+ int64x1_t __ret; \
+ __ret = (int64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 35); \
+ __ret; \
+})
+#else
+#define splat_laneq_s64(__p0, __p1) __extension__ ({ \
+ int64x2_t __s0 = __p0; \
+ int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ int64x1_t __ret; \
+ __ret = (int64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 35); \
+ __ret; \
+})
+#define __noswap_splat_laneq_s64(__p0, __p1) __extension__ ({ \
+ int64x2_t __s0 = __p0; \
+ int64x1_t __ret; \
+ __ret = (int64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 35); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define splat_laneq_s16(__p0, __p1) __extension__ ({ \
+ int16x8_t __s0 = __p0; \
+ int16x4_t __ret; \
+ __ret = (int16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 33); \
+ __ret; \
+})
+#else
+#define splat_laneq_s16(__p0, __p1) __extension__ ({ \
+ int16x8_t __s0 = __p0; \
+ int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __ret; \
+ __ret = (int16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 33); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_laneq_s16(__p0, __p1) __extension__ ({ \
+ int16x8_t __s0 = __p0; \
+ int16x4_t __ret; \
+ __ret = (int16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 33); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai uint8x16_t vabdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
+ uint8x16_t __ret;
+ __ret = (uint8x16_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
+ return __ret;
+}
+#else
+__ai uint8x16_t vabdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
+ uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x16_t __ret;
+ __ret = (uint8x16_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
+ __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+__ai uint8x16_t __noswap_vabdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
+ uint8x16_t __ret;
+ __ret = (uint8x16_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai uint32x4_t vabdq_u32(uint32x4_t __p0, uint32x4_t __p1) {
+ uint32x4_t __ret;
+ __ret = (uint32x4_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
+ return __ret;
+}
+#else
+__ai uint32x4_t vabdq_u32(uint32x4_t __p0, uint32x4_t __p1) {
+ uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ uint32x4_t __ret;
+ __ret = (uint32x4_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai uint32x4_t __noswap_vabdq_u32(uint32x4_t __p0, uint32x4_t __p1) {
+ uint32x4_t __ret;
+ __ret = (uint32x4_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai uint16x8_t vabdq_u16(uint16x8_t __p0, uint16x8_t __p1) {
+ uint16x8_t __ret;
+ __ret = (uint16x8_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
+ return __ret;
+}
+#else
+__ai uint16x8_t vabdq_u16(uint16x8_t __p0, uint16x8_t __p1) {
+ uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint16x8_t __ret;
+ __ret = (uint16x8_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+__ai uint16x8_t __noswap_vabdq_u16(uint16x8_t __p0, uint16x8_t __p1) {
+ uint16x8_t __ret;
+ __ret = (uint16x8_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai int8x16_t vabdq_s8(int8x16_t __p0, int8x16_t __p1) {
+ int8x16_t __ret;
+ __ret = (int8x16_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
+ return __ret;
+}
+#else
+__ai int8x16_t vabdq_s8(int8x16_t __p0, int8x16_t __p1) {
+ int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ int8x16_t __ret;
+ __ret = (int8x16_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
+ __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+__ai int8x16_t __noswap_vabdq_s8(int8x16_t __p0, int8x16_t __p1) {
+ int8x16_t __ret;
+ __ret = (int8x16_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float32x4_t vabdq_f32(float32x4_t __p0, float32x4_t __p1) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
+ return __ret;
+}
+#else
+__ai float32x4_t vabdq_f32(float32x4_t __p0, float32x4_t __p1) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai int32x4_t vabdq_s32(int32x4_t __p0, int32x4_t __p1) {
+ int32x4_t __ret;
+ __ret = (int32x4_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
+ return __ret;
+}
+#else
+__ai int32x4_t vabdq_s32(int32x4_t __p0, int32x4_t __p1) {
+ int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ int32x4_t __ret;
+ __ret = (int32x4_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai int32x4_t __noswap_vabdq_s32(int32x4_t __p0, int32x4_t __p1) {
+ int32x4_t __ret;
+ __ret = (int32x4_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai int16x8_t vabdq_s16(int16x8_t __p0, int16x8_t __p1) {
+ int16x8_t __ret;
+ __ret = (int16x8_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
+ return __ret;
+}
+#else
+__ai int16x8_t vabdq_s16(int16x8_t __p0, int16x8_t __p1) {
+ int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ int16x8_t __ret;
+ __ret = (int16x8_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+__ai int16x8_t __noswap_vabdq_s16(int16x8_t __p0, int16x8_t __p1) {
+ int16x8_t __ret;
+ __ret = (int16x8_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai uint8x8_t vabd_u8(uint8x8_t __p0, uint8x8_t __p1) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
+ return __ret;
+}
+#else
+__ai uint8x8_t vabd_u8(uint8x8_t __p0, uint8x8_t __p1) {
+ uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x8_t __ret;
+ __ret = (uint8x8_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+__ai uint8x8_t __noswap_vabd_u8(uint8x8_t __p0, uint8x8_t __p1) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai uint32x2_t vabd_u32(uint32x2_t __p0, uint32x2_t __p1) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
+ return __ret;
+}
+#else
+__ai uint32x2_t vabd_u32(uint32x2_t __p0, uint32x2_t __p1) {
+ uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
+ uint32x2_t __ret;
+ __ret = (uint32x2_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
+ return __ret;
+}
+__ai uint32x2_t __noswap_vabd_u32(uint32x2_t __p0, uint32x2_t __p1) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai uint16x4_t vabd_u16(uint16x4_t __p0, uint16x4_t __p1) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
+ return __ret;
+}
+#else
+__ai uint16x4_t vabd_u16(uint16x4_t __p0, uint16x4_t __p1) {
+ uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ uint16x4_t __ret;
+ __ret = (uint16x4_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai uint16x4_t __noswap_vabd_u16(uint16x4_t __p0, uint16x4_t __p1) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai int8x8_t vabd_s8(int8x8_t __p0, int8x8_t __p1) {
+ int8x8_t __ret;
+ __ret = (int8x8_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
+ return __ret;
+}
+#else
+__ai int8x8_t vabd_s8(int8x8_t __p0, int8x8_t __p1) {
+ int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ int8x8_t __ret;
+ __ret = (int8x8_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+__ai int8x8_t __noswap_vabd_s8(int8x8_t __p0, int8x8_t __p1) {
+ int8x8_t __ret;
+ __ret = (int8x8_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float32x2_t vabd_f32(float32x2_t __p0, float32x2_t __p1) {
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
+ return __ret;
+}
+#else
+__ai float32x2_t vabd_f32(float32x2_t __p0, float32x2_t __p1) {
+ float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai int32x2_t vabd_s32(int32x2_t __p0, int32x2_t __p1) {
+ int32x2_t __ret;
+ __ret = (int32x2_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
+ return __ret;
+}
+#else
+__ai int32x2_t vabd_s32(int32x2_t __p0, int32x2_t __p1) {
+ int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
+ int32x2_t __ret;
+ __ret = (int32x2_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
+ return __ret;
+}
+__ai int32x2_t __noswap_vabd_s32(int32x2_t __p0, int32x2_t __p1) {
+ int32x2_t __ret;
+ __ret = (int32x2_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai int16x4_t vabd_s16(int16x4_t __p0, int16x4_t __p1) {
+ int16x4_t __ret;
+ __ret = (int16x4_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
+ return __ret;
+}
+#else
+__ai int16x4_t vabd_s16(int16x4_t __p0, int16x4_t __p1) {
+ int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ int16x4_t __ret;
+ __ret = (int16x4_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai int16x4_t __noswap_vabd_s16(int16x4_t __p0, int16x4_t __p1) {
+ int16x4_t __ret;
+ __ret = (int16x4_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai int8x16_t vabsq_s8(int8x16_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t) __builtin_neon_vabsq_v((int8x16_t)__p0, 32);
+ return __ret;
+}
+#else
+__ai int8x16_t vabsq_s8(int8x16_t __p0) {
+ int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ int8x16_t __ret;
+ __ret = (int8x16_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 32);
+ __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float32x4_t vabsq_f32(float32x4_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vabsq_v((int8x16_t)__p0, 41);
+ return __ret;
+}
+#else
+__ai float32x4_t vabsq_f32(float32x4_t __p0) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 41);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai int32x4_t vabsq_s32(int32x4_t __p0) {
+ int32x4_t __ret;
__ret = (int32x4_t) __builtin_neon_vabsq_v((int8x16_t)__p0, 34);
return __ret;
}
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = __p0 + __p1;
- return __ret;
-}
-#else
__ai uint64x1_t vadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
uint64x1_t __ret;
__ret = __p0 + __p1;
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vadd_s64(int64x1_t __p0, int64x1_t __p1) {
- int64x1_t __ret;
- __ret = __p0 + __p1;
- return __ret;
-}
-#else
__ai int64x1_t vadd_s64(int64x1_t __p0, int64x1_t __p1) {
int64x1_t __ret;
__ret = __p0 + __p1;
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vadd_s16(int16x4_t __p0, int16x4_t __p1) {
int16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vand_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = __p0 & __p1;
- return __ret;
-}
-#else
__ai uint64x1_t vand_u64(uint64x1_t __p0, uint64x1_t __p1) {
uint64x1_t __ret;
__ret = __p0 & __p1;
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vand_u16(uint16x4_t __p0, uint16x4_t __p1) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai int64x1_t vand_s64(int64x1_t __p0, int64x1_t __p1) {
int64x1_t __ret;
__ret = __p0 & __p1;
return __ret;
}
-#else
-__ai int64x1_t vand_s64(int64x1_t __p0, int64x1_t __p1) {
- int64x1_t __ret;
- __ret = __p0 & __p1;
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vand_s16(int16x4_t __p0, int16x4_t __p1) {
int16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vbic_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = __p0 & ~__p1;
- return __ret;
-}
-#else
__ai uint64x1_t vbic_u64(uint64x1_t __p0, uint64x1_t __p1) {
uint64x1_t __ret;
__ret = __p0 & ~__p1;
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vbic_u16(uint16x4_t __p0, uint16x4_t __p1) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vbic_s64(int64x1_t __p0, int64x1_t __p1) {
- int64x1_t __ret;
- __ret = __p0 & ~__p1;
- return __ret;
-}
-#else
__ai int64x1_t vbic_s64(int64x1_t __p0, int64x1_t __p1) {
int64x1_t __ret;
__ret = __p0 & ~__p1;
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vbic_s16(int16x4_t __p0, int16x4_t __p1) {
int16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vbsl_u64(uint64x1_t __p0, uint64x1_t __p1, uint64x1_t __p2) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 19);
- return __ret;
-}
-#else
__ai uint64x1_t vbsl_u64(uint64x1_t __p0, uint64x1_t __p1, uint64x1_t __p2) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 19);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vbsl_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vbsl_s64(uint64x1_t __p0, int64x1_t __p1, int64x1_t __p2) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 3);
- return __ret;
-}
-#else
__ai int64x1_t vbsl_s64(uint64x1_t __p0, int64x1_t __p1, int64x1_t __p2) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 3);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vbsl_s16(uint16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
int16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vcreate_p8(uint64_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
-#else
-__ai poly8x8_t vcreate_p8(uint64_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vcreate_p16(uint64_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
-__ai poly16x4_t vcreate_p16(uint64_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vcreate_u8(uint64_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
- return __ret;
-}
-#else
-__ai uint8x8_t vcreate_u8(uint64_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vcreate_u32(uint64_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
- return __ret;
-}
-#else
-__ai uint32x2_t vcreate_u32(uint64_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vcreate_u64(uint64_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
- return __ret;
-}
-#else
-__ai uint64x1_t vcreate_u64(uint64_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vcreate_u16(uint64_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#else
-__ai uint16x4_t vcreate_u16(uint64_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vcreate_s8(uint64_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
-#else
-__ai int8x8_t vcreate_s8(uint64_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vcreate_f32(uint64_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
-#else
-__ai float32x2_t vcreate_f32(uint64_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vcreate_f16(uint64_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
-#else
-__ai float16x4_t vcreate_f16(uint64_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vcreate_s32(uint64_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
-#else
-__ai int32x2_t vcreate_s32(uint64_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vcreate_s64(uint64_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
-#else
-__ai int64x1_t vcreate_s64(uint64_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vcreate_s16(uint64_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
-#else
-__ai int16x4_t vcreate_s16(uint64_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
-#endif
-
+#define vcreate_p8(__p0) __extension__ ({ \
+ poly8x8_t __ret; \
+ uint64_t __promote = __p0; \
+ __ret = (poly8x8_t)(__promote); \
+ __ret; \
+})
+#define vcreate_p16(__p0) __extension__ ({ \
+ poly16x4_t __ret; \
+ uint64_t __promote = __p0; \
+ __ret = (poly16x4_t)(__promote); \
+ __ret; \
+})
+#define vcreate_u8(__p0) __extension__ ({ \
+ uint8x8_t __ret; \
+ uint64_t __promote = __p0; \
+ __ret = (uint8x8_t)(__promote); \
+ __ret; \
+})
+#define vcreate_u32(__p0) __extension__ ({ \
+ uint32x2_t __ret; \
+ uint64_t __promote = __p0; \
+ __ret = (uint32x2_t)(__promote); \
+ __ret; \
+})
+#define vcreate_u64(__p0) __extension__ ({ \
+ uint64x1_t __ret; \
+ uint64_t __promote = __p0; \
+ __ret = (uint64x1_t)(__promote); \
+ __ret; \
+})
+#define vcreate_u16(__p0) __extension__ ({ \
+ uint16x4_t __ret; \
+ uint64_t __promote = __p0; \
+ __ret = (uint16x4_t)(__promote); \
+ __ret; \
+})
+#define vcreate_s8(__p0) __extension__ ({ \
+ int8x8_t __ret; \
+ uint64_t __promote = __p0; \
+ __ret = (int8x8_t)(__promote); \
+ __ret; \
+})
+#define vcreate_f32(__p0) __extension__ ({ \
+ float32x2_t __ret; \
+ uint64_t __promote = __p0; \
+ __ret = (float32x2_t)(__promote); \
+ __ret; \
+})
+#define vcreate_f16(__p0) __extension__ ({ \
+ float16x4_t __ret; \
+ uint64_t __promote = __p0; \
+ __ret = (float16x4_t)(__promote); \
+ __ret; \
+})
+#define vcreate_s32(__p0) __extension__ ({ \
+ int32x2_t __ret; \
+ uint64_t __promote = __p0; \
+ __ret = (int32x2_t)(__promote); \
+ __ret; \
+})
+#define vcreate_s64(__p0) __extension__ ({ \
+ int64x1_t __ret; \
+ uint64_t __promote = __p0; \
+ __ret = (int64x1_t)(__promote); \
+ __ret; \
+})
+#define vcreate_s16(__p0) __extension__ ({ \
+ int16x4_t __ret; \
+ uint64_t __promote = __p0; \
+ __ret = (int16x4_t)(__promote); \
+ __ret; \
+})
#ifdef __LITTLE_ENDIAN__
__ai float32x4_t vcvtq_f32_u32(uint32x4_t __p0) {
float32x4_t __ret;
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_lane_p8(__p0, __p1) __extension__ ({ \
- poly8x8_t __s0 = __p0; \
- poly8x8_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdup_lane_p8(__p0_0, __p1_0) __extension__ ({ \
+ poly8x8_t __s0_0 = __p0_0; \
+ poly8x8_t __ret_0; \
+ __ret_0 = splat_lane_p8(__s0_0, __p1_0); \
+ __ret_0; \
})
#else
-#define vdup_lane_p8(__p0, __p1) __extension__ ({ \
- poly8x8_t __s0 = __p0; \
- poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly8x8_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdup_lane_p8(__p0_1, __p1_1) __extension__ ({ \
+ poly8x8_t __s0_1 = __p0_1; \
+ poly8x8_t __rev0_1; __rev0_1 = __builtin_shufflevector(__s0_1, __s0_1, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x8_t __ret_1; \
+ __ret_1 = __noswap_splat_lane_p8(__rev0_1, __p1_1); \
+ __ret_1 = __builtin_shufflevector(__ret_1, __ret_1, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_1; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_lane_p16(__p0, __p1) __extension__ ({ \
- poly16x4_t __s0 = __p0; \
- poly16x4_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdup_lane_p16(__p0_2, __p1_2) __extension__ ({ \
+ poly16x4_t __s0_2 = __p0_2; \
+ poly16x4_t __ret_2; \
+ __ret_2 = splat_lane_p16(__s0_2, __p1_2); \
+ __ret_2; \
})
#else
-#define vdup_lane_p16(__p0, __p1) __extension__ ({ \
- poly16x4_t __s0 = __p0; \
- poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- poly16x4_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdup_lane_p16(__p0_3, __p1_3) __extension__ ({ \
+ poly16x4_t __s0_3 = __p0_3; \
+ poly16x4_t __rev0_3; __rev0_3 = __builtin_shufflevector(__s0_3, __s0_3, 3, 2, 1, 0); \
+ poly16x4_t __ret_3; \
+ __ret_3 = __noswap_splat_lane_p16(__rev0_3, __p1_3); \
+ __ret_3 = __builtin_shufflevector(__ret_3, __ret_3, 3, 2, 1, 0); \
+ __ret_3; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_lane_p8(__p0, __p1) __extension__ ({ \
- poly8x8_t __s0 = __p0; \
- poly8x16_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_lane_p8(__p0_4, __p1_4) __extension__ ({ \
+ poly8x8_t __s0_4 = __p0_4; \
+ poly8x16_t __ret_4; \
+ __ret_4 = splatq_lane_p8(__s0_4, __p1_4); \
+ __ret_4; \
})
#else
-#define vdupq_lane_p8(__p0, __p1) __extension__ ({ \
- poly8x8_t __s0 = __p0; \
- poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly8x16_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_lane_p8(__p0_5, __p1_5) __extension__ ({ \
+ poly8x8_t __s0_5 = __p0_5; \
+ poly8x8_t __rev0_5; __rev0_5 = __builtin_shufflevector(__s0_5, __s0_5, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x16_t __ret_5; \
+ __ret_5 = __noswap_splatq_lane_p8(__rev0_5, __p1_5); \
+ __ret_5 = __builtin_shufflevector(__ret_5, __ret_5, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_5; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_lane_p16(__p0, __p1) __extension__ ({ \
- poly16x4_t __s0 = __p0; \
- poly16x8_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_lane_p16(__p0_6, __p1_6) __extension__ ({ \
+ poly16x4_t __s0_6 = __p0_6; \
+ poly16x8_t __ret_6; \
+ __ret_6 = splatq_lane_p16(__s0_6, __p1_6); \
+ __ret_6; \
})
#else
-#define vdupq_lane_p16(__p0, __p1) __extension__ ({ \
- poly16x4_t __s0 = __p0; \
- poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- poly16x8_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_lane_p16(__p0_7, __p1_7) __extension__ ({ \
+ poly16x4_t __s0_7 = __p0_7; \
+ poly16x4_t __rev0_7; __rev0_7 = __builtin_shufflevector(__s0_7, __s0_7, 3, 2, 1, 0); \
+ poly16x8_t __ret_7; \
+ __ret_7 = __noswap_splatq_lane_p16(__rev0_7, __p1_7); \
+ __ret_7 = __builtin_shufflevector(__ret_7, __ret_7, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_7; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_lane_u8(__p0, __p1) __extension__ ({ \
- uint8x8_t __s0 = __p0; \
- uint8x16_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_lane_u8(__p0_8, __p1_8) __extension__ ({ \
+ uint8x8_t __s0_8 = __p0_8; \
+ uint8x16_t __ret_8; \
+ __ret_8 = splatq_lane_u8(__s0_8, __p1_8); \
+ __ret_8; \
})
#else
-#define vdupq_lane_u8(__p0, __p1) __extension__ ({ \
- uint8x8_t __s0 = __p0; \
- uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x16_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_lane_u8(__p0_9, __p1_9) __extension__ ({ \
+ uint8x8_t __s0_9 = __p0_9; \
+ uint8x8_t __rev0_9; __rev0_9 = __builtin_shufflevector(__s0_9, __s0_9, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __ret_9; \
+ __ret_9 = __noswap_splatq_lane_u8(__rev0_9, __p1_9); \
+ __ret_9 = __builtin_shufflevector(__ret_9, __ret_9, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_9; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_lane_u32(__p0, __p1) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x4_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_lane_u32(__p0_10, __p1_10) __extension__ ({ \
+ uint32x2_t __s0_10 = __p0_10; \
+ uint32x4_t __ret_10; \
+ __ret_10 = splatq_lane_u32(__s0_10, __p1_10); \
+ __ret_10; \
})
#else
-#define vdupq_lane_u32(__p0, __p1) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x4_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_lane_u32(__p0_11, __p1_11) __extension__ ({ \
+ uint32x2_t __s0_11 = __p0_11; \
+ uint32x2_t __rev0_11; __rev0_11 = __builtin_shufflevector(__s0_11, __s0_11, 1, 0); \
+ uint32x4_t __ret_11; \
+ __ret_11 = __noswap_splatq_lane_u32(__rev0_11, __p1_11); \
+ __ret_11 = __builtin_shufflevector(__ret_11, __ret_11, 3, 2, 1, 0); \
+ __ret_11; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_lane_u64(__p0, __p1) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret; \
+#define vdupq_lane_u64(__p0_12, __p1_12) __extension__ ({ \
+ uint64x1_t __s0_12 = __p0_12; \
+ uint64x2_t __ret_12; \
+ __ret_12 = splatq_lane_u64(__s0_12, __p1_12); \
+ __ret_12; \
})
#else
-#define vdupq_lane_u64(__p0, __p1) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdupq_lane_u64(__p0_13, __p1_13) __extension__ ({ \
+ uint64x1_t __s0_13 = __p0_13; \
+ uint64x2_t __ret_13; \
+ __ret_13 = __noswap_splatq_lane_u64(__s0_13, __p1_13); \
+ __ret_13 = __builtin_shufflevector(__ret_13, __ret_13, 1, 0); \
+ __ret_13; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_lane_u16(__p0, __p1) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x8_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_lane_u16(__p0_14, __p1_14) __extension__ ({ \
+ uint16x4_t __s0_14 = __p0_14; \
+ uint16x8_t __ret_14; \
+ __ret_14 = splatq_lane_u16(__s0_14, __p1_14); \
+ __ret_14; \
})
#else
-#define vdupq_lane_u16(__p0, __p1) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x8_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_lane_u16(__p0_15, __p1_15) __extension__ ({ \
+ uint16x4_t __s0_15 = __p0_15; \
+ uint16x4_t __rev0_15; __rev0_15 = __builtin_shufflevector(__s0_15, __s0_15, 3, 2, 1, 0); \
+ uint16x8_t __ret_15; \
+ __ret_15 = __noswap_splatq_lane_u16(__rev0_15, __p1_15); \
+ __ret_15 = __builtin_shufflevector(__ret_15, __ret_15, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_15; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_lane_s8(__p0, __p1) __extension__ ({ \
- int8x8_t __s0 = __p0; \
- int8x16_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_lane_s8(__p0_16, __p1_16) __extension__ ({ \
+ int8x8_t __s0_16 = __p0_16; \
+ int8x16_t __ret_16; \
+ __ret_16 = splatq_lane_s8(__s0_16, __p1_16); \
+ __ret_16; \
})
#else
-#define vdupq_lane_s8(__p0, __p1) __extension__ ({ \
- int8x8_t __s0 = __p0; \
- int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x16_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_lane_s8(__p0_17, __p1_17) __extension__ ({ \
+ int8x8_t __s0_17 = __p0_17; \
+ int8x8_t __rev0_17; __rev0_17 = __builtin_shufflevector(__s0_17, __s0_17, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __ret_17; \
+ __ret_17 = __noswap_splatq_lane_s8(__rev0_17, __p1_17); \
+ __ret_17 = __builtin_shufflevector(__ret_17, __ret_17, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_17; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_lane_f32(__p0, __p1) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x4_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_lane_f32(__p0_18, __p1_18) __extension__ ({ \
+ float32x2_t __s0_18 = __p0_18; \
+ float32x4_t __ret_18; \
+ __ret_18 = splatq_lane_f32(__s0_18, __p1_18); \
+ __ret_18; \
})
#else
-#define vdupq_lane_f32(__p0, __p1) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float32x4_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_lane_f32(__p0_19, __p1_19) __extension__ ({ \
+ float32x2_t __s0_19 = __p0_19; \
+ float32x2_t __rev0_19; __rev0_19 = __builtin_shufflevector(__s0_19, __s0_19, 1, 0); \
+ float32x4_t __ret_19; \
+ __ret_19 = __noswap_splatq_lane_f32(__rev0_19, __p1_19); \
+ __ret_19 = __builtin_shufflevector(__ret_19, __ret_19, 3, 2, 1, 0); \
+ __ret_19; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_lane_s32(__p0, __p1) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x4_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_lane_s32(__p0_20, __p1_20) __extension__ ({ \
+ int32x2_t __s0_20 = __p0_20; \
+ int32x4_t __ret_20; \
+ __ret_20 = splatq_lane_s32(__s0_20, __p1_20); \
+ __ret_20; \
})
#else
-#define vdupq_lane_s32(__p0, __p1) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x4_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_lane_s32(__p0_21, __p1_21) __extension__ ({ \
+ int32x2_t __s0_21 = __p0_21; \
+ int32x2_t __rev0_21; __rev0_21 = __builtin_shufflevector(__s0_21, __s0_21, 1, 0); \
+ int32x4_t __ret_21; \
+ __ret_21 = __noswap_splatq_lane_s32(__rev0_21, __p1_21); \
+ __ret_21 = __builtin_shufflevector(__ret_21, __ret_21, 3, 2, 1, 0); \
+ __ret_21; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_lane_s64(__p0, __p1) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret; \
+#define vdupq_lane_s64(__p0_22, __p1_22) __extension__ ({ \
+ int64x1_t __s0_22 = __p0_22; \
+ int64x2_t __ret_22; \
+ __ret_22 = splatq_lane_s64(__s0_22, __p1_22); \
+ __ret_22; \
})
#else
-#define vdupq_lane_s64(__p0, __p1) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdupq_lane_s64(__p0_23, __p1_23) __extension__ ({ \
+ int64x1_t __s0_23 = __p0_23; \
+ int64x2_t __ret_23; \
+ __ret_23 = __noswap_splatq_lane_s64(__s0_23, __p1_23); \
+ __ret_23 = __builtin_shufflevector(__ret_23, __ret_23, 1, 0); \
+ __ret_23; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_lane_s16(__p0, __p1) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x8_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_lane_s16(__p0_24, __p1_24) __extension__ ({ \
+ int16x4_t __s0_24 = __p0_24; \
+ int16x8_t __ret_24; \
+ __ret_24 = splatq_lane_s16(__s0_24, __p1_24); \
+ __ret_24; \
})
#else
-#define vdupq_lane_s16(__p0, __p1) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x8_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_lane_s16(__p0_25, __p1_25) __extension__ ({ \
+ int16x4_t __s0_25 = __p0_25; \
+ int16x4_t __rev0_25; __rev0_25 = __builtin_shufflevector(__s0_25, __s0_25, 3, 2, 1, 0); \
+ int16x8_t __ret_25; \
+ __ret_25 = __noswap_splatq_lane_s16(__rev0_25, __p1_25); \
+ __ret_25 = __builtin_shufflevector(__ret_25, __ret_25, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_25; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_lane_u8(__p0, __p1) __extension__ ({ \
- uint8x8_t __s0 = __p0; \
- uint8x8_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdup_lane_u8(__p0_26, __p1_26) __extension__ ({ \
+ uint8x8_t __s0_26 = __p0_26; \
+ uint8x8_t __ret_26; \
+ __ret_26 = splat_lane_u8(__s0_26, __p1_26); \
+ __ret_26; \
})
#else
-#define vdup_lane_u8(__p0, __p1) __extension__ ({ \
- uint8x8_t __s0 = __p0; \
- uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x8_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdup_lane_u8(__p0_27, __p1_27) __extension__ ({ \
+ uint8x8_t __s0_27 = __p0_27; \
+ uint8x8_t __rev0_27; __rev0_27 = __builtin_shufflevector(__s0_27, __s0_27, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x8_t __ret_27; \
+ __ret_27 = __noswap_splat_lane_u8(__rev0_27, __p1_27); \
+ __ret_27 = __builtin_shufflevector(__ret_27, __ret_27, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_27; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_lane_u32(__p0, __p1) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret; \
+#define vdup_lane_u32(__p0_28, __p1_28) __extension__ ({ \
+ uint32x2_t __s0_28 = __p0_28; \
+ uint32x2_t __ret_28; \
+ __ret_28 = splat_lane_u32(__s0_28, __p1_28); \
+ __ret_28; \
})
#else
-#define vdup_lane_u32(__p0, __p1) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x2_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdup_lane_u32(__p0_29, __p1_29) __extension__ ({ \
+ uint32x2_t __s0_29 = __p0_29; \
+ uint32x2_t __rev0_29; __rev0_29 = __builtin_shufflevector(__s0_29, __s0_29, 1, 0); \
+ uint32x2_t __ret_29; \
+ __ret_29 = __noswap_splat_lane_u32(__rev0_29, __p1_29); \
+ __ret_29 = __builtin_shufflevector(__ret_29, __ret_29, 1, 0); \
+ __ret_29; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vdup_lane_u64(__p0, __p1) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64x1_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1); \
- __ret; \
-})
-#else
-#define vdup_lane_u64(__p0, __p1) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64x1_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1); \
- __ret; \
+#define vdup_lane_u64(__p0_30, __p1_30) __extension__ ({ \
+ uint64x1_t __s0_30 = __p0_30; \
+ uint64x1_t __ret_30; \
+ __ret_30 = splat_lane_u64(__s0_30, __p1_30); \
+ __ret_30; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vdup_lane_u16(__p0, __p1) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x4_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdup_lane_u16(__p0_31, __p1_31) __extension__ ({ \
+ uint16x4_t __s0_31 = __p0_31; \
+ uint16x4_t __ret_31; \
+ __ret_31 = splat_lane_u16(__s0_31, __p1_31); \
+ __ret_31; \
})
#else
-#define vdup_lane_u16(__p0, __p1) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x4_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdup_lane_u16(__p0_32, __p1_32) __extension__ ({ \
+ uint16x4_t __s0_32 = __p0_32; \
+ uint16x4_t __rev0_32; __rev0_32 = __builtin_shufflevector(__s0_32, __s0_32, 3, 2, 1, 0); \
+ uint16x4_t __ret_32; \
+ __ret_32 = __noswap_splat_lane_u16(__rev0_32, __p1_32); \
+ __ret_32 = __builtin_shufflevector(__ret_32, __ret_32, 3, 2, 1, 0); \
+ __ret_32; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_lane_s8(__p0, __p1) __extension__ ({ \
- int8x8_t __s0 = __p0; \
- int8x8_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdup_lane_s8(__p0_33, __p1_33) __extension__ ({ \
+ int8x8_t __s0_33 = __p0_33; \
+ int8x8_t __ret_33; \
+ __ret_33 = splat_lane_s8(__s0_33, __p1_33); \
+ __ret_33; \
})
#else
-#define vdup_lane_s8(__p0, __p1) __extension__ ({ \
- int8x8_t __s0 = __p0; \
- int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x8_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdup_lane_s8(__p0_34, __p1_34) __extension__ ({ \
+ int8x8_t __s0_34 = __p0_34; \
+ int8x8_t __rev0_34; __rev0_34 = __builtin_shufflevector(__s0_34, __s0_34, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x8_t __ret_34; \
+ __ret_34 = __noswap_splat_lane_s8(__rev0_34, __p1_34); \
+ __ret_34 = __builtin_shufflevector(__ret_34, __ret_34, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_34; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_lane_f32(__p0, __p1) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret; \
+#define vdup_lane_f32(__p0_35, __p1_35) __extension__ ({ \
+ float32x2_t __s0_35 = __p0_35; \
+ float32x2_t __ret_35; \
+ __ret_35 = splat_lane_f32(__s0_35, __p1_35); \
+ __ret_35; \
})
#else
-#define vdup_lane_f32(__p0, __p1) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float32x2_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdup_lane_f32(__p0_36, __p1_36) __extension__ ({ \
+ float32x2_t __s0_36 = __p0_36; \
+ float32x2_t __rev0_36; __rev0_36 = __builtin_shufflevector(__s0_36, __s0_36, 1, 0); \
+ float32x2_t __ret_36; \
+ __ret_36 = __noswap_splat_lane_f32(__rev0_36, __p1_36); \
+ __ret_36 = __builtin_shufflevector(__ret_36, __ret_36, 1, 0); \
+ __ret_36; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_lane_s32(__p0, __p1) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret; \
+#define vdup_lane_s32(__p0_37, __p1_37) __extension__ ({ \
+ int32x2_t __s0_37 = __p0_37; \
+ int32x2_t __ret_37; \
+ __ret_37 = splat_lane_s32(__s0_37, __p1_37); \
+ __ret_37; \
})
#else
-#define vdup_lane_s32(__p0, __p1) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdup_lane_s32(__p0_38, __p1_38) __extension__ ({ \
+ int32x2_t __s0_38 = __p0_38; \
+ int32x2_t __rev0_38; __rev0_38 = __builtin_shufflevector(__s0_38, __s0_38, 1, 0); \
+ int32x2_t __ret_38; \
+ __ret_38 = __noswap_splat_lane_s32(__rev0_38, __p1_38); \
+ __ret_38 = __builtin_shufflevector(__ret_38, __ret_38, 1, 0); \
+ __ret_38; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vdup_lane_s64(__p0, __p1) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64x1_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1); \
- __ret; \
-})
-#else
-#define vdup_lane_s64(__p0, __p1) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64x1_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1); \
- __ret; \
+#define vdup_lane_s64(__p0_39, __p1_39) __extension__ ({ \
+ int64x1_t __s0_39 = __p0_39; \
+ int64x1_t __ret_39; \
+ __ret_39 = splat_lane_s64(__s0_39, __p1_39); \
+ __ret_39; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vdup_lane_s16(__p0, __p1) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdup_lane_s16(__p0_40, __p1_40) __extension__ ({ \
+ int16x4_t __s0_40 = __p0_40; \
+ int16x4_t __ret_40; \
+ __ret_40 = splat_lane_s16(__s0_40, __p1_40); \
+ __ret_40; \
})
#else
-#define vdup_lane_s16(__p0, __p1) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdup_lane_s16(__p0_41, __p1_41) __extension__ ({ \
+ int16x4_t __s0_41 = __p0_41; \
+ int16x4_t __rev0_41; __rev0_41 = __builtin_shufflevector(__s0_41, __s0_41, 3, 2, 1, 0); \
+ int16x4_t __ret_41; \
+ __ret_41 = __noswap_splat_lane_s16(__rev0_41, __p1_41); \
+ __ret_41 = __builtin_shufflevector(__ret_41, __ret_41, 3, 2, 1, 0); \
+ __ret_41; \
})
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vdup_n_u64(uint64_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) {__p0};
return __ret;
}
-#else
-__ai uint64x1_t vdup_n_u64(uint64_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) {__p0};
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vdup_n_u16(uint16_t __p0) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vdup_n_s64(int64_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t) {__p0};
- return __ret;
-}
-#else
__ai int64x1_t vdup_n_s64(int64_t __p0) {
int64x1_t __ret;
__ret = (int64x1_t) {__p0};
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vdup_n_s16(int16_t __p0) {
int16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t veor_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = __p0 ^ __p1;
- return __ret;
-}
-#else
__ai uint64x1_t veor_u64(uint64x1_t __p0, uint64x1_t __p1) {
uint64x1_t __ret;
__ret = __p0 ^ __p1;
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t veor_u16(uint16x4_t __p0, uint16x4_t __p1) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai int64x1_t veor_s64(int64x1_t __p0, int64x1_t __p1) {
int64x1_t __ret;
__ret = __p0 ^ __p1;
return __ret;
}
-#else
-__ai int64x1_t veor_s64(int64x1_t __p0, int64x1_t __p1) {
- int64x1_t __ret;
- __ret = __p0 ^ __p1;
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
int16x4_t __ret;
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vext_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64x1_t __s1 = __p1; \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 19); \
- __ret; \
-})
-#else
#define vext_u64(__p0, __p1, __p2) __extension__ ({ \
uint64x1_t __s0 = __p0; \
uint64x1_t __s1 = __p1; \
__ret = (uint64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 19); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vext_u16(__p0, __p1, __p2) __extension__ ({ \
uint16x4_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vext_s64(__p0, __p1, __p2) __extension__ ({ \
int64x1_t __s0 = __p0; \
int64x1_t __s1 = __p1; \
__ret = (int64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 3); \
__ret; \
})
-#else
-#define vext_s64(__p0, __p1, __p2) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64x1_t __s1 = __p1; \
- int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 3); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vext_s16(__p0, __p1, __p2) __extension__ ({ \
int16x4_t __s0 = __p0; \
#define vget_lane_p8(__p0, __p1) __extension__ ({ \
poly8x8_t __s0 = __p0; \
poly8_t __ret; \
- __ret = (poly8_t) __builtin_neon_vget_lane_i8((int8x8_t)__s0, __p1); \
+ __ret = (poly8_t) __builtin_neon_vget_lane_i8((poly8x8_t)__s0, __p1); \
__ret; \
})
#else
poly8x8_t __s0 = __p0; \
poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
poly8_t __ret; \
- __ret = (poly8_t) __builtin_neon_vget_lane_i8((int8x8_t)__rev0, __p1); \
+ __ret = (poly8_t) __builtin_neon_vget_lane_i8((poly8x8_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vget_lane_p8(__p0, __p1) __extension__ ({ \
poly8x8_t __s0 = __p0; \
poly8_t __ret; \
- __ret = (poly8_t) __builtin_neon_vget_lane_i8((int8x8_t)__s0, __p1); \
+ __ret = (poly8_t) __builtin_neon_vget_lane_i8((poly8x8_t)__s0, __p1); \
__ret; \
})
#endif
#define vget_lane_p16(__p0, __p1) __extension__ ({ \
poly16x4_t __s0 = __p0; \
poly16_t __ret; \
- __ret = (poly16_t) __builtin_neon_vget_lane_i16((int8x8_t)__s0, __p1); \
+ __ret = (poly16_t) __builtin_neon_vget_lane_i16((poly16x4_t)__s0, __p1); \
__ret; \
})
#else
poly16x4_t __s0 = __p0; \
poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
poly16_t __ret; \
- __ret = (poly16_t) __builtin_neon_vget_lane_i16((int8x8_t)__rev0, __p1); \
+ __ret = (poly16_t) __builtin_neon_vget_lane_i16((poly16x4_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vget_lane_p16(__p0, __p1) __extension__ ({ \
poly16x4_t __s0 = __p0; \
poly16_t __ret; \
- __ret = (poly16_t) __builtin_neon_vget_lane_i16((int8x8_t)__s0, __p1); \
+ __ret = (poly16_t) __builtin_neon_vget_lane_i16((poly16x4_t)__s0, __p1); \
__ret; \
})
#endif
#define vgetq_lane_p8(__p0, __p1) __extension__ ({ \
poly8x16_t __s0 = __p0; \
poly8_t __ret; \
- __ret = (poly8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__s0, __p1); \
+ __ret = (poly8_t) __builtin_neon_vgetq_lane_i8((poly8x16_t)__s0, __p1); \
__ret; \
})
#else
poly8x16_t __s0 = __p0; \
poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
poly8_t __ret; \
- __ret = (poly8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__rev0, __p1); \
+ __ret = (poly8_t) __builtin_neon_vgetq_lane_i8((poly8x16_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vgetq_lane_p8(__p0, __p1) __extension__ ({ \
poly8x16_t __s0 = __p0; \
poly8_t __ret; \
- __ret = (poly8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__s0, __p1); \
+ __ret = (poly8_t) __builtin_neon_vgetq_lane_i8((poly8x16_t)__s0, __p1); \
__ret; \
})
#endif
#define vgetq_lane_p16(__p0, __p1) __extension__ ({ \
poly16x8_t __s0 = __p0; \
poly16_t __ret; \
- __ret = (poly16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__s0, __p1); \
+ __ret = (poly16_t) __builtin_neon_vgetq_lane_i16((poly16x8_t)__s0, __p1); \
__ret; \
})
#else
poly16x8_t __s0 = __p0; \
poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
poly16_t __ret; \
- __ret = (poly16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__rev0, __p1); \
+ __ret = (poly16_t) __builtin_neon_vgetq_lane_i16((poly16x8_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vgetq_lane_p16(__p0, __p1) __extension__ ({ \
poly16x8_t __s0 = __p0; \
poly16_t __ret; \
- __ret = (poly16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__s0, __p1); \
+ __ret = (poly16_t) __builtin_neon_vgetq_lane_i16((poly16x8_t)__s0, __p1); \
__ret; \
})
#endif
#define vgetq_lane_u32(__p0, __p1) __extension__ ({ \
uint32x4_t __s0 = __p0; \
uint32_t __ret; \
- __ret = (uint32_t) __builtin_neon_vgetq_lane_i32((int8x16_t)__s0, __p1); \
+ __ret = (uint32_t) __builtin_neon_vgetq_lane_i32((int32x4_t)__s0, __p1); \
__ret; \
})
#else
uint32x4_t __s0 = __p0; \
uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
uint32_t __ret; \
- __ret = (uint32_t) __builtin_neon_vgetq_lane_i32((int8x16_t)__rev0, __p1); \
+ __ret = (uint32_t) __builtin_neon_vgetq_lane_i32((int32x4_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vgetq_lane_u32(__p0, __p1) __extension__ ({ \
uint32x4_t __s0 = __p0; \
uint32_t __ret; \
- __ret = (uint32_t) __builtin_neon_vgetq_lane_i32((int8x16_t)__s0, __p1); \
+ __ret = (uint32_t) __builtin_neon_vgetq_lane_i32((int32x4_t)__s0, __p1); \
__ret; \
})
#endif
#define vgetq_lane_u64(__p0, __p1) __extension__ ({ \
uint64x2_t __s0 = __p0; \
uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__s0, __p1); \
+ __ret = (uint64_t) __builtin_neon_vgetq_lane_i64((int64x2_t)__s0, __p1); \
__ret; \
})
#else
uint64x2_t __s0 = __p0; \
uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__rev0, __p1); \
+ __ret = (uint64_t) __builtin_neon_vgetq_lane_i64((int64x2_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vgetq_lane_u64(__p0, __p1) __extension__ ({ \
uint64x2_t __s0 = __p0; \
uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__s0, __p1); \
+ __ret = (uint64_t) __builtin_neon_vgetq_lane_i64((int64x2_t)__s0, __p1); \
__ret; \
})
#endif
#define vgetq_lane_u16(__p0, __p1) __extension__ ({ \
uint16x8_t __s0 = __p0; \
uint16_t __ret; \
- __ret = (uint16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__s0, __p1); \
+ __ret = (uint16_t) __builtin_neon_vgetq_lane_i16((int16x8_t)__s0, __p1); \
__ret; \
})
#else
uint16x8_t __s0 = __p0; \
uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
uint16_t __ret; \
- __ret = (uint16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__rev0, __p1); \
+ __ret = (uint16_t) __builtin_neon_vgetq_lane_i16((int16x8_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vgetq_lane_u16(__p0, __p1) __extension__ ({ \
uint16x8_t __s0 = __p0; \
uint16_t __ret; \
- __ret = (uint16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__s0, __p1); \
+ __ret = (uint16_t) __builtin_neon_vgetq_lane_i16((int16x8_t)__s0, __p1); \
__ret; \
})
#endif
#define vgetq_lane_f32(__p0, __p1) __extension__ ({ \
float32x4_t __s0 = __p0; \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vgetq_lane_f32((int8x16_t)__s0, __p1); \
+ __ret = (float32_t) __builtin_neon_vgetq_lane_f32((float32x4_t)__s0, __p1); \
__ret; \
})
#else
float32x4_t __s0 = __p0; \
float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vgetq_lane_f32((int8x16_t)__rev0, __p1); \
+ __ret = (float32_t) __builtin_neon_vgetq_lane_f32((float32x4_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vgetq_lane_f32(__p0, __p1) __extension__ ({ \
float32x4_t __s0 = __p0; \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vgetq_lane_f32((int8x16_t)__s0, __p1); \
+ __ret = (float32_t) __builtin_neon_vgetq_lane_f32((float32x4_t)__s0, __p1); \
__ret; \
})
#endif
#define vgetq_lane_s32(__p0, __p1) __extension__ ({ \
int32x4_t __s0 = __p0; \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vgetq_lane_i32((int8x16_t)__s0, __p1); \
+ __ret = (int32_t) __builtin_neon_vgetq_lane_i32((int32x4_t)__s0, __p1); \
__ret; \
})
#else
int32x4_t __s0 = __p0; \
int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vgetq_lane_i32((int8x16_t)__rev0, __p1); \
+ __ret = (int32_t) __builtin_neon_vgetq_lane_i32((int32x4_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vgetq_lane_s32(__p0, __p1) __extension__ ({ \
int32x4_t __s0 = __p0; \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vgetq_lane_i32((int8x16_t)__s0, __p1); \
+ __ret = (int32_t) __builtin_neon_vgetq_lane_i32((int32x4_t)__s0, __p1); \
__ret; \
})
#endif
#define vgetq_lane_s64(__p0, __p1) __extension__ ({ \
int64x2_t __s0 = __p0; \
int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__s0, __p1); \
+ __ret = (int64_t) __builtin_neon_vgetq_lane_i64((int64x2_t)__s0, __p1); \
__ret; \
})
#else
int64x2_t __s0 = __p0; \
int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__rev0, __p1); \
+ __ret = (int64_t) __builtin_neon_vgetq_lane_i64((int64x2_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vgetq_lane_s64(__p0, __p1) __extension__ ({ \
int64x2_t __s0 = __p0; \
int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__s0, __p1); \
+ __ret = (int64_t) __builtin_neon_vgetq_lane_i64((int64x2_t)__s0, __p1); \
__ret; \
})
#endif
#define vgetq_lane_s16(__p0, __p1) __extension__ ({ \
int16x8_t __s0 = __p0; \
int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__s0, __p1); \
+ __ret = (int16_t) __builtin_neon_vgetq_lane_i16((int16x8_t)__s0, __p1); \
__ret; \
})
#else
int16x8_t __s0 = __p0; \
int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__rev0, __p1); \
+ __ret = (int16_t) __builtin_neon_vgetq_lane_i16((int16x8_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vgetq_lane_s16(__p0, __p1) __extension__ ({ \
int16x8_t __s0 = __p0; \
int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__s0, __p1); \
+ __ret = (int16_t) __builtin_neon_vgetq_lane_i16((int16x8_t)__s0, __p1); \
__ret; \
})
#endif
#define vget_lane_u32(__p0, __p1) __extension__ ({ \
uint32x2_t __s0 = __p0; \
uint32_t __ret; \
- __ret = (uint32_t) __builtin_neon_vget_lane_i32((int8x8_t)__s0, __p1); \
+ __ret = (uint32_t) __builtin_neon_vget_lane_i32((int32x2_t)__s0, __p1); \
__ret; \
})
#else
uint32x2_t __s0 = __p0; \
uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
uint32_t __ret; \
- __ret = (uint32_t) __builtin_neon_vget_lane_i32((int8x8_t)__rev0, __p1); \
+ __ret = (uint32_t) __builtin_neon_vget_lane_i32((int32x2_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vget_lane_u32(__p0, __p1) __extension__ ({ \
uint32x2_t __s0 = __p0; \
uint32_t __ret; \
- __ret = (uint32_t) __builtin_neon_vget_lane_i32((int8x8_t)__s0, __p1); \
+ __ret = (uint32_t) __builtin_neon_vget_lane_i32((int32x2_t)__s0, __p1); \
__ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vget_lane_u64(__p0, __p1) __extension__ ({ \
uint64x1_t __s0 = __p0; \
uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
+ __ret = (uint64_t) __builtin_neon_vget_lane_i64((int64x1_t)__s0, __p1); \
__ret; \
})
-#else
-#define vget_lane_u64(__p0, __p1) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
- __ret; \
-})
-#define __noswap_vget_lane_u64(__p0, __p1) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vget_lane_u16(__p0, __p1) __extension__ ({ \
uint16x4_t __s0 = __p0; \
uint16_t __ret; \
- __ret = (uint16_t) __builtin_neon_vget_lane_i16((int8x8_t)__s0, __p1); \
+ __ret = (uint16_t) __builtin_neon_vget_lane_i16((int16x4_t)__s0, __p1); \
__ret; \
})
#else
uint16x4_t __s0 = __p0; \
uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
uint16_t __ret; \
- __ret = (uint16_t) __builtin_neon_vget_lane_i16((int8x8_t)__rev0, __p1); \
+ __ret = (uint16_t) __builtin_neon_vget_lane_i16((int16x4_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vget_lane_u16(__p0, __p1) __extension__ ({ \
uint16x4_t __s0 = __p0; \
uint16_t __ret; \
- __ret = (uint16_t) __builtin_neon_vget_lane_i16((int8x8_t)__s0, __p1); \
+ __ret = (uint16_t) __builtin_neon_vget_lane_i16((int16x4_t)__s0, __p1); \
__ret; \
})
#endif
#define vget_lane_f32(__p0, __p1) __extension__ ({ \
float32x2_t __s0 = __p0; \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vget_lane_f32((int8x8_t)__s0, __p1); \
+ __ret = (float32_t) __builtin_neon_vget_lane_f32((float32x2_t)__s0, __p1); \
__ret; \
})
#else
float32x2_t __s0 = __p0; \
float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vget_lane_f32((int8x8_t)__rev0, __p1); \
+ __ret = (float32_t) __builtin_neon_vget_lane_f32((float32x2_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vget_lane_f32(__p0, __p1) __extension__ ({ \
float32x2_t __s0 = __p0; \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vget_lane_f32((int8x8_t)__s0, __p1); \
+ __ret = (float32_t) __builtin_neon_vget_lane_f32((float32x2_t)__s0, __p1); \
__ret; \
})
#endif
#define vget_lane_s32(__p0, __p1) __extension__ ({ \
int32x2_t __s0 = __p0; \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vget_lane_i32((int8x8_t)__s0, __p1); \
+ __ret = (int32_t) __builtin_neon_vget_lane_i32((int32x2_t)__s0, __p1); \
__ret; \
})
#else
int32x2_t __s0 = __p0; \
int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vget_lane_i32((int8x8_t)__rev0, __p1); \
+ __ret = (int32_t) __builtin_neon_vget_lane_i32((int32x2_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vget_lane_s32(__p0, __p1) __extension__ ({ \
int32x2_t __s0 = __p0; \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vget_lane_i32((int8x8_t)__s0, __p1); \
+ __ret = (int32_t) __builtin_neon_vget_lane_i32((int32x2_t)__s0, __p1); \
__ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vget_lane_s64(__p0, __p1) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
- __ret; \
-})
-#else
#define vget_lane_s64(__p0, __p1) __extension__ ({ \
int64x1_t __s0 = __p0; \
int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
+ __ret = (int64_t) __builtin_neon_vget_lane_i64((int64x1_t)__s0, __p1); \
__ret; \
})
-#define __noswap_vget_lane_s64(__p0, __p1) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vget_lane_s16(__p0, __p1) __extension__ ({ \
int16x4_t __s0 = __p0; \
int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vget_lane_i16((int8x8_t)__s0, __p1); \
+ __ret = (int16_t) __builtin_neon_vget_lane_i16((int16x4_t)__s0, __p1); \
__ret; \
})
#else
int16x4_t __s0 = __p0; \
int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vget_lane_i16((int8x8_t)__rev0, __p1); \
+ __ret = (int16_t) __builtin_neon_vget_lane_i16((int16x4_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vget_lane_s16(__p0, __p1) __extension__ ({ \
int16x4_t __s0 = __p0; \
int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vget_lane_i16((int8x8_t)__s0, __p1); \
+ __ret = (int16_t) __builtin_neon_vget_lane_i16((int16x4_t)__s0, __p1); \
__ret; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld1q_f16(__p0) __extension__ ({ \
- float16x8_t __ret; \
- __ret = (float16x8_t) __builtin_neon_vld1q_v(__p0, 40); \
- __ret; \
-})
-#else
-#define vld1q_f16(__p0) __extension__ ({ \
- float16x8_t __ret; \
- __ret = (float16x8_t) __builtin_neon_vld1q_v(__p0, 40); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld1q_s32(__p0) __extension__ ({ \
int32x4_t __ret; \
__ret = (int32x4_t) __builtin_neon_vld1q_v(__p0, 34); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vld1_u64(__p0) __extension__ ({ \
uint64x1_t __ret; \
__ret = (uint64x1_t) __builtin_neon_vld1_v(__p0, 19); \
__ret; \
})
-#else
-#define vld1_u64(__p0) __extension__ ({ \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vld1_v(__p0, 19); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1_u16(__p0) __extension__ ({ \
uint16x4_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld1_f16(__p0) __extension__ ({ \
- float16x4_t __ret; \
- __ret = (float16x4_t) __builtin_neon_vld1_v(__p0, 8); \
- __ret; \
-})
-#else
-#define vld1_f16(__p0) __extension__ ({ \
- float16x4_t __ret; \
- __ret = (float16x4_t) __builtin_neon_vld1_v(__p0, 8); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld1_s32(__p0) __extension__ ({ \
int32x2_t __ret; \
__ret = (int32x2_t) __builtin_neon_vld1_v(__p0, 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld1_s64(__p0) __extension__ ({ \
- int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vld1_v(__p0, 3); \
- __ret; \
-})
-#else
#define vld1_s64(__p0) __extension__ ({ \
int64x1_t __ret; \
__ret = (int64x1_t) __builtin_neon_vld1_v(__p0, 3); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1_s16(__p0) __extension__ ({ \
int16x4_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld1q_dup_f16(__p0) __extension__ ({ \
- float16x8_t __ret; \
- __ret = (float16x8_t) __builtin_neon_vld1q_dup_v(__p0, 40); \
- __ret; \
-})
-#else
-#define vld1q_dup_f16(__p0) __extension__ ({ \
- float16x8_t __ret; \
- __ret = (float16x8_t) __builtin_neon_vld1q_dup_v(__p0, 40); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld1q_dup_s32(__p0) __extension__ ({ \
int32x4_t __ret; \
__ret = (int32x4_t) __builtin_neon_vld1q_dup_v(__p0, 34); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld1_dup_u64(__p0) __extension__ ({ \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vld1_dup_v(__p0, 19); \
- __ret; \
-})
-#else
#define vld1_dup_u64(__p0) __extension__ ({ \
uint64x1_t __ret; \
__ret = (uint64x1_t) __builtin_neon_vld1_dup_v(__p0, 19); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1_dup_u16(__p0) __extension__ ({ \
uint16x4_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld1_dup_f16(__p0) __extension__ ({ \
- float16x4_t __ret; \
- __ret = (float16x4_t) __builtin_neon_vld1_dup_v(__p0, 8); \
- __ret; \
-})
-#else
-#define vld1_dup_f16(__p0) __extension__ ({ \
- float16x4_t __ret; \
- __ret = (float16x4_t) __builtin_neon_vld1_dup_v(__p0, 8); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld1_dup_s32(__p0) __extension__ ({ \
int32x2_t __ret; \
__ret = (int32x2_t) __builtin_neon_vld1_dup_v(__p0, 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld1_dup_s64(__p0) __extension__ ({ \
- int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vld1_dup_v(__p0, 3); \
- __ret; \
-})
-#else
#define vld1_dup_s64(__p0) __extension__ ({ \
int64x1_t __ret; \
__ret = (int64x1_t) __builtin_neon_vld1_dup_v(__p0, 3); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1_dup_s16(__p0) __extension__ ({ \
int16x4_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8_t __s1 = __p1; \
- float16x8_t __ret; \
- __ret = (float16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 40); \
- __ret; \
-})
-#else
-#define vld1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8_t __s1 = __p1; \
- float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x8_t __ret; \
- __ret = (float16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 40); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld1q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x4_t __s1 = __p1; \
int32x4_t __ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld1_lane_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64x1_t __s1 = __p1; \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 19); \
- __ret; \
-})
-#else
#define vld1_lane_u64(__p0, __p1, __p2) __extension__ ({ \
uint64x1_t __s1 = __p1; \
uint64x1_t __ret; \
__ret = (uint64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 19); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1_lane_u16(__p0, __p1, __p2) __extension__ ({ \
uint16x4_t __s1 = __p1; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4_t __s1 = __p1; \
- float16x4_t __ret; \
- __ret = (float16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 8); \
- __ret; \
-})
-#else
-#define vld1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4_t __s1 = __p1; \
- float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- float16x4_t __ret; \
- __ret = (float16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 8); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld1_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x2_t __s1 = __p1; \
int32x2_t __ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld1_lane_s64(__p0, __p1, __p2) __extension__ ({ \
- int64x1_t __s1 = __p1; \
- int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 3); \
- __ret; \
-})
-#else
#define vld1_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64x1_t __s1 = __p1; \
int64x1_t __ret; \
__ret = (int64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 3); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x4_t __s1 = __p1; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld1q_f16_x2(__p0) __extension__ ({ \
- float16x8x2_t __ret; \
- __builtin_neon_vld1q_x2_v(&__ret, __p0, 40); \
- __ret; \
-})
-#else
-#define vld1q_f16_x2(__p0) __extension__ ({ \
- float16x8x2_t __ret; \
- __builtin_neon_vld1q_x2_v(&__ret, __p0, 40); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld1q_s32_x2(__p0) __extension__ ({ \
int32x4x2_t __ret; \
__builtin_neon_vld1q_x2_v(&__ret, __p0, 34); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vld1_u64_x2(__p0) __extension__ ({ \
uint64x1x2_t __ret; \
__builtin_neon_vld1_x2_v(&__ret, __p0, 19); \
__ret; \
})
-#else
-#define vld1_u64_x2(__p0) __extension__ ({ \
- uint64x1x2_t __ret; \
- __builtin_neon_vld1_x2_v(&__ret, __p0, 19); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1_u16_x2(__p0) __extension__ ({ \
uint16x4x2_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld1_f16_x2(__p0) __extension__ ({ \
- float16x4x2_t __ret; \
- __builtin_neon_vld1_x2_v(&__ret, __p0, 8); \
- __ret; \
-})
-#else
-#define vld1_f16_x2(__p0) __extension__ ({ \
- float16x4x2_t __ret; \
- __builtin_neon_vld1_x2_v(&__ret, __p0, 8); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld1_s32_x2(__p0) __extension__ ({ \
int32x2x2_t __ret; \
__builtin_neon_vld1_x2_v(&__ret, __p0, 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vld1_s64_x2(__p0) __extension__ ({ \
int64x1x2_t __ret; \
__builtin_neon_vld1_x2_v(&__ret, __p0, 3); \
__ret; \
})
-#else
-#define vld1_s64_x2(__p0) __extension__ ({ \
- int64x1x2_t __ret; \
- __builtin_neon_vld1_x2_v(&__ret, __p0, 3); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1_s16_x2(__p0) __extension__ ({ \
int16x4x2_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld1q_f16_x3(__p0) __extension__ ({ \
- float16x8x3_t __ret; \
- __builtin_neon_vld1q_x3_v(&__ret, __p0, 40); \
- __ret; \
-})
-#else
-#define vld1q_f16_x3(__p0) __extension__ ({ \
- float16x8x3_t __ret; \
- __builtin_neon_vld1q_x3_v(&__ret, __p0, 40); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld1q_s32_x3(__p0) __extension__ ({ \
int32x4x3_t __ret; \
__builtin_neon_vld1q_x3_v(&__ret, __p0, 34); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vld1_u64_x3(__p0) __extension__ ({ \
uint64x1x3_t __ret; \
__builtin_neon_vld1_x3_v(&__ret, __p0, 19); \
__ret; \
})
-#else
-#define vld1_u64_x3(__p0) __extension__ ({ \
- uint64x1x3_t __ret; \
- __builtin_neon_vld1_x3_v(&__ret, __p0, 19); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1_u16_x3(__p0) __extension__ ({ \
uint16x4x3_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld1_f16_x3(__p0) __extension__ ({ \
- float16x4x3_t __ret; \
- __builtin_neon_vld1_x3_v(&__ret, __p0, 8); \
- __ret; \
-})
-#else
-#define vld1_f16_x3(__p0) __extension__ ({ \
- float16x4x3_t __ret; \
- __builtin_neon_vld1_x3_v(&__ret, __p0, 8); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld1_s32_x3(__p0) __extension__ ({ \
int32x2x3_t __ret; \
__builtin_neon_vld1_x3_v(&__ret, __p0, 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vld1_s64_x3(__p0) __extension__ ({ \
int64x1x3_t __ret; \
__builtin_neon_vld1_x3_v(&__ret, __p0, 3); \
__ret; \
})
-#else
-#define vld1_s64_x3(__p0) __extension__ ({ \
- int64x1x3_t __ret; \
- __builtin_neon_vld1_x3_v(&__ret, __p0, 3); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1_s16_x3(__p0) __extension__ ({ \
int16x4x3_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld1q_f16_x4(__p0) __extension__ ({ \
- float16x8x4_t __ret; \
- __builtin_neon_vld1q_x4_v(&__ret, __p0, 40); \
- __ret; \
-})
-#else
-#define vld1q_f16_x4(__p0) __extension__ ({ \
- float16x8x4_t __ret; \
- __builtin_neon_vld1q_x4_v(&__ret, __p0, 40); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld1q_s32_x4(__p0) __extension__ ({ \
int32x4x4_t __ret; \
__builtin_neon_vld1q_x4_v(&__ret, __p0, 34); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld1_u64_x4(__p0) __extension__ ({ \
- uint64x1x4_t __ret; \
- __builtin_neon_vld1_x4_v(&__ret, __p0, 19); \
- __ret; \
-})
-#else
#define vld1_u64_x4(__p0) __extension__ ({ \
uint64x1x4_t __ret; \
__builtin_neon_vld1_x4_v(&__ret, __p0, 19); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1_u16_x4(__p0) __extension__ ({ \
uint16x4x4_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld1_f16_x4(__p0) __extension__ ({ \
- float16x4x4_t __ret; \
- __builtin_neon_vld1_x4_v(&__ret, __p0, 8); \
- __ret; \
-})
-#else
-#define vld1_f16_x4(__p0) __extension__ ({ \
- float16x4x4_t __ret; \
- __builtin_neon_vld1_x4_v(&__ret, __p0, 8); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld1_s32_x4(__p0) __extension__ ({ \
int32x2x4_t __ret; \
__builtin_neon_vld1_x4_v(&__ret, __p0, 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vld1_s64_x4(__p0) __extension__ ({ \
int64x1x4_t __ret; \
__builtin_neon_vld1_x4_v(&__ret, __p0, 3); \
__ret; \
})
-#else
-#define vld1_s64_x4(__p0) __extension__ ({ \
- int64x1x4_t __ret; \
- __builtin_neon_vld1_x4_v(&__ret, __p0, 3); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1_s16_x4(__p0) __extension__ ({ \
int16x4x4_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld2q_f16(__p0) __extension__ ({ \
- float16x8x2_t __ret; \
- __builtin_neon_vld2q_v(&__ret, __p0, 40); \
- __ret; \
-})
-#else
-#define vld2q_f16(__p0) __extension__ ({ \
- float16x8x2_t __ret; \
- __builtin_neon_vld2q_v(&__ret, __p0, 40); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld2q_s32(__p0) __extension__ ({ \
int32x4x2_t __ret; \
__builtin_neon_vld2q_v(&__ret, __p0, 34); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vld2_u64(__p0) __extension__ ({ \
uint64x1x2_t __ret; \
__builtin_neon_vld2_v(&__ret, __p0, 19); \
__ret; \
})
-#else
-#define vld2_u64(__p0) __extension__ ({ \
- uint64x1x2_t __ret; \
- __builtin_neon_vld2_v(&__ret, __p0, 19); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld2_u16(__p0) __extension__ ({ \
uint16x4x2_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld2_f16(__p0) __extension__ ({ \
- float16x4x2_t __ret; \
- __builtin_neon_vld2_v(&__ret, __p0, 8); \
- __ret; \
-})
-#else
-#define vld2_f16(__p0) __extension__ ({ \
- float16x4x2_t __ret; \
- __builtin_neon_vld2_v(&__ret, __p0, 8); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld2_s32(__p0) __extension__ ({ \
int32x2x2_t __ret; \
__builtin_neon_vld2_v(&__ret, __p0, 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vld2_s64(__p0) __extension__ ({ \
int64x1x2_t __ret; \
__builtin_neon_vld2_v(&__ret, __p0, 3); \
__ret; \
})
-#else
-#define vld2_s64(__p0) __extension__ ({ \
- int64x1x2_t __ret; \
- __builtin_neon_vld2_v(&__ret, __p0, 3); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld2_s16(__p0) __extension__ ({ \
int16x4x2_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
+#define vld2q_dup_p8(__p0) __extension__ ({ \
+ poly8x16x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 36); \
+ __ret; \
+})
+#else
+#define vld2q_dup_p8(__p0) __extension__ ({ \
+ poly8x16x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 36); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld2q_dup_p16(__p0) __extension__ ({ \
+ poly16x8x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 37); \
+ __ret; \
+})
+#else
+#define vld2q_dup_p16(__p0) __extension__ ({ \
+ poly16x8x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 37); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld2q_dup_u8(__p0) __extension__ ({ \
+ uint8x16x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 48); \
+ __ret; \
+})
+#else
+#define vld2q_dup_u8(__p0) __extension__ ({ \
+ uint8x16x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 48); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld2q_dup_u32(__p0) __extension__ ({ \
+ uint32x4x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 50); \
+ __ret; \
+})
+#else
+#define vld2q_dup_u32(__p0) __extension__ ({ \
+ uint32x4x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 50); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld2q_dup_u64(__p0) __extension__ ({ \
+ uint64x2x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 51); \
+ __ret; \
+})
+#else
+#define vld2q_dup_u64(__p0) __extension__ ({ \
+ uint64x2x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 51); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld2q_dup_u16(__p0) __extension__ ({ \
+ uint16x8x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 49); \
+ __ret; \
+})
+#else
+#define vld2q_dup_u16(__p0) __extension__ ({ \
+ uint16x8x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 49); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld2q_dup_s8(__p0) __extension__ ({ \
+ int8x16x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 32); \
+ __ret; \
+})
+#else
+#define vld2q_dup_s8(__p0) __extension__ ({ \
+ int8x16x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 32); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld2q_dup_f32(__p0) __extension__ ({ \
+ float32x4x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 41); \
+ __ret; \
+})
+#else
+#define vld2q_dup_f32(__p0) __extension__ ({ \
+ float32x4x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 41); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld2q_dup_s32(__p0) __extension__ ({ \
+ int32x4x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 34); \
+ __ret; \
+})
+#else
+#define vld2q_dup_s32(__p0) __extension__ ({ \
+ int32x4x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 34); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld2q_dup_s64(__p0) __extension__ ({ \
+ int64x2x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 35); \
+ __ret; \
+})
+#else
+#define vld2q_dup_s64(__p0) __extension__ ({ \
+ int64x2x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 35); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld2q_dup_s16(__p0) __extension__ ({ \
+ int16x8x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 33); \
+ __ret; \
+})
+#else
+#define vld2q_dup_s16(__p0) __extension__ ({ \
+ int16x8x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 33); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
#define vld2_dup_u8(__p0) __extension__ ({ \
uint8x8x2_t __ret; \
__builtin_neon_vld2_dup_v(&__ret, __p0, 16); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vld2_dup_u64(__p0) __extension__ ({ \
uint64x1x2_t __ret; \
__builtin_neon_vld2_dup_v(&__ret, __p0, 19); \
__ret; \
})
-#else
-#define vld2_dup_u64(__p0) __extension__ ({ \
- uint64x1x2_t __ret; \
- __builtin_neon_vld2_dup_v(&__ret, __p0, 19); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld2_dup_u16(__p0) __extension__ ({ \
uint16x4x2_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld2_dup_f16(__p0) __extension__ ({ \
- float16x4x2_t __ret; \
- __builtin_neon_vld2_dup_v(&__ret, __p0, 8); \
- __ret; \
-})
-#else
-#define vld2_dup_f16(__p0) __extension__ ({ \
- float16x4x2_t __ret; \
- __builtin_neon_vld2_dup_v(&__ret, __p0, 8); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld2_dup_s32(__p0) __extension__ ({ \
int32x2x2_t __ret; \
__builtin_neon_vld2_dup_v(&__ret, __p0, 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vld2_dup_s64(__p0) __extension__ ({ \
int64x1x2_t __ret; \
__builtin_neon_vld2_dup_v(&__ret, __p0, 3); \
__ret; \
})
-#else
-#define vld2_dup_s64(__p0) __extension__ ({ \
- int64x1x2_t __ret; \
- __builtin_neon_vld2_dup_v(&__ret, __p0, 3); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld2_dup_s16(__p0) __extension__ ({ \
int16x4x2_t __ret; \
#define vld2q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
float32x4x2_t __s1 = __p1; \
float32x4x2_t __ret; \
- __builtin_neon_vld2q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __p2, 41); \
+ __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 41); \
__ret; \
})
#else
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
float32x4x2_t __ret; \
- __builtin_neon_vld2q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __p2, 41); \
+ __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 41); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8x2_t __s1 = __p1; \
- float16x8x2_t __ret; \
- __builtin_neon_vld2q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __p2, 40); \
- __ret; \
-})
-#else
-#define vld2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8x2_t __s1 = __p1; \
- float16x8x2_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x8x2_t __ret; \
- __builtin_neon_vld2q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __p2, 40); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld2q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x4x2_t __s1 = __p1; \
int32x4x2_t __ret; \
- __builtin_neon_vld2q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __p2, 34); \
+ __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 34); \
__ret; \
})
#else
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
int32x4x2_t __ret; \
- __builtin_neon_vld2q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __p2, 34); \
+ __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 34); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
#define vld2q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x8x2_t __s1 = __p1; \
int16x8x2_t __ret; \
- __builtin_neon_vld2q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __p2, 33); \
+ __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 33); \
__ret; \
})
#else
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
int16x8x2_t __ret; \
- __builtin_neon_vld2q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __p2, 33); \
+ __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 33); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
#define vld2_lane_f32(__p0, __p1, __p2) __extension__ ({ \
float32x2x2_t __s1 = __p1; \
float32x2x2_t __ret; \
- __builtin_neon_vld2_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __p2, 9); \
+ __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 9); \
__ret; \
})
#else
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
float32x2x2_t __ret; \
- __builtin_neon_vld2_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __p2, 9); \
+ __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 9); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4x2_t __s1 = __p1; \
- float16x4x2_t __ret; \
- __builtin_neon_vld2_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __p2, 8); \
- __ret; \
-})
-#else
-#define vld2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4x2_t __s1 = __p1; \
- float16x4x2_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- float16x4x2_t __ret; \
- __builtin_neon_vld2_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __p2, 8); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld2_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x2x2_t __s1 = __p1; \
int32x2x2_t __ret; \
- __builtin_neon_vld2_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __p2, 2); \
+ __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 2); \
__ret; \
})
#else
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
int32x2x2_t __ret; \
- __builtin_neon_vld2_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __p2, 2); \
+ __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 2); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
#define vld2_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x4x2_t __s1 = __p1; \
int16x4x2_t __ret; \
- __builtin_neon_vld2_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __p2, 1); \
+ __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 1); \
__ret; \
})
#else
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
int16x4x2_t __ret; \
- __builtin_neon_vld2_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __p2, 1); \
+ __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 1); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld3q_f16(__p0) __extension__ ({ \
- float16x8x3_t __ret; \
- __builtin_neon_vld3q_v(&__ret, __p0, 40); \
- __ret; \
-})
-#else
-#define vld3q_f16(__p0) __extension__ ({ \
- float16x8x3_t __ret; \
- __builtin_neon_vld3q_v(&__ret, __p0, 40); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld3q_s32(__p0) __extension__ ({ \
int32x4x3_t __ret; \
__builtin_neon_vld3q_v(&__ret, __p0, 34); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld3_u64(__p0) __extension__ ({ \
- uint64x1x3_t __ret; \
- __builtin_neon_vld3_v(&__ret, __p0, 19); \
- __ret; \
-})
-#else
#define vld3_u64(__p0) __extension__ ({ \
uint64x1x3_t __ret; \
__builtin_neon_vld3_v(&__ret, __p0, 19); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld3_u16(__p0) __extension__ ({ \
uint16x4x3_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld3_f16(__p0) __extension__ ({ \
- float16x4x3_t __ret; \
- __builtin_neon_vld3_v(&__ret, __p0, 8); \
- __ret; \
-})
-#else
-#define vld3_f16(__p0) __extension__ ({ \
- float16x4x3_t __ret; \
- __builtin_neon_vld3_v(&__ret, __p0, 8); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld3_s32(__p0) __extension__ ({ \
int32x2x3_t __ret; \
__builtin_neon_vld3_v(&__ret, __p0, 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vld3_s64(__p0) __extension__ ({ \
int64x1x3_t __ret; \
__builtin_neon_vld3_v(&__ret, __p0, 3); \
__ret; \
})
-#else
-#define vld3_s64(__p0) __extension__ ({ \
- int64x1x3_t __ret; \
- __builtin_neon_vld3_v(&__ret, __p0, 3); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld3_s16(__p0) __extension__ ({ \
int16x4x3_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
+#define vld3q_dup_p8(__p0) __extension__ ({ \
+ poly8x16x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 36); \
+ __ret; \
+})
+#else
+#define vld3q_dup_p8(__p0) __extension__ ({ \
+ poly8x16x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 36); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld3q_dup_p16(__p0) __extension__ ({ \
+ poly16x8x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 37); \
+ __ret; \
+})
+#else
+#define vld3q_dup_p16(__p0) __extension__ ({ \
+ poly16x8x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 37); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld3q_dup_u8(__p0) __extension__ ({ \
+ uint8x16x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 48); \
+ __ret; \
+})
+#else
+#define vld3q_dup_u8(__p0) __extension__ ({ \
+ uint8x16x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 48); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld3q_dup_u32(__p0) __extension__ ({ \
+ uint32x4x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 50); \
+ __ret; \
+})
+#else
+#define vld3q_dup_u32(__p0) __extension__ ({ \
+ uint32x4x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 50); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld3q_dup_u64(__p0) __extension__ ({ \
+ uint64x2x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 51); \
+ __ret; \
+})
+#else
+#define vld3q_dup_u64(__p0) __extension__ ({ \
+ uint64x2x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 51); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld3q_dup_u16(__p0) __extension__ ({ \
+ uint16x8x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 49); \
+ __ret; \
+})
+#else
+#define vld3q_dup_u16(__p0) __extension__ ({ \
+ uint16x8x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 49); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld3q_dup_s8(__p0) __extension__ ({ \
+ int8x16x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 32); \
+ __ret; \
+})
+#else
+#define vld3q_dup_s8(__p0) __extension__ ({ \
+ int8x16x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 32); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld3q_dup_f32(__p0) __extension__ ({ \
+ float32x4x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 41); \
+ __ret; \
+})
+#else
+#define vld3q_dup_f32(__p0) __extension__ ({ \
+ float32x4x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 41); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld3q_dup_s32(__p0) __extension__ ({ \
+ int32x4x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 34); \
+ __ret; \
+})
+#else
+#define vld3q_dup_s32(__p0) __extension__ ({ \
+ int32x4x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 34); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld3q_dup_s64(__p0) __extension__ ({ \
+ int64x2x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 35); \
+ __ret; \
+})
+#else
+#define vld3q_dup_s64(__p0) __extension__ ({ \
+ int64x2x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 35); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld3q_dup_s16(__p0) __extension__ ({ \
+ int16x8x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 33); \
+ __ret; \
+})
+#else
+#define vld3q_dup_s16(__p0) __extension__ ({ \
+ int16x8x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 33); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
#define vld3_dup_u8(__p0) __extension__ ({ \
uint8x8x3_t __ret; \
__builtin_neon_vld3_dup_v(&__ret, __p0, 16); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vld3_dup_u64(__p0) __extension__ ({ \
uint64x1x3_t __ret; \
__builtin_neon_vld3_dup_v(&__ret, __p0, 19); \
__ret; \
})
-#else
-#define vld3_dup_u64(__p0) __extension__ ({ \
- uint64x1x3_t __ret; \
- __builtin_neon_vld3_dup_v(&__ret, __p0, 19); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld3_dup_u16(__p0) __extension__ ({ \
uint16x4x3_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld3_dup_f16(__p0) __extension__ ({ \
- float16x4x3_t __ret; \
- __builtin_neon_vld3_dup_v(&__ret, __p0, 8); \
- __ret; \
-})
-#else
-#define vld3_dup_f16(__p0) __extension__ ({ \
- float16x4x3_t __ret; \
- __builtin_neon_vld3_dup_v(&__ret, __p0, 8); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld3_dup_s32(__p0) __extension__ ({ \
int32x2x3_t __ret; \
__builtin_neon_vld3_dup_v(&__ret, __p0, 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld3_dup_s64(__p0) __extension__ ({ \
- int64x1x3_t __ret; \
- __builtin_neon_vld3_dup_v(&__ret, __p0, 3); \
- __ret; \
-})
-#else
#define vld3_dup_s64(__p0) __extension__ ({ \
int64x1x3_t __ret; \
__builtin_neon_vld3_dup_v(&__ret, __p0, 3); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld3_dup_s16(__p0) __extension__ ({ \
int16x4x3_t __ret; \
#define vld3q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
float32x4x3_t __s1 = __p1; \
float32x4x3_t __ret; \
- __builtin_neon_vld3q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 41); \
+ __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 41); \
__ret; \
})
#else
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
float32x4x3_t __ret; \
- __builtin_neon_vld3q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 41); \
+ __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 41); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8x3_t __s1 = __p1; \
- float16x8x3_t __ret; \
- __builtin_neon_vld3q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 40); \
- __ret; \
-})
-#else
-#define vld3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8x3_t __s1 = __p1; \
- float16x8x3_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x8x3_t __ret; \
- __builtin_neon_vld3q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 40); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld3q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x4x3_t __s1 = __p1; \
int32x4x3_t __ret; \
- __builtin_neon_vld3q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 34); \
+ __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 34); \
__ret; \
})
#else
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
int32x4x3_t __ret; \
- __builtin_neon_vld3q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 34); \
+ __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 34); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
#define vld3q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x8x3_t __s1 = __p1; \
int16x8x3_t __ret; \
- __builtin_neon_vld3q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 33); \
+ __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 33); \
__ret; \
})
#else
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
int16x8x3_t __ret; \
- __builtin_neon_vld3q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 33); \
+ __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 33); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
#define vld3_lane_f32(__p0, __p1, __p2) __extension__ ({ \
float32x2x3_t __s1 = __p1; \
float32x2x3_t __ret; \
- __builtin_neon_vld3_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 9); \
+ __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 9); \
__ret; \
})
#else
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
float32x2x3_t __ret; \
- __builtin_neon_vld3_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 9); \
+ __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 9); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4x3_t __s1 = __p1; \
- float16x4x3_t __ret; \
- __builtin_neon_vld3_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 8); \
- __ret; \
-})
-#else
-#define vld3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4x3_t __s1 = __p1; \
- float16x4x3_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- float16x4x3_t __ret; \
- __builtin_neon_vld3_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 8); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld3_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x2x3_t __s1 = __p1; \
int32x2x3_t __ret; \
- __builtin_neon_vld3_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 2); \
+ __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 2); \
__ret; \
})
#else
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
int32x2x3_t __ret; \
- __builtin_neon_vld3_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 2); \
+ __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 2); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
#define vld3_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x4x3_t __s1 = __p1; \
int16x4x3_t __ret; \
- __builtin_neon_vld3_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 1); \
+ __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 1); \
__ret; \
})
#else
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
int16x4x3_t __ret; \
- __builtin_neon_vld3_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 1); \
+ __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 1); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld4q_f16(__p0) __extension__ ({ \
- float16x8x4_t __ret; \
- __builtin_neon_vld4q_v(&__ret, __p0, 40); \
- __ret; \
-})
-#else
-#define vld4q_f16(__p0) __extension__ ({ \
- float16x8x4_t __ret; \
- __builtin_neon_vld4q_v(&__ret, __p0, 40); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld4q_s32(__p0) __extension__ ({ \
int32x4x4_t __ret; \
__builtin_neon_vld4q_v(&__ret, __p0, 34); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld4_u64(__p0) __extension__ ({ \
- uint64x1x4_t __ret; \
- __builtin_neon_vld4_v(&__ret, __p0, 19); \
- __ret; \
-})
-#else
#define vld4_u64(__p0) __extension__ ({ \
uint64x1x4_t __ret; \
__builtin_neon_vld4_v(&__ret, __p0, 19); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld4_u16(__p0) __extension__ ({ \
uint16x4x4_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld4_f16(__p0) __extension__ ({ \
- float16x4x4_t __ret; \
- __builtin_neon_vld4_v(&__ret, __p0, 8); \
- __ret; \
-})
-#else
-#define vld4_f16(__p0) __extension__ ({ \
- float16x4x4_t __ret; \
- __builtin_neon_vld4_v(&__ret, __p0, 8); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld4_s32(__p0) __extension__ ({ \
int32x2x4_t __ret; \
__builtin_neon_vld4_v(&__ret, __p0, 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld4_s64(__p0) __extension__ ({ \
- int64x1x4_t __ret; \
- __builtin_neon_vld4_v(&__ret, __p0, 3); \
- __ret; \
-})
-#else
#define vld4_s64(__p0) __extension__ ({ \
int64x1x4_t __ret; \
__builtin_neon_vld4_v(&__ret, __p0, 3); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld4_s16(__p0) __extension__ ({ \
int16x4x4_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
+#define vld4q_dup_p8(__p0) __extension__ ({ \
+ poly8x16x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 36); \
+ __ret; \
+})
+#else
+#define vld4q_dup_p8(__p0) __extension__ ({ \
+ poly8x16x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 36); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld4q_dup_p16(__p0) __extension__ ({ \
+ poly16x8x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 37); \
+ __ret; \
+})
+#else
+#define vld4q_dup_p16(__p0) __extension__ ({ \
+ poly16x8x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 37); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld4q_dup_u8(__p0) __extension__ ({ \
+ uint8x16x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 48); \
+ __ret; \
+})
+#else
+#define vld4q_dup_u8(__p0) __extension__ ({ \
+ uint8x16x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 48); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld4q_dup_u32(__p0) __extension__ ({ \
+ uint32x4x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 50); \
+ __ret; \
+})
+#else
+#define vld4q_dup_u32(__p0) __extension__ ({ \
+ uint32x4x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 50); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld4q_dup_u64(__p0) __extension__ ({ \
+ uint64x2x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 51); \
+ __ret; \
+})
+#else
+#define vld4q_dup_u64(__p0) __extension__ ({ \
+ uint64x2x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 51); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld4q_dup_u16(__p0) __extension__ ({ \
+ uint16x8x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 49); \
+ __ret; \
+})
+#else
+#define vld4q_dup_u16(__p0) __extension__ ({ \
+ uint16x8x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 49); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld4q_dup_s8(__p0) __extension__ ({ \
+ int8x16x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 32); \
+ __ret; \
+})
+#else
+#define vld4q_dup_s8(__p0) __extension__ ({ \
+ int8x16x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 32); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld4q_dup_f32(__p0) __extension__ ({ \
+ float32x4x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 41); \
+ __ret; \
+})
+#else
+#define vld4q_dup_f32(__p0) __extension__ ({ \
+ float32x4x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 41); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld4q_dup_s32(__p0) __extension__ ({ \
+ int32x4x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 34); \
+ __ret; \
+})
+#else
+#define vld4q_dup_s32(__p0) __extension__ ({ \
+ int32x4x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 34); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld4q_dup_s64(__p0) __extension__ ({ \
+ int64x2x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 35); \
+ __ret; \
+})
+#else
+#define vld4q_dup_s64(__p0) __extension__ ({ \
+ int64x2x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 35); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld4q_dup_s16(__p0) __extension__ ({ \
+ int16x8x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 33); \
+ __ret; \
+})
+#else
+#define vld4q_dup_s16(__p0) __extension__ ({ \
+ int16x8x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 33); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
#define vld4_dup_u8(__p0) __extension__ ({ \
uint8x8x4_t __ret; \
__builtin_neon_vld4_dup_v(&__ret, __p0, 16); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vld4_dup_u64(__p0) __extension__ ({ \
uint64x1x4_t __ret; \
__builtin_neon_vld4_dup_v(&__ret, __p0, 19); \
__ret; \
})
-#else
-#define vld4_dup_u64(__p0) __extension__ ({ \
- uint64x1x4_t __ret; \
- __builtin_neon_vld4_dup_v(&__ret, __p0, 19); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld4_dup_u16(__p0) __extension__ ({ \
uint16x4x4_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld4_dup_f16(__p0) __extension__ ({ \
- float16x4x4_t __ret; \
- __builtin_neon_vld4_dup_v(&__ret, __p0, 8); \
- __ret; \
-})
-#else
-#define vld4_dup_f16(__p0) __extension__ ({ \
- float16x4x4_t __ret; \
- __builtin_neon_vld4_dup_v(&__ret, __p0, 8); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld4_dup_s32(__p0) __extension__ ({ \
int32x2x4_t __ret; \
__builtin_neon_vld4_dup_v(&__ret, __p0, 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld4_dup_s64(__p0) __extension__ ({ \
- int64x1x4_t __ret; \
- __builtin_neon_vld4_dup_v(&__ret, __p0, 3); \
- __ret; \
-})
-#else
#define vld4_dup_s64(__p0) __extension__ ({ \
int64x1x4_t __ret; \
__builtin_neon_vld4_dup_v(&__ret, __p0, 3); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld4_dup_s16(__p0) __extension__ ({ \
int16x4x4_t __ret; \
#define vld4q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
float32x4x4_t __s1 = __p1; \
float32x4x4_t __ret; \
- __builtin_neon_vld4q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 41); \
+ __builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 41); \
__ret; \
})
#else
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
float32x4x4_t __ret; \
- __builtin_neon_vld4q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 41); \
+ __builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 41); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8x4_t __s1 = __p1; \
- float16x8x4_t __ret; \
- __builtin_neon_vld4q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 40); \
- __ret; \
-})
-#else
-#define vld4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8x4_t __s1 = __p1; \
- float16x8x4_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x8x4_t __ret; \
- __builtin_neon_vld4q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 40); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld4q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x4x4_t __s1 = __p1; \
int32x4x4_t __ret; \
- __builtin_neon_vld4q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 34); \
+ __builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 34); \
__ret; \
})
#else
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
int32x4x4_t __ret; \
- __builtin_neon_vld4q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 34); \
+ __builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 34); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
#define vld4q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x8x4_t __s1 = __p1; \
int16x8x4_t __ret; \
- __builtin_neon_vld4q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 33); \
+ __builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 33); \
__ret; \
})
#else
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
int16x8x4_t __ret; \
- __builtin_neon_vld4q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 33); \
+ __builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 33); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
#define vld4_lane_f32(__p0, __p1, __p2) __extension__ ({ \
float32x2x4_t __s1 = __p1; \
float32x2x4_t __ret; \
- __builtin_neon_vld4_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 9); \
+ __builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 9); \
__ret; \
})
#else
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
float32x2x4_t __ret; \
- __builtin_neon_vld4_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 9); \
+ __builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 9); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4x4_t __s1 = __p1; \
- float16x4x4_t __ret; \
- __builtin_neon_vld4_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 8); \
- __ret; \
-})
-#else
-#define vld4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4x4_t __s1 = __p1; \
- float16x4x4_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
- float16x4x4_t __ret; \
- __builtin_neon_vld4_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 8); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld4_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x2x4_t __s1 = __p1; \
int32x2x4_t __ret; \
- __builtin_neon_vld4_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 2); \
+ __builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 2); \
__ret; \
})
#else
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
int32x2x4_t __ret; \
- __builtin_neon_vld4_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 2); \
+ __builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 2); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
#define vld4_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x4x4_t __s1 = __p1; \
int16x4x4_t __ret; \
- __builtin_neon_vld4_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 1); \
+ __builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 1); \
__ret; \
})
#else
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
int16x4x4_t __ret; \
- __builtin_neon_vld4_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 1); \
+ __builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 1); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlaq_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint32x4_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlaq_lane_u32(__p0_42, __p1_42, __p2_42, __p3_42) __extension__ ({ \
+ uint32x4_t __s0_42 = __p0_42; \
+ uint32x4_t __s1_42 = __p1_42; \
+ uint32x2_t __s2_42 = __p2_42; \
+ uint32x4_t __ret_42; \
+ __ret_42 = __s0_42 + __s1_42 * splatq_lane_u32(__s2_42, __p3_42); \
+ __ret_42; \
})
#else
-#define vmlaq_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlaq_lane_u32(__p0_43, __p1_43, __p2_43, __p3_43) __extension__ ({ \
+ uint32x4_t __s0_43 = __p0_43; \
+ uint32x4_t __s1_43 = __p1_43; \
+ uint32x2_t __s2_43 = __p2_43; \
+ uint32x4_t __rev0_43; __rev0_43 = __builtin_shufflevector(__s0_43, __s0_43, 3, 2, 1, 0); \
+ uint32x4_t __rev1_43; __rev1_43 = __builtin_shufflevector(__s1_43, __s1_43, 3, 2, 1, 0); \
+ uint32x2_t __rev2_43; __rev2_43 = __builtin_shufflevector(__s2_43, __s2_43, 1, 0); \
+ uint32x4_t __ret_43; \
+ __ret_43 = __rev0_43 + __rev1_43 * __noswap_splatq_lane_u32(__rev2_43, __p3_43); \
+ __ret_43 = __builtin_shufflevector(__ret_43, __ret_43, 3, 2, 1, 0); \
+ __ret_43; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlaq_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint16x8_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlaq_lane_u16(__p0_44, __p1_44, __p2_44, __p3_44) __extension__ ({ \
+ uint16x8_t __s0_44 = __p0_44; \
+ uint16x8_t __s1_44 = __p1_44; \
+ uint16x4_t __s2_44 = __p2_44; \
+ uint16x8_t __ret_44; \
+ __ret_44 = __s0_44 + __s1_44 * splatq_lane_u16(__s2_44, __p3_44); \
+ __ret_44; \
})
#else
-#define vmlaq_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint16x8_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmlaq_lane_u16(__p0_45, __p1_45, __p2_45, __p3_45) __extension__ ({ \
+ uint16x8_t __s0_45 = __p0_45; \
+ uint16x8_t __s1_45 = __p1_45; \
+ uint16x4_t __s2_45 = __p2_45; \
+ uint16x8_t __rev0_45; __rev0_45 = __builtin_shufflevector(__s0_45, __s0_45, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __rev1_45; __rev1_45 = __builtin_shufflevector(__s1_45, __s1_45, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x4_t __rev2_45; __rev2_45 = __builtin_shufflevector(__s2_45, __s2_45, 3, 2, 1, 0); \
+ uint16x8_t __ret_45; \
+ __ret_45 = __rev0_45 + __rev1_45 * __noswap_splatq_lane_u16(__rev2_45, __p3_45); \
+ __ret_45 = __builtin_shufflevector(__ret_45, __ret_45, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_45; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x2_t __s2 = __p2; \
- float32x4_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlaq_lane_f32(__p0_46, __p1_46, __p2_46, __p3_46) __extension__ ({ \
+ float32x4_t __s0_46 = __p0_46; \
+ float32x4_t __s1_46 = __p1_46; \
+ float32x2_t __s2_46 = __p2_46; \
+ float32x4_t __ret_46; \
+ __ret_46 = __s0_46 + __s1_46 * splatq_lane_f32(__s2_46, __p3_46); \
+ __ret_46; \
})
#else
-#define vmlaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x2_t __s2 = __p2; \
- float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- float32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- float32x4_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlaq_lane_f32(__p0_47, __p1_47, __p2_47, __p3_47) __extension__ ({ \
+ float32x4_t __s0_47 = __p0_47; \
+ float32x4_t __s1_47 = __p1_47; \
+ float32x2_t __s2_47 = __p2_47; \
+ float32x4_t __rev0_47; __rev0_47 = __builtin_shufflevector(__s0_47, __s0_47, 3, 2, 1, 0); \
+ float32x4_t __rev1_47; __rev1_47 = __builtin_shufflevector(__s1_47, __s1_47, 3, 2, 1, 0); \
+ float32x2_t __rev2_47; __rev2_47 = __builtin_shufflevector(__s2_47, __s2_47, 1, 0); \
+ float32x4_t __ret_47; \
+ __ret_47 = __rev0_47 + __rev1_47 * __noswap_splatq_lane_f32(__rev2_47, __p3_47); \
+ __ret_47 = __builtin_shufflevector(__ret_47, __ret_47, 3, 2, 1, 0); \
+ __ret_47; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlaq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlaq_lane_s32(__p0_48, __p1_48, __p2_48, __p3_48) __extension__ ({ \
+ int32x4_t __s0_48 = __p0_48; \
+ int32x4_t __s1_48 = __p1_48; \
+ int32x2_t __s2_48 = __p2_48; \
+ int32x4_t __ret_48; \
+ __ret_48 = __s0_48 + __s1_48 * splatq_lane_s32(__s2_48, __p3_48); \
+ __ret_48; \
})
#else
-#define vmlaq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int32x4_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlaq_lane_s32(__p0_49, __p1_49, __p2_49, __p3_49) __extension__ ({ \
+ int32x4_t __s0_49 = __p0_49; \
+ int32x4_t __s1_49 = __p1_49; \
+ int32x2_t __s2_49 = __p2_49; \
+ int32x4_t __rev0_49; __rev0_49 = __builtin_shufflevector(__s0_49, __s0_49, 3, 2, 1, 0); \
+ int32x4_t __rev1_49; __rev1_49 = __builtin_shufflevector(__s1_49, __s1_49, 3, 2, 1, 0); \
+ int32x2_t __rev2_49; __rev2_49 = __builtin_shufflevector(__s2_49, __s2_49, 1, 0); \
+ int32x4_t __ret_49; \
+ __ret_49 = __rev0_49 + __rev1_49 * __noswap_splatq_lane_s32(__rev2_49, __p3_49); \
+ __ret_49 = __builtin_shufflevector(__ret_49, __ret_49, 3, 2, 1, 0); \
+ __ret_49; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlaq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x8_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlaq_lane_s16(__p0_50, __p1_50, __p2_50, __p3_50) __extension__ ({ \
+ int16x8_t __s0_50 = __p0_50; \
+ int16x8_t __s1_50 = __p1_50; \
+ int16x4_t __s2_50 = __p2_50; \
+ int16x8_t __ret_50; \
+ __ret_50 = __s0_50 + __s1_50 * splatq_lane_s16(__s2_50, __p3_50); \
+ __ret_50; \
})
#else
-#define vmlaq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int16x8_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmlaq_lane_s16(__p0_51, __p1_51, __p2_51, __p3_51) __extension__ ({ \
+ int16x8_t __s0_51 = __p0_51; \
+ int16x8_t __s1_51 = __p1_51; \
+ int16x4_t __s2_51 = __p2_51; \
+ int16x8_t __rev0_51; __rev0_51 = __builtin_shufflevector(__s0_51, __s0_51, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_51; __rev1_51 = __builtin_shufflevector(__s1_51, __s1_51, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev2_51; __rev2_51 = __builtin_shufflevector(__s2_51, __s2_51, 3, 2, 1, 0); \
+ int16x8_t __ret_51; \
+ __ret_51 = __rev0_51 + __rev1_51 * __noswap_splatq_lane_s16(__rev2_51, __p3_51); \
+ __ret_51 = __builtin_shufflevector(__ret_51, __ret_51, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_51; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmla_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint32x2_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3); \
- __ret; \
+#define vmla_lane_u32(__p0_52, __p1_52, __p2_52, __p3_52) __extension__ ({ \
+ uint32x2_t __s0_52 = __p0_52; \
+ uint32x2_t __s1_52 = __p1_52; \
+ uint32x2_t __s2_52 = __p2_52; \
+ uint32x2_t __ret_52; \
+ __ret_52 = __s0_52 + __s1_52 * splat_lane_u32(__s2_52, __p3_52); \
+ __ret_52; \
})
#else
-#define vmla_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- uint32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- uint32x2_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmla_lane_u32(__p0_53, __p1_53, __p2_53, __p3_53) __extension__ ({ \
+ uint32x2_t __s0_53 = __p0_53; \
+ uint32x2_t __s1_53 = __p1_53; \
+ uint32x2_t __s2_53 = __p2_53; \
+ uint32x2_t __rev0_53; __rev0_53 = __builtin_shufflevector(__s0_53, __s0_53, 1, 0); \
+ uint32x2_t __rev1_53; __rev1_53 = __builtin_shufflevector(__s1_53, __s1_53, 1, 0); \
+ uint32x2_t __rev2_53; __rev2_53 = __builtin_shufflevector(__s2_53, __s2_53, 1, 0); \
+ uint32x2_t __ret_53; \
+ __ret_53 = __rev0_53 + __rev1_53 * __noswap_splat_lane_u32(__rev2_53, __p3_53); \
+ __ret_53 = __builtin_shufflevector(__ret_53, __ret_53, 1, 0); \
+ __ret_53; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmla_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint16x4_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmla_lane_u16(__p0_54, __p1_54, __p2_54, __p3_54) __extension__ ({ \
+ uint16x4_t __s0_54 = __p0_54; \
+ uint16x4_t __s1_54 = __p1_54; \
+ uint16x4_t __s2_54 = __p2_54; \
+ uint16x4_t __ret_54; \
+ __ret_54 = __s0_54 + __s1_54 * splat_lane_u16(__s2_54, __p3_54); \
+ __ret_54; \
})
#else
-#define vmla_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint16x4_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmla_lane_u16(__p0_55, __p1_55, __p2_55, __p3_55) __extension__ ({ \
+ uint16x4_t __s0_55 = __p0_55; \
+ uint16x4_t __s1_55 = __p1_55; \
+ uint16x4_t __s2_55 = __p2_55; \
+ uint16x4_t __rev0_55; __rev0_55 = __builtin_shufflevector(__s0_55, __s0_55, 3, 2, 1, 0); \
+ uint16x4_t __rev1_55; __rev1_55 = __builtin_shufflevector(__s1_55, __s1_55, 3, 2, 1, 0); \
+ uint16x4_t __rev2_55; __rev2_55 = __builtin_shufflevector(__s2_55, __s2_55, 3, 2, 1, 0); \
+ uint16x4_t __ret_55; \
+ __ret_55 = __rev0_55 + __rev1_55 * __noswap_splat_lane_u16(__rev2_55, __p3_55); \
+ __ret_55 = __builtin_shufflevector(__ret_55, __ret_55, 3, 2, 1, 0); \
+ __ret_55; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmla_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x2_t __s2 = __p2; \
- float32x2_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3); \
- __ret; \
+#define vmla_lane_f32(__p0_56, __p1_56, __p2_56, __p3_56) __extension__ ({ \
+ float32x2_t __s0_56 = __p0_56; \
+ float32x2_t __s1_56 = __p1_56; \
+ float32x2_t __s2_56 = __p2_56; \
+ float32x2_t __ret_56; \
+ __ret_56 = __s0_56 + __s1_56 * splat_lane_f32(__s2_56, __p3_56); \
+ __ret_56; \
})
#else
-#define vmla_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x2_t __s2 = __p2; \
- float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- float32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- float32x2_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmla_lane_f32(__p0_57, __p1_57, __p2_57, __p3_57) __extension__ ({ \
+ float32x2_t __s0_57 = __p0_57; \
+ float32x2_t __s1_57 = __p1_57; \
+ float32x2_t __s2_57 = __p2_57; \
+ float32x2_t __rev0_57; __rev0_57 = __builtin_shufflevector(__s0_57, __s0_57, 1, 0); \
+ float32x2_t __rev1_57; __rev1_57 = __builtin_shufflevector(__s1_57, __s1_57, 1, 0); \
+ float32x2_t __rev2_57; __rev2_57 = __builtin_shufflevector(__s2_57, __s2_57, 1, 0); \
+ float32x2_t __ret_57; \
+ __ret_57 = __rev0_57 + __rev1_57 * __noswap_splat_lane_f32(__rev2_57, __p3_57); \
+ __ret_57 = __builtin_shufflevector(__ret_57, __ret_57, 1, 0); \
+ __ret_57; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmla_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x2_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3); \
- __ret; \
+#define vmla_lane_s32(__p0_58, __p1_58, __p2_58, __p3_58) __extension__ ({ \
+ int32x2_t __s0_58 = __p0_58; \
+ int32x2_t __s1_58 = __p1_58; \
+ int32x2_t __s2_58 = __p2_58; \
+ int32x2_t __ret_58; \
+ __ret_58 = __s0_58 + __s1_58 * splat_lane_s32(__s2_58, __p3_58); \
+ __ret_58; \
})
#else
-#define vmla_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int32x2_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmla_lane_s32(__p0_59, __p1_59, __p2_59, __p3_59) __extension__ ({ \
+ int32x2_t __s0_59 = __p0_59; \
+ int32x2_t __s1_59 = __p1_59; \
+ int32x2_t __s2_59 = __p2_59; \
+ int32x2_t __rev0_59; __rev0_59 = __builtin_shufflevector(__s0_59, __s0_59, 1, 0); \
+ int32x2_t __rev1_59; __rev1_59 = __builtin_shufflevector(__s1_59, __s1_59, 1, 0); \
+ int32x2_t __rev2_59; __rev2_59 = __builtin_shufflevector(__s2_59, __s2_59, 1, 0); \
+ int32x2_t __ret_59; \
+ __ret_59 = __rev0_59 + __rev1_59 * __noswap_splat_lane_s32(__rev2_59, __p3_59); \
+ __ret_59 = __builtin_shufflevector(__ret_59, __ret_59, 1, 0); \
+ __ret_59; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmla_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x4_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmla_lane_s16(__p0_60, __p1_60, __p2_60, __p3_60) __extension__ ({ \
+ int16x4_t __s0_60 = __p0_60; \
+ int16x4_t __s1_60 = __p1_60; \
+ int16x4_t __s2_60 = __p2_60; \
+ int16x4_t __ret_60; \
+ __ret_60 = __s0_60 + __s1_60 * splat_lane_s16(__s2_60, __p3_60); \
+ __ret_60; \
})
#else
-#define vmla_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int16x4_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmla_lane_s16(__p0_61, __p1_61, __p2_61, __p3_61) __extension__ ({ \
+ int16x4_t __s0_61 = __p0_61; \
+ int16x4_t __s1_61 = __p1_61; \
+ int16x4_t __s2_61 = __p2_61; \
+ int16x4_t __rev0_61; __rev0_61 = __builtin_shufflevector(__s0_61, __s0_61, 3, 2, 1, 0); \
+ int16x4_t __rev1_61; __rev1_61 = __builtin_shufflevector(__s1_61, __s1_61, 3, 2, 1, 0); \
+ int16x4_t __rev2_61; __rev2_61 = __builtin_shufflevector(__s2_61, __s2_61, 3, 2, 1, 0); \
+ int16x4_t __ret_61; \
+ __ret_61 = __rev0_61 + __rev1_61 * __noswap_splat_lane_s16(__rev2_61, __p3_61); \
+ __ret_61 = __builtin_shufflevector(__ret_61, __ret_61, 3, 2, 1, 0); \
+ __ret_61; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsq_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint32x4_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlsq_lane_u32(__p0_62, __p1_62, __p2_62, __p3_62) __extension__ ({ \
+ uint32x4_t __s0_62 = __p0_62; \
+ uint32x4_t __s1_62 = __p1_62; \
+ uint32x2_t __s2_62 = __p2_62; \
+ uint32x4_t __ret_62; \
+ __ret_62 = __s0_62 - __s1_62 * splatq_lane_u32(__s2_62, __p3_62); \
+ __ret_62; \
})
#else
-#define vmlsq_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlsq_lane_u32(__p0_63, __p1_63, __p2_63, __p3_63) __extension__ ({ \
+ uint32x4_t __s0_63 = __p0_63; \
+ uint32x4_t __s1_63 = __p1_63; \
+ uint32x2_t __s2_63 = __p2_63; \
+ uint32x4_t __rev0_63; __rev0_63 = __builtin_shufflevector(__s0_63, __s0_63, 3, 2, 1, 0); \
+ uint32x4_t __rev1_63; __rev1_63 = __builtin_shufflevector(__s1_63, __s1_63, 3, 2, 1, 0); \
+ uint32x2_t __rev2_63; __rev2_63 = __builtin_shufflevector(__s2_63, __s2_63, 1, 0); \
+ uint32x4_t __ret_63; \
+ __ret_63 = __rev0_63 - __rev1_63 * __noswap_splatq_lane_u32(__rev2_63, __p3_63); \
+ __ret_63 = __builtin_shufflevector(__ret_63, __ret_63, 3, 2, 1, 0); \
+ __ret_63; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsq_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint16x8_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlsq_lane_u16(__p0_64, __p1_64, __p2_64, __p3_64) __extension__ ({ \
+ uint16x8_t __s0_64 = __p0_64; \
+ uint16x8_t __s1_64 = __p1_64; \
+ uint16x4_t __s2_64 = __p2_64; \
+ uint16x8_t __ret_64; \
+ __ret_64 = __s0_64 - __s1_64 * splatq_lane_u16(__s2_64, __p3_64); \
+ __ret_64; \
})
#else
-#define vmlsq_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint16x8_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmlsq_lane_u16(__p0_65, __p1_65, __p2_65, __p3_65) __extension__ ({ \
+ uint16x8_t __s0_65 = __p0_65; \
+ uint16x8_t __s1_65 = __p1_65; \
+ uint16x4_t __s2_65 = __p2_65; \
+ uint16x8_t __rev0_65; __rev0_65 = __builtin_shufflevector(__s0_65, __s0_65, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __rev1_65; __rev1_65 = __builtin_shufflevector(__s1_65, __s1_65, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x4_t __rev2_65; __rev2_65 = __builtin_shufflevector(__s2_65, __s2_65, 3, 2, 1, 0); \
+ uint16x8_t __ret_65; \
+ __ret_65 = __rev0_65 - __rev1_65 * __noswap_splatq_lane_u16(__rev2_65, __p3_65); \
+ __ret_65 = __builtin_shufflevector(__ret_65, __ret_65, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_65; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x2_t __s2 = __p2; \
- float32x4_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlsq_lane_f32(__p0_66, __p1_66, __p2_66, __p3_66) __extension__ ({ \
+ float32x4_t __s0_66 = __p0_66; \
+ float32x4_t __s1_66 = __p1_66; \
+ float32x2_t __s2_66 = __p2_66; \
+ float32x4_t __ret_66; \
+ __ret_66 = __s0_66 - __s1_66 * splatq_lane_f32(__s2_66, __p3_66); \
+ __ret_66; \
})
#else
-#define vmlsq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x2_t __s2 = __p2; \
- float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- float32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- float32x4_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlsq_lane_f32(__p0_67, __p1_67, __p2_67, __p3_67) __extension__ ({ \
+ float32x4_t __s0_67 = __p0_67; \
+ float32x4_t __s1_67 = __p1_67; \
+ float32x2_t __s2_67 = __p2_67; \
+ float32x4_t __rev0_67; __rev0_67 = __builtin_shufflevector(__s0_67, __s0_67, 3, 2, 1, 0); \
+ float32x4_t __rev1_67; __rev1_67 = __builtin_shufflevector(__s1_67, __s1_67, 3, 2, 1, 0); \
+ float32x2_t __rev2_67; __rev2_67 = __builtin_shufflevector(__s2_67, __s2_67, 1, 0); \
+ float32x4_t __ret_67; \
+ __ret_67 = __rev0_67 - __rev1_67 * __noswap_splatq_lane_f32(__rev2_67, __p3_67); \
+ __ret_67 = __builtin_shufflevector(__ret_67, __ret_67, 3, 2, 1, 0); \
+ __ret_67; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlsq_lane_s32(__p0_68, __p1_68, __p2_68, __p3_68) __extension__ ({ \
+ int32x4_t __s0_68 = __p0_68; \
+ int32x4_t __s1_68 = __p1_68; \
+ int32x2_t __s2_68 = __p2_68; \
+ int32x4_t __ret_68; \
+ __ret_68 = __s0_68 - __s1_68 * splatq_lane_s32(__s2_68, __p3_68); \
+ __ret_68; \
})
#else
-#define vmlsq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int32x4_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlsq_lane_s32(__p0_69, __p1_69, __p2_69, __p3_69) __extension__ ({ \
+ int32x4_t __s0_69 = __p0_69; \
+ int32x4_t __s1_69 = __p1_69; \
+ int32x2_t __s2_69 = __p2_69; \
+ int32x4_t __rev0_69; __rev0_69 = __builtin_shufflevector(__s0_69, __s0_69, 3, 2, 1, 0); \
+ int32x4_t __rev1_69; __rev1_69 = __builtin_shufflevector(__s1_69, __s1_69, 3, 2, 1, 0); \
+ int32x2_t __rev2_69; __rev2_69 = __builtin_shufflevector(__s2_69, __s2_69, 1, 0); \
+ int32x4_t __ret_69; \
+ __ret_69 = __rev0_69 - __rev1_69 * __noswap_splatq_lane_s32(__rev2_69, __p3_69); \
+ __ret_69 = __builtin_shufflevector(__ret_69, __ret_69, 3, 2, 1, 0); \
+ __ret_69; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x8_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlsq_lane_s16(__p0_70, __p1_70, __p2_70, __p3_70) __extension__ ({ \
+ int16x8_t __s0_70 = __p0_70; \
+ int16x8_t __s1_70 = __p1_70; \
+ int16x4_t __s2_70 = __p2_70; \
+ int16x8_t __ret_70; \
+ __ret_70 = __s0_70 - __s1_70 * splatq_lane_s16(__s2_70, __p3_70); \
+ __ret_70; \
})
#else
-#define vmlsq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int16x8_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmlsq_lane_s16(__p0_71, __p1_71, __p2_71, __p3_71) __extension__ ({ \
+ int16x8_t __s0_71 = __p0_71; \
+ int16x8_t __s1_71 = __p1_71; \
+ int16x4_t __s2_71 = __p2_71; \
+ int16x8_t __rev0_71; __rev0_71 = __builtin_shufflevector(__s0_71, __s0_71, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_71; __rev1_71 = __builtin_shufflevector(__s1_71, __s1_71, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev2_71; __rev2_71 = __builtin_shufflevector(__s2_71, __s2_71, 3, 2, 1, 0); \
+ int16x8_t __ret_71; \
+ __ret_71 = __rev0_71 - __rev1_71 * __noswap_splatq_lane_s16(__rev2_71, __p3_71); \
+ __ret_71 = __builtin_shufflevector(__ret_71, __ret_71, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_71; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmls_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint32x2_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3); \
- __ret; \
+#define vmls_lane_u32(__p0_72, __p1_72, __p2_72, __p3_72) __extension__ ({ \
+ uint32x2_t __s0_72 = __p0_72; \
+ uint32x2_t __s1_72 = __p1_72; \
+ uint32x2_t __s2_72 = __p2_72; \
+ uint32x2_t __ret_72; \
+ __ret_72 = __s0_72 - __s1_72 * splat_lane_u32(__s2_72, __p3_72); \
+ __ret_72; \
})
#else
-#define vmls_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- uint32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- uint32x2_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmls_lane_u32(__p0_73, __p1_73, __p2_73, __p3_73) __extension__ ({ \
+ uint32x2_t __s0_73 = __p0_73; \
+ uint32x2_t __s1_73 = __p1_73; \
+ uint32x2_t __s2_73 = __p2_73; \
+ uint32x2_t __rev0_73; __rev0_73 = __builtin_shufflevector(__s0_73, __s0_73, 1, 0); \
+ uint32x2_t __rev1_73; __rev1_73 = __builtin_shufflevector(__s1_73, __s1_73, 1, 0); \
+ uint32x2_t __rev2_73; __rev2_73 = __builtin_shufflevector(__s2_73, __s2_73, 1, 0); \
+ uint32x2_t __ret_73; \
+ __ret_73 = __rev0_73 - __rev1_73 * __noswap_splat_lane_u32(__rev2_73, __p3_73); \
+ __ret_73 = __builtin_shufflevector(__ret_73, __ret_73, 1, 0); \
+ __ret_73; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmls_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint16x4_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmls_lane_u16(__p0_74, __p1_74, __p2_74, __p3_74) __extension__ ({ \
+ uint16x4_t __s0_74 = __p0_74; \
+ uint16x4_t __s1_74 = __p1_74; \
+ uint16x4_t __s2_74 = __p2_74; \
+ uint16x4_t __ret_74; \
+ __ret_74 = __s0_74 - __s1_74 * splat_lane_u16(__s2_74, __p3_74); \
+ __ret_74; \
})
#else
-#define vmls_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint16x4_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmls_lane_u16(__p0_75, __p1_75, __p2_75, __p3_75) __extension__ ({ \
+ uint16x4_t __s0_75 = __p0_75; \
+ uint16x4_t __s1_75 = __p1_75; \
+ uint16x4_t __s2_75 = __p2_75; \
+ uint16x4_t __rev0_75; __rev0_75 = __builtin_shufflevector(__s0_75, __s0_75, 3, 2, 1, 0); \
+ uint16x4_t __rev1_75; __rev1_75 = __builtin_shufflevector(__s1_75, __s1_75, 3, 2, 1, 0); \
+ uint16x4_t __rev2_75; __rev2_75 = __builtin_shufflevector(__s2_75, __s2_75, 3, 2, 1, 0); \
+ uint16x4_t __ret_75; \
+ __ret_75 = __rev0_75 - __rev1_75 * __noswap_splat_lane_u16(__rev2_75, __p3_75); \
+ __ret_75 = __builtin_shufflevector(__ret_75, __ret_75, 3, 2, 1, 0); \
+ __ret_75; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmls_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x2_t __s2 = __p2; \
- float32x2_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3); \
- __ret; \
+#define vmls_lane_f32(__p0_76, __p1_76, __p2_76, __p3_76) __extension__ ({ \
+ float32x2_t __s0_76 = __p0_76; \
+ float32x2_t __s1_76 = __p1_76; \
+ float32x2_t __s2_76 = __p2_76; \
+ float32x2_t __ret_76; \
+ __ret_76 = __s0_76 - __s1_76 * splat_lane_f32(__s2_76, __p3_76); \
+ __ret_76; \
})
#else
-#define vmls_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x2_t __s2 = __p2; \
- float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- float32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- float32x2_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmls_lane_f32(__p0_77, __p1_77, __p2_77, __p3_77) __extension__ ({ \
+ float32x2_t __s0_77 = __p0_77; \
+ float32x2_t __s1_77 = __p1_77; \
+ float32x2_t __s2_77 = __p2_77; \
+ float32x2_t __rev0_77; __rev0_77 = __builtin_shufflevector(__s0_77, __s0_77, 1, 0); \
+ float32x2_t __rev1_77; __rev1_77 = __builtin_shufflevector(__s1_77, __s1_77, 1, 0); \
+ float32x2_t __rev2_77; __rev2_77 = __builtin_shufflevector(__s2_77, __s2_77, 1, 0); \
+ float32x2_t __ret_77; \
+ __ret_77 = __rev0_77 - __rev1_77 * __noswap_splat_lane_f32(__rev2_77, __p3_77); \
+ __ret_77 = __builtin_shufflevector(__ret_77, __ret_77, 1, 0); \
+ __ret_77; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmls_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x2_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3); \
- __ret; \
+#define vmls_lane_s32(__p0_78, __p1_78, __p2_78, __p3_78) __extension__ ({ \
+ int32x2_t __s0_78 = __p0_78; \
+ int32x2_t __s1_78 = __p1_78; \
+ int32x2_t __s2_78 = __p2_78; \
+ int32x2_t __ret_78; \
+ __ret_78 = __s0_78 - __s1_78 * splat_lane_s32(__s2_78, __p3_78); \
+ __ret_78; \
})
#else
-#define vmls_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int32x2_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmls_lane_s32(__p0_79, __p1_79, __p2_79, __p3_79) __extension__ ({ \
+ int32x2_t __s0_79 = __p0_79; \
+ int32x2_t __s1_79 = __p1_79; \
+ int32x2_t __s2_79 = __p2_79; \
+ int32x2_t __rev0_79; __rev0_79 = __builtin_shufflevector(__s0_79, __s0_79, 1, 0); \
+ int32x2_t __rev1_79; __rev1_79 = __builtin_shufflevector(__s1_79, __s1_79, 1, 0); \
+ int32x2_t __rev2_79; __rev2_79 = __builtin_shufflevector(__s2_79, __s2_79, 1, 0); \
+ int32x2_t __ret_79; \
+ __ret_79 = __rev0_79 - __rev1_79 * __noswap_splat_lane_s32(__rev2_79, __p3_79); \
+ __ret_79 = __builtin_shufflevector(__ret_79, __ret_79, 1, 0); \
+ __ret_79; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmls_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x4_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmls_lane_s16(__p0_80, __p1_80, __p2_80, __p3_80) __extension__ ({ \
+ int16x4_t __s0_80 = __p0_80; \
+ int16x4_t __s1_80 = __p1_80; \
+ int16x4_t __s2_80 = __p2_80; \
+ int16x4_t __ret_80; \
+ __ret_80 = __s0_80 - __s1_80 * splat_lane_s16(__s2_80, __p3_80); \
+ __ret_80; \
})
#else
-#define vmls_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int16x4_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmls_lane_s16(__p0_81, __p1_81, __p2_81, __p3_81) __extension__ ({ \
+ int16x4_t __s0_81 = __p0_81; \
+ int16x4_t __s1_81 = __p1_81; \
+ int16x4_t __s2_81 = __p2_81; \
+ int16x4_t __rev0_81; __rev0_81 = __builtin_shufflevector(__s0_81, __s0_81, 3, 2, 1, 0); \
+ int16x4_t __rev1_81; __rev1_81 = __builtin_shufflevector(__s1_81, __s1_81, 3, 2, 1, 0); \
+ int16x4_t __rev2_81; __rev2_81 = __builtin_shufflevector(__s2_81, __s2_81, 3, 2, 1, 0); \
+ int16x4_t __ret_81; \
+ __ret_81 = __rev0_81 - __rev1_81 * __noswap_splat_lane_s16(__rev2_81, __p3_81); \
+ __ret_81 = __builtin_shufflevector(__ret_81, __ret_81, 3, 2, 1, 0); \
+ __ret_81; \
})
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vmov_n_u64(uint64_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) {__p0};
- return __ret;
-}
-#else
__ai uint64x1_t vmov_n_u64(uint64_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) {__p0};
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vmov_n_u16(uint16_t __p0) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vmov_n_s64(int64_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t) {__p0};
- return __ret;
-}
-#else
__ai int64x1_t vmov_n_s64(int64_t __p0) {
int64x1_t __ret;
__ret = (int64x1_t) {__p0};
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vmov_n_s16(int16_t __p0) {
int16x4_t __ret;
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulq_lane_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x4_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmulq_lane_u32(__p0_82, __p1_82, __p2_82) __extension__ ({ \
+ uint32x4_t __s0_82 = __p0_82; \
+ uint32x2_t __s1_82 = __p1_82; \
+ uint32x4_t __ret_82; \
+ __ret_82 = __s0_82 * splatq_lane_u32(__s1_82, __p2_82); \
+ __ret_82; \
})
#else
-#define vmulq_lane_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- uint32x4_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmulq_lane_u32(__p0_83, __p1_83, __p2_83) __extension__ ({ \
+ uint32x4_t __s0_83 = __p0_83; \
+ uint32x2_t __s1_83 = __p1_83; \
+ uint32x4_t __rev0_83; __rev0_83 = __builtin_shufflevector(__s0_83, __s0_83, 3, 2, 1, 0); \
+ uint32x2_t __rev1_83; __rev1_83 = __builtin_shufflevector(__s1_83, __s1_83, 1, 0); \
+ uint32x4_t __ret_83; \
+ __ret_83 = __rev0_83 * __noswap_splatq_lane_u32(__rev1_83, __p2_83); \
+ __ret_83 = __builtin_shufflevector(__ret_83, __ret_83, 3, 2, 1, 0); \
+ __ret_83; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulq_lane_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x8_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmulq_lane_u16(__p0_84, __p1_84, __p2_84) __extension__ ({ \
+ uint16x8_t __s0_84 = __p0_84; \
+ uint16x4_t __s1_84 = __p1_84; \
+ uint16x8_t __ret_84; \
+ __ret_84 = __s0_84 * splatq_lane_u16(__s1_84, __p2_84); \
+ __ret_84; \
})
#else
-#define vmulq_lane_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint16x8_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmulq_lane_u16(__p0_85, __p1_85, __p2_85) __extension__ ({ \
+ uint16x8_t __s0_85 = __p0_85; \
+ uint16x4_t __s1_85 = __p1_85; \
+ uint16x8_t __rev0_85; __rev0_85 = __builtin_shufflevector(__s0_85, __s0_85, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x4_t __rev1_85; __rev1_85 = __builtin_shufflevector(__s1_85, __s1_85, 3, 2, 1, 0); \
+ uint16x8_t __ret_85; \
+ __ret_85 = __rev0_85 * __noswap_splatq_lane_u16(__rev1_85, __p2_85); \
+ __ret_85 = __builtin_shufflevector(__ret_85, __ret_85, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_85; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x4_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmulq_lane_f32(__p0_86, __p1_86, __p2_86) __extension__ ({ \
+ float32x4_t __s0_86 = __p0_86; \
+ float32x2_t __s1_86 = __p1_86; \
+ float32x4_t __ret_86; \
+ __ret_86 = __s0_86 * splatq_lane_f32(__s1_86, __p2_86); \
+ __ret_86; \
})
#else
-#define vmulq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- float32x4_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmulq_lane_f32(__p0_87, __p1_87, __p2_87) __extension__ ({ \
+ float32x4_t __s0_87 = __p0_87; \
+ float32x2_t __s1_87 = __p1_87; \
+ float32x4_t __rev0_87; __rev0_87 = __builtin_shufflevector(__s0_87, __s0_87, 3, 2, 1, 0); \
+ float32x2_t __rev1_87; __rev1_87 = __builtin_shufflevector(__s1_87, __s1_87, 1, 0); \
+ float32x4_t __ret_87; \
+ __ret_87 = __rev0_87 * __noswap_splatq_lane_f32(__rev1_87, __p2_87); \
+ __ret_87 = __builtin_shufflevector(__ret_87, __ret_87, 3, 2, 1, 0); \
+ __ret_87; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmulq_lane_s32(__p0_88, __p1_88, __p2_88) __extension__ ({ \
+ int32x4_t __s0_88 = __p0_88; \
+ int32x2_t __s1_88 = __p1_88; \
+ int32x4_t __ret_88; \
+ __ret_88 = __s0_88 * splatq_lane_s32(__s1_88, __p2_88); \
+ __ret_88; \
})
#else
-#define vmulq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x4_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmulq_lane_s32(__p0_89, __p1_89, __p2_89) __extension__ ({ \
+ int32x4_t __s0_89 = __p0_89; \
+ int32x2_t __s1_89 = __p1_89; \
+ int32x4_t __rev0_89; __rev0_89 = __builtin_shufflevector(__s0_89, __s0_89, 3, 2, 1, 0); \
+ int32x2_t __rev1_89; __rev1_89 = __builtin_shufflevector(__s1_89, __s1_89, 1, 0); \
+ int32x4_t __ret_89; \
+ __ret_89 = __rev0_89 * __noswap_splatq_lane_s32(__rev1_89, __p2_89); \
+ __ret_89 = __builtin_shufflevector(__ret_89, __ret_89, 3, 2, 1, 0); \
+ __ret_89; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmulq_lane_s16(__p0_90, __p1_90, __p2_90) __extension__ ({ \
+ int16x8_t __s0_90 = __p0_90; \
+ int16x4_t __s1_90 = __p1_90; \
+ int16x8_t __ret_90; \
+ __ret_90 = __s0_90 * splatq_lane_s16(__s1_90, __p2_90); \
+ __ret_90; \
})
#else
-#define vmulq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x8_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmulq_lane_s16(__p0_91, __p1_91, __p2_91) __extension__ ({ \
+ int16x8_t __s0_91 = __p0_91; \
+ int16x4_t __s1_91 = __p1_91; \
+ int16x8_t __rev0_91; __rev0_91 = __builtin_shufflevector(__s0_91, __s0_91, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev1_91; __rev1_91 = __builtin_shufflevector(__s1_91, __s1_91, 3, 2, 1, 0); \
+ int16x8_t __ret_91; \
+ __ret_91 = __rev0_91 * __noswap_splatq_lane_s16(__rev1_91, __p2_91); \
+ __ret_91 = __builtin_shufflevector(__ret_91, __ret_91, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_91; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmul_lane_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x2_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2); \
- __ret; \
+#define vmul_lane_u32(__p0_92, __p1_92, __p2_92) __extension__ ({ \
+ uint32x2_t __s0_92 = __p0_92; \
+ uint32x2_t __s1_92 = __p1_92; \
+ uint32x2_t __ret_92; \
+ __ret_92 = __s0_92 * splat_lane_u32(__s1_92, __p2_92); \
+ __ret_92; \
})
#else
-#define vmul_lane_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- uint32x2_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmul_lane_u32(__p0_93, __p1_93, __p2_93) __extension__ ({ \
+ uint32x2_t __s0_93 = __p0_93; \
+ uint32x2_t __s1_93 = __p1_93; \
+ uint32x2_t __rev0_93; __rev0_93 = __builtin_shufflevector(__s0_93, __s0_93, 1, 0); \
+ uint32x2_t __rev1_93; __rev1_93 = __builtin_shufflevector(__s1_93, __s1_93, 1, 0); \
+ uint32x2_t __ret_93; \
+ __ret_93 = __rev0_93 * __noswap_splat_lane_u32(__rev1_93, __p2_93); \
+ __ret_93 = __builtin_shufflevector(__ret_93, __ret_93, 1, 0); \
+ __ret_93; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmul_lane_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x4_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmul_lane_u16(__p0_94, __p1_94, __p2_94) __extension__ ({ \
+ uint16x4_t __s0_94 = __p0_94; \
+ uint16x4_t __s1_94 = __p1_94; \
+ uint16x4_t __ret_94; \
+ __ret_94 = __s0_94 * splat_lane_u16(__s1_94, __p2_94); \
+ __ret_94; \
})
#else
-#define vmul_lane_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint16x4_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmul_lane_u16(__p0_95, __p1_95, __p2_95) __extension__ ({ \
+ uint16x4_t __s0_95 = __p0_95; \
+ uint16x4_t __s1_95 = __p1_95; \
+ uint16x4_t __rev0_95; __rev0_95 = __builtin_shufflevector(__s0_95, __s0_95, 3, 2, 1, 0); \
+ uint16x4_t __rev1_95; __rev1_95 = __builtin_shufflevector(__s1_95, __s1_95, 3, 2, 1, 0); \
+ uint16x4_t __ret_95; \
+ __ret_95 = __rev0_95 * __noswap_splat_lane_u16(__rev1_95, __p2_95); \
+ __ret_95 = __builtin_shufflevector(__ret_95, __ret_95, 3, 2, 1, 0); \
+ __ret_95; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmul_lane_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x2_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2); \
- __ret; \
+#define vmul_lane_f32(__p0_96, __p1_96, __p2_96) __extension__ ({ \
+ float32x2_t __s0_96 = __p0_96; \
+ float32x2_t __s1_96 = __p1_96; \
+ float32x2_t __ret_96; \
+ __ret_96 = __s0_96 * splat_lane_f32(__s1_96, __p2_96); \
+ __ret_96; \
})
#else
-#define vmul_lane_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- float32x2_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmul_lane_f32(__p0_97, __p1_97, __p2_97) __extension__ ({ \
+ float32x2_t __s0_97 = __p0_97; \
+ float32x2_t __s1_97 = __p1_97; \
+ float32x2_t __rev0_97; __rev0_97 = __builtin_shufflevector(__s0_97, __s0_97, 1, 0); \
+ float32x2_t __rev1_97; __rev1_97 = __builtin_shufflevector(__s1_97, __s1_97, 1, 0); \
+ float32x2_t __ret_97; \
+ __ret_97 = __rev0_97 * __noswap_splat_lane_f32(__rev1_97, __p2_97); \
+ __ret_97 = __builtin_shufflevector(__ret_97, __ret_97, 1, 0); \
+ __ret_97; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmul_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2); \
- __ret; \
+#define vmul_lane_s32(__p0_98, __p1_98, __p2_98) __extension__ ({ \
+ int32x2_t __s0_98 = __p0_98; \
+ int32x2_t __s1_98 = __p1_98; \
+ int32x2_t __ret_98; \
+ __ret_98 = __s0_98 * splat_lane_s32(__s1_98, __p2_98); \
+ __ret_98; \
})
#else
-#define vmul_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x2_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmul_lane_s32(__p0_99, __p1_99, __p2_99) __extension__ ({ \
+ int32x2_t __s0_99 = __p0_99; \
+ int32x2_t __s1_99 = __p1_99; \
+ int32x2_t __rev0_99; __rev0_99 = __builtin_shufflevector(__s0_99, __s0_99, 1, 0); \
+ int32x2_t __rev1_99; __rev1_99 = __builtin_shufflevector(__s1_99, __s1_99, 1, 0); \
+ int32x2_t __ret_99; \
+ __ret_99 = __rev0_99 * __noswap_splat_lane_s32(__rev1_99, __p2_99); \
+ __ret_99 = __builtin_shufflevector(__ret_99, __ret_99, 1, 0); \
+ __ret_99; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmul_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmul_lane_s16(__p0_100, __p1_100, __p2_100) __extension__ ({ \
+ int16x4_t __s0_100 = __p0_100; \
+ int16x4_t __s1_100 = __p1_100; \
+ int16x4_t __ret_100; \
+ __ret_100 = __s0_100 * splat_lane_s16(__s1_100, __p2_100); \
+ __ret_100; \
})
#else
-#define vmul_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x4_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmul_lane_s16(__p0_101, __p1_101, __p2_101) __extension__ ({ \
+ int16x4_t __s0_101 = __p0_101; \
+ int16x4_t __s1_101 = __p1_101; \
+ int16x4_t __rev0_101; __rev0_101 = __builtin_shufflevector(__s0_101, __s0_101, 3, 2, 1, 0); \
+ int16x4_t __rev1_101; __rev1_101 = __builtin_shufflevector(__s1_101, __s1_101, 3, 2, 1, 0); \
+ int16x4_t __ret_101; \
+ __ret_101 = __rev0_101 * __noswap_splat_lane_s16(__rev1_101, __p2_101); \
+ __ret_101 = __builtin_shufflevector(__ret_101, __ret_101, 3, 2, 1, 0); \
+ __ret_101; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_lane_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint64x2_t __ret; \
- __ret = vmull_u32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vmull_lane_u32(__p0_102, __p1_102, __p2_102) __extension__ ({ \
+ uint32x2_t __s0_102 = __p0_102; \
+ uint32x2_t __s1_102 = __p1_102; \
+ uint64x2_t __ret_102; \
+ __ret_102 = vmull_u32(__s0_102, splat_lane_u32(__s1_102, __p2_102)); \
+ __ret_102; \
})
#else
-#define vmull_lane_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- uint64x2_t __ret; \
- __ret = __noswap_vmull_u32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmull_lane_u32(__p0_103, __p1_103, __p2_103) __extension__ ({ \
+ uint32x2_t __s0_103 = __p0_103; \
+ uint32x2_t __s1_103 = __p1_103; \
+ uint32x2_t __rev0_103; __rev0_103 = __builtin_shufflevector(__s0_103, __s0_103, 1, 0); \
+ uint32x2_t __rev1_103; __rev1_103 = __builtin_shufflevector(__s1_103, __s1_103, 1, 0); \
+ uint64x2_t __ret_103; \
+ __ret_103 = __noswap_vmull_u32(__rev0_103, __noswap_splat_lane_u32(__rev1_103, __p2_103)); \
+ __ret_103 = __builtin_shufflevector(__ret_103, __ret_103, 1, 0); \
+ __ret_103; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_lane_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint32x4_t __ret; \
- __ret = vmull_u16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vmull_lane_u16(__p0_104, __p1_104, __p2_104) __extension__ ({ \
+ uint16x4_t __s0_104 = __p0_104; \
+ uint16x4_t __s1_104 = __p1_104; \
+ uint32x4_t __ret_104; \
+ __ret_104 = vmull_u16(__s0_104, splat_lane_u16(__s1_104, __p2_104)); \
+ __ret_104; \
})
#else
-#define vmull_lane_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __noswap_vmull_u16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmull_lane_u16(__p0_105, __p1_105, __p2_105) __extension__ ({ \
+ uint16x4_t __s0_105 = __p0_105; \
+ uint16x4_t __s1_105 = __p1_105; \
+ uint16x4_t __rev0_105; __rev0_105 = __builtin_shufflevector(__s0_105, __s0_105, 3, 2, 1, 0); \
+ uint16x4_t __rev1_105; __rev1_105 = __builtin_shufflevector(__s1_105, __s1_105, 3, 2, 1, 0); \
+ uint32x4_t __ret_105; \
+ __ret_105 = __noswap_vmull_u16(__rev0_105, __noswap_splat_lane_u16(__rev1_105, __p2_105)); \
+ __ret_105 = __builtin_shufflevector(__ret_105, __ret_105, 3, 2, 1, 0); \
+ __ret_105; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int64x2_t __ret; \
- __ret = vmull_s32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vmull_lane_s32(__p0_106, __p1_106, __p2_106) __extension__ ({ \
+ int32x2_t __s0_106 = __p0_106; \
+ int32x2_t __s1_106 = __p1_106; \
+ int64x2_t __ret_106; \
+ __ret_106 = vmull_s32(__s0_106, splat_lane_s32(__s1_106, __p2_106)); \
+ __ret_106; \
})
#else
-#define vmull_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vmull_s32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmull_lane_s32(__p0_107, __p1_107, __p2_107) __extension__ ({ \
+ int32x2_t __s0_107 = __p0_107; \
+ int32x2_t __s1_107 = __p1_107; \
+ int32x2_t __rev0_107; __rev0_107 = __builtin_shufflevector(__s0_107, __s0_107, 1, 0); \
+ int32x2_t __rev1_107; __rev1_107 = __builtin_shufflevector(__s1_107, __s1_107, 1, 0); \
+ int64x2_t __ret_107; \
+ __ret_107 = __noswap_vmull_s32(__rev0_107, __noswap_splat_lane_s32(__rev1_107, __p2_107)); \
+ __ret_107 = __builtin_shufflevector(__ret_107, __ret_107, 1, 0); \
+ __ret_107; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int32x4_t __ret; \
- __ret = vmull_s16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vmull_lane_s16(__p0_108, __p1_108, __p2_108) __extension__ ({ \
+ int16x4_t __s0_108 = __p0_108; \
+ int16x4_t __s1_108 = __p1_108; \
+ int32x4_t __ret_108; \
+ __ret_108 = vmull_s16(__s0_108, splat_lane_s16(__s1_108, __p2_108)); \
+ __ret_108; \
})
#else
-#define vmull_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vmull_s16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmull_lane_s16(__p0_109, __p1_109, __p2_109) __extension__ ({ \
+ int16x4_t __s0_109 = __p0_109; \
+ int16x4_t __s1_109 = __p1_109; \
+ int16x4_t __rev0_109; __rev0_109 = __builtin_shufflevector(__s0_109, __s0_109, 3, 2, 1, 0); \
+ int16x4_t __rev1_109; __rev1_109 = __builtin_shufflevector(__s1_109, __s1_109, 3, 2, 1, 0); \
+ int32x4_t __ret_109; \
+ __ret_109 = __noswap_vmull_s16(__rev0_109, __noswap_splat_lane_s16(__rev1_109, __p2_109)); \
+ __ret_109 = __builtin_shufflevector(__ret_109, __ret_109, 3, 2, 1, 0); \
+ __ret_109; \
})
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vmull_n_u32(uint32x2_t __p0, uint32_t __p1) {
uint64x2_t __ret;
- __ret = (uint64x2_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(uint32x2_t) {__p1, __p1}, 51);
+ __ret = vmull_u32(__p0, (uint32x2_t) {__p1, __p1});
return __ret;
}
#else
__ai uint64x2_t vmull_n_u32(uint32x2_t __p0, uint32_t __p1) {
uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
uint64x2_t __ret;
- __ret = (uint64x2_t) __builtin_neon_vmull_v((int8x8_t)__rev0, (int8x8_t)(uint32x2_t) {__p1, __p1}, 51);
+ __ret = __noswap_vmull_u32(__rev0, (uint32x2_t) {__p1, __p1});
__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
__ai uint64x2_t __noswap_vmull_n_u32(uint32x2_t __p0, uint32_t __p1) {
uint64x2_t __ret;
- __ret = (uint64x2_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(uint32x2_t) {__p1, __p1}, 51);
+ __ret = __noswap_vmull_u32(__p0, (uint32x2_t) {__p1, __p1});
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint32x4_t vmull_n_u16(uint16x4_t __p0, uint16_t __p1) {
uint32x4_t __ret;
- __ret = (uint32x4_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(uint16x4_t) {__p1, __p1, __p1, __p1}, 50);
+ __ret = vmull_u16(__p0, (uint16x4_t) {__p1, __p1, __p1, __p1});
return __ret;
}
#else
__ai uint32x4_t vmull_n_u16(uint16x4_t __p0, uint16_t __p1) {
uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
uint32x4_t __ret;
- __ret = (uint32x4_t) __builtin_neon_vmull_v((int8x8_t)__rev0, (int8x8_t)(uint16x4_t) {__p1, __p1, __p1, __p1}, 50);
+ __ret = __noswap_vmull_u16(__rev0, (uint16x4_t) {__p1, __p1, __p1, __p1});
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
__ai uint32x4_t __noswap_vmull_n_u16(uint16x4_t __p0, uint16_t __p1) {
uint32x4_t __ret;
- __ret = (uint32x4_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(uint16x4_t) {__p1, __p1, __p1, __p1}, 50);
+ __ret = __noswap_vmull_u16(__p0, (uint16x4_t) {__p1, __p1, __p1, __p1});
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int64x2_t vmull_n_s32(int32x2_t __p0, int32_t __p1) {
int64x2_t __ret;
- __ret = (int64x2_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 35);
+ __ret = vmull_s32(__p0, (int32x2_t) {__p1, __p1});
return __ret;
}
#else
__ai int64x2_t vmull_n_s32(int32x2_t __p0, int32_t __p1) {
int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
int64x2_t __ret;
- __ret = (int64x2_t) __builtin_neon_vmull_v((int8x8_t)__rev0, (int8x8_t)(int32x2_t) {__p1, __p1}, 35);
+ __ret = __noswap_vmull_s32(__rev0, (int32x2_t) {__p1, __p1});
__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
__ai int64x2_t __noswap_vmull_n_s32(int32x2_t __p0, int32_t __p1) {
int64x2_t __ret;
- __ret = (int64x2_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 35);
+ __ret = __noswap_vmull_s32(__p0, (int32x2_t) {__p1, __p1});
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int32x4_t vmull_n_s16(int16x4_t __p0, int16_t __p1) {
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 34);
+ __ret = vmull_s16(__p0, (int16x4_t) {__p1, __p1, __p1, __p1});
return __ret;
}
#else
__ai int32x4_t vmull_n_s16(int16x4_t __p0, int16_t __p1) {
int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vmull_v((int8x8_t)__rev0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 34);
+ __ret = __noswap_vmull_s16(__rev0, (int16x4_t) {__p1, __p1, __p1, __p1});
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
__ai int32x4_t __noswap_vmull_n_s16(int16x4_t __p0, int16_t __p1) {
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 34);
+ __ret = __noswap_vmull_s16(__p0, (int16x4_t) {__p1, __p1, __p1, __p1});
return __ret;
}
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vorn_u64(uint64x1_t __p0, uint64x1_t __p1) {
uint64x1_t __ret;
__ret = __p0 | ~__p1;
return __ret;
}
-#else
-__ai uint64x1_t vorn_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = __p0 | ~__p1;
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vorn_u16(uint16x4_t __p0, uint16x4_t __p1) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vorn_s64(int64x1_t __p0, int64x1_t __p1) {
- int64x1_t __ret;
- __ret = __p0 | ~__p1;
- return __ret;
-}
-#else
__ai int64x1_t vorn_s64(int64x1_t __p0, int64x1_t __p1) {
int64x1_t __ret;
__ret = __p0 | ~__p1;
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vorn_s16(int16x4_t __p0, int16x4_t __p1) {
int16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vorr_u64(uint64x1_t __p0, uint64x1_t __p1) {
uint64x1_t __ret;
__ret = __p0 | __p1;
return __ret;
}
-#else
-__ai uint64x1_t vorr_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = __p0 | __p1;
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vorr_u16(uint16x4_t __p0, uint16x4_t __p1) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vorr_s64(int64x1_t __p0, int64x1_t __p1) {
- int64x1_t __ret;
- __ret = __p0 | __p1;
- return __ret;
-}
-#else
__ai int64x1_t vorr_s64(int64x1_t __p0, int64x1_t __p1) {
int64x1_t __ret;
__ret = __p0 | __p1;
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vorr_s16(int16x4_t __p0, int16x4_t __p1) {
int16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vqadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
- return __ret;
-}
-#else
__ai uint64x1_t vqadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vqadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vqadd_s64(int64x1_t __p0, int64x1_t __p1) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
- return __ret;
-}
-#else
__ai int64x1_t vqadd_s64(int64x1_t __p0, int64x1_t __p1) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vqadd_s16(int16x4_t __p0, int16x4_t __p1) {
int16x4_t __ret;
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlal_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = vqdmlal_s32(__s0, __s1, __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vqdmlal_lane_s32(__p0_110, __p1_110, __p2_110, __p3_110) __extension__ ({ \
+ int64x2_t __s0_110 = __p0_110; \
+ int32x2_t __s1_110 = __p1_110; \
+ int32x2_t __s2_110 = __p2_110; \
+ int64x2_t __ret_110; \
+ __ret_110 = vqdmlal_s32(__s0_110, __s1_110, splat_lane_s32(__s2_110, __p3_110)); \
+ __ret_110; \
})
#else
-#define vqdmlal_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vqdmlal_s32(__rev0, __rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vqdmlal_lane_s32(__p0_111, __p1_111, __p2_111, __p3_111) __extension__ ({ \
+ int64x2_t __s0_111 = __p0_111; \
+ int32x2_t __s1_111 = __p1_111; \
+ int32x2_t __s2_111 = __p2_111; \
+ int64x2_t __rev0_111; __rev0_111 = __builtin_shufflevector(__s0_111, __s0_111, 1, 0); \
+ int32x2_t __rev1_111; __rev1_111 = __builtin_shufflevector(__s1_111, __s1_111, 1, 0); \
+ int32x2_t __rev2_111; __rev2_111 = __builtin_shufflevector(__s2_111, __s2_111, 1, 0); \
+ int64x2_t __ret_111; \
+ __ret_111 = __noswap_vqdmlal_s32(__rev0_111, __rev1_111, __noswap_splat_lane_s32(__rev2_111, __p3_111)); \
+ __ret_111 = __builtin_shufflevector(__ret_111, __ret_111, 1, 0); \
+ __ret_111; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlal_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = vqdmlal_s16(__s0, __s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vqdmlal_lane_s16(__p0_112, __p1_112, __p2_112, __p3_112) __extension__ ({ \
+ int32x4_t __s0_112 = __p0_112; \
+ int16x4_t __s1_112 = __p1_112; \
+ int16x4_t __s2_112 = __p2_112; \
+ int32x4_t __ret_112; \
+ __ret_112 = vqdmlal_s16(__s0_112, __s1_112, splat_lane_s16(__s2_112, __p3_112)); \
+ __ret_112; \
})
#else
-#define vqdmlal_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqdmlal_s16(__rev0, __rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqdmlal_lane_s16(__p0_113, __p1_113, __p2_113, __p3_113) __extension__ ({ \
+ int32x4_t __s0_113 = __p0_113; \
+ int16x4_t __s1_113 = __p1_113; \
+ int16x4_t __s2_113 = __p2_113; \
+ int32x4_t __rev0_113; __rev0_113 = __builtin_shufflevector(__s0_113, __s0_113, 3, 2, 1, 0); \
+ int16x4_t __rev1_113; __rev1_113 = __builtin_shufflevector(__s1_113, __s1_113, 3, 2, 1, 0); \
+ int16x4_t __rev2_113; __rev2_113 = __builtin_shufflevector(__s2_113, __s2_113, 3, 2, 1, 0); \
+ int32x4_t __ret_113; \
+ __ret_113 = __noswap_vqdmlal_s16(__rev0_113, __rev1_113, __noswap_splat_lane_s16(__rev2_113, __p3_113)); \
+ __ret_113 = __builtin_shufflevector(__ret_113, __ret_113, 3, 2, 1, 0); \
+ __ret_113; \
})
#endif
#ifdef __LITTLE_ENDIAN__
__ai int64x2_t vqdmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
int64x2_t __ret;
- __ret = (int64x2_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int32x2_t) {__p2, __p2}, 35);
+ __ret = vqdmlal_s32(__p0, __p1, (int32x2_t) {__p2, __p2});
return __ret;
}
#else
int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
int64x2_t __ret;
- __ret = (int64x2_t) __builtin_neon_vqdmlal_v((int8x16_t)__rev0, (int8x8_t)__rev1, (int8x8_t)(int32x2_t) {__p2, __p2}, 35);
+ __ret = __noswap_vqdmlal_s32(__rev0, __rev1, (int32x2_t) {__p2, __p2});
__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
__ai int64x2_t __noswap_vqdmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
int64x2_t __ret;
- __ret = (int64x2_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int32x2_t) {__p2, __p2}, 35);
+ __ret = __noswap_vqdmlal_s32(__p0, __p1, (int32x2_t) {__p2, __p2});
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int32x4_t vqdmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int16x4_t) {__p2, __p2, __p2, __p2}, 34);
+ __ret = vqdmlal_s16(__p0, __p1, (int16x4_t) {__p2, __p2, __p2, __p2});
return __ret;
}
#else
int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vqdmlal_v((int8x16_t)__rev0, (int8x8_t)__rev1, (int8x8_t)(int16x4_t) {__p2, __p2, __p2, __p2}, 34);
+ __ret = __noswap_vqdmlal_s16(__rev0, __rev1, (int16x4_t) {__p2, __p2, __p2, __p2});
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
__ai int32x4_t __noswap_vqdmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int16x4_t) {__p2, __p2, __p2, __p2}, 34);
+ __ret = __noswap_vqdmlal_s16(__p0, __p1, (int16x4_t) {__p2, __p2, __p2, __p2});
return __ret;
}
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlsl_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = vqdmlsl_s32(__s0, __s1, __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vqdmlsl_lane_s32(__p0_114, __p1_114, __p2_114, __p3_114) __extension__ ({ \
+ int64x2_t __s0_114 = __p0_114; \
+ int32x2_t __s1_114 = __p1_114; \
+ int32x2_t __s2_114 = __p2_114; \
+ int64x2_t __ret_114; \
+ __ret_114 = vqdmlsl_s32(__s0_114, __s1_114, splat_lane_s32(__s2_114, __p3_114)); \
+ __ret_114; \
})
#else
-#define vqdmlsl_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vqdmlsl_s32(__rev0, __rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vqdmlsl_lane_s32(__p0_115, __p1_115, __p2_115, __p3_115) __extension__ ({ \
+ int64x2_t __s0_115 = __p0_115; \
+ int32x2_t __s1_115 = __p1_115; \
+ int32x2_t __s2_115 = __p2_115; \
+ int64x2_t __rev0_115; __rev0_115 = __builtin_shufflevector(__s0_115, __s0_115, 1, 0); \
+ int32x2_t __rev1_115; __rev1_115 = __builtin_shufflevector(__s1_115, __s1_115, 1, 0); \
+ int32x2_t __rev2_115; __rev2_115 = __builtin_shufflevector(__s2_115, __s2_115, 1, 0); \
+ int64x2_t __ret_115; \
+ __ret_115 = __noswap_vqdmlsl_s32(__rev0_115, __rev1_115, __noswap_splat_lane_s32(__rev2_115, __p3_115)); \
+ __ret_115 = __builtin_shufflevector(__ret_115, __ret_115, 1, 0); \
+ __ret_115; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlsl_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = vqdmlsl_s16(__s0, __s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vqdmlsl_lane_s16(__p0_116, __p1_116, __p2_116, __p3_116) __extension__ ({ \
+ int32x4_t __s0_116 = __p0_116; \
+ int16x4_t __s1_116 = __p1_116; \
+ int16x4_t __s2_116 = __p2_116; \
+ int32x4_t __ret_116; \
+ __ret_116 = vqdmlsl_s16(__s0_116, __s1_116, splat_lane_s16(__s2_116, __p3_116)); \
+ __ret_116; \
})
#else
-#define vqdmlsl_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqdmlsl_s16(__rev0, __rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqdmlsl_lane_s16(__p0_117, __p1_117, __p2_117, __p3_117) __extension__ ({ \
+ int32x4_t __s0_117 = __p0_117; \
+ int16x4_t __s1_117 = __p1_117; \
+ int16x4_t __s2_117 = __p2_117; \
+ int32x4_t __rev0_117; __rev0_117 = __builtin_shufflevector(__s0_117, __s0_117, 3, 2, 1, 0); \
+ int16x4_t __rev1_117; __rev1_117 = __builtin_shufflevector(__s1_117, __s1_117, 3, 2, 1, 0); \
+ int16x4_t __rev2_117; __rev2_117 = __builtin_shufflevector(__s2_117, __s2_117, 3, 2, 1, 0); \
+ int32x4_t __ret_117; \
+ __ret_117 = __noswap_vqdmlsl_s16(__rev0_117, __rev1_117, __noswap_splat_lane_s16(__rev2_117, __p3_117)); \
+ __ret_117 = __builtin_shufflevector(__ret_117, __ret_117, 3, 2, 1, 0); \
+ __ret_117; \
})
#endif
#ifdef __LITTLE_ENDIAN__
__ai int64x2_t vqdmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
int64x2_t __ret;
- __ret = (int64x2_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int32x2_t) {__p2, __p2}, 35);
+ __ret = vqdmlsl_s32(__p0, __p1, (int32x2_t) {__p2, __p2});
return __ret;
}
#else
int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
int64x2_t __ret;
- __ret = (int64x2_t) __builtin_neon_vqdmlsl_v((int8x16_t)__rev0, (int8x8_t)__rev1, (int8x8_t)(int32x2_t) {__p2, __p2}, 35);
+ __ret = __noswap_vqdmlsl_s32(__rev0, __rev1, (int32x2_t) {__p2, __p2});
__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
__ai int64x2_t __noswap_vqdmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
int64x2_t __ret;
- __ret = (int64x2_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int32x2_t) {__p2, __p2}, 35);
+ __ret = __noswap_vqdmlsl_s32(__p0, __p1, (int32x2_t) {__p2, __p2});
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int32x4_t vqdmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int16x4_t) {__p2, __p2, __p2, __p2}, 34);
+ __ret = vqdmlsl_s16(__p0, __p1, (int16x4_t) {__p2, __p2, __p2, __p2});
return __ret;
}
#else
int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vqdmlsl_v((int8x16_t)__rev0, (int8x8_t)__rev1, (int8x8_t)(int16x4_t) {__p2, __p2, __p2, __p2}, 34);
+ __ret = __noswap_vqdmlsl_s16(__rev0, __rev1, (int16x4_t) {__p2, __p2, __p2, __p2});
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
__ai int32x4_t __noswap_vqdmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int16x4_t) {__p2, __p2, __p2, __p2}, 34);
+ __ret = __noswap_vqdmlsl_s16(__p0, __p1, (int16x4_t) {__p2, __p2, __p2, __p2});
return __ret;
}
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __ret; \
- __ret = vqdmulhq_s32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
-})
-#else
-#define vqdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqdmulhq_s32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __ret; \
- __ret = vqdmulhq_s16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2)); \
- __ret; \
-})
-#else
-#define vqdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x8_t __ret; \
- __ret = __noswap_vqdmulhq_s16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __ret; \
- __ret = vqdmulh_s32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
-})
-#else
-#define vqdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x2_t __ret; \
- __ret = __noswap_vqdmulh_s32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __ret; \
- __ret = vqdmulh_s16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
-})
-#else
-#define vqdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x4_t __ret; \
- __ret = __noswap_vqdmulh_s16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai int32x4_t vqdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vqdmulhq_v((int8x16_t)__p0, (int8x16_t)(int32x4_t) {__p1, __p1, __p1, __p1}, 34);
+ __ret = vqdmulhq_s32(__p0, (int32x4_t) {__p1, __p1, __p1, __p1});
return __ret;
}
#else
__ai int32x4_t vqdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vqdmulhq_v((int8x16_t)__rev0, (int8x16_t)(int32x4_t) {__p1, __p1, __p1, __p1}, 34);
+ __ret = __noswap_vqdmulhq_s32(__rev0, (int32x4_t) {__p1, __p1, __p1, __p1});
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#ifdef __LITTLE_ENDIAN__
__ai int16x8_t vqdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
int16x8_t __ret;
- __ret = (int16x8_t) __builtin_neon_vqdmulhq_v((int8x16_t)__p0, (int8x16_t)(int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1}, 33);
+ __ret = vqdmulhq_s16(__p0, (int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1});
return __ret;
}
#else
__ai int16x8_t vqdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
int16x8_t __ret;
- __ret = (int16x8_t) __builtin_neon_vqdmulhq_v((int8x16_t)__rev0, (int8x16_t)(int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1}, 33);
+ __ret = __noswap_vqdmulhq_s16(__rev0, (int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1});
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
return __ret;
}
#ifdef __LITTLE_ENDIAN__
__ai int32x2_t vqdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
int32x2_t __ret;
- __ret = (int32x2_t) __builtin_neon_vqdmulh_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 2);
+ __ret = vqdmulh_s32(__p0, (int32x2_t) {__p1, __p1});
return __ret;
}
#else
__ai int32x2_t vqdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
int32x2_t __ret;
- __ret = (int32x2_t) __builtin_neon_vqdmulh_v((int8x8_t)__rev0, (int8x8_t)(int32x2_t) {__p1, __p1}, 2);
+ __ret = __noswap_vqdmulh_s32(__rev0, (int32x2_t) {__p1, __p1});
__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vqdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
int16x4_t __ret;
- __ret = (int16x4_t) __builtin_neon_vqdmulh_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 1);
+ __ret = vqdmulh_s16(__p0, (int16x4_t) {__p1, __p1, __p1, __p1});
return __ret;
}
#else
__ai int16x4_t vqdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int16x4_t __ret;
- __ret = (int16x4_t) __builtin_neon_vqdmulh_v((int8x8_t)__rev0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 1);
+ __ret = __noswap_vqdmulh_s16(__rev0, (int16x4_t) {__p1, __p1, __p1, __p1});
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmull_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int64x2_t __ret; \
- __ret = vqdmull_s32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vqdmull_lane_s32(__p0_118, __p1_118, __p2_118) __extension__ ({ \
+ int32x2_t __s0_118 = __p0_118; \
+ int32x2_t __s1_118 = __p1_118; \
+ int64x2_t __ret_118; \
+ __ret_118 = vqdmull_s32(__s0_118, splat_lane_s32(__s1_118, __p2_118)); \
+ __ret_118; \
})
#else
-#define vqdmull_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vqdmull_s32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vqdmull_lane_s32(__p0_119, __p1_119, __p2_119) __extension__ ({ \
+ int32x2_t __s0_119 = __p0_119; \
+ int32x2_t __s1_119 = __p1_119; \
+ int32x2_t __rev0_119; __rev0_119 = __builtin_shufflevector(__s0_119, __s0_119, 1, 0); \
+ int32x2_t __rev1_119; __rev1_119 = __builtin_shufflevector(__s1_119, __s1_119, 1, 0); \
+ int64x2_t __ret_119; \
+ __ret_119 = __noswap_vqdmull_s32(__rev0_119, __noswap_splat_lane_s32(__rev1_119, __p2_119)); \
+ __ret_119 = __builtin_shufflevector(__ret_119, __ret_119, 1, 0); \
+ __ret_119; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmull_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int32x4_t __ret; \
- __ret = vqdmull_s16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vqdmull_lane_s16(__p0_120, __p1_120, __p2_120) __extension__ ({ \
+ int16x4_t __s0_120 = __p0_120; \
+ int16x4_t __s1_120 = __p1_120; \
+ int32x4_t __ret_120; \
+ __ret_120 = vqdmull_s16(__s0_120, splat_lane_s16(__s1_120, __p2_120)); \
+ __ret_120; \
})
#else
-#define vqdmull_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqdmull_s16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqdmull_lane_s16(__p0_121, __p1_121, __p2_121) __extension__ ({ \
+ int16x4_t __s0_121 = __p0_121; \
+ int16x4_t __s1_121 = __p1_121; \
+ int16x4_t __rev0_121; __rev0_121 = __builtin_shufflevector(__s0_121, __s0_121, 3, 2, 1, 0); \
+ int16x4_t __rev1_121; __rev1_121 = __builtin_shufflevector(__s1_121, __s1_121, 3, 2, 1, 0); \
+ int32x4_t __ret_121; \
+ __ret_121 = __noswap_vqdmull_s16(__rev0_121, __noswap_splat_lane_s16(__rev1_121, __p2_121)); \
+ __ret_121 = __builtin_shufflevector(__ret_121, __ret_121, 3, 2, 1, 0); \
+ __ret_121; \
})
#endif
#ifdef __LITTLE_ENDIAN__
__ai int64x2_t vqdmull_n_s32(int32x2_t __p0, int32_t __p1) {
int64x2_t __ret;
- __ret = (int64x2_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 35);
+ __ret = vqdmull_s32(__p0, (int32x2_t) {__p1, __p1});
return __ret;
}
#else
__ai int64x2_t vqdmull_n_s32(int32x2_t __p0, int32_t __p1) {
int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
int64x2_t __ret;
- __ret = (int64x2_t) __builtin_neon_vqdmull_v((int8x8_t)__rev0, (int8x8_t)(int32x2_t) {__p1, __p1}, 35);
+ __ret = __noswap_vqdmull_s32(__rev0, (int32x2_t) {__p1, __p1});
__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
__ai int64x2_t __noswap_vqdmull_n_s32(int32x2_t __p0, int32_t __p1) {
int64x2_t __ret;
- __ret = (int64x2_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 35);
+ __ret = __noswap_vqdmull_s32(__p0, (int32x2_t) {__p1, __p1});
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int32x4_t vqdmull_n_s16(int16x4_t __p0, int16_t __p1) {
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 34);
+ __ret = vqdmull_s16(__p0, (int16x4_t) {__p1, __p1, __p1, __p1});
return __ret;
}
#else
__ai int32x4_t vqdmull_n_s16(int16x4_t __p0, int16_t __p1) {
int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vqdmull_v((int8x8_t)__rev0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 34);
+ __ret = __noswap_vqdmull_s16(__rev0, (int16x4_t) {__p1, __p1, __p1, __p1});
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
__ai int32x4_t __noswap_vqdmull_n_s16(int16x4_t __p0, int16_t __p1) {
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 34);
+ __ret = __noswap_vqdmull_s16(__p0, (int16x4_t) {__p1, __p1, __p1, __p1});
return __ret;
}
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __ret; \
- __ret = vqrdmulhq_s32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
-})
-#else
-#define vqrdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqrdmulhq_s32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqrdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __ret; \
- __ret = vqrdmulhq_s16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2)); \
- __ret; \
-})
-#else
-#define vqrdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x8_t __ret; \
- __ret = __noswap_vqrdmulhq_s16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqrdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __ret; \
- __ret = vqrdmulh_s32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
-})
-#else
-#define vqrdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x2_t __ret; \
- __ret = __noswap_vqrdmulh_s32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqrdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __ret; \
- __ret = vqrdmulh_s16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
-})
-#else
-#define vqrdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x4_t __ret; \
- __ret = __noswap_vqrdmulh_s16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai int32x4_t vqrdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__p0, (int8x16_t)(int32x4_t) {__p1, __p1, __p1, __p1}, 34);
+ __ret = vqrdmulhq_s32(__p0, (int32x4_t) {__p1, __p1, __p1, __p1});
return __ret;
}
#else
__ai int32x4_t vqrdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int32x4_t __ret;
- __ret = (int32x4_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__rev0, (int8x16_t)(int32x4_t) {__p1, __p1, __p1, __p1}, 34);
+ __ret = __noswap_vqrdmulhq_s32(__rev0, (int32x4_t) {__p1, __p1, __p1, __p1});
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#ifdef __LITTLE_ENDIAN__
__ai int16x8_t vqrdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
int16x8_t __ret;
- __ret = (int16x8_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__p0, (int8x16_t)(int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1}, 33);
+ __ret = vqrdmulhq_s16(__p0, (int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1});
return __ret;
}
#else
__ai int16x8_t vqrdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
int16x8_t __ret;
- __ret = (int16x8_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__rev0, (int8x16_t)(int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1}, 33);
+ __ret = __noswap_vqrdmulhq_s16(__rev0, (int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1});
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
return __ret;
}
#ifdef __LITTLE_ENDIAN__
__ai int32x2_t vqrdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
int32x2_t __ret;
- __ret = (int32x2_t) __builtin_neon_vqrdmulh_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 2);
+ __ret = vqrdmulh_s32(__p0, (int32x2_t) {__p1, __p1});
return __ret;
}
#else
__ai int32x2_t vqrdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
int32x2_t __ret;
- __ret = (int32x2_t) __builtin_neon_vqrdmulh_v((int8x8_t)__rev0, (int8x8_t)(int32x2_t) {__p1, __p1}, 2);
+ __ret = __noswap_vqrdmulh_s32(__rev0, (int32x2_t) {__p1, __p1});
__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vqrdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
int16x4_t __ret;
- __ret = (int16x4_t) __builtin_neon_vqrdmulh_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 1);
+ __ret = vqrdmulh_s16(__p0, (int16x4_t) {__p1, __p1, __p1, __p1});
return __ret;
}
#else
__ai int16x4_t vqrdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int16x4_t __ret;
- __ret = (int16x4_t) __builtin_neon_vqrdmulh_v((int8x8_t)__rev0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 1);
+ __ret = __noswap_vqrdmulh_s16(__rev0, (int16x4_t) {__p1, __p1, __p1, __p1});
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vqrshl_u64(uint64x1_t __p0, int64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
- return __ret;
-}
-#else
__ai uint64x1_t vqrshl_u64(uint64x1_t __p0, int64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vqrshl_u16(uint16x4_t __p0, int16x4_t __p1) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai int64x1_t vqrshl_s64(int64x1_t __p0, int64x1_t __p1) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
return __ret;
}
-#else
-__ai int64x1_t vqrshl_s64(int64x1_t __p0, int64x1_t __p1) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
int16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vqshl_u64(uint64x1_t __p0, int64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
return __ret;
}
-#else
-__ai uint64x1_t vqshl_u64(uint64x1_t __p0, int64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vqshl_u16(uint16x4_t __p0, int16x4_t __p1) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai int64x1_t vqshl_s64(int64x1_t __p0, int64x1_t __p1) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
return __ret;
}
-#else
-__ai int64x1_t vqshl_s64(int64x1_t __p0, int64x1_t __p1) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
int16x4_t __ret;
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vqshl_n_u64(__p0, __p1) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 19); \
- __ret; \
-})
-#else
#define vqshl_n_u64(__p0, __p1) __extension__ ({ \
uint64x1_t __s0 = __p0; \
uint64x1_t __ret; \
__ret = (uint64x1_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 19); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vqshl_n_u16(__p0, __p1) __extension__ ({ \
uint16x4_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vqshl_n_s64(__p0, __p1) __extension__ ({ \
int64x1_t __s0 = __p0; \
int64x1_t __ret; \
__ret = (int64x1_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 3); \
__ret; \
})
-#else
-#define vqshl_n_s64(__p0, __p1) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 3); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vqshl_n_s16(__p0, __p1) __extension__ ({ \
int16x4_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vqshlu_n_s64(__p0, __p1) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 19); \
- __ret; \
-})
-#else
#define vqshlu_n_s64(__p0, __p1) __extension__ ({ \
int64x1_t __s0 = __p0; \
uint64x1_t __ret; \
__ret = (uint64x1_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 19); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vqshlu_n_s16(__p0, __p1) __extension__ ({ \
int16x4_t __s0 = __p0; \
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vqsub_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
- return __ret;
-}
-#else
__ai uint64x1_t vqsub_u64(uint64x1_t __p0, uint64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vqsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vqsub_s64(int64x1_t __p0, int64x1_t __p1) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
- return __ret;
-}
-#else
__ai int64x1_t vqsub_s64(int64x1_t __p0, int64x1_t __p1) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vqsub_s16(int16x4_t __p0, int16x4_t __p1) {
int16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vrshl_u64(uint64x1_t __p0, int64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
return __ret;
}
-#else
-__ai uint64x1_t vrshl_u64(uint64x1_t __p0, int64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vrshl_u16(uint16x4_t __p0, int16x4_t __p1) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vrshl_s64(int64x1_t __p0, int64x1_t __p1) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
- return __ret;
-}
-#else
__ai int64x1_t vrshl_s64(int64x1_t __p0, int64x1_t __p1) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
int16x4_t __ret;
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vrshr_n_u64(__p0, __p1) __extension__ ({ \
uint64x1_t __s0 = __p0; \
uint64x1_t __ret; \
__ret = (uint64x1_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 19); \
__ret; \
})
-#else
-#define vrshr_n_u64(__p0, __p1) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 19); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vrshr_n_u16(__p0, __p1) __extension__ ({ \
uint16x4_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vrshr_n_s64(__p0, __p1) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 3); \
- __ret; \
-})
-#else
#define vrshr_n_s64(__p0, __p1) __extension__ ({ \
int64x1_t __s0 = __p0; \
int64x1_t __ret; \
__ret = (int64x1_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 3); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vrshr_n_s16(__p0, __p1) __extension__ ({ \
int16x4_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vrsra_n_u64(__p0, __p1, __p2) __extension__ ({ \
uint64x1_t __s0 = __p0; \
uint64x1_t __s1 = __p1; \
__ret = (uint64x1_t) __builtin_neon_vrsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 19); \
__ret; \
})
-#else
-#define vrsra_n_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64x1_t __s1 = __p1; \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vrsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 19); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vrsra_n_u16(__p0, __p1, __p2) __extension__ ({ \
uint16x4_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vrsra_n_s64(__p0, __p1, __p2) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64x1_t __s1 = __p1; \
- int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vrsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 3); \
- __ret; \
-})
-#else
#define vrsra_n_s64(__p0, __p1, __p2) __extension__ ({ \
int64x1_t __s0 = __p0; \
int64x1_t __s1 = __p1; \
__ret = (int64x1_t) __builtin_neon_vrsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 3); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vrsra_n_s16(__p0, __p1, __p2) __extension__ ({ \
int16x4_t __s0 = __p0; \
poly8_t __s0 = __p0; \
poly8x8_t __s1 = __p1; \
poly8x8_t __ret; \
- __ret = (poly8x8_t) __builtin_neon_vset_lane_i8(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (poly8x8_t) __builtin_neon_vset_lane_i8(__s0, (poly8x8_t)__s1, __p2); \
__ret; \
})
#else
poly8x8_t __s1 = __p1; \
poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
poly8x8_t __ret; \
- __ret = (poly8x8_t) __builtin_neon_vset_lane_i8(__s0, (int8x8_t)__rev1, __p2); \
+ __ret = (poly8x8_t) __builtin_neon_vset_lane_i8(__s0, (poly8x8_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
__ret; \
})
poly8_t __s0 = __p0; \
poly8x8_t __s1 = __p1; \
poly8x8_t __ret; \
- __ret = (poly8x8_t) __builtin_neon_vset_lane_i8(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (poly8x8_t) __builtin_neon_vset_lane_i8(__s0, (poly8x8_t)__s1, __p2); \
__ret; \
})
#endif
poly16_t __s0 = __p0; \
poly16x4_t __s1 = __p1; \
poly16x4_t __ret; \
- __ret = (poly16x4_t) __builtin_neon_vset_lane_i16(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (poly16x4_t) __builtin_neon_vset_lane_i16(__s0, (poly16x4_t)__s1, __p2); \
__ret; \
})
#else
poly16x4_t __s1 = __p1; \
poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
poly16x4_t __ret; \
- __ret = (poly16x4_t) __builtin_neon_vset_lane_i16(__s0, (int8x8_t)__rev1, __p2); \
+ __ret = (poly16x4_t) __builtin_neon_vset_lane_i16(__s0, (poly16x4_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
__ret; \
})
poly16_t __s0 = __p0; \
poly16x4_t __s1 = __p1; \
poly16x4_t __ret; \
- __ret = (poly16x4_t) __builtin_neon_vset_lane_i16(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (poly16x4_t) __builtin_neon_vset_lane_i16(__s0, (poly16x4_t)__s1, __p2); \
__ret; \
})
#endif
poly8_t __s0 = __p0; \
poly8x16_t __s1 = __p1; \
poly8x16_t __ret; \
- __ret = (poly8x16_t) __builtin_neon_vsetq_lane_i8(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (poly8x16_t) __builtin_neon_vsetq_lane_i8(__s0, (poly8x16_t)__s1, __p2); \
__ret; \
})
#else
poly8x16_t __s1 = __p1; \
poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
poly8x16_t __ret; \
- __ret = (poly8x16_t) __builtin_neon_vsetq_lane_i8(__s0, (int8x16_t)__rev1, __p2); \
+ __ret = (poly8x16_t) __builtin_neon_vsetq_lane_i8(__s0, (poly8x16_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
__ret; \
})
poly8_t __s0 = __p0; \
poly8x16_t __s1 = __p1; \
poly8x16_t __ret; \
- __ret = (poly8x16_t) __builtin_neon_vsetq_lane_i8(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (poly8x16_t) __builtin_neon_vsetq_lane_i8(__s0, (poly8x16_t)__s1, __p2); \
__ret; \
})
#endif
poly16_t __s0 = __p0; \
poly16x8_t __s1 = __p1; \
poly16x8_t __ret; \
- __ret = (poly16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (poly16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (poly16x8_t)__s1, __p2); \
__ret; \
})
#else
poly16x8_t __s1 = __p1; \
poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
poly16x8_t __ret; \
- __ret = (poly16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int8x16_t)__rev1, __p2); \
+ __ret = (poly16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (poly16x8_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
__ret; \
})
poly16_t __s0 = __p0; \
poly16x8_t __s1 = __p1; \
poly16x8_t __ret; \
- __ret = (poly16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (poly16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (poly16x8_t)__s1, __p2); \
__ret; \
})
#endif
uint32_t __s0 = __p0; \
uint32x4_t __s1 = __p1; \
uint32x4_t __ret; \
- __ret = (uint32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (uint32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int32x4_t)__s1, __p2); \
__ret; \
})
#else
uint32x4_t __s1 = __p1; \
uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
uint32x4_t __ret; \
- __ret = (uint32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int8x16_t)__rev1, __p2); \
+ __ret = (uint32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int32x4_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
__ret; \
})
uint32_t __s0 = __p0; \
uint32x4_t __s1 = __p1; \
uint32x4_t __ret; \
- __ret = (uint32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (uint32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int32x4_t)__s1, __p2); \
__ret; \
})
#endif
uint64_t __s0 = __p0; \
uint64x2_t __s1 = __p1; \
uint64x2_t __ret; \
- __ret = (uint64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (uint64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int64x2_t)__s1, __p2); \
__ret; \
})
#else
uint64x2_t __s1 = __p1; \
uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
uint64x2_t __ret; \
- __ret = (uint64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int8x16_t)__rev1, __p2); \
+ __ret = (uint64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int64x2_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
__ret; \
})
uint64_t __s0 = __p0; \
uint64x2_t __s1 = __p1; \
uint64x2_t __ret; \
- __ret = (uint64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (uint64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int64x2_t)__s1, __p2); \
__ret; \
})
#endif
uint16_t __s0 = __p0; \
uint16x8_t __s1 = __p1; \
uint16x8_t __ret; \
- __ret = (uint16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (uint16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int16x8_t)__s1, __p2); \
__ret; \
})
#else
uint16x8_t __s1 = __p1; \
uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
uint16x8_t __ret; \
- __ret = (uint16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int8x16_t)__rev1, __p2); \
+ __ret = (uint16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int16x8_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
__ret; \
})
uint16_t __s0 = __p0; \
uint16x8_t __s1 = __p1; \
uint16x8_t __ret; \
- __ret = (uint16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (uint16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int16x8_t)__s1, __p2); \
__ret; \
})
#endif
float32_t __s0 = __p0; \
float32x4_t __s1 = __p1; \
float32x4_t __ret; \
- __ret = (float32x4_t) __builtin_neon_vsetq_lane_f32(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (float32x4_t) __builtin_neon_vsetq_lane_f32(__s0, (float32x4_t)__s1, __p2); \
__ret; \
})
#else
float32x4_t __s1 = __p1; \
float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
float32x4_t __ret; \
- __ret = (float32x4_t) __builtin_neon_vsetq_lane_f32(__s0, (int8x16_t)__rev1, __p2); \
+ __ret = (float32x4_t) __builtin_neon_vsetq_lane_f32(__s0, (float32x4_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
__ret; \
})
float32_t __s0 = __p0; \
float32x4_t __s1 = __p1; \
float32x4_t __ret; \
- __ret = (float32x4_t) __builtin_neon_vsetq_lane_f32(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (float32x4_t) __builtin_neon_vsetq_lane_f32(__s0, (float32x4_t)__s1, __p2); \
__ret; \
})
#endif
int32_t __s0 = __p0; \
int32x4_t __s1 = __p1; \
int32x4_t __ret; \
- __ret = (int32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (int32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int32x4_t)__s1, __p2); \
__ret; \
})
#else
int32x4_t __s1 = __p1; \
int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
int32x4_t __ret; \
- __ret = (int32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int8x16_t)__rev1, __p2); \
+ __ret = (int32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int32x4_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
__ret; \
})
int32_t __s0 = __p0; \
int32x4_t __s1 = __p1; \
int32x4_t __ret; \
- __ret = (int32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (int32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int32x4_t)__s1, __p2); \
__ret; \
})
#endif
int64_t __s0 = __p0; \
int64x2_t __s1 = __p1; \
int64x2_t __ret; \
- __ret = (int64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (int64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int64x2_t)__s1, __p2); \
__ret; \
})
#else
int64x2_t __s1 = __p1; \
int64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
int64x2_t __ret; \
- __ret = (int64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int8x16_t)__rev1, __p2); \
+ __ret = (int64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int64x2_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
__ret; \
})
int64_t __s0 = __p0; \
int64x2_t __s1 = __p1; \
int64x2_t __ret; \
- __ret = (int64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (int64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int64x2_t)__s1, __p2); \
__ret; \
})
#endif
int16_t __s0 = __p0; \
int16x8_t __s1 = __p1; \
int16x8_t __ret; \
- __ret = (int16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (int16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int16x8_t)__s1, __p2); \
__ret; \
})
#else
int16x8_t __s1 = __p1; \
int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
int16x8_t __ret; \
- __ret = (int16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int8x16_t)__rev1, __p2); \
+ __ret = (int16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int16x8_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
__ret; \
})
int16_t __s0 = __p0; \
int16x8_t __s1 = __p1; \
int16x8_t __ret; \
- __ret = (int16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (int16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int16x8_t)__s1, __p2); \
__ret; \
})
#endif
uint32_t __s0 = __p0; \
uint32x2_t __s1 = __p1; \
uint32x2_t __ret; \
- __ret = (uint32x2_t) __builtin_neon_vset_lane_i32(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (uint32x2_t) __builtin_neon_vset_lane_i32(__s0, (int32x2_t)__s1, __p2); \
__ret; \
})
#else
uint32x2_t __s1 = __p1; \
uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
uint32x2_t __ret; \
- __ret = (uint32x2_t) __builtin_neon_vset_lane_i32(__s0, (int8x8_t)__rev1, __p2); \
+ __ret = (uint32x2_t) __builtin_neon_vset_lane_i32(__s0, (int32x2_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
__ret; \
})
uint32_t __s0 = __p0; \
uint32x2_t __s1 = __p1; \
uint32x2_t __ret; \
- __ret = (uint32x2_t) __builtin_neon_vset_lane_i32(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (uint32x2_t) __builtin_neon_vset_lane_i32(__s0, (int32x2_t)__s1, __p2); \
__ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vset_lane_u64(__p0, __p1, __p2) __extension__ ({ \
uint64_t __s0 = __p0; \
uint64x1_t __s1 = __p1; \
uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vset_lane_i64(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (uint64x1_t) __builtin_neon_vset_lane_i64(__s0, (int64x1_t)__s1, __p2); \
__ret; \
})
-#else
-#define vset_lane_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64_t __s0 = __p0; \
- uint64x1_t __s1 = __p1; \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vset_lane_i64(__s0, (int8x8_t)__s1, __p2); \
- __ret; \
-})
-#define __noswap_vset_lane_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64_t __s0 = __p0; \
- uint64x1_t __s1 = __p1; \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vset_lane_i64(__s0, (int8x8_t)__s1, __p2); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vset_lane_u16(__p0, __p1, __p2) __extension__ ({ \
uint16_t __s0 = __p0; \
uint16x4_t __s1 = __p1; \
uint16x4_t __ret; \
- __ret = (uint16x4_t) __builtin_neon_vset_lane_i16(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (uint16x4_t) __builtin_neon_vset_lane_i16(__s0, (int16x4_t)__s1, __p2); \
__ret; \
})
#else
uint16x4_t __s1 = __p1; \
uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
uint16x4_t __ret; \
- __ret = (uint16x4_t) __builtin_neon_vset_lane_i16(__s0, (int8x8_t)__rev1, __p2); \
+ __ret = (uint16x4_t) __builtin_neon_vset_lane_i16(__s0, (int16x4_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
__ret; \
})
uint16_t __s0 = __p0; \
uint16x4_t __s1 = __p1; \
uint16x4_t __ret; \
- __ret = (uint16x4_t) __builtin_neon_vset_lane_i16(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (uint16x4_t) __builtin_neon_vset_lane_i16(__s0, (int16x4_t)__s1, __p2); \
__ret; \
})
#endif
float32_t __s0 = __p0; \
float32x2_t __s1 = __p1; \
float32x2_t __ret; \
- __ret = (float32x2_t) __builtin_neon_vset_lane_f32(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (float32x2_t) __builtin_neon_vset_lane_f32(__s0, (float32x2_t)__s1, __p2); \
__ret; \
})
#else
float32x2_t __s1 = __p1; \
float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
float32x2_t __ret; \
- __ret = (float32x2_t) __builtin_neon_vset_lane_f32(__s0, (int8x8_t)__rev1, __p2); \
+ __ret = (float32x2_t) __builtin_neon_vset_lane_f32(__s0, (float32x2_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
__ret; \
})
float32_t __s0 = __p0; \
float32x2_t __s1 = __p1; \
float32x2_t __ret; \
- __ret = (float32x2_t) __builtin_neon_vset_lane_f32(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (float32x2_t) __builtin_neon_vset_lane_f32(__s0, (float32x2_t)__s1, __p2); \
__ret; \
})
#endif
int32_t __s0 = __p0; \
int32x2_t __s1 = __p1; \
int32x2_t __ret; \
- __ret = (int32x2_t) __builtin_neon_vset_lane_i32(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (int32x2_t) __builtin_neon_vset_lane_i32(__s0, (int32x2_t)__s1, __p2); \
__ret; \
})
#else
int32x2_t __s1 = __p1; \
int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
int32x2_t __ret; \
- __ret = (int32x2_t) __builtin_neon_vset_lane_i32(__s0, (int8x8_t)__rev1, __p2); \
+ __ret = (int32x2_t) __builtin_neon_vset_lane_i32(__s0, (int32x2_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
__ret; \
})
int32_t __s0 = __p0; \
int32x2_t __s1 = __p1; \
int32x2_t __ret; \
- __ret = (int32x2_t) __builtin_neon_vset_lane_i32(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (int32x2_t) __builtin_neon_vset_lane_i32(__s0, (int32x2_t)__s1, __p2); \
__ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vset_lane_s64(__p0, __p1, __p2) __extension__ ({ \
- int64_t __s0 = __p0; \
- int64x1_t __s1 = __p1; \
- int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vset_lane_i64(__s0, (int8x8_t)__s1, __p2); \
- __ret; \
-})
-#else
#define vset_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64_t __s0 = __p0; \
int64x1_t __s1 = __p1; \
int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vset_lane_i64(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (int64x1_t) __builtin_neon_vset_lane_i64(__s0, (int64x1_t)__s1, __p2); \
__ret; \
})
-#define __noswap_vset_lane_s64(__p0, __p1, __p2) __extension__ ({ \
- int64_t __s0 = __p0; \
- int64x1_t __s1 = __p1; \
- int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vset_lane_i64(__s0, (int8x8_t)__s1, __p2); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vset_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16_t __s0 = __p0; \
int16x4_t __s1 = __p1; \
int16x4_t __ret; \
- __ret = (int16x4_t) __builtin_neon_vset_lane_i16(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (int16x4_t) __builtin_neon_vset_lane_i16(__s0, (int16x4_t)__s1, __p2); \
__ret; \
})
#else
int16x4_t __s1 = __p1; \
int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
int16x4_t __ret; \
- __ret = (int16x4_t) __builtin_neon_vset_lane_i16(__s0, (int8x8_t)__rev1, __p2); \
+ __ret = (int16x4_t) __builtin_neon_vset_lane_i16(__s0, (int16x4_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
__ret; \
})
int16_t __s0 = __p0; \
int16x4_t __s1 = __p1; \
int16x4_t __ret; \
- __ret = (int16x4_t) __builtin_neon_vset_lane_i16(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (int16x4_t) __builtin_neon_vset_lane_i16(__s0, (int16x4_t)__s1, __p2); \
__ret; \
})
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vshl_u64(uint64x1_t __p0, int64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
- return __ret;
-}
-#else
__ai uint64x1_t vshl_u64(uint64x1_t __p0, int64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vshl_u16(uint16x4_t __p0, int16x4_t __p1) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vshl_s64(int64x1_t __p0, int64x1_t __p1) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
- return __ret;
-}
-#else
__ai int64x1_t vshl_s64(int64x1_t __p0, int64x1_t __p1) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
int16x4_t __ret;
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vshl_n_u64(__p0, __p1) __extension__ ({ \
uint64x1_t __s0 = __p0; \
uint64x1_t __ret; \
__ret = (uint64x1_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 19); \
__ret; \
})
-#else
-#define vshl_n_u64(__p0, __p1) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 19); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vshl_n_u16(__p0, __p1) __extension__ ({ \
uint16x4_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vshl_n_s64(__p0, __p1) __extension__ ({ \
int64x1_t __s0 = __p0; \
int64x1_t __ret; \
__ret = (int64x1_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 3); \
__ret; \
})
-#else
-#define vshl_n_s64(__p0, __p1) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 3); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vshl_n_s16(__p0, __p1) __extension__ ({ \
int16x4_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vshr_n_u64(__p0, __p1) __extension__ ({ \
uint64x1_t __s0 = __p0; \
uint64x1_t __ret; \
__ret = (uint64x1_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 19); \
__ret; \
})
-#else
-#define vshr_n_u64(__p0, __p1) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 19); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vshr_n_u16(__p0, __p1) __extension__ ({ \
uint16x4_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vshr_n_s64(__p0, __p1) __extension__ ({ \
int64x1_t __s0 = __p0; \
int64x1_t __ret; \
__ret = (int64x1_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 3); \
__ret; \
})
-#else
-#define vshr_n_s64(__p0, __p1) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 3); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vshr_n_s16(__p0, __p1) __extension__ ({ \
int16x4_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vsli_n_u64(__p0, __p1, __p2) __extension__ ({ \
uint64x1_t __s0 = __p0; \
uint64x1_t __s1 = __p1; \
__ret = (uint64x1_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 19); \
__ret; \
})
-#else
-#define vsli_n_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64x1_t __s1 = __p1; \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 19); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vsli_n_u16(__p0, __p1, __p2) __extension__ ({ \
uint16x4_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vsli_n_s64(__p0, __p1, __p2) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64x1_t __s1 = __p1; \
- int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 3); \
- __ret; \
-})
-#else
#define vsli_n_s64(__p0, __p1, __p2) __extension__ ({ \
int64x1_t __s0 = __p0; \
int64x1_t __s1 = __p1; \
__ret = (int64x1_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 3); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vsli_n_s16(__p0, __p1, __p2) __extension__ ({ \
int16x4_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vsra_n_u64(__p0, __p1, __p2) __extension__ ({ \
uint64x1_t __s0 = __p0; \
uint64x1_t __s1 = __p1; \
__ret = (uint64x1_t) __builtin_neon_vsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 19); \
__ret; \
})
-#else
-#define vsra_n_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64x1_t __s1 = __p1; \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 19); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vsra_n_u16(__p0, __p1, __p2) __extension__ ({ \
uint16x4_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vsra_n_s64(__p0, __p1, __p2) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64x1_t __s1 = __p1; \
- int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 3); \
- __ret; \
-})
-#else
#define vsra_n_s64(__p0, __p1, __p2) __extension__ ({ \
int64x1_t __s0 = __p0; \
int64x1_t __s1 = __p1; \
__ret = (int64x1_t) __builtin_neon_vsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 3); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vsra_n_s16(__p0, __p1, __p2) __extension__ ({ \
int16x4_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vsri_n_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64x1_t __s1 = __p1; \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 19); \
- __ret; \
-})
-#else
#define vsri_n_u64(__p0, __p1, __p2) __extension__ ({ \
uint64x1_t __s0 = __p0; \
uint64x1_t __s1 = __p1; \
__ret = (uint64x1_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 19); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vsri_n_u16(__p0, __p1, __p2) __extension__ ({ \
uint16x4_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vsri_n_s64(__p0, __p1, __p2) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64x1_t __s1 = __p1; \
- int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 3); \
- __ret; \
-})
-#else
#define vsri_n_s64(__p0, __p1, __p2) __extension__ ({ \
int64x1_t __s0 = __p0; \
int64x1_t __s1 = __p1; \
__ret = (int64x1_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 3); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vsri_n_s16(__p0, __p1, __p2) __extension__ ({ \
int16x4_t __s0 = __p0; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vst1q_f16(__p0, __p1) __extension__ ({ \
- float16x8_t __s1 = __p1; \
- __builtin_neon_vst1q_v(__p0, (int8x16_t)__s1, 40); \
-})
-#else
-#define vst1q_f16(__p0, __p1) __extension__ ({ \
- float16x8_t __s1 = __p1; \
- float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst1q_v(__p0, (int8x16_t)__rev1, 40); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vst1q_s32(__p0, __p1) __extension__ ({ \
int32x4_t __s1 = __p1; \
__builtin_neon_vst1q_v(__p0, (int8x16_t)__s1, 34); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vst1_u64(__p0, __p1) __extension__ ({ \
uint64x1_t __s1 = __p1; \
__builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 19); \
})
-#else
-#define vst1_u64(__p0, __p1) __extension__ ({ \
- uint64x1_t __s1 = __p1; \
- __builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 19); \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst1_u16(__p0, __p1) __extension__ ({ \
uint16x4_t __s1 = __p1; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vst1_f16(__p0, __p1) __extension__ ({ \
- float16x4_t __s1 = __p1; \
- __builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 8); \
-})
-#else
-#define vst1_f16(__p0, __p1) __extension__ ({ \
- float16x4_t __s1 = __p1; \
- float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- __builtin_neon_vst1_v(__p0, (int8x8_t)__rev1, 8); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vst1_s32(__p0, __p1) __extension__ ({ \
int32x2_t __s1 = __p1; \
__builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst1_s64(__p0, __p1) __extension__ ({ \
- int64x1_t __s1 = __p1; \
- __builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 3); \
-})
-#else
#define vst1_s64(__p0, __p1) __extension__ ({ \
int64x1_t __s1 = __p1; \
__builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 3); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst1_s16(__p0, __p1) __extension__ ({ \
int16x4_t __s1 = __p1; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vst1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8_t __s1 = __p1; \
- __builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__s1, __p2, 40); \
-})
-#else
-#define vst1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8_t __s1 = __p1; \
- float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 40); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vst1q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x4_t __s1 = __p1; \
__builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__s1, __p2, 34); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vst1_lane_u64(__p0, __p1, __p2) __extension__ ({ \
uint64x1_t __s1 = __p1; \
__builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 19); \
})
-#else
-#define vst1_lane_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64x1_t __s1 = __p1; \
- __builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 19); \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst1_lane_u16(__p0, __p1, __p2) __extension__ ({ \
uint16x4_t __s1 = __p1; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vst1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4_t __s1 = __p1; \
- __builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 8); \
-})
-#else
-#define vst1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4_t __s1 = __p1; \
- float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- __builtin_neon_vst1_lane_v(__p0, (int8x8_t)__rev1, __p2, 8); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vst1_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x2_t __s1 = __p1; \
__builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst1_lane_s64(__p0, __p1, __p2) __extension__ ({ \
- int64x1_t __s1 = __p1; \
- __builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 3); \
-})
-#else
#define vst1_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64x1_t __s1 = __p1; \
__builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 3); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst1_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x4_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst1q_f32_x2(__p0, __p1) __extension__ ({ \
float32x4x2_t __s1 = __p1; \
- __builtin_neon_vst1q_x2_v(__p0, __s1.val[0], __s1.val[1], 41); \
+ __builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 41); \
})
#else
#define vst1q_f32_x2(__p0, __p1) __extension__ ({ \
float32x4x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __builtin_neon_vst1q_x2_v(__p0, __rev1.val[0], __rev1.val[1], 41); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst1q_f16_x2(__p0, __p1) __extension__ ({ \
- float16x8x2_t __s1 = __p1; \
- __builtin_neon_vst1q_x2_v(__p0, __s1.val[0], __s1.val[1], 40); \
-})
-#else
-#define vst1q_f16_x2(__p0, __p1) __extension__ ({ \
- float16x8x2_t __s1 = __p1; \
- float16x8x2_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst1q_x2_v(__p0, __rev1.val[0], __rev1.val[1], 40); \
+ __builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 41); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst1q_s32_x2(__p0, __p1) __extension__ ({ \
int32x4x2_t __s1 = __p1; \
- __builtin_neon_vst1q_x2_v(__p0, __s1.val[0], __s1.val[1], 34); \
+ __builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 34); \
})
#else
#define vst1q_s32_x2(__p0, __p1) __extension__ ({ \
int32x4x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __builtin_neon_vst1q_x2_v(__p0, __rev1.val[0], __rev1.val[1], 34); \
+ __builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 34); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst1q_s64_x2(__p0, __p1) __extension__ ({ \
int64x2x2_t __s1 = __p1; \
- __builtin_neon_vst1q_x2_v(__p0, __s1.val[0], __s1.val[1], 35); \
+ __builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 35); \
})
#else
#define vst1q_s64_x2(__p0, __p1) __extension__ ({ \
int64x2x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
- __builtin_neon_vst1q_x2_v(__p0, __rev1.val[0], __rev1.val[1], 35); \
+ __builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 35); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst1q_s16_x2(__p0, __p1) __extension__ ({ \
int16x8x2_t __s1 = __p1; \
- __builtin_neon_vst1q_x2_v(__p0, __s1.val[0], __s1.val[1], 33); \
+ __builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 33); \
})
#else
#define vst1q_s16_x2(__p0, __p1) __extension__ ({ \
int16x8x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst1q_x2_v(__p0, __rev1.val[0], __rev1.val[1], 33); \
+ __builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 33); \
})
#endif
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst1_u64_x2(__p0, __p1) __extension__ ({ \
- uint64x1x2_t __s1 = __p1; \
- __builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 19); \
-})
-#else
#define vst1_u64_x2(__p0, __p1) __extension__ ({ \
uint64x1x2_t __s1 = __p1; \
__builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 19); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst1_u16_x2(__p0, __p1) __extension__ ({ \
uint16x4x2_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst1_f32_x2(__p0, __p1) __extension__ ({ \
float32x2x2_t __s1 = __p1; \
- __builtin_neon_vst1_x2_v(__p0, __s1.val[0], __s1.val[1], 9); \
+ __builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 9); \
})
#else
#define vst1_f32_x2(__p0, __p1) __extension__ ({ \
float32x2x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
- __builtin_neon_vst1_x2_v(__p0, __rev1.val[0], __rev1.val[1], 9); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst1_f16_x2(__p0, __p1) __extension__ ({ \
- float16x4x2_t __s1 = __p1; \
- __builtin_neon_vst1_x2_v(__p0, __s1.val[0], __s1.val[1], 8); \
-})
-#else
-#define vst1_f16_x2(__p0, __p1) __extension__ ({ \
- float16x4x2_t __s1 = __p1; \
- float16x4x2_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __builtin_neon_vst1_x2_v(__p0, __rev1.val[0], __rev1.val[1], 8); \
+ __builtin_neon_vst1_x2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 9); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst1_s32_x2(__p0, __p1) __extension__ ({ \
int32x2x2_t __s1 = __p1; \
- __builtin_neon_vst1_x2_v(__p0, __s1.val[0], __s1.val[1], 2); \
+ __builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 2); \
})
#else
#define vst1_s32_x2(__p0, __p1) __extension__ ({ \
int32x2x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
- __builtin_neon_vst1_x2_v(__p0, __rev1.val[0], __rev1.val[1], 2); \
+ __builtin_neon_vst1_x2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vst1_s64_x2(__p0, __p1) __extension__ ({ \
int64x1x2_t __s1 = __p1; \
- __builtin_neon_vst1_x2_v(__p0, __s1.val[0], __s1.val[1], 3); \
+ __builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 3); \
})
-#else
-#define vst1_s64_x2(__p0, __p1) __extension__ ({ \
- int64x1x2_t __s1 = __p1; \
- __builtin_neon_vst1_x2_v(__p0, __s1.val[0], __s1.val[1], 3); \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst1_s16_x2(__p0, __p1) __extension__ ({ \
int16x4x2_t __s1 = __p1; \
- __builtin_neon_vst1_x2_v(__p0, __s1.val[0], __s1.val[1], 1); \
+ __builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 1); \
})
#else
#define vst1_s16_x2(__p0, __p1) __extension__ ({ \
int16x4x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __builtin_neon_vst1_x2_v(__p0, __rev1.val[0], __rev1.val[1], 1); \
+ __builtin_neon_vst1_x2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 1); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst1q_f32_x3(__p0, __p1) __extension__ ({ \
float32x4x3_t __s1 = __p1; \
- __builtin_neon_vst1q_x3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 41); \
+ __builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 41); \
})
#else
#define vst1q_f32_x3(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __builtin_neon_vst1q_x3_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 41); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst1q_f16_x3(__p0, __p1) __extension__ ({ \
- float16x8x3_t __s1 = __p1; \
- __builtin_neon_vst1q_x3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 40); \
-})
-#else
-#define vst1q_f16_x3(__p0, __p1) __extension__ ({ \
- float16x8x3_t __s1 = __p1; \
- float16x8x3_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst1q_x3_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 40); \
+ __builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 41); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst1q_s32_x3(__p0, __p1) __extension__ ({ \
int32x4x3_t __s1 = __p1; \
- __builtin_neon_vst1q_x3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 34); \
+ __builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 34); \
})
#else
#define vst1q_s32_x3(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __builtin_neon_vst1q_x3_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 34); \
+ __builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 34); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst1q_s64_x3(__p0, __p1) __extension__ ({ \
int64x2x3_t __s1 = __p1; \
- __builtin_neon_vst1q_x3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 35); \
+ __builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 35); \
})
#else
#define vst1q_s64_x3(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
- __builtin_neon_vst1q_x3_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 35); \
+ __builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 35); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst1q_s16_x3(__p0, __p1) __extension__ ({ \
int16x8x3_t __s1 = __p1; \
- __builtin_neon_vst1q_x3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 33); \
+ __builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 33); \
})
#else
#define vst1q_s16_x3(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst1q_x3_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 33); \
+ __builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 33); \
})
#endif
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst1_u64_x3(__p0, __p1) __extension__ ({ \
- uint64x1x3_t __s1 = __p1; \
- __builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 19); \
-})
-#else
#define vst1_u64_x3(__p0, __p1) __extension__ ({ \
uint64x1x3_t __s1 = __p1; \
__builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 19); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst1_u16_x3(__p0, __p1) __extension__ ({ \
uint16x4x3_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst1_f32_x3(__p0, __p1) __extension__ ({ \
float32x2x3_t __s1 = __p1; \
- __builtin_neon_vst1_x3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 9); \
+ __builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 9); \
})
#else
#define vst1_f32_x3(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
- __builtin_neon_vst1_x3_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 9); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst1_f16_x3(__p0, __p1) __extension__ ({ \
- float16x4x3_t __s1 = __p1; \
- __builtin_neon_vst1_x3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 8); \
-})
-#else
-#define vst1_f16_x3(__p0, __p1) __extension__ ({ \
- float16x4x3_t __s1 = __p1; \
- float16x4x3_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __builtin_neon_vst1_x3_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 8); \
+ __builtin_neon_vst1_x3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 9); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst1_s32_x3(__p0, __p1) __extension__ ({ \
int32x2x3_t __s1 = __p1; \
- __builtin_neon_vst1_x3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 2); \
+ __builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 2); \
})
#else
#define vst1_s32_x3(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
- __builtin_neon_vst1_x3_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 2); \
+ __builtin_neon_vst1_x3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst1_s64_x3(__p0, __p1) __extension__ ({ \
- int64x1x3_t __s1 = __p1; \
- __builtin_neon_vst1_x3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 3); \
-})
-#else
#define vst1_s64_x3(__p0, __p1) __extension__ ({ \
int64x1x3_t __s1 = __p1; \
- __builtin_neon_vst1_x3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 3); \
+ __builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 3); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst1_s16_x3(__p0, __p1) __extension__ ({ \
int16x4x3_t __s1 = __p1; \
- __builtin_neon_vst1_x3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 1); \
+ __builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 1); \
})
#else
#define vst1_s16_x3(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __builtin_neon_vst1_x3_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 1); \
+ __builtin_neon_vst1_x3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 1); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst1q_f32_x4(__p0, __p1) __extension__ ({ \
float32x4x4_t __s1 = __p1; \
- __builtin_neon_vst1q_x4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 41); \
+ __builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 41); \
})
#else
#define vst1q_f32_x4(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
- __builtin_neon_vst1q_x4_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 41); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst1q_f16_x4(__p0, __p1) __extension__ ({ \
- float16x8x4_t __s1 = __p1; \
- __builtin_neon_vst1q_x4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 40); \
-})
-#else
-#define vst1q_f16_x4(__p0, __p1) __extension__ ({ \
- float16x8x4_t __s1 = __p1; \
- float16x8x4_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst1q_x4_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 40); \
+ __builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 41); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst1q_s32_x4(__p0, __p1) __extension__ ({ \
int32x4x4_t __s1 = __p1; \
- __builtin_neon_vst1q_x4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 34); \
+ __builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 34); \
})
#else
#define vst1q_s32_x4(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
- __builtin_neon_vst1q_x4_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 34); \
+ __builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 34); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst1q_s64_x4(__p0, __p1) __extension__ ({ \
int64x2x4_t __s1 = __p1; \
- __builtin_neon_vst1q_x4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 35); \
+ __builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 35); \
})
#else
#define vst1q_s64_x4(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
- __builtin_neon_vst1q_x4_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 35); \
+ __builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 35); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst1q_s16_x4(__p0, __p1) __extension__ ({ \
int16x8x4_t __s1 = __p1; \
- __builtin_neon_vst1q_x4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 33); \
+ __builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 33); \
})
#else
#define vst1q_s16_x4(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst1q_x4_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 33); \
+ __builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 33); \
})
#endif
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vst1_u64_x4(__p0, __p1) __extension__ ({ \
uint64x1x4_t __s1 = __p1; \
__builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 19); \
})
-#else
-#define vst1_u64_x4(__p0, __p1) __extension__ ({ \
- uint64x1x4_t __s1 = __p1; \
- __builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 19); \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst1_u16_x4(__p0, __p1) __extension__ ({ \
uint16x4x4_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst1_f32_x4(__p0, __p1) __extension__ ({ \
float32x2x4_t __s1 = __p1; \
- __builtin_neon_vst1_x4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 9); \
+ __builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 9); \
})
#else
#define vst1_f32_x4(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
- __builtin_neon_vst1_x4_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 9); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst1_f16_x4(__p0, __p1) __extension__ ({ \
- float16x4x4_t __s1 = __p1; \
- __builtin_neon_vst1_x4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 8); \
-})
-#else
-#define vst1_f16_x4(__p0, __p1) __extension__ ({ \
- float16x4x4_t __s1 = __p1; \
- float16x4x4_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
- __builtin_neon_vst1_x4_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 8); \
+ __builtin_neon_vst1_x4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 9); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst1_s32_x4(__p0, __p1) __extension__ ({ \
int32x2x4_t __s1 = __p1; \
- __builtin_neon_vst1_x4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 2); \
+ __builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 2); \
})
#else
#define vst1_s32_x4(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
- __builtin_neon_vst1_x4_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 2); \
+ __builtin_neon_vst1_x4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst1_s64_x4(__p0, __p1) __extension__ ({ \
- int64x1x4_t __s1 = __p1; \
- __builtin_neon_vst1_x4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 3); \
-})
-#else
#define vst1_s64_x4(__p0, __p1) __extension__ ({ \
int64x1x4_t __s1 = __p1; \
- __builtin_neon_vst1_x4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 3); \
+ __builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 3); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst1_s16_x4(__p0, __p1) __extension__ ({ \
int16x4x4_t __s1 = __p1; \
- __builtin_neon_vst1_x4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 1); \
+ __builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 1); \
})
#else
#define vst1_s16_x4(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
- __builtin_neon_vst1_x4_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 1); \
+ __builtin_neon_vst1_x4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 1); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst2q_f32(__p0, __p1) __extension__ ({ \
float32x4x2_t __s1 = __p1; \
- __builtin_neon_vst2q_v(__p0, __s1.val[0], __s1.val[1], 41); \
+ __builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 41); \
})
#else
#define vst2q_f32(__p0, __p1) __extension__ ({ \
float32x4x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __builtin_neon_vst2q_v(__p0, __rev1.val[0], __rev1.val[1], 41); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst2q_f16(__p0, __p1) __extension__ ({ \
- float16x8x2_t __s1 = __p1; \
- __builtin_neon_vst2q_v(__p0, __s1.val[0], __s1.val[1], 40); \
-})
-#else
-#define vst2q_f16(__p0, __p1) __extension__ ({ \
- float16x8x2_t __s1 = __p1; \
- float16x8x2_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst2q_v(__p0, __rev1.val[0], __rev1.val[1], 40); \
+ __builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 41); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst2q_s32(__p0, __p1) __extension__ ({ \
int32x4x2_t __s1 = __p1; \
- __builtin_neon_vst2q_v(__p0, __s1.val[0], __s1.val[1], 34); \
+ __builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 34); \
})
#else
#define vst2q_s32(__p0, __p1) __extension__ ({ \
int32x4x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __builtin_neon_vst2q_v(__p0, __rev1.val[0], __rev1.val[1], 34); \
+ __builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 34); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst2q_s16(__p0, __p1) __extension__ ({ \
int16x8x2_t __s1 = __p1; \
- __builtin_neon_vst2q_v(__p0, __s1.val[0], __s1.val[1], 33); \
+ __builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 33); \
})
#else
#define vst2q_s16(__p0, __p1) __extension__ ({ \
int16x8x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst2q_v(__p0, __rev1.val[0], __rev1.val[1], 33); \
+ __builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 33); \
})
#endif
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst2_u64(__p0, __p1) __extension__ ({ \
- uint64x1x2_t __s1 = __p1; \
- __builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 19); \
-})
-#else
#define vst2_u64(__p0, __p1) __extension__ ({ \
uint64x1x2_t __s1 = __p1; \
__builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 19); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst2_u16(__p0, __p1) __extension__ ({ \
uint16x4x2_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst2_f32(__p0, __p1) __extension__ ({ \
float32x2x2_t __s1 = __p1; \
- __builtin_neon_vst2_v(__p0, __s1.val[0], __s1.val[1], 9); \
+ __builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 9); \
})
#else
#define vst2_f32(__p0, __p1) __extension__ ({ \
float32x2x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
- __builtin_neon_vst2_v(__p0, __rev1.val[0], __rev1.val[1], 9); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst2_f16(__p0, __p1) __extension__ ({ \
- float16x4x2_t __s1 = __p1; \
- __builtin_neon_vst2_v(__p0, __s1.val[0], __s1.val[1], 8); \
-})
-#else
-#define vst2_f16(__p0, __p1) __extension__ ({ \
- float16x4x2_t __s1 = __p1; \
- float16x4x2_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __builtin_neon_vst2_v(__p0, __rev1.val[0], __rev1.val[1], 8); \
+ __builtin_neon_vst2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 9); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst2_s32(__p0, __p1) __extension__ ({ \
int32x2x2_t __s1 = __p1; \
- __builtin_neon_vst2_v(__p0, __s1.val[0], __s1.val[1], 2); \
+ __builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 2); \
})
#else
#define vst2_s32(__p0, __p1) __extension__ ({ \
int32x2x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
- __builtin_neon_vst2_v(__p0, __rev1.val[0], __rev1.val[1], 2); \
+ __builtin_neon_vst2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vst2_s64(__p0, __p1) __extension__ ({ \
int64x1x2_t __s1 = __p1; \
- __builtin_neon_vst2_v(__p0, __s1.val[0], __s1.val[1], 3); \
+ __builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 3); \
})
-#else
-#define vst2_s64(__p0, __p1) __extension__ ({ \
- int64x1x2_t __s1 = __p1; \
- __builtin_neon_vst2_v(__p0, __s1.val[0], __s1.val[1], 3); \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst2_s16(__p0, __p1) __extension__ ({ \
int16x4x2_t __s1 = __p1; \
- __builtin_neon_vst2_v(__p0, __s1.val[0], __s1.val[1], 1); \
+ __builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 1); \
})
#else
#define vst2_s16(__p0, __p1) __extension__ ({ \
int16x4x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __builtin_neon_vst2_v(__p0, __rev1.val[0], __rev1.val[1], 1); \
+ __builtin_neon_vst2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 1); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst2q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
float32x4x2_t __s1 = __p1; \
- __builtin_neon_vst2q_lane_v(__p0, __s1.val[0], __s1.val[1], __p2, 41); \
+ __builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 41); \
})
#else
#define vst2q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
float32x4x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __builtin_neon_vst2q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __p2, 41); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8x2_t __s1 = __p1; \
- __builtin_neon_vst2q_lane_v(__p0, __s1.val[0], __s1.val[1], __p2, 40); \
-})
-#else
-#define vst2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8x2_t __s1 = __p1; \
- float16x8x2_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst2q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __p2, 40); \
+ __builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 41); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst2q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x4x2_t __s1 = __p1; \
- __builtin_neon_vst2q_lane_v(__p0, __s1.val[0], __s1.val[1], __p2, 34); \
+ __builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 34); \
})
#else
#define vst2q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x4x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __builtin_neon_vst2q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __p2, 34); \
+ __builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 34); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst2q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x8x2_t __s1 = __p1; \
- __builtin_neon_vst2q_lane_v(__p0, __s1.val[0], __s1.val[1], __p2, 33); \
+ __builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 33); \
})
#else
#define vst2q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x8x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst2q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __p2, 33); \
+ __builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 33); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst2_lane_f32(__p0, __p1, __p2) __extension__ ({ \
float32x2x2_t __s1 = __p1; \
- __builtin_neon_vst2_lane_v(__p0, __s1.val[0], __s1.val[1], __p2, 9); \
+ __builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 9); \
})
#else
#define vst2_lane_f32(__p0, __p1, __p2) __extension__ ({ \
float32x2x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
- __builtin_neon_vst2_lane_v(__p0, __rev1.val[0], __rev1.val[1], __p2, 9); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4x2_t __s1 = __p1; \
- __builtin_neon_vst2_lane_v(__p0, __s1.val[0], __s1.val[1], __p2, 8); \
-})
-#else
-#define vst2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4x2_t __s1 = __p1; \
- float16x4x2_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __builtin_neon_vst2_lane_v(__p0, __rev1.val[0], __rev1.val[1], __p2, 8); \
+ __builtin_neon_vst2_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 9); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst2_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x2x2_t __s1 = __p1; \
- __builtin_neon_vst2_lane_v(__p0, __s1.val[0], __s1.val[1], __p2, 2); \
+ __builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 2); \
})
#else
#define vst2_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x2x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
- __builtin_neon_vst2_lane_v(__p0, __rev1.val[0], __rev1.val[1], __p2, 2); \
+ __builtin_neon_vst2_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 2); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst2_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x4x2_t __s1 = __p1; \
- __builtin_neon_vst2_lane_v(__p0, __s1.val[0], __s1.val[1], __p2, 1); \
+ __builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 1); \
})
#else
#define vst2_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x4x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __builtin_neon_vst2_lane_v(__p0, __rev1.val[0], __rev1.val[1], __p2, 1); \
+ __builtin_neon_vst2_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 1); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst3q_f32(__p0, __p1) __extension__ ({ \
float32x4x3_t __s1 = __p1; \
- __builtin_neon_vst3q_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 41); \
+ __builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 41); \
})
#else
#define vst3q_f32(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __builtin_neon_vst3q_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 41); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst3q_f16(__p0, __p1) __extension__ ({ \
- float16x8x3_t __s1 = __p1; \
- __builtin_neon_vst3q_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 40); \
-})
-#else
-#define vst3q_f16(__p0, __p1) __extension__ ({ \
- float16x8x3_t __s1 = __p1; \
- float16x8x3_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst3q_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 40); \
+ __builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 41); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst3q_s32(__p0, __p1) __extension__ ({ \
int32x4x3_t __s1 = __p1; \
- __builtin_neon_vst3q_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 34); \
+ __builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 34); \
})
#else
#define vst3q_s32(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __builtin_neon_vst3q_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 34); \
+ __builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 34); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst3q_s16(__p0, __p1) __extension__ ({ \
int16x8x3_t __s1 = __p1; \
- __builtin_neon_vst3q_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 33); \
+ __builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 33); \
})
#else
#define vst3q_s16(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst3q_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 33); \
+ __builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 33); \
})
#endif
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst3_u64(__p0, __p1) __extension__ ({ \
- uint64x1x3_t __s1 = __p1; \
- __builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 19); \
-})
-#else
#define vst3_u64(__p0, __p1) __extension__ ({ \
uint64x1x3_t __s1 = __p1; \
__builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 19); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst3_u16(__p0, __p1) __extension__ ({ \
uint16x4x3_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst3_f32(__p0, __p1) __extension__ ({ \
float32x2x3_t __s1 = __p1; \
- __builtin_neon_vst3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 9); \
+ __builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 9); \
})
#else
#define vst3_f32(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
- __builtin_neon_vst3_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 9); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst3_f16(__p0, __p1) __extension__ ({ \
- float16x4x3_t __s1 = __p1; \
- __builtin_neon_vst3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 8); \
-})
-#else
-#define vst3_f16(__p0, __p1) __extension__ ({ \
- float16x4x3_t __s1 = __p1; \
- float16x4x3_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __builtin_neon_vst3_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 8); \
+ __builtin_neon_vst3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 9); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst3_s32(__p0, __p1) __extension__ ({ \
int32x2x3_t __s1 = __p1; \
- __builtin_neon_vst3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 2); \
+ __builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 2); \
})
#else
#define vst3_s32(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
- __builtin_neon_vst3_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 2); \
+ __builtin_neon_vst3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vst3_s64(__p0, __p1) __extension__ ({ \
int64x1x3_t __s1 = __p1; \
- __builtin_neon_vst3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 3); \
+ __builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 3); \
})
-#else
-#define vst3_s64(__p0, __p1) __extension__ ({ \
- int64x1x3_t __s1 = __p1; \
- __builtin_neon_vst3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 3); \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst3_s16(__p0, __p1) __extension__ ({ \
int16x4x3_t __s1 = __p1; \
- __builtin_neon_vst3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 1); \
+ __builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 1); \
})
#else
#define vst3_s16(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __builtin_neon_vst3_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 1); \
+ __builtin_neon_vst3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 1); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst3q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
float32x4x3_t __s1 = __p1; \
- __builtin_neon_vst3q_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 41); \
+ __builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 41); \
})
#else
#define vst3q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __builtin_neon_vst3q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 41); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8x3_t __s1 = __p1; \
- __builtin_neon_vst3q_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 40); \
-})
-#else
-#define vst3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8x3_t __s1 = __p1; \
- float16x8x3_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst3q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 40); \
+ __builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 41); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst3q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x4x3_t __s1 = __p1; \
- __builtin_neon_vst3q_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 34); \
+ __builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 34); \
})
#else
#define vst3q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __builtin_neon_vst3q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 34); \
+ __builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 34); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst3q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x8x3_t __s1 = __p1; \
- __builtin_neon_vst3q_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 33); \
+ __builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 33); \
})
#else
#define vst3q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst3q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 33); \
+ __builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 33); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst3_lane_f32(__p0, __p1, __p2) __extension__ ({ \
float32x2x3_t __s1 = __p1; \
- __builtin_neon_vst3_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 9); \
+ __builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 9); \
})
#else
#define vst3_lane_f32(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
- __builtin_neon_vst3_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 9); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4x3_t __s1 = __p1; \
- __builtin_neon_vst3_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 8); \
-})
-#else
-#define vst3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4x3_t __s1 = __p1; \
- float16x4x3_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __builtin_neon_vst3_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 8); \
+ __builtin_neon_vst3_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 9); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst3_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x2x3_t __s1 = __p1; \
- __builtin_neon_vst3_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 2); \
+ __builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 2); \
})
#else
#define vst3_lane_s32(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
- __builtin_neon_vst3_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 2); \
+ __builtin_neon_vst3_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 2); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst3_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x4x3_t __s1 = __p1; \
- __builtin_neon_vst3_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 1); \
+ __builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 1); \
})
#else
#define vst3_lane_s16(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __builtin_neon_vst3_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 1); \
+ __builtin_neon_vst3_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 1); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst4q_f32(__p0, __p1) __extension__ ({ \
float32x4x4_t __s1 = __p1; \
- __builtin_neon_vst4q_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 41); \
+ __builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 41); \
})
#else
#define vst4q_f32(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
- __builtin_neon_vst4q_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 41); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst4q_f16(__p0, __p1) __extension__ ({ \
- float16x8x4_t __s1 = __p1; \
- __builtin_neon_vst4q_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 40); \
-})
-#else
-#define vst4q_f16(__p0, __p1) __extension__ ({ \
- float16x8x4_t __s1 = __p1; \
- float16x8x4_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst4q_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 40); \
+ __builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 41); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst4q_s32(__p0, __p1) __extension__ ({ \
int32x4x4_t __s1 = __p1; \
- __builtin_neon_vst4q_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 34); \
+ __builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 34); \
})
#else
#define vst4q_s32(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
- __builtin_neon_vst4q_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 34); \
+ __builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 34); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst4q_s16(__p0, __p1) __extension__ ({ \
int16x8x4_t __s1 = __p1; \
- __builtin_neon_vst4q_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 33); \
+ __builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 33); \
})
#else
#define vst4q_s16(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst4q_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 33); \
+ __builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 33); \
})
#endif
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst4_u64(__p0, __p1) __extension__ ({ \
- uint64x1x4_t __s1 = __p1; \
- __builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 19); \
-})
-#else
#define vst4_u64(__p0, __p1) __extension__ ({ \
uint64x1x4_t __s1 = __p1; \
__builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 19); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst4_u16(__p0, __p1) __extension__ ({ \
uint16x4x4_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst4_f32(__p0, __p1) __extension__ ({ \
float32x2x4_t __s1 = __p1; \
- __builtin_neon_vst4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 9); \
+ __builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 9); \
})
#else
#define vst4_f32(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
- __builtin_neon_vst4_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 9); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst4_f16(__p0, __p1) __extension__ ({ \
- float16x4x4_t __s1 = __p1; \
- __builtin_neon_vst4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 8); \
-})
-#else
-#define vst4_f16(__p0, __p1) __extension__ ({ \
- float16x4x4_t __s1 = __p1; \
- float16x4x4_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
- __builtin_neon_vst4_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 8); \
+ __builtin_neon_vst4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 9); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst4_s32(__p0, __p1) __extension__ ({ \
int32x2x4_t __s1 = __p1; \
- __builtin_neon_vst4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 2); \
+ __builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 2); \
})
#else
#define vst4_s32(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
- __builtin_neon_vst4_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 2); \
+ __builtin_neon_vst4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 2); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst4_s64(__p0, __p1) __extension__ ({ \
- int64x1x4_t __s1 = __p1; \
- __builtin_neon_vst4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 3); \
-})
-#else
#define vst4_s64(__p0, __p1) __extension__ ({ \
int64x1x4_t __s1 = __p1; \
- __builtin_neon_vst4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 3); \
+ __builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 3); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst4_s16(__p0, __p1) __extension__ ({ \
int16x4x4_t __s1 = __p1; \
- __builtin_neon_vst4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 1); \
+ __builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 1); \
})
#else
#define vst4_s16(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
- __builtin_neon_vst4_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 1); \
+ __builtin_neon_vst4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 1); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst4q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
float32x4x4_t __s1 = __p1; \
- __builtin_neon_vst4q_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 41); \
+ __builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 41); \
})
#else
#define vst4q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
- __builtin_neon_vst4q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 41); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8x4_t __s1 = __p1; \
- __builtin_neon_vst4q_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 40); \
-})
-#else
-#define vst4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8x4_t __s1 = __p1; \
- float16x8x4_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst4q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 40); \
+ __builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 41); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst4q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x4x4_t __s1 = __p1; \
- __builtin_neon_vst4q_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 34); \
+ __builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 34); \
})
#else
#define vst4q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
- __builtin_neon_vst4q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 34); \
+ __builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 34); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst4q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x8x4_t __s1 = __p1; \
- __builtin_neon_vst4q_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 33); \
+ __builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 33); \
})
#else
#define vst4q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
- __builtin_neon_vst4q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 33); \
+ __builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 33); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst4_lane_f32(__p0, __p1, __p2) __extension__ ({ \
float32x2x4_t __s1 = __p1; \
- __builtin_neon_vst4_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 9); \
+ __builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 9); \
})
#else
#define vst4_lane_f32(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
- __builtin_neon_vst4_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 9); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4x4_t __s1 = __p1; \
- __builtin_neon_vst4_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 8); \
-})
-#else
-#define vst4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4x4_t __s1 = __p1; \
- float16x4x4_t __rev1; \
- __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
- __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
- __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
- __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
- __builtin_neon_vst4_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 8); \
+ __builtin_neon_vst4_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 9); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst4_lane_s32(__p0, __p1, __p2) __extension__ ({ \
int32x2x4_t __s1 = __p1; \
- __builtin_neon_vst4_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 2); \
+ __builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 2); \
})
#else
#define vst4_lane_s32(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
- __builtin_neon_vst4_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 2); \
+ __builtin_neon_vst4_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 2); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst4_lane_s16(__p0, __p1, __p2) __extension__ ({ \
int16x4x4_t __s1 = __p1; \
- __builtin_neon_vst4_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 1); \
+ __builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 1); \
})
#else
#define vst4_lane_s16(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
- __builtin_neon_vst4_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 1); \
+ __builtin_neon_vst4_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 1); \
})
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vsub_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = __p0 - __p1;
- return __ret;
-}
-#else
__ai uint64x1_t vsub_u64(uint64x1_t __p0, uint64x1_t __p1) {
uint64x1_t __ret;
__ret = __p0 - __p1;
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai int64x1_t vsub_s64(int64x1_t __p0, int64x1_t __p1) {
int64x1_t __ret;
__ret = __p0 - __p1;
return __ret;
}
-#else
-__ai int64x1_t vsub_s64(int64x1_t __p0, int64x1_t __p1) {
- int64x1_t __ret;
- __ret = __p0 - __p1;
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int16x4_t vsub_s16(int16x4_t __p0, int16x4_t __p1) {
int16x4_t __ret;
#if !defined(__aarch64__)
#ifdef __LITTLE_ENDIAN__
-#define vdupq_lane_f16(__p0, __p1) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x8_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_lane_f16(__p0_122, __p1_122) __extension__ ({ \
+ float16x4_t __s0_122 = __p0_122; \
+ float16x8_t __ret_122; \
+ __ret_122 = splatq_lane_f16(__s0_122, __p1_122); \
+ __ret_122; \
})
#else
-#define vdupq_lane_f16(__p0, __p1) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float16x8_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_lane_f16(__p0_123, __p1_123) __extension__ ({ \
+ float16x4_t __s0_123 = __p0_123; \
+ float16x4_t __rev0_123; __rev0_123 = __builtin_shufflevector(__s0_123, __s0_123, 3, 2, 1, 0); \
+ float16x8_t __ret_123; \
+ __ret_123 = __noswap_splatq_lane_f16(__rev0_123, __p1_123); \
+ __ret_123 = __builtin_shufflevector(__ret_123, __ret_123, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_123; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_lane_f16(__p0, __p1) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x4_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdup_lane_f16(__p0_124, __p1_124) __extension__ ({ \
+ float16x4_t __s0_124 = __p0_124; \
+ float16x4_t __ret_124; \
+ __ret_124 = splat_lane_f16(__s0_124, __p1_124); \
+ __ret_124; \
})
#else
-#define vdup_lane_f16(__p0, __p1) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float16x4_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdup_lane_f16(__p0_125, __p1_125) __extension__ ({ \
+ float16x4_t __s0_125 = __p0_125; \
+ float16x4_t __rev0_125; __rev0_125 = __builtin_shufflevector(__s0_125, __s0_125, 3, 2, 1, 0); \
+ float16x4_t __ret_125; \
+ __ret_125 = __noswap_splat_lane_f16(__rev0_125, __p1_125); \
+ __ret_125 = __builtin_shufflevector(__ret_125, __ret_125, 3, 2, 1, 0); \
+ __ret_125; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_p16(poly16x4_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
+#define vqdmulhq_lane_s32(__p0_126, __p1_126, __p2_126) __extension__ ({ \
+ int32x4_t __s0_126 = __p0_126; \
+ int32x2_t __s1_126 = __p1_126; \
+ int32x4_t __ret_126; \
+ __ret_126 = vqdmulhq_s32(__s0_126, splatq_lane_s32(__s1_126, __p2_126)); \
+ __ret_126; \
+})
#else
-__ai poly8x8_t vreinterpret_p8_p16(poly16x4_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
+#define vqdmulhq_lane_s32(__p0_127, __p1_127, __p2_127) __extension__ ({ \
+ int32x4_t __s0_127 = __p0_127; \
+ int32x2_t __s1_127 = __p1_127; \
+ int32x4_t __rev0_127; __rev0_127 = __builtin_shufflevector(__s0_127, __s0_127, 3, 2, 1, 0); \
+ int32x2_t __rev1_127; __rev1_127 = __builtin_shufflevector(__s1_127, __s1_127, 1, 0); \
+ int32x4_t __ret_127; \
+ __ret_127 = __noswap_vqdmulhq_s32(__rev0_127, __noswap_splatq_lane_s32(__rev1_127, __p2_127)); \
+ __ret_127 = __builtin_shufflevector(__ret_127, __ret_127, 3, 2, 1, 0); \
+ __ret_127; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_u8(uint8x8_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
+#define vqdmulhq_lane_s16(__p0_128, __p1_128, __p2_128) __extension__ ({ \
+ int16x8_t __s0_128 = __p0_128; \
+ int16x4_t __s1_128 = __p1_128; \
+ int16x8_t __ret_128; \
+ __ret_128 = vqdmulhq_s16(__s0_128, splatq_lane_s16(__s1_128, __p2_128)); \
+ __ret_128; \
+})
#else
-__ai poly8x8_t vreinterpret_p8_u8(uint8x8_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
+#define vqdmulhq_lane_s16(__p0_129, __p1_129, __p2_129) __extension__ ({ \
+ int16x8_t __s0_129 = __p0_129; \
+ int16x4_t __s1_129 = __p1_129; \
+ int16x8_t __rev0_129; __rev0_129 = __builtin_shufflevector(__s0_129, __s0_129, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev1_129; __rev1_129 = __builtin_shufflevector(__s1_129, __s1_129, 3, 2, 1, 0); \
+ int16x8_t __ret_129; \
+ __ret_129 = __noswap_vqdmulhq_s16(__rev0_129, __noswap_splatq_lane_s16(__rev1_129, __p2_129)); \
+ __ret_129 = __builtin_shufflevector(__ret_129, __ret_129, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_129; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_u32(uint32x2_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
+#define vqdmulh_lane_s32(__p0_130, __p1_130, __p2_130) __extension__ ({ \
+ int32x2_t __s0_130 = __p0_130; \
+ int32x2_t __s1_130 = __p1_130; \
+ int32x2_t __ret_130; \
+ __ret_130 = vqdmulh_s32(__s0_130, splat_lane_s32(__s1_130, __p2_130)); \
+ __ret_130; \
+})
#else
-__ai poly8x8_t vreinterpret_p8_u32(uint32x2_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
+#define vqdmulh_lane_s32(__p0_131, __p1_131, __p2_131) __extension__ ({ \
+ int32x2_t __s0_131 = __p0_131; \
+ int32x2_t __s1_131 = __p1_131; \
+ int32x2_t __rev0_131; __rev0_131 = __builtin_shufflevector(__s0_131, __s0_131, 1, 0); \
+ int32x2_t __rev1_131; __rev1_131 = __builtin_shufflevector(__s1_131, __s1_131, 1, 0); \
+ int32x2_t __ret_131; \
+ __ret_131 = __noswap_vqdmulh_s32(__rev0_131, __noswap_splat_lane_s32(__rev1_131, __p2_131)); \
+ __ret_131 = __builtin_shufflevector(__ret_131, __ret_131, 1, 0); \
+ __ret_131; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_u64(uint64x1_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
+#define vqdmulh_lane_s16(__p0_132, __p1_132, __p2_132) __extension__ ({ \
+ int16x4_t __s0_132 = __p0_132; \
+ int16x4_t __s1_132 = __p1_132; \
+ int16x4_t __ret_132; \
+ __ret_132 = vqdmulh_s16(__s0_132, splat_lane_s16(__s1_132, __p2_132)); \
+ __ret_132; \
+})
#else
-__ai poly8x8_t vreinterpret_p8_u64(uint64x1_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
+#define vqdmulh_lane_s16(__p0_133, __p1_133, __p2_133) __extension__ ({ \
+ int16x4_t __s0_133 = __p0_133; \
+ int16x4_t __s1_133 = __p1_133; \
+ int16x4_t __rev0_133; __rev0_133 = __builtin_shufflevector(__s0_133, __s0_133, 3, 2, 1, 0); \
+ int16x4_t __rev1_133; __rev1_133 = __builtin_shufflevector(__s1_133, __s1_133, 3, 2, 1, 0); \
+ int16x4_t __ret_133; \
+ __ret_133 = __noswap_vqdmulh_s16(__rev0_133, __noswap_splat_lane_s16(__rev1_133, __p2_133)); \
+ __ret_133 = __builtin_shufflevector(__ret_133, __ret_133, 3, 2, 1, 0); \
+ __ret_133; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_u16(uint16x4_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
+#define vqrdmulhq_lane_s32(__p0_134, __p1_134, __p2_134) __extension__ ({ \
+ int32x4_t __s0_134 = __p0_134; \
+ int32x2_t __s1_134 = __p1_134; \
+ int32x4_t __ret_134; \
+ __ret_134 = vqrdmulhq_s32(__s0_134, splatq_lane_s32(__s1_134, __p2_134)); \
+ __ret_134; \
+})
#else
-__ai poly8x8_t vreinterpret_p8_u16(uint16x4_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
+#define vqrdmulhq_lane_s32(__p0_135, __p1_135, __p2_135) __extension__ ({ \
+ int32x4_t __s0_135 = __p0_135; \
+ int32x2_t __s1_135 = __p1_135; \
+ int32x4_t __rev0_135; __rev0_135 = __builtin_shufflevector(__s0_135, __s0_135, 3, 2, 1, 0); \
+ int32x2_t __rev1_135; __rev1_135 = __builtin_shufflevector(__s1_135, __s1_135, 1, 0); \
+ int32x4_t __ret_135; \
+ __ret_135 = __noswap_vqrdmulhq_s32(__rev0_135, __noswap_splatq_lane_s32(__rev1_135, __p2_135)); \
+ __ret_135 = __builtin_shufflevector(__ret_135, __ret_135, 3, 2, 1, 0); \
+ __ret_135; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_s8(int8x8_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
+#define vqrdmulhq_lane_s16(__p0_136, __p1_136, __p2_136) __extension__ ({ \
+ int16x8_t __s0_136 = __p0_136; \
+ int16x4_t __s1_136 = __p1_136; \
+ int16x8_t __ret_136; \
+ __ret_136 = vqrdmulhq_s16(__s0_136, splatq_lane_s16(__s1_136, __p2_136)); \
+ __ret_136; \
+})
#else
-__ai poly8x8_t vreinterpret_p8_s8(int8x8_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
+#define vqrdmulhq_lane_s16(__p0_137, __p1_137, __p2_137) __extension__ ({ \
+ int16x8_t __s0_137 = __p0_137; \
+ int16x4_t __s1_137 = __p1_137; \
+ int16x8_t __rev0_137; __rev0_137 = __builtin_shufflevector(__s0_137, __s0_137, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev1_137; __rev1_137 = __builtin_shufflevector(__s1_137, __s1_137, 3, 2, 1, 0); \
+ int16x8_t __ret_137; \
+ __ret_137 = __noswap_vqrdmulhq_s16(__rev0_137, __noswap_splatq_lane_s16(__rev1_137, __p2_137)); \
+ __ret_137 = __builtin_shufflevector(__ret_137, __ret_137, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_137; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_f32(float32x2_t __p0) {
+#define vqrdmulh_lane_s32(__p0_138, __p1_138, __p2_138) __extension__ ({ \
+ int32x2_t __s0_138 = __p0_138; \
+ int32x2_t __s1_138 = __p1_138; \
+ int32x2_t __ret_138; \
+ __ret_138 = vqrdmulh_s32(__s0_138, splat_lane_s32(__s1_138, __p2_138)); \
+ __ret_138; \
+})
+#else
+#define vqrdmulh_lane_s32(__p0_139, __p1_139, __p2_139) __extension__ ({ \
+ int32x2_t __s0_139 = __p0_139; \
+ int32x2_t __s1_139 = __p1_139; \
+ int32x2_t __rev0_139; __rev0_139 = __builtin_shufflevector(__s0_139, __s0_139, 1, 0); \
+ int32x2_t __rev1_139; __rev1_139 = __builtin_shufflevector(__s1_139, __s1_139, 1, 0); \
+ int32x2_t __ret_139; \
+ __ret_139 = __noswap_vqrdmulh_s32(__rev0_139, __noswap_splat_lane_s32(__rev1_139, __p2_139)); \
+ __ret_139 = __builtin_shufflevector(__ret_139, __ret_139, 1, 0); \
+ __ret_139; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vqrdmulh_lane_s16(__p0_140, __p1_140, __p2_140) __extension__ ({ \
+ int16x4_t __s0_140 = __p0_140; \
+ int16x4_t __s1_140 = __p1_140; \
+ int16x4_t __ret_140; \
+ __ret_140 = vqrdmulh_s16(__s0_140, splat_lane_s16(__s1_140, __p2_140)); \
+ __ret_140; \
+})
+#else
+#define vqrdmulh_lane_s16(__p0_141, __p1_141, __p2_141) __extension__ ({ \
+ int16x4_t __s0_141 = __p0_141; \
+ int16x4_t __s1_141 = __p1_141; \
+ int16x4_t __rev0_141; __rev0_141 = __builtin_shufflevector(__s0_141, __s0_141, 3, 2, 1, 0); \
+ int16x4_t __rev1_141; __rev1_141 = __builtin_shufflevector(__s1_141, __s1_141, 3, 2, 1, 0); \
+ int16x4_t __ret_141; \
+ __ret_141 = __noswap_vqrdmulh_s16(__rev0_141, __noswap_splat_lane_s16(__rev1_141, __p2_141)); \
+ __ret_141 = __builtin_shufflevector(__ret_141, __ret_141, 3, 2, 1, 0); \
+ __ret_141; \
+})
+#endif
+
+__ai poly8x8_t vreinterpret_p8_p16(poly16x4_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#else
-__ai poly8x8_t vreinterpret_p8_f32(float32x2_t __p0) {
+__ai poly8x8_t vreinterpret_p8_u8(uint8x8_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_f16(float16x4_t __p0) {
+__ai poly8x8_t vreinterpret_p8_u32(uint32x2_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#else
-__ai poly8x8_t vreinterpret_p8_f16(float16x4_t __p0) {
+__ai poly8x8_t vreinterpret_p8_u64(uint64x1_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_s32(int32x2_t __p0) {
+__ai poly8x8_t vreinterpret_p8_u16(uint16x4_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#else
-__ai poly8x8_t vreinterpret_p8_s32(int32x2_t __p0) {
+__ai poly8x8_t vreinterpret_p8_s8(int8x8_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_s64(int64x1_t __p0) {
+__ai poly8x8_t vreinterpret_p8_f32(float32x2_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#else
-__ai poly8x8_t vreinterpret_p8_s64(int64x1_t __p0) {
+__ai poly8x8_t vreinterpret_p8_f16(float16x4_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_s16(int16x4_t __p0) {
+__ai poly8x8_t vreinterpret_p8_s32(int32x2_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#else
-__ai poly8x8_t vreinterpret_p8_s16(int16x4_t __p0) {
+__ai poly8x8_t vreinterpret_p8_s64(int64x1_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_p8(poly8x8_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
+__ai poly8x8_t vreinterpret_p8_s16(int16x4_t __p0) {
+ poly8x8_t __ret;
+ __ret = (poly8x8_t)(__p0);
return __ret;
}
-#else
__ai poly16x4_t vreinterpret_p16_p8(poly8x8_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly16x4_t vreinterpret_p16_u8(uint8x8_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#else
-__ai poly16x4_t vreinterpret_p16_u8(uint8x8_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly16x4_t vreinterpret_p16_u32(uint32x2_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#else
-__ai poly16x4_t vreinterpret_p16_u32(uint32x2_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_u64(uint64x1_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_u64(uint64x1_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly16x4_t vreinterpret_p16_u16(uint16x4_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#else
-__ai poly16x4_t vreinterpret_p16_u16(uint16x4_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly16x4_t vreinterpret_p16_s8(int8x8_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#else
-__ai poly16x4_t vreinterpret_p16_s8(int8x8_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_f32(float32x2_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_f32(float32x2_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly16x4_t vreinterpret_p16_f16(float16x4_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#else
-__ai poly16x4_t vreinterpret_p16_f16(float16x4_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_s32(int32x2_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_s32(int32x2_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly16x4_t vreinterpret_p16_s64(int64x1_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#else
-__ai poly16x4_t vreinterpret_p16_s64(int64x1_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly16x4_t vreinterpret_p16_s16(int16x4_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#else
-__ai poly16x4_t vreinterpret_p16_s16(int16x4_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_p16(poly16x8_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_p16(poly16x8_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly8x16_t vreinterpretq_p8_u8(uint8x16_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#else
-__ai poly8x16_t vreinterpretq_p8_u8(uint8x16_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly8x16_t vreinterpretq_p8_u32(uint32x4_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#else
-__ai poly8x16_t vreinterpretq_p8_u32(uint32x4_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly8x16_t vreinterpretq_p8_u64(uint64x2_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#else
-__ai poly8x16_t vreinterpretq_p8_u64(uint64x2_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly8x16_t vreinterpretq_p8_u16(uint16x8_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#else
-__ai poly8x16_t vreinterpretq_p8_u16(uint16x8_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly8x16_t vreinterpretq_p8_s8(int8x16_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#else
-__ai poly8x16_t vreinterpretq_p8_s8(int8x16_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly8x16_t vreinterpretq_p8_f32(float32x4_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#else
-__ai poly8x16_t vreinterpretq_p8_f32(float32x4_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_f16(float16x8_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_f16(float16x8_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly8x16_t vreinterpretq_p8_s32(int32x4_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#else
-__ai poly8x16_t vreinterpretq_p8_s32(int32x4_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_s64(int64x2_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_s64(int64x2_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly8x16_t vreinterpretq_p8_s16(int16x8_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#else
-__ai poly8x16_t vreinterpretq_p8_s16(int16x8_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly16x8_t vreinterpretq_p16_p8(poly8x16_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#else
-__ai poly16x8_t vreinterpretq_p16_p8(poly8x16_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_u8(uint8x16_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x8_t vreinterpretq_p16_u8(uint8x16_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly16x8_t vreinterpretq_p16_u32(uint32x4_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#else
-__ai poly16x8_t vreinterpretq_p16_u32(uint32x4_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_u64(uint64x2_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
-__ai poly16x8_t vreinterpretq_p16_u64(uint64x2_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_u16(uint16x8_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
-__ai poly16x8_t vreinterpretq_p16_u16(uint16x8_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_s8(int8x16_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
-__ai poly16x8_t vreinterpretq_p16_s8(int8x16_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_f32(float32x4_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
-__ai poly16x8_t vreinterpretq_p16_f32(float32x4_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_f16(float16x8_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
-__ai poly16x8_t vreinterpretq_p16_f16(float16x8_t __p0) {
+__ai poly16x8_t vreinterpretq_p16_u64(uint64x2_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_s32(int32x4_t __p0) {
+__ai poly16x8_t vreinterpretq_p16_u16(uint16x8_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#else
-__ai poly16x8_t vreinterpretq_p16_s32(int32x4_t __p0) {
+__ai poly16x8_t vreinterpretq_p16_s8(int8x16_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_s64(int64x2_t __p0) {
+__ai poly16x8_t vreinterpretq_p16_f32(float32x4_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#else
-__ai poly16x8_t vreinterpretq_p16_s64(int64x2_t __p0) {
+__ai poly16x8_t vreinterpretq_p16_f16(float16x8_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_s16(int16x8_t __p0) {
+__ai poly16x8_t vreinterpretq_p16_s32(int32x4_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#else
-__ai poly16x8_t vreinterpretq_p16_s16(int16x8_t __p0) {
+__ai poly16x8_t vreinterpretq_p16_s64(int64x2_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_p8(poly8x16_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
+__ai poly16x8_t vreinterpretq_p16_s16(int16x8_t __p0) {
+ poly16x8_t __ret;
+ __ret = (poly16x8_t)(__p0);
return __ret;
}
-#else
__ai uint8x16_t vreinterpretq_u8_p8(poly8x16_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_p16(poly16x8_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_p16(poly16x8_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_u32(uint32x4_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_u32(uint32x4_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_u64(uint64x2_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_u64(uint64x2_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_u16(uint16x8_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_u16(uint16x8_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_s8(int8x16_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_s8(int8x16_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_f32(float32x4_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_f32(float32x4_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_f16(float16x8_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_f16(float16x8_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_s32(int32x4_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_s32(int32x4_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_s64(int64x2_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_s64(int64x2_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_s16(int16x8_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_s16(int16x8_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_p8(poly8x16_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x4_t vreinterpretq_u32_p8(poly8x16_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_p16(poly16x8_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x4_t vreinterpretq_u32_p16(poly16x8_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_u8(uint8x16_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x4_t vreinterpretq_u32_u8(uint8x16_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_u64(uint64x2_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x4_t vreinterpretq_u32_u64(uint64x2_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_u16(uint16x8_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x4_t vreinterpretq_u32_u16(uint16x8_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_s8(int8x16_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x4_t vreinterpretq_u32_s8(int8x16_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_f32(float32x4_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x4_t vreinterpretq_u32_f32(float32x4_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_f16(float16x8_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x4_t vreinterpretq_u32_f16(float16x8_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_s32(int32x4_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x4_t vreinterpretq_u32_s32(int32x4_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_s64(int64x2_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x4_t vreinterpretq_u32_s64(int64x2_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_s16(int16x8_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x4_t vreinterpretq_u32_s16(int16x8_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_p8(poly8x16_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_p8(poly8x16_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_p16(poly16x8_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_p16(poly16x8_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_u8(uint8x16_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_u8(uint8x16_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_u32(uint32x4_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_u32(uint32x4_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_u16(uint16x8_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_u16(uint16x8_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_s8(int8x16_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_s8(int8x16_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_f32(float32x4_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_f32(float32x4_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_f16(float16x8_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_f16(float16x8_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_s32(int32x4_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_s32(int32x4_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_s64(int64x2_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_s64(int64x2_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_s16(int16x8_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_s16(int16x8_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_p8(poly8x16_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x8_t vreinterpretq_u16_p8(poly8x16_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_p16(poly16x8_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x8_t vreinterpretq_u16_p16(poly16x8_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_u8(uint8x16_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x8_t vreinterpretq_u16_u8(uint8x16_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_u32(uint32x4_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x8_t vreinterpretq_u16_u32(uint32x4_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_u64(uint64x2_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x8_t vreinterpretq_u16_u64(uint64x2_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_s8(int8x16_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x8_t vreinterpretq_u16_s8(int8x16_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_f32(float32x4_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x8_t vreinterpretq_u16_f32(float32x4_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_f16(float16x8_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x8_t vreinterpretq_u16_f16(float16x8_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_s32(int32x4_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x8_t vreinterpretq_u16_s32(int32x4_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_s64(int64x2_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x8_t vreinterpretq_u16_s64(int64x2_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_s16(int16x8_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x8_t vreinterpretq_u16_s16(int16x8_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_p8(poly8x16_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
- return __ret;
-}
-#else
-__ai int8x16_t vreinterpretq_s8_p8(poly8x16_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_p16(poly16x8_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
- return __ret;
-}
-#else
-__ai int8x16_t vreinterpretq_s8_p16(poly16x8_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_u8(uint8x16_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
- return __ret;
-}
-#else
-__ai int8x16_t vreinterpretq_s8_u8(uint8x16_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_u32(uint32x4_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
- return __ret;
-}
-#else
-__ai int8x16_t vreinterpretq_s8_u32(uint32x4_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_u64(uint64x2_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
- return __ret;
-}
-#else
-__ai int8x16_t vreinterpretq_s8_u64(uint64x2_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_u16(uint16x8_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
- return __ret;
-}
-#else
-__ai int8x16_t vreinterpretq_s8_u16(uint16x8_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_f32(float32x4_t __p0) {
+__ai int8x16_t vreinterpretq_s8_p8(poly8x16_t __p0) {
int8x16_t __ret;
__ret = (int8x16_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_f32(float32x4_t __p0) {
+__ai int8x16_t vreinterpretq_s8_p16(poly16x8_t __p0) {
int8x16_t __ret;
__ret = (int8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_f16(float16x8_t __p0) {
+__ai int8x16_t vreinterpretq_s8_u8(uint8x16_t __p0) {
int8x16_t __ret;
__ret = (int8x16_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_f16(float16x8_t __p0) {
+__ai int8x16_t vreinterpretq_s8_u32(uint32x4_t __p0) {
int8x16_t __ret;
__ret = (int8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_s32(int32x4_t __p0) {
+__ai int8x16_t vreinterpretq_s8_u64(uint64x2_t __p0) {
int8x16_t __ret;
__ret = (int8x16_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_s32(int32x4_t __p0) {
+__ai int8x16_t vreinterpretq_s8_u16(uint16x8_t __p0) {
int8x16_t __ret;
__ret = (int8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_s64(int64x2_t __p0) {
+__ai int8x16_t vreinterpretq_s8_f32(float32x4_t __p0) {
int8x16_t __ret;
__ret = (int8x16_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_s64(int64x2_t __p0) {
+__ai int8x16_t vreinterpretq_s8_f16(float16x8_t __p0) {
int8x16_t __ret;
__ret = (int8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_s16(int16x8_t __p0) {
+__ai int8x16_t vreinterpretq_s8_s32(int32x4_t __p0) {
int8x16_t __ret;
__ret = (int8x16_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_s16(int16x8_t __p0) {
+__ai int8x16_t vreinterpretq_s8_s64(int64x2_t __p0) {
int8x16_t __ret;
__ret = (int8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_p8(poly8x16_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai int8x16_t vreinterpretq_s8_s16(int16x8_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
return __ret;
}
-#else
__ai float32x4_t vreinterpretq_f32_p8(poly8x16_t __p0) {
float32x4_t __ret;
__ret = (float32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_p16(poly16x8_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
- return __ret;
-}
-#else
__ai float32x4_t vreinterpretq_f32_p16(poly16x8_t __p0) {
float32x4_t __ret;
__ret = (float32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_u8(uint8x16_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
- return __ret;
-}
-#else
__ai float32x4_t vreinterpretq_f32_u8(uint8x16_t __p0) {
float32x4_t __ret;
__ret = (float32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_u32(uint32x4_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
- return __ret;
-}
-#else
__ai float32x4_t vreinterpretq_f32_u32(uint32x4_t __p0) {
float32x4_t __ret;
__ret = (float32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_u64(uint64x2_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
- return __ret;
-}
-#else
__ai float32x4_t vreinterpretq_f32_u64(uint64x2_t __p0) {
float32x4_t __ret;
__ret = (float32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_u16(uint16x8_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
- return __ret;
-}
-#else
__ai float32x4_t vreinterpretq_f32_u16(uint16x8_t __p0) {
float32x4_t __ret;
__ret = (float32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_s8(int8x16_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
- return __ret;
-}
-#else
__ai float32x4_t vreinterpretq_f32_s8(int8x16_t __p0) {
float32x4_t __ret;
__ret = (float32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_f16(float16x8_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
- return __ret;
-}
-#else
__ai float32x4_t vreinterpretq_f32_f16(float16x8_t __p0) {
float32x4_t __ret;
__ret = (float32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_s32(int32x4_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
- return __ret;
-}
-#else
__ai float32x4_t vreinterpretq_f32_s32(int32x4_t __p0) {
float32x4_t __ret;
__ret = (float32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_s64(int64x2_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
- return __ret;
-}
-#else
__ai float32x4_t vreinterpretq_f32_s64(int64x2_t __p0) {
float32x4_t __ret;
__ret = (float32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_s16(int16x8_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
- return __ret;
-}
-#else
__ai float32x4_t vreinterpretq_f32_s16(int16x8_t __p0) {
float32x4_t __ret;
__ret = (float32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_p8(poly8x16_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
- return __ret;
-}
-#else
__ai float16x8_t vreinterpretq_f16_p8(poly8x16_t __p0) {
float16x8_t __ret;
__ret = (float16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_p16(poly16x8_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
- return __ret;
-}
-#else
__ai float16x8_t vreinterpretq_f16_p16(poly16x8_t __p0) {
float16x8_t __ret;
__ret = (float16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_u8(uint8x16_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
- return __ret;
-}
-#else
__ai float16x8_t vreinterpretq_f16_u8(uint8x16_t __p0) {
float16x8_t __ret;
__ret = (float16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_u32(uint32x4_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
- return __ret;
-}
-#else
__ai float16x8_t vreinterpretq_f16_u32(uint32x4_t __p0) {
float16x8_t __ret;
__ret = (float16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_u64(uint64x2_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
- return __ret;
-}
-#else
__ai float16x8_t vreinterpretq_f16_u64(uint64x2_t __p0) {
float16x8_t __ret;
__ret = (float16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_u16(uint16x8_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
- return __ret;
-}
-#else
__ai float16x8_t vreinterpretq_f16_u16(uint16x8_t __p0) {
float16x8_t __ret;
__ret = (float16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_s8(int8x16_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
- return __ret;
-}
-#else
__ai float16x8_t vreinterpretq_f16_s8(int8x16_t __p0) {
float16x8_t __ret;
__ret = (float16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_f32(float32x4_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
- return __ret;
-}
-#else
__ai float16x8_t vreinterpretq_f16_f32(float32x4_t __p0) {
float16x8_t __ret;
__ret = (float16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_s32(int32x4_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
- return __ret;
-}
-#else
__ai float16x8_t vreinterpretq_f16_s32(int32x4_t __p0) {
float16x8_t __ret;
__ret = (float16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_s64(int64x2_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
- return __ret;
-}
-#else
__ai float16x8_t vreinterpretq_f16_s64(int64x2_t __p0) {
float16x8_t __ret;
__ret = (float16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_s16(int16x8_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
- return __ret;
-}
-#else
__ai float16x8_t vreinterpretq_f16_s16(int16x8_t __p0) {
float16x8_t __ret;
__ret = (float16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_p8(poly8x16_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
- return __ret;
-}
-#else
__ai int32x4_t vreinterpretq_s32_p8(poly8x16_t __p0) {
int32x4_t __ret;
__ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_p16(poly16x8_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
- return __ret;
-}
-#else
__ai int32x4_t vreinterpretq_s32_p16(poly16x8_t __p0) {
int32x4_t __ret;
__ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_u8(uint8x16_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
- return __ret;
-}
-#else
__ai int32x4_t vreinterpretq_s32_u8(uint8x16_t __p0) {
int32x4_t __ret;
__ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_u32(uint32x4_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
- return __ret;
-}
-#else
__ai int32x4_t vreinterpretq_s32_u32(uint32x4_t __p0) {
int32x4_t __ret;
__ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_u64(uint64x2_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
- return __ret;
-}
-#else
__ai int32x4_t vreinterpretq_s32_u64(uint64x2_t __p0) {
int32x4_t __ret;
__ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_u16(uint16x8_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
- return __ret;
-}
-#else
__ai int32x4_t vreinterpretq_s32_u16(uint16x8_t __p0) {
int32x4_t __ret;
__ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_s8(int8x16_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
- return __ret;
-}
-#else
__ai int32x4_t vreinterpretq_s32_s8(int8x16_t __p0) {
int32x4_t __ret;
__ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_f32(float32x4_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
- return __ret;
-}
-#else
__ai int32x4_t vreinterpretq_s32_f32(float32x4_t __p0) {
int32x4_t __ret;
__ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_f16(float16x8_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
- return __ret;
-}
-#else
__ai int32x4_t vreinterpretq_s32_f16(float16x8_t __p0) {
int32x4_t __ret;
__ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_s64(int64x2_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
- return __ret;
-}
-#else
__ai int32x4_t vreinterpretq_s32_s64(int64x2_t __p0) {
int32x4_t __ret;
__ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_s16(int16x8_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
- return __ret;
-}
-#else
__ai int32x4_t vreinterpretq_s32_s16(int16x8_t __p0) {
int32x4_t __ret;
__ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_p8(poly8x16_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
- return __ret;
-}
-#else
__ai int64x2_t vreinterpretq_s64_p8(poly8x16_t __p0) {
int64x2_t __ret;
__ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_p16(poly16x8_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
- return __ret;
-}
-#else
__ai int64x2_t vreinterpretq_s64_p16(poly16x8_t __p0) {
int64x2_t __ret;
__ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_u8(uint8x16_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
- return __ret;
-}
-#else
__ai int64x2_t vreinterpretq_s64_u8(uint8x16_t __p0) {
int64x2_t __ret;
__ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_u32(uint32x4_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
- return __ret;
-}
-#else
__ai int64x2_t vreinterpretq_s64_u32(uint32x4_t __p0) {
int64x2_t __ret;
__ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_u64(uint64x2_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
- return __ret;
-}
-#else
__ai int64x2_t vreinterpretq_s64_u64(uint64x2_t __p0) {
int64x2_t __ret;
__ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_u16(uint16x8_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
- return __ret;
-}
-#else
__ai int64x2_t vreinterpretq_s64_u16(uint16x8_t __p0) {
int64x2_t __ret;
__ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_s8(int8x16_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
- return __ret;
-}
-#else
__ai int64x2_t vreinterpretq_s64_s8(int8x16_t __p0) {
int64x2_t __ret;
__ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_f32(float32x4_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
- return __ret;
-}
-#else
__ai int64x2_t vreinterpretq_s64_f32(float32x4_t __p0) {
int64x2_t __ret;
__ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_f16(float16x8_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
- return __ret;
-}
-#else
__ai int64x2_t vreinterpretq_s64_f16(float16x8_t __p0) {
int64x2_t __ret;
__ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_s32(int32x4_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
- return __ret;
-}
-#else
__ai int64x2_t vreinterpretq_s64_s32(int32x4_t __p0) {
int64x2_t __ret;
__ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_s16(int16x8_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
- return __ret;
-}
-#else
__ai int64x2_t vreinterpretq_s64_s16(int16x8_t __p0) {
int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_p8(poly8x16_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
- return __ret;
-}
-#else
-__ai int16x8_t vreinterpretq_s16_p8(poly8x16_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_p16(poly16x8_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
- return __ret;
-}
-#else
-__ai int16x8_t vreinterpretq_s16_p16(poly16x8_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_u8(uint8x16_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
- return __ret;
-}
-#else
-__ai int16x8_t vreinterpretq_s16_u8(uint8x16_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_u32(uint32x4_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
- return __ret;
-}
-#else
-__ai int16x8_t vreinterpretq_s16_u32(uint32x4_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_u64(uint64x2_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
- return __ret;
-}
-#else
-__ai int16x8_t vreinterpretq_s16_u64(uint64x2_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_u16(uint16x8_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#else
-__ai int16x8_t vreinterpretq_s16_u16(uint16x8_t __p0) {
+__ai int16x8_t vreinterpretq_s16_p8(poly8x16_t __p0) {
int16x8_t __ret;
__ret = (int16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_s8(int8x16_t __p0) {
+__ai int16x8_t vreinterpretq_s16_p16(poly16x8_t __p0) {
int16x8_t __ret;
__ret = (int16x8_t)(__p0);
return __ret;
}
-#else
-__ai int16x8_t vreinterpretq_s16_s8(int8x16_t __p0) {
+__ai int16x8_t vreinterpretq_s16_u8(uint8x16_t __p0) {
int16x8_t __ret;
__ret = (int16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_f32(float32x4_t __p0) {
+__ai int16x8_t vreinterpretq_s16_u32(uint32x4_t __p0) {
int16x8_t __ret;
__ret = (int16x8_t)(__p0);
return __ret;
}
-#else
-__ai int16x8_t vreinterpretq_s16_f32(float32x4_t __p0) {
+__ai int16x8_t vreinterpretq_s16_u64(uint64x2_t __p0) {
int16x8_t __ret;
__ret = (int16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_f16(float16x8_t __p0) {
+__ai int16x8_t vreinterpretq_s16_u16(uint16x8_t __p0) {
int16x8_t __ret;
__ret = (int16x8_t)(__p0);
return __ret;
}
-#else
-__ai int16x8_t vreinterpretq_s16_f16(float16x8_t __p0) {
+__ai int16x8_t vreinterpretq_s16_s8(int8x16_t __p0) {
int16x8_t __ret;
__ret = (int16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_s32(int32x4_t __p0) {
+__ai int16x8_t vreinterpretq_s16_f32(float32x4_t __p0) {
int16x8_t __ret;
__ret = (int16x8_t)(__p0);
return __ret;
}
-#else
-__ai int16x8_t vreinterpretq_s16_s32(int32x4_t __p0) {
+__ai int16x8_t vreinterpretq_s16_f16(float16x8_t __p0) {
int16x8_t __ret;
__ret = (int16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_s64(int64x2_t __p0) {
+__ai int16x8_t vreinterpretq_s16_s32(int32x4_t __p0) {
int16x8_t __ret;
__ret = (int16x8_t)(__p0);
return __ret;
}
-#else
__ai int16x8_t vreinterpretq_s16_s64(int64x2_t __p0) {
int16x8_t __ret;
__ret = (int16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_p8(poly8x8_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x8_t vreinterpret_u8_p8(poly8x8_t __p0) {
uint8x8_t __ret;
__ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_p16(poly16x4_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x8_t vreinterpret_u8_p16(poly16x4_t __p0) {
uint8x8_t __ret;
__ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_u32(uint32x2_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x8_t vreinterpret_u8_u32(uint32x2_t __p0) {
uint8x8_t __ret;
__ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_u64(uint64x1_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x8_t vreinterpret_u8_u64(uint64x1_t __p0) {
uint8x8_t __ret;
__ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_u16(uint16x4_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x8_t vreinterpret_u8_u16(uint16x4_t __p0) {
uint8x8_t __ret;
__ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_s8(int8x8_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x8_t vreinterpret_u8_s8(int8x8_t __p0) {
uint8x8_t __ret;
__ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_f32(float32x2_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x8_t vreinterpret_u8_f32(float32x2_t __p0) {
uint8x8_t __ret;
__ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_f16(float16x4_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x8_t vreinterpret_u8_f16(float16x4_t __p0) {
uint8x8_t __ret;
__ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_s32(int32x2_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x8_t vreinterpret_u8_s32(int32x2_t __p0) {
uint8x8_t __ret;
__ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_s64(int64x1_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x8_t vreinterpret_u8_s64(int64x1_t __p0) {
uint8x8_t __ret;
__ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_s16(int16x4_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x8_t vreinterpret_u8_s16(int16x4_t __p0) {
uint8x8_t __ret;
__ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_p8(poly8x8_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x2_t vreinterpret_u32_p8(poly8x8_t __p0) {
uint32x2_t __ret;
__ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_p16(poly16x4_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x2_t vreinterpret_u32_p16(poly16x4_t __p0) {
uint32x2_t __ret;
__ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_u8(uint8x8_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x2_t vreinterpret_u32_u8(uint8x8_t __p0) {
uint32x2_t __ret;
__ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_u64(uint64x1_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x2_t vreinterpret_u32_u64(uint64x1_t __p0) {
uint32x2_t __ret;
__ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_u16(uint16x4_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x2_t vreinterpret_u32_u16(uint16x4_t __p0) {
uint32x2_t __ret;
__ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_s8(int8x8_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x2_t vreinterpret_u32_s8(int8x8_t __p0) {
uint32x2_t __ret;
__ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_f32(float32x2_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x2_t vreinterpret_u32_f32(float32x2_t __p0) {
uint32x2_t __ret;
__ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_f16(float16x4_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x2_t vreinterpret_u32_f16(float16x4_t __p0) {
uint32x2_t __ret;
__ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_s32(int32x2_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x2_t vreinterpret_u32_s32(int32x2_t __p0) {
uint32x2_t __ret;
__ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_s64(int64x1_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x2_t vreinterpret_u32_s64(int64x1_t __p0) {
uint32x2_t __ret;
__ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_s16(int16x4_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x2_t vreinterpret_u32_s16(int16x4_t __p0) {
uint32x2_t __ret;
__ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_p8(poly8x8_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x1_t vreinterpret_u64_p8(poly8x8_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_p16(poly16x4_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x1_t vreinterpret_u64_p16(poly16x4_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_u8(uint8x8_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x1_t vreinterpret_u64_u8(uint8x8_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_u32(uint32x2_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x1_t vreinterpret_u64_u32(uint32x2_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_u16(uint16x4_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x1_t vreinterpret_u64_u16(uint16x4_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_s8(int8x8_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x1_t vreinterpret_u64_s8(int8x8_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_f32(float32x2_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x1_t vreinterpret_u64_f32(float32x2_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_f16(float16x4_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x1_t vreinterpret_u64_f16(float16x4_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_s32(int32x2_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x1_t vreinterpret_u64_s32(int32x2_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_s64(int64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x1_t vreinterpret_u64_s64(int64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_s16(int16x4_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x1_t vreinterpret_u64_s16(int16x4_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_p8(poly8x8_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x4_t vreinterpret_u16_p8(poly8x8_t __p0) {
uint16x4_t __ret;
__ret = (uint16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_p16(poly16x4_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x4_t vreinterpret_u16_p16(poly16x4_t __p0) {
uint16x4_t __ret;
__ret = (uint16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_u8(uint8x8_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x4_t vreinterpret_u16_u8(uint8x8_t __p0) {
uint16x4_t __ret;
__ret = (uint16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_u32(uint32x2_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x4_t vreinterpret_u16_u32(uint32x2_t __p0) {
uint16x4_t __ret;
__ret = (uint16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_u64(uint64x1_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x4_t vreinterpret_u16_u64(uint64x1_t __p0) {
uint16x4_t __ret;
__ret = (uint16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_s8(int8x8_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint16x4_t vreinterpret_u16_s8(int8x8_t __p0) {
uint16x4_t __ret;
__ret = (uint16x4_t)(__p0);
return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_f32(float32x2_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#else
-__ai uint16x4_t vreinterpret_u16_f32(float32x2_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_f16(float16x4_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#else
-__ai uint16x4_t vreinterpret_u16_f16(float16x4_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_s32(int32x2_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#else
-__ai uint16x4_t vreinterpret_u16_s32(int32x2_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_s64(int64x1_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#else
-__ai uint16x4_t vreinterpret_u16_s64(int64x1_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_s16(int16x4_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#else
-__ai uint16x4_t vreinterpret_u16_s16(int16x4_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_p8(poly8x8_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
-#else
-__ai int8x8_t vreinterpret_s8_p8(poly8x8_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_p16(poly16x4_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
-#else
-__ai int8x8_t vreinterpret_s8_p16(poly16x4_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_u8(uint8x8_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
-#else
-__ai int8x8_t vreinterpret_s8_u8(uint8x8_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_u32(uint32x2_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+}
+__ai uint16x4_t vreinterpret_u16_f32(float32x2_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#else
-__ai int8x8_t vreinterpret_s8_u32(uint32x2_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_f16(float16x4_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_u64(uint64x1_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_s32(int32x2_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#else
-__ai int8x8_t vreinterpret_s8_u64(uint64x1_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_s64(int64x1_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_u16(uint16x4_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_s16(int16x4_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#else
-__ai int8x8_t vreinterpret_s8_u16(uint16x4_t __p0) {
+__ai int8x8_t vreinterpret_s8_p8(poly8x8_t __p0) {
int8x8_t __ret;
__ret = (int8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_f32(float32x2_t __p0) {
+__ai int8x8_t vreinterpret_s8_p16(poly16x4_t __p0) {
int8x8_t __ret;
__ret = (int8x8_t)(__p0);
return __ret;
}
-#else
-__ai int8x8_t vreinterpret_s8_f32(float32x2_t __p0) {
+__ai int8x8_t vreinterpret_s8_u8(uint8x8_t __p0) {
int8x8_t __ret;
__ret = (int8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_f16(float16x4_t __p0) {
+__ai int8x8_t vreinterpret_s8_u32(uint32x2_t __p0) {
int8x8_t __ret;
__ret = (int8x8_t)(__p0);
return __ret;
}
-#else
-__ai int8x8_t vreinterpret_s8_f16(float16x4_t __p0) {
+__ai int8x8_t vreinterpret_s8_u64(uint64x1_t __p0) {
int8x8_t __ret;
__ret = (int8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_s32(int32x2_t __p0) {
+__ai int8x8_t vreinterpret_s8_u16(uint16x4_t __p0) {
int8x8_t __ret;
__ret = (int8x8_t)(__p0);
return __ret;
}
-#else
-__ai int8x8_t vreinterpret_s8_s32(int32x2_t __p0) {
+__ai int8x8_t vreinterpret_s8_f32(float32x2_t __p0) {
int8x8_t __ret;
__ret = (int8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_s64(int64x1_t __p0) {
+__ai int8x8_t vreinterpret_s8_f16(float16x4_t __p0) {
int8x8_t __ret;
__ret = (int8x8_t)(__p0);
return __ret;
}
-#else
-__ai int8x8_t vreinterpret_s8_s64(int64x1_t __p0) {
+__ai int8x8_t vreinterpret_s8_s32(int32x2_t __p0) {
int8x8_t __ret;
__ret = (int8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_s16(int16x4_t __p0) {
+__ai int8x8_t vreinterpret_s8_s64(int64x1_t __p0) {
int8x8_t __ret;
__ret = (int8x8_t)(__p0);
return __ret;
}
-#else
__ai int8x8_t vreinterpret_s8_s16(int16x4_t __p0) {
int8x8_t __ret;
__ret = (int8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai float32x2_t vreinterpret_f32_p8(poly8x8_t __p0) {
float32x2_t __ret;
__ret = (float32x2_t)(__p0);
return __ret;
}
-#else
-__ai float32x2_t vreinterpret_f32_p8(poly8x8_t __p0) {
+__ai float32x2_t vreinterpret_f32_p16(poly16x4_t __p0) {
float32x2_t __ret;
__ret = (float32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_p16(poly16x4_t __p0) {
+__ai float32x2_t vreinterpret_f32_u8(uint8x8_t __p0) {
float32x2_t __ret;
__ret = (float32x2_t)(__p0);
return __ret;
}
-#else
-__ai float32x2_t vreinterpret_f32_p16(poly16x4_t __p0) {
+__ai float32x2_t vreinterpret_f32_u32(uint32x2_t __p0) {
float32x2_t __ret;
__ret = (float32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_u8(uint8x8_t __p0) {
+__ai float32x2_t vreinterpret_f32_u64(uint64x1_t __p0) {
float32x2_t __ret;
__ret = (float32x2_t)(__p0);
return __ret;
}
-#else
-__ai float32x2_t vreinterpret_f32_u8(uint8x8_t __p0) {
+__ai float32x2_t vreinterpret_f32_u16(uint16x4_t __p0) {
float32x2_t __ret;
__ret = (float32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_u32(uint32x2_t __p0) {
+__ai float32x2_t vreinterpret_f32_s8(int8x8_t __p0) {
float32x2_t __ret;
__ret = (float32x2_t)(__p0);
return __ret;
}
-#else
-__ai float32x2_t vreinterpret_f32_u32(uint32x2_t __p0) {
+__ai float32x2_t vreinterpret_f32_f16(float16x4_t __p0) {
float32x2_t __ret;
__ret = (float32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_u64(uint64x1_t __p0) {
+__ai float32x2_t vreinterpret_f32_s32(int32x2_t __p0) {
float32x2_t __ret;
__ret = (float32x2_t)(__p0);
return __ret;
}
-#else
-__ai float32x2_t vreinterpret_f32_u64(uint64x1_t __p0) {
+__ai float32x2_t vreinterpret_f32_s64(int64x1_t __p0) {
float32x2_t __ret;
__ret = (float32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_u16(uint16x4_t __p0) {
+__ai float32x2_t vreinterpret_f32_s16(int16x4_t __p0) {
float32x2_t __ret;
__ret = (float32x2_t)(__p0);
return __ret;
}
-#else
-__ai float32x2_t vreinterpret_f32_u16(uint16x4_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_p8(poly8x8_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_s8(int8x8_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_p16(poly16x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#else
-__ai float32x2_t vreinterpret_f32_s8(int8x8_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_u8(uint8x8_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_f16(float16x4_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_u32(uint32x2_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#else
-__ai float32x2_t vreinterpret_f32_f16(float16x4_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_u64(uint64x1_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_s32(int32x2_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_u16(uint16x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#else
-__ai float32x2_t vreinterpret_f32_s32(int32x2_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_s8(int8x8_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_s64(int64x1_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_f32(float32x2_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#else
-__ai float32x2_t vreinterpret_f32_s64(int64x1_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_s32(int32x2_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
+ return __ret;
+}
+__ai float16x4_t vreinterpret_f16_s64(int64x1_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
+ return __ret;
+}
+__ai float16x4_t vreinterpret_f16_s16(int16x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
+ return __ret;
+}
+__ai int32x2_t vreinterpret_s32_p8(poly8x8_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
+ return __ret;
+}
+__ai int32x2_t vreinterpret_s32_p16(poly16x4_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
+ return __ret;
+}
+__ai int32x2_t vreinterpret_s32_u8(uint8x8_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
+ return __ret;
+}
+__ai int32x2_t vreinterpret_s32_u32(uint32x2_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
+ return __ret;
+}
+__ai int32x2_t vreinterpret_s32_u64(uint64x1_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
+ return __ret;
+}
+__ai int32x2_t vreinterpret_s32_u16(uint16x4_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
+ return __ret;
+}
+__ai int32x2_t vreinterpret_s32_s8(int8x8_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
+ return __ret;
+}
+__ai int32x2_t vreinterpret_s32_f32(float32x2_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
+ return __ret;
+}
+__ai int32x2_t vreinterpret_s32_f16(float16x4_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
+ return __ret;
+}
+__ai int32x2_t vreinterpret_s32_s64(int64x1_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
+ return __ret;
+}
+__ai int32x2_t vreinterpret_s32_s16(int16x4_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
+ return __ret;
+}
+__ai int64x1_t vreinterpret_s64_p8(poly8x8_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
+ return __ret;
+}
+__ai int64x1_t vreinterpret_s64_p16(poly16x4_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
+ return __ret;
+}
+__ai int64x1_t vreinterpret_s64_u8(uint8x8_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
+ return __ret;
+}
+__ai int64x1_t vreinterpret_s64_u32(uint32x2_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
+ return __ret;
+}
+__ai int64x1_t vreinterpret_s64_u64(uint64x1_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
+ return __ret;
+}
+__ai int64x1_t vreinterpret_s64_u16(uint16x4_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
+ return __ret;
+}
+__ai int64x1_t vreinterpret_s64_s8(int8x8_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
+ return __ret;
+}
+__ai int64x1_t vreinterpret_s64_f32(float32x2_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
+ return __ret;
+}
+__ai int64x1_t vreinterpret_s64_f16(float16x4_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
+ return __ret;
+}
+__ai int64x1_t vreinterpret_s64_s32(int32x2_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
+ return __ret;
+}
+__ai int64x1_t vreinterpret_s64_s16(int16x4_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
+ return __ret;
+}
+__ai int16x4_t vreinterpret_s16_p8(poly8x8_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
+ return __ret;
+}
+__ai int16x4_t vreinterpret_s16_p16(poly16x4_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
+ return __ret;
+}
+__ai int16x4_t vreinterpret_s16_u8(uint8x8_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
+ return __ret;
+}
+__ai int16x4_t vreinterpret_s16_u32(uint32x2_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
+ return __ret;
+}
+__ai int16x4_t vreinterpret_s16_u64(uint64x1_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
+ return __ret;
+}
+__ai int16x4_t vreinterpret_s16_u16(uint16x4_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
+ return __ret;
+}
+__ai int16x4_t vreinterpret_s16_s8(int8x8_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
+ return __ret;
+}
+__ai int16x4_t vreinterpret_s16_f32(float32x2_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
+ return __ret;
+}
+__ai int16x4_t vreinterpret_s16_f16(float16x4_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
+ return __ret;
+}
+__ai int16x4_t vreinterpret_s16_s32(int32x2_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
+ return __ret;
+}
+__ai int16x4_t vreinterpret_s16_s64(int64x1_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
#endif
-
+#if (__ARM_FP & 2)
#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_s16(int16x4_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai float16x4_t vcvt_f16_f32(float32x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vcvt_f16_f32((int8x16_t)__p0, 41);
return __ret;
}
#else
-__ai float32x2_t vreinterpret_f32_s16(int16x4_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai float16x4_t vcvt_f16_f32(float32x4_t __p0) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vcvt_f16_f32((int8x16_t)__rev0, 41);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai float16x4_t __noswap_vcvt_f16_f32(float32x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vcvt_f16_f32((int8x16_t)__p0, 41);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_p8(poly8x8_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
+__ai float32x4_t vcvt_f32_f16(float16x4_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vcvt_f32_f16((int8x8_t)__p0, 8);
return __ret;
}
#else
-__ai float16x4_t vreinterpret_f16_p8(poly8x8_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
+__ai float32x4_t vcvt_f32_f16(float16x4_t __p0) {
+ float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vcvt_f32_f16((int8x8_t)__rev0, 8);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai float32x4_t __noswap_vcvt_f32_f16(float16x4_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vcvt_f32_f16((int8x8_t)__p0, 8);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_p16(poly16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1q_f16(__p0) __extension__ ({ \
+ float16x8_t __ret; \
+ __ret = (float16x8_t) __builtin_neon_vld1q_v(__p0, 40); \
+ __ret; \
+})
+#else
+#define vld1q_f16(__p0) __extension__ ({ \
+ float16x8_t __ret; \
+ __ret = (float16x8_t) __builtin_neon_vld1q_v(__p0, 40); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld1_f16(__p0) __extension__ ({ \
+ float16x4_t __ret; \
+ __ret = (float16x4_t) __builtin_neon_vld1_v(__p0, 8); \
+ __ret; \
+})
+#else
+#define vld1_f16(__p0) __extension__ ({ \
+ float16x4_t __ret; \
+ __ret = (float16x4_t) __builtin_neon_vld1_v(__p0, 8); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld1q_dup_f16(__p0) __extension__ ({ \
+ float16x8_t __ret; \
+ __ret = (float16x8_t) __builtin_neon_vld1q_dup_v(__p0, 40); \
+ __ret; \
+})
+#else
+#define vld1q_dup_f16(__p0) __extension__ ({ \
+ float16x8_t __ret; \
+ __ret = (float16x8_t) __builtin_neon_vld1q_dup_v(__p0, 40); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld1_dup_f16(__p0) __extension__ ({ \
+ float16x4_t __ret; \
+ __ret = (float16x4_t) __builtin_neon_vld1_dup_v(__p0, 8); \
+ __ret; \
+})
+#else
+#define vld1_dup_f16(__p0) __extension__ ({ \
+ float16x4_t __ret; \
+ __ret = (float16x4_t) __builtin_neon_vld1_dup_v(__p0, 8); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8_t __s1 = __p1; \
+ float16x8_t __ret; \
+ __ret = (float16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 40); \
+ __ret; \
+})
+#else
+#define vld1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8_t __s1 = __p1; \
+ float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __ret; \
+ __ret = (float16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 40); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4_t __s1 = __p1; \
+ float16x4_t __ret; \
+ __ret = (float16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 8); \
+ __ret; \
+})
+#else
+#define vld1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4_t __s1 = __p1; \
+ float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
+ float16x4_t __ret; \
+ __ret = (float16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 8); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld1q_f16_x2(__p0) __extension__ ({ \
+ float16x8x2_t __ret; \
+ __builtin_neon_vld1q_x2_v(&__ret, __p0, 40); \
+ __ret; \
+})
+#else
+#define vld1q_f16_x2(__p0) __extension__ ({ \
+ float16x8x2_t __ret; \
+ __builtin_neon_vld1q_x2_v(&__ret, __p0, 40); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vld1_f16_x2(__p0) __extension__ ({ \
+ float16x4x2_t __ret; \
+ __builtin_neon_vld1_x2_v(&__ret, __p0, 8); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_p16(poly16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1_f16_x2(__p0) __extension__ ({ \
+ float16x4x2_t __ret; \
+ __builtin_neon_vld1_x2_v(&__ret, __p0, 8); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_u8(uint8x8_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1q_f16_x3(__p0) __extension__ ({ \
+ float16x8x3_t __ret; \
+ __builtin_neon_vld1q_x3_v(&__ret, __p0, 40); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_u8(uint8x8_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1q_f16_x3(__p0) __extension__ ({ \
+ float16x8x3_t __ret; \
+ __builtin_neon_vld1q_x3_v(&__ret, __p0, 40); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_u32(uint32x2_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1_f16_x3(__p0) __extension__ ({ \
+ float16x4x3_t __ret; \
+ __builtin_neon_vld1_x3_v(&__ret, __p0, 8); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_u32(uint32x2_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1_f16_x3(__p0) __extension__ ({ \
+ float16x4x3_t __ret; \
+ __builtin_neon_vld1_x3_v(&__ret, __p0, 8); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_u64(uint64x1_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1q_f16_x4(__p0) __extension__ ({ \
+ float16x8x4_t __ret; \
+ __builtin_neon_vld1q_x4_v(&__ret, __p0, 40); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_u64(uint64x1_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1q_f16_x4(__p0) __extension__ ({ \
+ float16x8x4_t __ret; \
+ __builtin_neon_vld1q_x4_v(&__ret, __p0, 40); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_u16(uint16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1_f16_x4(__p0) __extension__ ({ \
+ float16x4x4_t __ret; \
+ __builtin_neon_vld1_x4_v(&__ret, __p0, 8); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_u16(uint16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1_f16_x4(__p0) __extension__ ({ \
+ float16x4x4_t __ret; \
+ __builtin_neon_vld1_x4_v(&__ret, __p0, 8); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_s8(int8x8_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2q_f16(__p0) __extension__ ({ \
+ float16x8x2_t __ret; \
+ __builtin_neon_vld2q_v(&__ret, __p0, 40); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_s8(int8x8_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2q_f16(__p0) __extension__ ({ \
+ float16x8x2_t __ret; \
+ __builtin_neon_vld2q_v(&__ret, __p0, 40); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_f32(float32x2_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2_f16(__p0) __extension__ ({ \
+ float16x4x2_t __ret; \
+ __builtin_neon_vld2_v(&__ret, __p0, 8); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_f32(float32x2_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2_f16(__p0) __extension__ ({ \
+ float16x4x2_t __ret; \
+ __builtin_neon_vld2_v(&__ret, __p0, 8); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_s32(int32x2_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2q_dup_f16(__p0) __extension__ ({ \
+ float16x8x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 40); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_s32(int32x2_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2q_dup_f16(__p0) __extension__ ({ \
+ float16x8x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 40); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_s64(int64x1_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2_dup_f16(__p0) __extension__ ({ \
+ float16x4x2_t __ret; \
+ __builtin_neon_vld2_dup_v(&__ret, __p0, 8); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_s64(int64x1_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2_dup_f16(__p0) __extension__ ({ \
+ float16x4x2_t __ret; \
+ __builtin_neon_vld2_dup_v(&__ret, __p0, 8); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_s16(int16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8x2_t __s1 = __p1; \
+ float16x8x2_t __ret; \
+ __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 40); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_s16(int16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8x2_t __s1 = __p1; \
+ float16x8x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8x2_t __ret; \
+ __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 40); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_p8(poly8x8_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4x2_t __s1 = __p1; \
+ float16x4x2_t __ret; \
+ __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 8); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_p8(poly8x8_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4x2_t __s1 = __p1; \
+ float16x4x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ float16x4x2_t __ret; \
+ __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 8); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_p16(poly16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3q_f16(__p0) __extension__ ({ \
+ float16x8x3_t __ret; \
+ __builtin_neon_vld3q_v(&__ret, __p0, 40); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_p16(poly16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3q_f16(__p0) __extension__ ({ \
+ float16x8x3_t __ret; \
+ __builtin_neon_vld3q_v(&__ret, __p0, 40); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_u8(uint8x8_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3_f16(__p0) __extension__ ({ \
+ float16x4x3_t __ret; \
+ __builtin_neon_vld3_v(&__ret, __p0, 8); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_u8(uint8x8_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3_f16(__p0) __extension__ ({ \
+ float16x4x3_t __ret; \
+ __builtin_neon_vld3_v(&__ret, __p0, 8); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_u32(uint32x2_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3q_dup_f16(__p0) __extension__ ({ \
+ float16x8x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 40); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_u32(uint32x2_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3q_dup_f16(__p0) __extension__ ({ \
+ float16x8x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 40); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_u64(uint64x1_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3_dup_f16(__p0) __extension__ ({ \
+ float16x4x3_t __ret; \
+ __builtin_neon_vld3_dup_v(&__ret, __p0, 8); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_u64(uint64x1_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3_dup_f16(__p0) __extension__ ({ \
+ float16x4x3_t __ret; \
+ __builtin_neon_vld3_dup_v(&__ret, __p0, 8); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_u16(uint16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8x3_t __s1 = __p1; \
+ float16x8x3_t __ret; \
+ __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 40); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_u16(uint16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8x3_t __s1 = __p1; \
+ float16x8x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8x3_t __ret; \
+ __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 40); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_s8(int8x8_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4x3_t __s1 = __p1; \
+ float16x4x3_t __ret; \
+ __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 8); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_s8(int8x8_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4x3_t __s1 = __p1; \
+ float16x4x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ float16x4x3_t __ret; \
+ __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 8); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_f32(float32x2_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4q_f16(__p0) __extension__ ({ \
+ float16x8x4_t __ret; \
+ __builtin_neon_vld4q_v(&__ret, __p0, 40); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_f32(float32x2_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4q_f16(__p0) __extension__ ({ \
+ float16x8x4_t __ret; \
+ __builtin_neon_vld4q_v(&__ret, __p0, 40); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_f16(float16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4_f16(__p0) __extension__ ({ \
+ float16x4x4_t __ret; \
+ __builtin_neon_vld4_v(&__ret, __p0, 8); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_f16(float16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4_f16(__p0) __extension__ ({ \
+ float16x4x4_t __ret; \
+ __builtin_neon_vld4_v(&__ret, __p0, 8); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_s64(int64x1_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4q_dup_f16(__p0) __extension__ ({ \
+ float16x8x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 40); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_s64(int64x1_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4q_dup_f16(__p0) __extension__ ({ \
+ float16x8x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 40); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_s16(int16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4_dup_f16(__p0) __extension__ ({ \
+ float16x4x4_t __ret; \
+ __builtin_neon_vld4_dup_v(&__ret, __p0, 8); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_s16(int16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4_dup_f16(__p0) __extension__ ({ \
+ float16x4x4_t __ret; \
+ __builtin_neon_vld4_dup_v(&__ret, __p0, 8); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_p8(poly8x8_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vld4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8x4_t __s1 = __p1; \
+ float16x8x4_t __ret; \
+ __builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 40); \
+ __ret; \
+})
#else
-__ai int64x1_t vreinterpret_s64_p8(poly8x8_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vld4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8x4_t __s1 = __p1; \
+ float16x8x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8x4_t __ret; \
+ __builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 40); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_p16(poly16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vld4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4x4_t __s1 = __p1; \
+ float16x4x4_t __ret; \
+ __builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 8); \
+ __ret; \
+})
#else
-__ai int64x1_t vreinterpret_s64_p16(poly16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vld4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4x4_t __s1 = __p1; \
+ float16x4x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
+ float16x4x4_t __ret; \
+ __builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 8); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_u8(uint8x8_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_f16(__p0, __p1) __extension__ ({ \
+ float16x8_t __s1 = __p1; \
+ __builtin_neon_vst1q_v(__p0, (int8x16_t)__s1, 40); \
+})
#else
-__ai int64x1_t vreinterpret_s64_u8(uint8x8_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_f16(__p0, __p1) __extension__ ({ \
+ float16x8_t __s1 = __p1; \
+ float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst1q_v(__p0, (int8x16_t)__rev1, 40); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_u32(uint32x2_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_f16(__p0, __p1) __extension__ ({ \
+ float16x4_t __s1 = __p1; \
+ __builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 8); \
+})
#else
-__ai int64x1_t vreinterpret_s64_u32(uint32x2_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_f16(__p0, __p1) __extension__ ({ \
+ float16x4_t __s1 = __p1; \
+ float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
+ __builtin_neon_vst1_v(__p0, (int8x8_t)__rev1, 8); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_u64(uint64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8_t __s1 = __p1; \
+ __builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__s1, __p2, 40); \
+})
#else
-__ai int64x1_t vreinterpret_s64_u64(uint64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8_t __s1 = __p1; \
+ float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 40); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_u16(uint16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4_t __s1 = __p1; \
+ __builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 8); \
+})
#else
-__ai int64x1_t vreinterpret_s64_u16(uint16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4_t __s1 = __p1; \
+ float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
+ __builtin_neon_vst1_lane_v(__p0, (int8x8_t)__rev1, __p2, 8); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_s8(int8x8_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_f16_x2(__p0, __p1) __extension__ ({ \
+ float16x8x2_t __s1 = __p1; \
+ __builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 40); \
+})
#else
-__ai int64x1_t vreinterpret_s64_s8(int8x8_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_f16_x2(__p0, __p1) __extension__ ({ \
+ float16x8x2_t __s1 = __p1; \
+ float16x8x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 40); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_f32(float32x2_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_f16_x2(__p0, __p1) __extension__ ({ \
+ float16x4x2_t __s1 = __p1; \
+ __builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 8); \
+})
#else
-__ai int64x1_t vreinterpret_s64_f32(float32x2_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_f16_x2(__p0, __p1) __extension__ ({ \
+ float16x4x2_t __s1 = __p1; \
+ float16x4x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __builtin_neon_vst1_x2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 8); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_f16(float16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_f16_x3(__p0, __p1) __extension__ ({ \
+ float16x8x3_t __s1 = __p1; \
+ __builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 40); \
+})
#else
-__ai int64x1_t vreinterpret_s64_f16(float16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_f16_x3(__p0, __p1) __extension__ ({ \
+ float16x8x3_t __s1 = __p1; \
+ float16x8x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 40); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_s32(int32x2_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_f16_x3(__p0, __p1) __extension__ ({ \
+ float16x4x3_t __s1 = __p1; \
+ __builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 8); \
+})
#else
-__ai int64x1_t vreinterpret_s64_s32(int32x2_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_f16_x3(__p0, __p1) __extension__ ({ \
+ float16x4x3_t __s1 = __p1; \
+ float16x4x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ __builtin_neon_vst1_x3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 8); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_s16(int16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_f16_x4(__p0, __p1) __extension__ ({ \
+ float16x8x4_t __s1 = __p1; \
+ __builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 40); \
+})
#else
-__ai int64x1_t vreinterpret_s64_s16(int16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_f16_x4(__p0, __p1) __extension__ ({ \
+ float16x8x4_t __s1 = __p1; \
+ float16x8x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 40); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_p8(poly8x8_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst1_f16_x4(__p0, __p1) __extension__ ({ \
+ float16x4x4_t __s1 = __p1; \
+ __builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 8); \
+})
#else
-__ai int16x4_t vreinterpret_s16_p8(poly8x8_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst1_f16_x4(__p0, __p1) __extension__ ({ \
+ float16x4x4_t __s1 = __p1; \
+ float16x4x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
+ __builtin_neon_vst1_x4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 8); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_p16(poly16x4_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst2q_f16(__p0, __p1) __extension__ ({ \
+ float16x8x2_t __s1 = __p1; \
+ __builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 40); \
+})
#else
-__ai int16x4_t vreinterpret_s16_p16(poly16x4_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst2q_f16(__p0, __p1) __extension__ ({ \
+ float16x8x2_t __s1 = __p1; \
+ float16x8x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 40); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_u8(uint8x8_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst2_f16(__p0, __p1) __extension__ ({ \
+ float16x4x2_t __s1 = __p1; \
+ __builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 8); \
+})
#else
-__ai int16x4_t vreinterpret_s16_u8(uint8x8_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst2_f16(__p0, __p1) __extension__ ({ \
+ float16x4x2_t __s1 = __p1; \
+ float16x4x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __builtin_neon_vst2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 8); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_u32(uint32x2_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8x2_t __s1 = __p1; \
+ __builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 40); \
+})
#else
-__ai int16x4_t vreinterpret_s16_u32(uint32x2_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8x2_t __s1 = __p1; \
+ float16x8x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 40); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_u64(uint64x1_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4x2_t __s1 = __p1; \
+ __builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 8); \
+})
#else
-__ai int16x4_t vreinterpret_s16_u64(uint64x1_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4x2_t __s1 = __p1; \
+ float16x4x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __builtin_neon_vst2_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 8); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_u16(uint16x4_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3q_f16(__p0, __p1) __extension__ ({ \
+ float16x8x3_t __s1 = __p1; \
+ __builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 40); \
+})
#else
-__ai int16x4_t vreinterpret_s16_u16(uint16x4_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3q_f16(__p0, __p1) __extension__ ({ \
+ float16x8x3_t __s1 = __p1; \
+ float16x8x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 40); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_s8(int8x8_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3_f16(__p0, __p1) __extension__ ({ \
+ float16x4x3_t __s1 = __p1; \
+ __builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 8); \
+})
#else
-__ai int16x4_t vreinterpret_s16_s8(int8x8_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3_f16(__p0, __p1) __extension__ ({ \
+ float16x4x3_t __s1 = __p1; \
+ float16x4x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ __builtin_neon_vst3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 8); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_f32(float32x2_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8x3_t __s1 = __p1; \
+ __builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 40); \
+})
#else
-__ai int16x4_t vreinterpret_s16_f32(float32x2_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8x3_t __s1 = __p1; \
+ float16x8x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 40); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_f16(float16x4_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4x3_t __s1 = __p1; \
+ __builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 8); \
+})
#else
-__ai int16x4_t vreinterpret_s16_f16(float16x4_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4x3_t __s1 = __p1; \
+ float16x4x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ __builtin_neon_vst3_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 8); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_s32(int32x2_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst4q_f16(__p0, __p1) __extension__ ({ \
+ float16x8x4_t __s1 = __p1; \
+ __builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 40); \
+})
#else
-__ai int16x4_t vreinterpret_s16_s32(int32x2_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst4q_f16(__p0, __p1) __extension__ ({ \
+ float16x8x4_t __s1 = __p1; \
+ float16x8x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 40); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_s64(int64x1_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst4_f16(__p0, __p1) __extension__ ({ \
+ float16x4x4_t __s1 = __p1; \
+ __builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 8); \
+})
#else
-__ai int16x4_t vreinterpret_s16_s64(int64x1_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst4_f16(__p0, __p1) __extension__ ({ \
+ float16x4x4_t __s1 = __p1; \
+ float16x4x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
+ __builtin_neon_vst4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 8); \
+})
#endif
-#endif
-#if (__ARM_FP & 2)
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vcvt_f16_f32(float32x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vcvt_f16_f32((int8x16_t)__p0, 8);
- return __ret;
-}
+#define vst4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8x4_t __s1 = __p1; \
+ __builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 40); \
+})
#else
-__ai float16x4_t vcvt_f16_f32(float32x4_t __p0) {
- float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vcvt_f16_f32((int8x16_t)__rev0, 8);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
-__ai float16x4_t __noswap_vcvt_f16_f32(float32x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vcvt_f16_f32((int8x16_t)__p0, 8);
- return __ret;
-}
+#define vst4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x8x4_t __s1 = __p1; \
+ float16x8x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 40); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vcvt_f32_f16(float16x4_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t) __builtin_neon_vcvt_f32_f16((int8x8_t)__p0, 41);
- return __ret;
-}
+#define vst4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4x4_t __s1 = __p1; \
+ __builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 8); \
+})
#else
-__ai float32x4_t vcvt_f32_f16(float16x4_t __p0) {
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- float32x4_t __ret;
- __ret = (float32x4_t) __builtin_neon_vcvt_f32_f16((int8x8_t)__rev0, 41);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
-__ai float32x4_t __noswap_vcvt_f32_f16(float16x4_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t) __builtin_neon_vcvt_f32_f16((int8x8_t)__p0, 41);
- return __ret;
-}
+#define vst4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
+ float16x4x4_t __s1 = __p1; \
+ float16x4x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
+ __builtin_neon_vst4_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 8); \
+})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint32x4_t vsha1cq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
uint32x4_t __ret;
- __ret = (uint32x4_t) __builtin_neon_vsha1cq_u32((int8x16_t)__p0, __p1, (int8x16_t)__p2);
+ __ret = (uint32x4_t) __builtin_neon_vsha1cq_u32(__p0, __p1, __p2);
return __ret;
}
#else
uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
uint32x4_t __ret;
- __ret = (uint32x4_t) __builtin_neon_vsha1cq_u32((int8x16_t)__rev0, __p1, (int8x16_t)__rev2);
+ __ret = (uint32x4_t) __builtin_neon_vsha1cq_u32(__rev0, __p1, __rev2);
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vsha1h_u32(uint32_t __p0) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vsha1h_u32(__p0);
- return __ret;
-}
-#else
__ai uint32_t vsha1h_u32(uint32_t __p0) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vsha1h_u32(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint32x4_t vsha1mq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
uint32x4_t __ret;
- __ret = (uint32x4_t) __builtin_neon_vsha1mq_u32((int8x16_t)__p0, __p1, (int8x16_t)__p2);
+ __ret = (uint32x4_t) __builtin_neon_vsha1mq_u32(__p0, __p1, __p2);
return __ret;
}
#else
uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
uint32x4_t __ret;
- __ret = (uint32x4_t) __builtin_neon_vsha1mq_u32((int8x16_t)__rev0, __p1, (int8x16_t)__rev2);
+ __ret = (uint32x4_t) __builtin_neon_vsha1mq_u32(__rev0, __p1, __rev2);
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#ifdef __LITTLE_ENDIAN__
__ai uint32x4_t vsha1pq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
uint32x4_t __ret;
- __ret = (uint32x4_t) __builtin_neon_vsha1pq_u32((int8x16_t)__p0, __p1, (int8x16_t)__p2);
+ __ret = (uint32x4_t) __builtin_neon_vsha1pq_u32(__p0, __p1, __p2);
return __ret;
}
#else
uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
uint32x4_t __ret;
- __ret = (uint32x4_t) __builtin_neon_vsha1pq_u32((int8x16_t)__rev0, __p1, (int8x16_t)__rev2);
+ __ret = (uint32x4_t) __builtin_neon_vsha1pq_u32(__rev0, __p1, __rev2);
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
+__ai float32x4_t vrndiq_f32(float32x4_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vrndiq_v((int8x16_t)__p0, 41);
+ return __ret;
+}
+#else
+__ai float32x4_t vrndiq_f32(float32x4_t __p0) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vrndiq_v((int8x16_t)__rev0, 41);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float32x2_t vrndi_f32(float32x2_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vrndi_v((int8x8_t)__p0, 9);
+ return __ret;
+}
+#else
+__ai float32x2_t vrndi_f32(float32x2_t __p0) {
+ float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vrndi_v((int8x8_t)__rev0, 9);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
__ai float32x4_t vrndmq_f32(float32x4_t __p0) {
float32x4_t __ret;
__ret = (float32x4_t) __builtin_neon_vrndmq_v((int8x16_t)__p0, 41);
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float32_t vrndns_f32(float32_t __p0) {
- float32_t __ret;
- __ret = (float32_t) __builtin_neon_vrndns_f32(__p0);
- return __ret;
-}
-#else
__ai float32_t vrndns_f32(float32_t __p0) {
float32_t __ret;
__ret = (float32_t) __builtin_neon_vrndns_f32(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float32x4_t vrndpq_f32(float32x4_t __p0) {
float32x4_t __ret;
#endif
#endif
+#if __ARM_ARCH >= 8 && defined(__ARM_FEATURE_DIRECTED_ROUNDING) && defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC)
+#ifdef __LITTLE_ENDIAN__
+__ai float16x8_t vrndq_f16(float16x8_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vrndq_v((int8x16_t)__p0, 40);
+ return __ret;
+}
+#else
+__ai float16x8_t vrndq_f16(float16x8_t __p0) {
+ float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vrndq_v((int8x16_t)__rev0, 40);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float16x4_t vrnd_f16(float16x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vrnd_v((int8x8_t)__p0, 8);
+ return __ret;
+}
+#else
+__ai float16x4_t vrnd_f16(float16x4_t __p0) {
+ float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vrnd_v((int8x8_t)__rev0, 8);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float16x8_t vrndaq_f16(float16x8_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vrndaq_v((int8x16_t)__p0, 40);
+ return __ret;
+}
+#else
+__ai float16x8_t vrndaq_f16(float16x8_t __p0) {
+ float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vrndaq_v((int8x16_t)__rev0, 40);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float16x4_t vrnda_f16(float16x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vrnda_v((int8x8_t)__p0, 8);
+ return __ret;
+}
+#else
+__ai float16x4_t vrnda_f16(float16x4_t __p0) {
+ float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vrnda_v((int8x8_t)__rev0, 8);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float16x8_t vrndmq_f16(float16x8_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vrndmq_v((int8x16_t)__p0, 40);
+ return __ret;
+}
+#else
+__ai float16x8_t vrndmq_f16(float16x8_t __p0) {
+ float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vrndmq_v((int8x16_t)__rev0, 40);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float16x4_t vrndm_f16(float16x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vrndm_v((int8x8_t)__p0, 8);
+ return __ret;
+}
+#else
+__ai float16x4_t vrndm_f16(float16x4_t __p0) {
+ float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vrndm_v((int8x8_t)__rev0, 8);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float16x8_t vrndnq_f16(float16x8_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vrndnq_v((int8x16_t)__p0, 40);
+ return __ret;
+}
+#else
+__ai float16x8_t vrndnq_f16(float16x8_t __p0) {
+ float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vrndnq_v((int8x16_t)__rev0, 40);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float16x4_t vrndn_f16(float16x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vrndn_v((int8x8_t)__p0, 8);
+ return __ret;
+}
+#else
+__ai float16x4_t vrndn_f16(float16x4_t __p0) {
+ float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vrndn_v((int8x8_t)__rev0, 8);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float16x8_t vrndpq_f16(float16x8_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vrndpq_v((int8x16_t)__p0, 40);
+ return __ret;
+}
+#else
+__ai float16x8_t vrndpq_f16(float16x8_t __p0) {
+ float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vrndpq_v((int8x16_t)__rev0, 40);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float16x4_t vrndp_f16(float16x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vrndp_v((int8x8_t)__p0, 8);
+ return __ret;
+}
+#else
+__ai float16x4_t vrndp_f16(float16x4_t __p0) {
+ float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vrndp_v((int8x8_t)__rev0, 8);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float16x8_t vrndxq_f16(float16x8_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vrndxq_v((int8x16_t)__p0, 40);
+ return __ret;
+}
+#else
+__ai float16x8_t vrndxq_f16(float16x8_t __p0) {
+ float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vrndxq_v((int8x16_t)__rev0, 40);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float16x4_t vrndx_f16(float16x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vrndx_v((int8x8_t)__p0, 8);
+ return __ret;
+}
+#else
+__ai float16x4_t vrndx_f16(float16x4_t __p0) {
+ float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vrndx_v((int8x8_t)__rev0, 8);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#endif
#if __ARM_ARCH >= 8 && defined(__ARM_FEATURE_NUMERIC_MAXMIN)
#ifdef __LITTLE_ENDIAN__
__ai float32x4_t vmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
#endif
#endif
+#if __ARM_ARCH >= 8 && defined(__ARM_FEATURE_NUMERIC_MAXMIN) && defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC)
+#ifdef __LITTLE_ENDIAN__
+__ai float16x8_t vmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
+ return __ret;
+}
+#else
+__ai float16x8_t vmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
+ float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float16x4_t vmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
+ return __ret;
+}
+#else
+__ai float16x4_t vmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
+ float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vmaxnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float16x8_t vminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
+ return __ret;
+}
+#else
+__ai float16x8_t vminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
+ float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float16x4_t vminnm_f16(float16x4_t __p0, float16x4_t __p1) {
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
+ return __ret;
+}
+#else
+__ai float16x4_t vminnm_f16(float16x4_t __p0, float16x4_t __p1) {
+ float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vminnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#endif
#if __ARM_ARCH >= 8 && defined(__aarch64__)
#ifdef __LITTLE_ENDIAN__
__ai int64x2_t vcvtaq_s64_f64(float64x2_t __p0) {
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vcvta_s64_f64(float64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vcvta_s64_v((int8x8_t)__p0, 3);
- return __ret;
-}
-#else
__ai int64x1_t vcvta_s64_f64(float64x1_t __p0) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vcvta_s64_v((int8x8_t)__p0, 3);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vcvtaq_u64_f64(float64x2_t __p0) {
uint64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vcvta_u64_f64(float64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vcvta_u64_v((int8x8_t)__p0, 19);
return __ret;
}
-#else
-__ai uint64x1_t vcvta_u64_f64(float64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vcvta_u64_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int64x2_t vcvtmq_s64_f64(float64x2_t __p0) {
int64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vcvtm_s64_f64(float64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vcvtm_s64_v((int8x8_t)__p0, 3);
- return __ret;
-}
-#else
__ai int64x1_t vcvtm_s64_f64(float64x1_t __p0) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vcvtm_s64_v((int8x8_t)__p0, 3);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vcvtmq_u64_f64(float64x2_t __p0) {
uint64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vcvtm_u64_f64(float64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vcvtm_u64_v((int8x8_t)__p0, 19);
return __ret;
}
-#else
-__ai uint64x1_t vcvtm_u64_f64(float64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vcvtm_u64_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int64x2_t vcvtnq_s64_f64(float64x2_t __p0) {
int64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vcvtn_s64_f64(float64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vcvtn_s64_v((int8x8_t)__p0, 3);
- return __ret;
-}
-#else
__ai int64x1_t vcvtn_s64_f64(float64x1_t __p0) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vcvtn_s64_v((int8x8_t)__p0, 3);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vcvtnq_u64_f64(float64x2_t __p0) {
uint64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vcvtn_u64_f64(float64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vcvtn_u64_v((int8x8_t)__p0, 19);
return __ret;
}
-#else
-__ai uint64x1_t vcvtn_u64_f64(float64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vcvtn_u64_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int64x2_t vcvtpq_s64_f64(float64x2_t __p0) {
int64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai int64x1_t vcvtp_s64_f64(float64x1_t __p0) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vcvtp_s64_v((int8x8_t)__p0, 3);
return __ret;
}
-#else
-__ai int64x1_t vcvtp_s64_f64(float64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vcvtp_s64_v((int8x8_t)__p0, 3);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vcvtpq_u64_f64(float64x2_t __p0) {
uint64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vcvtp_u64_f64(float64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vcvtp_u64_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#else
__ai uint64x1_t vcvtp_u64_f64(float64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vcvtp_u64_v((int8x8_t)__p0, 19);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly8x8_t vreinterpret_p8_p64(poly64x1_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#else
-__ai poly8x8_t vreinterpret_p8_p64(poly64x1_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_p16(poly16x4_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x8_t vreinterpret_p8_p16(poly16x4_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_u8(uint8x8_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x8_t vreinterpret_p8_u8(uint8x8_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai poly8x8_t vreinterpret_p8_u32(uint32x2_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#else
-__ai poly8x8_t vreinterpret_p8_u32(uint32x2_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_u64(uint64x1_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
-#else
-__ai poly8x8_t vreinterpret_p8_u64(uint64x1_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_u16(uint16x4_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
-#else
-__ai poly8x8_t vreinterpret_p8_u16(uint16x4_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_s8(int8x8_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
-#else
-__ai poly8x8_t vreinterpret_p8_s8(int8x8_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_f64(float64x1_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
-#else
-__ai poly8x8_t vreinterpret_p8_f64(float64x1_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_f32(float32x2_t __p0) {
- poly8x8_t __ret;
- __ret = (poly8x8_t)(__p0);
- return __ret;
-}
-#else
-__ai poly8x8_t vreinterpret_p8_f32(float32x2_t __p0) {
+__ai poly8x8_t vreinterpret_p8_u64(uint64x1_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_f16(float16x4_t __p0) {
+__ai poly8x8_t vreinterpret_p8_u16(uint16x4_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#else
-__ai poly8x8_t vreinterpret_p8_f16(float16x4_t __p0) {
+__ai poly8x8_t vreinterpret_p8_s8(int8x8_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_s32(int32x2_t __p0) {
+__ai poly8x8_t vreinterpret_p8_f64(float64x1_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#else
-__ai poly8x8_t vreinterpret_p8_s32(int32x2_t __p0) {
+__ai poly8x8_t vreinterpret_p8_f32(float32x2_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_s64(int64x1_t __p0) {
+__ai poly8x8_t vreinterpret_p8_f16(float16x4_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#else
-__ai poly8x8_t vreinterpret_p8_s64(int64x1_t __p0) {
+__ai poly8x8_t vreinterpret_p8_s32(int32x2_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x8_t vreinterpret_p8_s16(int16x4_t __p0) {
+__ai poly8x8_t vreinterpret_p8_s64(int64x1_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#else
__ai poly8x8_t vreinterpret_p8_s16(int16x4_t __p0) {
poly8x8_t __ret;
__ret = (poly8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vreinterpret_p64_p8(poly8x8_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x1_t vreinterpret_p64_p8(poly8x8_t __p0) {
poly64x1_t __ret;
__ret = (poly64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vreinterpret_p64_p16(poly16x4_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x1_t vreinterpret_p64_p16(poly16x4_t __p0) {
poly64x1_t __ret;
__ret = (poly64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vreinterpret_p64_u8(uint8x8_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x1_t vreinterpret_p64_u8(uint8x8_t __p0) {
poly64x1_t __ret;
__ret = (poly64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vreinterpret_p64_u32(uint32x2_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x1_t vreinterpret_p64_u32(uint32x2_t __p0) {
poly64x1_t __ret;
__ret = (poly64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vreinterpret_p64_u64(uint64x1_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x1_t vreinterpret_p64_u64(uint64x1_t __p0) {
poly64x1_t __ret;
__ret = (poly64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vreinterpret_p64_u16(uint16x4_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x1_t vreinterpret_p64_u16(uint16x4_t __p0) {
poly64x1_t __ret;
__ret = (poly64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vreinterpret_p64_s8(int8x8_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x1_t vreinterpret_p64_s8(int8x8_t __p0) {
poly64x1_t __ret;
__ret = (poly64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vreinterpret_p64_f64(float64x1_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x1_t vreinterpret_p64_f64(float64x1_t __p0) {
poly64x1_t __ret;
__ret = (poly64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vreinterpret_p64_f32(float32x2_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x1_t vreinterpret_p64_f32(float32x2_t __p0) {
poly64x1_t __ret;
__ret = (poly64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vreinterpret_p64_f16(float16x4_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x1_t vreinterpret_p64_f16(float16x4_t __p0) {
poly64x1_t __ret;
__ret = (poly64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vreinterpret_p64_s32(int32x2_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x1_t vreinterpret_p64_s32(int32x2_t __p0) {
poly64x1_t __ret;
__ret = (poly64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vreinterpret_p64_s64(int64x1_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x1_t vreinterpret_p64_s64(int64x1_t __p0) {
poly64x1_t __ret;
__ret = (poly64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vreinterpret_p64_s16(int16x4_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x1_t vreinterpret_p64_s16(int16x4_t __p0) {
poly64x1_t __ret;
__ret = (poly64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_p8(poly8x8_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_p8(poly8x8_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_p64(poly64x1_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_p64(poly64x1_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_u8(uint8x8_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_u8(uint8x8_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_u32(uint32x2_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_u32(uint32x2_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_u64(uint64x1_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_u64(uint64x1_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_u16(uint16x4_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_u16(uint16x4_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_s8(int8x8_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_s8(int8x8_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_f64(float64x1_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_f64(float64x1_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_f32(float32x2_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_f32(float32x2_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_f16(float16x4_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_f16(float16x4_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_s32(int32x2_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_s32(int32x2_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_s64(int64x1_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_s64(int64x1_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x4_t vreinterpret_p16_s16(int16x4_t __p0) {
- poly16x4_t __ret;
- __ret = (poly16x4_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x4_t vreinterpret_p16_s16(int16x4_t __p0) {
poly16x4_t __ret;
__ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_p128(poly128_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_p128(poly128_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_p64(poly64x2_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_p64(poly64x2_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_p16(poly16x8_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_p16(poly16x8_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_u8(uint8x16_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_u8(uint8x16_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_u32(uint32x4_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_u32(uint32x4_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_u64(uint64x2_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_u64(uint64x2_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_u16(uint16x8_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_u16(uint16x8_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_s8(int8x16_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_s8(int8x16_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_f64(float64x2_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_f64(float64x2_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_f32(float32x4_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_f32(float32x4_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_f16(float16x8_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_f16(float16x8_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_s32(int32x4_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_s32(int32x4_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_s64(int64x2_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_s64(int64x2_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly8x16_t vreinterpretq_p8_s16(int16x8_t __p0) {
- poly8x16_t __ret;
- __ret = (poly8x16_t)(__p0);
- return __ret;
-}
-#else
__ai poly8x16_t vreinterpretq_p8_s16(int16x8_t __p0) {
poly8x16_t __ret;
__ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly128_t vreinterpretq_p128_p8(poly8x16_t __p0) {
- poly128_t __ret;
- __ret = (poly128_t)(__p0);
- return __ret;
-}
-#else
__ai poly128_t vreinterpretq_p128_p8(poly8x16_t __p0) {
poly128_t __ret;
__ret = (poly128_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly128_t vreinterpretq_p128_p64(poly64x2_t __p0) {
- poly128_t __ret;
- __ret = (poly128_t)(__p0);
- return __ret;
-}
-#else
__ai poly128_t vreinterpretq_p128_p64(poly64x2_t __p0) {
poly128_t __ret;
__ret = (poly128_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly128_t vreinterpretq_p128_p16(poly16x8_t __p0) {
- poly128_t __ret;
- __ret = (poly128_t)(__p0);
- return __ret;
-}
-#else
__ai poly128_t vreinterpretq_p128_p16(poly16x8_t __p0) {
poly128_t __ret;
__ret = (poly128_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly128_t vreinterpretq_p128_u8(uint8x16_t __p0) {
- poly128_t __ret;
- __ret = (poly128_t)(__p0);
- return __ret;
-}
-#else
__ai poly128_t vreinterpretq_p128_u8(uint8x16_t __p0) {
poly128_t __ret;
__ret = (poly128_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly128_t vreinterpretq_p128_u32(uint32x4_t __p0) {
- poly128_t __ret;
- __ret = (poly128_t)(__p0);
- return __ret;
-}
-#else
__ai poly128_t vreinterpretq_p128_u32(uint32x4_t __p0) {
poly128_t __ret;
__ret = (poly128_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly128_t vreinterpretq_p128_u64(uint64x2_t __p0) {
- poly128_t __ret;
- __ret = (poly128_t)(__p0);
- return __ret;
-}
-#else
-__ai poly128_t vreinterpretq_p128_u64(uint64x2_t __p0) {
- poly128_t __ret;
- __ret = (poly128_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly128_t vreinterpretq_p128_u16(uint16x8_t __p0) {
- poly128_t __ret;
- __ret = (poly128_t)(__p0);
- return __ret;
-}
-#else
-__ai poly128_t vreinterpretq_p128_u16(uint16x8_t __p0) {
- poly128_t __ret;
- __ret = (poly128_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly128_t vreinterpretq_p128_s8(int8x16_t __p0) {
- poly128_t __ret;
- __ret = (poly128_t)(__p0);
- return __ret;
-}
-#else
-__ai poly128_t vreinterpretq_p128_s8(int8x16_t __p0) {
- poly128_t __ret;
- __ret = (poly128_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly128_t vreinterpretq_p128_f64(float64x2_t __p0) {
- poly128_t __ret;
- __ret = (poly128_t)(__p0);
- return __ret;
-}
-#else
-__ai poly128_t vreinterpretq_p128_f64(float64x2_t __p0) {
- poly128_t __ret;
- __ret = (poly128_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly128_t vreinterpretq_p128_f32(float32x4_t __p0) {
- poly128_t __ret;
- __ret = (poly128_t)(__p0);
- return __ret;
-}
-#else
-__ai poly128_t vreinterpretq_p128_f32(float32x4_t __p0) {
- poly128_t __ret;
- __ret = (poly128_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly128_t vreinterpretq_p128_f16(float16x8_t __p0) {
+__ai poly128_t vreinterpretq_p128_u64(uint64x2_t __p0) {
poly128_t __ret;
__ret = (poly128_t)(__p0);
return __ret;
}
-#else
-__ai poly128_t vreinterpretq_p128_f16(float16x8_t __p0) {
+__ai poly128_t vreinterpretq_p128_u16(uint16x8_t __p0) {
poly128_t __ret;
__ret = (poly128_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly128_t vreinterpretq_p128_s32(int32x4_t __p0) {
+__ai poly128_t vreinterpretq_p128_s8(int8x16_t __p0) {
poly128_t __ret;
__ret = (poly128_t)(__p0);
return __ret;
}
-#else
-__ai poly128_t vreinterpretq_p128_s32(int32x4_t __p0) {
+__ai poly128_t vreinterpretq_p128_f64(float64x2_t __p0) {
poly128_t __ret;
__ret = (poly128_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly128_t vreinterpretq_p128_s64(int64x2_t __p0) {
+__ai poly128_t vreinterpretq_p128_f32(float32x4_t __p0) {
poly128_t __ret;
__ret = (poly128_t)(__p0);
return __ret;
}
-#else
-__ai poly128_t vreinterpretq_p128_s64(int64x2_t __p0) {
+__ai poly128_t vreinterpretq_p128_f16(float16x8_t __p0) {
poly128_t __ret;
__ret = (poly128_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly128_t vreinterpretq_p128_s16(int16x8_t __p0) {
+__ai poly128_t vreinterpretq_p128_s32(int32x4_t __p0) {
poly128_t __ret;
__ret = (poly128_t)(__p0);
return __ret;
}
-#else
-__ai poly128_t vreinterpretq_p128_s16(int16x8_t __p0) {
+__ai poly128_t vreinterpretq_p128_s64(int64x2_t __p0) {
poly128_t __ret;
__ret = (poly128_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x2_t vreinterpretq_p64_p8(poly8x16_t __p0) {
- poly64x2_t __ret;
- __ret = (poly64x2_t)(__p0);
+__ai poly128_t vreinterpretq_p128_s16(int16x8_t __p0) {
+ poly128_t __ret;
+ __ret = (poly128_t)(__p0);
return __ret;
}
-#else
__ai poly64x2_t vreinterpretq_p64_p8(poly8x16_t __p0) {
poly64x2_t __ret;
__ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x2_t vreinterpretq_p64_p128(poly128_t __p0) {
- poly64x2_t __ret;
- __ret = (poly64x2_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x2_t vreinterpretq_p64_p128(poly128_t __p0) {
poly64x2_t __ret;
__ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x2_t vreinterpretq_p64_p16(poly16x8_t __p0) {
- poly64x2_t __ret;
- __ret = (poly64x2_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x2_t vreinterpretq_p64_p16(poly16x8_t __p0) {
poly64x2_t __ret;
__ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x2_t vreinterpretq_p64_u8(uint8x16_t __p0) {
- poly64x2_t __ret;
- __ret = (poly64x2_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x2_t vreinterpretq_p64_u8(uint8x16_t __p0) {
poly64x2_t __ret;
__ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x2_t vreinterpretq_p64_u32(uint32x4_t __p0) {
- poly64x2_t __ret;
- __ret = (poly64x2_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x2_t vreinterpretq_p64_u32(uint32x4_t __p0) {
poly64x2_t __ret;
__ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x2_t vreinterpretq_p64_u64(uint64x2_t __p0) {
- poly64x2_t __ret;
- __ret = (poly64x2_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x2_t vreinterpretq_p64_u64(uint64x2_t __p0) {
poly64x2_t __ret;
__ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x2_t vreinterpretq_p64_u16(uint16x8_t __p0) {
- poly64x2_t __ret;
- __ret = (poly64x2_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x2_t vreinterpretq_p64_u16(uint16x8_t __p0) {
poly64x2_t __ret;
__ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x2_t vreinterpretq_p64_s8(int8x16_t __p0) {
- poly64x2_t __ret;
- __ret = (poly64x2_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x2_t vreinterpretq_p64_s8(int8x16_t __p0) {
poly64x2_t __ret;
__ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x2_t vreinterpretq_p64_f64(float64x2_t __p0) {
- poly64x2_t __ret;
- __ret = (poly64x2_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x2_t vreinterpretq_p64_f64(float64x2_t __p0) {
poly64x2_t __ret;
__ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x2_t vreinterpretq_p64_f32(float32x4_t __p0) {
- poly64x2_t __ret;
- __ret = (poly64x2_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x2_t vreinterpretq_p64_f32(float32x4_t __p0) {
poly64x2_t __ret;
__ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x2_t vreinterpretq_p64_f16(float16x8_t __p0) {
- poly64x2_t __ret;
- __ret = (poly64x2_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x2_t vreinterpretq_p64_f16(float16x8_t __p0) {
poly64x2_t __ret;
__ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x2_t vreinterpretq_p64_s32(int32x4_t __p0) {
- poly64x2_t __ret;
- __ret = (poly64x2_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x2_t vreinterpretq_p64_s32(int32x4_t __p0) {
poly64x2_t __ret;
__ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x2_t vreinterpretq_p64_s64(int64x2_t __p0) {
- poly64x2_t __ret;
- __ret = (poly64x2_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x2_t vreinterpretq_p64_s64(int64x2_t __p0) {
poly64x2_t __ret;
__ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x2_t vreinterpretq_p64_s16(int16x8_t __p0) {
- poly64x2_t __ret;
- __ret = (poly64x2_t)(__p0);
- return __ret;
-}
-#else
__ai poly64x2_t vreinterpretq_p64_s16(int16x8_t __p0) {
poly64x2_t __ret;
__ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_p8(poly8x16_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x8_t vreinterpretq_p16_p8(poly8x16_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_p128(poly128_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x8_t vreinterpretq_p16_p128(poly128_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_p64(poly64x2_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x8_t vreinterpretq_p16_p64(poly64x2_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_u8(uint8x16_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x8_t vreinterpretq_p16_u8(uint8x16_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_u32(uint32x4_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x8_t vreinterpretq_p16_u32(uint32x4_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_u64(uint64x2_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x8_t vreinterpretq_p16_u64(uint64x2_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_u16(uint16x8_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x8_t vreinterpretq_p16_u16(uint16x8_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_s8(int8x16_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x8_t vreinterpretq_p16_s8(int8x16_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_f64(float64x2_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x8_t vreinterpretq_p16_f64(float64x2_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_f32(float32x4_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x8_t vreinterpretq_p16_f32(float32x4_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_f16(float16x8_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x8_t vreinterpretq_p16_f16(float16x8_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_s32(int32x4_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x8_t vreinterpretq_p16_s32(int32x4_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_s64(int64x2_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x8_t vreinterpretq_p16_s64(int64x2_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly16x8_t vreinterpretq_p16_s16(int16x8_t __p0) {
- poly16x8_t __ret;
- __ret = (poly16x8_t)(__p0);
- return __ret;
-}
-#else
__ai poly16x8_t vreinterpretq_p16_s16(int16x8_t __p0) {
poly16x8_t __ret;
__ret = (poly16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_p8(poly8x16_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_p8(poly8x16_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_p128(poly128_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_p128(poly128_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_p64(poly64x2_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_p64(poly64x2_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_p16(poly16x8_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_p16(poly16x8_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_u32(uint32x4_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_u32(uint32x4_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_u64(uint64x2_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_u64(uint64x2_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_u16(uint16x8_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_u16(uint16x8_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_s8(int8x16_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_s8(int8x16_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_f64(float64x2_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_f64(float64x2_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_f32(float32x4_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_f32(float32x4_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_f16(float16x8_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_f16(float16x8_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_s32(int32x4_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_s32(int32x4_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_s64(int64x2_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_s64(int64x2_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vreinterpretq_u8_s16(int16x8_t __p0) {
- uint8x16_t __ret;
- __ret = (uint8x16_t)(__p0);
- return __ret;
-}
-#else
__ai uint8x16_t vreinterpretq_u8_s16(int16x8_t __p0) {
uint8x16_t __ret;
__ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_p8(poly8x16_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x4_t vreinterpretq_u32_p8(poly8x16_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_p128(poly128_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x4_t vreinterpretq_u32_p128(poly128_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_p64(poly64x2_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
__ai uint32x4_t vreinterpretq_u32_p64(poly64x2_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_p16(poly16x8_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
-__ai uint32x4_t vreinterpretq_u32_p16(poly16x8_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_u8(uint8x16_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
-__ai uint32x4_t vreinterpretq_u32_u8(uint8x16_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_u64(uint64x2_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
-__ai uint32x4_t vreinterpretq_u32_u64(uint64x2_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_u16(uint16x8_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
-__ai uint32x4_t vreinterpretq_u32_u16(uint16x8_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_s8(int8x16_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
-__ai uint32x4_t vreinterpretq_u32_s8(int8x16_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_f64(float64x2_t __p0) {
- uint32x4_t __ret;
- __ret = (uint32x4_t)(__p0);
- return __ret;
-}
-#else
-__ai uint32x4_t vreinterpretq_u32_f64(float64x2_t __p0) {
+__ai uint32x4_t vreinterpretq_u32_p16(poly16x8_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_f32(float32x4_t __p0) {
+__ai uint32x4_t vreinterpretq_u32_u8(uint8x16_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#else
-__ai uint32x4_t vreinterpretq_u32_f32(float32x4_t __p0) {
+__ai uint32x4_t vreinterpretq_u32_u64(uint64x2_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_f16(float16x8_t __p0) {
+__ai uint32x4_t vreinterpretq_u32_u16(uint16x8_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#else
-__ai uint32x4_t vreinterpretq_u32_f16(float16x8_t __p0) {
+__ai uint32x4_t vreinterpretq_u32_s8(int8x16_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_s32(int32x4_t __p0) {
+__ai uint32x4_t vreinterpretq_u32_f64(float64x2_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#else
-__ai uint32x4_t vreinterpretq_u32_s32(int32x4_t __p0) {
+__ai uint32x4_t vreinterpretq_u32_f32(float32x4_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_s64(int64x2_t __p0) {
+__ai uint32x4_t vreinterpretq_u32_f16(float16x8_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#else
-__ai uint32x4_t vreinterpretq_u32_s64(int64x2_t __p0) {
+__ai uint32x4_t vreinterpretq_u32_s32(int32x4_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vreinterpretq_u32_s16(int16x8_t __p0) {
+__ai uint32x4_t vreinterpretq_u32_s64(int64x2_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#else
__ai uint32x4_t vreinterpretq_u32_s16(int16x8_t __p0) {
uint32x4_t __ret;
__ret = (uint32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_p8(poly8x16_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_p8(poly8x16_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_p128(poly128_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_p128(poly128_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_p64(poly64x2_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_p64(poly64x2_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_p16(poly16x8_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_p16(poly16x8_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_u8(uint8x16_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_u8(uint8x16_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_u32(uint32x4_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_u32(uint32x4_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_u16(uint16x8_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_u16(uint16x8_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_s8(int8x16_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_s8(int8x16_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_f64(float64x2_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_f64(float64x2_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_f32(float32x4_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_f32(float32x4_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_f16(float16x8_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_f16(float16x8_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_s32(int32x4_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_s32(int32x4_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_s64(int64x2_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_s64(int64x2_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vreinterpretq_u64_s16(int16x8_t __p0) {
- uint64x2_t __ret;
- __ret = (uint64x2_t)(__p0);
- return __ret;
-}
-#else
__ai uint64x2_t vreinterpretq_u64_s16(int16x8_t __p0) {
uint64x2_t __ret;
__ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint16x8_t vreinterpretq_u16_p8(poly8x16_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x8_t vreinterpretq_u16_p8(poly8x16_t __p0) {
+__ai uint16x8_t vreinterpretq_u16_p128(poly128_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_p128(poly128_t __p0) {
+__ai uint16x8_t vreinterpretq_u16_p64(poly64x2_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x8_t vreinterpretq_u16_p128(poly128_t __p0) {
+__ai uint16x8_t vreinterpretq_u16_p16(poly16x8_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_p64(poly64x2_t __p0) {
+__ai uint16x8_t vreinterpretq_u16_u8(uint8x16_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x8_t vreinterpretq_u16_p64(poly64x2_t __p0) {
+__ai uint16x8_t vreinterpretq_u16_u32(uint32x4_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_p16(poly16x8_t __p0) {
+__ai uint16x8_t vreinterpretq_u16_u64(uint64x2_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x8_t vreinterpretq_u16_p16(poly16x8_t __p0) {
+__ai uint16x8_t vreinterpretq_u16_s8(int8x16_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_u8(uint8x16_t __p0) {
+__ai uint16x8_t vreinterpretq_u16_f64(float64x2_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x8_t vreinterpretq_u16_u8(uint8x16_t __p0) {
+__ai uint16x8_t vreinterpretq_u16_f32(float32x4_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_u32(uint32x4_t __p0) {
+__ai uint16x8_t vreinterpretq_u16_f16(float16x8_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x8_t vreinterpretq_u16_u32(uint32x4_t __p0) {
+__ai uint16x8_t vreinterpretq_u16_s32(int32x4_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_u64(uint64x2_t __p0) {
+__ai uint16x8_t vreinterpretq_u16_s64(int64x2_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x8_t vreinterpretq_u16_u64(uint64x2_t __p0) {
+__ai uint16x8_t vreinterpretq_u16_s16(int16x8_t __p0) {
uint16x8_t __ret;
__ret = (uint16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_s8(int8x16_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
+__ai int8x16_t vreinterpretq_s8_p8(poly8x16_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
return __ret;
}
-#else
-__ai uint16x8_t vreinterpretq_u16_s8(int8x16_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
+__ai int8x16_t vreinterpretq_s8_p128(poly128_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_f64(float64x2_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
+__ai int8x16_t vreinterpretq_s8_p64(poly64x2_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
return __ret;
}
-#else
-__ai uint16x8_t vreinterpretq_u16_f64(float64x2_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
+__ai int8x16_t vreinterpretq_s8_p16(poly16x8_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_f32(float32x4_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
+__ai int8x16_t vreinterpretq_s8_u8(uint8x16_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
return __ret;
}
-#else
-__ai uint16x8_t vreinterpretq_u16_f32(float32x4_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
+__ai int8x16_t vreinterpretq_s8_u32(uint32x4_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_f16(float16x8_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
+__ai int8x16_t vreinterpretq_s8_u64(uint64x2_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
return __ret;
}
-#else
-__ai uint16x8_t vreinterpretq_u16_f16(float16x8_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
+__ai int8x16_t vreinterpretq_s8_u16(uint16x8_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_s32(int32x4_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
+__ai int8x16_t vreinterpretq_s8_f64(float64x2_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
return __ret;
}
-#else
-__ai uint16x8_t vreinterpretq_u16_s32(int32x4_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
+__ai int8x16_t vreinterpretq_s8_f32(float32x4_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_s64(int64x2_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
+__ai int8x16_t vreinterpretq_s8_f16(float16x8_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
return __ret;
}
-#else
-__ai uint16x8_t vreinterpretq_u16_s64(int64x2_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
+__ai int8x16_t vreinterpretq_s8_s32(int32x4_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
+ return __ret;
+}
+__ai int8x16_t vreinterpretq_s8_s64(int64x2_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
+ return __ret;
+}
+__ai int8x16_t vreinterpretq_s8_s16(int16x8_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
+ return __ret;
+}
+__ai float64x2_t vreinterpretq_f64_p8(poly8x16_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t)(__p0);
+ return __ret;
+}
+__ai float64x2_t vreinterpretq_f64_p128(poly128_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t)(__p0);
+ return __ret;
+}
+__ai float64x2_t vreinterpretq_f64_p64(poly64x2_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t)(__p0);
+ return __ret;
+}
+__ai float64x2_t vreinterpretq_f64_p16(poly16x8_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t)(__p0);
+ return __ret;
+}
+__ai float64x2_t vreinterpretq_f64_u8(uint8x16_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t)(__p0);
+ return __ret;
+}
+__ai float64x2_t vreinterpretq_f64_u32(uint32x4_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t)(__p0);
+ return __ret;
+}
+__ai float64x2_t vreinterpretq_f64_u64(uint64x2_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t)(__p0);
+ return __ret;
+}
+__ai float64x2_t vreinterpretq_f64_u16(uint16x8_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t)(__p0);
+ return __ret;
+}
+__ai float64x2_t vreinterpretq_f64_s8(int8x16_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t)(__p0);
+ return __ret;
+}
+__ai float64x2_t vreinterpretq_f64_f32(float32x4_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t)(__p0);
+ return __ret;
+}
+__ai float64x2_t vreinterpretq_f64_f16(float16x8_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t)(__p0);
+ return __ret;
+}
+__ai float64x2_t vreinterpretq_f64_s32(int32x4_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t)(__p0);
+ return __ret;
+}
+__ai float64x2_t vreinterpretq_f64_s64(int64x2_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t)(__p0);
+ return __ret;
+}
+__ai float64x2_t vreinterpretq_f64_s16(int16x8_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t)(__p0);
+ return __ret;
+}
+__ai float32x4_t vreinterpretq_f32_p8(poly8x16_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
+ return __ret;
+}
+__ai float32x4_t vreinterpretq_f32_p128(poly128_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
+ return __ret;
+}
+__ai float32x4_t vreinterpretq_f32_p64(poly64x2_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
+ return __ret;
+}
+__ai float32x4_t vreinterpretq_f32_p16(poly16x8_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
+ return __ret;
+}
+__ai float32x4_t vreinterpretq_f32_u8(uint8x16_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
+ return __ret;
+}
+__ai float32x4_t vreinterpretq_f32_u32(uint32x4_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
+ return __ret;
+}
+__ai float32x4_t vreinterpretq_f32_u64(uint64x2_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
+ return __ret;
+}
+__ai float32x4_t vreinterpretq_f32_u16(uint16x8_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
+ return __ret;
+}
+__ai float32x4_t vreinterpretq_f32_s8(int8x16_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
+ return __ret;
+}
+__ai float32x4_t vreinterpretq_f32_f64(float64x2_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
+ return __ret;
+}
+__ai float32x4_t vreinterpretq_f32_f16(float16x8_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
+ return __ret;
+}
+__ai float32x4_t vreinterpretq_f32_s32(int32x4_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
+ return __ret;
+}
+__ai float32x4_t vreinterpretq_f32_s64(int64x2_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
+ return __ret;
+}
+__ai float32x4_t vreinterpretq_f32_s16(int16x8_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
+ return __ret;
+}
+__ai float16x8_t vreinterpretq_f16_p8(poly8x16_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
+ return __ret;
+}
+__ai float16x8_t vreinterpretq_f16_p128(poly128_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
+ return __ret;
+}
+__ai float16x8_t vreinterpretq_f16_p64(poly64x2_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
+ return __ret;
+}
+__ai float16x8_t vreinterpretq_f16_p16(poly16x8_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
+ return __ret;
+}
+__ai float16x8_t vreinterpretq_f16_u8(uint8x16_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
+ return __ret;
+}
+__ai float16x8_t vreinterpretq_f16_u32(uint32x4_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
+ return __ret;
+}
+__ai float16x8_t vreinterpretq_f16_u64(uint64x2_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
+ return __ret;
+}
+__ai float16x8_t vreinterpretq_f16_u16(uint16x8_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
+ return __ret;
+}
+__ai float16x8_t vreinterpretq_f16_s8(int8x16_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vreinterpretq_u16_s16(int16x8_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
+__ai float16x8_t vreinterpretq_f16_f64(float64x2_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x8_t vreinterpretq_u16_s16(int16x8_t __p0) {
- uint16x8_t __ret;
- __ret = (uint16x8_t)(__p0);
+__ai float16x8_t vreinterpretq_f16_f32(float32x4_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_p8(poly8x16_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai float16x8_t vreinterpretq_f16_s32(int32x4_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_p8(poly8x16_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai float16x8_t vreinterpretq_f16_s64(int64x2_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_p128(poly128_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai float16x8_t vreinterpretq_f16_s16(int16x8_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_p128(poly128_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_p8(poly8x16_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_p64(poly64x2_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_p128(poly128_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_p64(poly64x2_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_p64(poly64x2_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_p16(poly16x8_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_p16(poly16x8_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_p16(poly16x8_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_u8(uint8x16_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_u8(uint8x16_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_u32(uint32x4_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_u8(uint8x16_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_u64(uint64x2_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_u32(uint32x4_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_u16(uint16x8_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_u32(uint32x4_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_s8(int8x16_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_u64(uint64x2_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_f64(float64x2_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_u64(uint64x2_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_f32(float32x4_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_u16(uint16x8_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_f16(float16x8_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_u16(uint16x8_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_s64(int64x2_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_f64(float64x2_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_s16(int16x8_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_f64(float64x2_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int64x2_t vreinterpretq_s64_p8(poly8x16_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_f32(float32x4_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int64x2_t vreinterpretq_s64_p128(poly128_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_f32(float32x4_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int64x2_t vreinterpretq_s64_p64(poly64x2_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_f16(float16x8_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int64x2_t vreinterpretq_s64_p16(poly16x8_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_f16(float16x8_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int64x2_t vreinterpretq_s64_u8(uint8x16_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_s32(int32x4_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int64x2_t vreinterpretq_s64_u32(uint32x4_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_s32(int32x4_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int64x2_t vreinterpretq_s64_u64(uint64x2_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_s64(int64x2_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int64x2_t vreinterpretq_s64_u16(uint16x8_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_s64(int64x2_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int64x2_t vreinterpretq_s64_s8(int8x16_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vreinterpretq_s8_s16(int16x8_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int64x2_t vreinterpretq_s64_f64(float64x2_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#else
-__ai int8x16_t vreinterpretq_s8_s16(int16x8_t __p0) {
- int8x16_t __ret;
- __ret = (int8x16_t)(__p0);
+__ai int64x2_t vreinterpretq_s64_f32(float32x4_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vreinterpretq_f64_p8(poly8x16_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int64x2_t vreinterpretq_s64_f16(float16x8_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#else
-__ai float64x2_t vreinterpretq_f64_p8(poly8x16_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int64x2_t vreinterpretq_s64_s32(int32x4_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vreinterpretq_f64_p128(poly128_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int64x2_t vreinterpretq_s64_s16(int16x8_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#else
-__ai float64x2_t vreinterpretq_f64_p128(poly128_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int16x8_t vreinterpretq_s16_p8(poly8x16_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vreinterpretq_f64_p64(poly64x2_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int16x8_t vreinterpretq_s16_p128(poly128_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
return __ret;
}
-#else
-__ai float64x2_t vreinterpretq_f64_p64(poly64x2_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int16x8_t vreinterpretq_s16_p64(poly64x2_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vreinterpretq_f64_p16(poly16x8_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int16x8_t vreinterpretq_s16_p16(poly16x8_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
return __ret;
}
-#else
-__ai float64x2_t vreinterpretq_f64_p16(poly16x8_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int16x8_t vreinterpretq_s16_u8(uint8x16_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vreinterpretq_f64_u8(uint8x16_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int16x8_t vreinterpretq_s16_u32(uint32x4_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
return __ret;
}
-#else
-__ai float64x2_t vreinterpretq_f64_u8(uint8x16_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int16x8_t vreinterpretq_s16_u64(uint64x2_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vreinterpretq_f64_u32(uint32x4_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int16x8_t vreinterpretq_s16_u16(uint16x8_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
return __ret;
}
-#else
-__ai float64x2_t vreinterpretq_f64_u32(uint32x4_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int16x8_t vreinterpretq_s16_s8(int8x16_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vreinterpretq_f64_u64(uint64x2_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int16x8_t vreinterpretq_s16_f64(float64x2_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
return __ret;
}
-#else
-__ai float64x2_t vreinterpretq_f64_u64(uint64x2_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int16x8_t vreinterpretq_s16_f32(float32x4_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vreinterpretq_f64_u16(uint16x8_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int16x8_t vreinterpretq_s16_f16(float16x8_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
return __ret;
}
-#else
-__ai float64x2_t vreinterpretq_f64_u16(uint16x8_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int16x8_t vreinterpretq_s16_s32(int32x4_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vreinterpretq_f64_s8(int8x16_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai int16x8_t vreinterpretq_s16_s64(int64x2_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
return __ret;
}
-#else
-__ai float64x2_t vreinterpretq_f64_s8(int8x16_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai uint8x8_t vreinterpret_u8_p8(poly8x8_t __p0) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vreinterpretq_f64_f32(float32x4_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai uint8x8_t vreinterpret_u8_p64(poly64x1_t __p0) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t)(__p0);
return __ret;
}
-#else
-__ai float64x2_t vreinterpretq_f64_f32(float32x4_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai uint8x8_t vreinterpret_u8_p16(poly16x4_t __p0) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vreinterpretq_f64_f16(float16x8_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai uint8x8_t vreinterpret_u8_u32(uint32x2_t __p0) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t)(__p0);
return __ret;
}
-#else
-__ai float64x2_t vreinterpretq_f64_f16(float16x8_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai uint8x8_t vreinterpret_u8_u64(uint64x1_t __p0) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vreinterpretq_f64_s32(int32x4_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai uint8x8_t vreinterpret_u8_u16(uint16x4_t __p0) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t)(__p0);
return __ret;
}
-#else
-__ai float64x2_t vreinterpretq_f64_s32(int32x4_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai uint8x8_t vreinterpret_u8_s8(int8x8_t __p0) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vreinterpretq_f64_s64(int64x2_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai uint8x8_t vreinterpret_u8_f64(float64x1_t __p0) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t)(__p0);
return __ret;
}
-#else
-__ai float64x2_t vreinterpretq_f64_s64(int64x2_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai uint8x8_t vreinterpret_u8_f32(float32x2_t __p0) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vreinterpretq_f64_s16(int16x8_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai uint8x8_t vreinterpret_u8_f16(float16x4_t __p0) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t)(__p0);
return __ret;
}
-#else
-__ai float64x2_t vreinterpretq_f64_s16(int16x8_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t)(__p0);
+__ai uint8x8_t vreinterpret_u8_s32(int32x2_t __p0) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_p8(poly8x16_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint8x8_t vreinterpret_u8_s64(int64x1_t __p0) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t)(__p0);
return __ret;
}
-#else
-__ai float32x4_t vreinterpretq_f32_p8(poly8x16_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint8x8_t vreinterpret_u8_s16(int16x4_t __p0) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_p128(poly128_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint32x2_t vreinterpret_u32_p8(poly8x8_t __p0) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t)(__p0);
return __ret;
}
-#else
-__ai float32x4_t vreinterpretq_f32_p128(poly128_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint32x2_t vreinterpret_u32_p64(poly64x1_t __p0) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_p64(poly64x2_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint32x2_t vreinterpret_u32_p16(poly16x4_t __p0) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t)(__p0);
return __ret;
}
-#else
-__ai float32x4_t vreinterpretq_f32_p64(poly64x2_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint32x2_t vreinterpret_u32_u8(uint8x8_t __p0) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_p16(poly16x8_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint32x2_t vreinterpret_u32_u64(uint64x1_t __p0) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t)(__p0);
return __ret;
}
-#else
-__ai float32x4_t vreinterpretq_f32_p16(poly16x8_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint32x2_t vreinterpret_u32_u16(uint16x4_t __p0) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_u8(uint8x16_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint32x2_t vreinterpret_u32_s8(int8x8_t __p0) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t)(__p0);
return __ret;
}
-#else
-__ai float32x4_t vreinterpretq_f32_u8(uint8x16_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint32x2_t vreinterpret_u32_f64(float64x1_t __p0) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_u32(uint32x4_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint32x2_t vreinterpret_u32_f32(float32x2_t __p0) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t)(__p0);
return __ret;
}
-#else
-__ai float32x4_t vreinterpretq_f32_u32(uint32x4_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint32x2_t vreinterpret_u32_f16(float16x4_t __p0) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_u64(uint64x2_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint32x2_t vreinterpret_u32_s32(int32x2_t __p0) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t)(__p0);
return __ret;
}
-#else
-__ai float32x4_t vreinterpretq_f32_u64(uint64x2_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint32x2_t vreinterpret_u32_s64(int64x1_t __p0) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_u16(uint16x8_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint32x2_t vreinterpret_u32_s16(int16x4_t __p0) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t)(__p0);
return __ret;
}
-#else
-__ai float32x4_t vreinterpretq_f32_u16(uint16x8_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint64x1_t vreinterpret_u64_p8(poly8x8_t __p0) {
+ uint64x1_t __ret;
+ __ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_s8(int8x16_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint64x1_t vreinterpret_u64_p64(poly64x1_t __p0) {
+ uint64x1_t __ret;
+ __ret = (uint64x1_t)(__p0);
return __ret;
}
-#else
-__ai float32x4_t vreinterpretq_f32_s8(int8x16_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint64x1_t vreinterpret_u64_p16(poly16x4_t __p0) {
+ uint64x1_t __ret;
+ __ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_f64(float64x2_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint64x1_t vreinterpret_u64_u8(uint8x8_t __p0) {
+ uint64x1_t __ret;
+ __ret = (uint64x1_t)(__p0);
return __ret;
}
-#else
-__ai float32x4_t vreinterpretq_f32_f64(float64x2_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint64x1_t vreinterpret_u64_u32(uint32x2_t __p0) {
+ uint64x1_t __ret;
+ __ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_f16(float16x8_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint64x1_t vreinterpret_u64_u16(uint16x4_t __p0) {
+ uint64x1_t __ret;
+ __ret = (uint64x1_t)(__p0);
return __ret;
}
-#else
-__ai float32x4_t vreinterpretq_f32_f16(float16x8_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint64x1_t vreinterpret_u64_s8(int8x8_t __p0) {
+ uint64x1_t __ret;
+ __ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_s32(int32x4_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint64x1_t vreinterpret_u64_f64(float64x1_t __p0) {
+ uint64x1_t __ret;
+ __ret = (uint64x1_t)(__p0);
return __ret;
}
-#else
-__ai float32x4_t vreinterpretq_f32_s32(int32x4_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint64x1_t vreinterpret_u64_f32(float32x2_t __p0) {
+ uint64x1_t __ret;
+ __ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_s64(int64x2_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint64x1_t vreinterpret_u64_f16(float16x4_t __p0) {
+ uint64x1_t __ret;
+ __ret = (uint64x1_t)(__p0);
return __ret;
}
-#else
-__ai float32x4_t vreinterpretq_f32_s64(int64x2_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint64x1_t vreinterpret_u64_s32(int32x2_t __p0) {
+ uint64x1_t __ret;
+ __ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vreinterpretq_f32_s16(int16x8_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint64x1_t vreinterpret_u64_s64(int64x1_t __p0) {
+ uint64x1_t __ret;
+ __ret = (uint64x1_t)(__p0);
return __ret;
}
-#else
-__ai float32x4_t vreinterpretq_f32_s16(int16x8_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t)(__p0);
+__ai uint64x1_t vreinterpret_u64_s16(int16x4_t __p0) {
+ uint64x1_t __ret;
+ __ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_p8(poly8x16_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_p8(poly8x8_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#else
-__ai float16x8_t vreinterpretq_f16_p8(poly8x16_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_p64(poly64x1_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_p128(poly128_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_p16(poly16x4_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#else
-__ai float16x8_t vreinterpretq_f16_p128(poly128_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_u8(uint8x8_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_p64(poly64x2_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_u32(uint32x2_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#else
-__ai float16x8_t vreinterpretq_f16_p64(poly64x2_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_u64(uint64x1_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_p16(poly16x8_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_s8(int8x8_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#else
-__ai float16x8_t vreinterpretq_f16_p16(poly16x8_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_f64(float64x1_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_u8(uint8x16_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_f32(float32x2_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#else
-__ai float16x8_t vreinterpretq_f16_u8(uint8x16_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_f16(float16x4_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_u32(uint32x4_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_s32(int32x2_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#else
-__ai float16x8_t vreinterpretq_f16_u32(uint32x4_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_s64(int64x1_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_u64(uint64x2_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_s16(int16x4_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#else
-__ai float16x8_t vreinterpretq_f16_u64(uint64x2_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai int8x8_t vreinterpret_s8_p8(poly8x8_t __p0) {
+ int8x8_t __ret;
+ __ret = (int8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_u16(uint16x8_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai int8x8_t vreinterpret_s8_p64(poly64x1_t __p0) {
+ int8x8_t __ret;
+ __ret = (int8x8_t)(__p0);
return __ret;
}
-#else
-__ai float16x8_t vreinterpretq_f16_u16(uint16x8_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai int8x8_t vreinterpret_s8_p16(poly16x4_t __p0) {
+ int8x8_t __ret;
+ __ret = (int8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_s8(int8x16_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai int8x8_t vreinterpret_s8_u8(uint8x8_t __p0) {
+ int8x8_t __ret;
+ __ret = (int8x8_t)(__p0);
return __ret;
}
-#else
-__ai float16x8_t vreinterpretq_f16_s8(int8x16_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai int8x8_t vreinterpret_s8_u32(uint32x2_t __p0) {
+ int8x8_t __ret;
+ __ret = (int8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_f64(float64x2_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai int8x8_t vreinterpret_s8_u64(uint64x1_t __p0) {
+ int8x8_t __ret;
+ __ret = (int8x8_t)(__p0);
return __ret;
}
-#else
-__ai float16x8_t vreinterpretq_f16_f64(float64x2_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai int8x8_t vreinterpret_s8_u16(uint16x4_t __p0) {
+ int8x8_t __ret;
+ __ret = (int8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_f32(float32x4_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai int8x8_t vreinterpret_s8_f64(float64x1_t __p0) {
+ int8x8_t __ret;
+ __ret = (int8x8_t)(__p0);
return __ret;
}
-#else
-__ai float16x8_t vreinterpretq_f16_f32(float32x4_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai int8x8_t vreinterpret_s8_f32(float32x2_t __p0) {
+ int8x8_t __ret;
+ __ret = (int8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_s32(int32x4_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai int8x8_t vreinterpret_s8_f16(float16x4_t __p0) {
+ int8x8_t __ret;
+ __ret = (int8x8_t)(__p0);
return __ret;
}
-#else
-__ai float16x8_t vreinterpretq_f16_s32(int32x4_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai int8x8_t vreinterpret_s8_s32(int32x2_t __p0) {
+ int8x8_t __ret;
+ __ret = (int8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_s64(int64x2_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai int8x8_t vreinterpret_s8_s64(int64x1_t __p0) {
+ int8x8_t __ret;
+ __ret = (int8x8_t)(__p0);
return __ret;
}
-#else
-__ai float16x8_t vreinterpretq_f16_s64(int64x2_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai int8x8_t vreinterpret_s8_s16(int16x4_t __p0) {
+ int8x8_t __ret;
+ __ret = (int8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vreinterpretq_f16_s16(int16x8_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai float64x1_t vreinterpret_f64_p8(poly8x8_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t)(__p0);
return __ret;
}
-#else
-__ai float16x8_t vreinterpretq_f16_s16(int16x8_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t)(__p0);
+__ai float64x1_t vreinterpret_f64_p64(poly64x1_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_p8(poly8x16_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float64x1_t vreinterpret_f64_p16(poly16x4_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t)(__p0);
return __ret;
}
-#else
-__ai int32x4_t vreinterpretq_s32_p8(poly8x16_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float64x1_t vreinterpret_f64_u8(uint8x8_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_p128(poly128_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float64x1_t vreinterpret_f64_u32(uint32x2_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t)(__p0);
return __ret;
}
-#else
-__ai int32x4_t vreinterpretq_s32_p128(poly128_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float64x1_t vreinterpret_f64_u64(uint64x1_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_p64(poly64x2_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float64x1_t vreinterpret_f64_u16(uint16x4_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t)(__p0);
return __ret;
}
-#else
-__ai int32x4_t vreinterpretq_s32_p64(poly64x2_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float64x1_t vreinterpret_f64_s8(int8x8_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_p16(poly16x8_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float64x1_t vreinterpret_f64_f32(float32x2_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t)(__p0);
return __ret;
}
-#else
-__ai int32x4_t vreinterpretq_s32_p16(poly16x8_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float64x1_t vreinterpret_f64_f16(float16x4_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_u8(uint8x16_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float64x1_t vreinterpret_f64_s32(int32x2_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t)(__p0);
return __ret;
}
-#else
-__ai int32x4_t vreinterpretq_s32_u8(uint8x16_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float64x1_t vreinterpret_f64_s64(int64x1_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_u32(uint32x4_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float64x1_t vreinterpret_f64_s16(int16x4_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t)(__p0);
return __ret;
}
-#else
-__ai int32x4_t vreinterpretq_s32_u32(uint32x4_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float32x2_t vreinterpret_f32_p8(poly8x8_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_u64(uint64x2_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float32x2_t vreinterpret_f32_p64(poly64x1_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t)(__p0);
return __ret;
}
-#else
-__ai int32x4_t vreinterpretq_s32_u64(uint64x2_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float32x2_t vreinterpret_f32_p16(poly16x4_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_u16(uint16x8_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float32x2_t vreinterpret_f32_u8(uint8x8_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t)(__p0);
return __ret;
}
-#else
-__ai int32x4_t vreinterpretq_s32_u16(uint16x8_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float32x2_t vreinterpret_f32_u32(uint32x2_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_s8(int8x16_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float32x2_t vreinterpret_f32_u64(uint64x1_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t)(__p0);
return __ret;
}
-#else
-__ai int32x4_t vreinterpretq_s32_s8(int8x16_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float32x2_t vreinterpret_f32_u16(uint16x4_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_f64(float64x2_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float32x2_t vreinterpret_f32_s8(int8x8_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t)(__p0);
return __ret;
}
-#else
-__ai int32x4_t vreinterpretq_s32_f64(float64x2_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float32x2_t vreinterpret_f32_f64(float64x1_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_f32(float32x4_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float32x2_t vreinterpret_f32_f16(float16x4_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t)(__p0);
return __ret;
}
-#else
-__ai int32x4_t vreinterpretq_s32_f32(float32x4_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float32x2_t vreinterpret_f32_s32(int32x2_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_f16(float16x8_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float32x2_t vreinterpret_f32_s64(int64x1_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t)(__p0);
return __ret;
}
-#else
-__ai int32x4_t vreinterpretq_s32_f16(float16x8_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float32x2_t vreinterpret_f32_s16(int16x4_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_s64(int64x2_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float16x4_t vreinterpret_f16_p8(poly8x8_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#else
-__ai int32x4_t vreinterpretq_s32_s64(int64x2_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float16x4_t vreinterpret_f16_p64(poly64x1_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vreinterpretq_s32_s16(int16x8_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float16x4_t vreinterpret_f16_p16(poly16x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#else
-__ai int32x4_t vreinterpretq_s32_s16(int16x8_t __p0) {
- int32x4_t __ret;
- __ret = (int32x4_t)(__p0);
+__ai float16x4_t vreinterpret_f16_u8(uint8x8_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_p8(poly8x16_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_u32(uint32x2_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#else
-__ai int64x2_t vreinterpretq_s64_p8(poly8x16_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_u64(uint64x1_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_p128(poly128_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_u16(uint16x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#else
-__ai int64x2_t vreinterpretq_s64_p128(poly128_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_s8(int8x8_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_p64(poly64x2_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_f64(float64x1_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#else
-__ai int64x2_t vreinterpretq_s64_p64(poly64x2_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_f32(float32x2_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_p16(poly16x8_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_s32(int32x2_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#else
-__ai int64x2_t vreinterpretq_s64_p16(poly16x8_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_s64(int64x1_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_u8(uint8x16_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_s16(int16x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#else
-__ai int64x2_t vreinterpretq_s64_u8(uint8x16_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int32x2_t vreinterpret_s32_p8(poly8x8_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_u32(uint32x4_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int32x2_t vreinterpret_s32_p64(poly64x1_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
return __ret;
}
-#else
-__ai int64x2_t vreinterpretq_s64_u32(uint32x4_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int32x2_t vreinterpret_s32_p16(poly16x4_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_u64(uint64x2_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int32x2_t vreinterpret_s32_u8(uint8x8_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
return __ret;
}
-#else
-__ai int64x2_t vreinterpretq_s64_u64(uint64x2_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int32x2_t vreinterpret_s32_u32(uint32x2_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_u16(uint16x8_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int32x2_t vreinterpret_s32_u64(uint64x1_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
return __ret;
}
-#else
-__ai int64x2_t vreinterpretq_s64_u16(uint16x8_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int32x2_t vreinterpret_s32_u16(uint16x4_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_s8(int8x16_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int32x2_t vreinterpret_s32_s8(int8x8_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
return __ret;
}
-#else
-__ai int64x2_t vreinterpretq_s64_s8(int8x16_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int32x2_t vreinterpret_s32_f64(float64x1_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_f64(float64x2_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int32x2_t vreinterpret_s32_f32(float32x2_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
return __ret;
}
-#else
-__ai int64x2_t vreinterpretq_s64_f64(float64x2_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int32x2_t vreinterpret_s32_f16(float16x4_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_f32(float32x4_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int32x2_t vreinterpret_s32_s64(int64x1_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
return __ret;
}
-#else
-__ai int64x2_t vreinterpretq_s64_f32(float32x4_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int32x2_t vreinterpret_s32_s16(int16x4_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_f16(float16x8_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int64x1_t vreinterpret_s64_p8(poly8x8_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
return __ret;
}
-#else
-__ai int64x2_t vreinterpretq_s64_f16(float16x8_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int64x1_t vreinterpret_s64_p64(poly64x1_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_s32(int32x4_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int64x1_t vreinterpret_s64_p16(poly16x4_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
return __ret;
}
-#else
-__ai int64x2_t vreinterpretq_s64_s32(int32x4_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int64x1_t vreinterpret_s64_u8(uint8x8_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vreinterpretq_s64_s16(int16x8_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int64x1_t vreinterpret_s64_u32(uint32x2_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
return __ret;
}
-#else
-__ai int64x2_t vreinterpretq_s64_s16(int16x8_t __p0) {
- int64x2_t __ret;
- __ret = (int64x2_t)(__p0);
+__ai int64x1_t vreinterpret_s64_u64(uint64x1_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_p8(poly8x16_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int64x1_t vreinterpret_s64_u16(uint16x4_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
return __ret;
}
-#else
-__ai int16x8_t vreinterpretq_s16_p8(poly8x16_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int64x1_t vreinterpret_s64_s8(int8x8_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_p128(poly128_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int64x1_t vreinterpret_s64_f64(float64x1_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
return __ret;
}
-#else
-__ai int16x8_t vreinterpretq_s16_p128(poly128_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int64x1_t vreinterpret_s64_f32(float32x2_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_p64(poly64x2_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int64x1_t vreinterpret_s64_f16(float16x4_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
return __ret;
}
-#else
-__ai int16x8_t vreinterpretq_s16_p64(poly64x2_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int64x1_t vreinterpret_s64_s32(int32x2_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_p16(poly16x8_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int64x1_t vreinterpret_s64_s16(int16x4_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
return __ret;
}
-#else
-__ai int16x8_t vreinterpretq_s16_p16(poly16x8_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int16x4_t vreinterpret_s16_p8(poly8x8_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_u8(uint8x16_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int16x4_t vreinterpret_s16_p64(poly64x1_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
-#else
-__ai int16x8_t vreinterpretq_s16_u8(uint8x16_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int16x4_t vreinterpret_s16_p16(poly16x4_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_u32(uint32x4_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int16x4_t vreinterpret_s16_u8(uint8x8_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
-#else
-__ai int16x8_t vreinterpretq_s16_u32(uint32x4_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int16x4_t vreinterpret_s16_u32(uint32x2_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_u64(uint64x2_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int16x4_t vreinterpret_s16_u64(uint64x1_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
-#else
-__ai int16x8_t vreinterpretq_s16_u64(uint64x2_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int16x4_t vreinterpret_s16_u16(uint16x4_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_u16(uint16x8_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int16x4_t vreinterpret_s16_s8(int8x8_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
-#else
-__ai int16x8_t vreinterpretq_s16_u16(uint16x8_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int16x4_t vreinterpret_s16_f64(float64x1_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_s8(int8x16_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int16x4_t vreinterpret_s16_f32(float32x2_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
-#else
-__ai int16x8_t vreinterpretq_s16_s8(int8x16_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int16x4_t vreinterpret_s16_f16(float16x4_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_f64(float64x2_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int16x4_t vreinterpret_s16_s32(int32x2_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
-#else
-__ai int16x8_t vreinterpretq_s16_f64(float64x2_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai int16x4_t vreinterpret_s16_s64(int64x1_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
#endif
-
+#if __ARM_ARCH >= 8 && defined(__aarch64__) && defined(__ARM_FEATURE_DIRECTED_ROUNDING)
#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_f32(float32x4_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai float64x2_t vrndq_f64(float64x2_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vrndq_v((int8x16_t)__p0, 42);
return __ret;
}
#else
-__ai int16x8_t vreinterpretq_s16_f32(float32x4_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai float64x2_t vrndq_f64(float64x2_t __p0) {
+ float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vrndq_v((int8x16_t)__rev0, 42);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
#endif
+__ai float64x1_t vrnd_f64(float64x1_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t) __builtin_neon_vrnd_v((int8x8_t)__p0, 10);
+ return __ret;
+}
#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_f16(float16x8_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai float64x2_t vrndaq_f64(float64x2_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vrndaq_v((int8x16_t)__p0, 42);
return __ret;
}
#else
-__ai int16x8_t vreinterpretq_s16_f16(float16x8_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai float64x2_t vrndaq_f64(float64x2_t __p0) {
+ float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vrndaq_v((int8x16_t)__rev0, 42);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
#endif
+__ai float64x1_t vrnda_f64(float64x1_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t) __builtin_neon_vrnda_v((int8x8_t)__p0, 10);
+ return __ret;
+}
#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_s32(int32x4_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai float64x2_t vrndiq_f64(float64x2_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vrndiq_v((int8x16_t)__p0, 42);
return __ret;
}
#else
-__ai int16x8_t vreinterpretq_s16_s32(int32x4_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai float64x2_t vrndiq_f64(float64x2_t __p0) {
+ float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vrndiq_v((int8x16_t)__rev0, 42);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
#endif
+__ai float64x1_t vrndi_f64(float64x1_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t) __builtin_neon_vrndi_v((int8x8_t)__p0, 10);
+ return __ret;
+}
#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vreinterpretq_s16_s64(int64x2_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai float64x2_t vrndmq_f64(float64x2_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vrndmq_v((int8x16_t)__p0, 42);
return __ret;
}
#else
-__ai int16x8_t vreinterpretq_s16_s64(int64x2_t __p0) {
- int16x8_t __ret;
- __ret = (int16x8_t)(__p0);
+__ai float64x2_t vrndmq_f64(float64x2_t __p0) {
+ float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vrndmq_v((int8x16_t)__rev0, 42);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
#endif
+__ai float64x1_t vrndm_f64(float64x1_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t) __builtin_neon_vrndm_v((int8x8_t)__p0, 10);
+ return __ret;
+}
#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_p8(poly8x8_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai float64x2_t vrndnq_f64(float64x2_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vrndnq_v((int8x16_t)__p0, 42);
return __ret;
}
#else
-__ai uint8x8_t vreinterpret_u8_p8(poly8x8_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai float64x2_t vrndnq_f64(float64x2_t __p0) {
+ float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vrndnq_v((int8x16_t)__rev0, 42);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
#endif
+__ai float64x1_t vrndn_f64(float64x1_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t) __builtin_neon_vrndn_v((int8x8_t)__p0, 10);
+ return __ret;
+}
#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_p64(poly64x1_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai float64x2_t vrndpq_f64(float64x2_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vrndpq_v((int8x16_t)__p0, 42);
return __ret;
}
#else
-__ai uint8x8_t vreinterpret_u8_p64(poly64x1_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai float64x2_t vrndpq_f64(float64x2_t __p0) {
+ float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vrndpq_v((int8x16_t)__rev0, 42);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
#endif
+__ai float64x1_t vrndp_f64(float64x1_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t) __builtin_neon_vrndp_v((int8x8_t)__p0, 10);
+ return __ret;
+}
#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_p16(poly16x4_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai float64x2_t vrndxq_f64(float64x2_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vrndxq_v((int8x16_t)__p0, 42);
return __ret;
}
#else
-__ai uint8x8_t vreinterpret_u8_p16(poly16x4_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai float64x2_t vrndxq_f64(float64x2_t __p0) {
+ float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vrndxq_v((int8x16_t)__rev0, 42);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
#endif
+__ai float64x1_t vrndx_f64(float64x1_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t) __builtin_neon_vrndx_v((int8x8_t)__p0, 10);
+ return __ret;
+}
+#endif
+#if __ARM_ARCH >= 8 && defined(__aarch64__) && defined(__ARM_FEATURE_NUMERIC_MAXMIN)
#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_u32(uint32x2_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai float64x2_t vmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
return __ret;
}
#else
-__ai uint8x8_t vreinterpret_u8_u32(uint32x2_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai float64x2_t vmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
+ float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
#endif
+__ai float64x1_t vmaxnm_f64(float64x1_t __p0, float64x1_t __p1) {
+ float64x1_t __ret;
+ __ret = (float64x1_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
+ return __ret;
+}
#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_u64(uint64x1_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai float64x2_t vminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
return __ret;
}
#else
-__ai uint8x8_t vreinterpret_u8_u64(uint64x1_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai float64x2_t vminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
+ float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
+ float64x2_t __ret;
+ __ret = (float64x2_t) __builtin_neon_vminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_u16(uint16x4_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai float64x1_t vminnm_f64(float64x1_t __p0, float64x1_t __p1) {
+ float64x1_t __ret;
+ __ret = (float64x1_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
return __ret;
}
-#else
-__ai uint8x8_t vreinterpret_u8_u16(uint16x4_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+#endif
+#if defined(__ARM_FEATURE_BF16) && !defined(__aarch64__)
+__ai poly8x8_t vreinterpret_p8_bf16(bfloat16x4_t __p0) {
+ poly8x8_t __ret;
+ __ret = (poly8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_s8(int8x8_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai poly64x1_t vreinterpret_p64_bf16(bfloat16x4_t __p0) {
+ poly64x1_t __ret;
+ __ret = (poly64x1_t)(__p0);
return __ret;
}
-#else
-__ai uint8x8_t vreinterpret_u8_s8(int8x8_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai poly16x4_t vreinterpret_p16_bf16(bfloat16x4_t __p0) {
+ poly16x4_t __ret;
+ __ret = (poly16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_f64(float64x1_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai poly8x16_t vreinterpretq_p8_bf16(bfloat16x8_t __p0) {
+ poly8x16_t __ret;
+ __ret = (poly8x16_t)(__p0);
return __ret;
}
-#else
-__ai uint8x8_t vreinterpret_u8_f64(float64x1_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai poly64x2_t vreinterpretq_p64_bf16(bfloat16x8_t __p0) {
+ poly64x2_t __ret;
+ __ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_f32(float32x2_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai poly16x8_t vreinterpretq_p16_bf16(bfloat16x8_t __p0) {
+ poly16x8_t __ret;
+ __ret = (poly16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint8x8_t vreinterpret_u8_f32(float32x2_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai uint8x16_t vreinterpretq_u8_bf16(bfloat16x8_t __p0) {
+ uint8x16_t __ret;
+ __ret = (uint8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_f16(float16x4_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai uint32x4_t vreinterpretq_u32_bf16(bfloat16x8_t __p0) {
+ uint32x4_t __ret;
+ __ret = (uint32x4_t)(__p0);
return __ret;
}
-#else
-__ai uint8x8_t vreinterpret_u8_f16(float16x4_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai uint64x2_t vreinterpretq_u64_bf16(bfloat16x8_t __p0) {
+ uint64x2_t __ret;
+ __ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_s32(int32x2_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai uint16x8_t vreinterpretq_u16_bf16(bfloat16x8_t __p0) {
+ uint16x8_t __ret;
+ __ret = (uint16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint8x8_t vreinterpret_u8_s32(int32x2_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai int8x16_t vreinterpretq_s8_bf16(bfloat16x8_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_s64(int64x1_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai float32x4_t vreinterpretq_f32_bf16(bfloat16x8_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
return __ret;
}
-#else
-__ai uint8x8_t vreinterpret_u8_s64(int64x1_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai float16x8_t vreinterpretq_f16_bf16(bfloat16x8_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vreinterpret_u8_s16(int16x4_t __p0) {
- uint8x8_t __ret;
- __ret = (uint8x8_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_bf16(bfloat16x8_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#else
-__ai uint8x8_t vreinterpret_u8_s16(int16x4_t __p0) {
+__ai int64x2_t vreinterpretq_s64_bf16(bfloat16x8_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
+ return __ret;
+}
+__ai int16x8_t vreinterpretq_s16_bf16(bfloat16x8_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
+ return __ret;
+}
+__ai uint8x8_t vreinterpret_u8_bf16(bfloat16x4_t __p0) {
uint8x8_t __ret;
__ret = (uint8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_p8(poly8x8_t __p0) {
+__ai uint32x2_t vreinterpret_u32_bf16(bfloat16x4_t __p0) {
uint32x2_t __ret;
__ret = (uint32x2_t)(__p0);
return __ret;
}
-#else
-__ai uint32x2_t vreinterpret_u32_p8(poly8x8_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai uint64x1_t vreinterpret_u64_bf16(bfloat16x4_t __p0) {
+ uint64x1_t __ret;
+ __ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_p64(poly64x1_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai uint16x4_t vreinterpret_u16_bf16(bfloat16x4_t __p0) {
+ uint16x4_t __ret;
+ __ret = (uint16x4_t)(__p0);
return __ret;
}
-#else
-__ai uint32x2_t vreinterpret_u32_p64(poly64x1_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai int8x8_t vreinterpret_s8_bf16(bfloat16x4_t __p0) {
+ int8x8_t __ret;
+ __ret = (int8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_p16(poly16x4_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai float32x2_t vreinterpret_f32_bf16(bfloat16x4_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t)(__p0);
return __ret;
}
-#else
-__ai uint32x2_t vreinterpret_u32_p16(poly16x4_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai float16x4_t vreinterpret_f16_bf16(bfloat16x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_u8(uint8x8_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai int32x2_t vreinterpret_s32_bf16(bfloat16x4_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
return __ret;
}
-#else
-__ai uint32x2_t vreinterpret_u32_u8(uint8x8_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai int64x1_t vreinterpret_s64_bf16(bfloat16x4_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_u64(uint64x1_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai int16x4_t vreinterpret_s16_bf16(bfloat16x4_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
-#else
-__ai uint32x2_t vreinterpret_u32_u64(uint64x1_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_p8(poly8x16_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_u16(uint16x4_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_p64(poly64x2_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint32x2_t vreinterpret_u32_u16(uint16x4_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_p16(poly16x8_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_s8(int8x8_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_u8(uint8x16_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint32x2_t vreinterpret_u32_s8(int8x8_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_u32(uint32x4_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_f64(float64x1_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_u64(uint64x2_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint32x2_t vreinterpret_u32_f64(float64x1_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_u16(uint16x8_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_f32(float32x2_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_s8(int8x16_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint32x2_t vreinterpret_u32_f32(float32x2_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_f32(float32x4_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_f16(float16x4_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_f16(float16x8_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint32x2_t vreinterpret_u32_f16(float16x4_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_s32(int32x4_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_s32(int32x2_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_s64(int64x2_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint32x2_t vreinterpret_u32_s32(int32x2_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_s16(int16x8_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_s64(int64x1_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_p8(poly8x8_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#else
-__ai uint32x2_t vreinterpret_u32_s64(int64x1_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_p64(poly64x1_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vreinterpret_u32_s16(int16x4_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_p16(poly16x4_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#else
-__ai uint32x2_t vreinterpret_u32_s16(int16x4_t __p0) {
- uint32x2_t __ret;
- __ret = (uint32x2_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_u8(uint8x8_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_p8(poly8x8_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_u32(uint32x2_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#else
-__ai uint64x1_t vreinterpret_u64_p8(poly8x8_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_u64(uint64x1_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_p64(poly64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_u16(uint16x4_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#else
-__ai uint64x1_t vreinterpret_u64_p64(poly64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_s8(int8x8_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_p16(poly16x4_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_f32(float32x2_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#else
-__ai uint64x1_t vreinterpret_u64_p16(poly16x4_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_f16(float16x4_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_u8(uint8x8_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_s32(int32x2_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#else
-__ai uint64x1_t vreinterpret_u64_u8(uint8x8_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_s64(int64x1_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
+ return __ret;
+}
+__ai bfloat16x4_t vreinterpret_bf16_s16(int16x4_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_u32(uint32x2_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+#if defined(__ARM_FEATURE_BF16) && defined(__aarch64__)
+__ai poly8x8_t vreinterpret_p8_bf16(bfloat16x4_t __p0) {
+ poly8x8_t __ret;
+ __ret = (poly8x8_t)(__p0);
return __ret;
}
-#else
-__ai uint64x1_t vreinterpret_u64_u32(uint32x2_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai poly64x1_t vreinterpret_p64_bf16(bfloat16x4_t __p0) {
+ poly64x1_t __ret;
+ __ret = (poly64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_u16(uint16x4_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai poly16x4_t vreinterpret_p16_bf16(bfloat16x4_t __p0) {
+ poly16x4_t __ret;
+ __ret = (poly16x4_t)(__p0);
return __ret;
}
-#else
-__ai uint64x1_t vreinterpret_u64_u16(uint16x4_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai poly8x16_t vreinterpretq_p8_bf16(bfloat16x8_t __p0) {
+ poly8x16_t __ret;
+ __ret = (poly8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_s8(int8x8_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai poly128_t vreinterpretq_p128_bf16(bfloat16x8_t __p0) {
+ poly128_t __ret;
+ __ret = (poly128_t)(__p0);
return __ret;
}
-#else
-__ai uint64x1_t vreinterpret_u64_s8(int8x8_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai poly64x2_t vreinterpretq_p64_bf16(bfloat16x8_t __p0) {
+ poly64x2_t __ret;
+ __ret = (poly64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_f64(float64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai poly16x8_t vreinterpretq_p16_bf16(bfloat16x8_t __p0) {
+ poly16x8_t __ret;
+ __ret = (poly16x8_t)(__p0);
+ return __ret;
+}
+__ai uint8x16_t vreinterpretq_u8_bf16(bfloat16x8_t __p0) {
+ uint8x16_t __ret;
+ __ret = (uint8x16_t)(__p0);
+ return __ret;
+}
+__ai uint32x4_t vreinterpretq_u32_bf16(bfloat16x8_t __p0) {
+ uint32x4_t __ret;
+ __ret = (uint32x4_t)(__p0);
return __ret;
}
-#else
-__ai uint64x1_t vreinterpret_u64_f64(float64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai uint64x2_t vreinterpretq_u64_bf16(bfloat16x8_t __p0) {
+ uint64x2_t __ret;
+ __ret = (uint64x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_f32(float32x2_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai uint16x8_t vreinterpretq_u16_bf16(bfloat16x8_t __p0) {
+ uint16x8_t __ret;
+ __ret = (uint16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint64x1_t vreinterpret_u64_f32(float32x2_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai int8x16_t vreinterpretq_s8_bf16(bfloat16x8_t __p0) {
+ int8x16_t __ret;
+ __ret = (int8x16_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_f16(float16x4_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai float64x2_t vreinterpretq_f64_bf16(bfloat16x8_t __p0) {
+ float64x2_t __ret;
+ __ret = (float64x2_t)(__p0);
return __ret;
}
-#else
-__ai uint64x1_t vreinterpret_u64_f16(float16x4_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai float32x4_t vreinterpretq_f32_bf16(bfloat16x8_t __p0) {
+ float32x4_t __ret;
+ __ret = (float32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_s32(int32x2_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai float16x8_t vreinterpretq_f16_bf16(bfloat16x8_t __p0) {
+ float16x8_t __ret;
+ __ret = (float16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint64x1_t vreinterpret_u64_s32(int32x2_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai int32x4_t vreinterpretq_s32_bf16(bfloat16x8_t __p0) {
+ int32x4_t __ret;
+ __ret = (int32x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_s64(int64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai int64x2_t vreinterpretq_s64_bf16(bfloat16x8_t __p0) {
+ int64x2_t __ret;
+ __ret = (int64x2_t)(__p0);
return __ret;
}
-#else
-__ai uint64x1_t vreinterpret_u64_s64(int64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai int16x8_t vreinterpretq_s16_bf16(bfloat16x8_t __p0) {
+ int16x8_t __ret;
+ __ret = (int16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vreinterpret_u64_s16(int16x4_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0);
+__ai uint8x8_t vreinterpret_u8_bf16(bfloat16x4_t __p0) {
+ uint8x8_t __ret;
+ __ret = (uint8x8_t)(__p0);
return __ret;
}
-#else
-__ai uint64x1_t vreinterpret_u64_s16(int16x4_t __p0) {
+__ai uint32x2_t vreinterpret_u32_bf16(bfloat16x4_t __p0) {
+ uint32x2_t __ret;
+ __ret = (uint32x2_t)(__p0);
+ return __ret;
+}
+__ai uint64x1_t vreinterpret_u64_bf16(bfloat16x4_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_p8(poly8x8_t __p0) {
+__ai uint16x4_t vreinterpret_u16_bf16(bfloat16x4_t __p0) {
uint16x4_t __ret;
__ret = (uint16x4_t)(__p0);
return __ret;
}
-#else
-__ai uint16x4_t vreinterpret_u16_p8(poly8x8_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai int8x8_t vreinterpret_s8_bf16(bfloat16x4_t __p0) {
+ int8x8_t __ret;
+ __ret = (int8x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_p64(poly64x1_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai float64x1_t vreinterpret_f64_bf16(bfloat16x4_t __p0) {
+ float64x1_t __ret;
+ __ret = (float64x1_t)(__p0);
return __ret;
}
-#else
-__ai uint16x4_t vreinterpret_u16_p64(poly64x1_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai float32x2_t vreinterpret_f32_bf16(bfloat16x4_t __p0) {
+ float32x2_t __ret;
+ __ret = (float32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_p16(poly16x4_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai float16x4_t vreinterpret_f16_bf16(bfloat16x4_t __p0) {
+ float16x4_t __ret;
+ __ret = (float16x4_t)(__p0);
return __ret;
}
-#else
-__ai uint16x4_t vreinterpret_u16_p16(poly16x4_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai int32x2_t vreinterpret_s32_bf16(bfloat16x4_t __p0) {
+ int32x2_t __ret;
+ __ret = (int32x2_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_u8(uint8x8_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai int64x1_t vreinterpret_s64_bf16(bfloat16x4_t __p0) {
+ int64x1_t __ret;
+ __ret = (int64x1_t)(__p0);
return __ret;
}
-#else
-__ai uint16x4_t vreinterpret_u16_u8(uint8x8_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai int16x4_t vreinterpret_s16_bf16(bfloat16x4_t __p0) {
+ int16x4_t __ret;
+ __ret = (int16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_u32(uint32x2_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_p8(poly8x16_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x4_t vreinterpret_u16_u32(uint32x2_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_p128(poly128_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_u64(uint64x1_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_p64(poly64x2_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x4_t vreinterpret_u16_u64(uint64x1_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_p16(poly16x8_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_s8(int8x8_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_u8(uint8x16_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x4_t vreinterpret_u16_s8(int8x8_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_u32(uint32x4_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_f64(float64x1_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_u64(uint64x2_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x4_t vreinterpret_u16_f64(float64x1_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_u16(uint16x8_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_f32(float32x2_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_s8(int8x16_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x4_t vreinterpret_u16_f32(float32x2_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_f64(float64x2_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_f16(float16x4_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_f32(float32x4_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x4_t vreinterpret_u16_f16(float16x4_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_f16(float16x8_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_s32(int32x2_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_s32(int32x4_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x4_t vreinterpret_u16_s32(int32x2_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_s64(int64x2_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_s64(int64x1_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x8_t vreinterpretq_bf16_s16(int16x8_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t)(__p0);
return __ret;
}
-#else
-__ai uint16x4_t vreinterpret_u16_s64(int64x1_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_p8(poly8x8_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vreinterpret_u16_s16(int16x4_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_p64(poly64x1_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#else
-__ai uint16x4_t vreinterpret_u16_s16(int16x4_t __p0) {
- uint16x4_t __ret;
- __ret = (uint16x4_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_p16(poly16x4_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_p8(poly8x8_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_u8(uint8x8_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#else
-__ai int8x8_t vreinterpret_s8_p8(poly8x8_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_u32(uint32x2_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_p64(poly64x1_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_u64(uint64x1_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#else
-__ai int8x8_t vreinterpret_s8_p64(poly64x1_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_u16(uint16x4_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_p16(poly16x4_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_s8(int8x8_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#else
-__ai int8x8_t vreinterpret_s8_p16(poly16x4_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_f64(float64x1_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_u8(uint8x8_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_f32(float32x2_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#else
-__ai int8x8_t vreinterpret_s8_u8(uint8x8_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_f16(float16x4_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_u32(uint32x2_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_s32(int32x2_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#else
-__ai int8x8_t vreinterpret_s8_u32(uint32x2_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_s64(int64x1_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_u64(uint64x1_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai bfloat16x4_t vreinterpret_bf16_s16(int16x4_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t)(__p0);
return __ret;
}
+#endif
+#if defined(__ARM_FEATURE_BF16_VECTOR_ARITHMETIC)
+#ifdef __LITTLE_ENDIAN__
+#define splatq_lane_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4_t __s0 = __p0; \
+ bfloat16x8_t __ret; \
+ __ret = (bfloat16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 11); \
+ __ret; \
+})
#else
-__ai int8x8_t vreinterpret_s8_u64(uint64x1_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
+#define splatq_lane_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4_t __s0 = __p0; \
+ bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ bfloat16x8_t __ret; \
+ __ret = (bfloat16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 11); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_lane_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4_t __s0 = __p0; \
+ bfloat16x8_t __ret; \
+ __ret = (bfloat16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 11); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_u16(uint16x4_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
+#define splat_lane_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4_t __s0 = __p0; \
+ bfloat16x4_t __ret; \
+ __ret = (bfloat16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 11); \
+ __ret; \
+})
#else
-__ai int8x8_t vreinterpret_s8_u16(uint16x4_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
+#define splat_lane_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4_t __s0 = __p0; \
+ bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ bfloat16x4_t __ret; \
+ __ret = (bfloat16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 11); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_lane_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4_t __s0 = __p0; \
+ bfloat16x4_t __ret; \
+ __ret = (bfloat16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 11); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_f64(float64x1_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
+#define splatq_laneq_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8_t __s0 = __p0; \
+ bfloat16x8_t __ret; \
+ __ret = (bfloat16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 43); \
+ __ret; \
+})
#else
-__ai int8x8_t vreinterpret_s8_f64(float64x1_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
+#define splatq_laneq_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8_t __s0 = __p0; \
+ bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x8_t __ret; \
+ __ret = (bfloat16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 43); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splatq_laneq_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8_t __s0 = __p0; \
+ bfloat16x8_t __ret; \
+ __ret = (bfloat16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 43); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_f32(float32x2_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
+#define splat_laneq_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8_t __s0 = __p0; \
+ bfloat16x4_t __ret; \
+ __ret = (bfloat16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 43); \
+ __ret; \
+})
#else
-__ai int8x8_t vreinterpret_s8_f32(float32x2_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
+#define splat_laneq_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8_t __s0 = __p0; \
+ bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x4_t __ret; \
+ __ret = (bfloat16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 43); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_splat_laneq_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8_t __s0 = __p0; \
+ bfloat16x4_t __ret; \
+ __ret = (bfloat16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 43); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_f16(float16x4_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai float32x4_t vbfdotq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vbfdotq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
return __ret;
}
#else
-__ai int8x8_t vreinterpret_s8_f16(float16x4_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai float32x4_t vbfdotq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vbfdotq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai float32x4_t __noswap_vbfdotq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vbfdotq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_s32(int32x2_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai float32x2_t vbfdot_f32(float32x2_t __p0, bfloat16x4_t __p1, bfloat16x4_t __p2) {
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vbfdot_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
return __ret;
}
#else
-__ai int8x8_t vreinterpret_s8_s32(int32x2_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
+__ai float32x2_t vbfdot_f32(float32x2_t __p0, bfloat16x4_t __p1, bfloat16x4_t __p2) {
+ float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ bfloat16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vbfdot_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
+ return __ret;
+}
+__ai float32x2_t __noswap_vbfdot_f32(float32x2_t __p0, bfloat16x4_t __p1, bfloat16x4_t __p2) {
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vbfdot_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_s64(int64x1_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
+#define vbfdotq_lane_f32(__p0_142, __p1_142, __p2_142, __p3_142) __extension__ ({ \
+ float32x4_t __s0_142 = __p0_142; \
+ bfloat16x8_t __s1_142 = __p1_142; \
+ bfloat16x4_t __s2_142 = __p2_142; \
+ float32x4_t __ret_142; \
+bfloat16x4_t __reint_142 = __s2_142; \
+float32x4_t __reint1_142 = splatq_lane_f32(*(float32x2_t *) &__reint_142, __p3_142); \
+ __ret_142 = vbfdotq_f32(__s0_142, __s1_142, *(bfloat16x8_t *) &__reint1_142); \
+ __ret_142; \
+})
#else
-__ai int8x8_t vreinterpret_s8_s64(int64x1_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
+#define vbfdotq_lane_f32(__p0_143, __p1_143, __p2_143, __p3_143) __extension__ ({ \
+ float32x4_t __s0_143 = __p0_143; \
+ bfloat16x8_t __s1_143 = __p1_143; \
+ bfloat16x4_t __s2_143 = __p2_143; \
+ float32x4_t __rev0_143; __rev0_143 = __builtin_shufflevector(__s0_143, __s0_143, 3, 2, 1, 0); \
+ bfloat16x8_t __rev1_143; __rev1_143 = __builtin_shufflevector(__s1_143, __s1_143, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x4_t __rev2_143; __rev2_143 = __builtin_shufflevector(__s2_143, __s2_143, 3, 2, 1, 0); \
+ float32x4_t __ret_143; \
+bfloat16x4_t __reint_143 = __rev2_143; \
+float32x4_t __reint1_143 = __noswap_splatq_lane_f32(*(float32x2_t *) &__reint_143, __p3_143); \
+ __ret_143 = __noswap_vbfdotq_f32(__rev0_143, __rev1_143, *(bfloat16x8_t *) &__reint1_143); \
+ __ret_143 = __builtin_shufflevector(__ret_143, __ret_143, 3, 2, 1, 0); \
+ __ret_143; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vreinterpret_s8_s16(int16x4_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
+#define vbfdot_lane_f32(__p0_144, __p1_144, __p2_144, __p3_144) __extension__ ({ \
+ float32x2_t __s0_144 = __p0_144; \
+ bfloat16x4_t __s1_144 = __p1_144; \
+ bfloat16x4_t __s2_144 = __p2_144; \
+ float32x2_t __ret_144; \
+bfloat16x4_t __reint_144 = __s2_144; \
+float32x2_t __reint1_144 = splat_lane_f32(*(float32x2_t *) &__reint_144, __p3_144); \
+ __ret_144 = vbfdot_f32(__s0_144, __s1_144, *(bfloat16x4_t *) &__reint1_144); \
+ __ret_144; \
+})
#else
-__ai int8x8_t vreinterpret_s8_s16(int16x4_t __p0) {
- int8x8_t __ret;
- __ret = (int8x8_t)(__p0);
- return __ret;
-}
+#define vbfdot_lane_f32(__p0_145, __p1_145, __p2_145, __p3_145) __extension__ ({ \
+ float32x2_t __s0_145 = __p0_145; \
+ bfloat16x4_t __s1_145 = __p1_145; \
+ bfloat16x4_t __s2_145 = __p2_145; \
+ float32x2_t __rev0_145; __rev0_145 = __builtin_shufflevector(__s0_145, __s0_145, 1, 0); \
+ bfloat16x4_t __rev1_145; __rev1_145 = __builtin_shufflevector(__s1_145, __s1_145, 3, 2, 1, 0); \
+ bfloat16x4_t __rev2_145; __rev2_145 = __builtin_shufflevector(__s2_145, __s2_145, 3, 2, 1, 0); \
+ float32x2_t __ret_145; \
+bfloat16x4_t __reint_145 = __rev2_145; \
+float32x2_t __reint1_145 = __noswap_splat_lane_f32(*(float32x2_t *) &__reint_145, __p3_145); \
+ __ret_145 = __noswap_vbfdot_f32(__rev0_145, __rev1_145, *(bfloat16x4_t *) &__reint1_145); \
+ __ret_145 = __builtin_shufflevector(__ret_145, __ret_145, 1, 0); \
+ __ret_145; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vreinterpret_f64_p8(poly8x8_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
+#define vbfdotq_laneq_f32(__p0_146, __p1_146, __p2_146, __p3_146) __extension__ ({ \
+ float32x4_t __s0_146 = __p0_146; \
+ bfloat16x8_t __s1_146 = __p1_146; \
+ bfloat16x8_t __s2_146 = __p2_146; \
+ float32x4_t __ret_146; \
+bfloat16x8_t __reint_146 = __s2_146; \
+float32x4_t __reint1_146 = splatq_laneq_f32(*(float32x4_t *) &__reint_146, __p3_146); \
+ __ret_146 = vbfdotq_f32(__s0_146, __s1_146, *(bfloat16x8_t *) &__reint1_146); \
+ __ret_146; \
+})
#else
-__ai float64x1_t vreinterpret_f64_p8(poly8x8_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
+#define vbfdotq_laneq_f32(__p0_147, __p1_147, __p2_147, __p3_147) __extension__ ({ \
+ float32x4_t __s0_147 = __p0_147; \
+ bfloat16x8_t __s1_147 = __p1_147; \
+ bfloat16x8_t __s2_147 = __p2_147; \
+ float32x4_t __rev0_147; __rev0_147 = __builtin_shufflevector(__s0_147, __s0_147, 3, 2, 1, 0); \
+ bfloat16x8_t __rev1_147; __rev1_147 = __builtin_shufflevector(__s1_147, __s1_147, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x8_t __rev2_147; __rev2_147 = __builtin_shufflevector(__s2_147, __s2_147, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float32x4_t __ret_147; \
+bfloat16x8_t __reint_147 = __rev2_147; \
+float32x4_t __reint1_147 = __noswap_splatq_laneq_f32(*(float32x4_t *) &__reint_147, __p3_147); \
+ __ret_147 = __noswap_vbfdotq_f32(__rev0_147, __rev1_147, *(bfloat16x8_t *) &__reint1_147); \
+ __ret_147 = __builtin_shufflevector(__ret_147, __ret_147, 3, 2, 1, 0); \
+ __ret_147; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vreinterpret_f64_p64(poly64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
+#define vbfdot_laneq_f32(__p0_148, __p1_148, __p2_148, __p3_148) __extension__ ({ \
+ float32x2_t __s0_148 = __p0_148; \
+ bfloat16x4_t __s1_148 = __p1_148; \
+ bfloat16x8_t __s2_148 = __p2_148; \
+ float32x2_t __ret_148; \
+bfloat16x8_t __reint_148 = __s2_148; \
+float32x2_t __reint1_148 = splat_laneq_f32(*(float32x4_t *) &__reint_148, __p3_148); \
+ __ret_148 = vbfdot_f32(__s0_148, __s1_148, *(bfloat16x4_t *) &__reint1_148); \
+ __ret_148; \
+})
#else
-__ai float64x1_t vreinterpret_f64_p64(poly64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
+#define vbfdot_laneq_f32(__p0_149, __p1_149, __p2_149, __p3_149) __extension__ ({ \
+ float32x2_t __s0_149 = __p0_149; \
+ bfloat16x4_t __s1_149 = __p1_149; \
+ bfloat16x8_t __s2_149 = __p2_149; \
+ float32x2_t __rev0_149; __rev0_149 = __builtin_shufflevector(__s0_149, __s0_149, 1, 0); \
+ bfloat16x4_t __rev1_149; __rev1_149 = __builtin_shufflevector(__s1_149, __s1_149, 3, 2, 1, 0); \
+ bfloat16x8_t __rev2_149; __rev2_149 = __builtin_shufflevector(__s2_149, __s2_149, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float32x2_t __ret_149; \
+bfloat16x8_t __reint_149 = __rev2_149; \
+float32x2_t __reint1_149 = __noswap_splat_laneq_f32(*(float32x4_t *) &__reint_149, __p3_149); \
+ __ret_149 = __noswap_vbfdot_f32(__rev0_149, __rev1_149, *(bfloat16x4_t *) &__reint1_149); \
+ __ret_149 = __builtin_shufflevector(__ret_149, __ret_149, 1, 0); \
+ __ret_149; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vreinterpret_f64_p16(poly16x4_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
+__ai float32x4_t vbfmlalbq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vbfmlalbq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
return __ret;
}
#else
-__ai float64x1_t vreinterpret_f64_p16(poly16x4_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
+__ai float32x4_t vbfmlalbq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vbfmlalbq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai float32x4_t __noswap_vbfmlalbq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vbfmlalbq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vreinterpret_f64_u8(uint8x8_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
+__ai float32x4_t vbfmlaltq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vbfmlaltq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
return __ret;
}
#else
-__ai float64x1_t vreinterpret_f64_u8(uint8x8_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
+__ai float32x4_t vbfmlaltq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vbfmlaltq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai float32x4_t __noswap_vbfmlaltq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vbfmlaltq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vreinterpret_f64_u32(uint32x2_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
+__ai float32x4_t vbfmmlaq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vbfmmlaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
return __ret;
}
#else
-__ai float64x1_t vreinterpret_f64_u32(uint32x2_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
+__ai float32x4_t vbfmmlaq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vbfmmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vreinterpret_f64_u64(uint64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
+__ai bfloat16x8_t vcombine_bf16(bfloat16x4_t __p0, bfloat16x4_t __p1) {
+ bfloat16x8_t __ret;
+ __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7);
return __ret;
}
#else
-__ai float64x1_t vreinterpret_f64_u64(uint64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
+__ai bfloat16x8_t vcombine_bf16(bfloat16x4_t __p0, bfloat16x4_t __p1) {
+ bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ bfloat16x8_t __ret;
+ __ret = __builtin_shufflevector(__rev0, __rev1, 0, 1, 2, 3, 4, 5, 6, 7);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+__ai bfloat16x8_t __noswap_vcombine_bf16(bfloat16x4_t __p0, bfloat16x4_t __p1) {
+ bfloat16x8_t __ret;
+ __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7);
return __ret;
}
#endif
+#define vcreate_bf16(__p0) __extension__ ({ \
+ bfloat16x4_t __ret; \
+ uint64_t __promote = __p0; \
+ __ret = (bfloat16x4_t)(__promote); \
+ __ret; \
+})
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vreinterpret_f64_u16(uint16x4_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
+__ai float32x4_t vcvt_f32_bf16(bfloat16x4_t __p0_150) {
+ float32x4_t __ret_150;
+bfloat16x4_t __reint_150 = __p0_150;
+int32x4_t __reint1_150 = vshll_n_s16(*(int16x4_t *) &__reint_150, 16);
+ __ret_150 = *(float32x4_t *) &__reint1_150;
+ return __ret_150;
}
#else
-__ai float64x1_t vreinterpret_f64_u16(uint16x4_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
+__ai float32x4_t vcvt_f32_bf16(bfloat16x4_t __p0_151) {
+ bfloat16x4_t __rev0_151; __rev0_151 = __builtin_shufflevector(__p0_151, __p0_151, 3, 2, 1, 0);
+ float32x4_t __ret_151;
+bfloat16x4_t __reint_151 = __rev0_151;
+int32x4_t __reint1_151 = __noswap_vshll_n_s16(*(int16x4_t *) &__reint_151, 16);
+ __ret_151 = *(float32x4_t *) &__reint1_151;
+ __ret_151 = __builtin_shufflevector(__ret_151, __ret_151, 3, 2, 1, 0);
+ return __ret_151;
+}
+__ai float32x4_t __noswap_vcvt_f32_bf16(bfloat16x4_t __p0_152) {
+ float32x4_t __ret_152;
+bfloat16x4_t __reint_152 = __p0_152;
+int32x4_t __reint1_152 = __noswap_vshll_n_s16(*(int16x4_t *) &__reint_152, 16);
+ __ret_152 = *(float32x4_t *) &__reint1_152;
+ return __ret_152;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vreinterpret_f64_s8(int8x8_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
+__ai float32_t vcvtah_f32_bf16(bfloat16_t __p0) {
+ float32_t __ret;
+bfloat16_t __reint = __p0;
+int32_t __reint1 = *(int32_t *) &__reint << 16;
+ __ret = *(float32_t *) &__reint1;
return __ret;
}
-#else
-__ai float64x1_t vreinterpret_f64_s8(int8x8_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
+__ai bfloat16_t vcvth_bf16_f32(float32_t __p0) {
+ bfloat16_t __ret;
+ __ret = (bfloat16_t) __builtin_neon_vcvth_bf16_f32(__p0);
return __ret;
}
+#ifdef __LITTLE_ENDIAN__
+#define vduph_lane_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4_t __s0 = __p0; \
+ bfloat16_t __ret; \
+ __ret = (bfloat16_t) __builtin_neon_vduph_lane_bf16((bfloat16x4_t)__s0, __p1); \
+ __ret; \
+})
+#else
+#define vduph_lane_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4_t __s0 = __p0; \
+ bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ bfloat16_t __ret; \
+ __ret = (bfloat16_t) __builtin_neon_vduph_lane_bf16((bfloat16x4_t)__rev0, __p1); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vreinterpret_f64_f32(float32x2_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
+#define vdupq_lane_bf16(__p0_153, __p1_153) __extension__ ({ \
+ bfloat16x4_t __s0_153 = __p0_153; \
+ bfloat16x8_t __ret_153; \
+ __ret_153 = splatq_lane_bf16(__s0_153, __p1_153); \
+ __ret_153; \
+})
#else
-__ai float64x1_t vreinterpret_f64_f32(float32x2_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
+#define vdupq_lane_bf16(__p0_154, __p1_154) __extension__ ({ \
+ bfloat16x4_t __s0_154 = __p0_154; \
+ bfloat16x4_t __rev0_154; __rev0_154 = __builtin_shufflevector(__s0_154, __s0_154, 3, 2, 1, 0); \
+ bfloat16x8_t __ret_154; \
+ __ret_154 = __noswap_splatq_lane_bf16(__rev0_154, __p1_154); \
+ __ret_154 = __builtin_shufflevector(__ret_154, __ret_154, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_154; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vreinterpret_f64_f16(float16x4_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
+#define vdup_lane_bf16(__p0_155, __p1_155) __extension__ ({ \
+ bfloat16x4_t __s0_155 = __p0_155; \
+ bfloat16x4_t __ret_155; \
+ __ret_155 = splat_lane_bf16(__s0_155, __p1_155); \
+ __ret_155; \
+})
#else
-__ai float64x1_t vreinterpret_f64_f16(float16x4_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
+#define vdup_lane_bf16(__p0_156, __p1_156) __extension__ ({ \
+ bfloat16x4_t __s0_156 = __p0_156; \
+ bfloat16x4_t __rev0_156; __rev0_156 = __builtin_shufflevector(__s0_156, __s0_156, 3, 2, 1, 0); \
+ bfloat16x4_t __ret_156; \
+ __ret_156 = __noswap_splat_lane_bf16(__rev0_156, __p1_156); \
+ __ret_156 = __builtin_shufflevector(__ret_156, __ret_156, 3, 2, 1, 0); \
+ __ret_156; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vreinterpret_f64_s32(int32x2_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
+#define vduph_laneq_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8_t __s0 = __p0; \
+ bfloat16_t __ret; \
+ __ret = (bfloat16_t) __builtin_neon_vduph_laneq_bf16((bfloat16x8_t)__s0, __p1); \
+ __ret; \
+})
#else
-__ai float64x1_t vreinterpret_f64_s32(int32x2_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
+#define vduph_laneq_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8_t __s0 = __p0; \
+ bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16_t __ret; \
+ __ret = (bfloat16_t) __builtin_neon_vduph_laneq_bf16((bfloat16x8_t)__rev0, __p1); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vreinterpret_f64_s64(int64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
+#define vdupq_laneq_bf16(__p0_157, __p1_157) __extension__ ({ \
+ bfloat16x8_t __s0_157 = __p0_157; \
+ bfloat16x8_t __ret_157; \
+ __ret_157 = splatq_laneq_bf16(__s0_157, __p1_157); \
+ __ret_157; \
+})
#else
-__ai float64x1_t vreinterpret_f64_s64(int64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
+#define vdupq_laneq_bf16(__p0_158, __p1_158) __extension__ ({ \
+ bfloat16x8_t __s0_158 = __p0_158; \
+ bfloat16x8_t __rev0_158; __rev0_158 = __builtin_shufflevector(__s0_158, __s0_158, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x8_t __ret_158; \
+ __ret_158 = __noswap_splatq_laneq_bf16(__rev0_158, __p1_158); \
+ __ret_158 = __builtin_shufflevector(__ret_158, __ret_158, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_158; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vreinterpret_f64_s16(int16x4_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
+#define vdup_laneq_bf16(__p0_159, __p1_159) __extension__ ({ \
+ bfloat16x8_t __s0_159 = __p0_159; \
+ bfloat16x4_t __ret_159; \
+ __ret_159 = splat_laneq_bf16(__s0_159, __p1_159); \
+ __ret_159; \
+})
#else
-__ai float64x1_t vreinterpret_f64_s16(int16x4_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
+#define vdup_laneq_bf16(__p0_160, __p1_160) __extension__ ({ \
+ bfloat16x8_t __s0_160 = __p0_160; \
+ bfloat16x8_t __rev0_160; __rev0_160 = __builtin_shufflevector(__s0_160, __s0_160, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x4_t __ret_160; \
+ __ret_160 = __noswap_splat_laneq_bf16(__rev0_160, __p1_160); \
+ __ret_160 = __builtin_shufflevector(__ret_160, __ret_160, 3, 2, 1, 0); \
+ __ret_160; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_p8(poly8x8_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai bfloat16x8_t vdupq_n_bf16(bfloat16_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t) {__p0, __p0, __p0, __p0, __p0, __p0, __p0, __p0};
return __ret;
}
#else
-__ai float32x2_t vreinterpret_f32_p8(poly8x8_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai bfloat16x8_t vdupq_n_bf16(bfloat16_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t) {__p0, __p0, __p0, __p0, __p0, __p0, __p0, __p0};
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_p64(poly64x1_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai bfloat16x4_t vdup_n_bf16(bfloat16_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t) {__p0, __p0, __p0, __p0};
return __ret;
}
#else
-__ai float32x2_t vreinterpret_f32_p64(poly64x1_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai bfloat16x4_t vdup_n_bf16(bfloat16_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t) {__p0, __p0, __p0, __p0};
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_p16(poly16x4_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai bfloat16x4_t vget_high_bf16(bfloat16x8_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = __builtin_shufflevector(__p0, __p0, 4, 5, 6, 7);
return __ret;
}
#else
-__ai float32x2_t vreinterpret_f32_p16(poly16x4_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_u8(uint8x8_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai bfloat16x4_t vget_high_bf16(bfloat16x8_t __p0) {
+ bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ bfloat16x4_t __ret;
+ __ret = __builtin_shufflevector(__rev0, __rev0, 4, 5, 6, 7);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
-#else
-__ai float32x2_t vreinterpret_f32_u8(uint8x8_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai bfloat16x4_t __noswap_vget_high_bf16(bfloat16x8_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = __builtin_shufflevector(__p0, __p0, 4, 5, 6, 7);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_u32(uint32x2_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vgetq_lane_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8_t __s0 = __p0; \
+ bfloat16_t __ret; \
+ __ret = (bfloat16_t) __builtin_neon_vgetq_lane_bf16((bfloat16x8_t)__s0, __p1); \
+ __ret; \
+})
#else
-__ai float32x2_t vreinterpret_f32_u32(uint32x2_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vgetq_lane_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8_t __s0 = __p0; \
+ bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16_t __ret; \
+ __ret = (bfloat16_t) __builtin_neon_vgetq_lane_bf16((bfloat16x8_t)__rev0, __p1); \
+ __ret; \
+})
+#define __noswap_vgetq_lane_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8_t __s0 = __p0; \
+ bfloat16_t __ret; \
+ __ret = (bfloat16_t) __builtin_neon_vgetq_lane_bf16((bfloat16x8_t)__s0, __p1); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_u64(uint64x1_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vget_lane_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4_t __s0 = __p0; \
+ bfloat16_t __ret; \
+ __ret = (bfloat16_t) __builtin_neon_vget_lane_bf16((bfloat16x4_t)__s0, __p1); \
+ __ret; \
+})
#else
-__ai float32x2_t vreinterpret_f32_u64(uint64x1_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vget_lane_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4_t __s0 = __p0; \
+ bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ bfloat16_t __ret; \
+ __ret = (bfloat16_t) __builtin_neon_vget_lane_bf16((bfloat16x4_t)__rev0, __p1); \
+ __ret; \
+})
+#define __noswap_vget_lane_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4_t __s0 = __p0; \
+ bfloat16_t __ret; \
+ __ret = (bfloat16_t) __builtin_neon_vget_lane_bf16((bfloat16x4_t)__s0, __p1); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_u16(uint16x4_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai bfloat16x4_t vget_low_bf16(bfloat16x8_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = __builtin_shufflevector(__p0, __p0, 0, 1, 2, 3);
return __ret;
}
#else
-__ai float32x2_t vreinterpret_f32_u16(uint16x4_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
+__ai bfloat16x4_t vget_low_bf16(bfloat16x8_t __p0) {
+ bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ bfloat16x4_t __ret;
+ __ret = __builtin_shufflevector(__rev0, __rev0, 0, 1, 2, 3);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai bfloat16x4_t __noswap_vget_low_bf16(bfloat16x8_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = __builtin_shufflevector(__p0, __p0, 0, 1, 2, 3);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_s8(int8x8_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vld1q_bf16(__p0) __extension__ ({ \
+ bfloat16x8_t __ret; \
+ __ret = (bfloat16x8_t) __builtin_neon_vld1q_v(__p0, 43); \
+ __ret; \
+})
#else
-__ai float32x2_t vreinterpret_f32_s8(int8x8_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vld1q_bf16(__p0) __extension__ ({ \
+ bfloat16x8_t __ret; \
+ __ret = (bfloat16x8_t) __builtin_neon_vld1q_v(__p0, 43); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_f64(float64x1_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vld1_bf16(__p0) __extension__ ({ \
+ bfloat16x4_t __ret; \
+ __ret = (bfloat16x4_t) __builtin_neon_vld1_v(__p0, 11); \
+ __ret; \
+})
#else
-__ai float32x2_t vreinterpret_f32_f64(float64x1_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vld1_bf16(__p0) __extension__ ({ \
+ bfloat16x4_t __ret; \
+ __ret = (bfloat16x4_t) __builtin_neon_vld1_v(__p0, 11); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_f16(float16x4_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vld1q_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x8_t __ret; \
+ __ret = (bfloat16x8_t) __builtin_neon_vld1q_dup_v(__p0, 43); \
+ __ret; \
+})
#else
-__ai float32x2_t vreinterpret_f32_f16(float16x4_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vld1q_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x8_t __ret; \
+ __ret = (bfloat16x8_t) __builtin_neon_vld1q_dup_v(__p0, 43); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_s32(int32x2_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vld1_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x4_t __ret; \
+ __ret = (bfloat16x4_t) __builtin_neon_vld1_dup_v(__p0, 11); \
+ __ret; \
+})
#else
-__ai float32x2_t vreinterpret_f32_s32(int32x2_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vld1_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x4_t __ret; \
+ __ret = (bfloat16x4_t) __builtin_neon_vld1_dup_v(__p0, 11); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_s64(int64x1_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vld1q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8_t __s1 = __p1; \
+ bfloat16x8_t __ret; \
+ __ret = (bfloat16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 43); \
+ __ret; \
+})
#else
-__ai float32x2_t vreinterpret_f32_s64(int64x1_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vld1q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8_t __s1 = __p1; \
+ bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x8_t __ret; \
+ __ret = (bfloat16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 43); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vreinterpret_f32_s16(int16x4_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vld1_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4_t __s1 = __p1; \
+ bfloat16x4_t __ret; \
+ __ret = (bfloat16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 11); \
+ __ret; \
+})
#else
-__ai float32x2_t vreinterpret_f32_s16(int16x4_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t)(__p0);
- return __ret;
-}
+#define vld1_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4_t __s1 = __p1; \
+ bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
+ bfloat16x4_t __ret; \
+ __ret = (bfloat16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 11); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_p8(poly8x8_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1q_bf16_x2(__p0) __extension__ ({ \
+ bfloat16x8x2_t __ret; \
+ __builtin_neon_vld1q_x2_v(&__ret, __p0, 43); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_p8(poly8x8_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1q_bf16_x2(__p0) __extension__ ({ \
+ bfloat16x8x2_t __ret; \
+ __builtin_neon_vld1q_x2_v(&__ret, __p0, 43); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_p64(poly64x1_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1_bf16_x2(__p0) __extension__ ({ \
+ bfloat16x4x2_t __ret; \
+ __builtin_neon_vld1_x2_v(&__ret, __p0, 11); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_p64(poly64x1_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1_bf16_x2(__p0) __extension__ ({ \
+ bfloat16x4x2_t __ret; \
+ __builtin_neon_vld1_x2_v(&__ret, __p0, 11); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_p16(poly16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1q_bf16_x3(__p0) __extension__ ({ \
+ bfloat16x8x3_t __ret; \
+ __builtin_neon_vld1q_x3_v(&__ret, __p0, 43); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_p16(poly16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1q_bf16_x3(__p0) __extension__ ({ \
+ bfloat16x8x3_t __ret; \
+ __builtin_neon_vld1q_x3_v(&__ret, __p0, 43); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_u8(uint8x8_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1_bf16_x3(__p0) __extension__ ({ \
+ bfloat16x4x3_t __ret; \
+ __builtin_neon_vld1_x3_v(&__ret, __p0, 11); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_u8(uint8x8_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1_bf16_x3(__p0) __extension__ ({ \
+ bfloat16x4x3_t __ret; \
+ __builtin_neon_vld1_x3_v(&__ret, __p0, 11); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_u32(uint32x2_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1q_bf16_x4(__p0) __extension__ ({ \
+ bfloat16x8x4_t __ret; \
+ __builtin_neon_vld1q_x4_v(&__ret, __p0, 43); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_u32(uint32x2_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1q_bf16_x4(__p0) __extension__ ({ \
+ bfloat16x8x4_t __ret; \
+ __builtin_neon_vld1q_x4_v(&__ret, __p0, 43); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_u64(uint64x1_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1_bf16_x4(__p0) __extension__ ({ \
+ bfloat16x4x4_t __ret; \
+ __builtin_neon_vld1_x4_v(&__ret, __p0, 11); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_u64(uint64x1_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld1_bf16_x4(__p0) __extension__ ({ \
+ bfloat16x4x4_t __ret; \
+ __builtin_neon_vld1_x4_v(&__ret, __p0, 11); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_u16(uint16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2q_bf16(__p0) __extension__ ({ \
+ bfloat16x8x2_t __ret; \
+ __builtin_neon_vld2q_v(&__ret, __p0, 43); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_u16(uint16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2q_bf16(__p0) __extension__ ({ \
+ bfloat16x8x2_t __ret; \
+ __builtin_neon_vld2q_v(&__ret, __p0, 43); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_s8(int8x8_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2_bf16(__p0) __extension__ ({ \
+ bfloat16x4x2_t __ret; \
+ __builtin_neon_vld2_v(&__ret, __p0, 11); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_s8(int8x8_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2_bf16(__p0) __extension__ ({ \
+ bfloat16x4x2_t __ret; \
+ __builtin_neon_vld2_v(&__ret, __p0, 11); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_f64(float64x1_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2q_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x8x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 43); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_f64(float64x1_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2q_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x8x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 43); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_f32(float32x2_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x4x2_t __ret; \
+ __builtin_neon_vld2_dup_v(&__ret, __p0, 11); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_f32(float32x2_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x4x2_t __ret; \
+ __builtin_neon_vld2_dup_v(&__ret, __p0, 11); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_s32(int32x2_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8x2_t __s1 = __p1; \
+ bfloat16x8x2_t __ret; \
+ __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 43); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_s32(int32x2_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8x2_t __s1 = __p1; \
+ bfloat16x8x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x8x2_t __ret; \
+ __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 43); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_s64(int64x1_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4x2_t __s1 = __p1; \
+ bfloat16x4x2_t __ret; \
+ __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 11); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_s64(int64x1_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld2_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4x2_t __s1 = __p1; \
+ bfloat16x4x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ bfloat16x4x2_t __ret; \
+ __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 11); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vreinterpret_f16_s16(int16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld3q_bf16(__p0) __extension__ ({ \
+ bfloat16x8x3_t __ret; \
+ __builtin_neon_vld3q_v(&__ret, __p0, 43); \
+ __ret; \
+})
#else
-__ai float16x4_t vreinterpret_f16_s16(int16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t)(__p0);
- return __ret;
-}
+#define vld3q_bf16(__p0) __extension__ ({ \
+ bfloat16x8x3_t __ret; \
+ __builtin_neon_vld3q_v(&__ret, __p0, 43); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_p8(poly8x8_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3_bf16(__p0) __extension__ ({ \
+ bfloat16x4x3_t __ret; \
+ __builtin_neon_vld3_v(&__ret, __p0, 11); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_p8(poly8x8_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3_bf16(__p0) __extension__ ({ \
+ bfloat16x4x3_t __ret; \
+ __builtin_neon_vld3_v(&__ret, __p0, 11); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_p64(poly64x1_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3q_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x8x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 43); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_p64(poly64x1_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3q_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x8x3_t __ret; \
+ __builtin_neon_vld3q_dup_v(&__ret, __p0, 43); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_p16(poly16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x4x3_t __ret; \
+ __builtin_neon_vld3_dup_v(&__ret, __p0, 11); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_p16(poly16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x4x3_t __ret; \
+ __builtin_neon_vld3_dup_v(&__ret, __p0, 11); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_u8(uint8x8_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8x3_t __s1 = __p1; \
+ bfloat16x8x3_t __ret; \
+ __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 43); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_u8(uint8x8_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8x3_t __s1 = __p1; \
+ bfloat16x8x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x8x3_t __ret; \
+ __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 43); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_u32(uint32x2_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4x3_t __s1 = __p1; \
+ bfloat16x4x3_t __ret; \
+ __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 11); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_u32(uint32x2_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld3_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4x3_t __s1 = __p1; \
+ bfloat16x4x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ bfloat16x4x3_t __ret; \
+ __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 11); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_u64(uint64x1_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4q_bf16(__p0) __extension__ ({ \
+ bfloat16x8x4_t __ret; \
+ __builtin_neon_vld4q_v(&__ret, __p0, 43); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_u64(uint64x1_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4q_bf16(__p0) __extension__ ({ \
+ bfloat16x8x4_t __ret; \
+ __builtin_neon_vld4q_v(&__ret, __p0, 43); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_u16(uint16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4_bf16(__p0) __extension__ ({ \
+ bfloat16x4x4_t __ret; \
+ __builtin_neon_vld4_v(&__ret, __p0, 11); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_u16(uint16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4_bf16(__p0) __extension__ ({ \
+ bfloat16x4x4_t __ret; \
+ __builtin_neon_vld4_v(&__ret, __p0, 11); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_s8(int8x8_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4q_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x8x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 43); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_s8(int8x8_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4q_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x8x4_t __ret; \
+ __builtin_neon_vld4q_dup_v(&__ret, __p0, 43); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_f64(float64x1_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x4x4_t __ret; \
+ __builtin_neon_vld4_dup_v(&__ret, __p0, 11); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_f64(float64x1_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4_dup_bf16(__p0) __extension__ ({ \
+ bfloat16x4x4_t __ret; \
+ __builtin_neon_vld4_dup_v(&__ret, __p0, 11); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_f32(float32x2_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8x4_t __s1 = __p1; \
+ bfloat16x8x4_t __ret; \
+ __builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 43); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_f32(float32x2_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8x4_t __s1 = __p1; \
+ bfloat16x8x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x8x4_t __ret; \
+ __builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 43); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_f16(float16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4x4_t __s1 = __p1; \
+ bfloat16x4x4_t __ret; \
+ __builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 11); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_f16(float16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vld4_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4x4_t __s1 = __p1; \
+ bfloat16x4x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
+ bfloat16x4x4_t __ret; \
+ __builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 11); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_s64(int64x1_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vsetq_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16_t __s0 = __p0; \
+ bfloat16x8_t __s1 = __p1; \
+ bfloat16x8_t __ret; \
+ __ret = (bfloat16x8_t) __builtin_neon_vsetq_lane_bf16(__s0, (bfloat16x8_t)__s1, __p2); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_s64(int64x1_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vsetq_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16_t __s0 = __p0; \
+ bfloat16x8_t __s1 = __p1; \
+ bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x8_t __ret; \
+ __ret = (bfloat16x8_t) __builtin_neon_vsetq_lane_bf16(__s0, (bfloat16x8_t)__rev1, __p2); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_vsetq_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16_t __s0 = __p0; \
+ bfloat16x8_t __s1 = __p1; \
+ bfloat16x8_t __ret; \
+ __ret = (bfloat16x8_t) __builtin_neon_vsetq_lane_bf16(__s0, (bfloat16x8_t)__s1, __p2); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vreinterpret_s32_s16(int16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vset_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16_t __s0 = __p0; \
+ bfloat16x4_t __s1 = __p1; \
+ bfloat16x4_t __ret; \
+ __ret = (bfloat16x4_t) __builtin_neon_vset_lane_bf16(__s0, (bfloat16x4_t)__s1, __p2); \
+ __ret; \
+})
#else
-__ai int32x2_t vreinterpret_s32_s16(int16x4_t __p0) {
- int32x2_t __ret;
- __ret = (int32x2_t)(__p0);
- return __ret;
-}
+#define vset_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16_t __s0 = __p0; \
+ bfloat16x4_t __s1 = __p1; \
+ bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
+ bfloat16x4_t __ret; \
+ __ret = (bfloat16x4_t) __builtin_neon_vset_lane_bf16(__s0, (bfloat16x4_t)__rev1, __p2); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#define __noswap_vset_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16_t __s0 = __p0; \
+ bfloat16x4_t __s1 = __p1; \
+ bfloat16x4_t __ret; \
+ __ret = (bfloat16x4_t) __builtin_neon_vset_lane_bf16(__s0, (bfloat16x4_t)__s1, __p2); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_p8(poly8x8_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8_t __s1 = __p1; \
+ __builtin_neon_vst1q_v(__p0, (int8x16_t)__s1, 43); \
+})
#else
-__ai int64x1_t vreinterpret_s64_p8(poly8x8_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8_t __s1 = __p1; \
+ bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst1q_v(__p0, (int8x16_t)__rev1, 43); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_p64(poly64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4_t __s1 = __p1; \
+ __builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 11); \
+})
#else
-__ai int64x1_t vreinterpret_s64_p64(poly64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4_t __s1 = __p1; \
+ bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
+ __builtin_neon_vst1_v(__p0, (int8x8_t)__rev1, 11); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_p16(poly16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8_t __s1 = __p1; \
+ __builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__s1, __p2, 43); \
+})
#else
-__ai int64x1_t vreinterpret_s64_p16(poly16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8_t __s1 = __p1; \
+ bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 43); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_u8(uint8x8_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4_t __s1 = __p1; \
+ __builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 11); \
+})
#else
-__ai int64x1_t vreinterpret_s64_u8(uint8x8_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4_t __s1 = __p1; \
+ bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
+ __builtin_neon_vst1_lane_v(__p0, (int8x8_t)__rev1, __p2, 11); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_u32(uint32x2_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_bf16_x2(__p0, __p1) __extension__ ({ \
+ bfloat16x8x2_t __s1 = __p1; \
+ __builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 43); \
+})
#else
-__ai int64x1_t vreinterpret_s64_u32(uint32x2_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_bf16_x2(__p0, __p1) __extension__ ({ \
+ bfloat16x8x2_t __s1 = __p1; \
+ bfloat16x8x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 43); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_u64(uint64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_bf16_x2(__p0, __p1) __extension__ ({ \
+ bfloat16x4x2_t __s1 = __p1; \
+ __builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 11); \
+})
#else
-__ai int64x1_t vreinterpret_s64_u64(uint64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_bf16_x2(__p0, __p1) __extension__ ({ \
+ bfloat16x4x2_t __s1 = __p1; \
+ bfloat16x4x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __builtin_neon_vst1_x2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 11); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_u16(uint16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_bf16_x3(__p0, __p1) __extension__ ({ \
+ bfloat16x8x3_t __s1 = __p1; \
+ __builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 43); \
+})
#else
-__ai int64x1_t vreinterpret_s64_u16(uint16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_bf16_x3(__p0, __p1) __extension__ ({ \
+ bfloat16x8x3_t __s1 = __p1; \
+ bfloat16x8x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 43); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_s8(int8x8_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_bf16_x3(__p0, __p1) __extension__ ({ \
+ bfloat16x4x3_t __s1 = __p1; \
+ __builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 11); \
+})
#else
-__ai int64x1_t vreinterpret_s64_s8(int8x8_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_bf16_x3(__p0, __p1) __extension__ ({ \
+ bfloat16x4x3_t __s1 = __p1; \
+ bfloat16x4x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ __builtin_neon_vst1_x3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 11); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_f64(float64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_bf16_x4(__p0, __p1) __extension__ ({ \
+ bfloat16x8x4_t __s1 = __p1; \
+ __builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 43); \
+})
#else
-__ai int64x1_t vreinterpret_s64_f64(float64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1q_bf16_x4(__p0, __p1) __extension__ ({ \
+ bfloat16x8x4_t __s1 = __p1; \
+ bfloat16x8x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 43); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_f32(float32x2_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_bf16_x4(__p0, __p1) __extension__ ({ \
+ bfloat16x4x4_t __s1 = __p1; \
+ __builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 11); \
+})
#else
-__ai int64x1_t vreinterpret_s64_f32(float32x2_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst1_bf16_x4(__p0, __p1) __extension__ ({ \
+ bfloat16x4x4_t __s1 = __p1; \
+ bfloat16x4x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
+ __builtin_neon_vst1_x4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 11); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_f16(float16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst2q_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8x2_t __s1 = __p1; \
+ __builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 43); \
+})
#else
-__ai int64x1_t vreinterpret_s64_f16(float16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst2q_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8x2_t __s1 = __p1; \
+ bfloat16x8x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 43); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_s32(int32x2_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst2_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4x2_t __s1 = __p1; \
+ __builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 11); \
+})
#else
-__ai int64x1_t vreinterpret_s64_s32(int32x2_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst2_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4x2_t __s1 = __p1; \
+ bfloat16x4x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __builtin_neon_vst2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 11); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vreinterpret_s64_s16(int16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst2q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8x2_t __s1 = __p1; \
+ __builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 43); \
+})
#else
-__ai int64x1_t vreinterpret_s64_s16(int16x4_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t)(__p0);
- return __ret;
-}
+#define vst2q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8x2_t __s1 = __p1; \
+ bfloat16x8x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 43); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_p8(poly8x8_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst2_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4x2_t __s1 = __p1; \
+ __builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 11); \
+})
#else
-__ai int16x4_t vreinterpret_s16_p8(poly8x8_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst2_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4x2_t __s1 = __p1; \
+ bfloat16x4x2_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __builtin_neon_vst2_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 11); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_p64(poly64x1_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3q_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8x3_t __s1 = __p1; \
+ __builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 43); \
+})
#else
-__ai int16x4_t vreinterpret_s16_p64(poly64x1_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3q_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8x3_t __s1 = __p1; \
+ bfloat16x8x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 43); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_p16(poly16x4_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4x3_t __s1 = __p1; \
+ __builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 11); \
+})
#else
-__ai int16x4_t vreinterpret_s16_p16(poly16x4_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4x3_t __s1 = __p1; \
+ bfloat16x4x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ __builtin_neon_vst3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 11); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_u8(uint8x8_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8x3_t __s1 = __p1; \
+ __builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 43); \
+})
#else
-__ai int16x4_t vreinterpret_s16_u8(uint8x8_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8x3_t __s1 = __p1; \
+ bfloat16x8x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 43); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_u32(uint32x2_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4x3_t __s1 = __p1; \
+ __builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 11); \
+})
#else
-__ai int16x4_t vreinterpret_s16_u32(uint32x2_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst3_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4x3_t __s1 = __p1; \
+ bfloat16x4x3_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ __builtin_neon_vst3_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 11); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_u64(uint64x1_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst4q_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8x4_t __s1 = __p1; \
+ __builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 43); \
+})
#else
-__ai int16x4_t vreinterpret_s16_u64(uint64x1_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst4q_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x8x4_t __s1 = __p1; \
+ bfloat16x8x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 43); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_u16(uint16x4_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst4_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4x4_t __s1 = __p1; \
+ __builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 11); \
+})
#else
-__ai int16x4_t vreinterpret_s16_u16(uint16x4_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst4_bf16(__p0, __p1) __extension__ ({ \
+ bfloat16x4x4_t __s1 = __p1; \
+ bfloat16x4x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
+ __builtin_neon_vst4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 11); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_s8(int8x8_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst4q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8x4_t __s1 = __p1; \
+ __builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 43); \
+})
#else
-__ai int16x4_t vreinterpret_s16_s8(int8x8_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst4q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x8x4_t __s1 = __p1; \
+ bfloat16x8x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 43); \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_f64(float64x1_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
-#else
-__ai int16x4_t vreinterpret_s16_f64(float64x1_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
- return __ret;
-}
+#define vst4_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4x4_t __s1 = __p1; \
+ __builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 11); \
+})
+#else
+#define vst4_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
+ bfloat16x4x4_t __s1 = __p1; \
+ bfloat16x4x4_t __rev1; \
+ __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
+ __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
+ __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
+ __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
+ __builtin_neon_vst4_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 11); \
+})
#endif
+#endif
+#if defined(__ARM_FEATURE_BF16_VECTOR_ARITHMETIC) && !defined(__aarch64__)
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_f32(float32x2_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
+__ai bfloat16x4_t __a32_vcvt_bf16_f32(float32x4_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t) __builtin_neon___a32_vcvt_bf16_v((int8x16_t)__p0, 11);
return __ret;
}
#else
-__ai int16x4_t vreinterpret_s16_f32(float32x2_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
+__ai bfloat16x4_t __a32_vcvt_bf16_f32(float32x4_t __p0) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t) __builtin_neon___a32_vcvt_bf16_v((int8x16_t)__rev0, 11);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai bfloat16x4_t __noswap___a32_vcvt_bf16_f32(float32x4_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = (bfloat16x4_t) __builtin_neon___a32_vcvt_bf16_v((int8x16_t)__p0, 11);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_f16(float16x4_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
+__ai bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = __a32_vcvt_bf16_f32(__p0);
return __ret;
}
#else
-__ai int16x4_t vreinterpret_s16_f16(float16x4_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
+__ai bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ bfloat16x4_t __ret;
+ __ret = __noswap___a32_vcvt_bf16_f32(__rev0);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_s32(int32x2_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
+__ai bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
+ bfloat16x8_t __ret;
+ __ret = vcombine_bf16(__a32_vcvt_bf16_f32(__p1), vget_low_bf16(__p0));
return __ret;
}
#else
-__ai int16x4_t vreinterpret_s16_s32(int32x2_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
+__ai bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
+ bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ bfloat16x8_t __ret;
+ __ret = __noswap_vcombine_bf16(__noswap___a32_vcvt_bf16_f32(__rev1), __noswap_vget_low_bf16(__rev0));
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vreinterpret_s16_s64(int64x1_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
+__ai bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = vcombine_bf16((bfloat16x4_t)(0ULL), __a32_vcvt_bf16_f32(__p0));
return __ret;
}
#else
-__ai int16x4_t vreinterpret_s16_s64(int64x1_t __p0) {
- int16x4_t __ret;
- __ret = (int16x4_t)(__p0);
+__ai bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ bfloat16x8_t __ret;
+ __ret = __noswap_vcombine_bf16((bfloat16x4_t)(0ULL), __noswap___a32_vcvt_bf16_f32(__rev0));
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
return __ret;
}
#endif
#endif
-#if __ARM_ARCH >= 8 && defined(__aarch64__) && defined(__ARM_FEATURE_DIRECTED_ROUNDING)
+#if defined(__ARM_FEATURE_BF16_VECTOR_ARITHMETIC) && defined(__aarch64__)
#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vrndq_f64(float64x2_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vrndq_v((int8x16_t)__p0, 42);
+__ai bfloat16x8_t __a64_vcvtq_low_bf16_f32(float32x4_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t) __builtin_neon___a64_vcvtq_low_bf16_v((int8x16_t)__p0, 43);
return __ret;
}
#else
-__ai float64x2_t vrndq_f64(float64x2_t __p0) {
- float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
- float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vrndq_v((int8x16_t)__rev0, 42);
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
+__ai bfloat16x8_t __a64_vcvtq_low_bf16_f32(float32x4_t __p0) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t) __builtin_neon___a64_vcvtq_low_bf16_v((int8x16_t)__rev0, 43);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret;
+}
+__ai bfloat16x8_t __noswap___a64_vcvtq_low_bf16_f32(float32x4_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t) __builtin_neon___a64_vcvtq_low_bf16_v((int8x16_t)__p0, 43);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vrnd_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrnd_v((int8x8_t)__p0, 10);
- return __ret;
-}
+#define vcopyq_lane_bf16(__p0_161, __p1_161, __p2_161, __p3_161) __extension__ ({ \
+ bfloat16x8_t __s0_161 = __p0_161; \
+ bfloat16x4_t __s2_161 = __p2_161; \
+ bfloat16x8_t __ret_161; \
+ __ret_161 = vsetq_lane_bf16(vget_lane_bf16(__s2_161, __p3_161), __s0_161, __p1_161); \
+ __ret_161; \
+})
#else
-__ai float64x1_t vrnd_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrnd_v((int8x8_t)__p0, 10);
- return __ret;
-}
+#define vcopyq_lane_bf16(__p0_162, __p1_162, __p2_162, __p3_162) __extension__ ({ \
+ bfloat16x8_t __s0_162 = __p0_162; \
+ bfloat16x4_t __s2_162 = __p2_162; \
+ bfloat16x8_t __rev0_162; __rev0_162 = __builtin_shufflevector(__s0_162, __s0_162, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x4_t __rev2_162; __rev2_162 = __builtin_shufflevector(__s2_162, __s2_162, 3, 2, 1, 0); \
+ bfloat16x8_t __ret_162; \
+ __ret_162 = __noswap_vsetq_lane_bf16(__noswap_vget_lane_bf16(__rev2_162, __p3_162), __rev0_162, __p1_162); \
+ __ret_162 = __builtin_shufflevector(__ret_162, __ret_162, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_162; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vrndaq_f64(float64x2_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vrndaq_v((int8x16_t)__p0, 42);
- return __ret;
-}
+#define vcopy_lane_bf16(__p0_163, __p1_163, __p2_163, __p3_163) __extension__ ({ \
+ bfloat16x4_t __s0_163 = __p0_163; \
+ bfloat16x4_t __s2_163 = __p2_163; \
+ bfloat16x4_t __ret_163; \
+ __ret_163 = vset_lane_bf16(vget_lane_bf16(__s2_163, __p3_163), __s0_163, __p1_163); \
+ __ret_163; \
+})
#else
-__ai float64x2_t vrndaq_f64(float64x2_t __p0) {
- float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
- float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vrndaq_v((int8x16_t)__rev0, 42);
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
- return __ret;
-}
+#define vcopy_lane_bf16(__p0_164, __p1_164, __p2_164, __p3_164) __extension__ ({ \
+ bfloat16x4_t __s0_164 = __p0_164; \
+ bfloat16x4_t __s2_164 = __p2_164; \
+ bfloat16x4_t __rev0_164; __rev0_164 = __builtin_shufflevector(__s0_164, __s0_164, 3, 2, 1, 0); \
+ bfloat16x4_t __rev2_164; __rev2_164 = __builtin_shufflevector(__s2_164, __s2_164, 3, 2, 1, 0); \
+ bfloat16x4_t __ret_164; \
+ __ret_164 = __noswap_vset_lane_bf16(__noswap_vget_lane_bf16(__rev2_164, __p3_164), __rev0_164, __p1_164); \
+ __ret_164 = __builtin_shufflevector(__ret_164, __ret_164, 3, 2, 1, 0); \
+ __ret_164; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vrnda_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrnda_v((int8x8_t)__p0, 10);
- return __ret;
-}
+#define vcopyq_laneq_bf16(__p0_165, __p1_165, __p2_165, __p3_165) __extension__ ({ \
+ bfloat16x8_t __s0_165 = __p0_165; \
+ bfloat16x8_t __s2_165 = __p2_165; \
+ bfloat16x8_t __ret_165; \
+ __ret_165 = vsetq_lane_bf16(vgetq_lane_bf16(__s2_165, __p3_165), __s0_165, __p1_165); \
+ __ret_165; \
+})
#else
-__ai float64x1_t vrnda_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrnda_v((int8x8_t)__p0, 10);
- return __ret;
-}
+#define vcopyq_laneq_bf16(__p0_166, __p1_166, __p2_166, __p3_166) __extension__ ({ \
+ bfloat16x8_t __s0_166 = __p0_166; \
+ bfloat16x8_t __s2_166 = __p2_166; \
+ bfloat16x8_t __rev0_166; __rev0_166 = __builtin_shufflevector(__s0_166, __s0_166, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x8_t __rev2_166; __rev2_166 = __builtin_shufflevector(__s2_166, __s2_166, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x8_t __ret_166; \
+ __ret_166 = __noswap_vsetq_lane_bf16(__noswap_vgetq_lane_bf16(__rev2_166, __p3_166), __rev0_166, __p1_166); \
+ __ret_166 = __builtin_shufflevector(__ret_166, __ret_166, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_166; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vrndiq_f64(float64x2_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vrndiq_v((int8x16_t)__p0, 42);
- return __ret;
-}
+#define vcopy_laneq_bf16(__p0_167, __p1_167, __p2_167, __p3_167) __extension__ ({ \
+ bfloat16x4_t __s0_167 = __p0_167; \
+ bfloat16x8_t __s2_167 = __p2_167; \
+ bfloat16x4_t __ret_167; \
+ __ret_167 = vset_lane_bf16(vgetq_lane_bf16(__s2_167, __p3_167), __s0_167, __p1_167); \
+ __ret_167; \
+})
#else
-__ai float64x2_t vrndiq_f64(float64x2_t __p0) {
- float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
- float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vrndiq_v((int8x16_t)__rev0, 42);
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
- return __ret;
-}
+#define vcopy_laneq_bf16(__p0_168, __p1_168, __p2_168, __p3_168) __extension__ ({ \
+ bfloat16x4_t __s0_168 = __p0_168; \
+ bfloat16x8_t __s2_168 = __p2_168; \
+ bfloat16x4_t __rev0_168; __rev0_168 = __builtin_shufflevector(__s0_168, __s0_168, 3, 2, 1, 0); \
+ bfloat16x8_t __rev2_168; __rev2_168 = __builtin_shufflevector(__s2_168, __s2_168, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x4_t __ret_168; \
+ __ret_168 = __noswap_vset_lane_bf16(__noswap_vgetq_lane_bf16(__rev2_168, __p3_168), __rev0_168, __p1_168); \
+ __ret_168 = __builtin_shufflevector(__ret_168, __ret_168, 3, 2, 1, 0); \
+ __ret_168; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x4_t vrndiq_f32(float32x4_t __p0) {
- float32x4_t __ret;
- __ret = (float32x4_t) __builtin_neon_vrndiq_v((int8x16_t)__p0, 41);
+__ai bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
+ bfloat16x4_t __ret;
+ __ret = vget_low_bf16(__a64_vcvtq_low_bf16_f32(__p0));
return __ret;
}
#else
-__ai float32x4_t vrndiq_f32(float32x4_t __p0) {
+__ai bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- float32x4_t __ret;
- __ret = (float32x4_t) __builtin_neon_vrndiq_v((int8x16_t)__rev0, 41);
+ bfloat16x4_t __ret;
+ __ret = __noswap_vget_low_bf16(__noswap___a64_vcvtq_low_bf16_f32(__rev0));
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vrndi_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrndi_v((int8x8_t)__p0, 10);
+__ai bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t) __builtin_neon_vcvtq_high_bf16_v((int8x16_t)__p0, (int8x16_t)__p1, 43);
return __ret;
}
#else
-__ai float64x1_t vrndi_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrndi_v((int8x8_t)__p0, 10);
+__ai bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
+ bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ bfloat16x8_t __ret;
+ __ret = (bfloat16x8_t) __builtin_neon_vcvtq_high_bf16_v((int8x16_t)__rev0, (int8x16_t)__rev1, 43);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float32x2_t vrndi_f32(float32x2_t __p0) {
- float32x2_t __ret;
- __ret = (float32x2_t) __builtin_neon_vrndi_v((int8x8_t)__p0, 9);
+__ai bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
+ bfloat16x8_t __ret;
+ __ret = __a64_vcvtq_low_bf16_f32(__p0);
return __ret;
}
#else
-__ai float32x2_t vrndi_f32(float32x2_t __p0) {
- float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
- float32x2_t __ret;
- __ret = (float32x2_t) __builtin_neon_vrndi_v((int8x8_t)__rev0, 9);
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
+__ai bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ bfloat16x8_t __ret;
+ __ret = __noswap___a64_vcvtq_low_bf16_f32(__rev0);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
return __ret;
}
#endif
+#endif
+#if defined(__ARM_FEATURE_COMPLEX)
#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vrndmq_f64(float64x2_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vrndmq_v((int8x16_t)__p0, 42);
+__ai float32x2_t vcadd_rot270_f32(float32x2_t __p0, float32x2_t __p1) {
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vcadd_rot270_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
return __ret;
}
#else
-__ai float64x2_t vrndmq_f64(float64x2_t __p0) {
- float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
- float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vrndmq_v((int8x16_t)__rev0, 42);
+__ai float32x2_t vcadd_rot270_f32(float32x2_t __p0, float32x2_t __p1) {
+ float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vcadd_rot270_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vrndm_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrndm_v((int8x8_t)__p0, 10);
+__ai float32x2_t vcadd_rot90_f32(float32x2_t __p0, float32x2_t __p1) {
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vcadd_rot90_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
return __ret;
}
#else
-__ai float64x1_t vrndm_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrndm_v((int8x8_t)__p0, 10);
+__ai float32x2_t vcadd_rot90_f32(float32x2_t __p0, float32x2_t __p1) {
+ float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vcadd_rot90_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vrndnq_f64(float64x2_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vrndnq_v((int8x16_t)__p0, 42);
+__ai float32x4_t vcaddq_rot270_f32(float32x4_t __p0, float32x4_t __p1) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vcaddq_rot270_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
return __ret;
}
#else
-__ai float64x2_t vrndnq_f64(float64x2_t __p0) {
- float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
- float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vrndnq_v((int8x16_t)__rev0, 42);
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
+__ai float32x4_t vcaddq_rot270_f32(float32x4_t __p0, float32x4_t __p1) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vcaddq_rot270_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vrndn_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrndn_v((int8x8_t)__p0, 10);
+__ai float32x4_t vcaddq_rot90_f32(float32x4_t __p0, float32x4_t __p1) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vcaddq_rot90_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
return __ret;
}
#else
-__ai float64x1_t vrndn_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrndn_v((int8x8_t)__p0, 10);
+__ai float32x4_t vcaddq_rot90_f32(float32x4_t __p0, float32x4_t __p1) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vcaddq_rot90_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#endif
+#endif
+#if defined(__ARM_FEATURE_COMPLEX) && defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC)
#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vrndpq_f64(float64x2_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vrndpq_v((int8x16_t)__p0, 42);
+__ai float16x4_t vcadd_rot270_f16(float16x4_t __p0, float16x4_t __p1) {
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vcadd_rot270_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
return __ret;
}
#else
-__ai float64x2_t vrndpq_f64(float64x2_t __p0) {
- float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
- float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vrndpq_v((int8x16_t)__rev0, 42);
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
+__ai float16x4_t vcadd_rot270_f16(float16x4_t __p0, float16x4_t __p1) {
+ float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vcadd_rot270_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vrndp_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrndp_v((int8x8_t)__p0, 10);
+__ai float16x4_t vcadd_rot90_f16(float16x4_t __p0, float16x4_t __p1) {
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vcadd_rot90_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
return __ret;
}
#else
-__ai float64x1_t vrndp_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrndp_v((int8x8_t)__p0, 10);
+__ai float16x4_t vcadd_rot90_f16(float16x4_t __p0, float16x4_t __p1) {
+ float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ float16x4_t __ret;
+ __ret = (float16x4_t) __builtin_neon_vcadd_rot90_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vrndxq_f64(float64x2_t __p0) {
- float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vrndxq_v((int8x16_t)__p0, 42);
+__ai float16x8_t vcaddq_rot270_f16(float16x8_t __p0, float16x8_t __p1) {
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vcaddq_rot270_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
return __ret;
}
#else
-__ai float64x2_t vrndxq_f64(float64x2_t __p0) {
- float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
- float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vrndxq_v((int8x16_t)__rev0, 42);
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
+__ai float16x8_t vcaddq_rot270_f16(float16x8_t __p0, float16x8_t __p1) {
+ float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vcaddq_rot270_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vrndx_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrndx_v((int8x8_t)__p0, 10);
+__ai float16x8_t vcaddq_rot90_f16(float16x8_t __p0, float16x8_t __p1) {
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vcaddq_rot90_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
return __ret;
}
#else
-__ai float64x1_t vrndx_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrndx_v((int8x8_t)__p0, 10);
+__ai float16x8_t vcaddq_rot90_f16(float16x8_t __p0, float16x8_t __p1) {
+ float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __ret;
+ __ret = (float16x8_t) __builtin_neon_vcaddq_rot90_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
return __ret;
}
#endif
#endif
-#if __ARM_ARCH >= 8 && defined(__aarch64__) && defined(__ARM_FEATURE_NUMERIC_MAXMIN)
+#if defined(__ARM_FEATURE_COMPLEX) && defined(__aarch64__)
#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
+__ai float64x2_t vcaddq_rot270_f64(float64x2_t __p0, float64x2_t __p1) {
float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
+ __ret = (float64x2_t) __builtin_neon_vcaddq_rot270_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
return __ret;
}
#else
-__ai float64x2_t vmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
+__ai float64x2_t vcaddq_rot270_f64(float64x2_t __p0, float64x2_t __p1) {
float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
+ __ret = (float64x2_t) __builtin_neon_vcaddq_rot270_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vmaxnm_f64(float64x1_t __p0, float64x1_t __p1) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
- return __ret;
-}
-#else
-__ai float64x1_t vmaxnm_f64(float64x1_t __p0, float64x1_t __p1) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x2_t vminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
+__ai float64x2_t vcaddq_rot90_f64(float64x2_t __p0, float64x2_t __p1) {
float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
+ __ret = (float64x2_t) __builtin_neon_vcaddq_rot90_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
return __ret;
}
#else
-__ai float64x2_t vminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
+__ai float64x2_t vcaddq_rot90_f64(float64x2_t __p0, float64x2_t __p1) {
float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
float64x2_t __ret;
- __ret = (float64x2_t) __builtin_neon_vminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
+ __ret = (float64x2_t) __builtin_neon_vcaddq_rot90_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
return __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vminnm_f64(float64x1_t __p0, float64x1_t __p1) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
- return __ret;
-}
-#else
-__ai float64x1_t vminnm_f64(float64x1_t __p0, float64x1_t __p1) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
- return __ret;
-}
-#endif
-
#endif
#if defined(__ARM_FEATURE_DOTPROD)
#ifdef __LITTLE_ENDIAN__
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdotq_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint8x16_t __s1 = __p1; \
- uint8x8_t __s2 = __p2; \
- uint32x4_t __ret; \
-uint8x8_t __reint = __s2; \
-uint32x4_t __reint1 = __builtin_shufflevector(*(uint32x2_t *) &__reint, *(uint32x2_t *) &__reint, __p3, __p3, __p3, __p3); \
- __ret = vdotq_u32(__s0, __s1, *(uint8x16_t *) &__reint1); \
- __ret; \
+#define vdotq_lane_u32(__p0_169, __p1_169, __p2_169, __p3_169) __extension__ ({ \
+ uint32x4_t __s0_169 = __p0_169; \
+ uint8x16_t __s1_169 = __p1_169; \
+ uint8x8_t __s2_169 = __p2_169; \
+ uint32x4_t __ret_169; \
+uint8x8_t __reint_169 = __s2_169; \
+uint32x4_t __reint1_169 = splatq_lane_u32(*(uint32x2_t *) &__reint_169, __p3_169); \
+ __ret_169 = vdotq_u32(__s0_169, __s1_169, *(uint8x16_t *) &__reint1_169); \
+ __ret_169; \
})
#else
-#define vdotq_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint8x16_t __s1 = __p1; \
- uint8x8_t __s2 = __p2; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint32x4_t __ret; \
-uint8x8_t __reint = __rev2; \
-uint32x4_t __reint1 = __builtin_shufflevector(*(uint32x2_t *) &__reint, *(uint32x2_t *) &__reint, __p3, __p3, __p3, __p3); \
- __ret = __noswap_vdotq_u32(__rev0, __rev1, *(uint8x16_t *) &__reint1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdotq_lane_u32(__p0_170, __p1_170, __p2_170, __p3_170) __extension__ ({ \
+ uint32x4_t __s0_170 = __p0_170; \
+ uint8x16_t __s1_170 = __p1_170; \
+ uint8x8_t __s2_170 = __p2_170; \
+ uint32x4_t __rev0_170; __rev0_170 = __builtin_shufflevector(__s0_170, __s0_170, 3, 2, 1, 0); \
+ uint8x16_t __rev1_170; __rev1_170 = __builtin_shufflevector(__s1_170, __s1_170, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x8_t __rev2_170; __rev2_170 = __builtin_shufflevector(__s2_170, __s2_170, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint32x4_t __ret_170; \
+uint8x8_t __reint_170 = __rev2_170; \
+uint32x4_t __reint1_170 = __noswap_splatq_lane_u32(*(uint32x2_t *) &__reint_170, __p3_170); \
+ __ret_170 = __noswap_vdotq_u32(__rev0_170, __rev1_170, *(uint8x16_t *) &__reint1_170); \
+ __ret_170 = __builtin_shufflevector(__ret_170, __ret_170, 3, 2, 1, 0); \
+ __ret_170; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdotq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int8x16_t __s1 = __p1; \
- int8x8_t __s2 = __p2; \
- int32x4_t __ret; \
-int8x8_t __reint = __s2; \
-int32x4_t __reint1 = __builtin_shufflevector(*(uint32x2_t *) &__reint, *(uint32x2_t *) &__reint, __p3, __p3, __p3, __p3); \
- __ret = vdotq_s32(__s0, __s1, *(int8x16_t *) &__reint1); \
- __ret; \
+#define vdotq_lane_s32(__p0_171, __p1_171, __p2_171, __p3_171) __extension__ ({ \
+ int32x4_t __s0_171 = __p0_171; \
+ int8x16_t __s1_171 = __p1_171; \
+ int8x8_t __s2_171 = __p2_171; \
+ int32x4_t __ret_171; \
+int8x8_t __reint_171 = __s2_171; \
+int32x4_t __reint1_171 = splatq_lane_s32(*(int32x2_t *) &__reint_171, __p3_171); \
+ __ret_171 = vdotq_s32(__s0_171, __s1_171, *(int8x16_t *) &__reint1_171); \
+ __ret_171; \
})
#else
-#define vdotq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int8x16_t __s1 = __p1; \
- int8x8_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x4_t __ret; \
-int8x8_t __reint = __rev2; \
-int32x4_t __reint1 = __builtin_shufflevector(*(uint32x2_t *) &__reint, *(uint32x2_t *) &__reint, __p3, __p3, __p3, __p3); \
- __ret = __noswap_vdotq_s32(__rev0, __rev1, *(int8x16_t *) &__reint1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdotq_lane_s32(__p0_172, __p1_172, __p2_172, __p3_172) __extension__ ({ \
+ int32x4_t __s0_172 = __p0_172; \
+ int8x16_t __s1_172 = __p1_172; \
+ int8x8_t __s2_172 = __p2_172; \
+ int32x4_t __rev0_172; __rev0_172 = __builtin_shufflevector(__s0_172, __s0_172, 3, 2, 1, 0); \
+ int8x16_t __rev1_172; __rev1_172 = __builtin_shufflevector(__s1_172, __s1_172, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x8_t __rev2_172; __rev2_172 = __builtin_shufflevector(__s2_172, __s2_172, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_172; \
+int8x8_t __reint_172 = __rev2_172; \
+int32x4_t __reint1_172 = __noswap_splatq_lane_s32(*(int32x2_t *) &__reint_172, __p3_172); \
+ __ret_172 = __noswap_vdotq_s32(__rev0_172, __rev1_172, *(int8x16_t *) &__reint1_172); \
+ __ret_172 = __builtin_shufflevector(__ret_172, __ret_172, 3, 2, 1, 0); \
+ __ret_172; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdot_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint8x8_t __s1 = __p1; \
- uint8x8_t __s2 = __p2; \
- uint32x2_t __ret; \
-uint8x8_t __reint = __s2; \
-uint32x2_t __reint1 = __builtin_shufflevector(*(uint32x2_t *) &__reint, *(uint32x2_t *) &__reint, __p3, __p3); \
- __ret = vdot_u32(__s0, __s1, *(uint8x8_t *) &__reint1); \
- __ret; \
+#define vdot_lane_u32(__p0_173, __p1_173, __p2_173, __p3_173) __extension__ ({ \
+ uint32x2_t __s0_173 = __p0_173; \
+ uint8x8_t __s1_173 = __p1_173; \
+ uint8x8_t __s2_173 = __p2_173; \
+ uint32x2_t __ret_173; \
+uint8x8_t __reint_173 = __s2_173; \
+uint32x2_t __reint1_173 = splat_lane_u32(*(uint32x2_t *) &__reint_173, __p3_173); \
+ __ret_173 = vdot_u32(__s0_173, __s1_173, *(uint8x8_t *) &__reint1_173); \
+ __ret_173; \
})
#else
-#define vdot_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint8x8_t __s1 = __p1; \
- uint8x8_t __s2 = __p2; \
- uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint32x2_t __ret; \
-uint8x8_t __reint = __rev2; \
-uint32x2_t __reint1 = __builtin_shufflevector(*(uint32x2_t *) &__reint, *(uint32x2_t *) &__reint, __p3, __p3); \
- __ret = __noswap_vdot_u32(__rev0, __rev1, *(uint8x8_t *) &__reint1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdot_lane_u32(__p0_174, __p1_174, __p2_174, __p3_174) __extension__ ({ \
+ uint32x2_t __s0_174 = __p0_174; \
+ uint8x8_t __s1_174 = __p1_174; \
+ uint8x8_t __s2_174 = __p2_174; \
+ uint32x2_t __rev0_174; __rev0_174 = __builtin_shufflevector(__s0_174, __s0_174, 1, 0); \
+ uint8x8_t __rev1_174; __rev1_174 = __builtin_shufflevector(__s1_174, __s1_174, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x8_t __rev2_174; __rev2_174 = __builtin_shufflevector(__s2_174, __s2_174, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint32x2_t __ret_174; \
+uint8x8_t __reint_174 = __rev2_174; \
+uint32x2_t __reint1_174 = __noswap_splat_lane_u32(*(uint32x2_t *) &__reint_174, __p3_174); \
+ __ret_174 = __noswap_vdot_u32(__rev0_174, __rev1_174, *(uint8x8_t *) &__reint1_174); \
+ __ret_174 = __builtin_shufflevector(__ret_174, __ret_174, 1, 0); \
+ __ret_174; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdot_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int8x8_t __s1 = __p1; \
- int8x8_t __s2 = __p2; \
- int32x2_t __ret; \
-int8x8_t __reint = __s2; \
-int32x2_t __reint1 = __builtin_shufflevector(*(uint32x2_t *) &__reint, *(uint32x2_t *) &__reint, __p3, __p3); \
- __ret = vdot_s32(__s0, __s1, *(int8x8_t *) &__reint1); \
- __ret; \
+#define vdot_lane_s32(__p0_175, __p1_175, __p2_175, __p3_175) __extension__ ({ \
+ int32x2_t __s0_175 = __p0_175; \
+ int8x8_t __s1_175 = __p1_175; \
+ int8x8_t __s2_175 = __p2_175; \
+ int32x2_t __ret_175; \
+int8x8_t __reint_175 = __s2_175; \
+int32x2_t __reint1_175 = splat_lane_s32(*(int32x2_t *) &__reint_175, __p3_175); \
+ __ret_175 = vdot_s32(__s0_175, __s1_175, *(int8x8_t *) &__reint1_175); \
+ __ret_175; \
})
#else
-#define vdot_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int8x8_t __s1 = __p1; \
- int8x8_t __s2 = __p2; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x2_t __ret; \
-int8x8_t __reint = __rev2; \
-int32x2_t __reint1 = __builtin_shufflevector(*(uint32x2_t *) &__reint, *(uint32x2_t *) &__reint, __p3, __p3); \
- __ret = __noswap_vdot_s32(__rev0, __rev1, *(int8x8_t *) &__reint1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdot_lane_s32(__p0_176, __p1_176, __p2_176, __p3_176) __extension__ ({ \
+ int32x2_t __s0_176 = __p0_176; \
+ int8x8_t __s1_176 = __p1_176; \
+ int8x8_t __s2_176 = __p2_176; \
+ int32x2_t __rev0_176; __rev0_176 = __builtin_shufflevector(__s0_176, __s0_176, 1, 0); \
+ int8x8_t __rev1_176; __rev1_176 = __builtin_shufflevector(__s1_176, __s1_176, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x8_t __rev2_176; __rev2_176 = __builtin_shufflevector(__s2_176, __s2_176, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x2_t __ret_176; \
+int8x8_t __reint_176 = __rev2_176; \
+int32x2_t __reint1_176 = __noswap_splat_lane_s32(*(int32x2_t *) &__reint_176, __p3_176); \
+ __ret_176 = __noswap_vdot_s32(__rev0_176, __rev1_176, *(int8x8_t *) &__reint1_176); \
+ __ret_176 = __builtin_shufflevector(__ret_176, __ret_176, 1, 0); \
+ __ret_176; \
})
#endif
#endif
#if defined(__ARM_FEATURE_DOTPROD) && defined(__aarch64__)
#ifdef __LITTLE_ENDIAN__
-#define vdotq_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint8x16_t __s1 = __p1; \
- uint8x16_t __s2 = __p2; \
- uint32x4_t __ret; \
-uint8x16_t __reint = __s2; \
-uint32x4_t __reint1 = __builtin_shufflevector(*(uint32x4_t *) &__reint, *(uint32x4_t *) &__reint, __p3, __p3, __p3, __p3); \
- __ret = vdotq_u32(__s0, __s1, *(uint8x16_t *) &__reint1); \
- __ret; \
+#define vdotq_laneq_u32(__p0_177, __p1_177, __p2_177, __p3_177) __extension__ ({ \
+ uint32x4_t __s0_177 = __p0_177; \
+ uint8x16_t __s1_177 = __p1_177; \
+ uint8x16_t __s2_177 = __p2_177; \
+ uint32x4_t __ret_177; \
+uint8x16_t __reint_177 = __s2_177; \
+uint32x4_t __reint1_177 = splatq_laneq_u32(*(uint32x4_t *) &__reint_177, __p3_177); \
+ __ret_177 = vdotq_u32(__s0_177, __s1_177, *(uint8x16_t *) &__reint1_177); \
+ __ret_177; \
})
#else
-#define vdotq_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint8x16_t __s1 = __p1; \
- uint8x16_t __s2 = __p2; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint32x4_t __ret; \
-uint8x16_t __reint = __rev2; \
-uint32x4_t __reint1 = __builtin_shufflevector(*(uint32x4_t *) &__reint, *(uint32x4_t *) &__reint, __p3, __p3, __p3, __p3); \
- __ret = __noswap_vdotq_u32(__rev0, __rev1, *(uint8x16_t *) &__reint1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdotq_laneq_u32(__p0_178, __p1_178, __p2_178, __p3_178) __extension__ ({ \
+ uint32x4_t __s0_178 = __p0_178; \
+ uint8x16_t __s1_178 = __p1_178; \
+ uint8x16_t __s2_178 = __p2_178; \
+ uint32x4_t __rev0_178; __rev0_178 = __builtin_shufflevector(__s0_178, __s0_178, 3, 2, 1, 0); \
+ uint8x16_t __rev1_178; __rev1_178 = __builtin_shufflevector(__s1_178, __s1_178, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __rev2_178; __rev2_178 = __builtin_shufflevector(__s2_178, __s2_178, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint32x4_t __ret_178; \
+uint8x16_t __reint_178 = __rev2_178; \
+uint32x4_t __reint1_178 = __noswap_splatq_laneq_u32(*(uint32x4_t *) &__reint_178, __p3_178); \
+ __ret_178 = __noswap_vdotq_u32(__rev0_178, __rev1_178, *(uint8x16_t *) &__reint1_178); \
+ __ret_178 = __builtin_shufflevector(__ret_178, __ret_178, 3, 2, 1, 0); \
+ __ret_178; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdotq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int8x16_t __s1 = __p1; \
- int8x16_t __s2 = __p2; \
- int32x4_t __ret; \
-int8x16_t __reint = __s2; \
-int32x4_t __reint1 = __builtin_shufflevector(*(uint32x4_t *) &__reint, *(uint32x4_t *) &__reint, __p3, __p3, __p3, __p3); \
- __ret = vdotq_s32(__s0, __s1, *(int8x16_t *) &__reint1); \
- __ret; \
+#define vdotq_laneq_s32(__p0_179, __p1_179, __p2_179, __p3_179) __extension__ ({ \
+ int32x4_t __s0_179 = __p0_179; \
+ int8x16_t __s1_179 = __p1_179; \
+ int8x16_t __s2_179 = __p2_179; \
+ int32x4_t __ret_179; \
+int8x16_t __reint_179 = __s2_179; \
+int32x4_t __reint1_179 = splatq_laneq_s32(*(int32x4_t *) &__reint_179, __p3_179); \
+ __ret_179 = vdotq_s32(__s0_179, __s1_179, *(int8x16_t *) &__reint1_179); \
+ __ret_179; \
})
#else
-#define vdotq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int8x16_t __s1 = __p1; \
- int8x16_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x16_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x4_t __ret; \
-int8x16_t __reint = __rev2; \
-int32x4_t __reint1 = __builtin_shufflevector(*(uint32x4_t *) &__reint, *(uint32x4_t *) &__reint, __p3, __p3, __p3, __p3); \
- __ret = __noswap_vdotq_s32(__rev0, __rev1, *(int8x16_t *) &__reint1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdotq_laneq_s32(__p0_180, __p1_180, __p2_180, __p3_180) __extension__ ({ \
+ int32x4_t __s0_180 = __p0_180; \
+ int8x16_t __s1_180 = __p1_180; \
+ int8x16_t __s2_180 = __p2_180; \
+ int32x4_t __rev0_180; __rev0_180 = __builtin_shufflevector(__s0_180, __s0_180, 3, 2, 1, 0); \
+ int8x16_t __rev1_180; __rev1_180 = __builtin_shufflevector(__s1_180, __s1_180, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __rev2_180; __rev2_180 = __builtin_shufflevector(__s2_180, __s2_180, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_180; \
+int8x16_t __reint_180 = __rev2_180; \
+int32x4_t __reint1_180 = __noswap_splatq_laneq_s32(*(int32x4_t *) &__reint_180, __p3_180); \
+ __ret_180 = __noswap_vdotq_s32(__rev0_180, __rev1_180, *(int8x16_t *) &__reint1_180); \
+ __ret_180 = __builtin_shufflevector(__ret_180, __ret_180, 3, 2, 1, 0); \
+ __ret_180; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdot_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint8x8_t __s1 = __p1; \
- uint8x16_t __s2 = __p2; \
- uint32x2_t __ret; \
-uint8x16_t __reint = __s2; \
-uint32x2_t __reint1 = __builtin_shufflevector(*(uint32x4_t *) &__reint, *(uint32x4_t *) &__reint, __p3, __p3); \
- __ret = vdot_u32(__s0, __s1, *(uint8x8_t *) &__reint1); \
- __ret; \
+#define vdot_laneq_u32(__p0_181, __p1_181, __p2_181, __p3_181) __extension__ ({ \
+ uint32x2_t __s0_181 = __p0_181; \
+ uint8x8_t __s1_181 = __p1_181; \
+ uint8x16_t __s2_181 = __p2_181; \
+ uint32x2_t __ret_181; \
+uint8x16_t __reint_181 = __s2_181; \
+uint32x2_t __reint1_181 = splat_laneq_u32(*(uint32x4_t *) &__reint_181, __p3_181); \
+ __ret_181 = vdot_u32(__s0_181, __s1_181, *(uint8x8_t *) &__reint1_181); \
+ __ret_181; \
})
#else
-#define vdot_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint8x8_t __s1 = __p1; \
- uint8x16_t __s2 = __p2; \
- uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint32x2_t __ret; \
-uint8x16_t __reint = __rev2; \
-uint32x2_t __reint1 = __builtin_shufflevector(*(uint32x4_t *) &__reint, *(uint32x4_t *) &__reint, __p3, __p3); \
- __ret = __noswap_vdot_u32(__rev0, __rev1, *(uint8x8_t *) &__reint1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdot_laneq_u32(__p0_182, __p1_182, __p2_182, __p3_182) __extension__ ({ \
+ uint32x2_t __s0_182 = __p0_182; \
+ uint8x8_t __s1_182 = __p1_182; \
+ uint8x16_t __s2_182 = __p2_182; \
+ uint32x2_t __rev0_182; __rev0_182 = __builtin_shufflevector(__s0_182, __s0_182, 1, 0); \
+ uint8x8_t __rev1_182; __rev1_182 = __builtin_shufflevector(__s1_182, __s1_182, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __rev2_182; __rev2_182 = __builtin_shufflevector(__s2_182, __s2_182, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint32x2_t __ret_182; \
+uint8x16_t __reint_182 = __rev2_182; \
+uint32x2_t __reint1_182 = __noswap_splat_laneq_u32(*(uint32x4_t *) &__reint_182, __p3_182); \
+ __ret_182 = __noswap_vdot_u32(__rev0_182, __rev1_182, *(uint8x8_t *) &__reint1_182); \
+ __ret_182 = __builtin_shufflevector(__ret_182, __ret_182, 1, 0); \
+ __ret_182; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdot_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int8x8_t __s1 = __p1; \
- int8x16_t __s2 = __p2; \
- int32x2_t __ret; \
-int8x16_t __reint = __s2; \
-int32x2_t __reint1 = __builtin_shufflevector(*(uint32x4_t *) &__reint, *(uint32x4_t *) &__reint, __p3, __p3); \
- __ret = vdot_s32(__s0, __s1, *(int8x8_t *) &__reint1); \
- __ret; \
+#define vdot_laneq_s32(__p0_183, __p1_183, __p2_183, __p3_183) __extension__ ({ \
+ int32x2_t __s0_183 = __p0_183; \
+ int8x8_t __s1_183 = __p1_183; \
+ int8x16_t __s2_183 = __p2_183; \
+ int32x2_t __ret_183; \
+int8x16_t __reint_183 = __s2_183; \
+int32x2_t __reint1_183 = splat_laneq_s32(*(int32x4_t *) &__reint_183, __p3_183); \
+ __ret_183 = vdot_s32(__s0_183, __s1_183, *(int8x8_t *) &__reint1_183); \
+ __ret_183; \
})
#else
-#define vdot_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int8x8_t __s1 = __p1; \
- int8x16_t __s2 = __p2; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x16_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x2_t __ret; \
-int8x16_t __reint = __rev2; \
-int32x2_t __reint1 = __builtin_shufflevector(*(uint32x4_t *) &__reint, *(uint32x4_t *) &__reint, __p3, __p3); \
- __ret = __noswap_vdot_s32(__rev0, __rev1, *(int8x8_t *) &__reint1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdot_laneq_s32(__p0_184, __p1_184, __p2_184, __p3_184) __extension__ ({ \
+ int32x2_t __s0_184 = __p0_184; \
+ int8x8_t __s1_184 = __p1_184; \
+ int8x16_t __s2_184 = __p2_184; \
+ int32x2_t __rev0_184; __rev0_184 = __builtin_shufflevector(__s0_184, __s0_184, 1, 0); \
+ int8x8_t __rev1_184; __rev1_184 = __builtin_shufflevector(__s1_184, __s1_184, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __rev2_184; __rev2_184 = __builtin_shufflevector(__s2_184, __s2_184, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x2_t __ret_184; \
+int8x16_t __reint_184 = __rev2_184; \
+int32x2_t __reint1_184 = __noswap_splat_laneq_s32(*(int32x4_t *) &__reint_184, __p3_184); \
+ __ret_184 = __noswap_vdot_s32(__rev0_184, __rev1_184, *(int8x8_t *) &__reint1_184); \
+ __ret_184 = __builtin_shufflevector(__ret_184, __ret_184, 1, 0); \
+ __ret_184; \
})
#endif
#endif
#endif
+#if defined(__ARM_FEATURE_FP16FML) && defined(__aarch64__)
+#ifdef __LITTLE_ENDIAN__
+__ai float32x4_t vfmlalq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vfmlalq_high_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
+ return __ret;
+}
+#else
+__ai float32x4_t vfmlalq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vfmlalq_high_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai float32x4_t __noswap_vfmlalq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vfmlalq_high_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float32x2_t vfmlal_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vfmlal_high_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
+ return __ret;
+}
+#else
+__ai float32x2_t vfmlal_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
+ float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vfmlal_high_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
+ return __ret;
+}
+__ai float32x2_t __noswap_vfmlal_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vfmlal_high_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float32x4_t vfmlalq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vfmlalq_low_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
+ return __ret;
+}
+#else
+__ai float32x4_t vfmlalq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vfmlalq_low_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai float32x4_t __noswap_vfmlalq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vfmlalq_low_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float32x2_t vfmlal_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vfmlal_low_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
+ return __ret;
+}
+#else
+__ai float32x2_t vfmlal_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
+ float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vfmlal_low_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
+ return __ret;
+}
+__ai float32x2_t __noswap_vfmlal_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vfmlal_low_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float32x4_t vfmlslq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vfmlslq_high_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
+ return __ret;
+}
+#else
+__ai float32x4_t vfmlslq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vfmlslq_high_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai float32x4_t __noswap_vfmlslq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vfmlslq_high_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float32x2_t vfmlsl_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vfmlsl_high_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
+ return __ret;
+}
+#else
+__ai float32x2_t vfmlsl_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
+ float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vfmlsl_high_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
+ return __ret;
+}
+__ai float32x2_t __noswap_vfmlsl_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vfmlsl_high_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float32x4_t vfmlslq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vfmlslq_low_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
+ return __ret;
+}
+#else
+__ai float32x4_t vfmlslq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
+ float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vfmlslq_low_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai float32x4_t __noswap_vfmlslq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
+ float32x4_t __ret;
+ __ret = (float32x4_t) __builtin_neon_vfmlslq_low_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float32x2_t vfmlsl_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vfmlsl_low_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
+ return __ret;
+}
+#else
+__ai float32x2_t vfmlsl_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
+ float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vfmlsl_low_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
+ return __ret;
+}
+__ai float32x2_t __noswap_vfmlsl_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
+ float32x2_t __ret;
+ __ret = (float32x2_t) __builtin_neon_vfmlsl_low_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
+ return __ret;
+}
+#endif
+
+#endif
#if defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC)
#ifdef __LITTLE_ENDIAN__
__ai float16x8_t vabdq_f16(float16x8_t __p0, float16x8_t __p1) {
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
- return __ret;
-}
-#else
-__ai float16x8_t vmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
- float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
- return __ret;
-}
-#else
-__ai float16x4_t vmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vmaxnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai float16x8_t vminq_f16(float16x8_t __p0, float16x8_t __p1) {
float16x8_t __ret;
__ret = (float16x8_t) __builtin_neon_vminq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
- return __ret;
-}
-#else
-__ai float16x8_t vminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
- float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vminnm_f16(float16x4_t __p0, float16x4_t __p1) {
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
- return __ret;
-}
-#else
-__ai float16x4_t vminnm_f16(float16x4_t __p0, float16x4_t __p1) {
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vminnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai float16x8_t vmulq_f16(float16x8_t __p0, float16x8_t __p1) {
float16x8_t __ret;
__ret = __p0 * __p1;
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulq_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8_t __s0 = __p0; \
- float16x4_t __s1 = __p1; \
- float16x8_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmulq_lane_f16(__p0_185, __p1_185, __p2_185) __extension__ ({ \
+ float16x8_t __s0_185 = __p0_185; \
+ float16x4_t __s1_185 = __p1_185; \
+ float16x8_t __ret_185; \
+ __ret_185 = __s0_185 * splatq_lane_f16(__s1_185, __p2_185); \
+ __ret_185; \
})
#else
-#define vmulq_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8_t __s0 = __p0; \
- float16x4_t __s1 = __p1; \
- float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- float16x8_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmulq_lane_f16(__p0_186, __p1_186, __p2_186) __extension__ ({ \
+ float16x8_t __s0_186 = __p0_186; \
+ float16x4_t __s1_186 = __p1_186; \
+ float16x8_t __rev0_186; __rev0_186 = __builtin_shufflevector(__s0_186, __s0_186, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x4_t __rev1_186; __rev1_186 = __builtin_shufflevector(__s1_186, __s1_186, 3, 2, 1, 0); \
+ float16x8_t __ret_186; \
+ __ret_186 = __rev0_186 * __noswap_splatq_lane_f16(__rev1_186, __p2_186); \
+ __ret_186 = __builtin_shufflevector(__ret_186, __ret_186, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_186; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmul_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x4_t __s1 = __p1; \
- float16x4_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmul_lane_f16(__p0_187, __p1_187, __p2_187) __extension__ ({ \
+ float16x4_t __s0_187 = __p0_187; \
+ float16x4_t __s1_187 = __p1_187; \
+ float16x4_t __ret_187; \
+ __ret_187 = __s0_187 * splat_lane_f16(__s1_187, __p2_187); \
+ __ret_187; \
})
#else
-#define vmul_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x4_t __s1 = __p1; \
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- float16x4_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmul_lane_f16(__p0_188, __p1_188, __p2_188) __extension__ ({ \
+ float16x4_t __s0_188 = __p0_188; \
+ float16x4_t __s1_188 = __p1_188; \
+ float16x4_t __rev0_188; __rev0_188 = __builtin_shufflevector(__s0_188, __s0_188, 3, 2, 1, 0); \
+ float16x4_t __rev1_188; __rev1_188 = __builtin_shufflevector(__s1_188, __s1_188, 3, 2, 1, 0); \
+ float16x4_t __ret_188; \
+ __ret_188 = __rev0_188 * __noswap_splat_lane_f16(__rev1_188, __p2_188); \
+ __ret_188 = __builtin_shufflevector(__ret_188, __ret_188, 3, 2, 1, 0); \
+ __ret_188; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vrndq_f16(float16x8_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vrndq_v((int8x16_t)__p0, 40);
- return __ret;
-}
-#else
-__ai float16x8_t vrndq_f16(float16x8_t __p0) {
- float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vrndq_v((int8x16_t)__rev0, 40);
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vrnd_f16(float16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vrnd_v((int8x8_t)__p0, 8);
- return __ret;
-}
-#else
-__ai float16x4_t vrnd_f16(float16x4_t __p0) {
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vrnd_v((int8x8_t)__rev0, 8);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vrndaq_f16(float16x8_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vrndaq_v((int8x16_t)__p0, 40);
- return __ret;
-}
-#else
-__ai float16x8_t vrndaq_f16(float16x8_t __p0) {
- float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vrndaq_v((int8x16_t)__rev0, 40);
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vrnda_f16(float16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vrnda_v((int8x8_t)__p0, 8);
- return __ret;
-}
-#else
-__ai float16x4_t vrnda_f16(float16x4_t __p0) {
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vrnda_v((int8x8_t)__rev0, 8);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vrndmq_f16(float16x8_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vrndmq_v((int8x16_t)__p0, 40);
- return __ret;
-}
-#else
-__ai float16x8_t vrndmq_f16(float16x8_t __p0) {
- float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vrndmq_v((int8x16_t)__rev0, 40);
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vrndm_f16(float16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vrndm_v((int8x8_t)__p0, 8);
- return __ret;
-}
-#else
-__ai float16x4_t vrndm_f16(float16x4_t __p0) {
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vrndm_v((int8x8_t)__rev0, 8);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vrndnq_f16(float16x8_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vrndnq_v((int8x16_t)__p0, 40);
- return __ret;
-}
-#else
-__ai float16x8_t vrndnq_f16(float16x8_t __p0) {
- float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vrndnq_v((int8x16_t)__rev0, 40);
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vrndn_f16(float16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vrndn_v((int8x8_t)__p0, 8);
- return __ret;
-}
-#else
-__ai float16x4_t vrndn_f16(float16x4_t __p0) {
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vrndn_v((int8x8_t)__rev0, 8);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vrndpq_f16(float16x8_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vrndpq_v((int8x16_t)__p0, 40);
- return __ret;
-}
-#else
-__ai float16x8_t vrndpq_f16(float16x8_t __p0) {
- float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vrndpq_v((int8x16_t)__rev0, 40);
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vrndp_f16(float16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vrndp_v((int8x8_t)__p0, 8);
- return __ret;
-}
-#else
-__ai float16x4_t vrndp_f16(float16x4_t __p0) {
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vrndp_v((int8x8_t)__rev0, 8);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x8_t vrndxq_f16(float16x8_t __p0) {
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vrndxq_v((int8x16_t)__p0, 40);
- return __ret;
-}
-#else
-__ai float16x8_t vrndxq_f16(float16x8_t __p0) {
- float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- float16x8_t __ret;
- __ret = (float16x8_t) __builtin_neon_vrndxq_v((int8x16_t)__rev0, 40);
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float16x4_t vrndx_f16(float16x4_t __p0) {
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vrndx_v((int8x8_t)__p0, 8);
- return __ret;
-}
-#else
-__ai float16x4_t vrndx_f16(float16x4_t __p0) {
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- float16x4_t __ret;
- __ret = (float16x4_t) __builtin_neon_vrndx_v((int8x8_t)__rev0, 8);
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai float16x8_t vrsqrteq_f16(float16x8_t __p0) {
float16x8_t __ret;
__ret = (float16x8_t) __builtin_neon_vrsqrteq_v((int8x16_t)__p0, 40);
#define vduph_lane_f16(__p0, __p1) __extension__ ({ \
float16x4_t __s0 = __p0; \
float16_t __ret; \
- __ret = (float16_t) __builtin_neon_vduph_lane_f16((int8x8_t)__s0, __p1); \
+ __ret = (float16_t) __builtin_neon_vduph_lane_f16((float16x4_t)__s0, __p1); \
__ret; \
})
#else
float16x4_t __s0 = __p0; \
float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
float16_t __ret; \
- __ret = (float16_t) __builtin_neon_vduph_lane_f16((int8x8_t)__rev0, __p1); \
+ __ret = (float16_t) __builtin_neon_vduph_lane_f16((float16x4_t)__rev0, __p1); \
__ret; \
})
#endif
#define vduph_laneq_f16(__p0, __p1) __extension__ ({ \
float16x8_t __s0 = __p0; \
float16_t __ret; \
- __ret = (float16_t) __builtin_neon_vduph_laneq_f16((int8x16_t)__s0, __p1); \
+ __ret = (float16_t) __builtin_neon_vduph_laneq_f16((float16x8_t)__s0, __p1); \
__ret; \
})
#else
float16x8_t __s0 = __p0; \
float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
float16_t __ret; \
- __ret = (float16_t) __builtin_neon_vduph_laneq_f16((int8x16_t)__rev0, __p1); \
+ __ret = (float16_t) __builtin_neon_vduph_laneq_f16((float16x8_t)__rev0, __p1); \
__ret; \
})
#endif
float16_t __s1 = __p1; \
float16x4_t __s2 = __p2; \
float16_t __ret; \
- __ret = (float16_t) __builtin_neon_vfmah_lane_f16(__s0, __s1, (int8x8_t)__s2, __p3); \
+ __ret = (float16_t) __builtin_neon_vfmah_lane_f16(__s0, __s1, (float16x4_t)__s2, __p3); \
__ret; \
})
#else
float16x4_t __s2 = __p2; \
float16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
float16_t __ret; \
- __ret = (float16_t) __builtin_neon_vfmah_lane_f16(__s0, __s1, (int8x8_t)__rev2, __p3); \
+ __ret = (float16_t) __builtin_neon_vfmah_lane_f16(__s0, __s1, (float16x4_t)__rev2, __p3); \
__ret; \
})
#define __noswap_vfmah_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
float16_t __s1 = __p1; \
float16x4_t __s2 = __p2; \
float16_t __ret; \
- __ret = (float16_t) __builtin_neon_vfmah_lane_f16(__s0, __s1, (int8x8_t)__s2, __p3); \
+ __ret = (float16_t) __builtin_neon_vfmah_lane_f16(__s0, __s1, (float16x4_t)__s2, __p3); \
__ret; \
})
#endif
float16_t __s1 = __p1; \
float16x8_t __s2 = __p2; \
float16_t __ret; \
- __ret = (float16_t) __builtin_neon_vfmah_laneq_f16(__s0, __s1, (int8x16_t)__s2, __p3); \
+ __ret = (float16_t) __builtin_neon_vfmah_laneq_f16(__s0, __s1, (float16x8_t)__s2, __p3); \
__ret; \
})
#else
float16x8_t __s2 = __p2; \
float16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
float16_t __ret; \
- __ret = (float16_t) __builtin_neon_vfmah_laneq_f16(__s0, __s1, (int8x16_t)__rev2, __p3); \
+ __ret = (float16_t) __builtin_neon_vfmah_laneq_f16(__s0, __s1, (float16x8_t)__rev2, __p3); \
__ret; \
})
#define __noswap_vfmah_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
float16_t __s1 = __p1; \
float16x8_t __s2 = __p2; \
float16_t __ret; \
- __ret = (float16_t) __builtin_neon_vfmah_laneq_f16(__s0, __s1, (int8x16_t)__s2, __p3); \
+ __ret = (float16_t) __builtin_neon_vfmah_laneq_f16(__s0, __s1, (float16x8_t)__s2, __p3); \
__ret; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vfmsh_lane_f16(__p0_0, __p1_0, __p2_0, __p3_0) __extension__ ({ \
- float16_t __s0_0 = __p0_0; \
- float16_t __s1_0 = __p1_0; \
- float16x4_t __s2_0 = __p2_0; \
- float16_t __ret_0; \
- __ret_0 = vfmah_lane_f16(__s0_0, -__s1_0, __s2_0, __p3_0); \
- __ret_0; \
+#define vfmsh_lane_f16(__p0_189, __p1_189, __p2_189, __p3_189) __extension__ ({ \
+ float16_t __s0_189 = __p0_189; \
+ float16_t __s1_189 = __p1_189; \
+ float16x4_t __s2_189 = __p2_189; \
+ float16_t __ret_189; \
+ __ret_189 = vfmah_lane_f16(__s0_189, -__s1_189, __s2_189, __p3_189); \
+ __ret_189; \
})
#else
-#define vfmsh_lane_f16(__p0_1, __p1_1, __p2_1, __p3_1) __extension__ ({ \
- float16_t __s0_1 = __p0_1; \
- float16_t __s1_1 = __p1_1; \
- float16x4_t __s2_1 = __p2_1; \
- float16x4_t __rev2_1; __rev2_1 = __builtin_shufflevector(__s2_1, __s2_1, 3, 2, 1, 0); \
- float16_t __ret_1; \
- __ret_1 = __noswap_vfmah_lane_f16(__s0_1, -__s1_1, __rev2_1, __p3_1); \
- __ret_1; \
+#define vfmsh_lane_f16(__p0_190, __p1_190, __p2_190, __p3_190) __extension__ ({ \
+ float16_t __s0_190 = __p0_190; \
+ float16_t __s1_190 = __p1_190; \
+ float16x4_t __s2_190 = __p2_190; \
+ float16x4_t __rev2_190; __rev2_190 = __builtin_shufflevector(__s2_190, __s2_190, 3, 2, 1, 0); \
+ float16_t __ret_190; \
+ __ret_190 = __noswap_vfmah_lane_f16(__s0_190, -__s1_190, __rev2_190, __p3_190); \
+ __ret_190; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vfmsq_lane_f16(__p0_2, __p1_2, __p2_2, __p3_2) __extension__ ({ \
- float16x8_t __s0_2 = __p0_2; \
- float16x8_t __s1_2 = __p1_2; \
- float16x4_t __s2_2 = __p2_2; \
- float16x8_t __ret_2; \
- __ret_2 = vfmaq_lane_f16(__s0_2, -__s1_2, __s2_2, __p3_2); \
- __ret_2; \
+#define vfmsq_lane_f16(__p0_191, __p1_191, __p2_191, __p3_191) __extension__ ({ \
+ float16x8_t __s0_191 = __p0_191; \
+ float16x8_t __s1_191 = __p1_191; \
+ float16x4_t __s2_191 = __p2_191; \
+ float16x8_t __ret_191; \
+ __ret_191 = vfmaq_lane_f16(__s0_191, -__s1_191, __s2_191, __p3_191); \
+ __ret_191; \
})
#else
-#define vfmsq_lane_f16(__p0_3, __p1_3, __p2_3, __p3_3) __extension__ ({ \
- float16x8_t __s0_3 = __p0_3; \
- float16x8_t __s1_3 = __p1_3; \
- float16x4_t __s2_3 = __p2_3; \
- float16x8_t __rev0_3; __rev0_3 = __builtin_shufflevector(__s0_3, __s0_3, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x8_t __rev1_3; __rev1_3 = __builtin_shufflevector(__s1_3, __s1_3, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x4_t __rev2_3; __rev2_3 = __builtin_shufflevector(__s2_3, __s2_3, 3, 2, 1, 0); \
- float16x8_t __ret_3; \
- __ret_3 = __noswap_vfmaq_lane_f16(__rev0_3, -__rev1_3, __rev2_3, __p3_3); \
- __ret_3 = __builtin_shufflevector(__ret_3, __ret_3, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_3; \
+#define vfmsq_lane_f16(__p0_192, __p1_192, __p2_192, __p3_192) __extension__ ({ \
+ float16x8_t __s0_192 = __p0_192; \
+ float16x8_t __s1_192 = __p1_192; \
+ float16x4_t __s2_192 = __p2_192; \
+ float16x8_t __rev0_192; __rev0_192 = __builtin_shufflevector(__s0_192, __s0_192, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __rev1_192; __rev1_192 = __builtin_shufflevector(__s1_192, __s1_192, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x4_t __rev2_192; __rev2_192 = __builtin_shufflevector(__s2_192, __s2_192, 3, 2, 1, 0); \
+ float16x8_t __ret_192; \
+ __ret_192 = __noswap_vfmaq_lane_f16(__rev0_192, -__rev1_192, __rev2_192, __p3_192); \
+ __ret_192 = __builtin_shufflevector(__ret_192, __ret_192, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_192; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vfms_lane_f16(__p0_4, __p1_4, __p2_4, __p3_4) __extension__ ({ \
- float16x4_t __s0_4 = __p0_4; \
- float16x4_t __s1_4 = __p1_4; \
- float16x4_t __s2_4 = __p2_4; \
- float16x4_t __ret_4; \
- __ret_4 = vfma_lane_f16(__s0_4, -__s1_4, __s2_4, __p3_4); \
- __ret_4; \
+#define vfms_lane_f16(__p0_193, __p1_193, __p2_193, __p3_193) __extension__ ({ \
+ float16x4_t __s0_193 = __p0_193; \
+ float16x4_t __s1_193 = __p1_193; \
+ float16x4_t __s2_193 = __p2_193; \
+ float16x4_t __ret_193; \
+ __ret_193 = vfma_lane_f16(__s0_193, -__s1_193, __s2_193, __p3_193); \
+ __ret_193; \
})
#else
-#define vfms_lane_f16(__p0_5, __p1_5, __p2_5, __p3_5) __extension__ ({ \
- float16x4_t __s0_5 = __p0_5; \
- float16x4_t __s1_5 = __p1_5; \
- float16x4_t __s2_5 = __p2_5; \
- float16x4_t __rev0_5; __rev0_5 = __builtin_shufflevector(__s0_5, __s0_5, 3, 2, 1, 0); \
- float16x4_t __rev1_5; __rev1_5 = __builtin_shufflevector(__s1_5, __s1_5, 3, 2, 1, 0); \
- float16x4_t __rev2_5; __rev2_5 = __builtin_shufflevector(__s2_5, __s2_5, 3, 2, 1, 0); \
- float16x4_t __ret_5; \
- __ret_5 = __noswap_vfma_lane_f16(__rev0_5, -__rev1_5, __rev2_5, __p3_5); \
- __ret_5 = __builtin_shufflevector(__ret_5, __ret_5, 3, 2, 1, 0); \
- __ret_5; \
+#define vfms_lane_f16(__p0_194, __p1_194, __p2_194, __p3_194) __extension__ ({ \
+ float16x4_t __s0_194 = __p0_194; \
+ float16x4_t __s1_194 = __p1_194; \
+ float16x4_t __s2_194 = __p2_194; \
+ float16x4_t __rev0_194; __rev0_194 = __builtin_shufflevector(__s0_194, __s0_194, 3, 2, 1, 0); \
+ float16x4_t __rev1_194; __rev1_194 = __builtin_shufflevector(__s1_194, __s1_194, 3, 2, 1, 0); \
+ float16x4_t __rev2_194; __rev2_194 = __builtin_shufflevector(__s2_194, __s2_194, 3, 2, 1, 0); \
+ float16x4_t __ret_194; \
+ __ret_194 = __noswap_vfma_lane_f16(__rev0_194, -__rev1_194, __rev2_194, __p3_194); \
+ __ret_194 = __builtin_shufflevector(__ret_194, __ret_194, 3, 2, 1, 0); \
+ __ret_194; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vfmsh_laneq_f16(__p0_6, __p1_6, __p2_6, __p3_6) __extension__ ({ \
- float16_t __s0_6 = __p0_6; \
- float16_t __s1_6 = __p1_6; \
- float16x8_t __s2_6 = __p2_6; \
- float16_t __ret_6; \
- __ret_6 = vfmah_laneq_f16(__s0_6, -__s1_6, __s2_6, __p3_6); \
- __ret_6; \
+#define vfmsh_laneq_f16(__p0_195, __p1_195, __p2_195, __p3_195) __extension__ ({ \
+ float16_t __s0_195 = __p0_195; \
+ float16_t __s1_195 = __p1_195; \
+ float16x8_t __s2_195 = __p2_195; \
+ float16_t __ret_195; \
+ __ret_195 = vfmah_laneq_f16(__s0_195, -__s1_195, __s2_195, __p3_195); \
+ __ret_195; \
})
#else
-#define vfmsh_laneq_f16(__p0_7, __p1_7, __p2_7, __p3_7) __extension__ ({ \
- float16_t __s0_7 = __p0_7; \
- float16_t __s1_7 = __p1_7; \
- float16x8_t __s2_7 = __p2_7; \
- float16x8_t __rev2_7; __rev2_7 = __builtin_shufflevector(__s2_7, __s2_7, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16_t __ret_7; \
- __ret_7 = __noswap_vfmah_laneq_f16(__s0_7, -__s1_7, __rev2_7, __p3_7); \
- __ret_7; \
+#define vfmsh_laneq_f16(__p0_196, __p1_196, __p2_196, __p3_196) __extension__ ({ \
+ float16_t __s0_196 = __p0_196; \
+ float16_t __s1_196 = __p1_196; \
+ float16x8_t __s2_196 = __p2_196; \
+ float16x8_t __rev2_196; __rev2_196 = __builtin_shufflevector(__s2_196, __s2_196, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16_t __ret_196; \
+ __ret_196 = __noswap_vfmah_laneq_f16(__s0_196, -__s1_196, __rev2_196, __p3_196); \
+ __ret_196; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vfmsq_laneq_f16(__p0_8, __p1_8, __p2_8, __p3_8) __extension__ ({ \
- float16x8_t __s0_8 = __p0_8; \
- float16x8_t __s1_8 = __p1_8; \
- float16x8_t __s2_8 = __p2_8; \
- float16x8_t __ret_8; \
- __ret_8 = vfmaq_laneq_f16(__s0_8, -__s1_8, __s2_8, __p3_8); \
- __ret_8; \
+#define vfmsq_laneq_f16(__p0_197, __p1_197, __p2_197, __p3_197) __extension__ ({ \
+ float16x8_t __s0_197 = __p0_197; \
+ float16x8_t __s1_197 = __p1_197; \
+ float16x8_t __s2_197 = __p2_197; \
+ float16x8_t __ret_197; \
+ __ret_197 = vfmaq_laneq_f16(__s0_197, -__s1_197, __s2_197, __p3_197); \
+ __ret_197; \
})
#else
-#define vfmsq_laneq_f16(__p0_9, __p1_9, __p2_9, __p3_9) __extension__ ({ \
- float16x8_t __s0_9 = __p0_9; \
- float16x8_t __s1_9 = __p1_9; \
- float16x8_t __s2_9 = __p2_9; \
- float16x8_t __rev0_9; __rev0_9 = __builtin_shufflevector(__s0_9, __s0_9, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x8_t __rev1_9; __rev1_9 = __builtin_shufflevector(__s1_9, __s1_9, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x8_t __rev2_9; __rev2_9 = __builtin_shufflevector(__s2_9, __s2_9, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x8_t __ret_9; \
- __ret_9 = __noswap_vfmaq_laneq_f16(__rev0_9, -__rev1_9, __rev2_9, __p3_9); \
- __ret_9 = __builtin_shufflevector(__ret_9, __ret_9, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_9; \
+#define vfmsq_laneq_f16(__p0_198, __p1_198, __p2_198, __p3_198) __extension__ ({ \
+ float16x8_t __s0_198 = __p0_198; \
+ float16x8_t __s1_198 = __p1_198; \
+ float16x8_t __s2_198 = __p2_198; \
+ float16x8_t __rev0_198; __rev0_198 = __builtin_shufflevector(__s0_198, __s0_198, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __rev1_198; __rev1_198 = __builtin_shufflevector(__s1_198, __s1_198, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __rev2_198; __rev2_198 = __builtin_shufflevector(__s2_198, __s2_198, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __ret_198; \
+ __ret_198 = __noswap_vfmaq_laneq_f16(__rev0_198, -__rev1_198, __rev2_198, __p3_198); \
+ __ret_198 = __builtin_shufflevector(__ret_198, __ret_198, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_198; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vfms_laneq_f16(__p0_10, __p1_10, __p2_10, __p3_10) __extension__ ({ \
- float16x4_t __s0_10 = __p0_10; \
- float16x4_t __s1_10 = __p1_10; \
- float16x8_t __s2_10 = __p2_10; \
- float16x4_t __ret_10; \
- __ret_10 = vfma_laneq_f16(__s0_10, -__s1_10, __s2_10, __p3_10); \
- __ret_10; \
+#define vfms_laneq_f16(__p0_199, __p1_199, __p2_199, __p3_199) __extension__ ({ \
+ float16x4_t __s0_199 = __p0_199; \
+ float16x4_t __s1_199 = __p1_199; \
+ float16x8_t __s2_199 = __p2_199; \
+ float16x4_t __ret_199; \
+ __ret_199 = vfma_laneq_f16(__s0_199, -__s1_199, __s2_199, __p3_199); \
+ __ret_199; \
})
#else
-#define vfms_laneq_f16(__p0_11, __p1_11, __p2_11, __p3_11) __extension__ ({ \
- float16x4_t __s0_11 = __p0_11; \
- float16x4_t __s1_11 = __p1_11; \
- float16x8_t __s2_11 = __p2_11; \
- float16x4_t __rev0_11; __rev0_11 = __builtin_shufflevector(__s0_11, __s0_11, 3, 2, 1, 0); \
- float16x4_t __rev1_11; __rev1_11 = __builtin_shufflevector(__s1_11, __s1_11, 3, 2, 1, 0); \
- float16x8_t __rev2_11; __rev2_11 = __builtin_shufflevector(__s2_11, __s2_11, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x4_t __ret_11; \
- __ret_11 = __noswap_vfma_laneq_f16(__rev0_11, -__rev1_11, __rev2_11, __p3_11); \
- __ret_11 = __builtin_shufflevector(__ret_11, __ret_11, 3, 2, 1, 0); \
- __ret_11; \
+#define vfms_laneq_f16(__p0_200, __p1_200, __p2_200, __p3_200) __extension__ ({ \
+ float16x4_t __s0_200 = __p0_200; \
+ float16x4_t __s1_200 = __p1_200; \
+ float16x8_t __s2_200 = __p2_200; \
+ float16x4_t __rev0_200; __rev0_200 = __builtin_shufflevector(__s0_200, __s0_200, 3, 2, 1, 0); \
+ float16x4_t __rev1_200; __rev1_200 = __builtin_shufflevector(__s1_200, __s1_200, 3, 2, 1, 0); \
+ float16x8_t __rev2_200; __rev2_200 = __builtin_shufflevector(__s2_200, __s2_200, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x4_t __ret_200; \
+ __ret_200 = __noswap_vfma_laneq_f16(__rev0_200, -__rev1_200, __rev2_200, __p3_200); \
+ __ret_200 = __builtin_shufflevector(__ret_200, __ret_200, 3, 2, 1, 0); \
+ __ret_200; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulq_laneq_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8_t __s0 = __p0; \
- float16x8_t __s1 = __p1; \
- float16x8_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmulq_laneq_f16(__p0_201, __p1_201, __p2_201) __extension__ ({ \
+ float16x8_t __s0_201 = __p0_201; \
+ float16x8_t __s1_201 = __p1_201; \
+ float16x8_t __ret_201; \
+ __ret_201 = __s0_201 * splatq_laneq_f16(__s1_201, __p2_201); \
+ __ret_201; \
})
#else
-#define vmulq_laneq_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8_t __s0 = __p0; \
- float16x8_t __s1 = __p1; \
- float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x8_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmulq_laneq_f16(__p0_202, __p1_202, __p2_202) __extension__ ({ \
+ float16x8_t __s0_202 = __p0_202; \
+ float16x8_t __s1_202 = __p1_202; \
+ float16x8_t __rev0_202; __rev0_202 = __builtin_shufflevector(__s0_202, __s0_202, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __rev1_202; __rev1_202 = __builtin_shufflevector(__s1_202, __s1_202, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __ret_202; \
+ __ret_202 = __rev0_202 * __noswap_splatq_laneq_f16(__rev1_202, __p2_202); \
+ __ret_202 = __builtin_shufflevector(__ret_202, __ret_202, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_202; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmul_laneq_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x8_t __s1 = __p1; \
- float16x4_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmul_laneq_f16(__p0_203, __p1_203, __p2_203) __extension__ ({ \
+ float16x4_t __s0_203 = __p0_203; \
+ float16x8_t __s1_203 = __p1_203; \
+ float16x4_t __ret_203; \
+ __ret_203 = __s0_203 * splat_laneq_f16(__s1_203, __p2_203); \
+ __ret_203; \
})
#else
-#define vmul_laneq_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x8_t __s1 = __p1; \
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x4_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmul_laneq_f16(__p0_204, __p1_204, __p2_204) __extension__ ({ \
+ float16x4_t __s0_204 = __p0_204; \
+ float16x8_t __s1_204 = __p1_204; \
+ float16x4_t __rev0_204; __rev0_204 = __builtin_shufflevector(__s0_204, __s0_204, 3, 2, 1, 0); \
+ float16x8_t __rev1_204; __rev1_204 = __builtin_shufflevector(__s1_204, __s1_204, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x4_t __ret_204; \
+ __ret_204 = __rev0_204 * __noswap_splat_laneq_f16(__rev1_204, __p2_204); \
+ __ret_204 = __builtin_shufflevector(__ret_204, __ret_204, 3, 2, 1, 0); \
+ __ret_204; \
})
#endif
float16_t __s0 = __p0; \
float16x4_t __s1 = __p1; \
float16_t __ret; \
- __ret = (float16_t) __builtin_neon_vmulxh_lane_f16(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (float16_t) __builtin_neon_vmulxh_lane_f16(__s0, (float16x4_t)__s1, __p2); \
__ret; \
})
#else
float16x4_t __s1 = __p1; \
float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
float16_t __ret; \
- __ret = (float16_t) __builtin_neon_vmulxh_lane_f16(__s0, (int8x8_t)__rev1, __p2); \
+ __ret = (float16_t) __builtin_neon_vmulxh_lane_f16(__s0, (float16x4_t)__rev1, __p2); \
__ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulxq_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8_t __s0 = __p0; \
- float16x4_t __s1 = __p1; \
- float16x8_t __ret; \
- __ret = vmulxq_f16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vmulxq_lane_f16(__p0_205, __p1_205, __p2_205) __extension__ ({ \
+ float16x8_t __s0_205 = __p0_205; \
+ float16x4_t __s1_205 = __p1_205; \
+ float16x8_t __ret_205; \
+ __ret_205 = vmulxq_f16(__s0_205, splatq_lane_f16(__s1_205, __p2_205)); \
+ __ret_205; \
})
#else
-#define vmulxq_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8_t __s0 = __p0; \
- float16x4_t __s1 = __p1; \
- float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- float16x8_t __ret; \
- __ret = __noswap_vmulxq_f16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmulxq_lane_f16(__p0_206, __p1_206, __p2_206) __extension__ ({ \
+ float16x8_t __s0_206 = __p0_206; \
+ float16x4_t __s1_206 = __p1_206; \
+ float16x8_t __rev0_206; __rev0_206 = __builtin_shufflevector(__s0_206, __s0_206, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x4_t __rev1_206; __rev1_206 = __builtin_shufflevector(__s1_206, __s1_206, 3, 2, 1, 0); \
+ float16x8_t __ret_206; \
+ __ret_206 = __noswap_vmulxq_f16(__rev0_206, __noswap_splatq_lane_f16(__rev1_206, __p2_206)); \
+ __ret_206 = __builtin_shufflevector(__ret_206, __ret_206, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_206; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulx_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x4_t __s1 = __p1; \
- float16x4_t __ret; \
- __ret = vmulx_f16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vmulx_lane_f16(__p0_207, __p1_207, __p2_207) __extension__ ({ \
+ float16x4_t __s0_207 = __p0_207; \
+ float16x4_t __s1_207 = __p1_207; \
+ float16x4_t __ret_207; \
+ __ret_207 = vmulx_f16(__s0_207, splat_lane_f16(__s1_207, __p2_207)); \
+ __ret_207; \
})
#else
-#define vmulx_lane_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x4_t __s1 = __p1; \
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- float16x4_t __ret; \
- __ret = __noswap_vmulx_f16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmulx_lane_f16(__p0_208, __p1_208, __p2_208) __extension__ ({ \
+ float16x4_t __s0_208 = __p0_208; \
+ float16x4_t __s1_208 = __p1_208; \
+ float16x4_t __rev0_208; __rev0_208 = __builtin_shufflevector(__s0_208, __s0_208, 3, 2, 1, 0); \
+ float16x4_t __rev1_208; __rev1_208 = __builtin_shufflevector(__s1_208, __s1_208, 3, 2, 1, 0); \
+ float16x4_t __ret_208; \
+ __ret_208 = __noswap_vmulx_f16(__rev0_208, __noswap_splat_lane_f16(__rev1_208, __p2_208)); \
+ __ret_208 = __builtin_shufflevector(__ret_208, __ret_208, 3, 2, 1, 0); \
+ __ret_208; \
})
#endif
float16_t __s0 = __p0; \
float16x8_t __s1 = __p1; \
float16_t __ret; \
- __ret = (float16_t) __builtin_neon_vmulxh_laneq_f16(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (float16_t) __builtin_neon_vmulxh_laneq_f16(__s0, (float16x8_t)__s1, __p2); \
__ret; \
})
#else
float16x8_t __s1 = __p1; \
float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
float16_t __ret; \
- __ret = (float16_t) __builtin_neon_vmulxh_laneq_f16(__s0, (int8x16_t)__rev1, __p2); \
+ __ret = (float16_t) __builtin_neon_vmulxh_laneq_f16(__s0, (float16x8_t)__rev1, __p2); \
__ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulxq_laneq_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8_t __s0 = __p0; \
- float16x8_t __s1 = __p1; \
- float16x8_t __ret; \
- __ret = vmulxq_f16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vmulxq_laneq_f16(__p0_209, __p1_209, __p2_209) __extension__ ({ \
+ float16x8_t __s0_209 = __p0_209; \
+ float16x8_t __s1_209 = __p1_209; \
+ float16x8_t __ret_209; \
+ __ret_209 = vmulxq_f16(__s0_209, splatq_laneq_f16(__s1_209, __p2_209)); \
+ __ret_209; \
})
#else
-#define vmulxq_laneq_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x8_t __s0 = __p0; \
- float16x8_t __s1 = __p1; \
- float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x8_t __ret; \
- __ret = __noswap_vmulxq_f16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmulxq_laneq_f16(__p0_210, __p1_210, __p2_210) __extension__ ({ \
+ float16x8_t __s0_210 = __p0_210; \
+ float16x8_t __s1_210 = __p1_210; \
+ float16x8_t __rev0_210; __rev0_210 = __builtin_shufflevector(__s0_210, __s0_210, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __rev1_210; __rev1_210 = __builtin_shufflevector(__s1_210, __s1_210, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __ret_210; \
+ __ret_210 = __noswap_vmulxq_f16(__rev0_210, __noswap_splatq_laneq_f16(__rev1_210, __p2_210)); \
+ __ret_210 = __builtin_shufflevector(__ret_210, __ret_210, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_210; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulx_laneq_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x8_t __s1 = __p1; \
- float16x4_t __ret; \
- __ret = vmulx_f16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vmulx_laneq_f16(__p0_211, __p1_211, __p2_211) __extension__ ({ \
+ float16x4_t __s0_211 = __p0_211; \
+ float16x8_t __s1_211 = __p1_211; \
+ float16x4_t __ret_211; \
+ __ret_211 = vmulx_f16(__s0_211, splat_laneq_f16(__s1_211, __p2_211)); \
+ __ret_211; \
})
#else
-#define vmulx_laneq_f16(__p0, __p1, __p2) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x8_t __s1 = __p1; \
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x4_t __ret; \
- __ret = __noswap_vmulx_f16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmulx_laneq_f16(__p0_212, __p1_212, __p2_212) __extension__ ({ \
+ float16x4_t __s0_212 = __p0_212; \
+ float16x8_t __s1_212 = __p1_212; \
+ float16x4_t __rev0_212; __rev0_212 = __builtin_shufflevector(__s0_212, __s0_212, 3, 2, 1, 0); \
+ float16x8_t __rev1_212; __rev1_212 = __builtin_shufflevector(__s1_212, __s1_212, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x4_t __ret_212; \
+ __ret_212 = __noswap_vmulx_f16(__rev0_212, __noswap_splat_laneq_f16(__rev1_212, __p2_212)); \
+ __ret_212 = __builtin_shufflevector(__ret_212, __ret_212, 3, 2, 1, 0); \
+ __ret_212; \
})
#endif
#endif
#endif
+#if defined(__ARM_FEATURE_MATMUL_INT8)
+#ifdef __LITTLE_ENDIAN__
+__ai uint32x4_t vmmlaq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
+ uint32x4_t __ret;
+ __ret = (uint32x4_t) __builtin_neon_vmmlaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
+ return __ret;
+}
+#else
+__ai uint32x4_t vmmlaq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
+ uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint32x4_t __ret;
+ __ret = (uint32x4_t) __builtin_neon_vmmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai int32x4_t vmmlaq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
+ int32x4_t __ret;
+ __ret = (int32x4_t) __builtin_neon_vmmlaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
+ return __ret;
+}
+#else
+__ai int32x4_t vmmlaq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
+ int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ int32x4_t __ret;
+ __ret = (int32x4_t) __builtin_neon_vmmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai int32x4_t vusdotq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
+ int32x4_t __ret;
+ __ret = (int32x4_t) __builtin_neon_vusdotq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
+ return __ret;
+}
+#else
+__ai int32x4_t vusdotq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
+ int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ int32x4_t __ret;
+ __ret = (int32x4_t) __builtin_neon_vusdotq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+__ai int32x4_t __noswap_vusdotq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
+ int32x4_t __ret;
+ __ret = (int32x4_t) __builtin_neon_vusdotq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai int32x2_t vusdot_s32(int32x2_t __p0, uint8x8_t __p1, int8x8_t __p2) {
+ int32x2_t __ret;
+ __ret = (int32x2_t) __builtin_neon_vusdot_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
+ return __ret;
+}
+#else
+__ai int32x2_t vusdot_s32(int32x2_t __p0, uint8x8_t __p1, int8x8_t __p2) {
+ int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
+ uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
+ int32x2_t __ret;
+ __ret = (int32x2_t) __builtin_neon_vusdot_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 2);
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
+ return __ret;
+}
+__ai int32x2_t __noswap_vusdot_s32(int32x2_t __p0, uint8x8_t __p1, int8x8_t __p2) {
+ int32x2_t __ret;
+ __ret = (int32x2_t) __builtin_neon_vusdot_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
+ return __ret;
+}
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vusdotq_lane_s32(__p0_213, __p1_213, __p2_213, __p3_213) __extension__ ({ \
+ int32x4_t __s0_213 = __p0_213; \
+ uint8x16_t __s1_213 = __p1_213; \
+ int8x8_t __s2_213 = __p2_213; \
+ int32x4_t __ret_213; \
+int8x8_t __reint_213 = __s2_213; \
+ __ret_213 = vusdotq_s32(__s0_213, __s1_213, (int8x16_t)(splatq_lane_s32(*(int32x2_t *) &__reint_213, __p3_213))); \
+ __ret_213; \
+})
+#else
+#define vusdotq_lane_s32(__p0_214, __p1_214, __p2_214, __p3_214) __extension__ ({ \
+ int32x4_t __s0_214 = __p0_214; \
+ uint8x16_t __s1_214 = __p1_214; \
+ int8x8_t __s2_214 = __p2_214; \
+ int32x4_t __rev0_214; __rev0_214 = __builtin_shufflevector(__s0_214, __s0_214, 3, 2, 1, 0); \
+ uint8x16_t __rev1_214; __rev1_214 = __builtin_shufflevector(__s1_214, __s1_214, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x8_t __rev2_214; __rev2_214 = __builtin_shufflevector(__s2_214, __s2_214, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_214; \
+int8x8_t __reint_214 = __rev2_214; \
+ __ret_214 = __noswap_vusdotq_s32(__rev0_214, __rev1_214, (int8x16_t)(__noswap_splatq_lane_s32(*(int32x2_t *) &__reint_214, __p3_214))); \
+ __ret_214 = __builtin_shufflevector(__ret_214, __ret_214, 3, 2, 1, 0); \
+ __ret_214; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vusdot_lane_s32(__p0_215, __p1_215, __p2_215, __p3_215) __extension__ ({ \
+ int32x2_t __s0_215 = __p0_215; \
+ uint8x8_t __s1_215 = __p1_215; \
+ int8x8_t __s2_215 = __p2_215; \
+ int32x2_t __ret_215; \
+int8x8_t __reint_215 = __s2_215; \
+ __ret_215 = vusdot_s32(__s0_215, __s1_215, (int8x8_t)(splat_lane_s32(*(int32x2_t *) &__reint_215, __p3_215))); \
+ __ret_215; \
+})
+#else
+#define vusdot_lane_s32(__p0_216, __p1_216, __p2_216, __p3_216) __extension__ ({ \
+ int32x2_t __s0_216 = __p0_216; \
+ uint8x8_t __s1_216 = __p1_216; \
+ int8x8_t __s2_216 = __p2_216; \
+ int32x2_t __rev0_216; __rev0_216 = __builtin_shufflevector(__s0_216, __s0_216, 1, 0); \
+ uint8x8_t __rev1_216; __rev1_216 = __builtin_shufflevector(__s1_216, __s1_216, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x8_t __rev2_216; __rev2_216 = __builtin_shufflevector(__s2_216, __s2_216, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x2_t __ret_216; \
+int8x8_t __reint_216 = __rev2_216; \
+ __ret_216 = __noswap_vusdot_s32(__rev0_216, __rev1_216, (int8x8_t)(__noswap_splat_lane_s32(*(int32x2_t *) &__reint_216, __p3_216))); \
+ __ret_216 = __builtin_shufflevector(__ret_216, __ret_216, 1, 0); \
+ __ret_216; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai int32x4_t vusmmlaq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
+ int32x4_t __ret;
+ __ret = (int32x4_t) __builtin_neon_vusmmlaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
+ return __ret;
+}
+#else
+__ai int32x4_t vusmmlaq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
+ int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
+ uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ int32x4_t __ret;
+ __ret = (int32x4_t) __builtin_neon_vusmmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
+ return __ret;
+}
+#endif
+
+#endif
#if defined(__ARM_FEATURE_QRDMX)
#ifdef __LITTLE_ENDIAN__
__ai int32x4_t vqrdmlahq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlahq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = vqaddq_s32(__s0, vqrdmulhq_s32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3))); \
- __ret; \
+#define vqrdmlahq_lane_s32(__p0_217, __p1_217, __p2_217, __p3_217) __extension__ ({ \
+ int32x4_t __s0_217 = __p0_217; \
+ int32x4_t __s1_217 = __p1_217; \
+ int32x2_t __s2_217 = __p2_217; \
+ int32x4_t __ret_217; \
+ __ret_217 = vqaddq_s32(__s0_217, vqrdmulhq_s32(__s1_217, splatq_lane_s32(__s2_217, __p3_217))); \
+ __ret_217; \
})
#else
-#define vqrdmlahq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqaddq_s32(__rev0, __noswap_vqrdmulhq_s32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqrdmlahq_lane_s32(__p0_218, __p1_218, __p2_218, __p3_218) __extension__ ({ \
+ int32x4_t __s0_218 = __p0_218; \
+ int32x4_t __s1_218 = __p1_218; \
+ int32x2_t __s2_218 = __p2_218; \
+ int32x4_t __rev0_218; __rev0_218 = __builtin_shufflevector(__s0_218, __s0_218, 3, 2, 1, 0); \
+ int32x4_t __rev1_218; __rev1_218 = __builtin_shufflevector(__s1_218, __s1_218, 3, 2, 1, 0); \
+ int32x2_t __rev2_218; __rev2_218 = __builtin_shufflevector(__s2_218, __s2_218, 1, 0); \
+ int32x4_t __ret_218; \
+ __ret_218 = __noswap_vqaddq_s32(__rev0_218, __noswap_vqrdmulhq_s32(__rev1_218, __noswap_splatq_lane_s32(__rev2_218, __p3_218))); \
+ __ret_218 = __builtin_shufflevector(__ret_218, __ret_218, 3, 2, 1, 0); \
+ __ret_218; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlahq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x8_t __ret; \
- __ret = vqaddq_s16(__s0, vqrdmulhq_s16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3))); \
- __ret; \
+#define vqrdmlahq_lane_s16(__p0_219, __p1_219, __p2_219, __p3_219) __extension__ ({ \
+ int16x8_t __s0_219 = __p0_219; \
+ int16x8_t __s1_219 = __p1_219; \
+ int16x4_t __s2_219 = __p2_219; \
+ int16x8_t __ret_219; \
+ __ret_219 = vqaddq_s16(__s0_219, vqrdmulhq_s16(__s1_219, splatq_lane_s16(__s2_219, __p3_219))); \
+ __ret_219; \
})
#else
-#define vqrdmlahq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int16x8_t __ret; \
- __ret = __noswap_vqaddq_s16(__rev0, __noswap_vqrdmulhq_s16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vqrdmlahq_lane_s16(__p0_220, __p1_220, __p2_220, __p3_220) __extension__ ({ \
+ int16x8_t __s0_220 = __p0_220; \
+ int16x8_t __s1_220 = __p1_220; \
+ int16x4_t __s2_220 = __p2_220; \
+ int16x8_t __rev0_220; __rev0_220 = __builtin_shufflevector(__s0_220, __s0_220, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_220; __rev1_220 = __builtin_shufflevector(__s1_220, __s1_220, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev2_220; __rev2_220 = __builtin_shufflevector(__s2_220, __s2_220, 3, 2, 1, 0); \
+ int16x8_t __ret_220; \
+ __ret_220 = __noswap_vqaddq_s16(__rev0_220, __noswap_vqrdmulhq_s16(__rev1_220, __noswap_splatq_lane_s16(__rev2_220, __p3_220))); \
+ __ret_220 = __builtin_shufflevector(__ret_220, __ret_220, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_220; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlah_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x2_t __ret; \
- __ret = vqadd_s32(__s0, vqrdmulh_s32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3))); \
- __ret; \
+#define vqrdmlah_lane_s32(__p0_221, __p1_221, __p2_221, __p3_221) __extension__ ({ \
+ int32x2_t __s0_221 = __p0_221; \
+ int32x2_t __s1_221 = __p1_221; \
+ int32x2_t __s2_221 = __p2_221; \
+ int32x2_t __ret_221; \
+ __ret_221 = vqadd_s32(__s0_221, vqrdmulh_s32(__s1_221, splat_lane_s32(__s2_221, __p3_221))); \
+ __ret_221; \
})
#else
-#define vqrdmlah_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int32x2_t __ret; \
- __ret = __noswap_vqadd_s32(__rev0, __noswap_vqrdmulh_s32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vqrdmlah_lane_s32(__p0_222, __p1_222, __p2_222, __p3_222) __extension__ ({ \
+ int32x2_t __s0_222 = __p0_222; \
+ int32x2_t __s1_222 = __p1_222; \
+ int32x2_t __s2_222 = __p2_222; \
+ int32x2_t __rev0_222; __rev0_222 = __builtin_shufflevector(__s0_222, __s0_222, 1, 0); \
+ int32x2_t __rev1_222; __rev1_222 = __builtin_shufflevector(__s1_222, __s1_222, 1, 0); \
+ int32x2_t __rev2_222; __rev2_222 = __builtin_shufflevector(__s2_222, __s2_222, 1, 0); \
+ int32x2_t __ret_222; \
+ __ret_222 = __noswap_vqadd_s32(__rev0_222, __noswap_vqrdmulh_s32(__rev1_222, __noswap_splat_lane_s32(__rev2_222, __p3_222))); \
+ __ret_222 = __builtin_shufflevector(__ret_222, __ret_222, 1, 0); \
+ __ret_222; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlah_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x4_t __ret; \
- __ret = vqadd_s16(__s0, vqrdmulh_s16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3))); \
- __ret; \
+#define vqrdmlah_lane_s16(__p0_223, __p1_223, __p2_223, __p3_223) __extension__ ({ \
+ int16x4_t __s0_223 = __p0_223; \
+ int16x4_t __s1_223 = __p1_223; \
+ int16x4_t __s2_223 = __p2_223; \
+ int16x4_t __ret_223; \
+ __ret_223 = vqadd_s16(__s0_223, vqrdmulh_s16(__s1_223, splat_lane_s16(__s2_223, __p3_223))); \
+ __ret_223; \
})
#else
-#define vqrdmlah_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int16x4_t __ret; \
- __ret = __noswap_vqadd_s16(__rev0, __noswap_vqrdmulh_s16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqrdmlah_lane_s16(__p0_224, __p1_224, __p2_224, __p3_224) __extension__ ({ \
+ int16x4_t __s0_224 = __p0_224; \
+ int16x4_t __s1_224 = __p1_224; \
+ int16x4_t __s2_224 = __p2_224; \
+ int16x4_t __rev0_224; __rev0_224 = __builtin_shufflevector(__s0_224, __s0_224, 3, 2, 1, 0); \
+ int16x4_t __rev1_224; __rev1_224 = __builtin_shufflevector(__s1_224, __s1_224, 3, 2, 1, 0); \
+ int16x4_t __rev2_224; __rev2_224 = __builtin_shufflevector(__s2_224, __s2_224, 3, 2, 1, 0); \
+ int16x4_t __ret_224; \
+ __ret_224 = __noswap_vqadd_s16(__rev0_224, __noswap_vqrdmulh_s16(__rev1_224, __noswap_splat_lane_s16(__rev2_224, __p3_224))); \
+ __ret_224 = __builtin_shufflevector(__ret_224, __ret_224, 3, 2, 1, 0); \
+ __ret_224; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlshq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = vqsubq_s32(__s0, vqrdmulhq_s32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3))); \
- __ret; \
+#define vqrdmlshq_lane_s32(__p0_225, __p1_225, __p2_225, __p3_225) __extension__ ({ \
+ int32x4_t __s0_225 = __p0_225; \
+ int32x4_t __s1_225 = __p1_225; \
+ int32x2_t __s2_225 = __p2_225; \
+ int32x4_t __ret_225; \
+ __ret_225 = vqsubq_s32(__s0_225, vqrdmulhq_s32(__s1_225, splatq_lane_s32(__s2_225, __p3_225))); \
+ __ret_225; \
})
#else
-#define vqrdmlshq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqsubq_s32(__rev0, __noswap_vqrdmulhq_s32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqrdmlshq_lane_s32(__p0_226, __p1_226, __p2_226, __p3_226) __extension__ ({ \
+ int32x4_t __s0_226 = __p0_226; \
+ int32x4_t __s1_226 = __p1_226; \
+ int32x2_t __s2_226 = __p2_226; \
+ int32x4_t __rev0_226; __rev0_226 = __builtin_shufflevector(__s0_226, __s0_226, 3, 2, 1, 0); \
+ int32x4_t __rev1_226; __rev1_226 = __builtin_shufflevector(__s1_226, __s1_226, 3, 2, 1, 0); \
+ int32x2_t __rev2_226; __rev2_226 = __builtin_shufflevector(__s2_226, __s2_226, 1, 0); \
+ int32x4_t __ret_226; \
+ __ret_226 = __noswap_vqsubq_s32(__rev0_226, __noswap_vqrdmulhq_s32(__rev1_226, __noswap_splatq_lane_s32(__rev2_226, __p3_226))); \
+ __ret_226 = __builtin_shufflevector(__ret_226, __ret_226, 3, 2, 1, 0); \
+ __ret_226; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlshq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x8_t __ret; \
- __ret = vqsubq_s16(__s0, vqrdmulhq_s16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3))); \
- __ret; \
+#define vqrdmlshq_lane_s16(__p0_227, __p1_227, __p2_227, __p3_227) __extension__ ({ \
+ int16x8_t __s0_227 = __p0_227; \
+ int16x8_t __s1_227 = __p1_227; \
+ int16x4_t __s2_227 = __p2_227; \
+ int16x8_t __ret_227; \
+ __ret_227 = vqsubq_s16(__s0_227, vqrdmulhq_s16(__s1_227, splatq_lane_s16(__s2_227, __p3_227))); \
+ __ret_227; \
})
#else
-#define vqrdmlshq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int16x8_t __ret; \
- __ret = __noswap_vqsubq_s16(__rev0, __noswap_vqrdmulhq_s16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vqrdmlshq_lane_s16(__p0_228, __p1_228, __p2_228, __p3_228) __extension__ ({ \
+ int16x8_t __s0_228 = __p0_228; \
+ int16x8_t __s1_228 = __p1_228; \
+ int16x4_t __s2_228 = __p2_228; \
+ int16x8_t __rev0_228; __rev0_228 = __builtin_shufflevector(__s0_228, __s0_228, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_228; __rev1_228 = __builtin_shufflevector(__s1_228, __s1_228, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev2_228; __rev2_228 = __builtin_shufflevector(__s2_228, __s2_228, 3, 2, 1, 0); \
+ int16x8_t __ret_228; \
+ __ret_228 = __noswap_vqsubq_s16(__rev0_228, __noswap_vqrdmulhq_s16(__rev1_228, __noswap_splatq_lane_s16(__rev2_228, __p3_228))); \
+ __ret_228 = __builtin_shufflevector(__ret_228, __ret_228, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_228; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlsh_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x2_t __ret; \
- __ret = vqsub_s32(__s0, vqrdmulh_s32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3))); \
- __ret; \
+#define vqrdmlsh_lane_s32(__p0_229, __p1_229, __p2_229, __p3_229) __extension__ ({ \
+ int32x2_t __s0_229 = __p0_229; \
+ int32x2_t __s1_229 = __p1_229; \
+ int32x2_t __s2_229 = __p2_229; \
+ int32x2_t __ret_229; \
+ __ret_229 = vqsub_s32(__s0_229, vqrdmulh_s32(__s1_229, splat_lane_s32(__s2_229, __p3_229))); \
+ __ret_229; \
})
#else
-#define vqrdmlsh_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int32x2_t __ret; \
- __ret = __noswap_vqsub_s32(__rev0, __noswap_vqrdmulh_s32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vqrdmlsh_lane_s32(__p0_230, __p1_230, __p2_230, __p3_230) __extension__ ({ \
+ int32x2_t __s0_230 = __p0_230; \
+ int32x2_t __s1_230 = __p1_230; \
+ int32x2_t __s2_230 = __p2_230; \
+ int32x2_t __rev0_230; __rev0_230 = __builtin_shufflevector(__s0_230, __s0_230, 1, 0); \
+ int32x2_t __rev1_230; __rev1_230 = __builtin_shufflevector(__s1_230, __s1_230, 1, 0); \
+ int32x2_t __rev2_230; __rev2_230 = __builtin_shufflevector(__s2_230, __s2_230, 1, 0); \
+ int32x2_t __ret_230; \
+ __ret_230 = __noswap_vqsub_s32(__rev0_230, __noswap_vqrdmulh_s32(__rev1_230, __noswap_splat_lane_s32(__rev2_230, __p3_230))); \
+ __ret_230 = __builtin_shufflevector(__ret_230, __ret_230, 1, 0); \
+ __ret_230; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlsh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x4_t __ret; \
- __ret = vqsub_s16(__s0, vqrdmulh_s16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3))); \
- __ret; \
+#define vqrdmlsh_lane_s16(__p0_231, __p1_231, __p2_231, __p3_231) __extension__ ({ \
+ int16x4_t __s0_231 = __p0_231; \
+ int16x4_t __s1_231 = __p1_231; \
+ int16x4_t __s2_231 = __p2_231; \
+ int16x4_t __ret_231; \
+ __ret_231 = vqsub_s16(__s0_231, vqrdmulh_s16(__s1_231, splat_lane_s16(__s2_231, __p3_231))); \
+ __ret_231; \
})
#else
-#define vqrdmlsh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int16x4_t __ret; \
- __ret = __noswap_vqsub_s16(__rev0, __noswap_vqrdmulh_s16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqrdmlsh_lane_s16(__p0_232, __p1_232, __p2_232, __p3_232) __extension__ ({ \
+ int16x4_t __s0_232 = __p0_232; \
+ int16x4_t __s1_232 = __p1_232; \
+ int16x4_t __s2_232 = __p2_232; \
+ int16x4_t __rev0_232; __rev0_232 = __builtin_shufflevector(__s0_232, __s0_232, 3, 2, 1, 0); \
+ int16x4_t __rev1_232; __rev1_232 = __builtin_shufflevector(__s1_232, __s1_232, 3, 2, 1, 0); \
+ int16x4_t __rev2_232; __rev2_232 = __builtin_shufflevector(__s2_232, __s2_232, 3, 2, 1, 0); \
+ int16x4_t __ret_232; \
+ __ret_232 = __noswap_vqsub_s16(__rev0_232, __noswap_vqrdmulh_s16(__rev1_232, __noswap_splat_lane_s16(__rev2_232, __p3_232))); \
+ __ret_232 = __builtin_shufflevector(__ret_232, __ret_232, 3, 2, 1, 0); \
+ __ret_232; \
})
#endif
#endif
#if defined(__ARM_FEATURE_QRDMX) && defined(__aarch64__)
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlahq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = vqaddq_s32(__s0, vqrdmulhq_s32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3))); \
- __ret; \
+#define vqrdmlahq_laneq_s32(__p0_233, __p1_233, __p2_233, __p3_233) __extension__ ({ \
+ int32x4_t __s0_233 = __p0_233; \
+ int32x4_t __s1_233 = __p1_233; \
+ int32x4_t __s2_233 = __p2_233; \
+ int32x4_t __ret_233; \
+ __ret_233 = vqaddq_s32(__s0_233, vqrdmulhq_s32(__s1_233, splatq_laneq_s32(__s2_233, __p3_233))); \
+ __ret_233; \
})
#else
-#define vqrdmlahq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqaddq_s32(__rev0, __noswap_vqrdmulhq_s32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqrdmlahq_laneq_s32(__p0_234, __p1_234, __p2_234, __p3_234) __extension__ ({ \
+ int32x4_t __s0_234 = __p0_234; \
+ int32x4_t __s1_234 = __p1_234; \
+ int32x4_t __s2_234 = __p2_234; \
+ int32x4_t __rev0_234; __rev0_234 = __builtin_shufflevector(__s0_234, __s0_234, 3, 2, 1, 0); \
+ int32x4_t __rev1_234; __rev1_234 = __builtin_shufflevector(__s1_234, __s1_234, 3, 2, 1, 0); \
+ int32x4_t __rev2_234; __rev2_234 = __builtin_shufflevector(__s2_234, __s2_234, 3, 2, 1, 0); \
+ int32x4_t __ret_234; \
+ __ret_234 = __noswap_vqaddq_s32(__rev0_234, __noswap_vqrdmulhq_s32(__rev1_234, __noswap_splatq_laneq_s32(__rev2_234, __p3_234))); \
+ __ret_234 = __builtin_shufflevector(__ret_234, __ret_234, 3, 2, 1, 0); \
+ __ret_234; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlahq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x8_t __ret; \
- __ret = vqaddq_s16(__s0, vqrdmulhq_s16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3))); \
- __ret; \
+#define vqrdmlahq_laneq_s16(__p0_235, __p1_235, __p2_235, __p3_235) __extension__ ({ \
+ int16x8_t __s0_235 = __p0_235; \
+ int16x8_t __s1_235 = __p1_235; \
+ int16x8_t __s2_235 = __p2_235; \
+ int16x8_t __ret_235; \
+ __ret_235 = vqaddq_s16(__s0_235, vqrdmulhq_s16(__s1_235, splatq_laneq_s16(__s2_235, __p3_235))); \
+ __ret_235; \
})
#else
-#define vqrdmlahq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __ret; \
- __ret = __noswap_vqaddq_s16(__rev0, __noswap_vqrdmulhq_s16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vqrdmlahq_laneq_s16(__p0_236, __p1_236, __p2_236, __p3_236) __extension__ ({ \
+ int16x8_t __s0_236 = __p0_236; \
+ int16x8_t __s1_236 = __p1_236; \
+ int16x8_t __s2_236 = __p2_236; \
+ int16x8_t __rev0_236; __rev0_236 = __builtin_shufflevector(__s0_236, __s0_236, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_236; __rev1_236 = __builtin_shufflevector(__s1_236, __s1_236, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev2_236; __rev2_236 = __builtin_shufflevector(__s2_236, __s2_236, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __ret_236; \
+ __ret_236 = __noswap_vqaddq_s16(__rev0_236, __noswap_vqrdmulhq_s16(__rev1_236, __noswap_splatq_laneq_s16(__rev2_236, __p3_236))); \
+ __ret_236 = __builtin_shufflevector(__ret_236, __ret_236, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_236; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlah_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x2_t __ret; \
- __ret = vqadd_s32(__s0, vqrdmulh_s32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3))); \
- __ret; \
+#define vqrdmlah_laneq_s32(__p0_237, __p1_237, __p2_237, __p3_237) __extension__ ({ \
+ int32x2_t __s0_237 = __p0_237; \
+ int32x2_t __s1_237 = __p1_237; \
+ int32x4_t __s2_237 = __p2_237; \
+ int32x2_t __ret_237; \
+ __ret_237 = vqadd_s32(__s0_237, vqrdmulh_s32(__s1_237, splat_laneq_s32(__s2_237, __p3_237))); \
+ __ret_237; \
})
#else
-#define vqrdmlah_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x2_t __ret; \
- __ret = __noswap_vqadd_s32(__rev0, __noswap_vqrdmulh_s32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vqrdmlah_laneq_s32(__p0_238, __p1_238, __p2_238, __p3_238) __extension__ ({ \
+ int32x2_t __s0_238 = __p0_238; \
+ int32x2_t __s1_238 = __p1_238; \
+ int32x4_t __s2_238 = __p2_238; \
+ int32x2_t __rev0_238; __rev0_238 = __builtin_shufflevector(__s0_238, __s0_238, 1, 0); \
+ int32x2_t __rev1_238; __rev1_238 = __builtin_shufflevector(__s1_238, __s1_238, 1, 0); \
+ int32x4_t __rev2_238; __rev2_238 = __builtin_shufflevector(__s2_238, __s2_238, 3, 2, 1, 0); \
+ int32x2_t __ret_238; \
+ __ret_238 = __noswap_vqadd_s32(__rev0_238, __noswap_vqrdmulh_s32(__rev1_238, __noswap_splat_laneq_s32(__rev2_238, __p3_238))); \
+ __ret_238 = __builtin_shufflevector(__ret_238, __ret_238, 1, 0); \
+ __ret_238; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlah_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x4_t __ret; \
- __ret = vqadd_s16(__s0, vqrdmulh_s16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3))); \
- __ret; \
+#define vqrdmlah_laneq_s16(__p0_239, __p1_239, __p2_239, __p3_239) __extension__ ({ \
+ int16x4_t __s0_239 = __p0_239; \
+ int16x4_t __s1_239 = __p1_239; \
+ int16x8_t __s2_239 = __p2_239; \
+ int16x4_t __ret_239; \
+ __ret_239 = vqadd_s16(__s0_239, vqrdmulh_s16(__s1_239, splat_laneq_s16(__s2_239, __p3_239))); \
+ __ret_239; \
})
#else
-#define vqrdmlah_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __ret; \
- __ret = __noswap_vqadd_s16(__rev0, __noswap_vqrdmulh_s16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqrdmlah_laneq_s16(__p0_240, __p1_240, __p2_240, __p3_240) __extension__ ({ \
+ int16x4_t __s0_240 = __p0_240; \
+ int16x4_t __s1_240 = __p1_240; \
+ int16x8_t __s2_240 = __p2_240; \
+ int16x4_t __rev0_240; __rev0_240 = __builtin_shufflevector(__s0_240, __s0_240, 3, 2, 1, 0); \
+ int16x4_t __rev1_240; __rev1_240 = __builtin_shufflevector(__s1_240, __s1_240, 3, 2, 1, 0); \
+ int16x8_t __rev2_240; __rev2_240 = __builtin_shufflevector(__s2_240, __s2_240, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __ret_240; \
+ __ret_240 = __noswap_vqadd_s16(__rev0_240, __noswap_vqrdmulh_s16(__rev1_240, __noswap_splat_laneq_s16(__rev2_240, __p3_240))); \
+ __ret_240 = __builtin_shufflevector(__ret_240, __ret_240, 3, 2, 1, 0); \
+ __ret_240; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlshq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = vqsubq_s32(__s0, vqrdmulhq_s32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3))); \
- __ret; \
+#define vqrdmlshq_laneq_s32(__p0_241, __p1_241, __p2_241, __p3_241) __extension__ ({ \
+ int32x4_t __s0_241 = __p0_241; \
+ int32x4_t __s1_241 = __p1_241; \
+ int32x4_t __s2_241 = __p2_241; \
+ int32x4_t __ret_241; \
+ __ret_241 = vqsubq_s32(__s0_241, vqrdmulhq_s32(__s1_241, splatq_laneq_s32(__s2_241, __p3_241))); \
+ __ret_241; \
})
#else
-#define vqrdmlshq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqsubq_s32(__rev0, __noswap_vqrdmulhq_s32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqrdmlshq_laneq_s32(__p0_242, __p1_242, __p2_242, __p3_242) __extension__ ({ \
+ int32x4_t __s0_242 = __p0_242; \
+ int32x4_t __s1_242 = __p1_242; \
+ int32x4_t __s2_242 = __p2_242; \
+ int32x4_t __rev0_242; __rev0_242 = __builtin_shufflevector(__s0_242, __s0_242, 3, 2, 1, 0); \
+ int32x4_t __rev1_242; __rev1_242 = __builtin_shufflevector(__s1_242, __s1_242, 3, 2, 1, 0); \
+ int32x4_t __rev2_242; __rev2_242 = __builtin_shufflevector(__s2_242, __s2_242, 3, 2, 1, 0); \
+ int32x4_t __ret_242; \
+ __ret_242 = __noswap_vqsubq_s32(__rev0_242, __noswap_vqrdmulhq_s32(__rev1_242, __noswap_splatq_laneq_s32(__rev2_242, __p3_242))); \
+ __ret_242 = __builtin_shufflevector(__ret_242, __ret_242, 3, 2, 1, 0); \
+ __ret_242; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlshq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x8_t __ret; \
- __ret = vqsubq_s16(__s0, vqrdmulhq_s16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3))); \
- __ret; \
+#define vqrdmlshq_laneq_s16(__p0_243, __p1_243, __p2_243, __p3_243) __extension__ ({ \
+ int16x8_t __s0_243 = __p0_243; \
+ int16x8_t __s1_243 = __p1_243; \
+ int16x8_t __s2_243 = __p2_243; \
+ int16x8_t __ret_243; \
+ __ret_243 = vqsubq_s16(__s0_243, vqrdmulhq_s16(__s1_243, splatq_laneq_s16(__s2_243, __p3_243))); \
+ __ret_243; \
})
#else
-#define vqrdmlshq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __ret; \
- __ret = __noswap_vqsubq_s16(__rev0, __noswap_vqrdmulhq_s16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vqrdmlshq_laneq_s16(__p0_244, __p1_244, __p2_244, __p3_244) __extension__ ({ \
+ int16x8_t __s0_244 = __p0_244; \
+ int16x8_t __s1_244 = __p1_244; \
+ int16x8_t __s2_244 = __p2_244; \
+ int16x8_t __rev0_244; __rev0_244 = __builtin_shufflevector(__s0_244, __s0_244, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_244; __rev1_244 = __builtin_shufflevector(__s1_244, __s1_244, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev2_244; __rev2_244 = __builtin_shufflevector(__s2_244, __s2_244, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __ret_244; \
+ __ret_244 = __noswap_vqsubq_s16(__rev0_244, __noswap_vqrdmulhq_s16(__rev1_244, __noswap_splatq_laneq_s16(__rev2_244, __p3_244))); \
+ __ret_244 = __builtin_shufflevector(__ret_244, __ret_244, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_244; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlsh_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x2_t __ret; \
- __ret = vqsub_s32(__s0, vqrdmulh_s32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3))); \
- __ret; \
+#define vqrdmlsh_laneq_s32(__p0_245, __p1_245, __p2_245, __p3_245) __extension__ ({ \
+ int32x2_t __s0_245 = __p0_245; \
+ int32x2_t __s1_245 = __p1_245; \
+ int32x4_t __s2_245 = __p2_245; \
+ int32x2_t __ret_245; \
+ __ret_245 = vqsub_s32(__s0_245, vqrdmulh_s32(__s1_245, splat_laneq_s32(__s2_245, __p3_245))); \
+ __ret_245; \
})
#else
-#define vqrdmlsh_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x2_t __ret; \
- __ret = __noswap_vqsub_s32(__rev0, __noswap_vqrdmulh_s32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vqrdmlsh_laneq_s32(__p0_246, __p1_246, __p2_246, __p3_246) __extension__ ({ \
+ int32x2_t __s0_246 = __p0_246; \
+ int32x2_t __s1_246 = __p1_246; \
+ int32x4_t __s2_246 = __p2_246; \
+ int32x2_t __rev0_246; __rev0_246 = __builtin_shufflevector(__s0_246, __s0_246, 1, 0); \
+ int32x2_t __rev1_246; __rev1_246 = __builtin_shufflevector(__s1_246, __s1_246, 1, 0); \
+ int32x4_t __rev2_246; __rev2_246 = __builtin_shufflevector(__s2_246, __s2_246, 3, 2, 1, 0); \
+ int32x2_t __ret_246; \
+ __ret_246 = __noswap_vqsub_s32(__rev0_246, __noswap_vqrdmulh_s32(__rev1_246, __noswap_splat_laneq_s32(__rev2_246, __p3_246))); \
+ __ret_246 = __builtin_shufflevector(__ret_246, __ret_246, 1, 0); \
+ __ret_246; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlsh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x4_t __ret; \
- __ret = vqsub_s16(__s0, vqrdmulh_s16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3))); \
- __ret; \
+#define vqrdmlsh_laneq_s16(__p0_247, __p1_247, __p2_247, __p3_247) __extension__ ({ \
+ int16x4_t __s0_247 = __p0_247; \
+ int16x4_t __s1_247 = __p1_247; \
+ int16x8_t __s2_247 = __p2_247; \
+ int16x4_t __ret_247; \
+ __ret_247 = vqsub_s16(__s0_247, vqrdmulh_s16(__s1_247, splat_laneq_s16(__s2_247, __p3_247))); \
+ __ret_247; \
})
#else
-#define vqrdmlsh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __ret; \
- __ret = __noswap_vqsub_s16(__rev0, __noswap_vqrdmulh_s16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3))); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqrdmlsh_laneq_s16(__p0_248, __p1_248, __p2_248, __p3_248) __extension__ ({ \
+ int16x4_t __s0_248 = __p0_248; \
+ int16x4_t __s1_248 = __p1_248; \
+ int16x8_t __s2_248 = __p2_248; \
+ int16x4_t __rev0_248; __rev0_248 = __builtin_shufflevector(__s0_248, __s0_248, 3, 2, 1, 0); \
+ int16x4_t __rev1_248; __rev1_248 = __builtin_shufflevector(__s1_248, __s1_248, 3, 2, 1, 0); \
+ int16x8_t __rev2_248; __rev2_248 = __builtin_shufflevector(__s2_248, __s2_248, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __ret_248; \
+ __ret_248 = __noswap_vqsub_s16(__rev0_248, __noswap_vqrdmulh_s16(__rev1_248, __noswap_splat_laneq_s16(__rev2_248, __p3_248))); \
+ __ret_248 = __builtin_shufflevector(__ret_248, __ret_248, 3, 2, 1, 0); \
+ __ret_248; \
})
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai float64x1_t vabd_f64(float64x1_t __p0, float64x1_t __p1) {
float64x1_t __ret;
__ret = (float64x1_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
return __ret;
}
-#else
-__ai float64x1_t vabd_f64(float64x1_t __p0, float64x1_t __p1) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64_t vabdd_f64(float64_t __p0, float64_t __p1) {
- float64_t __ret;
- __ret = (float64_t) __builtin_neon_vabdd_f64(__p0, __p1);
- return __ret;
-}
-#else
__ai float64_t vabdd_f64(float64_t __p0, float64_t __p1) {
float64_t __ret;
__ret = (float64_t) __builtin_neon_vabdd_f64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32_t vabds_f32(float32_t __p0, float32_t __p1) {
- float32_t __ret;
- __ret = (float32_t) __builtin_neon_vabds_f32(__p0, __p1);
- return __ret;
-}
-#else
__ai float32_t vabds_f32(float32_t __p0, float32_t __p1) {
float32_t __ret;
__ret = (float32_t) __builtin_neon_vabds_f32(__p0, __p1);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float64x2_t vabsq_f64(float64x2_t __p0) {
float64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai float64x1_t vabs_f64(float64x1_t __p0) {
float64x1_t __ret;
__ret = (float64x1_t) __builtin_neon_vabs_v((int8x8_t)__p0, 10);
return __ret;
}
-#else
-__ai float64x1_t vabs_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vabs_v((int8x8_t)__p0, 10);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vabs_s64(int64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vabs_v((int8x8_t)__p0, 3);
- return __ret;
-}
-#else
__ai int64x1_t vabs_s64(int64x1_t __p0) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vabs_v((int8x8_t)__p0, 3);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vabsd_s64(int64_t __p0) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vabsd_s64(__p0);
- return __ret;
-}
-#else
__ai int64_t vabsd_s64(int64_t __p0) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vabsd_s64(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float64x2_t vaddq_f64(float64x2_t __p0, float64x2_t __p1) {
float64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai float64x1_t vadd_f64(float64x1_t __p0, float64x1_t __p1) {
float64x1_t __ret;
__ret = __p0 + __p1;
return __ret;
}
-#else
-__ai float64x1_t vadd_f64(float64x1_t __p0, float64x1_t __p1) {
- float64x1_t __ret;
- __ret = __p0 + __p1;
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vaddd_u64(uint64_t __p0, uint64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vaddd_u64(__p0, __p1);
- return __ret;
-}
-#else
__ai uint64_t vaddd_u64(uint64_t __p0, uint64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vaddd_u64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vaddd_s64(int64_t __p0, int64_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vaddd_s64(__p0, __p1);
- return __ret;
-}
-#else
__ai int64_t vaddd_s64(int64_t __p0, int64_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vaddd_s64(__p0, __p1);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x8_t vaddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
uint16x8_t __ret;
#ifdef __LITTLE_ENDIAN__
__ai uint16_t vaddlvq_u8(uint8x16_t __p0) {
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vaddlvq_u8((int8x16_t)__p0);
+ __ret = (uint16_t) __builtin_neon_vaddlvq_u8(__p0);
return __ret;
}
#else
__ai uint16_t vaddlvq_u8(uint8x16_t __p0) {
uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vaddlvq_u8((int8x16_t)__rev0);
+ __ret = (uint16_t) __builtin_neon_vaddlvq_u8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint64_t vaddlvq_u32(uint32x4_t __p0) {
uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vaddlvq_u32((int8x16_t)__p0);
+ __ret = (uint64_t) __builtin_neon_vaddlvq_u32(__p0);
return __ret;
}
#else
__ai uint64_t vaddlvq_u32(uint32x4_t __p0) {
uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vaddlvq_u32((int8x16_t)__rev0);
+ __ret = (uint64_t) __builtin_neon_vaddlvq_u32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint32_t vaddlvq_u16(uint16x8_t __p0) {
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vaddlvq_u16((int8x16_t)__p0);
+ __ret = (uint32_t) __builtin_neon_vaddlvq_u16(__p0);
return __ret;
}
#else
__ai uint32_t vaddlvq_u16(uint16x8_t __p0) {
uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vaddlvq_u16((int8x16_t)__rev0);
+ __ret = (uint32_t) __builtin_neon_vaddlvq_u16(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int16_t vaddlvq_s8(int8x16_t __p0) {
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vaddlvq_s8((int8x16_t)__p0);
+ __ret = (int16_t) __builtin_neon_vaddlvq_s8(__p0);
return __ret;
}
#else
__ai int16_t vaddlvq_s8(int8x16_t __p0) {
int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vaddlvq_s8((int8x16_t)__rev0);
+ __ret = (int16_t) __builtin_neon_vaddlvq_s8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int64_t vaddlvq_s32(int32x4_t __p0) {
int64_t __ret;
- __ret = (int64_t) __builtin_neon_vaddlvq_s32((int8x16_t)__p0);
+ __ret = (int64_t) __builtin_neon_vaddlvq_s32(__p0);
return __ret;
}
#else
__ai int64_t vaddlvq_s32(int32x4_t __p0) {
int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int64_t __ret;
- __ret = (int64_t) __builtin_neon_vaddlvq_s32((int8x16_t)__rev0);
+ __ret = (int64_t) __builtin_neon_vaddlvq_s32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int32_t vaddlvq_s16(int16x8_t __p0) {
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vaddlvq_s16((int8x16_t)__p0);
+ __ret = (int32_t) __builtin_neon_vaddlvq_s16(__p0);
return __ret;
}
#else
__ai int32_t vaddlvq_s16(int16x8_t __p0) {
int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vaddlvq_s16((int8x16_t)__rev0);
+ __ret = (int32_t) __builtin_neon_vaddlvq_s16(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint16_t vaddlv_u8(uint8x8_t __p0) {
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vaddlv_u8((int8x8_t)__p0);
+ __ret = (uint16_t) __builtin_neon_vaddlv_u8(__p0);
return __ret;
}
#else
__ai uint16_t vaddlv_u8(uint8x8_t __p0) {
uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vaddlv_u8((int8x8_t)__rev0);
+ __ret = (uint16_t) __builtin_neon_vaddlv_u8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint64_t vaddlv_u32(uint32x2_t __p0) {
uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vaddlv_u32((int8x8_t)__p0);
+ __ret = (uint64_t) __builtin_neon_vaddlv_u32(__p0);
return __ret;
}
#else
__ai uint64_t vaddlv_u32(uint32x2_t __p0) {
uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vaddlv_u32((int8x8_t)__rev0);
+ __ret = (uint64_t) __builtin_neon_vaddlv_u32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint32_t vaddlv_u16(uint16x4_t __p0) {
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vaddlv_u16((int8x8_t)__p0);
+ __ret = (uint32_t) __builtin_neon_vaddlv_u16(__p0);
return __ret;
}
#else
__ai uint32_t vaddlv_u16(uint16x4_t __p0) {
uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vaddlv_u16((int8x8_t)__rev0);
+ __ret = (uint32_t) __builtin_neon_vaddlv_u16(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int16_t vaddlv_s8(int8x8_t __p0) {
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vaddlv_s8((int8x8_t)__p0);
+ __ret = (int16_t) __builtin_neon_vaddlv_s8(__p0);
return __ret;
}
#else
__ai int16_t vaddlv_s8(int8x8_t __p0) {
int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vaddlv_s8((int8x8_t)__rev0);
+ __ret = (int16_t) __builtin_neon_vaddlv_s8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int64_t vaddlv_s32(int32x2_t __p0) {
int64_t __ret;
- __ret = (int64_t) __builtin_neon_vaddlv_s32((int8x8_t)__p0);
+ __ret = (int64_t) __builtin_neon_vaddlv_s32(__p0);
return __ret;
}
#else
__ai int64_t vaddlv_s32(int32x2_t __p0) {
int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
int64_t __ret;
- __ret = (int64_t) __builtin_neon_vaddlv_s32((int8x8_t)__rev0);
+ __ret = (int64_t) __builtin_neon_vaddlv_s32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int32_t vaddlv_s16(int16x4_t __p0) {
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vaddlv_s16((int8x8_t)__p0);
+ __ret = (int32_t) __builtin_neon_vaddlv_s16(__p0);
return __ret;
}
#else
__ai int32_t vaddlv_s16(int16x4_t __p0) {
int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vaddlv_s16((int8x8_t)__rev0);
+ __ret = (int32_t) __builtin_neon_vaddlv_s16(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint8_t vaddvq_u8(uint8x16_t __p0) {
uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vaddvq_u8((int8x16_t)__p0);
+ __ret = (uint8_t) __builtin_neon_vaddvq_u8(__p0);
return __ret;
}
#else
__ai uint8_t vaddvq_u8(uint8x16_t __p0) {
uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vaddvq_u8((int8x16_t)__rev0);
+ __ret = (uint8_t) __builtin_neon_vaddvq_u8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint32_t vaddvq_u32(uint32x4_t __p0) {
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vaddvq_u32((int8x16_t)__p0);
+ __ret = (uint32_t) __builtin_neon_vaddvq_u32(__p0);
return __ret;
}
#else
__ai uint32_t vaddvq_u32(uint32x4_t __p0) {
uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vaddvq_u32((int8x16_t)__rev0);
+ __ret = (uint32_t) __builtin_neon_vaddvq_u32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint64_t vaddvq_u64(uint64x2_t __p0) {
uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vaddvq_u64((int8x16_t)__p0);
+ __ret = (uint64_t) __builtin_neon_vaddvq_u64(__p0);
return __ret;
}
#else
__ai uint64_t vaddvq_u64(uint64x2_t __p0) {
uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vaddvq_u64((int8x16_t)__rev0);
+ __ret = (uint64_t) __builtin_neon_vaddvq_u64(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint16_t vaddvq_u16(uint16x8_t __p0) {
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vaddvq_u16((int8x16_t)__p0);
+ __ret = (uint16_t) __builtin_neon_vaddvq_u16(__p0);
return __ret;
}
#else
__ai uint16_t vaddvq_u16(uint16x8_t __p0) {
uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vaddvq_u16((int8x16_t)__rev0);
+ __ret = (uint16_t) __builtin_neon_vaddvq_u16(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int8_t vaddvq_s8(int8x16_t __p0) {
int8_t __ret;
- __ret = (int8_t) __builtin_neon_vaddvq_s8((int8x16_t)__p0);
+ __ret = (int8_t) __builtin_neon_vaddvq_s8(__p0);
return __ret;
}
#else
__ai int8_t vaddvq_s8(int8x16_t __p0) {
int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8_t __ret;
- __ret = (int8_t) __builtin_neon_vaddvq_s8((int8x16_t)__rev0);
+ __ret = (int8_t) __builtin_neon_vaddvq_s8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float64_t vaddvq_f64(float64x2_t __p0) {
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vaddvq_f64((int8x16_t)__p0);
+ __ret = (float64_t) __builtin_neon_vaddvq_f64(__p0);
return __ret;
}
#else
__ai float64_t vaddvq_f64(float64x2_t __p0) {
float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vaddvq_f64((int8x16_t)__rev0);
+ __ret = (float64_t) __builtin_neon_vaddvq_f64(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float32_t vaddvq_f32(float32x4_t __p0) {
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vaddvq_f32((int8x16_t)__p0);
+ __ret = (float32_t) __builtin_neon_vaddvq_f32(__p0);
return __ret;
}
#else
__ai float32_t vaddvq_f32(float32x4_t __p0) {
float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vaddvq_f32((int8x16_t)__rev0);
+ __ret = (float32_t) __builtin_neon_vaddvq_f32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int32_t vaddvq_s32(int32x4_t __p0) {
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vaddvq_s32((int8x16_t)__p0);
+ __ret = (int32_t) __builtin_neon_vaddvq_s32(__p0);
return __ret;
}
#else
__ai int32_t vaddvq_s32(int32x4_t __p0) {
int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vaddvq_s32((int8x16_t)__rev0);
+ __ret = (int32_t) __builtin_neon_vaddvq_s32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int64_t vaddvq_s64(int64x2_t __p0) {
int64_t __ret;
- __ret = (int64_t) __builtin_neon_vaddvq_s64((int8x16_t)__p0);
+ __ret = (int64_t) __builtin_neon_vaddvq_s64(__p0);
return __ret;
}
#else
__ai int64_t vaddvq_s64(int64x2_t __p0) {
int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
int64_t __ret;
- __ret = (int64_t) __builtin_neon_vaddvq_s64((int8x16_t)__rev0);
+ __ret = (int64_t) __builtin_neon_vaddvq_s64(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int16_t vaddvq_s16(int16x8_t __p0) {
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vaddvq_s16((int8x16_t)__p0);
+ __ret = (int16_t) __builtin_neon_vaddvq_s16(__p0);
return __ret;
}
#else
__ai int16_t vaddvq_s16(int16x8_t __p0) {
int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vaddvq_s16((int8x16_t)__rev0);
+ __ret = (int16_t) __builtin_neon_vaddvq_s16(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint8_t vaddv_u8(uint8x8_t __p0) {
uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vaddv_u8((int8x8_t)__p0);
+ __ret = (uint8_t) __builtin_neon_vaddv_u8(__p0);
return __ret;
}
#else
__ai uint8_t vaddv_u8(uint8x8_t __p0) {
uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vaddv_u8((int8x8_t)__rev0);
+ __ret = (uint8_t) __builtin_neon_vaddv_u8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint32_t vaddv_u32(uint32x2_t __p0) {
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vaddv_u32((int8x8_t)__p0);
+ __ret = (uint32_t) __builtin_neon_vaddv_u32(__p0);
return __ret;
}
#else
__ai uint32_t vaddv_u32(uint32x2_t __p0) {
uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vaddv_u32((int8x8_t)__rev0);
+ __ret = (uint32_t) __builtin_neon_vaddv_u32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint16_t vaddv_u16(uint16x4_t __p0) {
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vaddv_u16((int8x8_t)__p0);
+ __ret = (uint16_t) __builtin_neon_vaddv_u16(__p0);
return __ret;
}
#else
__ai uint16_t vaddv_u16(uint16x4_t __p0) {
uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vaddv_u16((int8x8_t)__rev0);
+ __ret = (uint16_t) __builtin_neon_vaddv_u16(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int8_t vaddv_s8(int8x8_t __p0) {
int8_t __ret;
- __ret = (int8_t) __builtin_neon_vaddv_s8((int8x8_t)__p0);
+ __ret = (int8_t) __builtin_neon_vaddv_s8(__p0);
return __ret;
}
#else
__ai int8_t vaddv_s8(int8x8_t __p0) {
int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
int8_t __ret;
- __ret = (int8_t) __builtin_neon_vaddv_s8((int8x8_t)__rev0);
+ __ret = (int8_t) __builtin_neon_vaddv_s8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float32_t vaddv_f32(float32x2_t __p0) {
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vaddv_f32((int8x8_t)__p0);
+ __ret = (float32_t) __builtin_neon_vaddv_f32(__p0);
return __ret;
}
#else
__ai float32_t vaddv_f32(float32x2_t __p0) {
float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vaddv_f32((int8x8_t)__rev0);
+ __ret = (float32_t) __builtin_neon_vaddv_f32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int32_t vaddv_s32(int32x2_t __p0) {
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vaddv_s32((int8x8_t)__p0);
+ __ret = (int32_t) __builtin_neon_vaddv_s32(__p0);
return __ret;
}
#else
__ai int32_t vaddv_s32(int32x2_t __p0) {
int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vaddv_s32((int8x8_t)__rev0);
+ __ret = (int32_t) __builtin_neon_vaddv_s32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int16_t vaddv_s16(int16x4_t __p0) {
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vaddv_s16((int8x8_t)__p0);
+ __ret = (int16_t) __builtin_neon_vaddv_s16(__p0);
return __ret;
}
#else
__ai int16_t vaddv_s16(int16x4_t __p0) {
int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vaddv_s16((int8x8_t)__rev0);
+ __ret = (int16_t) __builtin_neon_vaddv_s16(__rev0);
return __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai poly64x1_t vbsl_p64(uint64x1_t __p0, poly64x1_t __p1, poly64x1_t __p2) {
poly64x1_t __ret;
__ret = (poly64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 6);
return __ret;
}
-#else
-__ai poly64x1_t vbsl_p64(uint64x1_t __p0, poly64x1_t __p1, poly64x1_t __p2) {
- poly64x1_t __ret;
- __ret = (poly64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 6);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai poly64x2_t vbslq_p64(uint64x2_t __p0, poly64x2_t __p1, poly64x2_t __p2) {
poly64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vbsl_f64(uint64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
- return __ret;
-}
-#else
__ai float64x1_t vbsl_f64(uint64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
float64x1_t __ret;
__ret = (float64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vcageq_f64(float64x2_t __p0, float64x2_t __p1) {
uint64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vcage_f64(float64x1_t __p0, float64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vcage_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
- return __ret;
-}
-#else
__ai uint64x1_t vcage_f64(float64x1_t __p0, float64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vcage_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint64_t vcaged_f64(float64_t __p0, float64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcaged_f64(__p0, __p1);
return __ret;
}
-#else
-__ai uint64_t vcaged_f64(float64_t __p0, float64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcaged_f64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vcages_f32(float32_t __p0, float32_t __p1) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vcages_f32(__p0, __p1);
- return __ret;
-}
-#else
__ai uint32_t vcages_f32(float32_t __p0, float32_t __p1) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vcages_f32(__p0, __p1);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vcagtq_f64(float64x2_t __p0, float64x2_t __p1) {
uint64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vcagt_f64(float64x1_t __p0, float64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vcagt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
return __ret;
}
-#else
-__ai uint64x1_t vcagt_f64(float64x1_t __p0, float64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vcagt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint64_t vcagtd_f64(float64_t __p0, float64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcagtd_f64(__p0, __p1);
return __ret;
}
-#else
-__ai uint64_t vcagtd_f64(float64_t __p0, float64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcagtd_f64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint32_t vcagts_f32(float32_t __p0, float32_t __p1) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vcagts_f32(__p0, __p1);
return __ret;
}
-#else
-__ai uint32_t vcagts_f32(float32_t __p0, float32_t __p1) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vcagts_f32(__p0, __p1);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vcaleq_f64(float64x2_t __p0, float64x2_t __p1) {
uint64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vcale_f64(float64x1_t __p0, float64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vcale_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
- return __ret;
-}
-#else
__ai uint64x1_t vcale_f64(float64x1_t __p0, float64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vcale_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vcaled_f64(float64_t __p0, float64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcaled_f64(__p0, __p1);
- return __ret;
-}
-#else
__ai uint64_t vcaled_f64(float64_t __p0, float64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcaled_f64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint32_t vcales_f32(float32_t __p0, float32_t __p1) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vcales_f32(__p0, __p1);
return __ret;
}
-#else
-__ai uint32_t vcales_f32(float32_t __p0, float32_t __p1) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vcales_f32(__p0, __p1);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vcaltq_f64(float64x2_t __p0, float64x2_t __p1) {
uint64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vcalt_f64(float64x1_t __p0, float64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vcalt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
return __ret;
}
-#else
-__ai uint64x1_t vcalt_f64(float64x1_t __p0, float64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vcalt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint64_t vcaltd_f64(float64_t __p0, float64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcaltd_f64(__p0, __p1);
return __ret;
}
-#else
-__ai uint64_t vcaltd_f64(float64_t __p0, float64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcaltd_f64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint32_t vcalts_f32(float32_t __p0, float32_t __p1) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vcalts_f32(__p0, __p1);
return __ret;
}
-#else
-__ai uint32_t vcalts_f32(float32_t __p0, float32_t __p1) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vcalts_f32(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vceq_p64(poly64x1_t __p0, poly64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 == __p1);
- return __ret;
-}
-#else
__ai uint64x1_t vceq_p64(poly64x1_t __p0, poly64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 == __p1);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vceqq_p64(poly64x2_t __p0, poly64x2_t __p1) {
uint64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vceq_u64(uint64x1_t __p0, uint64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 == __p1);
return __ret;
}
-#else
-__ai uint64x1_t vceq_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 == __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vceq_f64(float64x1_t __p0, float64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 == __p1);
- return __ret;
-}
-#else
__ai uint64x1_t vceq_f64(float64x1_t __p0, float64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 == __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vceq_s64(int64x1_t __p0, int64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 == __p1);
- return __ret;
-}
-#else
__ai uint64x1_t vceq_s64(int64x1_t __p0, int64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 == __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint64_t vceqd_u64(uint64_t __p0, uint64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vceqd_u64(__p0, __p1);
return __ret;
}
-#else
-__ai uint64_t vceqd_u64(uint64_t __p0, uint64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vceqd_u64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vceqd_s64(int64_t __p0, int64_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vceqd_s64(__p0, __p1);
- return __ret;
-}
-#else
__ai int64_t vceqd_s64(int64_t __p0, int64_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vceqd_s64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint64_t vceqd_f64(float64_t __p0, float64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vceqd_f64(__p0, __p1);
return __ret;
}
-#else
-__ai uint64_t vceqd_f64(float64_t __p0, float64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vceqd_f64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint32_t vceqs_f32(float32_t __p0, float32_t __p1) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vceqs_f32(__p0, __p1);
return __ret;
}
-#else
-__ai uint32_t vceqs_f32(float32_t __p0, float32_t __p1) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vceqs_f32(__p0, __p1);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint8x8_t vceqz_p8(poly8x8_t __p0) {
uint8x8_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vceqz_p64(poly64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#else
__ai uint64x1_t vceqz_p64(poly64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vceqz_p16(poly16x4_t __p0) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vceqz_u64(uint64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
return __ret;
}
-#else
-__ai uint64x1_t vceqz_u64(uint64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vceqz_u16(uint16x4_t __p0) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vceqz_f64(float64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
return __ret;
}
-#else
-__ai uint64x1_t vceqz_f64(float64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint32x2_t vceqz_f32(float32x2_t __p0) {
uint32x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vceqz_s64(int64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
return __ret;
}
-#else
-__ai uint64x1_t vceqz_s64(int64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vceqz_s16(int16x4_t __p0) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vceqzd_u64(uint64_t __p0) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vceqzd_u64(__p0);
- return __ret;
-}
-#else
__ai uint64_t vceqzd_u64(uint64_t __p0) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vceqzd_u64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vceqzd_s64(int64_t __p0) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vceqzd_s64(__p0);
- return __ret;
-}
-#else
__ai int64_t vceqzd_s64(int64_t __p0) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vceqzd_s64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint64_t vceqzd_f64(float64_t __p0) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vceqzd_f64(__p0);
return __ret;
}
-#else
-__ai uint64_t vceqzd_f64(float64_t __p0) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vceqzd_f64(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vceqzs_f32(float32_t __p0) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vceqzs_f32(__p0);
- return __ret;
-}
-#else
__ai uint32_t vceqzs_f32(float32_t __p0) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vceqzs_f32(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vcgeq_u64(uint64x2_t __p0, uint64x2_t __p1) {
uint64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vcge_u64(uint64x1_t __p0, uint64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 >= __p1);
return __ret;
}
-#else
-__ai uint64x1_t vcge_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 >= __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vcge_f64(float64x1_t __p0, float64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 >= __p1);
- return __ret;
-}
-#else
__ai uint64x1_t vcge_f64(float64x1_t __p0, float64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 >= __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vcge_s64(int64x1_t __p0, int64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 >= __p1);
- return __ret;
-}
-#else
__ai uint64x1_t vcge_s64(int64x1_t __p0, int64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 >= __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai int64_t vcged_s64(int64_t __p0, int64_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vcged_s64(__p0, __p1);
return __ret;
}
-#else
-__ai int64_t vcged_s64(int64_t __p0, int64_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vcged_s64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vcged_u64(uint64_t __p0, uint64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcged_u64(__p0, __p1);
- return __ret;
-}
-#else
__ai uint64_t vcged_u64(uint64_t __p0, uint64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcged_u64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint64_t vcged_f64(float64_t __p0, float64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcged_f64(__p0, __p1);
return __ret;
}
-#else
-__ai uint64_t vcged_f64(float64_t __p0, float64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcged_f64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint32_t vcges_f32(float32_t __p0, float32_t __p1) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vcges_f32(__p0, __p1);
return __ret;
}
-#else
-__ai uint32_t vcges_f32(float32_t __p0, float32_t __p1) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vcges_f32(__p0, __p1);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint8x16_t vcgezq_s8(int8x16_t __p0) {
uint8x16_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vcgez_f64(float64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#else
__ai uint64x1_t vcgez_f64(float64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 19);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint32x2_t vcgez_f32(float32x2_t __p0) {
uint32x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vcgez_s64(int64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 19);
return __ret;
}
-#else
-__ai uint64x1_t vcgez_s64(int64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vcgez_s16(int16x4_t __p0) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vcgezd_s64(int64_t __p0) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vcgezd_s64(__p0);
- return __ret;
-}
-#else
__ai int64_t vcgezd_s64(int64_t __p0) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vcgezd_s64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vcgezd_f64(float64_t __p0) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcgezd_f64(__p0);
- return __ret;
-}
-#else
__ai uint64_t vcgezd_f64(float64_t __p0) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcgezd_f64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint32_t vcgezs_f32(float32_t __p0) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vcgezs_f32(__p0);
return __ret;
}
-#else
-__ai uint32_t vcgezs_f32(float32_t __p0) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vcgezs_f32(__p0);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vcgtq_u64(uint64x2_t __p0, uint64x2_t __p1) {
uint64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vcgt_u64(uint64x1_t __p0, uint64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 > __p1);
return __ret;
}
-#else
-__ai uint64x1_t vcgt_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 > __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vcgt_f64(float64x1_t __p0, float64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 > __p1);
- return __ret;
-}
-#else
__ai uint64x1_t vcgt_f64(float64x1_t __p0, float64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 > __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vcgt_s64(int64x1_t __p0, int64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 > __p1);
- return __ret;
-}
-#else
__ai uint64x1_t vcgt_s64(int64x1_t __p0, int64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 > __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai int64_t vcgtd_s64(int64_t __p0, int64_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vcgtd_s64(__p0, __p1);
return __ret;
}
-#else
-__ai int64_t vcgtd_s64(int64_t __p0, int64_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vcgtd_s64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vcgtd_u64(uint64_t __p0, uint64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcgtd_u64(__p0, __p1);
- return __ret;
-}
-#else
__ai uint64_t vcgtd_u64(uint64_t __p0, uint64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcgtd_u64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vcgtd_f64(float64_t __p0, float64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcgtd_f64(__p0, __p1);
- return __ret;
-}
-#else
__ai uint64_t vcgtd_f64(float64_t __p0, float64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcgtd_f64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vcgts_f32(float32_t __p0, float32_t __p1) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vcgts_f32(__p0, __p1);
- return __ret;
-}
-#else
__ai uint32_t vcgts_f32(float32_t __p0, float32_t __p1) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vcgts_f32(__p0, __p1);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint8x16_t vcgtzq_s8(int8x16_t __p0) {
uint8x16_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vcgtz_f64(float64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#else
__ai uint64x1_t vcgtz_f64(float64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 19);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint32x2_t vcgtz_f32(float32x2_t __p0) {
uint32x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vcgtz_s64(int64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 19);
return __ret;
}
-#else
-__ai uint64x1_t vcgtz_s64(int64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vcgtz_s16(int16x4_t __p0) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai int64_t vcgtzd_s64(int64_t __p0) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vcgtzd_s64(__p0);
return __ret;
}
-#else
-__ai int64_t vcgtzd_s64(int64_t __p0) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vcgtzd_s64(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vcgtzd_f64(float64_t __p0) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcgtzd_f64(__p0);
- return __ret;
-}
-#else
__ai uint64_t vcgtzd_f64(float64_t __p0) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcgtzd_f64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vcgtzs_f32(float32_t __p0) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vcgtzs_f32(__p0);
- return __ret;
-}
-#else
__ai uint32_t vcgtzs_f32(float32_t __p0) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vcgtzs_f32(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vcleq_u64(uint64x2_t __p0, uint64x2_t __p1) {
uint64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vcle_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 <= __p1);
- return __ret;
-}
-#else
__ai uint64x1_t vcle_u64(uint64x1_t __p0, uint64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 <= __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vcle_f64(float64x1_t __p0, float64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 <= __p1);
- return __ret;
-}
-#else
__ai uint64x1_t vcle_f64(float64x1_t __p0, float64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 <= __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vcle_s64(int64x1_t __p0, int64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 <= __p1);
return __ret;
}
-#else
-__ai uint64x1_t vcle_s64(int64x1_t __p0, int64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 <= __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint64_t vcled_u64(uint64_t __p0, uint64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcled_u64(__p0, __p1);
return __ret;
}
-#else
-__ai uint64_t vcled_u64(uint64_t __p0, uint64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcled_u64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai int64_t vcled_s64(int64_t __p0, int64_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vcled_s64(__p0, __p1);
return __ret;
}
-#else
-__ai int64_t vcled_s64(int64_t __p0, int64_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vcled_s64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vcled_f64(float64_t __p0, float64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcled_f64(__p0, __p1);
- return __ret;
-}
-#else
__ai uint64_t vcled_f64(float64_t __p0, float64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcled_f64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint32_t vcles_f32(float32_t __p0, float32_t __p1) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vcles_f32(__p0, __p1);
return __ret;
}
-#else
-__ai uint32_t vcles_f32(float32_t __p0, float32_t __p1) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vcles_f32(__p0, __p1);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint8x16_t vclezq_s8(int8x16_t __p0) {
uint8x16_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vclez_f64(float64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vclez_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#else
__ai uint64x1_t vclez_f64(float64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vclez_v((int8x8_t)__p0, 19);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint32x2_t vclez_f32(float32x2_t __p0) {
uint32x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vclez_s64(int64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vclez_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#else
__ai uint64x1_t vclez_s64(int64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vclez_v((int8x8_t)__p0, 19);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vclez_s16(int16x4_t __p0) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai int64_t vclezd_s64(int64_t __p0) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vclezd_s64(__p0);
return __ret;
}
-#else
-__ai int64_t vclezd_s64(int64_t __p0) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vclezd_s64(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vclezd_f64(float64_t __p0) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vclezd_f64(__p0);
- return __ret;
-}
-#else
__ai uint64_t vclezd_f64(float64_t __p0) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vclezd_f64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vclezs_f32(float32_t __p0) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vclezs_f32(__p0);
- return __ret;
-}
-#else
__ai uint32_t vclezs_f32(float32_t __p0) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vclezs_f32(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vcltq_u64(uint64x2_t __p0, uint64x2_t __p1) {
uint64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vclt_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 < __p1);
- return __ret;
-}
-#else
__ai uint64x1_t vclt_u64(uint64x1_t __p0, uint64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 < __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vclt_f64(float64x1_t __p0, float64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 < __p1);
- return __ret;
-}
-#else
__ai uint64x1_t vclt_f64(float64x1_t __p0, float64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 < __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vclt_s64(int64x1_t __p0, int64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t)(__p0 < __p1);
- return __ret;
-}
-#else
__ai uint64x1_t vclt_s64(int64x1_t __p0, int64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t)(__p0 < __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vcltd_u64(uint64_t __p0, uint64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcltd_u64(__p0, __p1);
- return __ret;
-}
-#else
__ai uint64_t vcltd_u64(uint64_t __p0, uint64_t __p1) {
uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcltd_u64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vcltd_s64(int64_t __p0, int64_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vcltd_s64(__p0, __p1);
+ __ret = (uint64_t) __builtin_neon_vcltd_u64(__p0, __p1);
return __ret;
}
-#else
__ai int64_t vcltd_s64(int64_t __p0, int64_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vcltd_s64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vcltd_f64(float64_t __p0, float64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcltd_f64(__p0, __p1);
- return __ret;
-}
-#else
__ai uint64_t vcltd_f64(float64_t __p0, float64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcltd_f64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vclts_f32(float32_t __p0, float32_t __p1) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vclts_f32(__p0, __p1);
- return __ret;
-}
-#else
__ai uint32_t vclts_f32(float32_t __p0, float32_t __p1) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vclts_f32(__p0, __p1);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint8x16_t vcltzq_s8(int8x16_t __p0) {
uint8x16_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vcltz_f64(float64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#else
__ai uint64x1_t vcltz_f64(float64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 19);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint32x2_t vcltz_f32(float32x2_t __p0) {
uint32x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vcltz_s64(int64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 19);
return __ret;
}
-#else
-__ai uint64x1_t vcltz_s64(int64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x4_t vcltz_s16(int16x4_t __p0) {
uint16x4_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vcltzd_s64(int64_t __p0) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vcltzd_s64(__p0);
- return __ret;
-}
-#else
__ai int64_t vcltzd_s64(int64_t __p0) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vcltzd_s64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vcltzd_f64(float64_t __p0) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcltzd_f64(__p0);
- return __ret;
-}
-#else
__ai uint64_t vcltzd_f64(float64_t __p0) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcltzd_f64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vcltzs_f32(float32_t __p0) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vcltzs_f32(__p0);
- return __ret;
-}
-#else
__ai uint32_t vcltzs_f32(float32_t __p0) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vcltzs_f32(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai poly64x2_t vcombine_p64(poly64x1_t __p0, poly64x1_t __p1) {
poly64x2_t __ret;
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_lane_p8(__p0_12, __p1_12, __p2_12, __p3_12) __extension__ ({ \
- poly8x16_t __s0_12 = __p0_12; \
- poly8x8_t __s2_12 = __p2_12; \
- poly8x16_t __ret_12; \
- __ret_12 = vsetq_lane_p8(vget_lane_p8(__s2_12, __p3_12), __s0_12, __p1_12); \
- __ret_12; \
+#define vcopyq_lane_p8(__p0_249, __p1_249, __p2_249, __p3_249) __extension__ ({ \
+ poly8x16_t __s0_249 = __p0_249; \
+ poly8x8_t __s2_249 = __p2_249; \
+ poly8x16_t __ret_249; \
+ __ret_249 = vsetq_lane_p8(vget_lane_p8(__s2_249, __p3_249), __s0_249, __p1_249); \
+ __ret_249; \
})
#else
-#define vcopyq_lane_p8(__p0_13, __p1_13, __p2_13, __p3_13) __extension__ ({ \
- poly8x16_t __s0_13 = __p0_13; \
- poly8x8_t __s2_13 = __p2_13; \
- poly8x16_t __rev0_13; __rev0_13 = __builtin_shufflevector(__s0_13, __s0_13, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly8x8_t __rev2_13; __rev2_13 = __builtin_shufflevector(__s2_13, __s2_13, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly8x16_t __ret_13; \
- __ret_13 = __noswap_vsetq_lane_p8(__noswap_vget_lane_p8(__rev2_13, __p3_13), __rev0_13, __p1_13); \
- __ret_13 = __builtin_shufflevector(__ret_13, __ret_13, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_13; \
+#define vcopyq_lane_p8(__p0_250, __p1_250, __p2_250, __p3_250) __extension__ ({ \
+ poly8x16_t __s0_250 = __p0_250; \
+ poly8x8_t __s2_250 = __p2_250; \
+ poly8x16_t __rev0_250; __rev0_250 = __builtin_shufflevector(__s0_250, __s0_250, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x8_t __rev2_250; __rev2_250 = __builtin_shufflevector(__s2_250, __s2_250, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x16_t __ret_250; \
+ __ret_250 = __noswap_vsetq_lane_p8(__noswap_vget_lane_p8(__rev2_250, __p3_250), __rev0_250, __p1_250); \
+ __ret_250 = __builtin_shufflevector(__ret_250, __ret_250, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_250; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_lane_p16(__p0_14, __p1_14, __p2_14, __p3_14) __extension__ ({ \
- poly16x8_t __s0_14 = __p0_14; \
- poly16x4_t __s2_14 = __p2_14; \
- poly16x8_t __ret_14; \
- __ret_14 = vsetq_lane_p16(vget_lane_p16(__s2_14, __p3_14), __s0_14, __p1_14); \
- __ret_14; \
+#define vcopyq_lane_p16(__p0_251, __p1_251, __p2_251, __p3_251) __extension__ ({ \
+ poly16x8_t __s0_251 = __p0_251; \
+ poly16x4_t __s2_251 = __p2_251; \
+ poly16x8_t __ret_251; \
+ __ret_251 = vsetq_lane_p16(vget_lane_p16(__s2_251, __p3_251), __s0_251, __p1_251); \
+ __ret_251; \
})
#else
-#define vcopyq_lane_p16(__p0_15, __p1_15, __p2_15, __p3_15) __extension__ ({ \
- poly16x8_t __s0_15 = __p0_15; \
- poly16x4_t __s2_15 = __p2_15; \
- poly16x8_t __rev0_15; __rev0_15 = __builtin_shufflevector(__s0_15, __s0_15, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly16x4_t __rev2_15; __rev2_15 = __builtin_shufflevector(__s2_15, __s2_15, 3, 2, 1, 0); \
- poly16x8_t __ret_15; \
- __ret_15 = __noswap_vsetq_lane_p16(__noswap_vget_lane_p16(__rev2_15, __p3_15), __rev0_15, __p1_15); \
- __ret_15 = __builtin_shufflevector(__ret_15, __ret_15, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_15; \
+#define vcopyq_lane_p16(__p0_252, __p1_252, __p2_252, __p3_252) __extension__ ({ \
+ poly16x8_t __s0_252 = __p0_252; \
+ poly16x4_t __s2_252 = __p2_252; \
+ poly16x8_t __rev0_252; __rev0_252 = __builtin_shufflevector(__s0_252, __s0_252, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly16x4_t __rev2_252; __rev2_252 = __builtin_shufflevector(__s2_252, __s2_252, 3, 2, 1, 0); \
+ poly16x8_t __ret_252; \
+ __ret_252 = __noswap_vsetq_lane_p16(__noswap_vget_lane_p16(__rev2_252, __p3_252), __rev0_252, __p1_252); \
+ __ret_252 = __builtin_shufflevector(__ret_252, __ret_252, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_252; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_lane_u8(__p0_16, __p1_16, __p2_16, __p3_16) __extension__ ({ \
- uint8x16_t __s0_16 = __p0_16; \
- uint8x8_t __s2_16 = __p2_16; \
- uint8x16_t __ret_16; \
- __ret_16 = vsetq_lane_u8(vget_lane_u8(__s2_16, __p3_16), __s0_16, __p1_16); \
- __ret_16; \
+#define vcopyq_lane_u8(__p0_253, __p1_253, __p2_253, __p3_253) __extension__ ({ \
+ uint8x16_t __s0_253 = __p0_253; \
+ uint8x8_t __s2_253 = __p2_253; \
+ uint8x16_t __ret_253; \
+ __ret_253 = vsetq_lane_u8(vget_lane_u8(__s2_253, __p3_253), __s0_253, __p1_253); \
+ __ret_253; \
})
#else
-#define vcopyq_lane_u8(__p0_17, __p1_17, __p2_17, __p3_17) __extension__ ({ \
- uint8x16_t __s0_17 = __p0_17; \
- uint8x8_t __s2_17 = __p2_17; \
- uint8x16_t __rev0_17; __rev0_17 = __builtin_shufflevector(__s0_17, __s0_17, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x8_t __rev2_17; __rev2_17 = __builtin_shufflevector(__s2_17, __s2_17, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x16_t __ret_17; \
- __ret_17 = __noswap_vsetq_lane_u8(__noswap_vget_lane_u8(__rev2_17, __p3_17), __rev0_17, __p1_17); \
- __ret_17 = __builtin_shufflevector(__ret_17, __ret_17, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_17; \
+#define vcopyq_lane_u8(__p0_254, __p1_254, __p2_254, __p3_254) __extension__ ({ \
+ uint8x16_t __s0_254 = __p0_254; \
+ uint8x8_t __s2_254 = __p2_254; \
+ uint8x16_t __rev0_254; __rev0_254 = __builtin_shufflevector(__s0_254, __s0_254, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x8_t __rev2_254; __rev2_254 = __builtin_shufflevector(__s2_254, __s2_254, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __ret_254; \
+ __ret_254 = __noswap_vsetq_lane_u8(__noswap_vget_lane_u8(__rev2_254, __p3_254), __rev0_254, __p1_254); \
+ __ret_254 = __builtin_shufflevector(__ret_254, __ret_254, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_254; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_lane_u32(__p0_18, __p1_18, __p2_18, __p3_18) __extension__ ({ \
- uint32x4_t __s0_18 = __p0_18; \
- uint32x2_t __s2_18 = __p2_18; \
- uint32x4_t __ret_18; \
- __ret_18 = vsetq_lane_u32(vget_lane_u32(__s2_18, __p3_18), __s0_18, __p1_18); \
- __ret_18; \
+#define vcopyq_lane_u32(__p0_255, __p1_255, __p2_255, __p3_255) __extension__ ({ \
+ uint32x4_t __s0_255 = __p0_255; \
+ uint32x2_t __s2_255 = __p2_255; \
+ uint32x4_t __ret_255; \
+ __ret_255 = vsetq_lane_u32(vget_lane_u32(__s2_255, __p3_255), __s0_255, __p1_255); \
+ __ret_255; \
})
#else
-#define vcopyq_lane_u32(__p0_19, __p1_19, __p2_19, __p3_19) __extension__ ({ \
- uint32x4_t __s0_19 = __p0_19; \
- uint32x2_t __s2_19 = __p2_19; \
- uint32x4_t __rev0_19; __rev0_19 = __builtin_shufflevector(__s0_19, __s0_19, 3, 2, 1, 0); \
- uint32x2_t __rev2_19; __rev2_19 = __builtin_shufflevector(__s2_19, __s2_19, 1, 0); \
- uint32x4_t __ret_19; \
- __ret_19 = __noswap_vsetq_lane_u32(__noswap_vget_lane_u32(__rev2_19, __p3_19), __rev0_19, __p1_19); \
- __ret_19 = __builtin_shufflevector(__ret_19, __ret_19, 3, 2, 1, 0); \
- __ret_19; \
+#define vcopyq_lane_u32(__p0_256, __p1_256, __p2_256, __p3_256) __extension__ ({ \
+ uint32x4_t __s0_256 = __p0_256; \
+ uint32x2_t __s2_256 = __p2_256; \
+ uint32x4_t __rev0_256; __rev0_256 = __builtin_shufflevector(__s0_256, __s0_256, 3, 2, 1, 0); \
+ uint32x2_t __rev2_256; __rev2_256 = __builtin_shufflevector(__s2_256, __s2_256, 1, 0); \
+ uint32x4_t __ret_256; \
+ __ret_256 = __noswap_vsetq_lane_u32(__noswap_vget_lane_u32(__rev2_256, __p3_256), __rev0_256, __p1_256); \
+ __ret_256 = __builtin_shufflevector(__ret_256, __ret_256, 3, 2, 1, 0); \
+ __ret_256; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_lane_u64(__p0_20, __p1_20, __p2_20, __p3_20) __extension__ ({ \
- uint64x2_t __s0_20 = __p0_20; \
- uint64x1_t __s2_20 = __p2_20; \
- uint64x2_t __ret_20; \
- __ret_20 = vsetq_lane_u64(vget_lane_u64(__s2_20, __p3_20), __s0_20, __p1_20); \
- __ret_20; \
+#define vcopyq_lane_u64(__p0_257, __p1_257, __p2_257, __p3_257) __extension__ ({ \
+ uint64x2_t __s0_257 = __p0_257; \
+ uint64x1_t __s2_257 = __p2_257; \
+ uint64x2_t __ret_257; \
+ __ret_257 = vsetq_lane_u64(vget_lane_u64(__s2_257, __p3_257), __s0_257, __p1_257); \
+ __ret_257; \
})
#else
-#define vcopyq_lane_u64(__p0_21, __p1_21, __p2_21, __p3_21) __extension__ ({ \
- uint64x2_t __s0_21 = __p0_21; \
- uint64x1_t __s2_21 = __p2_21; \
- uint64x2_t __rev0_21; __rev0_21 = __builtin_shufflevector(__s0_21, __s0_21, 1, 0); \
- uint64x2_t __ret_21; \
- __ret_21 = __noswap_vsetq_lane_u64(__noswap_vget_lane_u64(__s2_21, __p3_21), __rev0_21, __p1_21); \
- __ret_21 = __builtin_shufflevector(__ret_21, __ret_21, 1, 0); \
- __ret_21; \
+#define vcopyq_lane_u64(__p0_258, __p1_258, __p2_258, __p3_258) __extension__ ({ \
+ uint64x2_t __s0_258 = __p0_258; \
+ uint64x1_t __s2_258 = __p2_258; \
+ uint64x2_t __rev0_258; __rev0_258 = __builtin_shufflevector(__s0_258, __s0_258, 1, 0); \
+ uint64x2_t __ret_258; \
+ __ret_258 = __noswap_vsetq_lane_u64(vget_lane_u64(__s2_258, __p3_258), __rev0_258, __p1_258); \
+ __ret_258 = __builtin_shufflevector(__ret_258, __ret_258, 1, 0); \
+ __ret_258; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_lane_u16(__p0_22, __p1_22, __p2_22, __p3_22) __extension__ ({ \
- uint16x8_t __s0_22 = __p0_22; \
- uint16x4_t __s2_22 = __p2_22; \
- uint16x8_t __ret_22; \
- __ret_22 = vsetq_lane_u16(vget_lane_u16(__s2_22, __p3_22), __s0_22, __p1_22); \
- __ret_22; \
+#define vcopyq_lane_u16(__p0_259, __p1_259, __p2_259, __p3_259) __extension__ ({ \
+ uint16x8_t __s0_259 = __p0_259; \
+ uint16x4_t __s2_259 = __p2_259; \
+ uint16x8_t __ret_259; \
+ __ret_259 = vsetq_lane_u16(vget_lane_u16(__s2_259, __p3_259), __s0_259, __p1_259); \
+ __ret_259; \
})
#else
-#define vcopyq_lane_u16(__p0_23, __p1_23, __p2_23, __p3_23) __extension__ ({ \
- uint16x8_t __s0_23 = __p0_23; \
- uint16x4_t __s2_23 = __p2_23; \
- uint16x8_t __rev0_23; __rev0_23 = __builtin_shufflevector(__s0_23, __s0_23, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x4_t __rev2_23; __rev2_23 = __builtin_shufflevector(__s2_23, __s2_23, 3, 2, 1, 0); \
- uint16x8_t __ret_23; \
- __ret_23 = __noswap_vsetq_lane_u16(__noswap_vget_lane_u16(__rev2_23, __p3_23), __rev0_23, __p1_23); \
- __ret_23 = __builtin_shufflevector(__ret_23, __ret_23, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_23; \
+#define vcopyq_lane_u16(__p0_260, __p1_260, __p2_260, __p3_260) __extension__ ({ \
+ uint16x8_t __s0_260 = __p0_260; \
+ uint16x4_t __s2_260 = __p2_260; \
+ uint16x8_t __rev0_260; __rev0_260 = __builtin_shufflevector(__s0_260, __s0_260, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x4_t __rev2_260; __rev2_260 = __builtin_shufflevector(__s2_260, __s2_260, 3, 2, 1, 0); \
+ uint16x8_t __ret_260; \
+ __ret_260 = __noswap_vsetq_lane_u16(__noswap_vget_lane_u16(__rev2_260, __p3_260), __rev0_260, __p1_260); \
+ __ret_260 = __builtin_shufflevector(__ret_260, __ret_260, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_260; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_lane_s8(__p0_24, __p1_24, __p2_24, __p3_24) __extension__ ({ \
- int8x16_t __s0_24 = __p0_24; \
- int8x8_t __s2_24 = __p2_24; \
- int8x16_t __ret_24; \
- __ret_24 = vsetq_lane_s8(vget_lane_s8(__s2_24, __p3_24), __s0_24, __p1_24); \
- __ret_24; \
+#define vcopyq_lane_s8(__p0_261, __p1_261, __p2_261, __p3_261) __extension__ ({ \
+ int8x16_t __s0_261 = __p0_261; \
+ int8x8_t __s2_261 = __p2_261; \
+ int8x16_t __ret_261; \
+ __ret_261 = vsetq_lane_s8(vget_lane_s8(__s2_261, __p3_261), __s0_261, __p1_261); \
+ __ret_261; \
})
#else
-#define vcopyq_lane_s8(__p0_25, __p1_25, __p2_25, __p3_25) __extension__ ({ \
- int8x16_t __s0_25 = __p0_25; \
- int8x8_t __s2_25 = __p2_25; \
- int8x16_t __rev0_25; __rev0_25 = __builtin_shufflevector(__s0_25, __s0_25, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x8_t __rev2_25; __rev2_25 = __builtin_shufflevector(__s2_25, __s2_25, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x16_t __ret_25; \
- __ret_25 = __noswap_vsetq_lane_s8(__noswap_vget_lane_s8(__rev2_25, __p3_25), __rev0_25, __p1_25); \
- __ret_25 = __builtin_shufflevector(__ret_25, __ret_25, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_25; \
+#define vcopyq_lane_s8(__p0_262, __p1_262, __p2_262, __p3_262) __extension__ ({ \
+ int8x16_t __s0_262 = __p0_262; \
+ int8x8_t __s2_262 = __p2_262; \
+ int8x16_t __rev0_262; __rev0_262 = __builtin_shufflevector(__s0_262, __s0_262, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x8_t __rev2_262; __rev2_262 = __builtin_shufflevector(__s2_262, __s2_262, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __ret_262; \
+ __ret_262 = __noswap_vsetq_lane_s8(__noswap_vget_lane_s8(__rev2_262, __p3_262), __rev0_262, __p1_262); \
+ __ret_262 = __builtin_shufflevector(__ret_262, __ret_262, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_262; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_lane_f32(__p0_26, __p1_26, __p2_26, __p3_26) __extension__ ({ \
- float32x4_t __s0_26 = __p0_26; \
- float32x2_t __s2_26 = __p2_26; \
- float32x4_t __ret_26; \
- __ret_26 = vsetq_lane_f32(vget_lane_f32(__s2_26, __p3_26), __s0_26, __p1_26); \
- __ret_26; \
+#define vcopyq_lane_f32(__p0_263, __p1_263, __p2_263, __p3_263) __extension__ ({ \
+ float32x4_t __s0_263 = __p0_263; \
+ float32x2_t __s2_263 = __p2_263; \
+ float32x4_t __ret_263; \
+ __ret_263 = vsetq_lane_f32(vget_lane_f32(__s2_263, __p3_263), __s0_263, __p1_263); \
+ __ret_263; \
})
#else
-#define vcopyq_lane_f32(__p0_27, __p1_27, __p2_27, __p3_27) __extension__ ({ \
- float32x4_t __s0_27 = __p0_27; \
- float32x2_t __s2_27 = __p2_27; \
- float32x4_t __rev0_27; __rev0_27 = __builtin_shufflevector(__s0_27, __s0_27, 3, 2, 1, 0); \
- float32x2_t __rev2_27; __rev2_27 = __builtin_shufflevector(__s2_27, __s2_27, 1, 0); \
- float32x4_t __ret_27; \
- __ret_27 = __noswap_vsetq_lane_f32(__noswap_vget_lane_f32(__rev2_27, __p3_27), __rev0_27, __p1_27); \
- __ret_27 = __builtin_shufflevector(__ret_27, __ret_27, 3, 2, 1, 0); \
- __ret_27; \
+#define vcopyq_lane_f32(__p0_264, __p1_264, __p2_264, __p3_264) __extension__ ({ \
+ float32x4_t __s0_264 = __p0_264; \
+ float32x2_t __s2_264 = __p2_264; \
+ float32x4_t __rev0_264; __rev0_264 = __builtin_shufflevector(__s0_264, __s0_264, 3, 2, 1, 0); \
+ float32x2_t __rev2_264; __rev2_264 = __builtin_shufflevector(__s2_264, __s2_264, 1, 0); \
+ float32x4_t __ret_264; \
+ __ret_264 = __noswap_vsetq_lane_f32(__noswap_vget_lane_f32(__rev2_264, __p3_264), __rev0_264, __p1_264); \
+ __ret_264 = __builtin_shufflevector(__ret_264, __ret_264, 3, 2, 1, 0); \
+ __ret_264; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_lane_s32(__p0_28, __p1_28, __p2_28, __p3_28) __extension__ ({ \
- int32x4_t __s0_28 = __p0_28; \
- int32x2_t __s2_28 = __p2_28; \
- int32x4_t __ret_28; \
- __ret_28 = vsetq_lane_s32(vget_lane_s32(__s2_28, __p3_28), __s0_28, __p1_28); \
- __ret_28; \
+#define vcopyq_lane_s32(__p0_265, __p1_265, __p2_265, __p3_265) __extension__ ({ \
+ int32x4_t __s0_265 = __p0_265; \
+ int32x2_t __s2_265 = __p2_265; \
+ int32x4_t __ret_265; \
+ __ret_265 = vsetq_lane_s32(vget_lane_s32(__s2_265, __p3_265), __s0_265, __p1_265); \
+ __ret_265; \
})
#else
-#define vcopyq_lane_s32(__p0_29, __p1_29, __p2_29, __p3_29) __extension__ ({ \
- int32x4_t __s0_29 = __p0_29; \
- int32x2_t __s2_29 = __p2_29; \
- int32x4_t __rev0_29; __rev0_29 = __builtin_shufflevector(__s0_29, __s0_29, 3, 2, 1, 0); \
- int32x2_t __rev2_29; __rev2_29 = __builtin_shufflevector(__s2_29, __s2_29, 1, 0); \
- int32x4_t __ret_29; \
- __ret_29 = __noswap_vsetq_lane_s32(__noswap_vget_lane_s32(__rev2_29, __p3_29), __rev0_29, __p1_29); \
- __ret_29 = __builtin_shufflevector(__ret_29, __ret_29, 3, 2, 1, 0); \
- __ret_29; \
+#define vcopyq_lane_s32(__p0_266, __p1_266, __p2_266, __p3_266) __extension__ ({ \
+ int32x4_t __s0_266 = __p0_266; \
+ int32x2_t __s2_266 = __p2_266; \
+ int32x4_t __rev0_266; __rev0_266 = __builtin_shufflevector(__s0_266, __s0_266, 3, 2, 1, 0); \
+ int32x2_t __rev2_266; __rev2_266 = __builtin_shufflevector(__s2_266, __s2_266, 1, 0); \
+ int32x4_t __ret_266; \
+ __ret_266 = __noswap_vsetq_lane_s32(__noswap_vget_lane_s32(__rev2_266, __p3_266), __rev0_266, __p1_266); \
+ __ret_266 = __builtin_shufflevector(__ret_266, __ret_266, 3, 2, 1, 0); \
+ __ret_266; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_lane_s64(__p0_30, __p1_30, __p2_30, __p3_30) __extension__ ({ \
- int64x2_t __s0_30 = __p0_30; \
- int64x1_t __s2_30 = __p2_30; \
- int64x2_t __ret_30; \
- __ret_30 = vsetq_lane_s64(vget_lane_s64(__s2_30, __p3_30), __s0_30, __p1_30); \
- __ret_30; \
+#define vcopyq_lane_s64(__p0_267, __p1_267, __p2_267, __p3_267) __extension__ ({ \
+ int64x2_t __s0_267 = __p0_267; \
+ int64x1_t __s2_267 = __p2_267; \
+ int64x2_t __ret_267; \
+ __ret_267 = vsetq_lane_s64(vget_lane_s64(__s2_267, __p3_267), __s0_267, __p1_267); \
+ __ret_267; \
})
#else
-#define vcopyq_lane_s64(__p0_31, __p1_31, __p2_31, __p3_31) __extension__ ({ \
- int64x2_t __s0_31 = __p0_31; \
- int64x1_t __s2_31 = __p2_31; \
- int64x2_t __rev0_31; __rev0_31 = __builtin_shufflevector(__s0_31, __s0_31, 1, 0); \
- int64x2_t __ret_31; \
- __ret_31 = __noswap_vsetq_lane_s64(__noswap_vget_lane_s64(__s2_31, __p3_31), __rev0_31, __p1_31); \
- __ret_31 = __builtin_shufflevector(__ret_31, __ret_31, 1, 0); \
- __ret_31; \
+#define vcopyq_lane_s64(__p0_268, __p1_268, __p2_268, __p3_268) __extension__ ({ \
+ int64x2_t __s0_268 = __p0_268; \
+ int64x1_t __s2_268 = __p2_268; \
+ int64x2_t __rev0_268; __rev0_268 = __builtin_shufflevector(__s0_268, __s0_268, 1, 0); \
+ int64x2_t __ret_268; \
+ __ret_268 = __noswap_vsetq_lane_s64(vget_lane_s64(__s2_268, __p3_268), __rev0_268, __p1_268); \
+ __ret_268 = __builtin_shufflevector(__ret_268, __ret_268, 1, 0); \
+ __ret_268; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_lane_s16(__p0_32, __p1_32, __p2_32, __p3_32) __extension__ ({ \
- int16x8_t __s0_32 = __p0_32; \
- int16x4_t __s2_32 = __p2_32; \
- int16x8_t __ret_32; \
- __ret_32 = vsetq_lane_s16(vget_lane_s16(__s2_32, __p3_32), __s0_32, __p1_32); \
- __ret_32; \
+#define vcopyq_lane_s16(__p0_269, __p1_269, __p2_269, __p3_269) __extension__ ({ \
+ int16x8_t __s0_269 = __p0_269; \
+ int16x4_t __s2_269 = __p2_269; \
+ int16x8_t __ret_269; \
+ __ret_269 = vsetq_lane_s16(vget_lane_s16(__s2_269, __p3_269), __s0_269, __p1_269); \
+ __ret_269; \
})
#else
-#define vcopyq_lane_s16(__p0_33, __p1_33, __p2_33, __p3_33) __extension__ ({ \
- int16x8_t __s0_33 = __p0_33; \
- int16x4_t __s2_33 = __p2_33; \
- int16x8_t __rev0_33; __rev0_33 = __builtin_shufflevector(__s0_33, __s0_33, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __rev2_33; __rev2_33 = __builtin_shufflevector(__s2_33, __s2_33, 3, 2, 1, 0); \
- int16x8_t __ret_33; \
- __ret_33 = __noswap_vsetq_lane_s16(__noswap_vget_lane_s16(__rev2_33, __p3_33), __rev0_33, __p1_33); \
- __ret_33 = __builtin_shufflevector(__ret_33, __ret_33, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_33; \
+#define vcopyq_lane_s16(__p0_270, __p1_270, __p2_270, __p3_270) __extension__ ({ \
+ int16x8_t __s0_270 = __p0_270; \
+ int16x4_t __s2_270 = __p2_270; \
+ int16x8_t __rev0_270; __rev0_270 = __builtin_shufflevector(__s0_270, __s0_270, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev2_270; __rev2_270 = __builtin_shufflevector(__s2_270, __s2_270, 3, 2, 1, 0); \
+ int16x8_t __ret_270; \
+ __ret_270 = __noswap_vsetq_lane_s16(__noswap_vget_lane_s16(__rev2_270, __p3_270), __rev0_270, __p1_270); \
+ __ret_270 = __builtin_shufflevector(__ret_270, __ret_270, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_270; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_lane_p8(__p0_34, __p1_34, __p2_34, __p3_34) __extension__ ({ \
- poly8x8_t __s0_34 = __p0_34; \
- poly8x8_t __s2_34 = __p2_34; \
- poly8x8_t __ret_34; \
- __ret_34 = vset_lane_p8(vget_lane_p8(__s2_34, __p3_34), __s0_34, __p1_34); \
- __ret_34; \
+#define vcopy_lane_p8(__p0_271, __p1_271, __p2_271, __p3_271) __extension__ ({ \
+ poly8x8_t __s0_271 = __p0_271; \
+ poly8x8_t __s2_271 = __p2_271; \
+ poly8x8_t __ret_271; \
+ __ret_271 = vset_lane_p8(vget_lane_p8(__s2_271, __p3_271), __s0_271, __p1_271); \
+ __ret_271; \
})
#else
-#define vcopy_lane_p8(__p0_35, __p1_35, __p2_35, __p3_35) __extension__ ({ \
- poly8x8_t __s0_35 = __p0_35; \
- poly8x8_t __s2_35 = __p2_35; \
- poly8x8_t __rev0_35; __rev0_35 = __builtin_shufflevector(__s0_35, __s0_35, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly8x8_t __rev2_35; __rev2_35 = __builtin_shufflevector(__s2_35, __s2_35, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly8x8_t __ret_35; \
- __ret_35 = __noswap_vset_lane_p8(__noswap_vget_lane_p8(__rev2_35, __p3_35), __rev0_35, __p1_35); \
- __ret_35 = __builtin_shufflevector(__ret_35, __ret_35, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_35; \
+#define vcopy_lane_p8(__p0_272, __p1_272, __p2_272, __p3_272) __extension__ ({ \
+ poly8x8_t __s0_272 = __p0_272; \
+ poly8x8_t __s2_272 = __p2_272; \
+ poly8x8_t __rev0_272; __rev0_272 = __builtin_shufflevector(__s0_272, __s0_272, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x8_t __rev2_272; __rev2_272 = __builtin_shufflevector(__s2_272, __s2_272, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x8_t __ret_272; \
+ __ret_272 = __noswap_vset_lane_p8(__noswap_vget_lane_p8(__rev2_272, __p3_272), __rev0_272, __p1_272); \
+ __ret_272 = __builtin_shufflevector(__ret_272, __ret_272, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_272; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_lane_p16(__p0_36, __p1_36, __p2_36, __p3_36) __extension__ ({ \
- poly16x4_t __s0_36 = __p0_36; \
- poly16x4_t __s2_36 = __p2_36; \
- poly16x4_t __ret_36; \
- __ret_36 = vset_lane_p16(vget_lane_p16(__s2_36, __p3_36), __s0_36, __p1_36); \
- __ret_36; \
+#define vcopy_lane_p16(__p0_273, __p1_273, __p2_273, __p3_273) __extension__ ({ \
+ poly16x4_t __s0_273 = __p0_273; \
+ poly16x4_t __s2_273 = __p2_273; \
+ poly16x4_t __ret_273; \
+ __ret_273 = vset_lane_p16(vget_lane_p16(__s2_273, __p3_273), __s0_273, __p1_273); \
+ __ret_273; \
})
#else
-#define vcopy_lane_p16(__p0_37, __p1_37, __p2_37, __p3_37) __extension__ ({ \
- poly16x4_t __s0_37 = __p0_37; \
- poly16x4_t __s2_37 = __p2_37; \
- poly16x4_t __rev0_37; __rev0_37 = __builtin_shufflevector(__s0_37, __s0_37, 3, 2, 1, 0); \
- poly16x4_t __rev2_37; __rev2_37 = __builtin_shufflevector(__s2_37, __s2_37, 3, 2, 1, 0); \
- poly16x4_t __ret_37; \
- __ret_37 = __noswap_vset_lane_p16(__noswap_vget_lane_p16(__rev2_37, __p3_37), __rev0_37, __p1_37); \
- __ret_37 = __builtin_shufflevector(__ret_37, __ret_37, 3, 2, 1, 0); \
- __ret_37; \
+#define vcopy_lane_p16(__p0_274, __p1_274, __p2_274, __p3_274) __extension__ ({ \
+ poly16x4_t __s0_274 = __p0_274; \
+ poly16x4_t __s2_274 = __p2_274; \
+ poly16x4_t __rev0_274; __rev0_274 = __builtin_shufflevector(__s0_274, __s0_274, 3, 2, 1, 0); \
+ poly16x4_t __rev2_274; __rev2_274 = __builtin_shufflevector(__s2_274, __s2_274, 3, 2, 1, 0); \
+ poly16x4_t __ret_274; \
+ __ret_274 = __noswap_vset_lane_p16(__noswap_vget_lane_p16(__rev2_274, __p3_274), __rev0_274, __p1_274); \
+ __ret_274 = __builtin_shufflevector(__ret_274, __ret_274, 3, 2, 1, 0); \
+ __ret_274; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_lane_u8(__p0_38, __p1_38, __p2_38, __p3_38) __extension__ ({ \
- uint8x8_t __s0_38 = __p0_38; \
- uint8x8_t __s2_38 = __p2_38; \
- uint8x8_t __ret_38; \
- __ret_38 = vset_lane_u8(vget_lane_u8(__s2_38, __p3_38), __s0_38, __p1_38); \
- __ret_38; \
+#define vcopy_lane_u8(__p0_275, __p1_275, __p2_275, __p3_275) __extension__ ({ \
+ uint8x8_t __s0_275 = __p0_275; \
+ uint8x8_t __s2_275 = __p2_275; \
+ uint8x8_t __ret_275; \
+ __ret_275 = vset_lane_u8(vget_lane_u8(__s2_275, __p3_275), __s0_275, __p1_275); \
+ __ret_275; \
})
#else
-#define vcopy_lane_u8(__p0_39, __p1_39, __p2_39, __p3_39) __extension__ ({ \
- uint8x8_t __s0_39 = __p0_39; \
- uint8x8_t __s2_39 = __p2_39; \
- uint8x8_t __rev0_39; __rev0_39 = __builtin_shufflevector(__s0_39, __s0_39, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x8_t __rev2_39; __rev2_39 = __builtin_shufflevector(__s2_39, __s2_39, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x8_t __ret_39; \
- __ret_39 = __noswap_vset_lane_u8(__noswap_vget_lane_u8(__rev2_39, __p3_39), __rev0_39, __p1_39); \
- __ret_39 = __builtin_shufflevector(__ret_39, __ret_39, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_39; \
+#define vcopy_lane_u8(__p0_276, __p1_276, __p2_276, __p3_276) __extension__ ({ \
+ uint8x8_t __s0_276 = __p0_276; \
+ uint8x8_t __s2_276 = __p2_276; \
+ uint8x8_t __rev0_276; __rev0_276 = __builtin_shufflevector(__s0_276, __s0_276, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x8_t __rev2_276; __rev2_276 = __builtin_shufflevector(__s2_276, __s2_276, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x8_t __ret_276; \
+ __ret_276 = __noswap_vset_lane_u8(__noswap_vget_lane_u8(__rev2_276, __p3_276), __rev0_276, __p1_276); \
+ __ret_276 = __builtin_shufflevector(__ret_276, __ret_276, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_276; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_lane_u32(__p0_40, __p1_40, __p2_40, __p3_40) __extension__ ({ \
- uint32x2_t __s0_40 = __p0_40; \
- uint32x2_t __s2_40 = __p2_40; \
- uint32x2_t __ret_40; \
- __ret_40 = vset_lane_u32(vget_lane_u32(__s2_40, __p3_40), __s0_40, __p1_40); \
- __ret_40; \
+#define vcopy_lane_u32(__p0_277, __p1_277, __p2_277, __p3_277) __extension__ ({ \
+ uint32x2_t __s0_277 = __p0_277; \
+ uint32x2_t __s2_277 = __p2_277; \
+ uint32x2_t __ret_277; \
+ __ret_277 = vset_lane_u32(vget_lane_u32(__s2_277, __p3_277), __s0_277, __p1_277); \
+ __ret_277; \
})
#else
-#define vcopy_lane_u32(__p0_41, __p1_41, __p2_41, __p3_41) __extension__ ({ \
- uint32x2_t __s0_41 = __p0_41; \
- uint32x2_t __s2_41 = __p2_41; \
- uint32x2_t __rev0_41; __rev0_41 = __builtin_shufflevector(__s0_41, __s0_41, 1, 0); \
- uint32x2_t __rev2_41; __rev2_41 = __builtin_shufflevector(__s2_41, __s2_41, 1, 0); \
- uint32x2_t __ret_41; \
- __ret_41 = __noswap_vset_lane_u32(__noswap_vget_lane_u32(__rev2_41, __p3_41), __rev0_41, __p1_41); \
- __ret_41 = __builtin_shufflevector(__ret_41, __ret_41, 1, 0); \
- __ret_41; \
+#define vcopy_lane_u32(__p0_278, __p1_278, __p2_278, __p3_278) __extension__ ({ \
+ uint32x2_t __s0_278 = __p0_278; \
+ uint32x2_t __s2_278 = __p2_278; \
+ uint32x2_t __rev0_278; __rev0_278 = __builtin_shufflevector(__s0_278, __s0_278, 1, 0); \
+ uint32x2_t __rev2_278; __rev2_278 = __builtin_shufflevector(__s2_278, __s2_278, 1, 0); \
+ uint32x2_t __ret_278; \
+ __ret_278 = __noswap_vset_lane_u32(__noswap_vget_lane_u32(__rev2_278, __p3_278), __rev0_278, __p1_278); \
+ __ret_278 = __builtin_shufflevector(__ret_278, __ret_278, 1, 0); \
+ __ret_278; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vcopy_lane_u64(__p0_42, __p1_42, __p2_42, __p3_42) __extension__ ({ \
- uint64x1_t __s0_42 = __p0_42; \
- uint64x1_t __s2_42 = __p2_42; \
- uint64x1_t __ret_42; \
- __ret_42 = vset_lane_u64(vget_lane_u64(__s2_42, __p3_42), __s0_42, __p1_42); \
- __ret_42; \
-})
-#else
-#define vcopy_lane_u64(__p0_43, __p1_43, __p2_43, __p3_43) __extension__ ({ \
- uint64x1_t __s0_43 = __p0_43; \
- uint64x1_t __s2_43 = __p2_43; \
- uint64x1_t __ret_43; \
- __ret_43 = __noswap_vset_lane_u64(__noswap_vget_lane_u64(__s2_43, __p3_43), __s0_43, __p1_43); \
- __ret_43; \
+#define vcopy_lane_u64(__p0_279, __p1_279, __p2_279, __p3_279) __extension__ ({ \
+ uint64x1_t __s0_279 = __p0_279; \
+ uint64x1_t __s2_279 = __p2_279; \
+ uint64x1_t __ret_279; \
+ __ret_279 = vset_lane_u64(vget_lane_u64(__s2_279, __p3_279), __s0_279, __p1_279); \
+ __ret_279; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vcopy_lane_u16(__p0_44, __p1_44, __p2_44, __p3_44) __extension__ ({ \
- uint16x4_t __s0_44 = __p0_44; \
- uint16x4_t __s2_44 = __p2_44; \
- uint16x4_t __ret_44; \
- __ret_44 = vset_lane_u16(vget_lane_u16(__s2_44, __p3_44), __s0_44, __p1_44); \
- __ret_44; \
+#define vcopy_lane_u16(__p0_280, __p1_280, __p2_280, __p3_280) __extension__ ({ \
+ uint16x4_t __s0_280 = __p0_280; \
+ uint16x4_t __s2_280 = __p2_280; \
+ uint16x4_t __ret_280; \
+ __ret_280 = vset_lane_u16(vget_lane_u16(__s2_280, __p3_280), __s0_280, __p1_280); \
+ __ret_280; \
})
#else
-#define vcopy_lane_u16(__p0_45, __p1_45, __p2_45, __p3_45) __extension__ ({ \
- uint16x4_t __s0_45 = __p0_45; \
- uint16x4_t __s2_45 = __p2_45; \
- uint16x4_t __rev0_45; __rev0_45 = __builtin_shufflevector(__s0_45, __s0_45, 3, 2, 1, 0); \
- uint16x4_t __rev2_45; __rev2_45 = __builtin_shufflevector(__s2_45, __s2_45, 3, 2, 1, 0); \
- uint16x4_t __ret_45; \
- __ret_45 = __noswap_vset_lane_u16(__noswap_vget_lane_u16(__rev2_45, __p3_45), __rev0_45, __p1_45); \
- __ret_45 = __builtin_shufflevector(__ret_45, __ret_45, 3, 2, 1, 0); \
- __ret_45; \
+#define vcopy_lane_u16(__p0_281, __p1_281, __p2_281, __p3_281) __extension__ ({ \
+ uint16x4_t __s0_281 = __p0_281; \
+ uint16x4_t __s2_281 = __p2_281; \
+ uint16x4_t __rev0_281; __rev0_281 = __builtin_shufflevector(__s0_281, __s0_281, 3, 2, 1, 0); \
+ uint16x4_t __rev2_281; __rev2_281 = __builtin_shufflevector(__s2_281, __s2_281, 3, 2, 1, 0); \
+ uint16x4_t __ret_281; \
+ __ret_281 = __noswap_vset_lane_u16(__noswap_vget_lane_u16(__rev2_281, __p3_281), __rev0_281, __p1_281); \
+ __ret_281 = __builtin_shufflevector(__ret_281, __ret_281, 3, 2, 1, 0); \
+ __ret_281; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_lane_s8(__p0_46, __p1_46, __p2_46, __p3_46) __extension__ ({ \
- int8x8_t __s0_46 = __p0_46; \
- int8x8_t __s2_46 = __p2_46; \
- int8x8_t __ret_46; \
- __ret_46 = vset_lane_s8(vget_lane_s8(__s2_46, __p3_46), __s0_46, __p1_46); \
- __ret_46; \
+#define vcopy_lane_s8(__p0_282, __p1_282, __p2_282, __p3_282) __extension__ ({ \
+ int8x8_t __s0_282 = __p0_282; \
+ int8x8_t __s2_282 = __p2_282; \
+ int8x8_t __ret_282; \
+ __ret_282 = vset_lane_s8(vget_lane_s8(__s2_282, __p3_282), __s0_282, __p1_282); \
+ __ret_282; \
})
#else
-#define vcopy_lane_s8(__p0_47, __p1_47, __p2_47, __p3_47) __extension__ ({ \
- int8x8_t __s0_47 = __p0_47; \
- int8x8_t __s2_47 = __p2_47; \
- int8x8_t __rev0_47; __rev0_47 = __builtin_shufflevector(__s0_47, __s0_47, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x8_t __rev2_47; __rev2_47 = __builtin_shufflevector(__s2_47, __s2_47, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x8_t __ret_47; \
- __ret_47 = __noswap_vset_lane_s8(__noswap_vget_lane_s8(__rev2_47, __p3_47), __rev0_47, __p1_47); \
- __ret_47 = __builtin_shufflevector(__ret_47, __ret_47, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_47; \
+#define vcopy_lane_s8(__p0_283, __p1_283, __p2_283, __p3_283) __extension__ ({ \
+ int8x8_t __s0_283 = __p0_283; \
+ int8x8_t __s2_283 = __p2_283; \
+ int8x8_t __rev0_283; __rev0_283 = __builtin_shufflevector(__s0_283, __s0_283, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x8_t __rev2_283; __rev2_283 = __builtin_shufflevector(__s2_283, __s2_283, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x8_t __ret_283; \
+ __ret_283 = __noswap_vset_lane_s8(__noswap_vget_lane_s8(__rev2_283, __p3_283), __rev0_283, __p1_283); \
+ __ret_283 = __builtin_shufflevector(__ret_283, __ret_283, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_283; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_lane_f32(__p0_48, __p1_48, __p2_48, __p3_48) __extension__ ({ \
- float32x2_t __s0_48 = __p0_48; \
- float32x2_t __s2_48 = __p2_48; \
- float32x2_t __ret_48; \
- __ret_48 = vset_lane_f32(vget_lane_f32(__s2_48, __p3_48), __s0_48, __p1_48); \
- __ret_48; \
+#define vcopy_lane_f32(__p0_284, __p1_284, __p2_284, __p3_284) __extension__ ({ \
+ float32x2_t __s0_284 = __p0_284; \
+ float32x2_t __s2_284 = __p2_284; \
+ float32x2_t __ret_284; \
+ __ret_284 = vset_lane_f32(vget_lane_f32(__s2_284, __p3_284), __s0_284, __p1_284); \
+ __ret_284; \
})
#else
-#define vcopy_lane_f32(__p0_49, __p1_49, __p2_49, __p3_49) __extension__ ({ \
- float32x2_t __s0_49 = __p0_49; \
- float32x2_t __s2_49 = __p2_49; \
- float32x2_t __rev0_49; __rev0_49 = __builtin_shufflevector(__s0_49, __s0_49, 1, 0); \
- float32x2_t __rev2_49; __rev2_49 = __builtin_shufflevector(__s2_49, __s2_49, 1, 0); \
- float32x2_t __ret_49; \
- __ret_49 = __noswap_vset_lane_f32(__noswap_vget_lane_f32(__rev2_49, __p3_49), __rev0_49, __p1_49); \
- __ret_49 = __builtin_shufflevector(__ret_49, __ret_49, 1, 0); \
- __ret_49; \
+#define vcopy_lane_f32(__p0_285, __p1_285, __p2_285, __p3_285) __extension__ ({ \
+ float32x2_t __s0_285 = __p0_285; \
+ float32x2_t __s2_285 = __p2_285; \
+ float32x2_t __rev0_285; __rev0_285 = __builtin_shufflevector(__s0_285, __s0_285, 1, 0); \
+ float32x2_t __rev2_285; __rev2_285 = __builtin_shufflevector(__s2_285, __s2_285, 1, 0); \
+ float32x2_t __ret_285; \
+ __ret_285 = __noswap_vset_lane_f32(__noswap_vget_lane_f32(__rev2_285, __p3_285), __rev0_285, __p1_285); \
+ __ret_285 = __builtin_shufflevector(__ret_285, __ret_285, 1, 0); \
+ __ret_285; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_lane_s32(__p0_50, __p1_50, __p2_50, __p3_50) __extension__ ({ \
- int32x2_t __s0_50 = __p0_50; \
- int32x2_t __s2_50 = __p2_50; \
- int32x2_t __ret_50; \
- __ret_50 = vset_lane_s32(vget_lane_s32(__s2_50, __p3_50), __s0_50, __p1_50); \
- __ret_50; \
+#define vcopy_lane_s32(__p0_286, __p1_286, __p2_286, __p3_286) __extension__ ({ \
+ int32x2_t __s0_286 = __p0_286; \
+ int32x2_t __s2_286 = __p2_286; \
+ int32x2_t __ret_286; \
+ __ret_286 = vset_lane_s32(vget_lane_s32(__s2_286, __p3_286), __s0_286, __p1_286); \
+ __ret_286; \
})
#else
-#define vcopy_lane_s32(__p0_51, __p1_51, __p2_51, __p3_51) __extension__ ({ \
- int32x2_t __s0_51 = __p0_51; \
- int32x2_t __s2_51 = __p2_51; \
- int32x2_t __rev0_51; __rev0_51 = __builtin_shufflevector(__s0_51, __s0_51, 1, 0); \
- int32x2_t __rev2_51; __rev2_51 = __builtin_shufflevector(__s2_51, __s2_51, 1, 0); \
- int32x2_t __ret_51; \
- __ret_51 = __noswap_vset_lane_s32(__noswap_vget_lane_s32(__rev2_51, __p3_51), __rev0_51, __p1_51); \
- __ret_51 = __builtin_shufflevector(__ret_51, __ret_51, 1, 0); \
- __ret_51; \
+#define vcopy_lane_s32(__p0_287, __p1_287, __p2_287, __p3_287) __extension__ ({ \
+ int32x2_t __s0_287 = __p0_287; \
+ int32x2_t __s2_287 = __p2_287; \
+ int32x2_t __rev0_287; __rev0_287 = __builtin_shufflevector(__s0_287, __s0_287, 1, 0); \
+ int32x2_t __rev2_287; __rev2_287 = __builtin_shufflevector(__s2_287, __s2_287, 1, 0); \
+ int32x2_t __ret_287; \
+ __ret_287 = __noswap_vset_lane_s32(__noswap_vget_lane_s32(__rev2_287, __p3_287), __rev0_287, __p1_287); \
+ __ret_287 = __builtin_shufflevector(__ret_287, __ret_287, 1, 0); \
+ __ret_287; \
})
#endif
+#define vcopy_lane_s64(__p0_288, __p1_288, __p2_288, __p3_288) __extension__ ({ \
+ int64x1_t __s0_288 = __p0_288; \
+ int64x1_t __s2_288 = __p2_288; \
+ int64x1_t __ret_288; \
+ __ret_288 = vset_lane_s64(vget_lane_s64(__s2_288, __p3_288), __s0_288, __p1_288); \
+ __ret_288; \
+})
#ifdef __LITTLE_ENDIAN__
-#define vcopy_lane_s64(__p0_52, __p1_52, __p2_52, __p3_52) __extension__ ({ \
- int64x1_t __s0_52 = __p0_52; \
- int64x1_t __s2_52 = __p2_52; \
- int64x1_t __ret_52; \
- __ret_52 = vset_lane_s64(vget_lane_s64(__s2_52, __p3_52), __s0_52, __p1_52); \
- __ret_52; \
+#define vcopy_lane_s16(__p0_289, __p1_289, __p2_289, __p3_289) __extension__ ({ \
+ int16x4_t __s0_289 = __p0_289; \
+ int16x4_t __s2_289 = __p2_289; \
+ int16x4_t __ret_289; \
+ __ret_289 = vset_lane_s16(vget_lane_s16(__s2_289, __p3_289), __s0_289, __p1_289); \
+ __ret_289; \
})
#else
-#define vcopy_lane_s64(__p0_53, __p1_53, __p2_53, __p3_53) __extension__ ({ \
- int64x1_t __s0_53 = __p0_53; \
- int64x1_t __s2_53 = __p2_53; \
- int64x1_t __ret_53; \
- __ret_53 = __noswap_vset_lane_s64(__noswap_vget_lane_s64(__s2_53, __p3_53), __s0_53, __p1_53); \
- __ret_53; \
+#define vcopy_lane_s16(__p0_290, __p1_290, __p2_290, __p3_290) __extension__ ({ \
+ int16x4_t __s0_290 = __p0_290; \
+ int16x4_t __s2_290 = __p2_290; \
+ int16x4_t __rev0_290; __rev0_290 = __builtin_shufflevector(__s0_290, __s0_290, 3, 2, 1, 0); \
+ int16x4_t __rev2_290; __rev2_290 = __builtin_shufflevector(__s2_290, __s2_290, 3, 2, 1, 0); \
+ int16x4_t __ret_290; \
+ __ret_290 = __noswap_vset_lane_s16(__noswap_vget_lane_s16(__rev2_290, __p3_290), __rev0_290, __p1_290); \
+ __ret_290 = __builtin_shufflevector(__ret_290, __ret_290, 3, 2, 1, 0); \
+ __ret_290; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_lane_s16(__p0_54, __p1_54, __p2_54, __p3_54) __extension__ ({ \
- int16x4_t __s0_54 = __p0_54; \
- int16x4_t __s2_54 = __p2_54; \
- int16x4_t __ret_54; \
- __ret_54 = vset_lane_s16(vget_lane_s16(__s2_54, __p3_54), __s0_54, __p1_54); \
- __ret_54; \
+#define vcopyq_laneq_p8(__p0_291, __p1_291, __p2_291, __p3_291) __extension__ ({ \
+ poly8x16_t __s0_291 = __p0_291; \
+ poly8x16_t __s2_291 = __p2_291; \
+ poly8x16_t __ret_291; \
+ __ret_291 = vsetq_lane_p8(vgetq_lane_p8(__s2_291, __p3_291), __s0_291, __p1_291); \
+ __ret_291; \
})
#else
-#define vcopy_lane_s16(__p0_55, __p1_55, __p2_55, __p3_55) __extension__ ({ \
- int16x4_t __s0_55 = __p0_55; \
- int16x4_t __s2_55 = __p2_55; \
- int16x4_t __rev0_55; __rev0_55 = __builtin_shufflevector(__s0_55, __s0_55, 3, 2, 1, 0); \
- int16x4_t __rev2_55; __rev2_55 = __builtin_shufflevector(__s2_55, __s2_55, 3, 2, 1, 0); \
- int16x4_t __ret_55; \
- __ret_55 = __noswap_vset_lane_s16(__noswap_vget_lane_s16(__rev2_55, __p3_55), __rev0_55, __p1_55); \
- __ret_55 = __builtin_shufflevector(__ret_55, __ret_55, 3, 2, 1, 0); \
- __ret_55; \
+#define vcopyq_laneq_p8(__p0_292, __p1_292, __p2_292, __p3_292) __extension__ ({ \
+ poly8x16_t __s0_292 = __p0_292; \
+ poly8x16_t __s2_292 = __p2_292; \
+ poly8x16_t __rev0_292; __rev0_292 = __builtin_shufflevector(__s0_292, __s0_292, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x16_t __rev2_292; __rev2_292 = __builtin_shufflevector(__s2_292, __s2_292, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x16_t __ret_292; \
+ __ret_292 = __noswap_vsetq_lane_p8(__noswap_vgetq_lane_p8(__rev2_292, __p3_292), __rev0_292, __p1_292); \
+ __ret_292 = __builtin_shufflevector(__ret_292, __ret_292, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_292; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_laneq_p8(__p0_56, __p1_56, __p2_56, __p3_56) __extension__ ({ \
- poly8x16_t __s0_56 = __p0_56; \
- poly8x16_t __s2_56 = __p2_56; \
- poly8x16_t __ret_56; \
- __ret_56 = vsetq_lane_p8(vgetq_lane_p8(__s2_56, __p3_56), __s0_56, __p1_56); \
- __ret_56; \
+#define vcopyq_laneq_p16(__p0_293, __p1_293, __p2_293, __p3_293) __extension__ ({ \
+ poly16x8_t __s0_293 = __p0_293; \
+ poly16x8_t __s2_293 = __p2_293; \
+ poly16x8_t __ret_293; \
+ __ret_293 = vsetq_lane_p16(vgetq_lane_p16(__s2_293, __p3_293), __s0_293, __p1_293); \
+ __ret_293; \
})
#else
-#define vcopyq_laneq_p8(__p0_57, __p1_57, __p2_57, __p3_57) __extension__ ({ \
- poly8x16_t __s0_57 = __p0_57; \
- poly8x16_t __s2_57 = __p2_57; \
- poly8x16_t __rev0_57; __rev0_57 = __builtin_shufflevector(__s0_57, __s0_57, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly8x16_t __rev2_57; __rev2_57 = __builtin_shufflevector(__s2_57, __s2_57, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly8x16_t __ret_57; \
- __ret_57 = __noswap_vsetq_lane_p8(__noswap_vgetq_lane_p8(__rev2_57, __p3_57), __rev0_57, __p1_57); \
- __ret_57 = __builtin_shufflevector(__ret_57, __ret_57, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_57; \
+#define vcopyq_laneq_p16(__p0_294, __p1_294, __p2_294, __p3_294) __extension__ ({ \
+ poly16x8_t __s0_294 = __p0_294; \
+ poly16x8_t __s2_294 = __p2_294; \
+ poly16x8_t __rev0_294; __rev0_294 = __builtin_shufflevector(__s0_294, __s0_294, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly16x8_t __rev2_294; __rev2_294 = __builtin_shufflevector(__s2_294, __s2_294, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly16x8_t __ret_294; \
+ __ret_294 = __noswap_vsetq_lane_p16(__noswap_vgetq_lane_p16(__rev2_294, __p3_294), __rev0_294, __p1_294); \
+ __ret_294 = __builtin_shufflevector(__ret_294, __ret_294, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_294; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_laneq_p16(__p0_58, __p1_58, __p2_58, __p3_58) __extension__ ({ \
- poly16x8_t __s0_58 = __p0_58; \
- poly16x8_t __s2_58 = __p2_58; \
- poly16x8_t __ret_58; \
- __ret_58 = vsetq_lane_p16(vgetq_lane_p16(__s2_58, __p3_58), __s0_58, __p1_58); \
- __ret_58; \
+#define vcopyq_laneq_u8(__p0_295, __p1_295, __p2_295, __p3_295) __extension__ ({ \
+ uint8x16_t __s0_295 = __p0_295; \
+ uint8x16_t __s2_295 = __p2_295; \
+ uint8x16_t __ret_295; \
+ __ret_295 = vsetq_lane_u8(vgetq_lane_u8(__s2_295, __p3_295), __s0_295, __p1_295); \
+ __ret_295; \
})
#else
-#define vcopyq_laneq_p16(__p0_59, __p1_59, __p2_59, __p3_59) __extension__ ({ \
- poly16x8_t __s0_59 = __p0_59; \
- poly16x8_t __s2_59 = __p2_59; \
- poly16x8_t __rev0_59; __rev0_59 = __builtin_shufflevector(__s0_59, __s0_59, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly16x8_t __rev2_59; __rev2_59 = __builtin_shufflevector(__s2_59, __s2_59, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly16x8_t __ret_59; \
- __ret_59 = __noswap_vsetq_lane_p16(__noswap_vgetq_lane_p16(__rev2_59, __p3_59), __rev0_59, __p1_59); \
- __ret_59 = __builtin_shufflevector(__ret_59, __ret_59, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_59; \
+#define vcopyq_laneq_u8(__p0_296, __p1_296, __p2_296, __p3_296) __extension__ ({ \
+ uint8x16_t __s0_296 = __p0_296; \
+ uint8x16_t __s2_296 = __p2_296; \
+ uint8x16_t __rev0_296; __rev0_296 = __builtin_shufflevector(__s0_296, __s0_296, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __rev2_296; __rev2_296 = __builtin_shufflevector(__s2_296, __s2_296, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __ret_296; \
+ __ret_296 = __noswap_vsetq_lane_u8(__noswap_vgetq_lane_u8(__rev2_296, __p3_296), __rev0_296, __p1_296); \
+ __ret_296 = __builtin_shufflevector(__ret_296, __ret_296, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_296; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_laneq_u8(__p0_60, __p1_60, __p2_60, __p3_60) __extension__ ({ \
- uint8x16_t __s0_60 = __p0_60; \
- uint8x16_t __s2_60 = __p2_60; \
- uint8x16_t __ret_60; \
- __ret_60 = vsetq_lane_u8(vgetq_lane_u8(__s2_60, __p3_60), __s0_60, __p1_60); \
- __ret_60; \
+#define vcopyq_laneq_u32(__p0_297, __p1_297, __p2_297, __p3_297) __extension__ ({ \
+ uint32x4_t __s0_297 = __p0_297; \
+ uint32x4_t __s2_297 = __p2_297; \
+ uint32x4_t __ret_297; \
+ __ret_297 = vsetq_lane_u32(vgetq_lane_u32(__s2_297, __p3_297), __s0_297, __p1_297); \
+ __ret_297; \
})
#else
-#define vcopyq_laneq_u8(__p0_61, __p1_61, __p2_61, __p3_61) __extension__ ({ \
- uint8x16_t __s0_61 = __p0_61; \
- uint8x16_t __s2_61 = __p2_61; \
- uint8x16_t __rev0_61; __rev0_61 = __builtin_shufflevector(__s0_61, __s0_61, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x16_t __rev2_61; __rev2_61 = __builtin_shufflevector(__s2_61, __s2_61, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x16_t __ret_61; \
- __ret_61 = __noswap_vsetq_lane_u8(__noswap_vgetq_lane_u8(__rev2_61, __p3_61), __rev0_61, __p1_61); \
- __ret_61 = __builtin_shufflevector(__ret_61, __ret_61, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_61; \
+#define vcopyq_laneq_u32(__p0_298, __p1_298, __p2_298, __p3_298) __extension__ ({ \
+ uint32x4_t __s0_298 = __p0_298; \
+ uint32x4_t __s2_298 = __p2_298; \
+ uint32x4_t __rev0_298; __rev0_298 = __builtin_shufflevector(__s0_298, __s0_298, 3, 2, 1, 0); \
+ uint32x4_t __rev2_298; __rev2_298 = __builtin_shufflevector(__s2_298, __s2_298, 3, 2, 1, 0); \
+ uint32x4_t __ret_298; \
+ __ret_298 = __noswap_vsetq_lane_u32(__noswap_vgetq_lane_u32(__rev2_298, __p3_298), __rev0_298, __p1_298); \
+ __ret_298 = __builtin_shufflevector(__ret_298, __ret_298, 3, 2, 1, 0); \
+ __ret_298; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_laneq_u32(__p0_62, __p1_62, __p2_62, __p3_62) __extension__ ({ \
- uint32x4_t __s0_62 = __p0_62; \
- uint32x4_t __s2_62 = __p2_62; \
- uint32x4_t __ret_62; \
- __ret_62 = vsetq_lane_u32(vgetq_lane_u32(__s2_62, __p3_62), __s0_62, __p1_62); \
- __ret_62; \
+#define vcopyq_laneq_u64(__p0_299, __p1_299, __p2_299, __p3_299) __extension__ ({ \
+ uint64x2_t __s0_299 = __p0_299; \
+ uint64x2_t __s2_299 = __p2_299; \
+ uint64x2_t __ret_299; \
+ __ret_299 = vsetq_lane_u64(vgetq_lane_u64(__s2_299, __p3_299), __s0_299, __p1_299); \
+ __ret_299; \
})
#else
-#define vcopyq_laneq_u32(__p0_63, __p1_63, __p2_63, __p3_63) __extension__ ({ \
- uint32x4_t __s0_63 = __p0_63; \
- uint32x4_t __s2_63 = __p2_63; \
- uint32x4_t __rev0_63; __rev0_63 = __builtin_shufflevector(__s0_63, __s0_63, 3, 2, 1, 0); \
- uint32x4_t __rev2_63; __rev2_63 = __builtin_shufflevector(__s2_63, __s2_63, 3, 2, 1, 0); \
- uint32x4_t __ret_63; \
- __ret_63 = __noswap_vsetq_lane_u32(__noswap_vgetq_lane_u32(__rev2_63, __p3_63), __rev0_63, __p1_63); \
- __ret_63 = __builtin_shufflevector(__ret_63, __ret_63, 3, 2, 1, 0); \
- __ret_63; \
+#define vcopyq_laneq_u64(__p0_300, __p1_300, __p2_300, __p3_300) __extension__ ({ \
+ uint64x2_t __s0_300 = __p0_300; \
+ uint64x2_t __s2_300 = __p2_300; \
+ uint64x2_t __rev0_300; __rev0_300 = __builtin_shufflevector(__s0_300, __s0_300, 1, 0); \
+ uint64x2_t __rev2_300; __rev2_300 = __builtin_shufflevector(__s2_300, __s2_300, 1, 0); \
+ uint64x2_t __ret_300; \
+ __ret_300 = __noswap_vsetq_lane_u64(__noswap_vgetq_lane_u64(__rev2_300, __p3_300), __rev0_300, __p1_300); \
+ __ret_300 = __builtin_shufflevector(__ret_300, __ret_300, 1, 0); \
+ __ret_300; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_laneq_u64(__p0_64, __p1_64, __p2_64, __p3_64) __extension__ ({ \
- uint64x2_t __s0_64 = __p0_64; \
- uint64x2_t __s2_64 = __p2_64; \
- uint64x2_t __ret_64; \
- __ret_64 = vsetq_lane_u64(vgetq_lane_u64(__s2_64, __p3_64), __s0_64, __p1_64); \
- __ret_64; \
+#define vcopyq_laneq_u16(__p0_301, __p1_301, __p2_301, __p3_301) __extension__ ({ \
+ uint16x8_t __s0_301 = __p0_301; \
+ uint16x8_t __s2_301 = __p2_301; \
+ uint16x8_t __ret_301; \
+ __ret_301 = vsetq_lane_u16(vgetq_lane_u16(__s2_301, __p3_301), __s0_301, __p1_301); \
+ __ret_301; \
})
#else
-#define vcopyq_laneq_u64(__p0_65, __p1_65, __p2_65, __p3_65) __extension__ ({ \
- uint64x2_t __s0_65 = __p0_65; \
- uint64x2_t __s2_65 = __p2_65; \
- uint64x2_t __rev0_65; __rev0_65 = __builtin_shufflevector(__s0_65, __s0_65, 1, 0); \
- uint64x2_t __rev2_65; __rev2_65 = __builtin_shufflevector(__s2_65, __s2_65, 1, 0); \
- uint64x2_t __ret_65; \
- __ret_65 = __noswap_vsetq_lane_u64(__noswap_vgetq_lane_u64(__rev2_65, __p3_65), __rev0_65, __p1_65); \
- __ret_65 = __builtin_shufflevector(__ret_65, __ret_65, 1, 0); \
- __ret_65; \
+#define vcopyq_laneq_u16(__p0_302, __p1_302, __p2_302, __p3_302) __extension__ ({ \
+ uint16x8_t __s0_302 = __p0_302; \
+ uint16x8_t __s2_302 = __p2_302; \
+ uint16x8_t __rev0_302; __rev0_302 = __builtin_shufflevector(__s0_302, __s0_302, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __rev2_302; __rev2_302 = __builtin_shufflevector(__s2_302, __s2_302, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __ret_302; \
+ __ret_302 = __noswap_vsetq_lane_u16(__noswap_vgetq_lane_u16(__rev2_302, __p3_302), __rev0_302, __p1_302); \
+ __ret_302 = __builtin_shufflevector(__ret_302, __ret_302, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_302; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_laneq_u16(__p0_66, __p1_66, __p2_66, __p3_66) __extension__ ({ \
- uint16x8_t __s0_66 = __p0_66; \
- uint16x8_t __s2_66 = __p2_66; \
- uint16x8_t __ret_66; \
- __ret_66 = vsetq_lane_u16(vgetq_lane_u16(__s2_66, __p3_66), __s0_66, __p1_66); \
- __ret_66; \
+#define vcopyq_laneq_s8(__p0_303, __p1_303, __p2_303, __p3_303) __extension__ ({ \
+ int8x16_t __s0_303 = __p0_303; \
+ int8x16_t __s2_303 = __p2_303; \
+ int8x16_t __ret_303; \
+ __ret_303 = vsetq_lane_s8(vgetq_lane_s8(__s2_303, __p3_303), __s0_303, __p1_303); \
+ __ret_303; \
})
#else
-#define vcopyq_laneq_u16(__p0_67, __p1_67, __p2_67, __p3_67) __extension__ ({ \
- uint16x8_t __s0_67 = __p0_67; \
- uint16x8_t __s2_67 = __p2_67; \
- uint16x8_t __rev0_67; __rev0_67 = __builtin_shufflevector(__s0_67, __s0_67, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __rev2_67; __rev2_67 = __builtin_shufflevector(__s2_67, __s2_67, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __ret_67; \
- __ret_67 = __noswap_vsetq_lane_u16(__noswap_vgetq_lane_u16(__rev2_67, __p3_67), __rev0_67, __p1_67); \
- __ret_67 = __builtin_shufflevector(__ret_67, __ret_67, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_67; \
+#define vcopyq_laneq_s8(__p0_304, __p1_304, __p2_304, __p3_304) __extension__ ({ \
+ int8x16_t __s0_304 = __p0_304; \
+ int8x16_t __s2_304 = __p2_304; \
+ int8x16_t __rev0_304; __rev0_304 = __builtin_shufflevector(__s0_304, __s0_304, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __rev2_304; __rev2_304 = __builtin_shufflevector(__s2_304, __s2_304, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __ret_304; \
+ __ret_304 = __noswap_vsetq_lane_s8(__noswap_vgetq_lane_s8(__rev2_304, __p3_304), __rev0_304, __p1_304); \
+ __ret_304 = __builtin_shufflevector(__ret_304, __ret_304, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_304; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_laneq_s8(__p0_68, __p1_68, __p2_68, __p3_68) __extension__ ({ \
- int8x16_t __s0_68 = __p0_68; \
- int8x16_t __s2_68 = __p2_68; \
- int8x16_t __ret_68; \
- __ret_68 = vsetq_lane_s8(vgetq_lane_s8(__s2_68, __p3_68), __s0_68, __p1_68); \
- __ret_68; \
+#define vcopyq_laneq_f32(__p0_305, __p1_305, __p2_305, __p3_305) __extension__ ({ \
+ float32x4_t __s0_305 = __p0_305; \
+ float32x4_t __s2_305 = __p2_305; \
+ float32x4_t __ret_305; \
+ __ret_305 = vsetq_lane_f32(vgetq_lane_f32(__s2_305, __p3_305), __s0_305, __p1_305); \
+ __ret_305; \
})
#else
-#define vcopyq_laneq_s8(__p0_69, __p1_69, __p2_69, __p3_69) __extension__ ({ \
- int8x16_t __s0_69 = __p0_69; \
- int8x16_t __s2_69 = __p2_69; \
- int8x16_t __rev0_69; __rev0_69 = __builtin_shufflevector(__s0_69, __s0_69, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x16_t __rev2_69; __rev2_69 = __builtin_shufflevector(__s2_69, __s2_69, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x16_t __ret_69; \
- __ret_69 = __noswap_vsetq_lane_s8(__noswap_vgetq_lane_s8(__rev2_69, __p3_69), __rev0_69, __p1_69); \
- __ret_69 = __builtin_shufflevector(__ret_69, __ret_69, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_69; \
+#define vcopyq_laneq_f32(__p0_306, __p1_306, __p2_306, __p3_306) __extension__ ({ \
+ float32x4_t __s0_306 = __p0_306; \
+ float32x4_t __s2_306 = __p2_306; \
+ float32x4_t __rev0_306; __rev0_306 = __builtin_shufflevector(__s0_306, __s0_306, 3, 2, 1, 0); \
+ float32x4_t __rev2_306; __rev2_306 = __builtin_shufflevector(__s2_306, __s2_306, 3, 2, 1, 0); \
+ float32x4_t __ret_306; \
+ __ret_306 = __noswap_vsetq_lane_f32(__noswap_vgetq_lane_f32(__rev2_306, __p3_306), __rev0_306, __p1_306); \
+ __ret_306 = __builtin_shufflevector(__ret_306, __ret_306, 3, 2, 1, 0); \
+ __ret_306; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_laneq_f32(__p0_70, __p1_70, __p2_70, __p3_70) __extension__ ({ \
- float32x4_t __s0_70 = __p0_70; \
- float32x4_t __s2_70 = __p2_70; \
- float32x4_t __ret_70; \
- __ret_70 = vsetq_lane_f32(vgetq_lane_f32(__s2_70, __p3_70), __s0_70, __p1_70); \
- __ret_70; \
+#define vcopyq_laneq_s32(__p0_307, __p1_307, __p2_307, __p3_307) __extension__ ({ \
+ int32x4_t __s0_307 = __p0_307; \
+ int32x4_t __s2_307 = __p2_307; \
+ int32x4_t __ret_307; \
+ __ret_307 = vsetq_lane_s32(vgetq_lane_s32(__s2_307, __p3_307), __s0_307, __p1_307); \
+ __ret_307; \
})
#else
-#define vcopyq_laneq_f32(__p0_71, __p1_71, __p2_71, __p3_71) __extension__ ({ \
- float32x4_t __s0_71 = __p0_71; \
- float32x4_t __s2_71 = __p2_71; \
- float32x4_t __rev0_71; __rev0_71 = __builtin_shufflevector(__s0_71, __s0_71, 3, 2, 1, 0); \
- float32x4_t __rev2_71; __rev2_71 = __builtin_shufflevector(__s2_71, __s2_71, 3, 2, 1, 0); \
- float32x4_t __ret_71; \
- __ret_71 = __noswap_vsetq_lane_f32(__noswap_vgetq_lane_f32(__rev2_71, __p3_71), __rev0_71, __p1_71); \
- __ret_71 = __builtin_shufflevector(__ret_71, __ret_71, 3, 2, 1, 0); \
- __ret_71; \
+#define vcopyq_laneq_s32(__p0_308, __p1_308, __p2_308, __p3_308) __extension__ ({ \
+ int32x4_t __s0_308 = __p0_308; \
+ int32x4_t __s2_308 = __p2_308; \
+ int32x4_t __rev0_308; __rev0_308 = __builtin_shufflevector(__s0_308, __s0_308, 3, 2, 1, 0); \
+ int32x4_t __rev2_308; __rev2_308 = __builtin_shufflevector(__s2_308, __s2_308, 3, 2, 1, 0); \
+ int32x4_t __ret_308; \
+ __ret_308 = __noswap_vsetq_lane_s32(__noswap_vgetq_lane_s32(__rev2_308, __p3_308), __rev0_308, __p1_308); \
+ __ret_308 = __builtin_shufflevector(__ret_308, __ret_308, 3, 2, 1, 0); \
+ __ret_308; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_laneq_s32(__p0_72, __p1_72, __p2_72, __p3_72) __extension__ ({ \
- int32x4_t __s0_72 = __p0_72; \
- int32x4_t __s2_72 = __p2_72; \
- int32x4_t __ret_72; \
- __ret_72 = vsetq_lane_s32(vgetq_lane_s32(__s2_72, __p3_72), __s0_72, __p1_72); \
- __ret_72; \
+#define vcopyq_laneq_s64(__p0_309, __p1_309, __p2_309, __p3_309) __extension__ ({ \
+ int64x2_t __s0_309 = __p0_309; \
+ int64x2_t __s2_309 = __p2_309; \
+ int64x2_t __ret_309; \
+ __ret_309 = vsetq_lane_s64(vgetq_lane_s64(__s2_309, __p3_309), __s0_309, __p1_309); \
+ __ret_309; \
})
#else
-#define vcopyq_laneq_s32(__p0_73, __p1_73, __p2_73, __p3_73) __extension__ ({ \
- int32x4_t __s0_73 = __p0_73; \
- int32x4_t __s2_73 = __p2_73; \
- int32x4_t __rev0_73; __rev0_73 = __builtin_shufflevector(__s0_73, __s0_73, 3, 2, 1, 0); \
- int32x4_t __rev2_73; __rev2_73 = __builtin_shufflevector(__s2_73, __s2_73, 3, 2, 1, 0); \
- int32x4_t __ret_73; \
- __ret_73 = __noswap_vsetq_lane_s32(__noswap_vgetq_lane_s32(__rev2_73, __p3_73), __rev0_73, __p1_73); \
- __ret_73 = __builtin_shufflevector(__ret_73, __ret_73, 3, 2, 1, 0); \
- __ret_73; \
+#define vcopyq_laneq_s64(__p0_310, __p1_310, __p2_310, __p3_310) __extension__ ({ \
+ int64x2_t __s0_310 = __p0_310; \
+ int64x2_t __s2_310 = __p2_310; \
+ int64x2_t __rev0_310; __rev0_310 = __builtin_shufflevector(__s0_310, __s0_310, 1, 0); \
+ int64x2_t __rev2_310; __rev2_310 = __builtin_shufflevector(__s2_310, __s2_310, 1, 0); \
+ int64x2_t __ret_310; \
+ __ret_310 = __noswap_vsetq_lane_s64(__noswap_vgetq_lane_s64(__rev2_310, __p3_310), __rev0_310, __p1_310); \
+ __ret_310 = __builtin_shufflevector(__ret_310, __ret_310, 1, 0); \
+ __ret_310; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_laneq_s64(__p0_74, __p1_74, __p2_74, __p3_74) __extension__ ({ \
- int64x2_t __s0_74 = __p0_74; \
- int64x2_t __s2_74 = __p2_74; \
- int64x2_t __ret_74; \
- __ret_74 = vsetq_lane_s64(vgetq_lane_s64(__s2_74, __p3_74), __s0_74, __p1_74); \
- __ret_74; \
+#define vcopyq_laneq_s16(__p0_311, __p1_311, __p2_311, __p3_311) __extension__ ({ \
+ int16x8_t __s0_311 = __p0_311; \
+ int16x8_t __s2_311 = __p2_311; \
+ int16x8_t __ret_311; \
+ __ret_311 = vsetq_lane_s16(vgetq_lane_s16(__s2_311, __p3_311), __s0_311, __p1_311); \
+ __ret_311; \
})
#else
-#define vcopyq_laneq_s64(__p0_75, __p1_75, __p2_75, __p3_75) __extension__ ({ \
- int64x2_t __s0_75 = __p0_75; \
- int64x2_t __s2_75 = __p2_75; \
- int64x2_t __rev0_75; __rev0_75 = __builtin_shufflevector(__s0_75, __s0_75, 1, 0); \
- int64x2_t __rev2_75; __rev2_75 = __builtin_shufflevector(__s2_75, __s2_75, 1, 0); \
- int64x2_t __ret_75; \
- __ret_75 = __noswap_vsetq_lane_s64(__noswap_vgetq_lane_s64(__rev2_75, __p3_75), __rev0_75, __p1_75); \
- __ret_75 = __builtin_shufflevector(__ret_75, __ret_75, 1, 0); \
- __ret_75; \
+#define vcopyq_laneq_s16(__p0_312, __p1_312, __p2_312, __p3_312) __extension__ ({ \
+ int16x8_t __s0_312 = __p0_312; \
+ int16x8_t __s2_312 = __p2_312; \
+ int16x8_t __rev0_312; __rev0_312 = __builtin_shufflevector(__s0_312, __s0_312, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev2_312; __rev2_312 = __builtin_shufflevector(__s2_312, __s2_312, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __ret_312; \
+ __ret_312 = __noswap_vsetq_lane_s16(__noswap_vgetq_lane_s16(__rev2_312, __p3_312), __rev0_312, __p1_312); \
+ __ret_312 = __builtin_shufflevector(__ret_312, __ret_312, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_312; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_laneq_s16(__p0_76, __p1_76, __p2_76, __p3_76) __extension__ ({ \
- int16x8_t __s0_76 = __p0_76; \
- int16x8_t __s2_76 = __p2_76; \
- int16x8_t __ret_76; \
- __ret_76 = vsetq_lane_s16(vgetq_lane_s16(__s2_76, __p3_76), __s0_76, __p1_76); \
- __ret_76; \
+#define vcopy_laneq_p8(__p0_313, __p1_313, __p2_313, __p3_313) __extension__ ({ \
+ poly8x8_t __s0_313 = __p0_313; \
+ poly8x16_t __s2_313 = __p2_313; \
+ poly8x8_t __ret_313; \
+ __ret_313 = vset_lane_p8(vgetq_lane_p8(__s2_313, __p3_313), __s0_313, __p1_313); \
+ __ret_313; \
})
#else
-#define vcopyq_laneq_s16(__p0_77, __p1_77, __p2_77, __p3_77) __extension__ ({ \
- int16x8_t __s0_77 = __p0_77; \
- int16x8_t __s2_77 = __p2_77; \
- int16x8_t __rev0_77; __rev0_77 = __builtin_shufflevector(__s0_77, __s0_77, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev2_77; __rev2_77 = __builtin_shufflevector(__s2_77, __s2_77, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __ret_77; \
- __ret_77 = __noswap_vsetq_lane_s16(__noswap_vgetq_lane_s16(__rev2_77, __p3_77), __rev0_77, __p1_77); \
- __ret_77 = __builtin_shufflevector(__ret_77, __ret_77, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_77; \
+#define vcopy_laneq_p8(__p0_314, __p1_314, __p2_314, __p3_314) __extension__ ({ \
+ poly8x8_t __s0_314 = __p0_314; \
+ poly8x16_t __s2_314 = __p2_314; \
+ poly8x8_t __rev0_314; __rev0_314 = __builtin_shufflevector(__s0_314, __s0_314, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x16_t __rev2_314; __rev2_314 = __builtin_shufflevector(__s2_314, __s2_314, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x8_t __ret_314; \
+ __ret_314 = __noswap_vset_lane_p8(__noswap_vgetq_lane_p8(__rev2_314, __p3_314), __rev0_314, __p1_314); \
+ __ret_314 = __builtin_shufflevector(__ret_314, __ret_314, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_314; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_laneq_p8(__p0_78, __p1_78, __p2_78, __p3_78) __extension__ ({ \
- poly8x8_t __s0_78 = __p0_78; \
- poly8x16_t __s2_78 = __p2_78; \
- poly8x8_t __ret_78; \
- __ret_78 = vset_lane_p8(vgetq_lane_p8(__s2_78, __p3_78), __s0_78, __p1_78); \
- __ret_78; \
+#define vcopy_laneq_p16(__p0_315, __p1_315, __p2_315, __p3_315) __extension__ ({ \
+ poly16x4_t __s0_315 = __p0_315; \
+ poly16x8_t __s2_315 = __p2_315; \
+ poly16x4_t __ret_315; \
+ __ret_315 = vset_lane_p16(vgetq_lane_p16(__s2_315, __p3_315), __s0_315, __p1_315); \
+ __ret_315; \
})
#else
-#define vcopy_laneq_p8(__p0_79, __p1_79, __p2_79, __p3_79) __extension__ ({ \
- poly8x8_t __s0_79 = __p0_79; \
- poly8x16_t __s2_79 = __p2_79; \
- poly8x8_t __rev0_79; __rev0_79 = __builtin_shufflevector(__s0_79, __s0_79, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly8x16_t __rev2_79; __rev2_79 = __builtin_shufflevector(__s2_79, __s2_79, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly8x8_t __ret_79; \
- __ret_79 = __noswap_vset_lane_p8(__noswap_vgetq_lane_p8(__rev2_79, __p3_79), __rev0_79, __p1_79); \
- __ret_79 = __builtin_shufflevector(__ret_79, __ret_79, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_79; \
+#define vcopy_laneq_p16(__p0_316, __p1_316, __p2_316, __p3_316) __extension__ ({ \
+ poly16x4_t __s0_316 = __p0_316; \
+ poly16x8_t __s2_316 = __p2_316; \
+ poly16x4_t __rev0_316; __rev0_316 = __builtin_shufflevector(__s0_316, __s0_316, 3, 2, 1, 0); \
+ poly16x8_t __rev2_316; __rev2_316 = __builtin_shufflevector(__s2_316, __s2_316, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly16x4_t __ret_316; \
+ __ret_316 = __noswap_vset_lane_p16(__noswap_vgetq_lane_p16(__rev2_316, __p3_316), __rev0_316, __p1_316); \
+ __ret_316 = __builtin_shufflevector(__ret_316, __ret_316, 3, 2, 1, 0); \
+ __ret_316; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_laneq_p16(__p0_80, __p1_80, __p2_80, __p3_80) __extension__ ({ \
- poly16x4_t __s0_80 = __p0_80; \
- poly16x8_t __s2_80 = __p2_80; \
- poly16x4_t __ret_80; \
- __ret_80 = vset_lane_p16(vgetq_lane_p16(__s2_80, __p3_80), __s0_80, __p1_80); \
- __ret_80; \
+#define vcopy_laneq_u8(__p0_317, __p1_317, __p2_317, __p3_317) __extension__ ({ \
+ uint8x8_t __s0_317 = __p0_317; \
+ uint8x16_t __s2_317 = __p2_317; \
+ uint8x8_t __ret_317; \
+ __ret_317 = vset_lane_u8(vgetq_lane_u8(__s2_317, __p3_317), __s0_317, __p1_317); \
+ __ret_317; \
})
#else
-#define vcopy_laneq_p16(__p0_81, __p1_81, __p2_81, __p3_81) __extension__ ({ \
- poly16x4_t __s0_81 = __p0_81; \
- poly16x8_t __s2_81 = __p2_81; \
- poly16x4_t __rev0_81; __rev0_81 = __builtin_shufflevector(__s0_81, __s0_81, 3, 2, 1, 0); \
- poly16x8_t __rev2_81; __rev2_81 = __builtin_shufflevector(__s2_81, __s2_81, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly16x4_t __ret_81; \
- __ret_81 = __noswap_vset_lane_p16(__noswap_vgetq_lane_p16(__rev2_81, __p3_81), __rev0_81, __p1_81); \
- __ret_81 = __builtin_shufflevector(__ret_81, __ret_81, 3, 2, 1, 0); \
- __ret_81; \
+#define vcopy_laneq_u8(__p0_318, __p1_318, __p2_318, __p3_318) __extension__ ({ \
+ uint8x8_t __s0_318 = __p0_318; \
+ uint8x16_t __s2_318 = __p2_318; \
+ uint8x8_t __rev0_318; __rev0_318 = __builtin_shufflevector(__s0_318, __s0_318, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __rev2_318; __rev2_318 = __builtin_shufflevector(__s2_318, __s2_318, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x8_t __ret_318; \
+ __ret_318 = __noswap_vset_lane_u8(__noswap_vgetq_lane_u8(__rev2_318, __p3_318), __rev0_318, __p1_318); \
+ __ret_318 = __builtin_shufflevector(__ret_318, __ret_318, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_318; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_laneq_u8(__p0_82, __p1_82, __p2_82, __p3_82) __extension__ ({ \
- uint8x8_t __s0_82 = __p0_82; \
- uint8x16_t __s2_82 = __p2_82; \
- uint8x8_t __ret_82; \
- __ret_82 = vset_lane_u8(vgetq_lane_u8(__s2_82, __p3_82), __s0_82, __p1_82); \
- __ret_82; \
+#define vcopy_laneq_u32(__p0_319, __p1_319, __p2_319, __p3_319) __extension__ ({ \
+ uint32x2_t __s0_319 = __p0_319; \
+ uint32x4_t __s2_319 = __p2_319; \
+ uint32x2_t __ret_319; \
+ __ret_319 = vset_lane_u32(vgetq_lane_u32(__s2_319, __p3_319), __s0_319, __p1_319); \
+ __ret_319; \
})
#else
-#define vcopy_laneq_u8(__p0_83, __p1_83, __p2_83, __p3_83) __extension__ ({ \
- uint8x8_t __s0_83 = __p0_83; \
- uint8x16_t __s2_83 = __p2_83; \
- uint8x8_t __rev0_83; __rev0_83 = __builtin_shufflevector(__s0_83, __s0_83, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x16_t __rev2_83; __rev2_83 = __builtin_shufflevector(__s2_83, __s2_83, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x8_t __ret_83; \
- __ret_83 = __noswap_vset_lane_u8(__noswap_vgetq_lane_u8(__rev2_83, __p3_83), __rev0_83, __p1_83); \
- __ret_83 = __builtin_shufflevector(__ret_83, __ret_83, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_83; \
+#define vcopy_laneq_u32(__p0_320, __p1_320, __p2_320, __p3_320) __extension__ ({ \
+ uint32x2_t __s0_320 = __p0_320; \
+ uint32x4_t __s2_320 = __p2_320; \
+ uint32x2_t __rev0_320; __rev0_320 = __builtin_shufflevector(__s0_320, __s0_320, 1, 0); \
+ uint32x4_t __rev2_320; __rev2_320 = __builtin_shufflevector(__s2_320, __s2_320, 3, 2, 1, 0); \
+ uint32x2_t __ret_320; \
+ __ret_320 = __noswap_vset_lane_u32(__noswap_vgetq_lane_u32(__rev2_320, __p3_320), __rev0_320, __p1_320); \
+ __ret_320 = __builtin_shufflevector(__ret_320, __ret_320, 1, 0); \
+ __ret_320; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_laneq_u32(__p0_84, __p1_84, __p2_84, __p3_84) __extension__ ({ \
- uint32x2_t __s0_84 = __p0_84; \
- uint32x4_t __s2_84 = __p2_84; \
- uint32x2_t __ret_84; \
- __ret_84 = vset_lane_u32(vgetq_lane_u32(__s2_84, __p3_84), __s0_84, __p1_84); \
- __ret_84; \
+#define vcopy_laneq_u64(__p0_321, __p1_321, __p2_321, __p3_321) __extension__ ({ \
+ uint64x1_t __s0_321 = __p0_321; \
+ uint64x2_t __s2_321 = __p2_321; \
+ uint64x1_t __ret_321; \
+ __ret_321 = vset_lane_u64(vgetq_lane_u64(__s2_321, __p3_321), __s0_321, __p1_321); \
+ __ret_321; \
})
#else
-#define vcopy_laneq_u32(__p0_85, __p1_85, __p2_85, __p3_85) __extension__ ({ \
- uint32x2_t __s0_85 = __p0_85; \
- uint32x4_t __s2_85 = __p2_85; \
- uint32x2_t __rev0_85; __rev0_85 = __builtin_shufflevector(__s0_85, __s0_85, 1, 0); \
- uint32x4_t __rev2_85; __rev2_85 = __builtin_shufflevector(__s2_85, __s2_85, 3, 2, 1, 0); \
- uint32x2_t __ret_85; \
- __ret_85 = __noswap_vset_lane_u32(__noswap_vgetq_lane_u32(__rev2_85, __p3_85), __rev0_85, __p1_85); \
- __ret_85 = __builtin_shufflevector(__ret_85, __ret_85, 1, 0); \
- __ret_85; \
+#define vcopy_laneq_u64(__p0_322, __p1_322, __p2_322, __p3_322) __extension__ ({ \
+ uint64x1_t __s0_322 = __p0_322; \
+ uint64x2_t __s2_322 = __p2_322; \
+ uint64x2_t __rev2_322; __rev2_322 = __builtin_shufflevector(__s2_322, __s2_322, 1, 0); \
+ uint64x1_t __ret_322; \
+ __ret_322 = vset_lane_u64(__noswap_vgetq_lane_u64(__rev2_322, __p3_322), __s0_322, __p1_322); \
+ __ret_322; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_laneq_u64(__p0_86, __p1_86, __p2_86, __p3_86) __extension__ ({ \
- uint64x1_t __s0_86 = __p0_86; \
- uint64x2_t __s2_86 = __p2_86; \
- uint64x1_t __ret_86; \
- __ret_86 = vset_lane_u64(vgetq_lane_u64(__s2_86, __p3_86), __s0_86, __p1_86); \
- __ret_86; \
+#define vcopy_laneq_u16(__p0_323, __p1_323, __p2_323, __p3_323) __extension__ ({ \
+ uint16x4_t __s0_323 = __p0_323; \
+ uint16x8_t __s2_323 = __p2_323; \
+ uint16x4_t __ret_323; \
+ __ret_323 = vset_lane_u16(vgetq_lane_u16(__s2_323, __p3_323), __s0_323, __p1_323); \
+ __ret_323; \
})
#else
-#define vcopy_laneq_u64(__p0_87, __p1_87, __p2_87, __p3_87) __extension__ ({ \
- uint64x1_t __s0_87 = __p0_87; \
- uint64x2_t __s2_87 = __p2_87; \
- uint64x2_t __rev2_87; __rev2_87 = __builtin_shufflevector(__s2_87, __s2_87, 1, 0); \
- uint64x1_t __ret_87; \
- __ret_87 = __noswap_vset_lane_u64(__noswap_vgetq_lane_u64(__rev2_87, __p3_87), __s0_87, __p1_87); \
- __ret_87; \
+#define vcopy_laneq_u16(__p0_324, __p1_324, __p2_324, __p3_324) __extension__ ({ \
+ uint16x4_t __s0_324 = __p0_324; \
+ uint16x8_t __s2_324 = __p2_324; \
+ uint16x4_t __rev0_324; __rev0_324 = __builtin_shufflevector(__s0_324, __s0_324, 3, 2, 1, 0); \
+ uint16x8_t __rev2_324; __rev2_324 = __builtin_shufflevector(__s2_324, __s2_324, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x4_t __ret_324; \
+ __ret_324 = __noswap_vset_lane_u16(__noswap_vgetq_lane_u16(__rev2_324, __p3_324), __rev0_324, __p1_324); \
+ __ret_324 = __builtin_shufflevector(__ret_324, __ret_324, 3, 2, 1, 0); \
+ __ret_324; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_laneq_u16(__p0_88, __p1_88, __p2_88, __p3_88) __extension__ ({ \
- uint16x4_t __s0_88 = __p0_88; \
- uint16x8_t __s2_88 = __p2_88; \
- uint16x4_t __ret_88; \
- __ret_88 = vset_lane_u16(vgetq_lane_u16(__s2_88, __p3_88), __s0_88, __p1_88); \
- __ret_88; \
+#define vcopy_laneq_s8(__p0_325, __p1_325, __p2_325, __p3_325) __extension__ ({ \
+ int8x8_t __s0_325 = __p0_325; \
+ int8x16_t __s2_325 = __p2_325; \
+ int8x8_t __ret_325; \
+ __ret_325 = vset_lane_s8(vgetq_lane_s8(__s2_325, __p3_325), __s0_325, __p1_325); \
+ __ret_325; \
})
#else
-#define vcopy_laneq_u16(__p0_89, __p1_89, __p2_89, __p3_89) __extension__ ({ \
- uint16x4_t __s0_89 = __p0_89; \
- uint16x8_t __s2_89 = __p2_89; \
- uint16x4_t __rev0_89; __rev0_89 = __builtin_shufflevector(__s0_89, __s0_89, 3, 2, 1, 0); \
- uint16x8_t __rev2_89; __rev2_89 = __builtin_shufflevector(__s2_89, __s2_89, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x4_t __ret_89; \
- __ret_89 = __noswap_vset_lane_u16(__noswap_vgetq_lane_u16(__rev2_89, __p3_89), __rev0_89, __p1_89); \
- __ret_89 = __builtin_shufflevector(__ret_89, __ret_89, 3, 2, 1, 0); \
- __ret_89; \
+#define vcopy_laneq_s8(__p0_326, __p1_326, __p2_326, __p3_326) __extension__ ({ \
+ int8x8_t __s0_326 = __p0_326; \
+ int8x16_t __s2_326 = __p2_326; \
+ int8x8_t __rev0_326; __rev0_326 = __builtin_shufflevector(__s0_326, __s0_326, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __rev2_326; __rev2_326 = __builtin_shufflevector(__s2_326, __s2_326, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x8_t __ret_326; \
+ __ret_326 = __noswap_vset_lane_s8(__noswap_vgetq_lane_s8(__rev2_326, __p3_326), __rev0_326, __p1_326); \
+ __ret_326 = __builtin_shufflevector(__ret_326, __ret_326, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_326; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_laneq_s8(__p0_90, __p1_90, __p2_90, __p3_90) __extension__ ({ \
- int8x8_t __s0_90 = __p0_90; \
- int8x16_t __s2_90 = __p2_90; \
- int8x8_t __ret_90; \
- __ret_90 = vset_lane_s8(vgetq_lane_s8(__s2_90, __p3_90), __s0_90, __p1_90); \
- __ret_90; \
+#define vcopy_laneq_f32(__p0_327, __p1_327, __p2_327, __p3_327) __extension__ ({ \
+ float32x2_t __s0_327 = __p0_327; \
+ float32x4_t __s2_327 = __p2_327; \
+ float32x2_t __ret_327; \
+ __ret_327 = vset_lane_f32(vgetq_lane_f32(__s2_327, __p3_327), __s0_327, __p1_327); \
+ __ret_327; \
})
#else
-#define vcopy_laneq_s8(__p0_91, __p1_91, __p2_91, __p3_91) __extension__ ({ \
- int8x8_t __s0_91 = __p0_91; \
- int8x16_t __s2_91 = __p2_91; \
- int8x8_t __rev0_91; __rev0_91 = __builtin_shufflevector(__s0_91, __s0_91, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x16_t __rev2_91; __rev2_91 = __builtin_shufflevector(__s2_91, __s2_91, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x8_t __ret_91; \
- __ret_91 = __noswap_vset_lane_s8(__noswap_vgetq_lane_s8(__rev2_91, __p3_91), __rev0_91, __p1_91); \
- __ret_91 = __builtin_shufflevector(__ret_91, __ret_91, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_91; \
+#define vcopy_laneq_f32(__p0_328, __p1_328, __p2_328, __p3_328) __extension__ ({ \
+ float32x2_t __s0_328 = __p0_328; \
+ float32x4_t __s2_328 = __p2_328; \
+ float32x2_t __rev0_328; __rev0_328 = __builtin_shufflevector(__s0_328, __s0_328, 1, 0); \
+ float32x4_t __rev2_328; __rev2_328 = __builtin_shufflevector(__s2_328, __s2_328, 3, 2, 1, 0); \
+ float32x2_t __ret_328; \
+ __ret_328 = __noswap_vset_lane_f32(__noswap_vgetq_lane_f32(__rev2_328, __p3_328), __rev0_328, __p1_328); \
+ __ret_328 = __builtin_shufflevector(__ret_328, __ret_328, 1, 0); \
+ __ret_328; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_laneq_f32(__p0_92, __p1_92, __p2_92, __p3_92) __extension__ ({ \
- float32x2_t __s0_92 = __p0_92; \
- float32x4_t __s2_92 = __p2_92; \
- float32x2_t __ret_92; \
- __ret_92 = vset_lane_f32(vgetq_lane_f32(__s2_92, __p3_92), __s0_92, __p1_92); \
- __ret_92; \
+#define vcopy_laneq_s32(__p0_329, __p1_329, __p2_329, __p3_329) __extension__ ({ \
+ int32x2_t __s0_329 = __p0_329; \
+ int32x4_t __s2_329 = __p2_329; \
+ int32x2_t __ret_329; \
+ __ret_329 = vset_lane_s32(vgetq_lane_s32(__s2_329, __p3_329), __s0_329, __p1_329); \
+ __ret_329; \
})
#else
-#define vcopy_laneq_f32(__p0_93, __p1_93, __p2_93, __p3_93) __extension__ ({ \
- float32x2_t __s0_93 = __p0_93; \
- float32x4_t __s2_93 = __p2_93; \
- float32x2_t __rev0_93; __rev0_93 = __builtin_shufflevector(__s0_93, __s0_93, 1, 0); \
- float32x4_t __rev2_93; __rev2_93 = __builtin_shufflevector(__s2_93, __s2_93, 3, 2, 1, 0); \
- float32x2_t __ret_93; \
- __ret_93 = __noswap_vset_lane_f32(__noswap_vgetq_lane_f32(__rev2_93, __p3_93), __rev0_93, __p1_93); \
- __ret_93 = __builtin_shufflevector(__ret_93, __ret_93, 1, 0); \
- __ret_93; \
+#define vcopy_laneq_s32(__p0_330, __p1_330, __p2_330, __p3_330) __extension__ ({ \
+ int32x2_t __s0_330 = __p0_330; \
+ int32x4_t __s2_330 = __p2_330; \
+ int32x2_t __rev0_330; __rev0_330 = __builtin_shufflevector(__s0_330, __s0_330, 1, 0); \
+ int32x4_t __rev2_330; __rev2_330 = __builtin_shufflevector(__s2_330, __s2_330, 3, 2, 1, 0); \
+ int32x2_t __ret_330; \
+ __ret_330 = __noswap_vset_lane_s32(__noswap_vgetq_lane_s32(__rev2_330, __p3_330), __rev0_330, __p1_330); \
+ __ret_330 = __builtin_shufflevector(__ret_330, __ret_330, 1, 0); \
+ __ret_330; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_laneq_s32(__p0_94, __p1_94, __p2_94, __p3_94) __extension__ ({ \
- int32x2_t __s0_94 = __p0_94; \
- int32x4_t __s2_94 = __p2_94; \
- int32x2_t __ret_94; \
- __ret_94 = vset_lane_s32(vgetq_lane_s32(__s2_94, __p3_94), __s0_94, __p1_94); \
- __ret_94; \
+#define vcopy_laneq_s64(__p0_331, __p1_331, __p2_331, __p3_331) __extension__ ({ \
+ int64x1_t __s0_331 = __p0_331; \
+ int64x2_t __s2_331 = __p2_331; \
+ int64x1_t __ret_331; \
+ __ret_331 = vset_lane_s64(vgetq_lane_s64(__s2_331, __p3_331), __s0_331, __p1_331); \
+ __ret_331; \
})
#else
-#define vcopy_laneq_s32(__p0_95, __p1_95, __p2_95, __p3_95) __extension__ ({ \
- int32x2_t __s0_95 = __p0_95; \
- int32x4_t __s2_95 = __p2_95; \
- int32x2_t __rev0_95; __rev0_95 = __builtin_shufflevector(__s0_95, __s0_95, 1, 0); \
- int32x4_t __rev2_95; __rev2_95 = __builtin_shufflevector(__s2_95, __s2_95, 3, 2, 1, 0); \
- int32x2_t __ret_95; \
- __ret_95 = __noswap_vset_lane_s32(__noswap_vgetq_lane_s32(__rev2_95, __p3_95), __rev0_95, __p1_95); \
- __ret_95 = __builtin_shufflevector(__ret_95, __ret_95, 1, 0); \
- __ret_95; \
+#define vcopy_laneq_s64(__p0_332, __p1_332, __p2_332, __p3_332) __extension__ ({ \
+ int64x1_t __s0_332 = __p0_332; \
+ int64x2_t __s2_332 = __p2_332; \
+ int64x2_t __rev2_332; __rev2_332 = __builtin_shufflevector(__s2_332, __s2_332, 1, 0); \
+ int64x1_t __ret_332; \
+ __ret_332 = vset_lane_s64(__noswap_vgetq_lane_s64(__rev2_332, __p3_332), __s0_332, __p1_332); \
+ __ret_332; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_laneq_s64(__p0_96, __p1_96, __p2_96, __p3_96) __extension__ ({ \
- int64x1_t __s0_96 = __p0_96; \
- int64x2_t __s2_96 = __p2_96; \
- int64x1_t __ret_96; \
- __ret_96 = vset_lane_s64(vgetq_lane_s64(__s2_96, __p3_96), __s0_96, __p1_96); \
- __ret_96; \
+#define vcopy_laneq_s16(__p0_333, __p1_333, __p2_333, __p3_333) __extension__ ({ \
+ int16x4_t __s0_333 = __p0_333; \
+ int16x8_t __s2_333 = __p2_333; \
+ int16x4_t __ret_333; \
+ __ret_333 = vset_lane_s16(vgetq_lane_s16(__s2_333, __p3_333), __s0_333, __p1_333); \
+ __ret_333; \
})
#else
-#define vcopy_laneq_s64(__p0_97, __p1_97, __p2_97, __p3_97) __extension__ ({ \
- int64x1_t __s0_97 = __p0_97; \
- int64x2_t __s2_97 = __p2_97; \
- int64x2_t __rev2_97; __rev2_97 = __builtin_shufflevector(__s2_97, __s2_97, 1, 0); \
- int64x1_t __ret_97; \
- __ret_97 = __noswap_vset_lane_s64(__noswap_vgetq_lane_s64(__rev2_97, __p3_97), __s0_97, __p1_97); \
- __ret_97; \
+#define vcopy_laneq_s16(__p0_334, __p1_334, __p2_334, __p3_334) __extension__ ({ \
+ int16x4_t __s0_334 = __p0_334; \
+ int16x8_t __s2_334 = __p2_334; \
+ int16x4_t __rev0_334; __rev0_334 = __builtin_shufflevector(__s0_334, __s0_334, 3, 2, 1, 0); \
+ int16x8_t __rev2_334; __rev2_334 = __builtin_shufflevector(__s2_334, __s2_334, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __ret_334; \
+ __ret_334 = __noswap_vset_lane_s16(__noswap_vgetq_lane_s16(__rev2_334, __p3_334), __rev0_334, __p1_334); \
+ __ret_334 = __builtin_shufflevector(__ret_334, __ret_334, 3, 2, 1, 0); \
+ __ret_334; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vcopy_laneq_s16(__p0_98, __p1_98, __p2_98, __p3_98) __extension__ ({ \
- int16x4_t __s0_98 = __p0_98; \
- int16x8_t __s2_98 = __p2_98; \
- int16x4_t __ret_98; \
- __ret_98 = vset_lane_s16(vgetq_lane_s16(__s2_98, __p3_98), __s0_98, __p1_98); \
- __ret_98; \
+#define vcreate_p64(__p0) __extension__ ({ \
+ poly64x1_t __ret; \
+ uint64_t __promote = __p0; \
+ __ret = (poly64x1_t)(__promote); \
+ __ret; \
})
-#else
-#define vcopy_laneq_s16(__p0_99, __p1_99, __p2_99, __p3_99) __extension__ ({ \
- int16x4_t __s0_99 = __p0_99; \
- int16x8_t __s2_99 = __p2_99; \
- int16x4_t __rev0_99; __rev0_99 = __builtin_shufflevector(__s0_99, __s0_99, 3, 2, 1, 0); \
- int16x8_t __rev2_99; __rev2_99 = __builtin_shufflevector(__s2_99, __s2_99, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __ret_99; \
- __ret_99 = __noswap_vset_lane_s16(__noswap_vgetq_lane_s16(__rev2_99, __p3_99), __rev0_99, __p1_99); \
- __ret_99 = __builtin_shufflevector(__ret_99, __ret_99, 3, 2, 1, 0); \
- __ret_99; \
+#define vcreate_f64(__p0) __extension__ ({ \
+ float64x1_t __ret; \
+ uint64_t __promote = __p0; \
+ __ret = (float64x1_t)(__promote); \
+ __ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vcreate_p64(uint64_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t)(__p0);
- return __ret;
-}
-#else
-__ai poly64x1_t vcreate_p64(uint64_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vcreate_f64(uint64_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
-#else
-__ai float64x1_t vcreate_f64(uint64_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t)(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32_t vcvts_f32_s32(int32_t __p0) {
- float32_t __ret;
- __ret = (float32_t) __builtin_neon_vcvts_f32_s32(__p0);
- return __ret;
-}
-#else
__ai float32_t vcvts_f32_s32(int32_t __p0) {
float32_t __ret;
__ret = (float32_t) __builtin_neon_vcvts_f32_s32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32_t vcvts_f32_u32(uint32_t __p0) {
- float32_t __ret;
- __ret = (float32_t) __builtin_neon_vcvts_f32_u32(__p0);
- return __ret;
-}
-#else
__ai float32_t vcvts_f32_u32(uint32_t __p0) {
float32_t __ret;
__ret = (float32_t) __builtin_neon_vcvts_f32_u32(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float32x2_t vcvt_f32_f64(float64x2_t __p0) {
float32x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float64_t vcvtd_f64_s64(int64_t __p0) {
- float64_t __ret;
- __ret = (float64_t) __builtin_neon_vcvtd_f64_s64(__p0);
- return __ret;
-}
-#else
__ai float64_t vcvtd_f64_s64(int64_t __p0) {
float64_t __ret;
__ret = (float64_t) __builtin_neon_vcvtd_f64_s64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64_t vcvtd_f64_u64(uint64_t __p0) {
- float64_t __ret;
- __ret = (float64_t) __builtin_neon_vcvtd_f64_u64(__p0);
- return __ret;
-}
-#else
__ai float64_t vcvtd_f64_u64(uint64_t __p0) {
float64_t __ret;
__ret = (float64_t) __builtin_neon_vcvtd_f64_u64(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float64x2_t vcvtq_f64_u64(uint64x2_t __p0) {
float64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vcvt_f64_u64(uint64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vcvt_f64_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#else
__ai float64x1_t vcvt_f64_u64(uint64x1_t __p0) {
float64x1_t __ret;
__ret = (float64x1_t) __builtin_neon_vcvt_f64_v((int8x8_t)__p0, 19);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vcvt_f64_s64(int64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vcvt_f64_v((int8x8_t)__p0, 3);
- return __ret;
-}
-#else
__ai float64x1_t vcvt_f64_s64(int64x1_t __p0) {
float64x1_t __ret;
__ret = (float64x1_t) __builtin_neon_vcvt_f64_v((int8x8_t)__p0, 3);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float64x2_t vcvt_f64_f32(float32x2_t __p0) {
float64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vcvts_n_f32_u32(__p0, __p1) __extension__ ({ \
- uint32_t __s0 = __p0; \
- float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vcvts_n_f32_u32(__s0, __p1); \
- __ret; \
-})
-#else
#define vcvts_n_f32_u32(__p0, __p1) __extension__ ({ \
uint32_t __s0 = __p0; \
float32_t __ret; \
__ret = (float32_t) __builtin_neon_vcvts_n_f32_u32(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vcvts_n_f32_s32(__p0, __p1) __extension__ ({ \
- int32_t __s0 = __p0; \
- float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vcvts_n_f32_s32(__s0, __p1); \
- __ret; \
-})
-#else
#define vcvts_n_f32_s32(__p0, __p1) __extension__ ({ \
int32_t __s0 = __p0; \
float32_t __ret; \
__ret = (float32_t) __builtin_neon_vcvts_n_f32_s32(__s0, __p1); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vcvtq_n_f64_u64(__p0, __p1) __extension__ ({ \
uint64x2_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vcvt_n_f64_u64(__p0, __p1) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- float64x1_t __ret; \
- __ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 19); \
- __ret; \
-})
-#else
#define vcvt_n_f64_u64(__p0, __p1) __extension__ ({ \
uint64x1_t __s0 = __p0; \
float64x1_t __ret; \
__ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 19); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vcvt_n_f64_s64(__p0, __p1) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- float64x1_t __ret; \
- __ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 3); \
- __ret; \
-})
-#else
#define vcvt_n_f64_s64(__p0, __p1) __extension__ ({ \
int64x1_t __s0 = __p0; \
float64x1_t __ret; \
__ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 3); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vcvtd_n_f64_u64(__p0, __p1) __extension__ ({ \
- uint64_t __s0 = __p0; \
- float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vcvtd_n_f64_u64(__s0, __p1); \
- __ret; \
-})
-#else
#define vcvtd_n_f64_u64(__p0, __p1) __extension__ ({ \
uint64_t __s0 = __p0; \
float64_t __ret; \
__ret = (float64_t) __builtin_neon_vcvtd_n_f64_u64(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vcvtd_n_f64_s64(__p0, __p1) __extension__ ({ \
- int64_t __s0 = __p0; \
- float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vcvtd_n_f64_s64(__s0, __p1); \
- __ret; \
-})
-#else
#define vcvtd_n_f64_s64(__p0, __p1) __extension__ ({ \
int64_t __s0 = __p0; \
float64_t __ret; \
__ret = (float64_t) __builtin_neon_vcvtd_n_f64_s64(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vcvts_n_s32_f32(__p0, __p1) __extension__ ({ \
- float32_t __s0 = __p0; \
- int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vcvts_n_s32_f32(__s0, __p1); \
- __ret; \
-})
-#else
#define vcvts_n_s32_f32(__p0, __p1) __extension__ ({ \
float32_t __s0 = __p0; \
int32_t __ret; \
__ret = (int32_t) __builtin_neon_vcvts_n_s32_f32(__s0, __p1); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vcvtq_n_s64_f64(__p0, __p1) __extension__ ({ \
float64x2_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vcvt_n_s64_f64(__p0, __p1) __extension__ ({ \
- float64x1_t __s0 = __p0; \
- int64x1_t __ret; \
- __ret = (int64x1_t) __builtin_neon_vcvt_n_s64_v((int8x8_t)__s0, __p1, 3); \
- __ret; \
-})
-#else
#define vcvt_n_s64_f64(__p0, __p1) __extension__ ({ \
float64x1_t __s0 = __p0; \
int64x1_t __ret; \
__ret = (int64x1_t) __builtin_neon_vcvt_n_s64_v((int8x8_t)__s0, __p1, 3); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vcvtd_n_s64_f64(__p0, __p1) __extension__ ({ \
- float64_t __s0 = __p0; \
- int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vcvtd_n_s64_f64(__s0, __p1); \
- __ret; \
-})
-#else
#define vcvtd_n_s64_f64(__p0, __p1) __extension__ ({ \
float64_t __s0 = __p0; \
int64_t __ret; \
__ret = (int64_t) __builtin_neon_vcvtd_n_s64_f64(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vcvts_n_u32_f32(__p0, __p1) __extension__ ({ \
- float32_t __s0 = __p0; \
- uint32_t __ret; \
- __ret = (uint32_t) __builtin_neon_vcvts_n_u32_f32(__s0, __p1); \
- __ret; \
-})
-#else
#define vcvts_n_u32_f32(__p0, __p1) __extension__ ({ \
float32_t __s0 = __p0; \
uint32_t __ret; \
__ret = (uint32_t) __builtin_neon_vcvts_n_u32_f32(__s0, __p1); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vcvtq_n_u64_f64(__p0, __p1) __extension__ ({ \
float64x2_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vcvt_n_u64_f64(__p0, __p1) __extension__ ({ \
- float64x1_t __s0 = __p0; \
- uint64x1_t __ret; \
- __ret = (uint64x1_t) __builtin_neon_vcvt_n_u64_v((int8x8_t)__s0, __p1, 19); \
- __ret; \
-})
-#else
#define vcvt_n_u64_f64(__p0, __p1) __extension__ ({ \
float64x1_t __s0 = __p0; \
uint64x1_t __ret; \
__ret = (uint64x1_t) __builtin_neon_vcvt_n_u64_v((int8x8_t)__s0, __p1, 19); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vcvtd_n_u64_f64(__p0, __p1) __extension__ ({ \
- float64_t __s0 = __p0; \
- uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vcvtd_n_u64_f64(__s0, __p1); \
- __ret; \
-})
-#else
#define vcvtd_n_u64_f64(__p0, __p1) __extension__ ({ \
float64_t __s0 = __p0; \
uint64_t __ret; \
__ret = (uint64_t) __builtin_neon_vcvtd_n_u64_f64(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vcvts_s32_f32(float32_t __p0) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vcvts_s32_f32(__p0);
- return __ret;
-}
-#else
__ai int32_t vcvts_s32_f32(float32_t __p0) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vcvts_s32_f32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vcvtd_s64_f64(float64_t __p0) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vcvtd_s64_f64(__p0);
- return __ret;
-}
-#else
__ai int64_t vcvtd_s64_f64(float64_t __p0) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vcvtd_s64_f64(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int64x2_t vcvtq_s64_f64(float64x2_t __p0) {
int64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vcvt_s64_f64(float64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vcvt_s64_v((int8x8_t)__p0, 3);
- return __ret;
-}
-#else
__ai int64x1_t vcvt_s64_f64(float64x1_t __p0) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vcvt_s64_v((int8x8_t)__p0, 3);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vcvts_u32_f32(float32_t __p0) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vcvts_u32_f32(__p0);
- return __ret;
-}
-#else
__ai uint32_t vcvts_u32_f32(float32_t __p0) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vcvts_u32_f32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vcvtd_u64_f64(float64_t __p0) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcvtd_u64_f64(__p0);
- return __ret;
-}
-#else
__ai uint64_t vcvtd_u64_f64(float64_t __p0) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcvtd_u64_f64(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vcvtq_u64_f64(float64x2_t __p0) {
uint64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vcvt_u64_f64(float64x1_t __p0) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vcvt_u64_v((int8x8_t)__p0, 19);
return __ret;
}
-#else
-__ai uint64x1_t vcvt_u64_f64(float64x1_t __p0) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vcvt_u64_v((int8x8_t)__p0, 19);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vcvtas_s32_f32(float32_t __p0) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vcvtas_s32_f32(__p0);
- return __ret;
-}
-#else
__ai int32_t vcvtas_s32_f32(float32_t __p0) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vcvtas_s32_f32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vcvtad_s64_f64(float64_t __p0) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vcvtad_s64_f64(__p0);
- return __ret;
-}
-#else
__ai int64_t vcvtad_s64_f64(float64_t __p0) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vcvtad_s64_f64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vcvtas_u32_f32(float32_t __p0) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vcvtas_u32_f32(__p0);
- return __ret;
-}
-#else
__ai uint32_t vcvtas_u32_f32(float32_t __p0) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vcvtas_u32_f32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vcvtad_u64_f64(float64_t __p0) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcvtad_u64_f64(__p0);
- return __ret;
-}
-#else
__ai uint64_t vcvtad_u64_f64(float64_t __p0) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcvtad_u64_f64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vcvtms_s32_f32(float32_t __p0) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vcvtms_s32_f32(__p0);
- return __ret;
-}
-#else
__ai int32_t vcvtms_s32_f32(float32_t __p0) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vcvtms_s32_f32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vcvtmd_s64_f64(float64_t __p0) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vcvtmd_s64_f64(__p0);
- return __ret;
-}
-#else
__ai int64_t vcvtmd_s64_f64(float64_t __p0) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vcvtmd_s64_f64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vcvtms_u32_f32(float32_t __p0) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vcvtms_u32_f32(__p0);
- return __ret;
-}
-#else
__ai uint32_t vcvtms_u32_f32(float32_t __p0) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vcvtms_u32_f32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vcvtmd_u64_f64(float64_t __p0) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcvtmd_u64_f64(__p0);
- return __ret;
-}
-#else
__ai uint64_t vcvtmd_u64_f64(float64_t __p0) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcvtmd_u64_f64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vcvtns_s32_f32(float32_t __p0) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vcvtns_s32_f32(__p0);
- return __ret;
-}
-#else
__ai int32_t vcvtns_s32_f32(float32_t __p0) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vcvtns_s32_f32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vcvtnd_s64_f64(float64_t __p0) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vcvtnd_s64_f64(__p0);
- return __ret;
-}
-#else
__ai int64_t vcvtnd_s64_f64(float64_t __p0) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vcvtnd_s64_f64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vcvtns_u32_f32(float32_t __p0) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vcvtns_u32_f32(__p0);
- return __ret;
-}
-#else
__ai uint32_t vcvtns_u32_f32(float32_t __p0) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vcvtns_u32_f32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vcvtnd_u64_f64(float64_t __p0) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcvtnd_u64_f64(__p0);
- return __ret;
-}
-#else
__ai uint64_t vcvtnd_u64_f64(float64_t __p0) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcvtnd_u64_f64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vcvtps_s32_f32(float32_t __p0) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vcvtps_s32_f32(__p0);
- return __ret;
-}
-#else
__ai int32_t vcvtps_s32_f32(float32_t __p0) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vcvtps_s32_f32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vcvtpd_s64_f64(float64_t __p0) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vcvtpd_s64_f64(__p0);
- return __ret;
-}
-#else
__ai int64_t vcvtpd_s64_f64(float64_t __p0) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vcvtpd_s64_f64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vcvtps_u32_f32(float32_t __p0) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vcvtps_u32_f32(__p0);
- return __ret;
-}
-#else
__ai uint32_t vcvtps_u32_f32(float32_t __p0) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vcvtps_u32_f32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vcvtpd_u64_f64(float64_t __p0) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vcvtpd_u64_f64(__p0);
- return __ret;
-}
-#else
__ai uint64_t vcvtpd_u64_f64(float64_t __p0) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vcvtpd_u64_f64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32_t vcvtxd_f32_f64(float64_t __p0) {
- float32_t __ret;
- __ret = (float32_t) __builtin_neon_vcvtxd_f32_f64(__p0);
- return __ret;
-}
-#else
__ai float32_t vcvtxd_f32_f64(float64_t __p0) {
float32_t __ret;
__ret = (float32_t) __builtin_neon_vcvtxd_f32_f64(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float32x2_t vcvtx_f32_f64(float64x2_t __p0) {
float32x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vdiv_f64(float64x1_t __p0, float64x1_t __p1) {
- float64x1_t __ret;
- __ret = __p0 / __p1;
- return __ret;
-}
-#else
__ai float64x1_t vdiv_f64(float64x1_t __p0, float64x1_t __p1) {
float64x1_t __ret;
__ret = __p0 / __p1;
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
float32x2_t __ret;
#define vdupb_lane_p8(__p0, __p1) __extension__ ({ \
poly8x8_t __s0 = __p0; \
poly8_t __ret; \
- __ret = (poly8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__s0, __p1); \
+ __ret = (poly8_t) __builtin_neon_vdupb_lane_i8((poly8x8_t)__s0, __p1); \
__ret; \
})
#else
poly8x8_t __s0 = __p0; \
poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
poly8_t __ret; \
- __ret = (poly8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__rev0, __p1); \
+ __ret = (poly8_t) __builtin_neon_vdupb_lane_i8((poly8x8_t)__rev0, __p1); \
__ret; \
})
#endif
#define vduph_lane_p16(__p0, __p1) __extension__ ({ \
poly16x4_t __s0 = __p0; \
poly16_t __ret; \
- __ret = (poly16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__s0, __p1); \
+ __ret = (poly16_t) __builtin_neon_vduph_lane_i16((poly16x4_t)__s0, __p1); \
__ret; \
})
#else
poly16x4_t __s0 = __p0; \
poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
poly16_t __ret; \
- __ret = (poly16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__rev0, __p1); \
+ __ret = (poly16_t) __builtin_neon_vduph_lane_i16((poly16x4_t)__rev0, __p1); \
__ret; \
})
#endif
#define vdups_lane_u32(__p0, __p1) __extension__ ({ \
uint32x2_t __s0 = __p0; \
uint32_t __ret; \
- __ret = (uint32_t) __builtin_neon_vdups_lane_i32((int8x8_t)__s0, __p1); \
+ __ret = (uint32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__s0, __p1); \
__ret; \
})
#else
uint32x2_t __s0 = __p0; \
uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
uint32_t __ret; \
- __ret = (uint32_t) __builtin_neon_vdups_lane_i32((int8x8_t)__rev0, __p1); \
+ __ret = (uint32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__rev0, __p1); \
__ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vdupd_lane_u64(__p0, __p1) __extension__ ({ \
- uint64x1_t __s0 = __p0; \
- uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vdupd_lane_i64((int8x8_t)__s0, __p1); \
- __ret; \
-})
-#else
#define vdupd_lane_u64(__p0, __p1) __extension__ ({ \
uint64x1_t __s0 = __p0; \
uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vdupd_lane_i64((int8x8_t)__s0, __p1); \
+ __ret = (uint64_t) __builtin_neon_vdupd_lane_i64((int64x1_t)__s0, __p1); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vduph_lane_u16(__p0, __p1) __extension__ ({ \
uint16x4_t __s0 = __p0; \
uint16_t __ret; \
- __ret = (uint16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__s0, __p1); \
+ __ret = (uint16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__s0, __p1); \
__ret; \
})
#else
uint16x4_t __s0 = __p0; \
uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
uint16_t __ret; \
- __ret = (uint16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__rev0, __p1); \
+ __ret = (uint16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__rev0, __p1); \
__ret; \
})
#endif
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vdupd_lane_f64(__p0, __p1) __extension__ ({ \
- float64x1_t __s0 = __p0; \
- float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vdupd_lane_f64((int8x8_t)__s0, __p1); \
- __ret; \
-})
-#else
#define vdupd_lane_f64(__p0, __p1) __extension__ ({ \
float64x1_t __s0 = __p0; \
float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vdupd_lane_f64((int8x8_t)__s0, __p1); \
+ __ret = (float64_t) __builtin_neon_vdupd_lane_f64((float64x1_t)__s0, __p1); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vdups_lane_f32(__p0, __p1) __extension__ ({ \
float32x2_t __s0 = __p0; \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vdups_lane_f32((int8x8_t)__s0, __p1); \
+ __ret = (float32_t) __builtin_neon_vdups_lane_f32((float32x2_t)__s0, __p1); \
__ret; \
})
#else
float32x2_t __s0 = __p0; \
float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vdups_lane_f32((int8x8_t)__rev0, __p1); \
+ __ret = (float32_t) __builtin_neon_vdups_lane_f32((float32x2_t)__rev0, __p1); \
__ret; \
})
#endif
#define vdups_lane_s32(__p0, __p1) __extension__ ({ \
int32x2_t __s0 = __p0; \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vdups_lane_i32((int8x8_t)__s0, __p1); \
+ __ret = (int32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__s0, __p1); \
__ret; \
})
#else
int32x2_t __s0 = __p0; \
int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vdups_lane_i32((int8x8_t)__rev0, __p1); \
+ __ret = (int32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__rev0, __p1); \
__ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vdupd_lane_s64(__p0, __p1) __extension__ ({ \
- int64x1_t __s0 = __p0; \
- int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vdupd_lane_i64((int8x8_t)__s0, __p1); \
- __ret; \
-})
-#else
#define vdupd_lane_s64(__p0, __p1) __extension__ ({ \
int64x1_t __s0 = __p0; \
int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vdupd_lane_i64((int8x8_t)__s0, __p1); \
+ __ret = (int64_t) __builtin_neon_vdupd_lane_i64((int64x1_t)__s0, __p1); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vduph_lane_s16(__p0, __p1) __extension__ ({ \
int16x4_t __s0 = __p0; \
int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__s0, __p1); \
+ __ret = (int16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__s0, __p1); \
__ret; \
})
#else
int16x4_t __s0 = __p0; \
int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__rev0, __p1); \
+ __ret = (int16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__rev0, __p1); \
__ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vdup_lane_p64(__p0, __p1) __extension__ ({ \
- poly64x1_t __s0 = __p0; \
- poly64x1_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1); \
- __ret; \
-})
-#else
-#define vdup_lane_p64(__p0, __p1) __extension__ ({ \
- poly64x1_t __s0 = __p0; \
- poly64x1_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1); \
- __ret; \
+#define vdup_lane_p64(__p0_335, __p1_335) __extension__ ({ \
+ poly64x1_t __s0_335 = __p0_335; \
+ poly64x1_t __ret_335; \
+ __ret_335 = splat_lane_p64(__s0_335, __p1_335); \
+ __ret_335; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vdupq_lane_p64(__p0, __p1) __extension__ ({ \
- poly64x1_t __s0 = __p0; \
- poly64x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret; \
+#define vdupq_lane_p64(__p0_336, __p1_336) __extension__ ({ \
+ poly64x1_t __s0_336 = __p0_336; \
+ poly64x2_t __ret_336; \
+ __ret_336 = splatq_lane_p64(__s0_336, __p1_336); \
+ __ret_336; \
})
#else
-#define vdupq_lane_p64(__p0, __p1) __extension__ ({ \
- poly64x1_t __s0 = __p0; \
- poly64x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdupq_lane_p64(__p0_337, __p1_337) __extension__ ({ \
+ poly64x1_t __s0_337 = __p0_337; \
+ poly64x2_t __ret_337; \
+ __ret_337 = __noswap_splatq_lane_p64(__s0_337, __p1_337); \
+ __ret_337 = __builtin_shufflevector(__ret_337, __ret_337, 1, 0); \
+ __ret_337; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_lane_f64(__p0, __p1) __extension__ ({ \
- float64x1_t __s0 = __p0; \
- float64x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret; \
+#define vdupq_lane_f64(__p0_338, __p1_338) __extension__ ({ \
+ float64x1_t __s0_338 = __p0_338; \
+ float64x2_t __ret_338; \
+ __ret_338 = splatq_lane_f64(__s0_338, __p1_338); \
+ __ret_338; \
})
#else
-#define vdupq_lane_f64(__p0, __p1) __extension__ ({ \
- float64x1_t __s0 = __p0; \
- float64x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdupq_lane_f64(__p0_339, __p1_339) __extension__ ({ \
+ float64x1_t __s0_339 = __p0_339; \
+ float64x2_t __ret_339; \
+ __ret_339 = __noswap_splatq_lane_f64(__s0_339, __p1_339); \
+ __ret_339 = __builtin_shufflevector(__ret_339, __ret_339, 1, 0); \
+ __ret_339; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_lane_f16(__p0, __p1) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x8_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_lane_f16(__p0_340, __p1_340) __extension__ ({ \
+ float16x4_t __s0_340 = __p0_340; \
+ float16x8_t __ret_340; \
+ __ret_340 = splatq_lane_f16(__s0_340, __p1_340); \
+ __ret_340; \
})
#else
-#define vdupq_lane_f16(__p0, __p1) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float16x8_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_lane_f16(__p0_341, __p1_341) __extension__ ({ \
+ float16x4_t __s0_341 = __p0_341; \
+ float16x4_t __rev0_341; __rev0_341 = __builtin_shufflevector(__s0_341, __s0_341, 3, 2, 1, 0); \
+ float16x8_t __ret_341; \
+ __ret_341 = __noswap_splatq_lane_f16(__rev0_341, __p1_341); \
+ __ret_341 = __builtin_shufflevector(__ret_341, __ret_341, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_341; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vdup_lane_f64(__p0, __p1) __extension__ ({ \
- float64x1_t __s0 = __p0; \
- float64x1_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1); \
- __ret; \
-})
-#else
-#define vdup_lane_f64(__p0, __p1) __extension__ ({ \
- float64x1_t __s0 = __p0; \
- float64x1_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1); \
- __ret; \
+#define vdup_lane_f64(__p0_342, __p1_342) __extension__ ({ \
+ float64x1_t __s0_342 = __p0_342; \
+ float64x1_t __ret_342; \
+ __ret_342 = splat_lane_f64(__s0_342, __p1_342); \
+ __ret_342; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vdup_lane_f16(__p0, __p1) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x4_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdup_lane_f16(__p0_343, __p1_343) __extension__ ({ \
+ float16x4_t __s0_343 = __p0_343; \
+ float16x4_t __ret_343; \
+ __ret_343 = splat_lane_f16(__s0_343, __p1_343); \
+ __ret_343; \
})
#else
-#define vdup_lane_f16(__p0, __p1) __extension__ ({ \
- float16x4_t __s0 = __p0; \
- float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float16x4_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdup_lane_f16(__p0_344, __p1_344) __extension__ ({ \
+ float16x4_t __s0_344 = __p0_344; \
+ float16x4_t __rev0_344; __rev0_344 = __builtin_shufflevector(__s0_344, __s0_344, 3, 2, 1, 0); \
+ float16x4_t __ret_344; \
+ __ret_344 = __noswap_splat_lane_f16(__rev0_344, __p1_344); \
+ __ret_344 = __builtin_shufflevector(__ret_344, __ret_344, 3, 2, 1, 0); \
+ __ret_344; \
})
#endif
#define vdupb_laneq_p8(__p0, __p1) __extension__ ({ \
poly8x16_t __s0 = __p0; \
poly8_t __ret; \
- __ret = (poly8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__s0, __p1); \
+ __ret = (poly8_t) __builtin_neon_vdupb_laneq_i8((poly8x16_t)__s0, __p1); \
__ret; \
})
#else
poly8x16_t __s0 = __p0; \
poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
poly8_t __ret; \
- __ret = (poly8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__rev0, __p1); \
+ __ret = (poly8_t) __builtin_neon_vdupb_laneq_i8((poly8x16_t)__rev0, __p1); \
__ret; \
})
#endif
#define vduph_laneq_p16(__p0, __p1) __extension__ ({ \
poly16x8_t __s0 = __p0; \
poly16_t __ret; \
- __ret = (poly16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__s0, __p1); \
+ __ret = (poly16_t) __builtin_neon_vduph_laneq_i16((poly16x8_t)__s0, __p1); \
__ret; \
})
#else
poly16x8_t __s0 = __p0; \
poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
poly16_t __ret; \
- __ret = (poly16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__rev0, __p1); \
+ __ret = (poly16_t) __builtin_neon_vduph_laneq_i16((poly16x8_t)__rev0, __p1); \
__ret; \
})
#endif
#define vdups_laneq_u32(__p0, __p1) __extension__ ({ \
uint32x4_t __s0 = __p0; \
uint32_t __ret; \
- __ret = (uint32_t) __builtin_neon_vdups_laneq_i32((int8x16_t)__s0, __p1); \
+ __ret = (uint32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__s0, __p1); \
__ret; \
})
#else
uint32x4_t __s0 = __p0; \
uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
uint32_t __ret; \
- __ret = (uint32_t) __builtin_neon_vdups_laneq_i32((int8x16_t)__rev0, __p1); \
+ __ret = (uint32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__rev0, __p1); \
__ret; \
})
#endif
#define vdupd_laneq_u64(__p0, __p1) __extension__ ({ \
uint64x2_t __s0 = __p0; \
uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vdupd_laneq_i64((int8x16_t)__s0, __p1); \
+ __ret = (uint64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__s0, __p1); \
__ret; \
})
#else
uint64x2_t __s0 = __p0; \
uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vdupd_laneq_i64((int8x16_t)__rev0, __p1); \
+ __ret = (uint64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__rev0, __p1); \
__ret; \
})
#endif
#define vduph_laneq_u16(__p0, __p1) __extension__ ({ \
uint16x8_t __s0 = __p0; \
uint16_t __ret; \
- __ret = (uint16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__s0, __p1); \
+ __ret = (uint16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__s0, __p1); \
__ret; \
})
#else
uint16x8_t __s0 = __p0; \
uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
uint16_t __ret; \
- __ret = (uint16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__rev0, __p1); \
+ __ret = (uint16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__rev0, __p1); \
__ret; \
})
#endif
#define vdupd_laneq_f64(__p0, __p1) __extension__ ({ \
float64x2_t __s0 = __p0; \
float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vdupd_laneq_f64((int8x16_t)__s0, __p1); \
+ __ret = (float64_t) __builtin_neon_vdupd_laneq_f64((float64x2_t)__s0, __p1); \
__ret; \
})
#else
float64x2_t __s0 = __p0; \
float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vdupd_laneq_f64((int8x16_t)__rev0, __p1); \
+ __ret = (float64_t) __builtin_neon_vdupd_laneq_f64((float64x2_t)__rev0, __p1); \
__ret; \
})
#endif
#define vdups_laneq_f32(__p0, __p1) __extension__ ({ \
float32x4_t __s0 = __p0; \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vdups_laneq_f32((int8x16_t)__s0, __p1); \
+ __ret = (float32_t) __builtin_neon_vdups_laneq_f32((float32x4_t)__s0, __p1); \
__ret; \
})
#else
float32x4_t __s0 = __p0; \
float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vdups_laneq_f32((int8x16_t)__rev0, __p1); \
+ __ret = (float32_t) __builtin_neon_vdups_laneq_f32((float32x4_t)__rev0, __p1); \
__ret; \
})
#endif
#define vdups_laneq_s32(__p0, __p1) __extension__ ({ \
int32x4_t __s0 = __p0; \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vdups_laneq_i32((int8x16_t)__s0, __p1); \
+ __ret = (int32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__s0, __p1); \
__ret; \
})
#else
int32x4_t __s0 = __p0; \
int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vdups_laneq_i32((int8x16_t)__rev0, __p1); \
+ __ret = (int32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__rev0, __p1); \
__ret; \
})
#endif
#define vdupd_laneq_s64(__p0, __p1) __extension__ ({ \
int64x2_t __s0 = __p0; \
int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vdupd_laneq_i64((int8x16_t)__s0, __p1); \
+ __ret = (int64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__s0, __p1); \
__ret; \
})
#else
int64x2_t __s0 = __p0; \
int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vdupd_laneq_i64((int8x16_t)__rev0, __p1); \
+ __ret = (int64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__rev0, __p1); \
__ret; \
})
#endif
#define vduph_laneq_s16(__p0, __p1) __extension__ ({ \
int16x8_t __s0 = __p0; \
int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__s0, __p1); \
+ __ret = (int16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__s0, __p1); \
__ret; \
})
#else
int16x8_t __s0 = __p0; \
int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__rev0, __p1); \
+ __ret = (int16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__rev0, __p1); \
__ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_laneq_p8(__p0, __p1) __extension__ ({ \
- poly8x16_t __s0 = __p0; \
- poly8x8_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdup_laneq_p8(__p0_345, __p1_345) __extension__ ({ \
+ poly8x16_t __s0_345 = __p0_345; \
+ poly8x8_t __ret_345; \
+ __ret_345 = splat_laneq_p8(__s0_345, __p1_345); \
+ __ret_345; \
})
#else
-#define vdup_laneq_p8(__p0, __p1) __extension__ ({ \
- poly8x16_t __s0 = __p0; \
- poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly8x8_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdup_laneq_p8(__p0_346, __p1_346) __extension__ ({ \
+ poly8x16_t __s0_346 = __p0_346; \
+ poly8x16_t __rev0_346; __rev0_346 = __builtin_shufflevector(__s0_346, __s0_346, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x8_t __ret_346; \
+ __ret_346 = __noswap_splat_laneq_p8(__rev0_346, __p1_346); \
+ __ret_346 = __builtin_shufflevector(__ret_346, __ret_346, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_346; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_laneq_p64(__p0, __p1) __extension__ ({ \
- poly64x2_t __s0 = __p0; \
- poly64x1_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1); \
- __ret; \
+#define vdup_laneq_p64(__p0_347, __p1_347) __extension__ ({ \
+ poly64x2_t __s0_347 = __p0_347; \
+ poly64x1_t __ret_347; \
+ __ret_347 = splat_laneq_p64(__s0_347, __p1_347); \
+ __ret_347; \
})
#else
-#define vdup_laneq_p64(__p0, __p1) __extension__ ({ \
- poly64x2_t __s0 = __p0; \
- poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- poly64x1_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1); \
- __ret; \
+#define vdup_laneq_p64(__p0_348, __p1_348) __extension__ ({ \
+ poly64x2_t __s0_348 = __p0_348; \
+ poly64x2_t __rev0_348; __rev0_348 = __builtin_shufflevector(__s0_348, __s0_348, 1, 0); \
+ poly64x1_t __ret_348; \
+ __ret_348 = __noswap_splat_laneq_p64(__rev0_348, __p1_348); \
+ __ret_348; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_laneq_p16(__p0, __p1) __extension__ ({ \
- poly16x8_t __s0 = __p0; \
- poly16x4_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdup_laneq_p16(__p0_349, __p1_349) __extension__ ({ \
+ poly16x8_t __s0_349 = __p0_349; \
+ poly16x4_t __ret_349; \
+ __ret_349 = splat_laneq_p16(__s0_349, __p1_349); \
+ __ret_349; \
})
#else
-#define vdup_laneq_p16(__p0, __p1) __extension__ ({ \
- poly16x8_t __s0 = __p0; \
- poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly16x4_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdup_laneq_p16(__p0_350, __p1_350) __extension__ ({ \
+ poly16x8_t __s0_350 = __p0_350; \
+ poly16x8_t __rev0_350; __rev0_350 = __builtin_shufflevector(__s0_350, __s0_350, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly16x4_t __ret_350; \
+ __ret_350 = __noswap_splat_laneq_p16(__rev0_350, __p1_350); \
+ __ret_350 = __builtin_shufflevector(__ret_350, __ret_350, 3, 2, 1, 0); \
+ __ret_350; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_laneq_p8(__p0, __p1) __extension__ ({ \
- poly8x16_t __s0 = __p0; \
- poly8x16_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_laneq_p8(__p0_351, __p1_351) __extension__ ({ \
+ poly8x16_t __s0_351 = __p0_351; \
+ poly8x16_t __ret_351; \
+ __ret_351 = splatq_laneq_p8(__s0_351, __p1_351); \
+ __ret_351; \
})
#else
-#define vdupq_laneq_p8(__p0, __p1) __extension__ ({ \
- poly8x16_t __s0 = __p0; \
- poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly8x16_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_laneq_p8(__p0_352, __p1_352) __extension__ ({ \
+ poly8x16_t __s0_352 = __p0_352; \
+ poly8x16_t __rev0_352; __rev0_352 = __builtin_shufflevector(__s0_352, __s0_352, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly8x16_t __ret_352; \
+ __ret_352 = __noswap_splatq_laneq_p8(__rev0_352, __p1_352); \
+ __ret_352 = __builtin_shufflevector(__ret_352, __ret_352, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_352; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_laneq_p64(__p0, __p1) __extension__ ({ \
- poly64x2_t __s0 = __p0; \
- poly64x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret; \
+#define vdupq_laneq_p64(__p0_353, __p1_353) __extension__ ({ \
+ poly64x2_t __s0_353 = __p0_353; \
+ poly64x2_t __ret_353; \
+ __ret_353 = splatq_laneq_p64(__s0_353, __p1_353); \
+ __ret_353; \
})
#else
-#define vdupq_laneq_p64(__p0, __p1) __extension__ ({ \
- poly64x2_t __s0 = __p0; \
- poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- poly64x2_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdupq_laneq_p64(__p0_354, __p1_354) __extension__ ({ \
+ poly64x2_t __s0_354 = __p0_354; \
+ poly64x2_t __rev0_354; __rev0_354 = __builtin_shufflevector(__s0_354, __s0_354, 1, 0); \
+ poly64x2_t __ret_354; \
+ __ret_354 = __noswap_splatq_laneq_p64(__rev0_354, __p1_354); \
+ __ret_354 = __builtin_shufflevector(__ret_354, __ret_354, 1, 0); \
+ __ret_354; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_laneq_p16(__p0, __p1) __extension__ ({ \
- poly16x8_t __s0 = __p0; \
- poly16x8_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_laneq_p16(__p0_355, __p1_355) __extension__ ({ \
+ poly16x8_t __s0_355 = __p0_355; \
+ poly16x8_t __ret_355; \
+ __ret_355 = splatq_laneq_p16(__s0_355, __p1_355); \
+ __ret_355; \
})
#else
-#define vdupq_laneq_p16(__p0, __p1) __extension__ ({ \
- poly16x8_t __s0 = __p0; \
- poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- poly16x8_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_laneq_p16(__p0_356, __p1_356) __extension__ ({ \
+ poly16x8_t __s0_356 = __p0_356; \
+ poly16x8_t __rev0_356; __rev0_356 = __builtin_shufflevector(__s0_356, __s0_356, 7, 6, 5, 4, 3, 2, 1, 0); \
+ poly16x8_t __ret_356; \
+ __ret_356 = __noswap_splatq_laneq_p16(__rev0_356, __p1_356); \
+ __ret_356 = __builtin_shufflevector(__ret_356, __ret_356, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_356; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_laneq_u8(__p0, __p1) __extension__ ({ \
- uint8x16_t __s0 = __p0; \
- uint8x16_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_laneq_u8(__p0_357, __p1_357) __extension__ ({ \
+ uint8x16_t __s0_357 = __p0_357; \
+ uint8x16_t __ret_357; \
+ __ret_357 = splatq_laneq_u8(__s0_357, __p1_357); \
+ __ret_357; \
})
#else
-#define vdupq_laneq_u8(__p0, __p1) __extension__ ({ \
- uint8x16_t __s0 = __p0; \
- uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x16_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_laneq_u8(__p0_358, __p1_358) __extension__ ({ \
+ uint8x16_t __s0_358 = __p0_358; \
+ uint8x16_t __rev0_358; __rev0_358 = __builtin_shufflevector(__s0_358, __s0_358, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __ret_358; \
+ __ret_358 = __noswap_splatq_laneq_u8(__rev0_358, __p1_358); \
+ __ret_358 = __builtin_shufflevector(__ret_358, __ret_358, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_358; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_laneq_u32(__p0, __p1) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x4_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_laneq_u32(__p0_359, __p1_359) __extension__ ({ \
+ uint32x4_t __s0_359 = __p0_359; \
+ uint32x4_t __ret_359; \
+ __ret_359 = splatq_laneq_u32(__s0_359, __p1_359); \
+ __ret_359; \
})
#else
-#define vdupq_laneq_u32(__p0, __p1) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_laneq_u32(__p0_360, __p1_360) __extension__ ({ \
+ uint32x4_t __s0_360 = __p0_360; \
+ uint32x4_t __rev0_360; __rev0_360 = __builtin_shufflevector(__s0_360, __s0_360, 3, 2, 1, 0); \
+ uint32x4_t __ret_360; \
+ __ret_360 = __noswap_splatq_laneq_u32(__rev0_360, __p1_360); \
+ __ret_360 = __builtin_shufflevector(__ret_360, __ret_360, 3, 2, 1, 0); \
+ __ret_360; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_laneq_u64(__p0, __p1) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint64x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret; \
+#define vdupq_laneq_u64(__p0_361, __p1_361) __extension__ ({ \
+ uint64x2_t __s0_361 = __p0_361; \
+ uint64x2_t __ret_361; \
+ __ret_361 = splatq_laneq_u64(__s0_361, __p1_361); \
+ __ret_361; \
})
#else
-#define vdupq_laneq_u64(__p0, __p1) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint64x2_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdupq_laneq_u64(__p0_362, __p1_362) __extension__ ({ \
+ uint64x2_t __s0_362 = __p0_362; \
+ uint64x2_t __rev0_362; __rev0_362 = __builtin_shufflevector(__s0_362, __s0_362, 1, 0); \
+ uint64x2_t __ret_362; \
+ __ret_362 = __noswap_splatq_laneq_u64(__rev0_362, __p1_362); \
+ __ret_362 = __builtin_shufflevector(__ret_362, __ret_362, 1, 0); \
+ __ret_362; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_laneq_u16(__p0, __p1) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x8_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_laneq_u16(__p0_363, __p1_363) __extension__ ({ \
+ uint16x8_t __s0_363 = __p0_363; \
+ uint16x8_t __ret_363; \
+ __ret_363 = splatq_laneq_u16(__s0_363, __p1_363); \
+ __ret_363; \
})
#else
-#define vdupq_laneq_u16(__p0, __p1) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_laneq_u16(__p0_364, __p1_364) __extension__ ({ \
+ uint16x8_t __s0_364 = __p0_364; \
+ uint16x8_t __rev0_364; __rev0_364 = __builtin_shufflevector(__s0_364, __s0_364, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __ret_364; \
+ __ret_364 = __noswap_splatq_laneq_u16(__rev0_364, __p1_364); \
+ __ret_364 = __builtin_shufflevector(__ret_364, __ret_364, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_364; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_laneq_s8(__p0, __p1) __extension__ ({ \
- int8x16_t __s0 = __p0; \
- int8x16_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_laneq_s8(__p0_365, __p1_365) __extension__ ({ \
+ int8x16_t __s0_365 = __p0_365; \
+ int8x16_t __ret_365; \
+ __ret_365 = splatq_laneq_s8(__s0_365, __p1_365); \
+ __ret_365; \
})
#else
-#define vdupq_laneq_s8(__p0, __p1) __extension__ ({ \
- int8x16_t __s0 = __p0; \
- int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x16_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_laneq_s8(__p0_366, __p1_366) __extension__ ({ \
+ int8x16_t __s0_366 = __p0_366; \
+ int8x16_t __rev0_366; __rev0_366 = __builtin_shufflevector(__s0_366, __s0_366, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __ret_366; \
+ __ret_366 = __noswap_splatq_laneq_s8(__rev0_366, __p1_366); \
+ __ret_366 = __builtin_shufflevector(__ret_366, __ret_366, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_366; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_laneq_f64(__p0, __p1) __extension__ ({ \
- float64x2_t __s0 = __p0; \
- float64x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret; \
+#define vdupq_laneq_f64(__p0_367, __p1_367) __extension__ ({ \
+ float64x2_t __s0_367 = __p0_367; \
+ float64x2_t __ret_367; \
+ __ret_367 = splatq_laneq_f64(__s0_367, __p1_367); \
+ __ret_367; \
})
#else
-#define vdupq_laneq_f64(__p0, __p1) __extension__ ({ \
- float64x2_t __s0 = __p0; \
- float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float64x2_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdupq_laneq_f64(__p0_368, __p1_368) __extension__ ({ \
+ float64x2_t __s0_368 = __p0_368; \
+ float64x2_t __rev0_368; __rev0_368 = __builtin_shufflevector(__s0_368, __s0_368, 1, 0); \
+ float64x2_t __ret_368; \
+ __ret_368 = __noswap_splatq_laneq_f64(__rev0_368, __p1_368); \
+ __ret_368 = __builtin_shufflevector(__ret_368, __ret_368, 1, 0); \
+ __ret_368; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_laneq_f32(__p0, __p1) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x4_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_laneq_f32(__p0_369, __p1_369) __extension__ ({ \
+ float32x4_t __s0_369 = __p0_369; \
+ float32x4_t __ret_369; \
+ __ret_369 = splatq_laneq_f32(__s0_369, __p1_369); \
+ __ret_369; \
})
#else
-#define vdupq_laneq_f32(__p0, __p1) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float32x4_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_laneq_f32(__p0_370, __p1_370) __extension__ ({ \
+ float32x4_t __s0_370 = __p0_370; \
+ float32x4_t __rev0_370; __rev0_370 = __builtin_shufflevector(__s0_370, __s0_370, 3, 2, 1, 0); \
+ float32x4_t __ret_370; \
+ __ret_370 = __noswap_splatq_laneq_f32(__rev0_370, __p1_370); \
+ __ret_370 = __builtin_shufflevector(__ret_370, __ret_370, 3, 2, 1, 0); \
+ __ret_370; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_laneq_f16(__p0, __p1) __extension__ ({ \
- float16x8_t __s0 = __p0; \
- float16x8_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_laneq_f16(__p0_371, __p1_371) __extension__ ({ \
+ float16x8_t __s0_371 = __p0_371; \
+ float16x8_t __ret_371; \
+ __ret_371 = splatq_laneq_f16(__s0_371, __p1_371); \
+ __ret_371; \
})
#else
-#define vdupq_laneq_f16(__p0, __p1) __extension__ ({ \
- float16x8_t __s0 = __p0; \
- float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x8_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_laneq_f16(__p0_372, __p1_372) __extension__ ({ \
+ float16x8_t __s0_372 = __p0_372; \
+ float16x8_t __rev0_372; __rev0_372 = __builtin_shufflevector(__s0_372, __s0_372, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __ret_372; \
+ __ret_372 = __noswap_splatq_laneq_f16(__rev0_372, __p1_372); \
+ __ret_372 = __builtin_shufflevector(__ret_372, __ret_372, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_372; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_laneq_s32(__p0, __p1) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_laneq_s32(__p0_373, __p1_373) __extension__ ({ \
+ int32x4_t __s0_373 = __p0_373; \
+ int32x4_t __ret_373; \
+ __ret_373 = splatq_laneq_s32(__s0_373, __p1_373); \
+ __ret_373; \
})
#else
-#define vdupq_laneq_s32(__p0, __p1) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_laneq_s32(__p0_374, __p1_374) __extension__ ({ \
+ int32x4_t __s0_374 = __p0_374; \
+ int32x4_t __rev0_374; __rev0_374 = __builtin_shufflevector(__s0_374, __s0_374, 3, 2, 1, 0); \
+ int32x4_t __ret_374; \
+ __ret_374 = __noswap_splatq_laneq_s32(__rev0_374, __p1_374); \
+ __ret_374 = __builtin_shufflevector(__ret_374, __ret_374, 3, 2, 1, 0); \
+ __ret_374; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_laneq_s64(__p0, __p1) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int64x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret; \
+#define vdupq_laneq_s64(__p0_375, __p1_375) __extension__ ({ \
+ int64x2_t __s0_375 = __p0_375; \
+ int64x2_t __ret_375; \
+ __ret_375 = splatq_laneq_s64(__s0_375, __p1_375); \
+ __ret_375; \
})
#else
-#define vdupq_laneq_s64(__p0, __p1) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int64x2_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdupq_laneq_s64(__p0_376, __p1_376) __extension__ ({ \
+ int64x2_t __s0_376 = __p0_376; \
+ int64x2_t __rev0_376; __rev0_376 = __builtin_shufflevector(__s0_376, __s0_376, 1, 0); \
+ int64x2_t __ret_376; \
+ __ret_376 = __noswap_splatq_laneq_s64(__rev0_376, __p1_376); \
+ __ret_376 = __builtin_shufflevector(__ret_376, __ret_376, 1, 0); \
+ __ret_376; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdupq_laneq_s16(__p0, __p1) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdupq_laneq_s16(__p0_377, __p1_377) __extension__ ({ \
+ int16x8_t __s0_377 = __p0_377; \
+ int16x8_t __ret_377; \
+ __ret_377 = splatq_laneq_s16(__s0_377, __p1_377); \
+ __ret_377; \
})
#else
-#define vdupq_laneq_s16(__p0, __p1) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdupq_laneq_s16(__p0_378, __p1_378) __extension__ ({ \
+ int16x8_t __s0_378 = __p0_378; \
+ int16x8_t __rev0_378; __rev0_378 = __builtin_shufflevector(__s0_378, __s0_378, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __ret_378; \
+ __ret_378 = __noswap_splatq_laneq_s16(__rev0_378, __p1_378); \
+ __ret_378 = __builtin_shufflevector(__ret_378, __ret_378, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_378; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_laneq_u8(__p0, __p1) __extension__ ({ \
- uint8x16_t __s0 = __p0; \
- uint8x8_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdup_laneq_u8(__p0_379, __p1_379) __extension__ ({ \
+ uint8x16_t __s0_379 = __p0_379; \
+ uint8x8_t __ret_379; \
+ __ret_379 = splat_laneq_u8(__s0_379, __p1_379); \
+ __ret_379; \
})
#else
-#define vdup_laneq_u8(__p0, __p1) __extension__ ({ \
- uint8x16_t __s0 = __p0; \
- uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x8_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdup_laneq_u8(__p0_380, __p1_380) __extension__ ({ \
+ uint8x16_t __s0_380 = __p0_380; \
+ uint8x16_t __rev0_380; __rev0_380 = __builtin_shufflevector(__s0_380, __s0_380, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x8_t __ret_380; \
+ __ret_380 = __noswap_splat_laneq_u8(__rev0_380, __p1_380); \
+ __ret_380 = __builtin_shufflevector(__ret_380, __ret_380, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_380; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_laneq_u32(__p0, __p1) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret; \
+#define vdup_laneq_u32(__p0_381, __p1_381) __extension__ ({ \
+ uint32x4_t __s0_381 = __p0_381; \
+ uint32x2_t __ret_381; \
+ __ret_381 = splat_laneq_u32(__s0_381, __p1_381); \
+ __ret_381; \
})
#else
-#define vdup_laneq_u32(__p0, __p1) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint32x2_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdup_laneq_u32(__p0_382, __p1_382) __extension__ ({ \
+ uint32x4_t __s0_382 = __p0_382; \
+ uint32x4_t __rev0_382; __rev0_382 = __builtin_shufflevector(__s0_382, __s0_382, 3, 2, 1, 0); \
+ uint32x2_t __ret_382; \
+ __ret_382 = __noswap_splat_laneq_u32(__rev0_382, __p1_382); \
+ __ret_382 = __builtin_shufflevector(__ret_382, __ret_382, 1, 0); \
+ __ret_382; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_laneq_u64(__p0, __p1) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint64x1_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1); \
- __ret; \
+#define vdup_laneq_u64(__p0_383, __p1_383) __extension__ ({ \
+ uint64x2_t __s0_383 = __p0_383; \
+ uint64x1_t __ret_383; \
+ __ret_383 = splat_laneq_u64(__s0_383, __p1_383); \
+ __ret_383; \
})
#else
-#define vdup_laneq_u64(__p0, __p1) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint64x1_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1); \
- __ret; \
+#define vdup_laneq_u64(__p0_384, __p1_384) __extension__ ({ \
+ uint64x2_t __s0_384 = __p0_384; \
+ uint64x2_t __rev0_384; __rev0_384 = __builtin_shufflevector(__s0_384, __s0_384, 1, 0); \
+ uint64x1_t __ret_384; \
+ __ret_384 = __noswap_splat_laneq_u64(__rev0_384, __p1_384); \
+ __ret_384; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_laneq_u16(__p0, __p1) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x4_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdup_laneq_u16(__p0_385, __p1_385) __extension__ ({ \
+ uint16x8_t __s0_385 = __p0_385; \
+ uint16x4_t __ret_385; \
+ __ret_385 = splat_laneq_u16(__s0_385, __p1_385); \
+ __ret_385; \
})
#else
-#define vdup_laneq_u16(__p0, __p1) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x4_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdup_laneq_u16(__p0_386, __p1_386) __extension__ ({ \
+ uint16x8_t __s0_386 = __p0_386; \
+ uint16x8_t __rev0_386; __rev0_386 = __builtin_shufflevector(__s0_386, __s0_386, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x4_t __ret_386; \
+ __ret_386 = __noswap_splat_laneq_u16(__rev0_386, __p1_386); \
+ __ret_386 = __builtin_shufflevector(__ret_386, __ret_386, 3, 2, 1, 0); \
+ __ret_386; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_laneq_s8(__p0, __p1) __extension__ ({ \
- int8x16_t __s0 = __p0; \
- int8x8_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdup_laneq_s8(__p0_387, __p1_387) __extension__ ({ \
+ int8x16_t __s0_387 = __p0_387; \
+ int8x8_t __ret_387; \
+ __ret_387 = splat_laneq_s8(__s0_387, __p1_387); \
+ __ret_387; \
})
#else
-#define vdup_laneq_s8(__p0, __p1) __extension__ ({ \
- int8x16_t __s0 = __p0; \
- int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x8_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vdup_laneq_s8(__p0_388, __p1_388) __extension__ ({ \
+ int8x16_t __s0_388 = __p0_388; \
+ int8x16_t __rev0_388; __rev0_388 = __builtin_shufflevector(__s0_388, __s0_388, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x8_t __ret_388; \
+ __ret_388 = __noswap_splat_laneq_s8(__rev0_388, __p1_388); \
+ __ret_388 = __builtin_shufflevector(__ret_388, __ret_388, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_388; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_laneq_f64(__p0, __p1) __extension__ ({ \
- float64x2_t __s0 = __p0; \
- float64x1_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1); \
- __ret; \
+#define vdup_laneq_f64(__p0_389, __p1_389) __extension__ ({ \
+ float64x2_t __s0_389 = __p0_389; \
+ float64x1_t __ret_389; \
+ __ret_389 = splat_laneq_f64(__s0_389, __p1_389); \
+ __ret_389; \
})
#else
-#define vdup_laneq_f64(__p0, __p1) __extension__ ({ \
- float64x2_t __s0 = __p0; \
- float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float64x1_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1); \
- __ret; \
+#define vdup_laneq_f64(__p0_390, __p1_390) __extension__ ({ \
+ float64x2_t __s0_390 = __p0_390; \
+ float64x2_t __rev0_390; __rev0_390 = __builtin_shufflevector(__s0_390, __s0_390, 1, 0); \
+ float64x1_t __ret_390; \
+ __ret_390 = __noswap_splat_laneq_f64(__rev0_390, __p1_390); \
+ __ret_390; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_laneq_f32(__p0, __p1) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret; \
+#define vdup_laneq_f32(__p0_391, __p1_391) __extension__ ({ \
+ float32x4_t __s0_391 = __p0_391; \
+ float32x2_t __ret_391; \
+ __ret_391 = splat_laneq_f32(__s0_391, __p1_391); \
+ __ret_391; \
})
#else
-#define vdup_laneq_f32(__p0, __p1) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float32x2_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdup_laneq_f32(__p0_392, __p1_392) __extension__ ({ \
+ float32x4_t __s0_392 = __p0_392; \
+ float32x4_t __rev0_392; __rev0_392 = __builtin_shufflevector(__s0_392, __s0_392, 3, 2, 1, 0); \
+ float32x2_t __ret_392; \
+ __ret_392 = __noswap_splat_laneq_f32(__rev0_392, __p1_392); \
+ __ret_392 = __builtin_shufflevector(__ret_392, __ret_392, 1, 0); \
+ __ret_392; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_laneq_f16(__p0, __p1) __extension__ ({ \
- float16x8_t __s0 = __p0; \
- float16x4_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdup_laneq_f16(__p0_393, __p1_393) __extension__ ({ \
+ float16x8_t __s0_393 = __p0_393; \
+ float16x4_t __ret_393; \
+ __ret_393 = splat_laneq_f16(__s0_393, __p1_393); \
+ __ret_393; \
})
#else
-#define vdup_laneq_f16(__p0, __p1) __extension__ ({ \
- float16x8_t __s0 = __p0; \
- float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x4_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdup_laneq_f16(__p0_394, __p1_394) __extension__ ({ \
+ float16x8_t __s0_394 = __p0_394; \
+ float16x8_t __rev0_394; __rev0_394 = __builtin_shufflevector(__s0_394, __s0_394, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x4_t __ret_394; \
+ __ret_394 = __noswap_splat_laneq_f16(__rev0_394, __p1_394); \
+ __ret_394 = __builtin_shufflevector(__ret_394, __ret_394, 3, 2, 1, 0); \
+ __ret_394; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_laneq_s32(__p0, __p1) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x2_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
- __ret; \
+#define vdup_laneq_s32(__p0_395, __p1_395) __extension__ ({ \
+ int32x4_t __s0_395 = __p0_395; \
+ int32x2_t __ret_395; \
+ __ret_395 = splat_laneq_s32(__s0_395, __p1_395); \
+ __ret_395; \
})
#else
-#define vdup_laneq_s32(__p0, __p1) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x2_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vdup_laneq_s32(__p0_396, __p1_396) __extension__ ({ \
+ int32x4_t __s0_396 = __p0_396; \
+ int32x4_t __rev0_396; __rev0_396 = __builtin_shufflevector(__s0_396, __s0_396, 3, 2, 1, 0); \
+ int32x2_t __ret_396; \
+ __ret_396 = __noswap_splat_laneq_s32(__rev0_396, __p1_396); \
+ __ret_396 = __builtin_shufflevector(__ret_396, __ret_396, 1, 0); \
+ __ret_396; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_laneq_s64(__p0, __p1) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int64x1_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1); \
- __ret; \
+#define vdup_laneq_s64(__p0_397, __p1_397) __extension__ ({ \
+ int64x2_t __s0_397 = __p0_397; \
+ int64x1_t __ret_397; \
+ __ret_397 = splat_laneq_s64(__s0_397, __p1_397); \
+ __ret_397; \
})
#else
-#define vdup_laneq_s64(__p0, __p1) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int64x1_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1); \
- __ret; \
+#define vdup_laneq_s64(__p0_398, __p1_398) __extension__ ({ \
+ int64x2_t __s0_398 = __p0_398; \
+ int64x2_t __rev0_398; __rev0_398 = __builtin_shufflevector(__s0_398, __s0_398, 1, 0); \
+ int64x1_t __ret_398; \
+ __ret_398 = __noswap_splat_laneq_s64(__rev0_398, __p1_398); \
+ __ret_398; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vdup_laneq_s16(__p0, __p1) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x4_t __ret; \
- __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
- __ret; \
+#define vdup_laneq_s16(__p0_399, __p1_399) __extension__ ({ \
+ int16x8_t __s0_399 = __p0_399; \
+ int16x4_t __ret_399; \
+ __ret_399 = splat_laneq_s16(__s0_399, __p1_399); \
+ __ret_399; \
})
#else
-#define vdup_laneq_s16(__p0, __p1) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __ret; \
- __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vdup_laneq_s16(__p0_400, __p1_400) __extension__ ({ \
+ int16x8_t __s0_400 = __p0_400; \
+ int16x8_t __rev0_400; __rev0_400 = __builtin_shufflevector(__s0_400, __s0_400, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __ret_400; \
+ __ret_400 = __noswap_splat_laneq_s16(__rev0_400, __p1_400); \
+ __ret_400 = __builtin_shufflevector(__ret_400, __ret_400, 3, 2, 1, 0); \
+ __ret_400; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vdup_n_p64(poly64_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t) {__p0};
- return __ret;
-}
-#else
__ai poly64x1_t vdup_n_p64(poly64_t __p0) {
poly64x1_t __ret;
__ret = (poly64x1_t) {__p0};
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai poly64x2_t vdupq_n_p64(poly64_t __p0) {
poly64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vdup_n_f64(float64_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) {__p0};
- return __ret;
-}
-#else
__ai float64x1_t vdup_n_f64(float64_t __p0) {
float64x1_t __ret;
__ret = (float64x1_t) {__p0};
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vext_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x1_t __s0 = __p0; \
poly64x1_t __s1 = __p1; \
__ret = (poly64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 6); \
__ret; \
})
-#else
-#define vext_p64(__p0, __p1, __p2) __extension__ ({ \
- poly64x1_t __s0 = __p0; \
- poly64x1_t __s1 = __p1; \
- poly64x1_t __ret; \
- __ret = (poly64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 6); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vextq_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x2_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vext_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x1_t __s0 = __p0; \
- float64x1_t __s1 = __p1; \
- float64x1_t __ret; \
- __ret = (float64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 10); \
- __ret; \
-})
-#else
#define vext_f64(__p0, __p1, __p2) __extension__ ({ \
float64x1_t __s0 = __p0; \
float64x1_t __s1 = __p1; \
__ret = (float64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 10); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float64x2_t vfmaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
float64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
- return __ret;
-}
-#else
-__ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
- return __ret;
-}
-__ai float64x1_t __noswap_vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vfmad_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
- float64_t __s0 = __p0; \
- float64_t __s1 = __p1; \
- float64x1_t __s2 = __p2; \
- float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vfmad_lane_f64(__s0, __s1, (int8x8_t)__s2, __p3); \
- __ret; \
-})
-#else
+__ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
+ float64x1_t __ret;
+ __ret = (float64x1_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
+ return __ret;
+}
#define vfmad_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
float64_t __s0 = __p0; \
float64_t __s1 = __p1; \
float64x1_t __s2 = __p2; \
float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vfmad_lane_f64(__s0, __s1, (int8x8_t)__s2, __p3); \
- __ret; \
-})
-#define __noswap_vfmad_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
- float64_t __s0 = __p0; \
- float64_t __s1 = __p1; \
- float64x1_t __s2 = __p2; \
- float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vfmad_lane_f64(__s0, __s1, (int8x8_t)__s2, __p3); \
+ __ret = (float64_t) __builtin_neon_vfmad_lane_f64(__s0, __s1, (float64x1_t)__s2, __p3); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
float32_t __s0 = __p0; \
float32_t __s1 = __p1; \
float32x2_t __s2 = __p2; \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vfmas_lane_f32(__s0, __s1, (int8x8_t)__s2, __p3); \
+ __ret = (float32_t) __builtin_neon_vfmas_lane_f32(__s0, __s1, (float32x2_t)__s2, __p3); \
__ret; \
})
#else
float32x2_t __s2 = __p2; \
float32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vfmas_lane_f32(__s0, __s1, (int8x8_t)__rev2, __p3); \
+ __ret = (float32_t) __builtin_neon_vfmas_lane_f32(__s0, __s1, (float32x2_t)__rev2, __p3); \
__ret; \
})
#define __noswap_vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
float32_t __s1 = __p1; \
float32x2_t __s2 = __p2; \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vfmas_lane_f32(__s0, __s1, (int8x8_t)__s2, __p3); \
+ __ret = (float32_t) __builtin_neon_vfmas_lane_f32(__s0, __s1, (float32x2_t)__s2, __p3); \
__ret; \
})
#endif
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vfma_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
- float64x1_t __s0 = __p0; \
- float64x1_t __s1 = __p1; \
- float64x1_t __s2 = __p2; \
- float64x1_t __ret; \
- __ret = (float64x1_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 10); \
- __ret; \
-})
-#else
#define vfma_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
float64x1_t __s0 = __p0; \
float64x1_t __s1 = __p1; \
__ret = (float64x1_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 10); \
__ret; \
})
-#define __noswap_vfma_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
- float64x1_t __s0 = __p0; \
- float64x1_t __s1 = __p1; \
- float64x1_t __s2 = __p2; \
- float64x1_t __ret; \
- __ret = (float64x1_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 10); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
float32x2_t __s0 = __p0; \
float64_t __s1 = __p1; \
float64x2_t __s2 = __p2; \
float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vfmad_laneq_f64(__s0, __s1, (int8x16_t)__s2, __p3); \
+ __ret = (float64_t) __builtin_neon_vfmad_laneq_f64(__s0, __s1, (float64x2_t)__s2, __p3); \
__ret; \
})
#else
float64x2_t __s2 = __p2; \
float64x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vfmad_laneq_f64(__s0, __s1, (int8x16_t)__rev2, __p3); \
+ __ret = (float64_t) __builtin_neon_vfmad_laneq_f64(__s0, __s1, (float64x2_t)__rev2, __p3); \
__ret; \
})
#define __noswap_vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
float64_t __s1 = __p1; \
float64x2_t __s2 = __p2; \
float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vfmad_laneq_f64(__s0, __s1, (int8x16_t)__s2, __p3); \
+ __ret = (float64_t) __builtin_neon_vfmad_laneq_f64(__s0, __s1, (float64x2_t)__s2, __p3); \
__ret; \
})
#endif
float32_t __s1 = __p1; \
float32x4_t __s2 = __p2; \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vfmas_laneq_f32(__s0, __s1, (int8x16_t)__s2, __p3); \
+ __ret = (float32_t) __builtin_neon_vfmas_laneq_f32(__s0, __s1, (float32x4_t)__s2, __p3); \
__ret; \
})
#else
float32x4_t __s2 = __p2; \
float32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vfmas_laneq_f32(__s0, __s1, (int8x16_t)__rev2, __p3); \
+ __ret = (float32_t) __builtin_neon_vfmas_laneq_f32(__s0, __s1, (float32x4_t)__rev2, __p3); \
__ret; \
})
#define __noswap_vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
float32_t __s1 = __p1; \
float32x4_t __s2 = __p2; \
float32_t __ret; \
- __ret = (float32_t) __builtin_neon_vfmas_laneq_f32(__s0, __s1, (int8x16_t)__s2, __p3); \
+ __ret = (float32_t) __builtin_neon_vfmas_laneq_f32(__s0, __s1, (float32x4_t)__s2, __p3); \
__ret; \
})
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai float64x1_t vfma_n_f64(float64x1_t __p0, float64x1_t __p1, float64_t __p2) {
float64x1_t __ret;
__ret = vfma_f64(__p0, __p1, (float64x1_t) {__p2});
return __ret;
}
-#else
-__ai float64x1_t vfma_n_f64(float64x1_t __p0, float64x1_t __p1, float64_t __p2) {
- float64x1_t __ret;
- __ret = __noswap_vfma_f64(__p0, __p1, (float64x1_t) {__p2});
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float64x2_t vfmsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
float64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
float64x1_t __ret;
__ret = vfma_f64(__p0, -__p1, __p2);
return __ret;
}
-#else
-__ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
- float64x1_t __ret;
- __ret = __noswap_vfma_f64(__p0, -__p1, __p2);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vfmsd_lane_f64(__p0_100, __p1_100, __p2_100, __p3_100) __extension__ ({ \
- float64_t __s0_100 = __p0_100; \
- float64_t __s1_100 = __p1_100; \
- float64x1_t __s2_100 = __p2_100; \
- float64_t __ret_100; \
- __ret_100 = vfmad_lane_f64(__s0_100, -__s1_100, __s2_100, __p3_100); \
- __ret_100; \
-})
-#else
-#define vfmsd_lane_f64(__p0_101, __p1_101, __p2_101, __p3_101) __extension__ ({ \
- float64_t __s0_101 = __p0_101; \
- float64_t __s1_101 = __p1_101; \
- float64x1_t __s2_101 = __p2_101; \
- float64_t __ret_101; \
- __ret_101 = __noswap_vfmad_lane_f64(__s0_101, -__s1_101, __s2_101, __p3_101); \
- __ret_101; \
+#define vfmsd_lane_f64(__p0_401, __p1_401, __p2_401, __p3_401) __extension__ ({ \
+ float64_t __s0_401 = __p0_401; \
+ float64_t __s1_401 = __p1_401; \
+ float64x1_t __s2_401 = __p2_401; \
+ float64_t __ret_401; \
+ __ret_401 = vfmad_lane_f64(__s0_401, -__s1_401, __s2_401, __p3_401); \
+ __ret_401; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vfmss_lane_f32(__p0_102, __p1_102, __p2_102, __p3_102) __extension__ ({ \
- float32_t __s0_102 = __p0_102; \
- float32_t __s1_102 = __p1_102; \
- float32x2_t __s2_102 = __p2_102; \
- float32_t __ret_102; \
- __ret_102 = vfmas_lane_f32(__s0_102, -__s1_102, __s2_102, __p3_102); \
- __ret_102; \
+#define vfmss_lane_f32(__p0_402, __p1_402, __p2_402, __p3_402) __extension__ ({ \
+ float32_t __s0_402 = __p0_402; \
+ float32_t __s1_402 = __p1_402; \
+ float32x2_t __s2_402 = __p2_402; \
+ float32_t __ret_402; \
+ __ret_402 = vfmas_lane_f32(__s0_402, -__s1_402, __s2_402, __p3_402); \
+ __ret_402; \
})
#else
-#define vfmss_lane_f32(__p0_103, __p1_103, __p2_103, __p3_103) __extension__ ({ \
- float32_t __s0_103 = __p0_103; \
- float32_t __s1_103 = __p1_103; \
- float32x2_t __s2_103 = __p2_103; \
- float32x2_t __rev2_103; __rev2_103 = __builtin_shufflevector(__s2_103, __s2_103, 1, 0); \
- float32_t __ret_103; \
- __ret_103 = __noswap_vfmas_lane_f32(__s0_103, -__s1_103, __rev2_103, __p3_103); \
- __ret_103; \
+#define vfmss_lane_f32(__p0_403, __p1_403, __p2_403, __p3_403) __extension__ ({ \
+ float32_t __s0_403 = __p0_403; \
+ float32_t __s1_403 = __p1_403; \
+ float32x2_t __s2_403 = __p2_403; \
+ float32x2_t __rev2_403; __rev2_403 = __builtin_shufflevector(__s2_403, __s2_403, 1, 0); \
+ float32_t __ret_403; \
+ __ret_403 = __noswap_vfmas_lane_f32(__s0_403, -__s1_403, __rev2_403, __p3_403); \
+ __ret_403; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vfmsq_lane_f64(__p0_104, __p1_104, __p2_104, __p3_104) __extension__ ({ \
- float64x2_t __s0_104 = __p0_104; \
- float64x2_t __s1_104 = __p1_104; \
- float64x1_t __s2_104 = __p2_104; \
- float64x2_t __ret_104; \
- __ret_104 = vfmaq_lane_f64(__s0_104, -__s1_104, __s2_104, __p3_104); \
- __ret_104; \
+#define vfmsq_lane_f64(__p0_404, __p1_404, __p2_404, __p3_404) __extension__ ({ \
+ float64x2_t __s0_404 = __p0_404; \
+ float64x2_t __s1_404 = __p1_404; \
+ float64x1_t __s2_404 = __p2_404; \
+ float64x2_t __ret_404; \
+ __ret_404 = vfmaq_lane_f64(__s0_404, -__s1_404, __s2_404, __p3_404); \
+ __ret_404; \
})
#else
-#define vfmsq_lane_f64(__p0_105, __p1_105, __p2_105, __p3_105) __extension__ ({ \
- float64x2_t __s0_105 = __p0_105; \
- float64x2_t __s1_105 = __p1_105; \
- float64x1_t __s2_105 = __p2_105; \
- float64x2_t __rev0_105; __rev0_105 = __builtin_shufflevector(__s0_105, __s0_105, 1, 0); \
- float64x2_t __rev1_105; __rev1_105 = __builtin_shufflevector(__s1_105, __s1_105, 1, 0); \
- float64x2_t __ret_105; \
- __ret_105 = __noswap_vfmaq_lane_f64(__rev0_105, -__rev1_105, __s2_105, __p3_105); \
- __ret_105 = __builtin_shufflevector(__ret_105, __ret_105, 1, 0); \
- __ret_105; \
+#define vfmsq_lane_f64(__p0_405, __p1_405, __p2_405, __p3_405) __extension__ ({ \
+ float64x2_t __s0_405 = __p0_405; \
+ float64x2_t __s1_405 = __p1_405; \
+ float64x1_t __s2_405 = __p2_405; \
+ float64x2_t __rev0_405; __rev0_405 = __builtin_shufflevector(__s0_405, __s0_405, 1, 0); \
+ float64x2_t __rev1_405; __rev1_405 = __builtin_shufflevector(__s1_405, __s1_405, 1, 0); \
+ float64x2_t __ret_405; \
+ __ret_405 = __noswap_vfmaq_lane_f64(__rev0_405, -__rev1_405, __s2_405, __p3_405); \
+ __ret_405 = __builtin_shufflevector(__ret_405, __ret_405, 1, 0); \
+ __ret_405; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vfmsq_lane_f32(__p0_106, __p1_106, __p2_106, __p3_106) __extension__ ({ \
- float32x4_t __s0_106 = __p0_106; \
- float32x4_t __s1_106 = __p1_106; \
- float32x2_t __s2_106 = __p2_106; \
- float32x4_t __ret_106; \
- __ret_106 = vfmaq_lane_f32(__s0_106, -__s1_106, __s2_106, __p3_106); \
- __ret_106; \
+#define vfmsq_lane_f32(__p0_406, __p1_406, __p2_406, __p3_406) __extension__ ({ \
+ float32x4_t __s0_406 = __p0_406; \
+ float32x4_t __s1_406 = __p1_406; \
+ float32x2_t __s2_406 = __p2_406; \
+ float32x4_t __ret_406; \
+ __ret_406 = vfmaq_lane_f32(__s0_406, -__s1_406, __s2_406, __p3_406); \
+ __ret_406; \
})
#else
-#define vfmsq_lane_f32(__p0_107, __p1_107, __p2_107, __p3_107) __extension__ ({ \
- float32x4_t __s0_107 = __p0_107; \
- float32x4_t __s1_107 = __p1_107; \
- float32x2_t __s2_107 = __p2_107; \
- float32x4_t __rev0_107; __rev0_107 = __builtin_shufflevector(__s0_107, __s0_107, 3, 2, 1, 0); \
- float32x4_t __rev1_107; __rev1_107 = __builtin_shufflevector(__s1_107, __s1_107, 3, 2, 1, 0); \
- float32x2_t __rev2_107; __rev2_107 = __builtin_shufflevector(__s2_107, __s2_107, 1, 0); \
- float32x4_t __ret_107; \
- __ret_107 = __noswap_vfmaq_lane_f32(__rev0_107, -__rev1_107, __rev2_107, __p3_107); \
- __ret_107 = __builtin_shufflevector(__ret_107, __ret_107, 3, 2, 1, 0); \
- __ret_107; \
+#define vfmsq_lane_f32(__p0_407, __p1_407, __p2_407, __p3_407) __extension__ ({ \
+ float32x4_t __s0_407 = __p0_407; \
+ float32x4_t __s1_407 = __p1_407; \
+ float32x2_t __s2_407 = __p2_407; \
+ float32x4_t __rev0_407; __rev0_407 = __builtin_shufflevector(__s0_407, __s0_407, 3, 2, 1, 0); \
+ float32x4_t __rev1_407; __rev1_407 = __builtin_shufflevector(__s1_407, __s1_407, 3, 2, 1, 0); \
+ float32x2_t __rev2_407; __rev2_407 = __builtin_shufflevector(__s2_407, __s2_407, 1, 0); \
+ float32x4_t __ret_407; \
+ __ret_407 = __noswap_vfmaq_lane_f32(__rev0_407, -__rev1_407, __rev2_407, __p3_407); \
+ __ret_407 = __builtin_shufflevector(__ret_407, __ret_407, 3, 2, 1, 0); \
+ __ret_407; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vfms_lane_f64(__p0_108, __p1_108, __p2_108, __p3_108) __extension__ ({ \
- float64x1_t __s0_108 = __p0_108; \
- float64x1_t __s1_108 = __p1_108; \
- float64x1_t __s2_108 = __p2_108; \
- float64x1_t __ret_108; \
- __ret_108 = vfma_lane_f64(__s0_108, -__s1_108, __s2_108, __p3_108); \
- __ret_108; \
-})
-#else
-#define vfms_lane_f64(__p0_109, __p1_109, __p2_109, __p3_109) __extension__ ({ \
- float64x1_t __s0_109 = __p0_109; \
- float64x1_t __s1_109 = __p1_109; \
- float64x1_t __s2_109 = __p2_109; \
- float64x1_t __ret_109; \
- __ret_109 = __noswap_vfma_lane_f64(__s0_109, -__s1_109, __s2_109, __p3_109); \
- __ret_109; \
+#define vfms_lane_f64(__p0_408, __p1_408, __p2_408, __p3_408) __extension__ ({ \
+ float64x1_t __s0_408 = __p0_408; \
+ float64x1_t __s1_408 = __p1_408; \
+ float64x1_t __s2_408 = __p2_408; \
+ float64x1_t __ret_408; \
+ __ret_408 = vfma_lane_f64(__s0_408, -__s1_408, __s2_408, __p3_408); \
+ __ret_408; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vfms_lane_f32(__p0_110, __p1_110, __p2_110, __p3_110) __extension__ ({ \
- float32x2_t __s0_110 = __p0_110; \
- float32x2_t __s1_110 = __p1_110; \
- float32x2_t __s2_110 = __p2_110; \
- float32x2_t __ret_110; \
- __ret_110 = vfma_lane_f32(__s0_110, -__s1_110, __s2_110, __p3_110); \
- __ret_110; \
+#define vfms_lane_f32(__p0_409, __p1_409, __p2_409, __p3_409) __extension__ ({ \
+ float32x2_t __s0_409 = __p0_409; \
+ float32x2_t __s1_409 = __p1_409; \
+ float32x2_t __s2_409 = __p2_409; \
+ float32x2_t __ret_409; \
+ __ret_409 = vfma_lane_f32(__s0_409, -__s1_409, __s2_409, __p3_409); \
+ __ret_409; \
})
#else
-#define vfms_lane_f32(__p0_111, __p1_111, __p2_111, __p3_111) __extension__ ({ \
- float32x2_t __s0_111 = __p0_111; \
- float32x2_t __s1_111 = __p1_111; \
- float32x2_t __s2_111 = __p2_111; \
- float32x2_t __rev0_111; __rev0_111 = __builtin_shufflevector(__s0_111, __s0_111, 1, 0); \
- float32x2_t __rev1_111; __rev1_111 = __builtin_shufflevector(__s1_111, __s1_111, 1, 0); \
- float32x2_t __rev2_111; __rev2_111 = __builtin_shufflevector(__s2_111, __s2_111, 1, 0); \
- float32x2_t __ret_111; \
- __ret_111 = __noswap_vfma_lane_f32(__rev0_111, -__rev1_111, __rev2_111, __p3_111); \
- __ret_111 = __builtin_shufflevector(__ret_111, __ret_111, 1, 0); \
- __ret_111; \
+#define vfms_lane_f32(__p0_410, __p1_410, __p2_410, __p3_410) __extension__ ({ \
+ float32x2_t __s0_410 = __p0_410; \
+ float32x2_t __s1_410 = __p1_410; \
+ float32x2_t __s2_410 = __p2_410; \
+ float32x2_t __rev0_410; __rev0_410 = __builtin_shufflevector(__s0_410, __s0_410, 1, 0); \
+ float32x2_t __rev1_410; __rev1_410 = __builtin_shufflevector(__s1_410, __s1_410, 1, 0); \
+ float32x2_t __rev2_410; __rev2_410 = __builtin_shufflevector(__s2_410, __s2_410, 1, 0); \
+ float32x2_t __ret_410; \
+ __ret_410 = __noswap_vfma_lane_f32(__rev0_410, -__rev1_410, __rev2_410, __p3_410); \
+ __ret_410 = __builtin_shufflevector(__ret_410, __ret_410, 1, 0); \
+ __ret_410; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vfmsd_laneq_f64(__p0_112, __p1_112, __p2_112, __p3_112) __extension__ ({ \
- float64_t __s0_112 = __p0_112; \
- float64_t __s1_112 = __p1_112; \
- float64x2_t __s2_112 = __p2_112; \
- float64_t __ret_112; \
- __ret_112 = vfmad_laneq_f64(__s0_112, -__s1_112, __s2_112, __p3_112); \
- __ret_112; \
+#define vfmsd_laneq_f64(__p0_411, __p1_411, __p2_411, __p3_411) __extension__ ({ \
+ float64_t __s0_411 = __p0_411; \
+ float64_t __s1_411 = __p1_411; \
+ float64x2_t __s2_411 = __p2_411; \
+ float64_t __ret_411; \
+ __ret_411 = vfmad_laneq_f64(__s0_411, -__s1_411, __s2_411, __p3_411); \
+ __ret_411; \
})
#else
-#define vfmsd_laneq_f64(__p0_113, __p1_113, __p2_113, __p3_113) __extension__ ({ \
- float64_t __s0_113 = __p0_113; \
- float64_t __s1_113 = __p1_113; \
- float64x2_t __s2_113 = __p2_113; \
- float64x2_t __rev2_113; __rev2_113 = __builtin_shufflevector(__s2_113, __s2_113, 1, 0); \
- float64_t __ret_113; \
- __ret_113 = __noswap_vfmad_laneq_f64(__s0_113, -__s1_113, __rev2_113, __p3_113); \
- __ret_113; \
+#define vfmsd_laneq_f64(__p0_412, __p1_412, __p2_412, __p3_412) __extension__ ({ \
+ float64_t __s0_412 = __p0_412; \
+ float64_t __s1_412 = __p1_412; \
+ float64x2_t __s2_412 = __p2_412; \
+ float64x2_t __rev2_412; __rev2_412 = __builtin_shufflevector(__s2_412, __s2_412, 1, 0); \
+ float64_t __ret_412; \
+ __ret_412 = __noswap_vfmad_laneq_f64(__s0_412, -__s1_412, __rev2_412, __p3_412); \
+ __ret_412; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vfmss_laneq_f32(__p0_114, __p1_114, __p2_114, __p3_114) __extension__ ({ \
- float32_t __s0_114 = __p0_114; \
- float32_t __s1_114 = __p1_114; \
- float32x4_t __s2_114 = __p2_114; \
- float32_t __ret_114; \
- __ret_114 = vfmas_laneq_f32(__s0_114, -__s1_114, __s2_114, __p3_114); \
- __ret_114; \
+#define vfmss_laneq_f32(__p0_413, __p1_413, __p2_413, __p3_413) __extension__ ({ \
+ float32_t __s0_413 = __p0_413; \
+ float32_t __s1_413 = __p1_413; \
+ float32x4_t __s2_413 = __p2_413; \
+ float32_t __ret_413; \
+ __ret_413 = vfmas_laneq_f32(__s0_413, -__s1_413, __s2_413, __p3_413); \
+ __ret_413; \
})
#else
-#define vfmss_laneq_f32(__p0_115, __p1_115, __p2_115, __p3_115) __extension__ ({ \
- float32_t __s0_115 = __p0_115; \
- float32_t __s1_115 = __p1_115; \
- float32x4_t __s2_115 = __p2_115; \
- float32x4_t __rev2_115; __rev2_115 = __builtin_shufflevector(__s2_115, __s2_115, 3, 2, 1, 0); \
- float32_t __ret_115; \
- __ret_115 = __noswap_vfmas_laneq_f32(__s0_115, -__s1_115, __rev2_115, __p3_115); \
- __ret_115; \
+#define vfmss_laneq_f32(__p0_414, __p1_414, __p2_414, __p3_414) __extension__ ({ \
+ float32_t __s0_414 = __p0_414; \
+ float32_t __s1_414 = __p1_414; \
+ float32x4_t __s2_414 = __p2_414; \
+ float32x4_t __rev2_414; __rev2_414 = __builtin_shufflevector(__s2_414, __s2_414, 3, 2, 1, 0); \
+ float32_t __ret_414; \
+ __ret_414 = __noswap_vfmas_laneq_f32(__s0_414, -__s1_414, __rev2_414, __p3_414); \
+ __ret_414; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vfmsq_laneq_f64(__p0_116, __p1_116, __p2_116, __p3_116) __extension__ ({ \
- float64x2_t __s0_116 = __p0_116; \
- float64x2_t __s1_116 = __p1_116; \
- float64x2_t __s2_116 = __p2_116; \
- float64x2_t __ret_116; \
- __ret_116 = vfmaq_laneq_f64(__s0_116, -__s1_116, __s2_116, __p3_116); \
- __ret_116; \
+#define vfmsq_laneq_f64(__p0_415, __p1_415, __p2_415, __p3_415) __extension__ ({ \
+ float64x2_t __s0_415 = __p0_415; \
+ float64x2_t __s1_415 = __p1_415; \
+ float64x2_t __s2_415 = __p2_415; \
+ float64x2_t __ret_415; \
+ __ret_415 = vfmaq_laneq_f64(__s0_415, -__s1_415, __s2_415, __p3_415); \
+ __ret_415; \
})
#else
-#define vfmsq_laneq_f64(__p0_117, __p1_117, __p2_117, __p3_117) __extension__ ({ \
- float64x2_t __s0_117 = __p0_117; \
- float64x2_t __s1_117 = __p1_117; \
- float64x2_t __s2_117 = __p2_117; \
- float64x2_t __rev0_117; __rev0_117 = __builtin_shufflevector(__s0_117, __s0_117, 1, 0); \
- float64x2_t __rev1_117; __rev1_117 = __builtin_shufflevector(__s1_117, __s1_117, 1, 0); \
- float64x2_t __rev2_117; __rev2_117 = __builtin_shufflevector(__s2_117, __s2_117, 1, 0); \
- float64x2_t __ret_117; \
- __ret_117 = __noswap_vfmaq_laneq_f64(__rev0_117, -__rev1_117, __rev2_117, __p3_117); \
- __ret_117 = __builtin_shufflevector(__ret_117, __ret_117, 1, 0); \
- __ret_117; \
+#define vfmsq_laneq_f64(__p0_416, __p1_416, __p2_416, __p3_416) __extension__ ({ \
+ float64x2_t __s0_416 = __p0_416; \
+ float64x2_t __s1_416 = __p1_416; \
+ float64x2_t __s2_416 = __p2_416; \
+ float64x2_t __rev0_416; __rev0_416 = __builtin_shufflevector(__s0_416, __s0_416, 1, 0); \
+ float64x2_t __rev1_416; __rev1_416 = __builtin_shufflevector(__s1_416, __s1_416, 1, 0); \
+ float64x2_t __rev2_416; __rev2_416 = __builtin_shufflevector(__s2_416, __s2_416, 1, 0); \
+ float64x2_t __ret_416; \
+ __ret_416 = __noswap_vfmaq_laneq_f64(__rev0_416, -__rev1_416, __rev2_416, __p3_416); \
+ __ret_416 = __builtin_shufflevector(__ret_416, __ret_416, 1, 0); \
+ __ret_416; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vfmsq_laneq_f32(__p0_118, __p1_118, __p2_118, __p3_118) __extension__ ({ \
- float32x4_t __s0_118 = __p0_118; \
- float32x4_t __s1_118 = __p1_118; \
- float32x4_t __s2_118 = __p2_118; \
- float32x4_t __ret_118; \
- __ret_118 = vfmaq_laneq_f32(__s0_118, -__s1_118, __s2_118, __p3_118); \
- __ret_118; \
+#define vfmsq_laneq_f32(__p0_417, __p1_417, __p2_417, __p3_417) __extension__ ({ \
+ float32x4_t __s0_417 = __p0_417; \
+ float32x4_t __s1_417 = __p1_417; \
+ float32x4_t __s2_417 = __p2_417; \
+ float32x4_t __ret_417; \
+ __ret_417 = vfmaq_laneq_f32(__s0_417, -__s1_417, __s2_417, __p3_417); \
+ __ret_417; \
})
#else
-#define vfmsq_laneq_f32(__p0_119, __p1_119, __p2_119, __p3_119) __extension__ ({ \
- float32x4_t __s0_119 = __p0_119; \
- float32x4_t __s1_119 = __p1_119; \
- float32x4_t __s2_119 = __p2_119; \
- float32x4_t __rev0_119; __rev0_119 = __builtin_shufflevector(__s0_119, __s0_119, 3, 2, 1, 0); \
- float32x4_t __rev1_119; __rev1_119 = __builtin_shufflevector(__s1_119, __s1_119, 3, 2, 1, 0); \
- float32x4_t __rev2_119; __rev2_119 = __builtin_shufflevector(__s2_119, __s2_119, 3, 2, 1, 0); \
- float32x4_t __ret_119; \
- __ret_119 = __noswap_vfmaq_laneq_f32(__rev0_119, -__rev1_119, __rev2_119, __p3_119); \
- __ret_119 = __builtin_shufflevector(__ret_119, __ret_119, 3, 2, 1, 0); \
- __ret_119; \
+#define vfmsq_laneq_f32(__p0_418, __p1_418, __p2_418, __p3_418) __extension__ ({ \
+ float32x4_t __s0_418 = __p0_418; \
+ float32x4_t __s1_418 = __p1_418; \
+ float32x4_t __s2_418 = __p2_418; \
+ float32x4_t __rev0_418; __rev0_418 = __builtin_shufflevector(__s0_418, __s0_418, 3, 2, 1, 0); \
+ float32x4_t __rev1_418; __rev1_418 = __builtin_shufflevector(__s1_418, __s1_418, 3, 2, 1, 0); \
+ float32x4_t __rev2_418; __rev2_418 = __builtin_shufflevector(__s2_418, __s2_418, 3, 2, 1, 0); \
+ float32x4_t __ret_418; \
+ __ret_418 = __noswap_vfmaq_laneq_f32(__rev0_418, -__rev1_418, __rev2_418, __p3_418); \
+ __ret_418 = __builtin_shufflevector(__ret_418, __ret_418, 3, 2, 1, 0); \
+ __ret_418; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vfms_laneq_f64(__p0_120, __p1_120, __p2_120, __p3_120) __extension__ ({ \
- float64x1_t __s0_120 = __p0_120; \
- float64x1_t __s1_120 = __p1_120; \
- float64x2_t __s2_120 = __p2_120; \
- float64x1_t __ret_120; \
- __ret_120 = vfma_laneq_f64(__s0_120, -__s1_120, __s2_120, __p3_120); \
- __ret_120; \
+#define vfms_laneq_f64(__p0_419, __p1_419, __p2_419, __p3_419) __extension__ ({ \
+ float64x1_t __s0_419 = __p0_419; \
+ float64x1_t __s1_419 = __p1_419; \
+ float64x2_t __s2_419 = __p2_419; \
+ float64x1_t __ret_419; \
+ __ret_419 = vfma_laneq_f64(__s0_419, -__s1_419, __s2_419, __p3_419); \
+ __ret_419; \
})
#else
-#define vfms_laneq_f64(__p0_121, __p1_121, __p2_121, __p3_121) __extension__ ({ \
- float64x1_t __s0_121 = __p0_121; \
- float64x1_t __s1_121 = __p1_121; \
- float64x2_t __s2_121 = __p2_121; \
- float64x2_t __rev2_121; __rev2_121 = __builtin_shufflevector(__s2_121, __s2_121, 1, 0); \
- float64x1_t __ret_121; \
- __ret_121 = __noswap_vfma_laneq_f64(__s0_121, -__s1_121, __rev2_121, __p3_121); \
- __ret_121; \
+#define vfms_laneq_f64(__p0_420, __p1_420, __p2_420, __p3_420) __extension__ ({ \
+ float64x1_t __s0_420 = __p0_420; \
+ float64x1_t __s1_420 = __p1_420; \
+ float64x2_t __s2_420 = __p2_420; \
+ float64x2_t __rev2_420; __rev2_420 = __builtin_shufflevector(__s2_420, __s2_420, 1, 0); \
+ float64x1_t __ret_420; \
+ __ret_420 = __noswap_vfma_laneq_f64(__s0_420, -__s1_420, __rev2_420, __p3_420); \
+ __ret_420; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vfms_laneq_f32(__p0_122, __p1_122, __p2_122, __p3_122) __extension__ ({ \
- float32x2_t __s0_122 = __p0_122; \
- float32x2_t __s1_122 = __p1_122; \
- float32x4_t __s2_122 = __p2_122; \
- float32x2_t __ret_122; \
- __ret_122 = vfma_laneq_f32(__s0_122, -__s1_122, __s2_122, __p3_122); \
- __ret_122; \
+#define vfms_laneq_f32(__p0_421, __p1_421, __p2_421, __p3_421) __extension__ ({ \
+ float32x2_t __s0_421 = __p0_421; \
+ float32x2_t __s1_421 = __p1_421; \
+ float32x4_t __s2_421 = __p2_421; \
+ float32x2_t __ret_421; \
+ __ret_421 = vfma_laneq_f32(__s0_421, -__s1_421, __s2_421, __p3_421); \
+ __ret_421; \
})
#else
-#define vfms_laneq_f32(__p0_123, __p1_123, __p2_123, __p3_123) __extension__ ({ \
- float32x2_t __s0_123 = __p0_123; \
- float32x2_t __s1_123 = __p1_123; \
- float32x4_t __s2_123 = __p2_123; \
- float32x2_t __rev0_123; __rev0_123 = __builtin_shufflevector(__s0_123, __s0_123, 1, 0); \
- float32x2_t __rev1_123; __rev1_123 = __builtin_shufflevector(__s1_123, __s1_123, 1, 0); \
- float32x4_t __rev2_123; __rev2_123 = __builtin_shufflevector(__s2_123, __s2_123, 3, 2, 1, 0); \
- float32x2_t __ret_123; \
- __ret_123 = __noswap_vfma_laneq_f32(__rev0_123, -__rev1_123, __rev2_123, __p3_123); \
- __ret_123 = __builtin_shufflevector(__ret_123, __ret_123, 1, 0); \
- __ret_123; \
+#define vfms_laneq_f32(__p0_422, __p1_422, __p2_422, __p3_422) __extension__ ({ \
+ float32x2_t __s0_422 = __p0_422; \
+ float32x2_t __s1_422 = __p1_422; \
+ float32x4_t __s2_422 = __p2_422; \
+ float32x2_t __rev0_422; __rev0_422 = __builtin_shufflevector(__s0_422, __s0_422, 1, 0); \
+ float32x2_t __rev1_422; __rev1_422 = __builtin_shufflevector(__s1_422, __s1_422, 1, 0); \
+ float32x4_t __rev2_422; __rev2_422 = __builtin_shufflevector(__s2_422, __s2_422, 3, 2, 1, 0); \
+ float32x2_t __ret_422; \
+ __ret_422 = __noswap_vfma_laneq_f32(__rev0_422, -__rev1_422, __rev2_422, __p3_422); \
+ __ret_422 = __builtin_shufflevector(__ret_422, __ret_422, 1, 0); \
+ __ret_422; \
})
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai float64x1_t vfms_n_f64(float64x1_t __p0, float64x1_t __p1, float64_t __p2) {
float64x1_t __ret;
__ret = vfma_f64(__p0, -__p1, (float64x1_t) {__p2});
return __ret;
}
-#else
-__ai float64x1_t vfms_n_f64(float64x1_t __p0, float64x1_t __p1, float64_t __p2) {
- float64x1_t __ret;
- __ret = __noswap_vfma_f64(__p0, -__p1, (float64x1_t) {__p2});
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float32x2_t vfms_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
float32x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vget_lane_p64(__p0, __p1) __extension__ ({ \
- poly64x1_t __s0 = __p0; \
- poly64_t __ret; \
- __ret = (poly64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
- __ret; \
-})
-#else
#define vget_lane_p64(__p0, __p1) __extension__ ({ \
poly64x1_t __s0 = __p0; \
poly64_t __ret; \
- __ret = (poly64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
- __ret; \
-})
-#define __noswap_vget_lane_p64(__p0, __p1) __extension__ ({ \
- poly64x1_t __s0 = __p0; \
- poly64_t __ret; \
- __ret = (poly64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
+ __ret = (poly64_t) __builtin_neon_vget_lane_i64((poly64x1_t)__s0, __p1); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vgetq_lane_p64(__p0, __p1) __extension__ ({ \
poly64x2_t __s0 = __p0; \
poly64_t __ret; \
- __ret = (poly64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__s0, __p1); \
+ __ret = (poly64_t) __builtin_neon_vgetq_lane_i64((poly64x2_t)__s0, __p1); \
__ret; \
})
#else
poly64x2_t __s0 = __p0; \
poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
poly64_t __ret; \
- __ret = (poly64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__rev0, __p1); \
+ __ret = (poly64_t) __builtin_neon_vgetq_lane_i64((poly64x2_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vgetq_lane_p64(__p0, __p1) __extension__ ({ \
poly64x2_t __s0 = __p0; \
poly64_t __ret; \
- __ret = (poly64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__s0, __p1); \
+ __ret = (poly64_t) __builtin_neon_vgetq_lane_i64((poly64x2_t)__s0, __p1); \
__ret; \
})
#endif
#define vgetq_lane_f64(__p0, __p1) __extension__ ({ \
float64x2_t __s0 = __p0; \
float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vgetq_lane_f64((int8x16_t)__s0, __p1); \
+ __ret = (float64_t) __builtin_neon_vgetq_lane_f64((float64x2_t)__s0, __p1); \
__ret; \
})
#else
float64x2_t __s0 = __p0; \
float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vgetq_lane_f64((int8x16_t)__rev0, __p1); \
+ __ret = (float64_t) __builtin_neon_vgetq_lane_f64((float64x2_t)__rev0, __p1); \
__ret; \
})
#define __noswap_vgetq_lane_f64(__p0, __p1) __extension__ ({ \
float64x2_t __s0 = __p0; \
float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vgetq_lane_f64((int8x16_t)__s0, __p1); \
+ __ret = (float64_t) __builtin_neon_vgetq_lane_f64((float64x2_t)__s0, __p1); \
__ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vget_lane_f64(__p0, __p1) __extension__ ({ \
- float64x1_t __s0 = __p0; \
- float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vget_lane_f64((int8x8_t)__s0, __p1); \
- __ret; \
-})
-#else
#define vget_lane_f64(__p0, __p1) __extension__ ({ \
float64x1_t __s0 = __p0; \
float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vget_lane_f64((int8x8_t)__s0, __p1); \
- __ret; \
-})
-#define __noswap_vget_lane_f64(__p0, __p1) __extension__ ({ \
- float64x1_t __s0 = __p0; \
- float64_t __ret; \
- __ret = (float64_t) __builtin_neon_vget_lane_f64((int8x8_t)__s0, __p1); \
+ __ret = (float64_t) __builtin_neon_vget_lane_f64((float64x1_t)__s0, __p1); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai poly64x1_t vget_low_p64(poly64x2_t __p0) {
poly64x1_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld1_p64(__p0) __extension__ ({ \
- poly64x1_t __ret; \
- __ret = (poly64x1_t) __builtin_neon_vld1_v(__p0, 6); \
- __ret; \
-})
-#else
#define vld1_p64(__p0) __extension__ ({ \
poly64x1_t __ret; \
__ret = (poly64x1_t) __builtin_neon_vld1_v(__p0, 6); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1q_p64(__p0) __extension__ ({ \
poly64x2_t __ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld1_f64(__p0) __extension__ ({ \
- float64x1_t __ret; \
- __ret = (float64x1_t) __builtin_neon_vld1_v(__p0, 10); \
- __ret; \
-})
-#else
#define vld1_f64(__p0) __extension__ ({ \
float64x1_t __ret; \
__ret = (float64x1_t) __builtin_neon_vld1_v(__p0, 10); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld1_dup_p64(__p0) __extension__ ({ \
- poly64x1_t __ret; \
- __ret = (poly64x1_t) __builtin_neon_vld1_dup_v(__p0, 6); \
- __ret; \
-})
-#else
#define vld1_dup_p64(__p0) __extension__ ({ \
poly64x1_t __ret; \
__ret = (poly64x1_t) __builtin_neon_vld1_dup_v(__p0, 6); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1q_dup_p64(__p0) __extension__ ({ \
poly64x2_t __ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld1_dup_f64(__p0) __extension__ ({ \
- float64x1_t __ret; \
- __ret = (float64x1_t) __builtin_neon_vld1_dup_v(__p0, 10); \
- __ret; \
-})
-#else
#define vld1_dup_f64(__p0) __extension__ ({ \
float64x1_t __ret; \
__ret = (float64x1_t) __builtin_neon_vld1_dup_v(__p0, 10); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
- poly64x1_t __s1 = __p1; \
- poly64x1_t __ret; \
- __ret = (poly64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 6); \
- __ret; \
-})
-#else
#define vld1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x1_t __s1 = __p1; \
poly64x1_t __ret; \
__ret = (poly64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 6); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x2_t __s1 = __p1; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x1_t __s1 = __p1; \
- float64x1_t __ret; \
- __ret = (float64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 10); \
- __ret; \
-})
-#else
#define vld1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x1_t __s1 = __p1; \
float64x1_t __ret; \
__ret = (float64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 10); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld1_p64_x2(__p0) __extension__ ({ \
- poly64x1x2_t __ret; \
- __builtin_neon_vld1_x2_v(&__ret, __p0, 6); \
- __ret; \
-})
-#else
#define vld1_p64_x2(__p0) __extension__ ({ \
poly64x1x2_t __ret; \
__builtin_neon_vld1_x2_v(&__ret, __p0, 6); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1q_p64_x2(__p0) __extension__ ({ \
poly64x2x2_t __ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld1_f64_x2(__p0) __extension__ ({ \
- float64x1x2_t __ret; \
- __builtin_neon_vld1_x2_v(&__ret, __p0, 10); \
- __ret; \
-})
-#else
#define vld1_f64_x2(__p0) __extension__ ({ \
float64x1x2_t __ret; \
__builtin_neon_vld1_x2_v(&__ret, __p0, 10); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld1_p64_x3(__p0) __extension__ ({ \
- poly64x1x3_t __ret; \
- __builtin_neon_vld1_x3_v(&__ret, __p0, 6); \
- __ret; \
-})
-#else
#define vld1_p64_x3(__p0) __extension__ ({ \
poly64x1x3_t __ret; \
__builtin_neon_vld1_x3_v(&__ret, __p0, 6); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld1q_p64_x3(__p0) __extension__ ({ \
poly64x2x3_t __ret; \
- __builtin_neon_vld1q_x3_v(&__ret, __p0, 38); \
- __ret; \
-})
-#else
-#define vld1q_p64_x3(__p0) __extension__ ({ \
- poly64x2x3_t __ret; \
- __builtin_neon_vld1q_x3_v(&__ret, __p0, 38); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld1q_f64_x3(__p0) __extension__ ({ \
- float64x2x3_t __ret; \
- __builtin_neon_vld1q_x3_v(&__ret, __p0, 42); \
- __ret; \
-})
-#else
-#define vld1q_f64_x3(__p0) __extension__ ({ \
- float64x2x3_t __ret; \
- __builtin_neon_vld1q_x3_v(&__ret, __p0, 42); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld1_f64_x3(__p0) __extension__ ({ \
- float64x1x3_t __ret; \
- __builtin_neon_vld1_x3_v(&__ret, __p0, 10); \
- __ret; \
-})
-#else
-#define vld1_f64_x3(__p0) __extension__ ({ \
- float64x1x3_t __ret; \
- __builtin_neon_vld1_x3_v(&__ret, __p0, 10); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld1_p64_x4(__p0) __extension__ ({ \
- poly64x1x4_t __ret; \
- __builtin_neon_vld1_x4_v(&__ret, __p0, 6); \
- __ret; \
-})
-#else
-#define vld1_p64_x4(__p0) __extension__ ({ \
- poly64x1x4_t __ret; \
- __builtin_neon_vld1_x4_v(&__ret, __p0, 6); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld1q_p64_x4(__p0) __extension__ ({ \
- poly64x2x4_t __ret; \
- __builtin_neon_vld1q_x4_v(&__ret, __p0, 38); \
- __ret; \
-})
-#else
-#define vld1q_p64_x4(__p0) __extension__ ({ \
- poly64x2x4_t __ret; \
- __builtin_neon_vld1q_x4_v(&__ret, __p0, 38); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld1q_f64_x4(__p0) __extension__ ({ \
- float64x2x4_t __ret; \
- __builtin_neon_vld1q_x4_v(&__ret, __p0, 42); \
- __ret; \
-})
-#else
-#define vld1q_f64_x4(__p0) __extension__ ({ \
- float64x2x4_t __ret; \
- __builtin_neon_vld1q_x4_v(&__ret, __p0, 42); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld1_f64_x4(__p0) __extension__ ({ \
- float64x1x4_t __ret; \
- __builtin_neon_vld1_x4_v(&__ret, __p0, 10); \
- __ret; \
-})
-#else
-#define vld1_f64_x4(__p0) __extension__ ({ \
- float64x1x4_t __ret; \
- __builtin_neon_vld1_x4_v(&__ret, __p0, 10); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld2_p64(__p0) __extension__ ({ \
- poly64x1x2_t __ret; \
- __builtin_neon_vld2_v(&__ret, __p0, 6); \
- __ret; \
-})
-#else
-#define vld2_p64(__p0) __extension__ ({ \
- poly64x1x2_t __ret; \
- __builtin_neon_vld2_v(&__ret, __p0, 6); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld2q_p64(__p0) __extension__ ({ \
- poly64x2x2_t __ret; \
- __builtin_neon_vld2q_v(&__ret, __p0, 38); \
- __ret; \
-})
-#else
-#define vld2q_p64(__p0) __extension__ ({ \
- poly64x2x2_t __ret; \
- __builtin_neon_vld2q_v(&__ret, __p0, 38); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld2q_u64(__p0) __extension__ ({ \
- uint64x2x2_t __ret; \
- __builtin_neon_vld2q_v(&__ret, __p0, 51); \
- __ret; \
-})
-#else
-#define vld2q_u64(__p0) __extension__ ({ \
- uint64x2x2_t __ret; \
- __builtin_neon_vld2q_v(&__ret, __p0, 51); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld2q_f64(__p0) __extension__ ({ \
- float64x2x2_t __ret; \
- __builtin_neon_vld2q_v(&__ret, __p0, 42); \
- __ret; \
-})
-#else
-#define vld2q_f64(__p0) __extension__ ({ \
- float64x2x2_t __ret; \
- __builtin_neon_vld2q_v(&__ret, __p0, 42); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld2q_s64(__p0) __extension__ ({ \
- int64x2x2_t __ret; \
- __builtin_neon_vld2q_v(&__ret, __p0, 35); \
- __ret; \
-})
-#else
-#define vld2q_s64(__p0) __extension__ ({ \
- int64x2x2_t __ret; \
- __builtin_neon_vld2q_v(&__ret, __p0, 35); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld2_f64(__p0) __extension__ ({ \
- float64x1x2_t __ret; \
- __builtin_neon_vld2_v(&__ret, __p0, 10); \
- __ret; \
-})
-#else
-#define vld2_f64(__p0) __extension__ ({ \
- float64x1x2_t __ret; \
- __builtin_neon_vld2_v(&__ret, __p0, 10); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld2_dup_p64(__p0) __extension__ ({ \
- poly64x1x2_t __ret; \
- __builtin_neon_vld2_dup_v(&__ret, __p0, 6); \
- __ret; \
-})
-#else
-#define vld2_dup_p64(__p0) __extension__ ({ \
- poly64x1x2_t __ret; \
- __builtin_neon_vld2_dup_v(&__ret, __p0, 6); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld2q_dup_p8(__p0) __extension__ ({ \
- poly8x16x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 36); \
- __ret; \
-})
-#else
-#define vld2q_dup_p8(__p0) __extension__ ({ \
- poly8x16x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 36); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld2q_dup_p64(__p0) __extension__ ({ \
- poly64x2x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 38); \
- __ret; \
-})
-#else
-#define vld2q_dup_p64(__p0) __extension__ ({ \
- poly64x2x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 38); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld2q_dup_p16(__p0) __extension__ ({ \
- poly16x8x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 37); \
- __ret; \
-})
-#else
-#define vld2q_dup_p16(__p0) __extension__ ({ \
- poly16x8x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 37); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld2q_dup_u8(__p0) __extension__ ({ \
- uint8x16x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 48); \
- __ret; \
-})
-#else
-#define vld2q_dup_u8(__p0) __extension__ ({ \
- uint8x16x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 48); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld2q_dup_u32(__p0) __extension__ ({ \
- uint32x4x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 50); \
- __ret; \
-})
-#else
-#define vld2q_dup_u32(__p0) __extension__ ({ \
- uint32x4x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 50); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld2q_dup_u64(__p0) __extension__ ({ \
- uint64x2x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 51); \
+ __builtin_neon_vld1q_x3_v(&__ret, __p0, 38); \
__ret; \
})
#else
-#define vld2q_dup_u64(__p0) __extension__ ({ \
- uint64x2x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 51); \
+#define vld1q_p64_x3(__p0) __extension__ ({ \
+ poly64x2x3_t __ret; \
+ __builtin_neon_vld1q_x3_v(&__ret, __p0, 38); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
__ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld2q_dup_u16(__p0) __extension__ ({ \
- uint16x8x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 49); \
+#define vld1q_f64_x3(__p0) __extension__ ({ \
+ float64x2x3_t __ret; \
+ __builtin_neon_vld1q_x3_v(&__ret, __p0, 42); \
__ret; \
})
#else
-#define vld2q_dup_u16(__p0) __extension__ ({ \
- uint16x8x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 49); \
+#define vld1q_f64_x3(__p0) __extension__ ({ \
+ float64x2x3_t __ret; \
+ __builtin_neon_vld1q_x3_v(&__ret, __p0, 42); \
\
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
__ret; \
})
#endif
+#define vld1_f64_x3(__p0) __extension__ ({ \
+ float64x1x3_t __ret; \
+ __builtin_neon_vld1_x3_v(&__ret, __p0, 10); \
+ __ret; \
+})
+#define vld1_p64_x4(__p0) __extension__ ({ \
+ poly64x1x4_t __ret; \
+ __builtin_neon_vld1_x4_v(&__ret, __p0, 6); \
+ __ret; \
+})
#ifdef __LITTLE_ENDIAN__
-#define vld2q_dup_s8(__p0) __extension__ ({ \
- int8x16x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 32); \
+#define vld1q_p64_x4(__p0) __extension__ ({ \
+ poly64x2x4_t __ret; \
+ __builtin_neon_vld1q_x4_v(&__ret, __p0, 38); \
__ret; \
})
#else
-#define vld2q_dup_s8(__p0) __extension__ ({ \
- int8x16x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 32); \
+#define vld1q_p64_x4(__p0) __extension__ ({ \
+ poly64x2x4_t __ret; \
+ __builtin_neon_vld1q_x4_v(&__ret, __p0, 38); \
\
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
__ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld2q_dup_f64(__p0) __extension__ ({ \
- float64x2x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 42); \
+#define vld1q_f64_x4(__p0) __extension__ ({ \
+ float64x2x4_t __ret; \
+ __builtin_neon_vld1q_x4_v(&__ret, __p0, 42); \
__ret; \
})
#else
-#define vld2q_dup_f64(__p0) __extension__ ({ \
- float64x2x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 42); \
+#define vld1q_f64_x4(__p0) __extension__ ({ \
+ float64x2x4_t __ret; \
+ __builtin_neon_vld1q_x4_v(&__ret, __p0, 42); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
+ __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
+ __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
__ret; \
})
#endif
+#define vld1_f64_x4(__p0) __extension__ ({ \
+ float64x1x4_t __ret; \
+ __builtin_neon_vld1_x4_v(&__ret, __p0, 10); \
+ __ret; \
+})
+#define vld2_p64(__p0) __extension__ ({ \
+ poly64x1x2_t __ret; \
+ __builtin_neon_vld2_v(&__ret, __p0, 6); \
+ __ret; \
+})
#ifdef __LITTLE_ENDIAN__
-#define vld2q_dup_f32(__p0) __extension__ ({ \
- float32x4x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 41); \
+#define vld2q_p64(__p0) __extension__ ({ \
+ poly64x2x2_t __ret; \
+ __builtin_neon_vld2q_v(&__ret, __p0, 38); \
__ret; \
})
#else
-#define vld2q_dup_f32(__p0) __extension__ ({ \
- float32x4x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 41); \
+#define vld2q_p64(__p0) __extension__ ({ \
+ poly64x2x2_t __ret; \
+ __builtin_neon_vld2q_v(&__ret, __p0, 38); \
\
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
__ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld2q_dup_f16(__p0) __extension__ ({ \
- float16x8x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 40); \
+#define vld2q_u64(__p0) __extension__ ({ \
+ uint64x2x2_t __ret; \
+ __builtin_neon_vld2q_v(&__ret, __p0, 51); \
__ret; \
})
#else
-#define vld2q_dup_f16(__p0) __extension__ ({ \
- float16x8x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 40); \
+#define vld2q_u64(__p0) __extension__ ({ \
+ uint64x2x2_t __ret; \
+ __builtin_neon_vld2q_v(&__ret, __p0, 51); \
\
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
__ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld2q_dup_s32(__p0) __extension__ ({ \
- int32x4x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 34); \
+#define vld2q_f64(__p0) __extension__ ({ \
+ float64x2x2_t __ret; \
+ __builtin_neon_vld2q_v(&__ret, __p0, 42); \
__ret; \
})
#else
-#define vld2q_dup_s32(__p0) __extension__ ({ \
- int32x4x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 34); \
+#define vld2q_f64(__p0) __extension__ ({ \
+ float64x2x2_t __ret; \
+ __builtin_neon_vld2q_v(&__ret, __p0, 42); \
\
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
__ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld2q_dup_s64(__p0) __extension__ ({ \
+#define vld2q_s64(__p0) __extension__ ({ \
int64x2x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 35); \
+ __builtin_neon_vld2q_v(&__ret, __p0, 35); \
__ret; \
})
#else
-#define vld2q_dup_s64(__p0) __extension__ ({ \
+#define vld2q_s64(__p0) __extension__ ({ \
int64x2x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 35); \
+ __builtin_neon_vld2q_v(&__ret, __p0, 35); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
})
#endif
+#define vld2_f64(__p0) __extension__ ({ \
+ float64x1x2_t __ret; \
+ __builtin_neon_vld2_v(&__ret, __p0, 10); \
+ __ret; \
+})
+#define vld2_dup_p64(__p0) __extension__ ({ \
+ poly64x1x2_t __ret; \
+ __builtin_neon_vld2_dup_v(&__ret, __p0, 6); \
+ __ret; \
+})
#ifdef __LITTLE_ENDIAN__
-#define vld2q_dup_s16(__p0) __extension__ ({ \
- int16x8x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 33); \
+#define vld2q_dup_p64(__p0) __extension__ ({ \
+ poly64x2x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 38); \
__ret; \
})
#else
-#define vld2q_dup_s16(__p0) __extension__ ({ \
- int16x8x2_t __ret; \
- __builtin_neon_vld2q_dup_v(&__ret, __p0, 33); \
+#define vld2q_dup_p64(__p0) __extension__ ({ \
+ poly64x2x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 38); \
\
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
__ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld2_dup_f64(__p0) __extension__ ({ \
- float64x1x2_t __ret; \
- __builtin_neon_vld2_dup_v(&__ret, __p0, 10); \
+#define vld2q_dup_f64(__p0) __extension__ ({ \
+ float64x2x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 42); \
__ret; \
})
#else
-#define vld2_dup_f64(__p0) __extension__ ({ \
- float64x1x2_t __ret; \
- __builtin_neon_vld2_dup_v(&__ret, __p0, 10); \
+#define vld2q_dup_f64(__p0) __extension__ ({ \
+ float64x2x2_t __ret; \
+ __builtin_neon_vld2q_dup_v(&__ret, __p0, 42); \
+ \
+ __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
+ __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
__ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
- poly64x1x2_t __s1 = __p1; \
- poly64x1x2_t __ret; \
- __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 6); \
+#define vld2_dup_f64(__p0) __extension__ ({ \
+ float64x1x2_t __ret; \
+ __builtin_neon_vld2_dup_v(&__ret, __p0, 10); \
__ret; \
})
-#else
#define vld2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x1x2_t __s1 = __p1; \
poly64x1x2_t __ret; \
__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 6); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
poly8x16x2_t __s1 = __p1; \
#define vld2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x2x2_t __s1 = __p1; \
float64x2x2_t __ret; \
- __builtin_neon_vld2q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __p2, 42); \
+ __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 42); \
__ret; \
})
#else
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
float64x2x2_t __ret; \
- __builtin_neon_vld2q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __p2, 42); \
+ __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 42); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
#define vld2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64x2x2_t __s1 = __p1; \
int64x2x2_t __ret; \
- __builtin_neon_vld2q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __p2, 35); \
+ __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 35); \
__ret; \
})
#else
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
int64x2x2_t __ret; \
- __builtin_neon_vld2q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __p2, 35); \
+ __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 35); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64x1x2_t __s1 = __p1; \
- uint64x1x2_t __ret; \
- __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 19); \
- __ret; \
-})
-#else
#define vld2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
uint64x1x2_t __s1 = __p1; \
uint64x1x2_t __ret; \
__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 19); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x1x2_t __s1 = __p1; \
- float64x1x2_t __ret; \
- __builtin_neon_vld2_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __p2, 10); \
- __ret; \
-})
-#else
#define vld2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x1x2_t __s1 = __p1; \
float64x1x2_t __ret; \
- __builtin_neon_vld2_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __p2, 10); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
- int64x1x2_t __s1 = __p1; \
- int64x1x2_t __ret; \
- __builtin_neon_vld2_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __p2, 3); \
+ __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 10); \
__ret; \
})
-#else
#define vld2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64x1x2_t __s1 = __p1; \
int64x1x2_t __ret; \
- __builtin_neon_vld2_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __p2, 3); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3_p64(__p0) __extension__ ({ \
- poly64x1x3_t __ret; \
- __builtin_neon_vld3_v(&__ret, __p0, 6); \
+ __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 3); \
__ret; \
})
-#else
#define vld3_p64(__p0) __extension__ ({ \
poly64x1x3_t __ret; \
__builtin_neon_vld3_v(&__ret, __p0, 6); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld3q_p64(__p0) __extension__ ({ \
poly64x2x3_t __ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld3_f64(__p0) __extension__ ({ \
- float64x1x3_t __ret; \
- __builtin_neon_vld3_v(&__ret, __p0, 10); \
- __ret; \
-})
-#else
#define vld3_f64(__p0) __extension__ ({ \
float64x1x3_t __ret; \
__builtin_neon_vld3_v(&__ret, __p0, 10); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3_dup_p64(__p0) __extension__ ({ \
- poly64x1x3_t __ret; \
- __builtin_neon_vld3_dup_v(&__ret, __p0, 6); \
- __ret; \
-})
-#else
#define vld3_dup_p64(__p0) __extension__ ({ \
poly64x1x3_t __ret; \
__builtin_neon_vld3_dup_v(&__ret, __p0, 6); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3q_dup_p8(__p0) __extension__ ({ \
- poly8x16x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 36); \
- __ret; \
-})
-#else
-#define vld3q_dup_p8(__p0) __extension__ ({ \
- poly8x16x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 36); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld3q_dup_p64(__p0) __extension__ ({ \
poly64x2x3_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld3q_dup_p16(__p0) __extension__ ({ \
- poly16x8x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 37); \
- __ret; \
-})
-#else
-#define vld3q_dup_p16(__p0) __extension__ ({ \
- poly16x8x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 37); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3q_dup_u8(__p0) __extension__ ({ \
- uint8x16x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 48); \
- __ret; \
-})
-#else
-#define vld3q_dup_u8(__p0) __extension__ ({ \
- uint8x16x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 48); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3q_dup_u32(__p0) __extension__ ({ \
- uint32x4x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 50); \
- __ret; \
-})
-#else
-#define vld3q_dup_u32(__p0) __extension__ ({ \
- uint32x4x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 50); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3q_dup_u64(__p0) __extension__ ({ \
- uint64x2x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 51); \
- __ret; \
-})
-#else
-#define vld3q_dup_u64(__p0) __extension__ ({ \
- uint64x2x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 51); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3q_dup_u16(__p0) __extension__ ({ \
- uint16x8x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 49); \
- __ret; \
-})
-#else
-#define vld3q_dup_u16(__p0) __extension__ ({ \
- uint16x8x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 49); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3q_dup_s8(__p0) __extension__ ({ \
- int8x16x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 32); \
- __ret; \
-})
-#else
-#define vld3q_dup_s8(__p0) __extension__ ({ \
- int8x16x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 32); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld3q_dup_f64(__p0) __extension__ ({ \
float64x2x3_t __ret; \
__builtin_neon_vld3q_dup_v(&__ret, __p0, 42); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld3q_dup_f32(__p0) __extension__ ({ \
- float32x4x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 41); \
- __ret; \
-})
-#else
-#define vld3q_dup_f32(__p0) __extension__ ({ \
- float32x4x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 41); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3q_dup_f16(__p0) __extension__ ({ \
- float16x8x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 40); \
- __ret; \
-})
-#else
-#define vld3q_dup_f16(__p0) __extension__ ({ \
- float16x8x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 40); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3q_dup_s32(__p0) __extension__ ({ \
- int32x4x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 34); \
- __ret; \
-})
-#else
-#define vld3q_dup_s32(__p0) __extension__ ({ \
- int32x4x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 34); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3q_dup_s64(__p0) __extension__ ({ \
- int64x2x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 35); \
- __ret; \
-})
-#else
-#define vld3q_dup_s64(__p0) __extension__ ({ \
- int64x2x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 35); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3q_dup_s16(__p0) __extension__ ({ \
- int16x8x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 33); \
- __ret; \
-})
-#else
-#define vld3q_dup_s16(__p0) __extension__ ({ \
- int16x8x3_t __ret; \
- __builtin_neon_vld3q_dup_v(&__ret, __p0, 33); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3_dup_f64(__p0) __extension__ ({ \
- float64x1x3_t __ret; \
- __builtin_neon_vld3_dup_v(&__ret, __p0, 10); \
- __ret; \
-})
-#else
#define vld3_dup_f64(__p0) __extension__ ({ \
float64x1x3_t __ret; \
__builtin_neon_vld3_dup_v(&__ret, __p0, 10); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
- poly64x1x3_t __s1 = __p1; \
- poly64x1x3_t __ret; \
- __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 6); \
- __ret; \
-})
-#else
#define vld3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x1x3_t __s1 = __p1; \
poly64x1x3_t __ret; \
__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 6); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
poly8x16x3_t __s1 = __p1; \
#define vld3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x2x3_t __s1 = __p1; \
float64x2x3_t __ret; \
- __builtin_neon_vld3q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 42); \
+ __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 42); \
__ret; \
})
#else
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
float64x2x3_t __ret; \
- __builtin_neon_vld3q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 42); \
+ __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 42); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
#define vld3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64x2x3_t __s1 = __p1; \
int64x2x3_t __ret; \
- __builtin_neon_vld3q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 35); \
+ __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 35); \
__ret; \
})
#else
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
int64x2x3_t __ret; \
- __builtin_neon_vld3q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 35); \
+ __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 35); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64x1x3_t __s1 = __p1; \
- uint64x1x3_t __ret; \
- __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 19); \
- __ret; \
-})
-#else
#define vld3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
uint64x1x3_t __s1 = __p1; \
uint64x1x3_t __ret; \
__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 19); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x1x3_t __s1 = __p1; \
- float64x1x3_t __ret; \
- __builtin_neon_vld3_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 10); \
- __ret; \
-})
-#else
#define vld3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x1x3_t __s1 = __p1; \
float64x1x3_t __ret; \
- __builtin_neon_vld3_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 10); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
- int64x1x3_t __s1 = __p1; \
- int64x1x3_t __ret; \
- __builtin_neon_vld3_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 3); \
+ __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 10); \
__ret; \
})
-#else
#define vld3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64x1x3_t __s1 = __p1; \
int64x1x3_t __ret; \
- __builtin_neon_vld3_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 3); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld4_p64(__p0) __extension__ ({ \
- poly64x1x4_t __ret; \
- __builtin_neon_vld4_v(&__ret, __p0, 6); \
+ __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 3); \
__ret; \
})
-#else
#define vld4_p64(__p0) __extension__ ({ \
poly64x1x4_t __ret; \
__builtin_neon_vld4_v(&__ret, __p0, 6); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld4q_p64(__p0) __extension__ ({ \
poly64x2x4_t __ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vld4_f64(__p0) __extension__ ({ \
float64x1x4_t __ret; \
__builtin_neon_vld4_v(&__ret, __p0, 10); \
__ret; \
})
-#else
-#define vld4_f64(__p0) __extension__ ({ \
- float64x1x4_t __ret; \
- __builtin_neon_vld4_v(&__ret, __p0, 10); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld4_dup_p64(__p0) __extension__ ({ \
- poly64x1x4_t __ret; \
- __builtin_neon_vld4_dup_v(&__ret, __p0, 6); \
- __ret; \
-})
-#else
#define vld4_dup_p64(__p0) __extension__ ({ \
poly64x1x4_t __ret; \
__builtin_neon_vld4_dup_v(&__ret, __p0, 6); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld4q_dup_p8(__p0) __extension__ ({ \
- poly8x16x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 36); \
- __ret; \
-})
-#else
-#define vld4q_dup_p8(__p0) __extension__ ({ \
- poly8x16x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 36); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld4q_dup_p64(__p0) __extension__ ({ \
poly64x2x4_t __ret; \
#endif
#ifdef __LITTLE_ENDIAN__
-#define vld4q_dup_p16(__p0) __extension__ ({ \
- poly16x8x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 37); \
- __ret; \
-})
-#else
-#define vld4q_dup_p16(__p0) __extension__ ({ \
- poly16x8x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 37); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld4q_dup_u8(__p0) __extension__ ({ \
- uint8x16x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 48); \
- __ret; \
-})
-#else
-#define vld4q_dup_u8(__p0) __extension__ ({ \
- uint8x16x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 48); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld4q_dup_u32(__p0) __extension__ ({ \
- uint32x4x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 50); \
- __ret; \
-})
-#else
-#define vld4q_dup_u32(__p0) __extension__ ({ \
- uint32x4x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 50); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld4q_dup_u64(__p0) __extension__ ({ \
- uint64x2x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 51); \
- __ret; \
-})
-#else
-#define vld4q_dup_u64(__p0) __extension__ ({ \
- uint64x2x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 51); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld4q_dup_u16(__p0) __extension__ ({ \
- uint16x8x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 49); \
- __ret; \
-})
-#else
-#define vld4q_dup_u16(__p0) __extension__ ({ \
- uint16x8x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 49); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld4q_dup_s8(__p0) __extension__ ({ \
- int8x16x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 32); \
- __ret; \
-})
-#else
-#define vld4q_dup_s8(__p0) __extension__ ({ \
- int8x16x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 32); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld4q_dup_f64(__p0) __extension__ ({ \
float64x2x4_t __ret; \
__builtin_neon_vld4q_dup_v(&__ret, __p0, 42); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld4q_dup_f32(__p0) __extension__ ({ \
- float32x4x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 41); \
- __ret; \
-})
-#else
-#define vld4q_dup_f32(__p0) __extension__ ({ \
- float32x4x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 41); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld4q_dup_f16(__p0) __extension__ ({ \
- float16x8x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 40); \
- __ret; \
-})
-#else
-#define vld4q_dup_f16(__p0) __extension__ ({ \
- float16x8x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 40); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld4q_dup_s32(__p0) __extension__ ({ \
- int32x4x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 34); \
- __ret; \
-})
-#else
-#define vld4q_dup_s32(__p0) __extension__ ({ \
- int32x4x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 34); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld4q_dup_s64(__p0) __extension__ ({ \
- int64x2x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 35); \
- __ret; \
-})
-#else
-#define vld4q_dup_s64(__p0) __extension__ ({ \
- int64x2x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 35); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld4q_dup_s16(__p0) __extension__ ({ \
- int16x8x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 33); \
- __ret; \
-})
-#else
-#define vld4q_dup_s16(__p0) __extension__ ({ \
- int16x8x4_t __ret; \
- __builtin_neon_vld4q_dup_v(&__ret, __p0, 33); \
- \
- __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld4_dup_f64(__p0) __extension__ ({ \
float64x1x4_t __ret; \
__builtin_neon_vld4_dup_v(&__ret, __p0, 10); \
__ret; \
})
-#else
-#define vld4_dup_f64(__p0) __extension__ ({ \
- float64x1x4_t __ret; \
- __builtin_neon_vld4_dup_v(&__ret, __p0, 10); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
- poly64x1x4_t __s1 = __p1; \
- poly64x1x4_t __ret; \
- __builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 6); \
- __ret; \
-})
-#else
#define vld4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x1x4_t __s1 = __p1; \
poly64x1x4_t __ret; \
__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 6); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vld4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
poly8x16x4_t __s1 = __p1; \
#define vld4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x2x4_t __s1 = __p1; \
float64x2x4_t __ret; \
- __builtin_neon_vld4q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 42); \
+ __builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 42); \
__ret; \
})
#else
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
float64x2x4_t __ret; \
- __builtin_neon_vld4q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 42); \
+ __builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 42); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
#define vld4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64x2x4_t __s1 = __p1; \
int64x2x4_t __ret; \
- __builtin_neon_vld4q_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 35); \
+ __builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 35); \
__ret; \
})
#else
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
int64x2x4_t __ret; \
- __builtin_neon_vld4q_lane_v(&__ret, __p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 35); \
+ __builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 35); \
\
__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vld4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64x1x4_t __s1 = __p1; \
- uint64x1x4_t __ret; \
- __builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 19); \
- __ret; \
-})
-#else
#define vld4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
uint64x1x4_t __s1 = __p1; \
uint64x1x4_t __ret; \
__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 19); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vld4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x1x4_t __s1 = __p1; \
float64x1x4_t __ret; \
- __builtin_neon_vld4_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 10); \
+ __builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 10); \
__ret; \
})
-#else
-#define vld4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x1x4_t __s1 = __p1; \
- float64x1x4_t __ret; \
- __builtin_neon_vld4_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 10); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vld4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
- int64x1x4_t __s1 = __p1; \
- int64x1x4_t __ret; \
- __builtin_neon_vld4_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 3); \
- __ret; \
-})
-#else
#define vld4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64x1x4_t __s1 = __p1; \
int64x1x4_t __ret; \
- __builtin_neon_vld4_lane_v(&__ret, __p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 3); \
+ __builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 3); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vldrq_p128(__p0) __extension__ ({ \
- poly128_t __ret; \
- __ret = (poly128_t) __builtin_neon_vldrq_p128(__p0); \
- __ret; \
-})
-#else
#define vldrq_p128(__p0) __extension__ ({ \
poly128_t __ret; \
__ret = (poly128_t) __builtin_neon_vldrq_p128(__p0); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float64x2_t vmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
float64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai float64x1_t vmax_f64(float64x1_t __p0, float64x1_t __p1) {
float64x1_t __ret;
__ret = (float64x1_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
return __ret;
}
-#else
-__ai float64x1_t vmax_f64(float64x1_t __p0, float64x1_t __p1) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float64_t vmaxnmvq_f64(float64x2_t __p0) {
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vmaxnmvq_f64((int8x16_t)__p0);
+ __ret = (float64_t) __builtin_neon_vmaxnmvq_f64(__p0);
return __ret;
}
#else
__ai float64_t vmaxnmvq_f64(float64x2_t __p0) {
float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vmaxnmvq_f64((int8x16_t)__rev0);
+ __ret = (float64_t) __builtin_neon_vmaxnmvq_f64(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float32_t vmaxnmvq_f32(float32x4_t __p0) {
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vmaxnmvq_f32((int8x16_t)__p0);
+ __ret = (float32_t) __builtin_neon_vmaxnmvq_f32(__p0);
return __ret;
}
#else
__ai float32_t vmaxnmvq_f32(float32x4_t __p0) {
float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vmaxnmvq_f32((int8x16_t)__rev0);
+ __ret = (float32_t) __builtin_neon_vmaxnmvq_f32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float32_t vmaxnmv_f32(float32x2_t __p0) {
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vmaxnmv_f32((int8x8_t)__p0);
+ __ret = (float32_t) __builtin_neon_vmaxnmv_f32(__p0);
return __ret;
}
#else
__ai float32_t vmaxnmv_f32(float32x2_t __p0) {
float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vmaxnmv_f32((int8x8_t)__rev0);
+ __ret = (float32_t) __builtin_neon_vmaxnmv_f32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint8_t vmaxvq_u8(uint8x16_t __p0) {
uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vmaxvq_u8((int8x16_t)__p0);
+ __ret = (uint8_t) __builtin_neon_vmaxvq_u8(__p0);
return __ret;
}
#else
__ai uint8_t vmaxvq_u8(uint8x16_t __p0) {
uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vmaxvq_u8((int8x16_t)__rev0);
+ __ret = (uint8_t) __builtin_neon_vmaxvq_u8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint32_t vmaxvq_u32(uint32x4_t __p0) {
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vmaxvq_u32((int8x16_t)__p0);
+ __ret = (uint32_t) __builtin_neon_vmaxvq_u32(__p0);
return __ret;
}
#else
__ai uint32_t vmaxvq_u32(uint32x4_t __p0) {
uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vmaxvq_u32((int8x16_t)__rev0);
+ __ret = (uint32_t) __builtin_neon_vmaxvq_u32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint16_t vmaxvq_u16(uint16x8_t __p0) {
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vmaxvq_u16((int8x16_t)__p0);
+ __ret = (uint16_t) __builtin_neon_vmaxvq_u16(__p0);
return __ret;
}
#else
__ai uint16_t vmaxvq_u16(uint16x8_t __p0) {
uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vmaxvq_u16((int8x16_t)__rev0);
+ __ret = (uint16_t) __builtin_neon_vmaxvq_u16(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int8_t vmaxvq_s8(int8x16_t __p0) {
int8_t __ret;
- __ret = (int8_t) __builtin_neon_vmaxvq_s8((int8x16_t)__p0);
+ __ret = (int8_t) __builtin_neon_vmaxvq_s8(__p0);
return __ret;
}
#else
__ai int8_t vmaxvq_s8(int8x16_t __p0) {
int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8_t __ret;
- __ret = (int8_t) __builtin_neon_vmaxvq_s8((int8x16_t)__rev0);
+ __ret = (int8_t) __builtin_neon_vmaxvq_s8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float64_t vmaxvq_f64(float64x2_t __p0) {
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vmaxvq_f64((int8x16_t)__p0);
+ __ret = (float64_t) __builtin_neon_vmaxvq_f64(__p0);
return __ret;
}
#else
__ai float64_t vmaxvq_f64(float64x2_t __p0) {
float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vmaxvq_f64((int8x16_t)__rev0);
+ __ret = (float64_t) __builtin_neon_vmaxvq_f64(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float32_t vmaxvq_f32(float32x4_t __p0) {
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vmaxvq_f32((int8x16_t)__p0);
+ __ret = (float32_t) __builtin_neon_vmaxvq_f32(__p0);
return __ret;
}
#else
__ai float32_t vmaxvq_f32(float32x4_t __p0) {
float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vmaxvq_f32((int8x16_t)__rev0);
+ __ret = (float32_t) __builtin_neon_vmaxvq_f32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int32_t vmaxvq_s32(int32x4_t __p0) {
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vmaxvq_s32((int8x16_t)__p0);
+ __ret = (int32_t) __builtin_neon_vmaxvq_s32(__p0);
return __ret;
}
#else
__ai int32_t vmaxvq_s32(int32x4_t __p0) {
int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vmaxvq_s32((int8x16_t)__rev0);
+ __ret = (int32_t) __builtin_neon_vmaxvq_s32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int16_t vmaxvq_s16(int16x8_t __p0) {
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vmaxvq_s16((int8x16_t)__p0);
+ __ret = (int16_t) __builtin_neon_vmaxvq_s16(__p0);
return __ret;
}
#else
__ai int16_t vmaxvq_s16(int16x8_t __p0) {
int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vmaxvq_s16((int8x16_t)__rev0);
+ __ret = (int16_t) __builtin_neon_vmaxvq_s16(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint8_t vmaxv_u8(uint8x8_t __p0) {
uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vmaxv_u8((int8x8_t)__p0);
+ __ret = (uint8_t) __builtin_neon_vmaxv_u8(__p0);
return __ret;
}
#else
__ai uint8_t vmaxv_u8(uint8x8_t __p0) {
uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vmaxv_u8((int8x8_t)__rev0);
+ __ret = (uint8_t) __builtin_neon_vmaxv_u8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint32_t vmaxv_u32(uint32x2_t __p0) {
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vmaxv_u32((int8x8_t)__p0);
+ __ret = (uint32_t) __builtin_neon_vmaxv_u32(__p0);
return __ret;
}
#else
__ai uint32_t vmaxv_u32(uint32x2_t __p0) {
uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vmaxv_u32((int8x8_t)__rev0);
+ __ret = (uint32_t) __builtin_neon_vmaxv_u32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint16_t vmaxv_u16(uint16x4_t __p0) {
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vmaxv_u16((int8x8_t)__p0);
+ __ret = (uint16_t) __builtin_neon_vmaxv_u16(__p0);
return __ret;
}
#else
__ai uint16_t vmaxv_u16(uint16x4_t __p0) {
uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vmaxv_u16((int8x8_t)__rev0);
+ __ret = (uint16_t) __builtin_neon_vmaxv_u16(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int8_t vmaxv_s8(int8x8_t __p0) {
int8_t __ret;
- __ret = (int8_t) __builtin_neon_vmaxv_s8((int8x8_t)__p0);
+ __ret = (int8_t) __builtin_neon_vmaxv_s8(__p0);
return __ret;
}
#else
__ai int8_t vmaxv_s8(int8x8_t __p0) {
int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
int8_t __ret;
- __ret = (int8_t) __builtin_neon_vmaxv_s8((int8x8_t)__rev0);
+ __ret = (int8_t) __builtin_neon_vmaxv_s8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float32_t vmaxv_f32(float32x2_t __p0) {
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vmaxv_f32((int8x8_t)__p0);
+ __ret = (float32_t) __builtin_neon_vmaxv_f32(__p0);
return __ret;
}
#else
__ai float32_t vmaxv_f32(float32x2_t __p0) {
float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vmaxv_f32((int8x8_t)__rev0);
+ __ret = (float32_t) __builtin_neon_vmaxv_f32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int32_t vmaxv_s32(int32x2_t __p0) {
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vmaxv_s32((int8x8_t)__p0);
+ __ret = (int32_t) __builtin_neon_vmaxv_s32(__p0);
return __ret;
}
#else
__ai int32_t vmaxv_s32(int32x2_t __p0) {
int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vmaxv_s32((int8x8_t)__rev0);
+ __ret = (int32_t) __builtin_neon_vmaxv_s32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int16_t vmaxv_s16(int16x4_t __p0) {
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vmaxv_s16((int8x8_t)__p0);
+ __ret = (int16_t) __builtin_neon_vmaxv_s16(__p0);
return __ret;
}
#else
__ai int16_t vmaxv_s16(int16x4_t __p0) {
int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vmaxv_s16((int8x8_t)__rev0);
+ __ret = (int16_t) __builtin_neon_vmaxv_s16(__rev0);
return __ret;
}
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai float64x1_t vmin_f64(float64x1_t __p0, float64x1_t __p1) {
float64x1_t __ret;
__ret = (float64x1_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
return __ret;
}
-#else
-__ai float64x1_t vmin_f64(float64x1_t __p0, float64x1_t __p1) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float64_t vminnmvq_f64(float64x2_t __p0) {
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vminnmvq_f64((int8x16_t)__p0);
+ __ret = (float64_t) __builtin_neon_vminnmvq_f64(__p0);
return __ret;
}
#else
__ai float64_t vminnmvq_f64(float64x2_t __p0) {
float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vminnmvq_f64((int8x16_t)__rev0);
+ __ret = (float64_t) __builtin_neon_vminnmvq_f64(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float32_t vminnmvq_f32(float32x4_t __p0) {
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vminnmvq_f32((int8x16_t)__p0);
+ __ret = (float32_t) __builtin_neon_vminnmvq_f32(__p0);
return __ret;
}
#else
__ai float32_t vminnmvq_f32(float32x4_t __p0) {
float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vminnmvq_f32((int8x16_t)__rev0);
+ __ret = (float32_t) __builtin_neon_vminnmvq_f32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float32_t vminnmv_f32(float32x2_t __p0) {
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vminnmv_f32((int8x8_t)__p0);
+ __ret = (float32_t) __builtin_neon_vminnmv_f32(__p0);
return __ret;
}
#else
__ai float32_t vminnmv_f32(float32x2_t __p0) {
float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vminnmv_f32((int8x8_t)__rev0);
+ __ret = (float32_t) __builtin_neon_vminnmv_f32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint8_t vminvq_u8(uint8x16_t __p0) {
uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vminvq_u8((int8x16_t)__p0);
+ __ret = (uint8_t) __builtin_neon_vminvq_u8(__p0);
return __ret;
}
#else
__ai uint8_t vminvq_u8(uint8x16_t __p0) {
uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vminvq_u8((int8x16_t)__rev0);
+ __ret = (uint8_t) __builtin_neon_vminvq_u8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint32_t vminvq_u32(uint32x4_t __p0) {
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vminvq_u32((int8x16_t)__p0);
+ __ret = (uint32_t) __builtin_neon_vminvq_u32(__p0);
return __ret;
}
#else
__ai uint32_t vminvq_u32(uint32x4_t __p0) {
uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vminvq_u32((int8x16_t)__rev0);
+ __ret = (uint32_t) __builtin_neon_vminvq_u32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint16_t vminvq_u16(uint16x8_t __p0) {
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vminvq_u16((int8x16_t)__p0);
+ __ret = (uint16_t) __builtin_neon_vminvq_u16(__p0);
return __ret;
}
#else
__ai uint16_t vminvq_u16(uint16x8_t __p0) {
uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vminvq_u16((int8x16_t)__rev0);
+ __ret = (uint16_t) __builtin_neon_vminvq_u16(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int8_t vminvq_s8(int8x16_t __p0) {
int8_t __ret;
- __ret = (int8_t) __builtin_neon_vminvq_s8((int8x16_t)__p0);
+ __ret = (int8_t) __builtin_neon_vminvq_s8(__p0);
return __ret;
}
#else
__ai int8_t vminvq_s8(int8x16_t __p0) {
int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8_t __ret;
- __ret = (int8_t) __builtin_neon_vminvq_s8((int8x16_t)__rev0);
+ __ret = (int8_t) __builtin_neon_vminvq_s8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float64_t vminvq_f64(float64x2_t __p0) {
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vminvq_f64((int8x16_t)__p0);
+ __ret = (float64_t) __builtin_neon_vminvq_f64(__p0);
return __ret;
}
#else
__ai float64_t vminvq_f64(float64x2_t __p0) {
float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vminvq_f64((int8x16_t)__rev0);
+ __ret = (float64_t) __builtin_neon_vminvq_f64(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float32_t vminvq_f32(float32x4_t __p0) {
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vminvq_f32((int8x16_t)__p0);
+ __ret = (float32_t) __builtin_neon_vminvq_f32(__p0);
return __ret;
}
#else
__ai float32_t vminvq_f32(float32x4_t __p0) {
float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vminvq_f32((int8x16_t)__rev0);
+ __ret = (float32_t) __builtin_neon_vminvq_f32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int32_t vminvq_s32(int32x4_t __p0) {
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vminvq_s32((int8x16_t)__p0);
+ __ret = (int32_t) __builtin_neon_vminvq_s32(__p0);
return __ret;
}
#else
__ai int32_t vminvq_s32(int32x4_t __p0) {
int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vminvq_s32((int8x16_t)__rev0);
+ __ret = (int32_t) __builtin_neon_vminvq_s32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int16_t vminvq_s16(int16x8_t __p0) {
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vminvq_s16((int8x16_t)__p0);
+ __ret = (int16_t) __builtin_neon_vminvq_s16(__p0);
return __ret;
}
#else
__ai int16_t vminvq_s16(int16x8_t __p0) {
int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vminvq_s16((int8x16_t)__rev0);
+ __ret = (int16_t) __builtin_neon_vminvq_s16(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint8_t vminv_u8(uint8x8_t __p0) {
uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vminv_u8((int8x8_t)__p0);
+ __ret = (uint8_t) __builtin_neon_vminv_u8(__p0);
return __ret;
}
#else
__ai uint8_t vminv_u8(uint8x8_t __p0) {
uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vminv_u8((int8x8_t)__rev0);
+ __ret = (uint8_t) __builtin_neon_vminv_u8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint32_t vminv_u32(uint32x2_t __p0) {
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vminv_u32((int8x8_t)__p0);
+ __ret = (uint32_t) __builtin_neon_vminv_u32(__p0);
return __ret;
}
#else
__ai uint32_t vminv_u32(uint32x2_t __p0) {
uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vminv_u32((int8x8_t)__rev0);
+ __ret = (uint32_t) __builtin_neon_vminv_u32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai uint16_t vminv_u16(uint16x4_t __p0) {
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vminv_u16((int8x8_t)__p0);
+ __ret = (uint16_t) __builtin_neon_vminv_u16(__p0);
return __ret;
}
#else
__ai uint16_t vminv_u16(uint16x4_t __p0) {
uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vminv_u16((int8x8_t)__rev0);
+ __ret = (uint16_t) __builtin_neon_vminv_u16(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int8_t vminv_s8(int8x8_t __p0) {
int8_t __ret;
- __ret = (int8_t) __builtin_neon_vminv_s8((int8x8_t)__p0);
+ __ret = (int8_t) __builtin_neon_vminv_s8(__p0);
return __ret;
}
#else
__ai int8_t vminv_s8(int8x8_t __p0) {
int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
int8_t __ret;
- __ret = (int8_t) __builtin_neon_vminv_s8((int8x8_t)__rev0);
+ __ret = (int8_t) __builtin_neon_vminv_s8(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float32_t vminv_f32(float32x2_t __p0) {
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vminv_f32((int8x8_t)__p0);
+ __ret = (float32_t) __builtin_neon_vminv_f32(__p0);
return __ret;
}
#else
__ai float32_t vminv_f32(float32x2_t __p0) {
float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vminv_f32((int8x8_t)__rev0);
+ __ret = (float32_t) __builtin_neon_vminv_f32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int32_t vminv_s32(int32x2_t __p0) {
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vminv_s32((int8x8_t)__p0);
+ __ret = (int32_t) __builtin_neon_vminv_s32(__p0);
return __ret;
}
#else
__ai int32_t vminv_s32(int32x2_t __p0) {
int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
int32_t __ret;
- __ret = (int32_t) __builtin_neon_vminv_s32((int8x8_t)__rev0);
+ __ret = (int32_t) __builtin_neon_vminv_s32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int16_t vminv_s16(int16x4_t __p0) {
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vminv_s16((int8x8_t)__p0);
+ __ret = (int16_t) __builtin_neon_vminv_s16(__p0);
return __ret;
}
#else
__ai int16_t vminv_s16(int16x4_t __p0) {
int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
int16_t __ret;
- __ret = (int16_t) __builtin_neon_vminv_s16((int8x8_t)__rev0);
+ __ret = (int16_t) __builtin_neon_vminv_s16(__rev0);
return __ret;
}
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
float64x1_t __ret;
__ret = __p0 + __p1 * __p2;
return __ret;
}
-#else
-__ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
- float64x1_t __ret;
- __ret = __p0 + __p1 * __p2;
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vmlaq_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint32x4_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlaq_laneq_u32(__p0_423, __p1_423, __p2_423, __p3_423) __extension__ ({ \
+ uint32x4_t __s0_423 = __p0_423; \
+ uint32x4_t __s1_423 = __p1_423; \
+ uint32x4_t __s2_423 = __p2_423; \
+ uint32x4_t __ret_423; \
+ __ret_423 = __s0_423 + __s1_423 * splatq_laneq_u32(__s2_423, __p3_423); \
+ __ret_423; \
})
#else
-#define vmlaq_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlaq_laneq_u32(__p0_424, __p1_424, __p2_424, __p3_424) __extension__ ({ \
+ uint32x4_t __s0_424 = __p0_424; \
+ uint32x4_t __s1_424 = __p1_424; \
+ uint32x4_t __s2_424 = __p2_424; \
+ uint32x4_t __rev0_424; __rev0_424 = __builtin_shufflevector(__s0_424, __s0_424, 3, 2, 1, 0); \
+ uint32x4_t __rev1_424; __rev1_424 = __builtin_shufflevector(__s1_424, __s1_424, 3, 2, 1, 0); \
+ uint32x4_t __rev2_424; __rev2_424 = __builtin_shufflevector(__s2_424, __s2_424, 3, 2, 1, 0); \
+ uint32x4_t __ret_424; \
+ __ret_424 = __rev0_424 + __rev1_424 * __noswap_splatq_laneq_u32(__rev2_424, __p3_424); \
+ __ret_424 = __builtin_shufflevector(__ret_424, __ret_424, 3, 2, 1, 0); \
+ __ret_424; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlaq_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint16x8_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlaq_laneq_u16(__p0_425, __p1_425, __p2_425, __p3_425) __extension__ ({ \
+ uint16x8_t __s0_425 = __p0_425; \
+ uint16x8_t __s1_425 = __p1_425; \
+ uint16x8_t __s2_425 = __p2_425; \
+ uint16x8_t __ret_425; \
+ __ret_425 = __s0_425 + __s1_425 * splatq_laneq_u16(__s2_425, __p3_425); \
+ __ret_425; \
})
#else
-#define vmlaq_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmlaq_laneq_u16(__p0_426, __p1_426, __p2_426, __p3_426) __extension__ ({ \
+ uint16x8_t __s0_426 = __p0_426; \
+ uint16x8_t __s1_426 = __p1_426; \
+ uint16x8_t __s2_426 = __p2_426; \
+ uint16x8_t __rev0_426; __rev0_426 = __builtin_shufflevector(__s0_426, __s0_426, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __rev1_426; __rev1_426 = __builtin_shufflevector(__s1_426, __s1_426, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __rev2_426; __rev2_426 = __builtin_shufflevector(__s2_426, __s2_426, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __ret_426; \
+ __ret_426 = __rev0_426 + __rev1_426 * __noswap_splatq_laneq_u16(__rev2_426, __p3_426); \
+ __ret_426 = __builtin_shufflevector(__ret_426, __ret_426, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_426; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x4_t __s2 = __p2; \
- float32x4_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlaq_laneq_f32(__p0_427, __p1_427, __p2_427, __p3_427) __extension__ ({ \
+ float32x4_t __s0_427 = __p0_427; \
+ float32x4_t __s1_427 = __p1_427; \
+ float32x4_t __s2_427 = __p2_427; \
+ float32x4_t __ret_427; \
+ __ret_427 = __s0_427 + __s1_427 * splatq_laneq_f32(__s2_427, __p3_427); \
+ __ret_427; \
})
#else
-#define vmlaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x4_t __s2 = __p2; \
- float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- float32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- float32x4_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlaq_laneq_f32(__p0_428, __p1_428, __p2_428, __p3_428) __extension__ ({ \
+ float32x4_t __s0_428 = __p0_428; \
+ float32x4_t __s1_428 = __p1_428; \
+ float32x4_t __s2_428 = __p2_428; \
+ float32x4_t __rev0_428; __rev0_428 = __builtin_shufflevector(__s0_428, __s0_428, 3, 2, 1, 0); \
+ float32x4_t __rev1_428; __rev1_428 = __builtin_shufflevector(__s1_428, __s1_428, 3, 2, 1, 0); \
+ float32x4_t __rev2_428; __rev2_428 = __builtin_shufflevector(__s2_428, __s2_428, 3, 2, 1, 0); \
+ float32x4_t __ret_428; \
+ __ret_428 = __rev0_428 + __rev1_428 * __noswap_splatq_laneq_f32(__rev2_428, __p3_428); \
+ __ret_428 = __builtin_shufflevector(__ret_428, __ret_428, 3, 2, 1, 0); \
+ __ret_428; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlaq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlaq_laneq_s32(__p0_429, __p1_429, __p2_429, __p3_429) __extension__ ({ \
+ int32x4_t __s0_429 = __p0_429; \
+ int32x4_t __s1_429 = __p1_429; \
+ int32x4_t __s2_429 = __p2_429; \
+ int32x4_t __ret_429; \
+ __ret_429 = __s0_429 + __s1_429 * splatq_laneq_s32(__s2_429, __p3_429); \
+ __ret_429; \
})
#else
-#define vmlaq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlaq_laneq_s32(__p0_430, __p1_430, __p2_430, __p3_430) __extension__ ({ \
+ int32x4_t __s0_430 = __p0_430; \
+ int32x4_t __s1_430 = __p1_430; \
+ int32x4_t __s2_430 = __p2_430; \
+ int32x4_t __rev0_430; __rev0_430 = __builtin_shufflevector(__s0_430, __s0_430, 3, 2, 1, 0); \
+ int32x4_t __rev1_430; __rev1_430 = __builtin_shufflevector(__s1_430, __s1_430, 3, 2, 1, 0); \
+ int32x4_t __rev2_430; __rev2_430 = __builtin_shufflevector(__s2_430, __s2_430, 3, 2, 1, 0); \
+ int32x4_t __ret_430; \
+ __ret_430 = __rev0_430 + __rev1_430 * __noswap_splatq_laneq_s32(__rev2_430, __p3_430); \
+ __ret_430 = __builtin_shufflevector(__ret_430, __ret_430, 3, 2, 1, 0); \
+ __ret_430; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlaq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x8_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlaq_laneq_s16(__p0_431, __p1_431, __p2_431, __p3_431) __extension__ ({ \
+ int16x8_t __s0_431 = __p0_431; \
+ int16x8_t __s1_431 = __p1_431; \
+ int16x8_t __s2_431 = __p2_431; \
+ int16x8_t __ret_431; \
+ __ret_431 = __s0_431 + __s1_431 * splatq_laneq_s16(__s2_431, __p3_431); \
+ __ret_431; \
})
#else
-#define vmlaq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmlaq_laneq_s16(__p0_432, __p1_432, __p2_432, __p3_432) __extension__ ({ \
+ int16x8_t __s0_432 = __p0_432; \
+ int16x8_t __s1_432 = __p1_432; \
+ int16x8_t __s2_432 = __p2_432; \
+ int16x8_t __rev0_432; __rev0_432 = __builtin_shufflevector(__s0_432, __s0_432, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_432; __rev1_432 = __builtin_shufflevector(__s1_432, __s1_432, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev2_432; __rev2_432 = __builtin_shufflevector(__s2_432, __s2_432, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __ret_432; \
+ __ret_432 = __rev0_432 + __rev1_432 * __noswap_splatq_laneq_s16(__rev2_432, __p3_432); \
+ __ret_432 = __builtin_shufflevector(__ret_432, __ret_432, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_432; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmla_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint32x2_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3); \
- __ret; \
+#define vmla_laneq_u32(__p0_433, __p1_433, __p2_433, __p3_433) __extension__ ({ \
+ uint32x2_t __s0_433 = __p0_433; \
+ uint32x2_t __s1_433 = __p1_433; \
+ uint32x4_t __s2_433 = __p2_433; \
+ uint32x2_t __ret_433; \
+ __ret_433 = __s0_433 + __s1_433 * splat_laneq_u32(__s2_433, __p3_433); \
+ __ret_433; \
})
#else
-#define vmla_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint32x2_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmla_laneq_u32(__p0_434, __p1_434, __p2_434, __p3_434) __extension__ ({ \
+ uint32x2_t __s0_434 = __p0_434; \
+ uint32x2_t __s1_434 = __p1_434; \
+ uint32x4_t __s2_434 = __p2_434; \
+ uint32x2_t __rev0_434; __rev0_434 = __builtin_shufflevector(__s0_434, __s0_434, 1, 0); \
+ uint32x2_t __rev1_434; __rev1_434 = __builtin_shufflevector(__s1_434, __s1_434, 1, 0); \
+ uint32x4_t __rev2_434; __rev2_434 = __builtin_shufflevector(__s2_434, __s2_434, 3, 2, 1, 0); \
+ uint32x2_t __ret_434; \
+ __ret_434 = __rev0_434 + __rev1_434 * __noswap_splat_laneq_u32(__rev2_434, __p3_434); \
+ __ret_434 = __builtin_shufflevector(__ret_434, __ret_434, 1, 0); \
+ __ret_434; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmla_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint16x4_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmla_laneq_u16(__p0_435, __p1_435, __p2_435, __p3_435) __extension__ ({ \
+ uint16x4_t __s0_435 = __p0_435; \
+ uint16x4_t __s1_435 = __p1_435; \
+ uint16x8_t __s2_435 = __p2_435; \
+ uint16x4_t __ret_435; \
+ __ret_435 = __s0_435 + __s1_435 * splat_laneq_u16(__s2_435, __p3_435); \
+ __ret_435; \
})
#else
-#define vmla_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x4_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmla_laneq_u16(__p0_436, __p1_436, __p2_436, __p3_436) __extension__ ({ \
+ uint16x4_t __s0_436 = __p0_436; \
+ uint16x4_t __s1_436 = __p1_436; \
+ uint16x8_t __s2_436 = __p2_436; \
+ uint16x4_t __rev0_436; __rev0_436 = __builtin_shufflevector(__s0_436, __s0_436, 3, 2, 1, 0); \
+ uint16x4_t __rev1_436; __rev1_436 = __builtin_shufflevector(__s1_436, __s1_436, 3, 2, 1, 0); \
+ uint16x8_t __rev2_436; __rev2_436 = __builtin_shufflevector(__s2_436, __s2_436, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x4_t __ret_436; \
+ __ret_436 = __rev0_436 + __rev1_436 * __noswap_splat_laneq_u16(__rev2_436, __p3_436); \
+ __ret_436 = __builtin_shufflevector(__ret_436, __ret_436, 3, 2, 1, 0); \
+ __ret_436; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmla_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x4_t __s2 = __p2; \
- float32x2_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3); \
- __ret; \
+#define vmla_laneq_f32(__p0_437, __p1_437, __p2_437, __p3_437) __extension__ ({ \
+ float32x2_t __s0_437 = __p0_437; \
+ float32x2_t __s1_437 = __p1_437; \
+ float32x4_t __s2_437 = __p2_437; \
+ float32x2_t __ret_437; \
+ __ret_437 = __s0_437 + __s1_437 * splat_laneq_f32(__s2_437, __p3_437); \
+ __ret_437; \
})
#else
-#define vmla_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x4_t __s2 = __p2; \
- float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- float32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- float32x2_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmla_laneq_f32(__p0_438, __p1_438, __p2_438, __p3_438) __extension__ ({ \
+ float32x2_t __s0_438 = __p0_438; \
+ float32x2_t __s1_438 = __p1_438; \
+ float32x4_t __s2_438 = __p2_438; \
+ float32x2_t __rev0_438; __rev0_438 = __builtin_shufflevector(__s0_438, __s0_438, 1, 0); \
+ float32x2_t __rev1_438; __rev1_438 = __builtin_shufflevector(__s1_438, __s1_438, 1, 0); \
+ float32x4_t __rev2_438; __rev2_438 = __builtin_shufflevector(__s2_438, __s2_438, 3, 2, 1, 0); \
+ float32x2_t __ret_438; \
+ __ret_438 = __rev0_438 + __rev1_438 * __noswap_splat_laneq_f32(__rev2_438, __p3_438); \
+ __ret_438 = __builtin_shufflevector(__ret_438, __ret_438, 1, 0); \
+ __ret_438; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmla_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x2_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3); \
- __ret; \
+#define vmla_laneq_s32(__p0_439, __p1_439, __p2_439, __p3_439) __extension__ ({ \
+ int32x2_t __s0_439 = __p0_439; \
+ int32x2_t __s1_439 = __p1_439; \
+ int32x4_t __s2_439 = __p2_439; \
+ int32x2_t __ret_439; \
+ __ret_439 = __s0_439 + __s1_439 * splat_laneq_s32(__s2_439, __p3_439); \
+ __ret_439; \
})
#else
-#define vmla_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x2_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmla_laneq_s32(__p0_440, __p1_440, __p2_440, __p3_440) __extension__ ({ \
+ int32x2_t __s0_440 = __p0_440; \
+ int32x2_t __s1_440 = __p1_440; \
+ int32x4_t __s2_440 = __p2_440; \
+ int32x2_t __rev0_440; __rev0_440 = __builtin_shufflevector(__s0_440, __s0_440, 1, 0); \
+ int32x2_t __rev1_440; __rev1_440 = __builtin_shufflevector(__s1_440, __s1_440, 1, 0); \
+ int32x4_t __rev2_440; __rev2_440 = __builtin_shufflevector(__s2_440, __s2_440, 3, 2, 1, 0); \
+ int32x2_t __ret_440; \
+ __ret_440 = __rev0_440 + __rev1_440 * __noswap_splat_laneq_s32(__rev2_440, __p3_440); \
+ __ret_440 = __builtin_shufflevector(__ret_440, __ret_440, 1, 0); \
+ __ret_440; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmla_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x4_t __ret; \
- __ret = __s0 + __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmla_laneq_s16(__p0_441, __p1_441, __p2_441, __p3_441) __extension__ ({ \
+ int16x4_t __s0_441 = __p0_441; \
+ int16x4_t __s1_441 = __p1_441; \
+ int16x8_t __s2_441 = __p2_441; \
+ int16x4_t __ret_441; \
+ __ret_441 = __s0_441 + __s1_441 * splat_laneq_s16(__s2_441, __p3_441); \
+ __ret_441; \
})
#else
-#define vmla_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __ret; \
- __ret = __rev0 + __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmla_laneq_s16(__p0_442, __p1_442, __p2_442, __p3_442) __extension__ ({ \
+ int16x4_t __s0_442 = __p0_442; \
+ int16x4_t __s1_442 = __p1_442; \
+ int16x8_t __s2_442 = __p2_442; \
+ int16x4_t __rev0_442; __rev0_442 = __builtin_shufflevector(__s0_442, __s0_442, 3, 2, 1, 0); \
+ int16x4_t __rev1_442; __rev1_442 = __builtin_shufflevector(__s1_442, __s1_442, 3, 2, 1, 0); \
+ int16x8_t __rev2_442; __rev2_442 = __builtin_shufflevector(__s2_442, __s2_442, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __ret_442; \
+ __ret_442 = __rev0_442 + __rev1_442 * __noswap_splat_laneq_s16(__rev2_442, __p3_442); \
+ __ret_442 = __builtin_shufflevector(__ret_442, __ret_442, 3, 2, 1, 0); \
+ __ret_442; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_high_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint64x2_t __ret; \
- __ret = __s0 + vmull_u32(vget_high_u32(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlal_high_lane_u32(__p0_443, __p1_443, __p2_443, __p3_443) __extension__ ({ \
+ uint64x2_t __s0_443 = __p0_443; \
+ uint32x4_t __s1_443 = __p1_443; \
+ uint32x2_t __s2_443 = __p2_443; \
+ uint64x2_t __ret_443; \
+ __ret_443 = __s0_443 + vmull_u32(vget_high_u32(__s1_443), splat_lane_u32(__s2_443, __p3_443)); \
+ __ret_443; \
})
#else
-#define vmlal_high_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- uint64x2_t __ret; \
- __ret = __rev0 + __noswap_vmull_u32(__noswap_vget_high_u32(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmlal_high_lane_u32(__p0_444, __p1_444, __p2_444, __p3_444) __extension__ ({ \
+ uint64x2_t __s0_444 = __p0_444; \
+ uint32x4_t __s1_444 = __p1_444; \
+ uint32x2_t __s2_444 = __p2_444; \
+ uint64x2_t __rev0_444; __rev0_444 = __builtin_shufflevector(__s0_444, __s0_444, 1, 0); \
+ uint32x4_t __rev1_444; __rev1_444 = __builtin_shufflevector(__s1_444, __s1_444, 3, 2, 1, 0); \
+ uint32x2_t __rev2_444; __rev2_444 = __builtin_shufflevector(__s2_444, __s2_444, 1, 0); \
+ uint64x2_t __ret_444; \
+ __ret_444 = __rev0_444 + __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_444), __noswap_splat_lane_u32(__rev2_444, __p3_444)); \
+ __ret_444 = __builtin_shufflevector(__ret_444, __ret_444, 1, 0); \
+ __ret_444; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_high_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint32x4_t __ret; \
- __ret = __s0 + vmull_u16(vget_high_u16(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlal_high_lane_u16(__p0_445, __p1_445, __p2_445, __p3_445) __extension__ ({ \
+ uint32x4_t __s0_445 = __p0_445; \
+ uint16x8_t __s1_445 = __p1_445; \
+ uint16x4_t __s2_445 = __p2_445; \
+ uint32x4_t __ret_445; \
+ __ret_445 = __s0_445 + vmull_u16(vget_high_u16(__s1_445), splat_lane_u16(__s2_445, __p3_445)); \
+ __ret_445; \
})
#else
-#define vmlal_high_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __rev0 + __noswap_vmull_u16(__noswap_vget_high_u16(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlal_high_lane_u16(__p0_446, __p1_446, __p2_446, __p3_446) __extension__ ({ \
+ uint32x4_t __s0_446 = __p0_446; \
+ uint16x8_t __s1_446 = __p1_446; \
+ uint16x4_t __s2_446 = __p2_446; \
+ uint32x4_t __rev0_446; __rev0_446 = __builtin_shufflevector(__s0_446, __s0_446, 3, 2, 1, 0); \
+ uint16x8_t __rev1_446; __rev1_446 = __builtin_shufflevector(__s1_446, __s1_446, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x4_t __rev2_446; __rev2_446 = __builtin_shufflevector(__s2_446, __s2_446, 3, 2, 1, 0); \
+ uint32x4_t __ret_446; \
+ __ret_446 = __rev0_446 + __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_446), __noswap_splat_lane_u16(__rev2_446, __p3_446)); \
+ __ret_446 = __builtin_shufflevector(__ret_446, __ret_446, 3, 2, 1, 0); \
+ __ret_446; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = __s0 + vmull_s32(vget_high_s32(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlal_high_lane_s32(__p0_447, __p1_447, __p2_447, __p3_447) __extension__ ({ \
+ int64x2_t __s0_447 = __p0_447; \
+ int32x4_t __s1_447 = __p1_447; \
+ int32x2_t __s2_447 = __p2_447; \
+ int64x2_t __ret_447; \
+ __ret_447 = __s0_447 + vmull_s32(vget_high_s32(__s1_447), splat_lane_s32(__s2_447, __p3_447)); \
+ __ret_447; \
})
#else
-#define vmlal_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int64x2_t __ret; \
- __ret = __rev0 + __noswap_vmull_s32(__noswap_vget_high_s32(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmlal_high_lane_s32(__p0_448, __p1_448, __p2_448, __p3_448) __extension__ ({ \
+ int64x2_t __s0_448 = __p0_448; \
+ int32x4_t __s1_448 = __p1_448; \
+ int32x2_t __s2_448 = __p2_448; \
+ int64x2_t __rev0_448; __rev0_448 = __builtin_shufflevector(__s0_448, __s0_448, 1, 0); \
+ int32x4_t __rev1_448; __rev1_448 = __builtin_shufflevector(__s1_448, __s1_448, 3, 2, 1, 0); \
+ int32x2_t __rev2_448; __rev2_448 = __builtin_shufflevector(__s2_448, __s2_448, 1, 0); \
+ int64x2_t __ret_448; \
+ __ret_448 = __rev0_448 + __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_448), __noswap_splat_lane_s32(__rev2_448, __p3_448)); \
+ __ret_448 = __builtin_shufflevector(__ret_448, __ret_448, 1, 0); \
+ __ret_448; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = __s0 + vmull_s16(vget_high_s16(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlal_high_lane_s16(__p0_449, __p1_449, __p2_449, __p3_449) __extension__ ({ \
+ int32x4_t __s0_449 = __p0_449; \
+ int16x8_t __s1_449 = __p1_449; \
+ int16x4_t __s2_449 = __p2_449; \
+ int32x4_t __ret_449; \
+ __ret_449 = __s0_449 + vmull_s16(vget_high_s16(__s1_449), splat_lane_s16(__s2_449, __p3_449)); \
+ __ret_449; \
})
#else
-#define vmlal_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __rev0 + __noswap_vmull_s16(__noswap_vget_high_s16(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlal_high_lane_s16(__p0_450, __p1_450, __p2_450, __p3_450) __extension__ ({ \
+ int32x4_t __s0_450 = __p0_450; \
+ int16x8_t __s1_450 = __p1_450; \
+ int16x4_t __s2_450 = __p2_450; \
+ int32x4_t __rev0_450; __rev0_450 = __builtin_shufflevector(__s0_450, __s0_450, 3, 2, 1, 0); \
+ int16x8_t __rev1_450; __rev1_450 = __builtin_shufflevector(__s1_450, __s1_450, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev2_450; __rev2_450 = __builtin_shufflevector(__s2_450, __s2_450, 3, 2, 1, 0); \
+ int32x4_t __ret_450; \
+ __ret_450 = __rev0_450 + __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_450), __noswap_splat_lane_s16(__rev2_450, __p3_450)); \
+ __ret_450 = __builtin_shufflevector(__ret_450, __ret_450, 3, 2, 1, 0); \
+ __ret_450; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_high_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint64x2_t __ret; \
- __ret = __s0 + vmull_u32(vget_high_u32(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlal_high_laneq_u32(__p0_451, __p1_451, __p2_451, __p3_451) __extension__ ({ \
+ uint64x2_t __s0_451 = __p0_451; \
+ uint32x4_t __s1_451 = __p1_451; \
+ uint32x4_t __s2_451 = __p2_451; \
+ uint64x2_t __ret_451; \
+ __ret_451 = __s0_451 + vmull_u32(vget_high_u32(__s1_451), splat_laneq_u32(__s2_451, __p3_451)); \
+ __ret_451; \
})
#else
-#define vmlal_high_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint64x2_t __ret; \
- __ret = __rev0 + __noswap_vmull_u32(__noswap_vget_high_u32(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmlal_high_laneq_u32(__p0_452, __p1_452, __p2_452, __p3_452) __extension__ ({ \
+ uint64x2_t __s0_452 = __p0_452; \
+ uint32x4_t __s1_452 = __p1_452; \
+ uint32x4_t __s2_452 = __p2_452; \
+ uint64x2_t __rev0_452; __rev0_452 = __builtin_shufflevector(__s0_452, __s0_452, 1, 0); \
+ uint32x4_t __rev1_452; __rev1_452 = __builtin_shufflevector(__s1_452, __s1_452, 3, 2, 1, 0); \
+ uint32x4_t __rev2_452; __rev2_452 = __builtin_shufflevector(__s2_452, __s2_452, 3, 2, 1, 0); \
+ uint64x2_t __ret_452; \
+ __ret_452 = __rev0_452 + __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_452), __noswap_splat_laneq_u32(__rev2_452, __p3_452)); \
+ __ret_452 = __builtin_shufflevector(__ret_452, __ret_452, 1, 0); \
+ __ret_452; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_high_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint32x4_t __ret; \
- __ret = __s0 + vmull_u16(vget_high_u16(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlal_high_laneq_u16(__p0_453, __p1_453, __p2_453, __p3_453) __extension__ ({ \
+ uint32x4_t __s0_453 = __p0_453; \
+ uint16x8_t __s1_453 = __p1_453; \
+ uint16x8_t __s2_453 = __p2_453; \
+ uint32x4_t __ret_453; \
+ __ret_453 = __s0_453 + vmull_u16(vget_high_u16(__s1_453), splat_laneq_u16(__s2_453, __p3_453)); \
+ __ret_453; \
})
#else
-#define vmlal_high_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __rev0 + __noswap_vmull_u16(__noswap_vget_high_u16(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlal_high_laneq_u16(__p0_454, __p1_454, __p2_454, __p3_454) __extension__ ({ \
+ uint32x4_t __s0_454 = __p0_454; \
+ uint16x8_t __s1_454 = __p1_454; \
+ uint16x8_t __s2_454 = __p2_454; \
+ uint32x4_t __rev0_454; __rev0_454 = __builtin_shufflevector(__s0_454, __s0_454, 3, 2, 1, 0); \
+ uint16x8_t __rev1_454; __rev1_454 = __builtin_shufflevector(__s1_454, __s1_454, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __rev2_454; __rev2_454 = __builtin_shufflevector(__s2_454, __s2_454, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint32x4_t __ret_454; \
+ __ret_454 = __rev0_454 + __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_454), __noswap_splat_laneq_u16(__rev2_454, __p3_454)); \
+ __ret_454 = __builtin_shufflevector(__ret_454, __ret_454, 3, 2, 1, 0); \
+ __ret_454; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = __s0 + vmull_s32(vget_high_s32(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlal_high_laneq_s32(__p0_455, __p1_455, __p2_455, __p3_455) __extension__ ({ \
+ int64x2_t __s0_455 = __p0_455; \
+ int32x4_t __s1_455 = __p1_455; \
+ int32x4_t __s2_455 = __p2_455; \
+ int64x2_t __ret_455; \
+ __ret_455 = __s0_455 + vmull_s32(vget_high_s32(__s1_455), splat_laneq_s32(__s2_455, __p3_455)); \
+ __ret_455; \
})
#else
-#define vmlal_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int64x2_t __ret; \
- __ret = __rev0 + __noswap_vmull_s32(__noswap_vget_high_s32(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmlal_high_laneq_s32(__p0_456, __p1_456, __p2_456, __p3_456) __extension__ ({ \
+ int64x2_t __s0_456 = __p0_456; \
+ int32x4_t __s1_456 = __p1_456; \
+ int32x4_t __s2_456 = __p2_456; \
+ int64x2_t __rev0_456; __rev0_456 = __builtin_shufflevector(__s0_456, __s0_456, 1, 0); \
+ int32x4_t __rev1_456; __rev1_456 = __builtin_shufflevector(__s1_456, __s1_456, 3, 2, 1, 0); \
+ int32x4_t __rev2_456; __rev2_456 = __builtin_shufflevector(__s2_456, __s2_456, 3, 2, 1, 0); \
+ int64x2_t __ret_456; \
+ __ret_456 = __rev0_456 + __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_456), __noswap_splat_laneq_s32(__rev2_456, __p3_456)); \
+ __ret_456 = __builtin_shufflevector(__ret_456, __ret_456, 1, 0); \
+ __ret_456; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = __s0 + vmull_s16(vget_high_s16(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlal_high_laneq_s16(__p0_457, __p1_457, __p2_457, __p3_457) __extension__ ({ \
+ int32x4_t __s0_457 = __p0_457; \
+ int16x8_t __s1_457 = __p1_457; \
+ int16x8_t __s2_457 = __p2_457; \
+ int32x4_t __ret_457; \
+ __ret_457 = __s0_457 + vmull_s16(vget_high_s16(__s1_457), splat_laneq_s16(__s2_457, __p3_457)); \
+ __ret_457; \
})
#else
-#define vmlal_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __rev0 + __noswap_vmull_s16(__noswap_vget_high_s16(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlal_high_laneq_s16(__p0_458, __p1_458, __p2_458, __p3_458) __extension__ ({ \
+ int32x4_t __s0_458 = __p0_458; \
+ int16x8_t __s1_458 = __p1_458; \
+ int16x8_t __s2_458 = __p2_458; \
+ int32x4_t __rev0_458; __rev0_458 = __builtin_shufflevector(__s0_458, __s0_458, 3, 2, 1, 0); \
+ int16x8_t __rev1_458; __rev1_458 = __builtin_shufflevector(__s1_458, __s1_458, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev2_458; __rev2_458 = __builtin_shufflevector(__s2_458, __s2_458, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_458; \
+ __ret_458 = __rev0_458 + __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_458), __noswap_splat_laneq_s16(__rev2_458, __p3_458)); \
+ __ret_458 = __builtin_shufflevector(__ret_458, __ret_458, 3, 2, 1, 0); \
+ __ret_458; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint64x2_t __ret; \
- __ret = __s0 + vmull_u32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlal_laneq_u32(__p0_459, __p1_459, __p2_459, __p3_459) __extension__ ({ \
+ uint64x2_t __s0_459 = __p0_459; \
+ uint32x2_t __s1_459 = __p1_459; \
+ uint32x4_t __s2_459 = __p2_459; \
+ uint64x2_t __ret_459; \
+ __ret_459 = __s0_459 + vmull_u32(__s1_459, splat_laneq_u32(__s2_459, __p3_459)); \
+ __ret_459; \
})
#else
-#define vmlal_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint64x2_t __ret; \
- __ret = __rev0 + __noswap_vmull_u32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmlal_laneq_u32(__p0_460, __p1_460, __p2_460, __p3_460) __extension__ ({ \
+ uint64x2_t __s0_460 = __p0_460; \
+ uint32x2_t __s1_460 = __p1_460; \
+ uint32x4_t __s2_460 = __p2_460; \
+ uint64x2_t __rev0_460; __rev0_460 = __builtin_shufflevector(__s0_460, __s0_460, 1, 0); \
+ uint32x2_t __rev1_460; __rev1_460 = __builtin_shufflevector(__s1_460, __s1_460, 1, 0); \
+ uint32x4_t __rev2_460; __rev2_460 = __builtin_shufflevector(__s2_460, __s2_460, 3, 2, 1, 0); \
+ uint64x2_t __ret_460; \
+ __ret_460 = __rev0_460 + __noswap_vmull_u32(__rev1_460, __noswap_splat_laneq_u32(__rev2_460, __p3_460)); \
+ __ret_460 = __builtin_shufflevector(__ret_460, __ret_460, 1, 0); \
+ __ret_460; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint32x4_t __ret; \
- __ret = __s0 + vmull_u16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlal_laneq_u16(__p0_461, __p1_461, __p2_461, __p3_461) __extension__ ({ \
+ uint32x4_t __s0_461 = __p0_461; \
+ uint16x4_t __s1_461 = __p1_461; \
+ uint16x8_t __s2_461 = __p2_461; \
+ uint32x4_t __ret_461; \
+ __ret_461 = __s0_461 + vmull_u16(__s1_461, splat_laneq_u16(__s2_461, __p3_461)); \
+ __ret_461; \
})
#else
-#define vmlal_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __rev0 + __noswap_vmull_u16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlal_laneq_u16(__p0_462, __p1_462, __p2_462, __p3_462) __extension__ ({ \
+ uint32x4_t __s0_462 = __p0_462; \
+ uint16x4_t __s1_462 = __p1_462; \
+ uint16x8_t __s2_462 = __p2_462; \
+ uint32x4_t __rev0_462; __rev0_462 = __builtin_shufflevector(__s0_462, __s0_462, 3, 2, 1, 0); \
+ uint16x4_t __rev1_462; __rev1_462 = __builtin_shufflevector(__s1_462, __s1_462, 3, 2, 1, 0); \
+ uint16x8_t __rev2_462; __rev2_462 = __builtin_shufflevector(__s2_462, __s2_462, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint32x4_t __ret_462; \
+ __ret_462 = __rev0_462 + __noswap_vmull_u16(__rev1_462, __noswap_splat_laneq_u16(__rev2_462, __p3_462)); \
+ __ret_462 = __builtin_shufflevector(__ret_462, __ret_462, 3, 2, 1, 0); \
+ __ret_462; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = __s0 + vmull_s32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlal_laneq_s32(__p0_463, __p1_463, __p2_463, __p3_463) __extension__ ({ \
+ int64x2_t __s0_463 = __p0_463; \
+ int32x2_t __s1_463 = __p1_463; \
+ int32x4_t __s2_463 = __p2_463; \
+ int64x2_t __ret_463; \
+ __ret_463 = __s0_463 + vmull_s32(__s1_463, splat_laneq_s32(__s2_463, __p3_463)); \
+ __ret_463; \
})
#else
-#define vmlal_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int64x2_t __ret; \
- __ret = __rev0 + __noswap_vmull_s32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmlal_laneq_s32(__p0_464, __p1_464, __p2_464, __p3_464) __extension__ ({ \
+ int64x2_t __s0_464 = __p0_464; \
+ int32x2_t __s1_464 = __p1_464; \
+ int32x4_t __s2_464 = __p2_464; \
+ int64x2_t __rev0_464; __rev0_464 = __builtin_shufflevector(__s0_464, __s0_464, 1, 0); \
+ int32x2_t __rev1_464; __rev1_464 = __builtin_shufflevector(__s1_464, __s1_464, 1, 0); \
+ int32x4_t __rev2_464; __rev2_464 = __builtin_shufflevector(__s2_464, __s2_464, 3, 2, 1, 0); \
+ int64x2_t __ret_464; \
+ __ret_464 = __rev0_464 + __noswap_vmull_s32(__rev1_464, __noswap_splat_laneq_s32(__rev2_464, __p3_464)); \
+ __ret_464 = __builtin_shufflevector(__ret_464, __ret_464, 1, 0); \
+ __ret_464; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = __s0 + vmull_s16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlal_laneq_s16(__p0_465, __p1_465, __p2_465, __p3_465) __extension__ ({ \
+ int32x4_t __s0_465 = __p0_465; \
+ int16x4_t __s1_465 = __p1_465; \
+ int16x8_t __s2_465 = __p2_465; \
+ int32x4_t __ret_465; \
+ __ret_465 = __s0_465 + vmull_s16(__s1_465, splat_laneq_s16(__s2_465, __p3_465)); \
+ __ret_465; \
})
#else
-#define vmlal_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __rev0 + __noswap_vmull_s16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlal_laneq_s16(__p0_466, __p1_466, __p2_466, __p3_466) __extension__ ({ \
+ int32x4_t __s0_466 = __p0_466; \
+ int16x4_t __s1_466 = __p1_466; \
+ int16x8_t __s2_466 = __p2_466; \
+ int32x4_t __rev0_466; __rev0_466 = __builtin_shufflevector(__s0_466, __s0_466, 3, 2, 1, 0); \
+ int16x4_t __rev1_466; __rev1_466 = __builtin_shufflevector(__s1_466, __s1_466, 3, 2, 1, 0); \
+ int16x8_t __rev2_466; __rev2_466 = __builtin_shufflevector(__s2_466, __s2_466, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_466; \
+ __ret_466 = __rev0_466 + __noswap_vmull_s16(__rev1_466, __noswap_splat_laneq_s16(__rev2_466, __p3_466)); \
+ __ret_466 = __builtin_shufflevector(__ret_466, __ret_466, 3, 2, 1, 0); \
+ __ret_466; \
})
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
float64x1_t __ret;
__ret = __p0 - __p1 * __p2;
return __ret;
}
-#else
-__ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
- float64x1_t __ret;
- __ret = __p0 - __p1 * __p2;
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vmlsq_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint32x4_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlsq_laneq_u32(__p0_467, __p1_467, __p2_467, __p3_467) __extension__ ({ \
+ uint32x4_t __s0_467 = __p0_467; \
+ uint32x4_t __s1_467 = __p1_467; \
+ uint32x4_t __s2_467 = __p2_467; \
+ uint32x4_t __ret_467; \
+ __ret_467 = __s0_467 - __s1_467 * splatq_laneq_u32(__s2_467, __p3_467); \
+ __ret_467; \
})
#else
-#define vmlsq_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlsq_laneq_u32(__p0_468, __p1_468, __p2_468, __p3_468) __extension__ ({ \
+ uint32x4_t __s0_468 = __p0_468; \
+ uint32x4_t __s1_468 = __p1_468; \
+ uint32x4_t __s2_468 = __p2_468; \
+ uint32x4_t __rev0_468; __rev0_468 = __builtin_shufflevector(__s0_468, __s0_468, 3, 2, 1, 0); \
+ uint32x4_t __rev1_468; __rev1_468 = __builtin_shufflevector(__s1_468, __s1_468, 3, 2, 1, 0); \
+ uint32x4_t __rev2_468; __rev2_468 = __builtin_shufflevector(__s2_468, __s2_468, 3, 2, 1, 0); \
+ uint32x4_t __ret_468; \
+ __ret_468 = __rev0_468 - __rev1_468 * __noswap_splatq_laneq_u32(__rev2_468, __p3_468); \
+ __ret_468 = __builtin_shufflevector(__ret_468, __ret_468, 3, 2, 1, 0); \
+ __ret_468; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsq_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint16x8_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlsq_laneq_u16(__p0_469, __p1_469, __p2_469, __p3_469) __extension__ ({ \
+ uint16x8_t __s0_469 = __p0_469; \
+ uint16x8_t __s1_469 = __p1_469; \
+ uint16x8_t __s2_469 = __p2_469; \
+ uint16x8_t __ret_469; \
+ __ret_469 = __s0_469 - __s1_469 * splatq_laneq_u16(__s2_469, __p3_469); \
+ __ret_469; \
})
#else
-#define vmlsq_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmlsq_laneq_u16(__p0_470, __p1_470, __p2_470, __p3_470) __extension__ ({ \
+ uint16x8_t __s0_470 = __p0_470; \
+ uint16x8_t __s1_470 = __p1_470; \
+ uint16x8_t __s2_470 = __p2_470; \
+ uint16x8_t __rev0_470; __rev0_470 = __builtin_shufflevector(__s0_470, __s0_470, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __rev1_470; __rev1_470 = __builtin_shufflevector(__s1_470, __s1_470, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __rev2_470; __rev2_470 = __builtin_shufflevector(__s2_470, __s2_470, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __ret_470; \
+ __ret_470 = __rev0_470 - __rev1_470 * __noswap_splatq_laneq_u16(__rev2_470, __p3_470); \
+ __ret_470 = __builtin_shufflevector(__ret_470, __ret_470, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_470; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x4_t __s2 = __p2; \
- float32x4_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlsq_laneq_f32(__p0_471, __p1_471, __p2_471, __p3_471) __extension__ ({ \
+ float32x4_t __s0_471 = __p0_471; \
+ float32x4_t __s1_471 = __p1_471; \
+ float32x4_t __s2_471 = __p2_471; \
+ float32x4_t __ret_471; \
+ __ret_471 = __s0_471 - __s1_471 * splatq_laneq_f32(__s2_471, __p3_471); \
+ __ret_471; \
})
#else
-#define vmlsq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x4_t __s2 = __p2; \
- float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- float32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- float32x4_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlsq_laneq_f32(__p0_472, __p1_472, __p2_472, __p3_472) __extension__ ({ \
+ float32x4_t __s0_472 = __p0_472; \
+ float32x4_t __s1_472 = __p1_472; \
+ float32x4_t __s2_472 = __p2_472; \
+ float32x4_t __rev0_472; __rev0_472 = __builtin_shufflevector(__s0_472, __s0_472, 3, 2, 1, 0); \
+ float32x4_t __rev1_472; __rev1_472 = __builtin_shufflevector(__s1_472, __s1_472, 3, 2, 1, 0); \
+ float32x4_t __rev2_472; __rev2_472 = __builtin_shufflevector(__s2_472, __s2_472, 3, 2, 1, 0); \
+ float32x4_t __ret_472; \
+ __ret_472 = __rev0_472 - __rev1_472 * __noswap_splatq_laneq_f32(__rev2_472, __p3_472); \
+ __ret_472 = __builtin_shufflevector(__ret_472, __ret_472, 3, 2, 1, 0); \
+ __ret_472; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlsq_laneq_s32(__p0_473, __p1_473, __p2_473, __p3_473) __extension__ ({ \
+ int32x4_t __s0_473 = __p0_473; \
+ int32x4_t __s1_473 = __p1_473; \
+ int32x4_t __s2_473 = __p2_473; \
+ int32x4_t __ret_473; \
+ __ret_473 = __s0_473 - __s1_473 * splatq_laneq_s32(__s2_473, __p3_473); \
+ __ret_473; \
})
#else
-#define vmlsq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlsq_laneq_s32(__p0_474, __p1_474, __p2_474, __p3_474) __extension__ ({ \
+ int32x4_t __s0_474 = __p0_474; \
+ int32x4_t __s1_474 = __p1_474; \
+ int32x4_t __s2_474 = __p2_474; \
+ int32x4_t __rev0_474; __rev0_474 = __builtin_shufflevector(__s0_474, __s0_474, 3, 2, 1, 0); \
+ int32x4_t __rev1_474; __rev1_474 = __builtin_shufflevector(__s1_474, __s1_474, 3, 2, 1, 0); \
+ int32x4_t __rev2_474; __rev2_474 = __builtin_shufflevector(__s2_474, __s2_474, 3, 2, 1, 0); \
+ int32x4_t __ret_474; \
+ __ret_474 = __rev0_474 - __rev1_474 * __noswap_splatq_laneq_s32(__rev2_474, __p3_474); \
+ __ret_474 = __builtin_shufflevector(__ret_474, __ret_474, 3, 2, 1, 0); \
+ __ret_474; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x8_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmlsq_laneq_s16(__p0_475, __p1_475, __p2_475, __p3_475) __extension__ ({ \
+ int16x8_t __s0_475 = __p0_475; \
+ int16x8_t __s1_475 = __p1_475; \
+ int16x8_t __s2_475 = __p2_475; \
+ int16x8_t __ret_475; \
+ __ret_475 = __s0_475 - __s1_475 * splatq_laneq_s16(__s2_475, __p3_475); \
+ __ret_475; \
})
#else
-#define vmlsq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmlsq_laneq_s16(__p0_476, __p1_476, __p2_476, __p3_476) __extension__ ({ \
+ int16x8_t __s0_476 = __p0_476; \
+ int16x8_t __s1_476 = __p1_476; \
+ int16x8_t __s2_476 = __p2_476; \
+ int16x8_t __rev0_476; __rev0_476 = __builtin_shufflevector(__s0_476, __s0_476, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_476; __rev1_476 = __builtin_shufflevector(__s1_476, __s1_476, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev2_476; __rev2_476 = __builtin_shufflevector(__s2_476, __s2_476, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __ret_476; \
+ __ret_476 = __rev0_476 - __rev1_476 * __noswap_splatq_laneq_s16(__rev2_476, __p3_476); \
+ __ret_476 = __builtin_shufflevector(__ret_476, __ret_476, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_476; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmls_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint32x2_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3); \
- __ret; \
+#define vmls_laneq_u32(__p0_477, __p1_477, __p2_477, __p3_477) __extension__ ({ \
+ uint32x2_t __s0_477 = __p0_477; \
+ uint32x2_t __s1_477 = __p1_477; \
+ uint32x4_t __s2_477 = __p2_477; \
+ uint32x2_t __ret_477; \
+ __ret_477 = __s0_477 - __s1_477 * splat_laneq_u32(__s2_477, __p3_477); \
+ __ret_477; \
})
#else
-#define vmls_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint32x2_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmls_laneq_u32(__p0_478, __p1_478, __p2_478, __p3_478) __extension__ ({ \
+ uint32x2_t __s0_478 = __p0_478; \
+ uint32x2_t __s1_478 = __p1_478; \
+ uint32x4_t __s2_478 = __p2_478; \
+ uint32x2_t __rev0_478; __rev0_478 = __builtin_shufflevector(__s0_478, __s0_478, 1, 0); \
+ uint32x2_t __rev1_478; __rev1_478 = __builtin_shufflevector(__s1_478, __s1_478, 1, 0); \
+ uint32x4_t __rev2_478; __rev2_478 = __builtin_shufflevector(__s2_478, __s2_478, 3, 2, 1, 0); \
+ uint32x2_t __ret_478; \
+ __ret_478 = __rev0_478 - __rev1_478 * __noswap_splat_laneq_u32(__rev2_478, __p3_478); \
+ __ret_478 = __builtin_shufflevector(__ret_478, __ret_478, 1, 0); \
+ __ret_478; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmls_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint16x4_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmls_laneq_u16(__p0_479, __p1_479, __p2_479, __p3_479) __extension__ ({ \
+ uint16x4_t __s0_479 = __p0_479; \
+ uint16x4_t __s1_479 = __p1_479; \
+ uint16x8_t __s2_479 = __p2_479; \
+ uint16x4_t __ret_479; \
+ __ret_479 = __s0_479 - __s1_479 * splat_laneq_u16(__s2_479, __p3_479); \
+ __ret_479; \
})
#else
-#define vmls_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x4_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmls_laneq_u16(__p0_480, __p1_480, __p2_480, __p3_480) __extension__ ({ \
+ uint16x4_t __s0_480 = __p0_480; \
+ uint16x4_t __s1_480 = __p1_480; \
+ uint16x8_t __s2_480 = __p2_480; \
+ uint16x4_t __rev0_480; __rev0_480 = __builtin_shufflevector(__s0_480, __s0_480, 3, 2, 1, 0); \
+ uint16x4_t __rev1_480; __rev1_480 = __builtin_shufflevector(__s1_480, __s1_480, 3, 2, 1, 0); \
+ uint16x8_t __rev2_480; __rev2_480 = __builtin_shufflevector(__s2_480, __s2_480, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x4_t __ret_480; \
+ __ret_480 = __rev0_480 - __rev1_480 * __noswap_splat_laneq_u16(__rev2_480, __p3_480); \
+ __ret_480 = __builtin_shufflevector(__ret_480, __ret_480, 3, 2, 1, 0); \
+ __ret_480; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmls_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x4_t __s2 = __p2; \
- float32x2_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3); \
- __ret; \
+#define vmls_laneq_f32(__p0_481, __p1_481, __p2_481, __p3_481) __extension__ ({ \
+ float32x2_t __s0_481 = __p0_481; \
+ float32x2_t __s1_481 = __p1_481; \
+ float32x4_t __s2_481 = __p2_481; \
+ float32x2_t __ret_481; \
+ __ret_481 = __s0_481 - __s1_481 * splat_laneq_f32(__s2_481, __p3_481); \
+ __ret_481; \
})
#else
-#define vmls_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x4_t __s2 = __p2; \
- float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- float32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- float32x2_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmls_laneq_f32(__p0_482, __p1_482, __p2_482, __p3_482) __extension__ ({ \
+ float32x2_t __s0_482 = __p0_482; \
+ float32x2_t __s1_482 = __p1_482; \
+ float32x4_t __s2_482 = __p2_482; \
+ float32x2_t __rev0_482; __rev0_482 = __builtin_shufflevector(__s0_482, __s0_482, 1, 0); \
+ float32x2_t __rev1_482; __rev1_482 = __builtin_shufflevector(__s1_482, __s1_482, 1, 0); \
+ float32x4_t __rev2_482; __rev2_482 = __builtin_shufflevector(__s2_482, __s2_482, 3, 2, 1, 0); \
+ float32x2_t __ret_482; \
+ __ret_482 = __rev0_482 - __rev1_482 * __noswap_splat_laneq_f32(__rev2_482, __p3_482); \
+ __ret_482 = __builtin_shufflevector(__ret_482, __ret_482, 1, 0); \
+ __ret_482; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x2_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3); \
- __ret; \
+#define vmls_laneq_s32(__p0_483, __p1_483, __p2_483, __p3_483) __extension__ ({ \
+ int32x2_t __s0_483 = __p0_483; \
+ int32x2_t __s1_483 = __p1_483; \
+ int32x4_t __s2_483 = __p2_483; \
+ int32x2_t __ret_483; \
+ __ret_483 = __s0_483 - __s1_483 * splat_laneq_s32(__s2_483, __p3_483); \
+ __ret_483; \
})
#else
-#define vmls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x2_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmls_laneq_s32(__p0_484, __p1_484, __p2_484, __p3_484) __extension__ ({ \
+ int32x2_t __s0_484 = __p0_484; \
+ int32x2_t __s1_484 = __p1_484; \
+ int32x4_t __s2_484 = __p2_484; \
+ int32x2_t __rev0_484; __rev0_484 = __builtin_shufflevector(__s0_484, __s0_484, 1, 0); \
+ int32x2_t __rev1_484; __rev1_484 = __builtin_shufflevector(__s1_484, __s1_484, 1, 0); \
+ int32x4_t __rev2_484; __rev2_484 = __builtin_shufflevector(__s2_484, __s2_484, 3, 2, 1, 0); \
+ int32x2_t __ret_484; \
+ __ret_484 = __rev0_484 - __rev1_484 * __noswap_splat_laneq_s32(__rev2_484, __p3_484); \
+ __ret_484 = __builtin_shufflevector(__ret_484, __ret_484, 1, 0); \
+ __ret_484; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmls_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x4_t __ret; \
- __ret = __s0 - __s1 * __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3); \
- __ret; \
+#define vmls_laneq_s16(__p0_485, __p1_485, __p2_485, __p3_485) __extension__ ({ \
+ int16x4_t __s0_485 = __p0_485; \
+ int16x4_t __s1_485 = __p1_485; \
+ int16x8_t __s2_485 = __p2_485; \
+ int16x4_t __ret_485; \
+ __ret_485 = __s0_485 - __s1_485 * splat_laneq_s16(__s2_485, __p3_485); \
+ __ret_485; \
})
#else
-#define vmls_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __ret; \
- __ret = __rev0 - __rev1 * __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmls_laneq_s16(__p0_486, __p1_486, __p2_486, __p3_486) __extension__ ({ \
+ int16x4_t __s0_486 = __p0_486; \
+ int16x4_t __s1_486 = __p1_486; \
+ int16x8_t __s2_486 = __p2_486; \
+ int16x4_t __rev0_486; __rev0_486 = __builtin_shufflevector(__s0_486, __s0_486, 3, 2, 1, 0); \
+ int16x4_t __rev1_486; __rev1_486 = __builtin_shufflevector(__s1_486, __s1_486, 3, 2, 1, 0); \
+ int16x8_t __rev2_486; __rev2_486 = __builtin_shufflevector(__s2_486, __s2_486, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __ret_486; \
+ __ret_486 = __rev0_486 - __rev1_486 * __noswap_splat_laneq_s16(__rev2_486, __p3_486); \
+ __ret_486 = __builtin_shufflevector(__ret_486, __ret_486, 3, 2, 1, 0); \
+ __ret_486; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_high_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint64x2_t __ret; \
- __ret = __s0 - vmull_u32(vget_high_u32(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlsl_high_lane_u32(__p0_487, __p1_487, __p2_487, __p3_487) __extension__ ({ \
+ uint64x2_t __s0_487 = __p0_487; \
+ uint32x4_t __s1_487 = __p1_487; \
+ uint32x2_t __s2_487 = __p2_487; \
+ uint64x2_t __ret_487; \
+ __ret_487 = __s0_487 - vmull_u32(vget_high_u32(__s1_487), splat_lane_u32(__s2_487, __p3_487)); \
+ __ret_487; \
})
#else
-#define vmlsl_high_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- uint64x2_t __ret; \
- __ret = __rev0 - __noswap_vmull_u32(__noswap_vget_high_u32(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmlsl_high_lane_u32(__p0_488, __p1_488, __p2_488, __p3_488) __extension__ ({ \
+ uint64x2_t __s0_488 = __p0_488; \
+ uint32x4_t __s1_488 = __p1_488; \
+ uint32x2_t __s2_488 = __p2_488; \
+ uint64x2_t __rev0_488; __rev0_488 = __builtin_shufflevector(__s0_488, __s0_488, 1, 0); \
+ uint32x4_t __rev1_488; __rev1_488 = __builtin_shufflevector(__s1_488, __s1_488, 3, 2, 1, 0); \
+ uint32x2_t __rev2_488; __rev2_488 = __builtin_shufflevector(__s2_488, __s2_488, 1, 0); \
+ uint64x2_t __ret_488; \
+ __ret_488 = __rev0_488 - __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_488), __noswap_splat_lane_u32(__rev2_488, __p3_488)); \
+ __ret_488 = __builtin_shufflevector(__ret_488, __ret_488, 1, 0); \
+ __ret_488; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_high_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint32x4_t __ret; \
- __ret = __s0 - vmull_u16(vget_high_u16(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlsl_high_lane_u16(__p0_489, __p1_489, __p2_489, __p3_489) __extension__ ({ \
+ uint32x4_t __s0_489 = __p0_489; \
+ uint16x8_t __s1_489 = __p1_489; \
+ uint16x4_t __s2_489 = __p2_489; \
+ uint32x4_t __ret_489; \
+ __ret_489 = __s0_489 - vmull_u16(vget_high_u16(__s1_489), splat_lane_u16(__s2_489, __p3_489)); \
+ __ret_489; \
})
#else
-#define vmlsl_high_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __rev0 - __noswap_vmull_u16(__noswap_vget_high_u16(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlsl_high_lane_u16(__p0_490, __p1_490, __p2_490, __p3_490) __extension__ ({ \
+ uint32x4_t __s0_490 = __p0_490; \
+ uint16x8_t __s1_490 = __p1_490; \
+ uint16x4_t __s2_490 = __p2_490; \
+ uint32x4_t __rev0_490; __rev0_490 = __builtin_shufflevector(__s0_490, __s0_490, 3, 2, 1, 0); \
+ uint16x8_t __rev1_490; __rev1_490 = __builtin_shufflevector(__s1_490, __s1_490, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x4_t __rev2_490; __rev2_490 = __builtin_shufflevector(__s2_490, __s2_490, 3, 2, 1, 0); \
+ uint32x4_t __ret_490; \
+ __ret_490 = __rev0_490 - __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_490), __noswap_splat_lane_u16(__rev2_490, __p3_490)); \
+ __ret_490 = __builtin_shufflevector(__ret_490, __ret_490, 3, 2, 1, 0); \
+ __ret_490; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = __s0 - vmull_s32(vget_high_s32(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlsl_high_lane_s32(__p0_491, __p1_491, __p2_491, __p3_491) __extension__ ({ \
+ int64x2_t __s0_491 = __p0_491; \
+ int32x4_t __s1_491 = __p1_491; \
+ int32x2_t __s2_491 = __p2_491; \
+ int64x2_t __ret_491; \
+ __ret_491 = __s0_491 - vmull_s32(vget_high_s32(__s1_491), splat_lane_s32(__s2_491, __p3_491)); \
+ __ret_491; \
})
#else
-#define vmlsl_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int64x2_t __ret; \
- __ret = __rev0 - __noswap_vmull_s32(__noswap_vget_high_s32(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmlsl_high_lane_s32(__p0_492, __p1_492, __p2_492, __p3_492) __extension__ ({ \
+ int64x2_t __s0_492 = __p0_492; \
+ int32x4_t __s1_492 = __p1_492; \
+ int32x2_t __s2_492 = __p2_492; \
+ int64x2_t __rev0_492; __rev0_492 = __builtin_shufflevector(__s0_492, __s0_492, 1, 0); \
+ int32x4_t __rev1_492; __rev1_492 = __builtin_shufflevector(__s1_492, __s1_492, 3, 2, 1, 0); \
+ int32x2_t __rev2_492; __rev2_492 = __builtin_shufflevector(__s2_492, __s2_492, 1, 0); \
+ int64x2_t __ret_492; \
+ __ret_492 = __rev0_492 - __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_492), __noswap_splat_lane_s32(__rev2_492, __p3_492)); \
+ __ret_492 = __builtin_shufflevector(__ret_492, __ret_492, 1, 0); \
+ __ret_492; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = __s0 - vmull_s16(vget_high_s16(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlsl_high_lane_s16(__p0_493, __p1_493, __p2_493, __p3_493) __extension__ ({ \
+ int32x4_t __s0_493 = __p0_493; \
+ int16x8_t __s1_493 = __p1_493; \
+ int16x4_t __s2_493 = __p2_493; \
+ int32x4_t __ret_493; \
+ __ret_493 = __s0_493 - vmull_s16(vget_high_s16(__s1_493), splat_lane_s16(__s2_493, __p3_493)); \
+ __ret_493; \
})
#else
-#define vmlsl_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __rev0 - __noswap_vmull_s16(__noswap_vget_high_s16(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlsl_high_lane_s16(__p0_494, __p1_494, __p2_494, __p3_494) __extension__ ({ \
+ int32x4_t __s0_494 = __p0_494; \
+ int16x8_t __s1_494 = __p1_494; \
+ int16x4_t __s2_494 = __p2_494; \
+ int32x4_t __rev0_494; __rev0_494 = __builtin_shufflevector(__s0_494, __s0_494, 3, 2, 1, 0); \
+ int16x8_t __rev1_494; __rev1_494 = __builtin_shufflevector(__s1_494, __s1_494, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev2_494; __rev2_494 = __builtin_shufflevector(__s2_494, __s2_494, 3, 2, 1, 0); \
+ int32x4_t __ret_494; \
+ __ret_494 = __rev0_494 - __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_494), __noswap_splat_lane_s16(__rev2_494, __p3_494)); \
+ __ret_494 = __builtin_shufflevector(__ret_494, __ret_494, 3, 2, 1, 0); \
+ __ret_494; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_high_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint64x2_t __ret; \
- __ret = __s0 - vmull_u32(vget_high_u32(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlsl_high_laneq_u32(__p0_495, __p1_495, __p2_495, __p3_495) __extension__ ({ \
+ uint64x2_t __s0_495 = __p0_495; \
+ uint32x4_t __s1_495 = __p1_495; \
+ uint32x4_t __s2_495 = __p2_495; \
+ uint64x2_t __ret_495; \
+ __ret_495 = __s0_495 - vmull_u32(vget_high_u32(__s1_495), splat_laneq_u32(__s2_495, __p3_495)); \
+ __ret_495; \
})
#else
-#define vmlsl_high_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint64x2_t __ret; \
- __ret = __rev0 - __noswap_vmull_u32(__noswap_vget_high_u32(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmlsl_high_laneq_u32(__p0_496, __p1_496, __p2_496, __p3_496) __extension__ ({ \
+ uint64x2_t __s0_496 = __p0_496; \
+ uint32x4_t __s1_496 = __p1_496; \
+ uint32x4_t __s2_496 = __p2_496; \
+ uint64x2_t __rev0_496; __rev0_496 = __builtin_shufflevector(__s0_496, __s0_496, 1, 0); \
+ uint32x4_t __rev1_496; __rev1_496 = __builtin_shufflevector(__s1_496, __s1_496, 3, 2, 1, 0); \
+ uint32x4_t __rev2_496; __rev2_496 = __builtin_shufflevector(__s2_496, __s2_496, 3, 2, 1, 0); \
+ uint64x2_t __ret_496; \
+ __ret_496 = __rev0_496 - __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_496), __noswap_splat_laneq_u32(__rev2_496, __p3_496)); \
+ __ret_496 = __builtin_shufflevector(__ret_496, __ret_496, 1, 0); \
+ __ret_496; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_high_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint32x4_t __ret; \
- __ret = __s0 - vmull_u16(vget_high_u16(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlsl_high_laneq_u16(__p0_497, __p1_497, __p2_497, __p3_497) __extension__ ({ \
+ uint32x4_t __s0_497 = __p0_497; \
+ uint16x8_t __s1_497 = __p1_497; \
+ uint16x8_t __s2_497 = __p2_497; \
+ uint32x4_t __ret_497; \
+ __ret_497 = __s0_497 - vmull_u16(vget_high_u16(__s1_497), splat_laneq_u16(__s2_497, __p3_497)); \
+ __ret_497; \
})
#else
-#define vmlsl_high_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __rev0 - __noswap_vmull_u16(__noswap_vget_high_u16(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlsl_high_laneq_u16(__p0_498, __p1_498, __p2_498, __p3_498) __extension__ ({ \
+ uint32x4_t __s0_498 = __p0_498; \
+ uint16x8_t __s1_498 = __p1_498; \
+ uint16x8_t __s2_498 = __p2_498; \
+ uint32x4_t __rev0_498; __rev0_498 = __builtin_shufflevector(__s0_498, __s0_498, 3, 2, 1, 0); \
+ uint16x8_t __rev1_498; __rev1_498 = __builtin_shufflevector(__s1_498, __s1_498, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __rev2_498; __rev2_498 = __builtin_shufflevector(__s2_498, __s2_498, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint32x4_t __ret_498; \
+ __ret_498 = __rev0_498 - __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_498), __noswap_splat_laneq_u16(__rev2_498, __p3_498)); \
+ __ret_498 = __builtin_shufflevector(__ret_498, __ret_498, 3, 2, 1, 0); \
+ __ret_498; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = __s0 - vmull_s32(vget_high_s32(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlsl_high_laneq_s32(__p0_499, __p1_499, __p2_499, __p3_499) __extension__ ({ \
+ int64x2_t __s0_499 = __p0_499; \
+ int32x4_t __s1_499 = __p1_499; \
+ int32x4_t __s2_499 = __p2_499; \
+ int64x2_t __ret_499; \
+ __ret_499 = __s0_499 - vmull_s32(vget_high_s32(__s1_499), splat_laneq_s32(__s2_499, __p3_499)); \
+ __ret_499; \
})
#else
-#define vmlsl_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int64x2_t __ret; \
- __ret = __rev0 - __noswap_vmull_s32(__noswap_vget_high_s32(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmlsl_high_laneq_s32(__p0_500, __p1_500, __p2_500, __p3_500) __extension__ ({ \
+ int64x2_t __s0_500 = __p0_500; \
+ int32x4_t __s1_500 = __p1_500; \
+ int32x4_t __s2_500 = __p2_500; \
+ int64x2_t __rev0_500; __rev0_500 = __builtin_shufflevector(__s0_500, __s0_500, 1, 0); \
+ int32x4_t __rev1_500; __rev1_500 = __builtin_shufflevector(__s1_500, __s1_500, 3, 2, 1, 0); \
+ int32x4_t __rev2_500; __rev2_500 = __builtin_shufflevector(__s2_500, __s2_500, 3, 2, 1, 0); \
+ int64x2_t __ret_500; \
+ __ret_500 = __rev0_500 - __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_500), __noswap_splat_laneq_s32(__rev2_500, __p3_500)); \
+ __ret_500 = __builtin_shufflevector(__ret_500, __ret_500, 1, 0); \
+ __ret_500; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = __s0 - vmull_s16(vget_high_s16(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlsl_high_laneq_s16(__p0_501, __p1_501, __p2_501, __p3_501) __extension__ ({ \
+ int32x4_t __s0_501 = __p0_501; \
+ int16x8_t __s1_501 = __p1_501; \
+ int16x8_t __s2_501 = __p2_501; \
+ int32x4_t __ret_501; \
+ __ret_501 = __s0_501 - vmull_s16(vget_high_s16(__s1_501), splat_laneq_s16(__s2_501, __p3_501)); \
+ __ret_501; \
})
#else
-#define vmlsl_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __rev0 - __noswap_vmull_s16(__noswap_vget_high_s16(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlsl_high_laneq_s16(__p0_502, __p1_502, __p2_502, __p3_502) __extension__ ({ \
+ int32x4_t __s0_502 = __p0_502; \
+ int16x8_t __s1_502 = __p1_502; \
+ int16x8_t __s2_502 = __p2_502; \
+ int32x4_t __rev0_502; __rev0_502 = __builtin_shufflevector(__s0_502, __s0_502, 3, 2, 1, 0); \
+ int16x8_t __rev1_502; __rev1_502 = __builtin_shufflevector(__s1_502, __s1_502, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev2_502; __rev2_502 = __builtin_shufflevector(__s2_502, __s2_502, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_502; \
+ __ret_502 = __rev0_502 - __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_502), __noswap_splat_laneq_s16(__rev2_502, __p3_502)); \
+ __ret_502 = __builtin_shufflevector(__ret_502, __ret_502, 3, 2, 1, 0); \
+ __ret_502; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint64x2_t __ret; \
- __ret = __s0 - vmull_u32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlsl_laneq_u32(__p0_503, __p1_503, __p2_503, __p3_503) __extension__ ({ \
+ uint64x2_t __s0_503 = __p0_503; \
+ uint32x2_t __s1_503 = __p1_503; \
+ uint32x4_t __s2_503 = __p2_503; \
+ uint64x2_t __ret_503; \
+ __ret_503 = __s0_503 - vmull_u32(__s1_503, splat_laneq_u32(__s2_503, __p3_503)); \
+ __ret_503; \
})
-#else
-#define vmlsl_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x4_t __s2 = __p2; \
- uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint64x2_t __ret; \
- __ret = __rev0 - __noswap_vmull_u32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#else
+#define vmlsl_laneq_u32(__p0_504, __p1_504, __p2_504, __p3_504) __extension__ ({ \
+ uint64x2_t __s0_504 = __p0_504; \
+ uint32x2_t __s1_504 = __p1_504; \
+ uint32x4_t __s2_504 = __p2_504; \
+ uint64x2_t __rev0_504; __rev0_504 = __builtin_shufflevector(__s0_504, __s0_504, 1, 0); \
+ uint32x2_t __rev1_504; __rev1_504 = __builtin_shufflevector(__s1_504, __s1_504, 1, 0); \
+ uint32x4_t __rev2_504; __rev2_504 = __builtin_shufflevector(__s2_504, __s2_504, 3, 2, 1, 0); \
+ uint64x2_t __ret_504; \
+ __ret_504 = __rev0_504 - __noswap_vmull_u32(__rev1_504, __noswap_splat_laneq_u32(__rev2_504, __p3_504)); \
+ __ret_504 = __builtin_shufflevector(__ret_504, __ret_504, 1, 0); \
+ __ret_504; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint32x4_t __ret; \
- __ret = __s0 - vmull_u16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlsl_laneq_u16(__p0_505, __p1_505, __p2_505, __p3_505) __extension__ ({ \
+ uint32x4_t __s0_505 = __p0_505; \
+ uint16x4_t __s1_505 = __p1_505; \
+ uint16x8_t __s2_505 = __p2_505; \
+ uint32x4_t __ret_505; \
+ __ret_505 = __s0_505 - vmull_u16(__s1_505, splat_laneq_u16(__s2_505, __p3_505)); \
+ __ret_505; \
})
#else
-#define vmlsl_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x8_t __s2 = __p2; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __rev0 - __noswap_vmull_u16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlsl_laneq_u16(__p0_506, __p1_506, __p2_506, __p3_506) __extension__ ({ \
+ uint32x4_t __s0_506 = __p0_506; \
+ uint16x4_t __s1_506 = __p1_506; \
+ uint16x8_t __s2_506 = __p2_506; \
+ uint32x4_t __rev0_506; __rev0_506 = __builtin_shufflevector(__s0_506, __s0_506, 3, 2, 1, 0); \
+ uint16x4_t __rev1_506; __rev1_506 = __builtin_shufflevector(__s1_506, __s1_506, 3, 2, 1, 0); \
+ uint16x8_t __rev2_506; __rev2_506 = __builtin_shufflevector(__s2_506, __s2_506, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint32x4_t __ret_506; \
+ __ret_506 = __rev0_506 - __noswap_vmull_u16(__rev1_506, __noswap_splat_laneq_u16(__rev2_506, __p3_506)); \
+ __ret_506 = __builtin_shufflevector(__ret_506, __ret_506, 3, 2, 1, 0); \
+ __ret_506; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = __s0 - vmull_s32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlsl_laneq_s32(__p0_507, __p1_507, __p2_507, __p3_507) __extension__ ({ \
+ int64x2_t __s0_507 = __p0_507; \
+ int32x2_t __s1_507 = __p1_507; \
+ int32x4_t __s2_507 = __p2_507; \
+ int64x2_t __ret_507; \
+ __ret_507 = __s0_507 - vmull_s32(__s1_507, splat_laneq_s32(__s2_507, __p3_507)); \
+ __ret_507; \
})
#else
-#define vmlsl_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int64x2_t __ret; \
- __ret = __rev0 - __noswap_vmull_s32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmlsl_laneq_s32(__p0_508, __p1_508, __p2_508, __p3_508) __extension__ ({ \
+ int64x2_t __s0_508 = __p0_508; \
+ int32x2_t __s1_508 = __p1_508; \
+ int32x4_t __s2_508 = __p2_508; \
+ int64x2_t __rev0_508; __rev0_508 = __builtin_shufflevector(__s0_508, __s0_508, 1, 0); \
+ int32x2_t __rev1_508; __rev1_508 = __builtin_shufflevector(__s1_508, __s1_508, 1, 0); \
+ int32x4_t __rev2_508; __rev2_508 = __builtin_shufflevector(__s2_508, __s2_508, 3, 2, 1, 0); \
+ int64x2_t __ret_508; \
+ __ret_508 = __rev0_508 - __noswap_vmull_s32(__rev1_508, __noswap_splat_laneq_s32(__rev2_508, __p3_508)); \
+ __ret_508 = __builtin_shufflevector(__ret_508, __ret_508, 1, 0); \
+ __ret_508; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = __s0 - vmull_s16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlsl_laneq_s16(__p0_509, __p1_509, __p2_509, __p3_509) __extension__ ({ \
+ int32x4_t __s0_509 = __p0_509; \
+ int16x4_t __s1_509 = __p1_509; \
+ int16x8_t __s2_509 = __p2_509; \
+ int32x4_t __ret_509; \
+ __ret_509 = __s0_509 - vmull_s16(__s1_509, splat_laneq_s16(__s2_509, __p3_509)); \
+ __ret_509; \
})
#else
-#define vmlsl_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __rev0 - __noswap_vmull_s16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlsl_laneq_s16(__p0_510, __p1_510, __p2_510, __p3_510) __extension__ ({ \
+ int32x4_t __s0_510 = __p0_510; \
+ int16x4_t __s1_510 = __p1_510; \
+ int16x8_t __s2_510 = __p2_510; \
+ int32x4_t __rev0_510; __rev0_510 = __builtin_shufflevector(__s0_510, __s0_510, 3, 2, 1, 0); \
+ int16x4_t __rev1_510; __rev1_510 = __builtin_shufflevector(__s1_510, __s1_510, 3, 2, 1, 0); \
+ int16x8_t __rev2_510; __rev2_510 = __builtin_shufflevector(__s2_510, __s2_510, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_510; \
+ __ret_510 = __rev0_510 - __noswap_vmull_s16(__rev1_510, __noswap_splat_laneq_s16(__rev2_510, __p3_510)); \
+ __ret_510 = __builtin_shufflevector(__ret_510, __ret_510, 3, 2, 1, 0); \
+ __ret_510; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai poly64x1_t vmov_n_p64(poly64_t __p0) {
- poly64x1_t __ret;
- __ret = (poly64x1_t) {__p0};
- return __ret;
-}
-#else
__ai poly64x1_t vmov_n_p64(poly64_t __p0) {
poly64x1_t __ret;
__ret = (poly64x1_t) {__p0};
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai poly64x2_t vmovq_n_p64(poly64_t __p0) {
poly64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vmov_n_f64(float64_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) {__p0};
- return __ret;
-}
-#else
__ai float64x1_t vmov_n_f64(float64_t __p0) {
float64x1_t __ret;
__ret = (float64x1_t) {__p0};
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vmovl_high_u8(uint8x16_t __p0_124) {
- uint16x8_t __ret_124;
- uint8x8_t __a1_124 = vget_high_u8(__p0_124);
- __ret_124 = (uint16x8_t)(vshll_n_u8(__a1_124, 0));
- return __ret_124;
+__ai uint16x8_t vmovl_high_u8(uint8x16_t __p0_511) {
+ uint16x8_t __ret_511;
+ uint8x8_t __a1_511 = vget_high_u8(__p0_511);
+ __ret_511 = (uint16x8_t)(vshll_n_u8(__a1_511, 0));
+ return __ret_511;
}
#else
-__ai uint16x8_t vmovl_high_u8(uint8x16_t __p0_125) {
- uint8x16_t __rev0_125; __rev0_125 = __builtin_shufflevector(__p0_125, __p0_125, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- uint16x8_t __ret_125;
- uint8x8_t __a1_125 = __noswap_vget_high_u8(__rev0_125);
- __ret_125 = (uint16x8_t)(__noswap_vshll_n_u8(__a1_125, 0));
- __ret_125 = __builtin_shufflevector(__ret_125, __ret_125, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret_125;
+__ai uint16x8_t vmovl_high_u8(uint8x16_t __p0_512) {
+ uint8x16_t __rev0_512; __rev0_512 = __builtin_shufflevector(__p0_512, __p0_512, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint16x8_t __ret_512;
+ uint8x8_t __a1_512 = __noswap_vget_high_u8(__rev0_512);
+ __ret_512 = (uint16x8_t)(__noswap_vshll_n_u8(__a1_512, 0));
+ __ret_512 = __builtin_shufflevector(__ret_512, __ret_512, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret_512;
}
-__ai uint16x8_t __noswap_vmovl_high_u8(uint8x16_t __p0_126) {
- uint16x8_t __ret_126;
- uint8x8_t __a1_126 = __noswap_vget_high_u8(__p0_126);
- __ret_126 = (uint16x8_t)(__noswap_vshll_n_u8(__a1_126, 0));
- return __ret_126;
+__ai uint16x8_t __noswap_vmovl_high_u8(uint8x16_t __p0_513) {
+ uint16x8_t __ret_513;
+ uint8x8_t __a1_513 = __noswap_vget_high_u8(__p0_513);
+ __ret_513 = (uint16x8_t)(__noswap_vshll_n_u8(__a1_513, 0));
+ return __ret_513;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vmovl_high_u32(uint32x4_t __p0_127) {
- uint64x2_t __ret_127;
- uint32x2_t __a1_127 = vget_high_u32(__p0_127);
- __ret_127 = (uint64x2_t)(vshll_n_u32(__a1_127, 0));
- return __ret_127;
+__ai uint64x2_t vmovl_high_u32(uint32x4_t __p0_514) {
+ uint64x2_t __ret_514;
+ uint32x2_t __a1_514 = vget_high_u32(__p0_514);
+ __ret_514 = (uint64x2_t)(vshll_n_u32(__a1_514, 0));
+ return __ret_514;
}
#else
-__ai uint64x2_t vmovl_high_u32(uint32x4_t __p0_128) {
- uint32x4_t __rev0_128; __rev0_128 = __builtin_shufflevector(__p0_128, __p0_128, 3, 2, 1, 0);
- uint64x2_t __ret_128;
- uint32x2_t __a1_128 = __noswap_vget_high_u32(__rev0_128);
- __ret_128 = (uint64x2_t)(__noswap_vshll_n_u32(__a1_128, 0));
- __ret_128 = __builtin_shufflevector(__ret_128, __ret_128, 1, 0);
- return __ret_128;
+__ai uint64x2_t vmovl_high_u32(uint32x4_t __p0_515) {
+ uint32x4_t __rev0_515; __rev0_515 = __builtin_shufflevector(__p0_515, __p0_515, 3, 2, 1, 0);
+ uint64x2_t __ret_515;
+ uint32x2_t __a1_515 = __noswap_vget_high_u32(__rev0_515);
+ __ret_515 = (uint64x2_t)(__noswap_vshll_n_u32(__a1_515, 0));
+ __ret_515 = __builtin_shufflevector(__ret_515, __ret_515, 1, 0);
+ return __ret_515;
}
-__ai uint64x2_t __noswap_vmovl_high_u32(uint32x4_t __p0_129) {
- uint64x2_t __ret_129;
- uint32x2_t __a1_129 = __noswap_vget_high_u32(__p0_129);
- __ret_129 = (uint64x2_t)(__noswap_vshll_n_u32(__a1_129, 0));
- return __ret_129;
+__ai uint64x2_t __noswap_vmovl_high_u32(uint32x4_t __p0_516) {
+ uint64x2_t __ret_516;
+ uint32x2_t __a1_516 = __noswap_vget_high_u32(__p0_516);
+ __ret_516 = (uint64x2_t)(__noswap_vshll_n_u32(__a1_516, 0));
+ return __ret_516;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vmovl_high_u16(uint16x8_t __p0_130) {
- uint32x4_t __ret_130;
- uint16x4_t __a1_130 = vget_high_u16(__p0_130);
- __ret_130 = (uint32x4_t)(vshll_n_u16(__a1_130, 0));
- return __ret_130;
+__ai uint32x4_t vmovl_high_u16(uint16x8_t __p0_517) {
+ uint32x4_t __ret_517;
+ uint16x4_t __a1_517 = vget_high_u16(__p0_517);
+ __ret_517 = (uint32x4_t)(vshll_n_u16(__a1_517, 0));
+ return __ret_517;
}
#else
-__ai uint32x4_t vmovl_high_u16(uint16x8_t __p0_131) {
- uint16x8_t __rev0_131; __rev0_131 = __builtin_shufflevector(__p0_131, __p0_131, 7, 6, 5, 4, 3, 2, 1, 0);
- uint32x4_t __ret_131;
- uint16x4_t __a1_131 = __noswap_vget_high_u16(__rev0_131);
- __ret_131 = (uint32x4_t)(__noswap_vshll_n_u16(__a1_131, 0));
- __ret_131 = __builtin_shufflevector(__ret_131, __ret_131, 3, 2, 1, 0);
- return __ret_131;
+__ai uint32x4_t vmovl_high_u16(uint16x8_t __p0_518) {
+ uint16x8_t __rev0_518; __rev0_518 = __builtin_shufflevector(__p0_518, __p0_518, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint32x4_t __ret_518;
+ uint16x4_t __a1_518 = __noswap_vget_high_u16(__rev0_518);
+ __ret_518 = (uint32x4_t)(__noswap_vshll_n_u16(__a1_518, 0));
+ __ret_518 = __builtin_shufflevector(__ret_518, __ret_518, 3, 2, 1, 0);
+ return __ret_518;
}
-__ai uint32x4_t __noswap_vmovl_high_u16(uint16x8_t __p0_132) {
- uint32x4_t __ret_132;
- uint16x4_t __a1_132 = __noswap_vget_high_u16(__p0_132);
- __ret_132 = (uint32x4_t)(__noswap_vshll_n_u16(__a1_132, 0));
- return __ret_132;
+__ai uint32x4_t __noswap_vmovl_high_u16(uint16x8_t __p0_519) {
+ uint32x4_t __ret_519;
+ uint16x4_t __a1_519 = __noswap_vget_high_u16(__p0_519);
+ __ret_519 = (uint32x4_t)(__noswap_vshll_n_u16(__a1_519, 0));
+ return __ret_519;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vmovl_high_s8(int8x16_t __p0_133) {
- int16x8_t __ret_133;
- int8x8_t __a1_133 = vget_high_s8(__p0_133);
- __ret_133 = (int16x8_t)(vshll_n_s8(__a1_133, 0));
- return __ret_133;
+__ai int16x8_t vmovl_high_s8(int8x16_t __p0_520) {
+ int16x8_t __ret_520;
+ int8x8_t __a1_520 = vget_high_s8(__p0_520);
+ __ret_520 = (int16x8_t)(vshll_n_s8(__a1_520, 0));
+ return __ret_520;
}
#else
-__ai int16x8_t vmovl_high_s8(int8x16_t __p0_134) {
- int8x16_t __rev0_134; __rev0_134 = __builtin_shufflevector(__p0_134, __p0_134, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int16x8_t __ret_134;
- int8x8_t __a1_134 = __noswap_vget_high_s8(__rev0_134);
- __ret_134 = (int16x8_t)(__noswap_vshll_n_s8(__a1_134, 0));
- __ret_134 = __builtin_shufflevector(__ret_134, __ret_134, 7, 6, 5, 4, 3, 2, 1, 0);
- return __ret_134;
+__ai int16x8_t vmovl_high_s8(int8x16_t __p0_521) {
+ int8x16_t __rev0_521; __rev0_521 = __builtin_shufflevector(__p0_521, __p0_521, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ int16x8_t __ret_521;
+ int8x8_t __a1_521 = __noswap_vget_high_s8(__rev0_521);
+ __ret_521 = (int16x8_t)(__noswap_vshll_n_s8(__a1_521, 0));
+ __ret_521 = __builtin_shufflevector(__ret_521, __ret_521, 7, 6, 5, 4, 3, 2, 1, 0);
+ return __ret_521;
}
-__ai int16x8_t __noswap_vmovl_high_s8(int8x16_t __p0_135) {
- int16x8_t __ret_135;
- int8x8_t __a1_135 = __noswap_vget_high_s8(__p0_135);
- __ret_135 = (int16x8_t)(__noswap_vshll_n_s8(__a1_135, 0));
- return __ret_135;
+__ai int16x8_t __noswap_vmovl_high_s8(int8x16_t __p0_522) {
+ int16x8_t __ret_522;
+ int8x8_t __a1_522 = __noswap_vget_high_s8(__p0_522);
+ __ret_522 = (int16x8_t)(__noswap_vshll_n_s8(__a1_522, 0));
+ return __ret_522;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vmovl_high_s32(int32x4_t __p0_136) {
- int64x2_t __ret_136;
- int32x2_t __a1_136 = vget_high_s32(__p0_136);
- __ret_136 = (int64x2_t)(vshll_n_s32(__a1_136, 0));
- return __ret_136;
+__ai int64x2_t vmovl_high_s32(int32x4_t __p0_523) {
+ int64x2_t __ret_523;
+ int32x2_t __a1_523 = vget_high_s32(__p0_523);
+ __ret_523 = (int64x2_t)(vshll_n_s32(__a1_523, 0));
+ return __ret_523;
}
#else
-__ai int64x2_t vmovl_high_s32(int32x4_t __p0_137) {
- int32x4_t __rev0_137; __rev0_137 = __builtin_shufflevector(__p0_137, __p0_137, 3, 2, 1, 0);
- int64x2_t __ret_137;
- int32x2_t __a1_137 = __noswap_vget_high_s32(__rev0_137);
- __ret_137 = (int64x2_t)(__noswap_vshll_n_s32(__a1_137, 0));
- __ret_137 = __builtin_shufflevector(__ret_137, __ret_137, 1, 0);
- return __ret_137;
+__ai int64x2_t vmovl_high_s32(int32x4_t __p0_524) {
+ int32x4_t __rev0_524; __rev0_524 = __builtin_shufflevector(__p0_524, __p0_524, 3, 2, 1, 0);
+ int64x2_t __ret_524;
+ int32x2_t __a1_524 = __noswap_vget_high_s32(__rev0_524);
+ __ret_524 = (int64x2_t)(__noswap_vshll_n_s32(__a1_524, 0));
+ __ret_524 = __builtin_shufflevector(__ret_524, __ret_524, 1, 0);
+ return __ret_524;
}
-__ai int64x2_t __noswap_vmovl_high_s32(int32x4_t __p0_138) {
- int64x2_t __ret_138;
- int32x2_t __a1_138 = __noswap_vget_high_s32(__p0_138);
- __ret_138 = (int64x2_t)(__noswap_vshll_n_s32(__a1_138, 0));
- return __ret_138;
+__ai int64x2_t __noswap_vmovl_high_s32(int32x4_t __p0_525) {
+ int64x2_t __ret_525;
+ int32x2_t __a1_525 = __noswap_vget_high_s32(__p0_525);
+ __ret_525 = (int64x2_t)(__noswap_vshll_n_s32(__a1_525, 0));
+ return __ret_525;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vmovl_high_s16(int16x8_t __p0_139) {
- int32x4_t __ret_139;
- int16x4_t __a1_139 = vget_high_s16(__p0_139);
- __ret_139 = (int32x4_t)(vshll_n_s16(__a1_139, 0));
- return __ret_139;
+__ai int32x4_t vmovl_high_s16(int16x8_t __p0_526) {
+ int32x4_t __ret_526;
+ int16x4_t __a1_526 = vget_high_s16(__p0_526);
+ __ret_526 = (int32x4_t)(vshll_n_s16(__a1_526, 0));
+ return __ret_526;
}
#else
-__ai int32x4_t vmovl_high_s16(int16x8_t __p0_140) {
- int16x8_t __rev0_140; __rev0_140 = __builtin_shufflevector(__p0_140, __p0_140, 7, 6, 5, 4, 3, 2, 1, 0);
- int32x4_t __ret_140;
- int16x4_t __a1_140 = __noswap_vget_high_s16(__rev0_140);
- __ret_140 = (int32x4_t)(__noswap_vshll_n_s16(__a1_140, 0));
- __ret_140 = __builtin_shufflevector(__ret_140, __ret_140, 3, 2, 1, 0);
- return __ret_140;
+__ai int32x4_t vmovl_high_s16(int16x8_t __p0_527) {
+ int16x8_t __rev0_527; __rev0_527 = __builtin_shufflevector(__p0_527, __p0_527, 7, 6, 5, 4, 3, 2, 1, 0);
+ int32x4_t __ret_527;
+ int16x4_t __a1_527 = __noswap_vget_high_s16(__rev0_527);
+ __ret_527 = (int32x4_t)(__noswap_vshll_n_s16(__a1_527, 0));
+ __ret_527 = __builtin_shufflevector(__ret_527, __ret_527, 3, 2, 1, 0);
+ return __ret_527;
}
-__ai int32x4_t __noswap_vmovl_high_s16(int16x8_t __p0_141) {
- int32x4_t __ret_141;
- int16x4_t __a1_141 = __noswap_vget_high_s16(__p0_141);
- __ret_141 = (int32x4_t)(__noswap_vshll_n_s16(__a1_141, 0));
- return __ret_141;
+__ai int32x4_t __noswap_vmovl_high_s16(int16x8_t __p0_528) {
+ int32x4_t __ret_528;
+ int16x4_t __a1_528 = __noswap_vget_high_s16(__p0_528);
+ __ret_528 = (int32x4_t)(__noswap_vshll_n_s16(__a1_528, 0));
+ return __ret_528;
}
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
- float64x1_t __ret;
- __ret = __p0 * __p1;
- return __ret;
-}
-#else
__ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
float64x1_t __ret;
__ret = __p0 * __p1;
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vmuld_lane_f64(__p0_142, __p1_142, __p2_142) __extension__ ({ \
- float64_t __s0_142 = __p0_142; \
- float64x1_t __s1_142 = __p1_142; \
- float64_t __ret_142; \
- __ret_142 = __s0_142 * vget_lane_f64(__s1_142, __p2_142); \
- __ret_142; \
-})
-#else
-#define vmuld_lane_f64(__p0_143, __p1_143, __p2_143) __extension__ ({ \
- float64_t __s0_143 = __p0_143; \
- float64x1_t __s1_143 = __p1_143; \
- float64_t __ret_143; \
- __ret_143 = __s0_143 * __noswap_vget_lane_f64(__s1_143, __p2_143); \
- __ret_143; \
+#define vmuld_lane_f64(__p0_529, __p1_529, __p2_529) __extension__ ({ \
+ float64_t __s0_529 = __p0_529; \
+ float64x1_t __s1_529 = __p1_529; \
+ float64_t __ret_529; \
+ __ret_529 = __s0_529 * vget_lane_f64(__s1_529, __p2_529); \
+ __ret_529; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vmuls_lane_f32(__p0_144, __p1_144, __p2_144) __extension__ ({ \
- float32_t __s0_144 = __p0_144; \
- float32x2_t __s1_144 = __p1_144; \
- float32_t __ret_144; \
- __ret_144 = __s0_144 * vget_lane_f32(__s1_144, __p2_144); \
- __ret_144; \
+#define vmuls_lane_f32(__p0_530, __p1_530, __p2_530) __extension__ ({ \
+ float32_t __s0_530 = __p0_530; \
+ float32x2_t __s1_530 = __p1_530; \
+ float32_t __ret_530; \
+ __ret_530 = __s0_530 * vget_lane_f32(__s1_530, __p2_530); \
+ __ret_530; \
})
#else
-#define vmuls_lane_f32(__p0_145, __p1_145, __p2_145) __extension__ ({ \
- float32_t __s0_145 = __p0_145; \
- float32x2_t __s1_145 = __p1_145; \
- float32x2_t __rev1_145; __rev1_145 = __builtin_shufflevector(__s1_145, __s1_145, 1, 0); \
- float32_t __ret_145; \
- __ret_145 = __s0_145 * __noswap_vget_lane_f32(__rev1_145, __p2_145); \
- __ret_145; \
+#define vmuls_lane_f32(__p0_531, __p1_531, __p2_531) __extension__ ({ \
+ float32_t __s0_531 = __p0_531; \
+ float32x2_t __s1_531 = __p1_531; \
+ float32x2_t __rev1_531; __rev1_531 = __builtin_shufflevector(__s1_531, __s1_531, 1, 0); \
+ float32_t __ret_531; \
+ __ret_531 = __s0_531 * __noswap_vget_lane_f32(__rev1_531, __p2_531); \
+ __ret_531; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vmul_lane_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x1_t __s0 = __p0; \
- float64x1_t __s1 = __p1; \
- float64x1_t __ret; \
- __ret = (float64x1_t) __builtin_neon_vmul_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 10); \
- __ret; \
-})
-#else
#define vmul_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x1_t __s0 = __p0; \
float64x1_t __s1 = __p1; \
__ret = (float64x1_t) __builtin_neon_vmul_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 10); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vmulq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x2_t __s0 = __p0; \
- float64x1_t __s1 = __p1; \
- float64x2_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2); \
- __ret; \
+#define vmulq_lane_f64(__p0_532, __p1_532, __p2_532) __extension__ ({ \
+ float64x2_t __s0_532 = __p0_532; \
+ float64x1_t __s1_532 = __p1_532; \
+ float64x2_t __ret_532; \
+ __ret_532 = __s0_532 * splatq_lane_f64(__s1_532, __p2_532); \
+ __ret_532; \
})
#else
-#define vmulq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x2_t __s0 = __p0; \
- float64x1_t __s1 = __p1; \
- float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float64x2_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__s1, __s1, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmulq_lane_f64(__p0_533, __p1_533, __p2_533) __extension__ ({ \
+ float64x2_t __s0_533 = __p0_533; \
+ float64x1_t __s1_533 = __p1_533; \
+ float64x2_t __rev0_533; __rev0_533 = __builtin_shufflevector(__s0_533, __s0_533, 1, 0); \
+ float64x2_t __ret_533; \
+ __ret_533 = __rev0_533 * __noswap_splatq_lane_f64(__s1_533, __p2_533); \
+ __ret_533 = __builtin_shufflevector(__ret_533, __ret_533, 1, 0); \
+ __ret_533; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmuld_laneq_f64(__p0_146, __p1_146, __p2_146) __extension__ ({ \
- float64_t __s0_146 = __p0_146; \
- float64x2_t __s1_146 = __p1_146; \
- float64_t __ret_146; \
- __ret_146 = __s0_146 * vgetq_lane_f64(__s1_146, __p2_146); \
- __ret_146; \
+#define vmuld_laneq_f64(__p0_534, __p1_534, __p2_534) __extension__ ({ \
+ float64_t __s0_534 = __p0_534; \
+ float64x2_t __s1_534 = __p1_534; \
+ float64_t __ret_534; \
+ __ret_534 = __s0_534 * vgetq_lane_f64(__s1_534, __p2_534); \
+ __ret_534; \
})
#else
-#define vmuld_laneq_f64(__p0_147, __p1_147, __p2_147) __extension__ ({ \
- float64_t __s0_147 = __p0_147; \
- float64x2_t __s1_147 = __p1_147; \
- float64x2_t __rev1_147; __rev1_147 = __builtin_shufflevector(__s1_147, __s1_147, 1, 0); \
- float64_t __ret_147; \
- __ret_147 = __s0_147 * __noswap_vgetq_lane_f64(__rev1_147, __p2_147); \
- __ret_147; \
+#define vmuld_laneq_f64(__p0_535, __p1_535, __p2_535) __extension__ ({ \
+ float64_t __s0_535 = __p0_535; \
+ float64x2_t __s1_535 = __p1_535; \
+ float64x2_t __rev1_535; __rev1_535 = __builtin_shufflevector(__s1_535, __s1_535, 1, 0); \
+ float64_t __ret_535; \
+ __ret_535 = __s0_535 * __noswap_vgetq_lane_f64(__rev1_535, __p2_535); \
+ __ret_535; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmuls_laneq_f32(__p0_148, __p1_148, __p2_148) __extension__ ({ \
- float32_t __s0_148 = __p0_148; \
- float32x4_t __s1_148 = __p1_148; \
- float32_t __ret_148; \
- __ret_148 = __s0_148 * vgetq_lane_f32(__s1_148, __p2_148); \
- __ret_148; \
+#define vmuls_laneq_f32(__p0_536, __p1_536, __p2_536) __extension__ ({ \
+ float32_t __s0_536 = __p0_536; \
+ float32x4_t __s1_536 = __p1_536; \
+ float32_t __ret_536; \
+ __ret_536 = __s0_536 * vgetq_lane_f32(__s1_536, __p2_536); \
+ __ret_536; \
})
#else
-#define vmuls_laneq_f32(__p0_149, __p1_149, __p2_149) __extension__ ({ \
- float32_t __s0_149 = __p0_149; \
- float32x4_t __s1_149 = __p1_149; \
- float32x4_t __rev1_149; __rev1_149 = __builtin_shufflevector(__s1_149, __s1_149, 3, 2, 1, 0); \
- float32_t __ret_149; \
- __ret_149 = __s0_149 * __noswap_vgetq_lane_f32(__rev1_149, __p2_149); \
- __ret_149; \
+#define vmuls_laneq_f32(__p0_537, __p1_537, __p2_537) __extension__ ({ \
+ float32_t __s0_537 = __p0_537; \
+ float32x4_t __s1_537 = __p1_537; \
+ float32x4_t __rev1_537; __rev1_537 = __builtin_shufflevector(__s1_537, __s1_537, 3, 2, 1, 0); \
+ float32_t __ret_537; \
+ __ret_537 = __s0_537 * __noswap_vgetq_lane_f32(__rev1_537, __p2_537); \
+ __ret_537; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulq_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x4_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmulq_laneq_u32(__p0_538, __p1_538, __p2_538) __extension__ ({ \
+ uint32x4_t __s0_538 = __p0_538; \
+ uint32x4_t __s1_538 = __p1_538; \
+ uint32x4_t __ret_538; \
+ __ret_538 = __s0_538 * splatq_laneq_u32(__s1_538, __p2_538); \
+ __ret_538; \
})
#else
-#define vmulq_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmulq_laneq_u32(__p0_539, __p1_539, __p2_539) __extension__ ({ \
+ uint32x4_t __s0_539 = __p0_539; \
+ uint32x4_t __s1_539 = __p1_539; \
+ uint32x4_t __rev0_539; __rev0_539 = __builtin_shufflevector(__s0_539, __s0_539, 3, 2, 1, 0); \
+ uint32x4_t __rev1_539; __rev1_539 = __builtin_shufflevector(__s1_539, __s1_539, 3, 2, 1, 0); \
+ uint32x4_t __ret_539; \
+ __ret_539 = __rev0_539 * __noswap_splatq_laneq_u32(__rev1_539, __p2_539); \
+ __ret_539 = __builtin_shufflevector(__ret_539, __ret_539, 3, 2, 1, 0); \
+ __ret_539; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulq_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x8_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmulq_laneq_u16(__p0_540, __p1_540, __p2_540) __extension__ ({ \
+ uint16x8_t __s0_540 = __p0_540; \
+ uint16x8_t __s1_540 = __p1_540; \
+ uint16x8_t __ret_540; \
+ __ret_540 = __s0_540 * splatq_laneq_u16(__s1_540, __p2_540); \
+ __ret_540; \
})
#else
-#define vmulq_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmulq_laneq_u16(__p0_541, __p1_541, __p2_541) __extension__ ({ \
+ uint16x8_t __s0_541 = __p0_541; \
+ uint16x8_t __s1_541 = __p1_541; \
+ uint16x8_t __rev0_541; __rev0_541 = __builtin_shufflevector(__s0_541, __s0_541, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __rev1_541; __rev1_541 = __builtin_shufflevector(__s1_541, __s1_541, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __ret_541; \
+ __ret_541 = __rev0_541 * __noswap_splatq_laneq_u16(__rev1_541, __p2_541); \
+ __ret_541 = __builtin_shufflevector(__ret_541, __ret_541, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_541; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulq_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x2_t __s0 = __p0; \
- float64x2_t __s1 = __p1; \
- float64x2_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2); \
- __ret; \
+#define vmulq_laneq_f64(__p0_542, __p1_542, __p2_542) __extension__ ({ \
+ float64x2_t __s0_542 = __p0_542; \
+ float64x2_t __s1_542 = __p1_542; \
+ float64x2_t __ret_542; \
+ __ret_542 = __s0_542 * splatq_laneq_f64(__s1_542, __p2_542); \
+ __ret_542; \
})
#else
-#define vmulq_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x2_t __s0 = __p0; \
- float64x2_t __s1 = __p1; \
- float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- float64x2_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmulq_laneq_f64(__p0_543, __p1_543, __p2_543) __extension__ ({ \
+ float64x2_t __s0_543 = __p0_543; \
+ float64x2_t __s1_543 = __p1_543; \
+ float64x2_t __rev0_543; __rev0_543 = __builtin_shufflevector(__s0_543, __s0_543, 1, 0); \
+ float64x2_t __rev1_543; __rev1_543 = __builtin_shufflevector(__s1_543, __s1_543, 1, 0); \
+ float64x2_t __ret_543; \
+ __ret_543 = __rev0_543 * __noswap_splatq_laneq_f64(__rev1_543, __p2_543); \
+ __ret_543 = __builtin_shufflevector(__ret_543, __ret_543, 1, 0); \
+ __ret_543; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulq_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x4_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmulq_laneq_f32(__p0_544, __p1_544, __p2_544) __extension__ ({ \
+ float32x4_t __s0_544 = __p0_544; \
+ float32x4_t __s1_544 = __p1_544; \
+ float32x4_t __ret_544; \
+ __ret_544 = __s0_544 * splatq_laneq_f32(__s1_544, __p2_544); \
+ __ret_544; \
})
#else
-#define vmulq_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- float32x4_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmulq_laneq_f32(__p0_545, __p1_545, __p2_545) __extension__ ({ \
+ float32x4_t __s0_545 = __p0_545; \
+ float32x4_t __s1_545 = __p1_545; \
+ float32x4_t __rev0_545; __rev0_545 = __builtin_shufflevector(__s0_545, __s0_545, 3, 2, 1, 0); \
+ float32x4_t __rev1_545; __rev1_545 = __builtin_shufflevector(__s1_545, __s1_545, 3, 2, 1, 0); \
+ float32x4_t __ret_545; \
+ __ret_545 = __rev0_545 * __noswap_splatq_laneq_f32(__rev1_545, __p2_545); \
+ __ret_545 = __builtin_shufflevector(__ret_545, __ret_545, 3, 2, 1, 0); \
+ __ret_545; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmulq_laneq_s32(__p0_546, __p1_546, __p2_546) __extension__ ({ \
+ int32x4_t __s0_546 = __p0_546; \
+ int32x4_t __s1_546 = __p1_546; \
+ int32x4_t __ret_546; \
+ __ret_546 = __s0_546 * splatq_laneq_s32(__s1_546, __p2_546); \
+ __ret_546; \
})
#else
-#define vmulq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmulq_laneq_s32(__p0_547, __p1_547, __p2_547) __extension__ ({ \
+ int32x4_t __s0_547 = __p0_547; \
+ int32x4_t __s1_547 = __p1_547; \
+ int32x4_t __rev0_547; __rev0_547 = __builtin_shufflevector(__s0_547, __s0_547, 3, 2, 1, 0); \
+ int32x4_t __rev1_547; __rev1_547 = __builtin_shufflevector(__s1_547, __s1_547, 3, 2, 1, 0); \
+ int32x4_t __ret_547; \
+ __ret_547 = __rev0_547 * __noswap_splatq_laneq_s32(__rev1_547, __p2_547); \
+ __ret_547 = __builtin_shufflevector(__ret_547, __ret_547, 3, 2, 1, 0); \
+ __ret_547; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmulq_laneq_s16(__p0_548, __p1_548, __p2_548) __extension__ ({ \
+ int16x8_t __s0_548 = __p0_548; \
+ int16x8_t __s1_548 = __p1_548; \
+ int16x8_t __ret_548; \
+ __ret_548 = __s0_548 * splatq_laneq_s16(__s1_548, __p2_548); \
+ __ret_548; \
})
#else
-#define vmulq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret; \
+#define vmulq_laneq_s16(__p0_549, __p1_549, __p2_549) __extension__ ({ \
+ int16x8_t __s0_549 = __p0_549; \
+ int16x8_t __s1_549 = __p1_549; \
+ int16x8_t __rev0_549; __rev0_549 = __builtin_shufflevector(__s0_549, __s0_549, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_549; __rev1_549 = __builtin_shufflevector(__s1_549, __s1_549, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __ret_549; \
+ __ret_549 = __rev0_549 * __noswap_splatq_laneq_s16(__rev1_549, __p2_549); \
+ __ret_549 = __builtin_shufflevector(__ret_549, __ret_549, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_549; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmul_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x2_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2); \
- __ret; \
+#define vmul_laneq_u32(__p0_550, __p1_550, __p2_550) __extension__ ({ \
+ uint32x2_t __s0_550 = __p0_550; \
+ uint32x4_t __s1_550 = __p1_550; \
+ uint32x2_t __ret_550; \
+ __ret_550 = __s0_550 * splat_laneq_u32(__s1_550, __p2_550); \
+ __ret_550; \
})
#else
-#define vmul_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint32x2_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmul_laneq_u32(__p0_551, __p1_551, __p2_551) __extension__ ({ \
+ uint32x2_t __s0_551 = __p0_551; \
+ uint32x4_t __s1_551 = __p1_551; \
+ uint32x2_t __rev0_551; __rev0_551 = __builtin_shufflevector(__s0_551, __s0_551, 1, 0); \
+ uint32x4_t __rev1_551; __rev1_551 = __builtin_shufflevector(__s1_551, __s1_551, 3, 2, 1, 0); \
+ uint32x2_t __ret_551; \
+ __ret_551 = __rev0_551 * __noswap_splat_laneq_u32(__rev1_551, __p2_551); \
+ __ret_551 = __builtin_shufflevector(__ret_551, __ret_551, 1, 0); \
+ __ret_551; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmul_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x4_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmul_laneq_u16(__p0_552, __p1_552, __p2_552) __extension__ ({ \
+ uint16x4_t __s0_552 = __p0_552; \
+ uint16x8_t __s1_552 = __p1_552; \
+ uint16x4_t __ret_552; \
+ __ret_552 = __s0_552 * splat_laneq_u16(__s1_552, __p2_552); \
+ __ret_552; \
})
#else
-#define vmul_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x4_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmul_laneq_u16(__p0_553, __p1_553, __p2_553) __extension__ ({ \
+ uint16x4_t __s0_553 = __p0_553; \
+ uint16x8_t __s1_553 = __p1_553; \
+ uint16x4_t __rev0_553; __rev0_553 = __builtin_shufflevector(__s0_553, __s0_553, 3, 2, 1, 0); \
+ uint16x8_t __rev1_553; __rev1_553 = __builtin_shufflevector(__s1_553, __s1_553, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x4_t __ret_553; \
+ __ret_553 = __rev0_553 * __noswap_splat_laneq_u16(__rev1_553, __p2_553); \
+ __ret_553 = __builtin_shufflevector(__ret_553, __ret_553, 3, 2, 1, 0); \
+ __ret_553; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmul_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x2_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2); \
- __ret; \
+#define vmul_laneq_f32(__p0_554, __p1_554, __p2_554) __extension__ ({ \
+ float32x2_t __s0_554 = __p0_554; \
+ float32x4_t __s1_554 = __p1_554; \
+ float32x2_t __ret_554; \
+ __ret_554 = __s0_554 * splat_laneq_f32(__s1_554, __p2_554); \
+ __ret_554; \
})
#else
-#define vmul_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- float32x2_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmul_laneq_f32(__p0_555, __p1_555, __p2_555) __extension__ ({ \
+ float32x2_t __s0_555 = __p0_555; \
+ float32x4_t __s1_555 = __p1_555; \
+ float32x2_t __rev0_555; __rev0_555 = __builtin_shufflevector(__s0_555, __s0_555, 1, 0); \
+ float32x4_t __rev1_555; __rev1_555 = __builtin_shufflevector(__s1_555, __s1_555, 3, 2, 1, 0); \
+ float32x2_t __ret_555; \
+ __ret_555 = __rev0_555 * __noswap_splat_laneq_f32(__rev1_555, __p2_555); \
+ __ret_555 = __builtin_shufflevector(__ret_555, __ret_555, 1, 0); \
+ __ret_555; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmul_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2); \
- __ret; \
+#define vmul_laneq_s32(__p0_556, __p1_556, __p2_556) __extension__ ({ \
+ int32x2_t __s0_556 = __p0_556; \
+ int32x4_t __s1_556 = __p1_556; \
+ int32x2_t __ret_556; \
+ __ret_556 = __s0_556 * splat_laneq_s32(__s1_556, __p2_556); \
+ __ret_556; \
})
#else
-#define vmul_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x2_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmul_laneq_s32(__p0_557, __p1_557, __p2_557) __extension__ ({ \
+ int32x2_t __s0_557 = __p0_557; \
+ int32x4_t __s1_557 = __p1_557; \
+ int32x2_t __rev0_557; __rev0_557 = __builtin_shufflevector(__s0_557, __s0_557, 1, 0); \
+ int32x4_t __rev1_557; __rev1_557 = __builtin_shufflevector(__s1_557, __s1_557, 3, 2, 1, 0); \
+ int32x2_t __ret_557; \
+ __ret_557 = __rev0_557 * __noswap_splat_laneq_s32(__rev1_557, __p2_557); \
+ __ret_557 = __builtin_shufflevector(__ret_557, __ret_557, 1, 0); \
+ __ret_557; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmul_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __ret; \
- __ret = __s0 * __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2); \
- __ret; \
+#define vmul_laneq_s16(__p0_558, __p1_558, __p2_558) __extension__ ({ \
+ int16x4_t __s0_558 = __p0_558; \
+ int16x8_t __s1_558 = __p1_558; \
+ int16x4_t __ret_558; \
+ __ret_558 = __s0_558 * splat_laneq_s16(__s1_558, __p2_558); \
+ __ret_558; \
})
#else
-#define vmul_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __ret; \
- __ret = __rev0 * __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmul_laneq_s16(__p0_559, __p1_559, __p2_559) __extension__ ({ \
+ int16x4_t __s0_559 = __p0_559; \
+ int16x8_t __s1_559 = __p1_559; \
+ int16x4_t __rev0_559; __rev0_559 = __builtin_shufflevector(__s0_559, __s0_559, 3, 2, 1, 0); \
+ int16x8_t __rev1_559; __rev1_559 = __builtin_shufflevector(__s1_559, __s1_559, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __ret_559; \
+ __ret_559 = __rev0_559 * __noswap_splat_laneq_s16(__rev1_559, __p2_559); \
+ __ret_559 = __builtin_shufflevector(__ret_559, __ret_559, 3, 2, 1, 0); \
+ __ret_559; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vmul_n_f64(float64x1_t __p0, float64_t __p1) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vmul_n_f64((int8x8_t)__p0, __p1);
- return __ret;
-}
-#else
__ai float64x1_t vmul_n_f64(float64x1_t __p0, float64_t __p1) {
float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vmul_n_f64((int8x8_t)__p0, __p1);
+ __ret = (float64x1_t) __builtin_neon_vmul_n_f64((float64x1_t)__p0, __p1);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float64x2_t vmulq_n_f64(float64x2_t __p0, float64_t __p1) {
float64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai poly128_t vmull_p64(poly64_t __p0, poly64_t __p1) {
- poly128_t __ret;
- __ret = (poly128_t) __builtin_neon_vmull_p64(__p0, __p1);
- return __ret;
-}
-#else
__ai poly128_t vmull_p64(poly64_t __p0, poly64_t __p1) {
poly128_t __ret;
__ret = (poly128_t) __builtin_neon_vmull_p64(__p0, __p1);
return __ret;
}
-__ai poly128_t __noswap_vmull_p64(poly64_t __p0, poly64_t __p1) {
- poly128_t __ret;
- __ret = (poly128_t) __builtin_neon_vmull_p64(__p0, __p1);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai poly16x8_t vmull_high_p8(poly8x16_t __p0, poly8x16_t __p1) {
poly16x8_t __ret;
poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
poly128_t __ret;
- __ret = __noswap_vmull_p64((poly64_t)(__noswap_vget_high_p64(__rev0)), (poly64_t)(__noswap_vget_high_p64(__rev1)));
+ __ret = vmull_p64((poly64_t)(__noswap_vget_high_p64(__rev0)), (poly64_t)(__noswap_vget_high_p64(__rev1)));
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_high_lane_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint64x2_t __ret; \
- __ret = vmull_u32(vget_high_u32(__s0), __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vmull_high_lane_u32(__p0_560, __p1_560, __p2_560) __extension__ ({ \
+ uint32x4_t __s0_560 = __p0_560; \
+ uint32x2_t __s1_560 = __p1_560; \
+ uint64x2_t __ret_560; \
+ __ret_560 = vmull_u32(vget_high_u32(__s0_560), splat_lane_u32(__s1_560, __p2_560)); \
+ __ret_560; \
})
#else
-#define vmull_high_lane_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- uint64x2_t __ret; \
- __ret = __noswap_vmull_u32(__noswap_vget_high_u32(__rev0), __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmull_high_lane_u32(__p0_561, __p1_561, __p2_561) __extension__ ({ \
+ uint32x4_t __s0_561 = __p0_561; \
+ uint32x2_t __s1_561 = __p1_561; \
+ uint32x4_t __rev0_561; __rev0_561 = __builtin_shufflevector(__s0_561, __s0_561, 3, 2, 1, 0); \
+ uint32x2_t __rev1_561; __rev1_561 = __builtin_shufflevector(__s1_561, __s1_561, 1, 0); \
+ uint64x2_t __ret_561; \
+ __ret_561 = __noswap_vmull_u32(__noswap_vget_high_u32(__rev0_561), __noswap_splat_lane_u32(__rev1_561, __p2_561)); \
+ __ret_561 = __builtin_shufflevector(__ret_561, __ret_561, 1, 0); \
+ __ret_561; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_high_lane_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint32x4_t __ret; \
- __ret = vmull_u16(vget_high_u16(__s0), __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vmull_high_lane_u16(__p0_562, __p1_562, __p2_562) __extension__ ({ \
+ uint16x8_t __s0_562 = __p0_562; \
+ uint16x4_t __s1_562 = __p1_562; \
+ uint32x4_t __ret_562; \
+ __ret_562 = vmull_u16(vget_high_u16(__s0_562), splat_lane_u16(__s1_562, __p2_562)); \
+ __ret_562; \
})
#else
-#define vmull_high_lane_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __noswap_vmull_u16(__noswap_vget_high_u16(__rev0), __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmull_high_lane_u16(__p0_563, __p1_563, __p2_563) __extension__ ({ \
+ uint16x8_t __s0_563 = __p0_563; \
+ uint16x4_t __s1_563 = __p1_563; \
+ uint16x8_t __rev0_563; __rev0_563 = __builtin_shufflevector(__s0_563, __s0_563, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x4_t __rev1_563; __rev1_563 = __builtin_shufflevector(__s1_563, __s1_563, 3, 2, 1, 0); \
+ uint32x4_t __ret_563; \
+ __ret_563 = __noswap_vmull_u16(__noswap_vget_high_u16(__rev0_563), __noswap_splat_lane_u16(__rev1_563, __p2_563)); \
+ __ret_563 = __builtin_shufflevector(__ret_563, __ret_563, 3, 2, 1, 0); \
+ __ret_563; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_high_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int64x2_t __ret; \
- __ret = vmull_s32(vget_high_s32(__s0), __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vmull_high_lane_s32(__p0_564, __p1_564, __p2_564) __extension__ ({ \
+ int32x4_t __s0_564 = __p0_564; \
+ int32x2_t __s1_564 = __p1_564; \
+ int64x2_t __ret_564; \
+ __ret_564 = vmull_s32(vget_high_s32(__s0_564), splat_lane_s32(__s1_564, __p2_564)); \
+ __ret_564; \
})
#else
-#define vmull_high_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vmull_s32(__noswap_vget_high_s32(__rev0), __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmull_high_lane_s32(__p0_565, __p1_565, __p2_565) __extension__ ({ \
+ int32x4_t __s0_565 = __p0_565; \
+ int32x2_t __s1_565 = __p1_565; \
+ int32x4_t __rev0_565; __rev0_565 = __builtin_shufflevector(__s0_565, __s0_565, 3, 2, 1, 0); \
+ int32x2_t __rev1_565; __rev1_565 = __builtin_shufflevector(__s1_565, __s1_565, 1, 0); \
+ int64x2_t __ret_565; \
+ __ret_565 = __noswap_vmull_s32(__noswap_vget_high_s32(__rev0_565), __noswap_splat_lane_s32(__rev1_565, __p2_565)); \
+ __ret_565 = __builtin_shufflevector(__ret_565, __ret_565, 1, 0); \
+ __ret_565; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_high_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int32x4_t __ret; \
- __ret = vmull_s16(vget_high_s16(__s0), __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vmull_high_lane_s16(__p0_566, __p1_566, __p2_566) __extension__ ({ \
+ int16x8_t __s0_566 = __p0_566; \
+ int16x4_t __s1_566 = __p1_566; \
+ int32x4_t __ret_566; \
+ __ret_566 = vmull_s16(vget_high_s16(__s0_566), splat_lane_s16(__s1_566, __p2_566)); \
+ __ret_566; \
})
#else
-#define vmull_high_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vmull_s16(__noswap_vget_high_s16(__rev0), __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmull_high_lane_s16(__p0_567, __p1_567, __p2_567) __extension__ ({ \
+ int16x8_t __s0_567 = __p0_567; \
+ int16x4_t __s1_567 = __p1_567; \
+ int16x8_t __rev0_567; __rev0_567 = __builtin_shufflevector(__s0_567, __s0_567, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev1_567; __rev1_567 = __builtin_shufflevector(__s1_567, __s1_567, 3, 2, 1, 0); \
+ int32x4_t __ret_567; \
+ __ret_567 = __noswap_vmull_s16(__noswap_vget_high_s16(__rev0_567), __noswap_splat_lane_s16(__rev1_567, __p2_567)); \
+ __ret_567 = __builtin_shufflevector(__ret_567, __ret_567, 3, 2, 1, 0); \
+ __ret_567; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_high_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint64x2_t __ret; \
- __ret = vmull_u32(vget_high_u32(__s0), __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vmull_high_laneq_u32(__p0_568, __p1_568, __p2_568) __extension__ ({ \
+ uint32x4_t __s0_568 = __p0_568; \
+ uint32x4_t __s1_568 = __p1_568; \
+ uint64x2_t __ret_568; \
+ __ret_568 = vmull_u32(vget_high_u32(__s0_568), splat_laneq_u32(__s1_568, __p2_568)); \
+ __ret_568; \
})
#else
-#define vmull_high_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint64x2_t __ret; \
- __ret = __noswap_vmull_u32(__noswap_vget_high_u32(__rev0), __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmull_high_laneq_u32(__p0_569, __p1_569, __p2_569) __extension__ ({ \
+ uint32x4_t __s0_569 = __p0_569; \
+ uint32x4_t __s1_569 = __p1_569; \
+ uint32x4_t __rev0_569; __rev0_569 = __builtin_shufflevector(__s0_569, __s0_569, 3, 2, 1, 0); \
+ uint32x4_t __rev1_569; __rev1_569 = __builtin_shufflevector(__s1_569, __s1_569, 3, 2, 1, 0); \
+ uint64x2_t __ret_569; \
+ __ret_569 = __noswap_vmull_u32(__noswap_vget_high_u32(__rev0_569), __noswap_splat_laneq_u32(__rev1_569, __p2_569)); \
+ __ret_569 = __builtin_shufflevector(__ret_569, __ret_569, 1, 0); \
+ __ret_569; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_high_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint32x4_t __ret; \
- __ret = vmull_u16(vget_high_u16(__s0), __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vmull_high_laneq_u16(__p0_570, __p1_570, __p2_570) __extension__ ({ \
+ uint16x8_t __s0_570 = __p0_570; \
+ uint16x8_t __s1_570 = __p1_570; \
+ uint32x4_t __ret_570; \
+ __ret_570 = vmull_u16(vget_high_u16(__s0_570), splat_laneq_u16(__s1_570, __p2_570)); \
+ __ret_570; \
})
#else
-#define vmull_high_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x8_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __noswap_vmull_u16(__noswap_vget_high_u16(__rev0), __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmull_high_laneq_u16(__p0_571, __p1_571, __p2_571) __extension__ ({ \
+ uint16x8_t __s0_571 = __p0_571; \
+ uint16x8_t __s1_571 = __p1_571; \
+ uint16x8_t __rev0_571; __rev0_571 = __builtin_shufflevector(__s0_571, __s0_571, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __rev1_571; __rev1_571 = __builtin_shufflevector(__s1_571, __s1_571, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint32x4_t __ret_571; \
+ __ret_571 = __noswap_vmull_u16(__noswap_vget_high_u16(__rev0_571), __noswap_splat_laneq_u16(__rev1_571, __p2_571)); \
+ __ret_571 = __builtin_shufflevector(__ret_571, __ret_571, 3, 2, 1, 0); \
+ __ret_571; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_high_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int64x2_t __ret; \
- __ret = vmull_s32(vget_high_s32(__s0), __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vmull_high_laneq_s32(__p0_572, __p1_572, __p2_572) __extension__ ({ \
+ int32x4_t __s0_572 = __p0_572; \
+ int32x4_t __s1_572 = __p1_572; \
+ int64x2_t __ret_572; \
+ __ret_572 = vmull_s32(vget_high_s32(__s0_572), splat_laneq_s32(__s1_572, __p2_572)); \
+ __ret_572; \
})
#else
-#define vmull_high_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vmull_s32(__noswap_vget_high_s32(__rev0), __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmull_high_laneq_s32(__p0_573, __p1_573, __p2_573) __extension__ ({ \
+ int32x4_t __s0_573 = __p0_573; \
+ int32x4_t __s1_573 = __p1_573; \
+ int32x4_t __rev0_573; __rev0_573 = __builtin_shufflevector(__s0_573, __s0_573, 3, 2, 1, 0); \
+ int32x4_t __rev1_573; __rev1_573 = __builtin_shufflevector(__s1_573, __s1_573, 3, 2, 1, 0); \
+ int64x2_t __ret_573; \
+ __ret_573 = __noswap_vmull_s32(__noswap_vget_high_s32(__rev0_573), __noswap_splat_laneq_s32(__rev1_573, __p2_573)); \
+ __ret_573 = __builtin_shufflevector(__ret_573, __ret_573, 1, 0); \
+ __ret_573; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_high_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int32x4_t __ret; \
- __ret = vmull_s16(vget_high_s16(__s0), __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vmull_high_laneq_s16(__p0_574, __p1_574, __p2_574) __extension__ ({ \
+ int16x8_t __s0_574 = __p0_574; \
+ int16x8_t __s1_574 = __p1_574; \
+ int32x4_t __ret_574; \
+ __ret_574 = vmull_s16(vget_high_s16(__s0_574), splat_laneq_s16(__s1_574, __p2_574)); \
+ __ret_574; \
})
#else
-#define vmull_high_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vmull_s16(__noswap_vget_high_s16(__rev0), __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmull_high_laneq_s16(__p0_575, __p1_575, __p2_575) __extension__ ({ \
+ int16x8_t __s0_575 = __p0_575; \
+ int16x8_t __s1_575 = __p1_575; \
+ int16x8_t __rev0_575; __rev0_575 = __builtin_shufflevector(__s0_575, __s0_575, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_575; __rev1_575 = __builtin_shufflevector(__s1_575, __s1_575, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_575; \
+ __ret_575 = __noswap_vmull_s16(__noswap_vget_high_s16(__rev0_575), __noswap_splat_laneq_s16(__rev1_575, __p2_575)); \
+ __ret_575 = __builtin_shufflevector(__ret_575, __ret_575, 3, 2, 1, 0); \
+ __ret_575; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint64x2_t __ret; \
- __ret = vmull_u32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vmull_laneq_u32(__p0_576, __p1_576, __p2_576) __extension__ ({ \
+ uint32x2_t __s0_576 = __p0_576; \
+ uint32x4_t __s1_576 = __p1_576; \
+ uint64x2_t __ret_576; \
+ __ret_576 = vmull_u32(__s0_576, splat_laneq_u32(__s1_576, __p2_576)); \
+ __ret_576; \
})
#else
-#define vmull_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
- uint32x2_t __s0 = __p0; \
- uint32x4_t __s1 = __p1; \
- uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint64x2_t __ret; \
- __ret = __noswap_vmull_u32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmull_laneq_u32(__p0_577, __p1_577, __p2_577) __extension__ ({ \
+ uint32x2_t __s0_577 = __p0_577; \
+ uint32x4_t __s1_577 = __p1_577; \
+ uint32x2_t __rev0_577; __rev0_577 = __builtin_shufflevector(__s0_577, __s0_577, 1, 0); \
+ uint32x4_t __rev1_577; __rev1_577 = __builtin_shufflevector(__s1_577, __s1_577, 3, 2, 1, 0); \
+ uint64x2_t __ret_577; \
+ __ret_577 = __noswap_vmull_u32(__rev0_577, __noswap_splat_laneq_u32(__rev1_577, __p2_577)); \
+ __ret_577 = __builtin_shufflevector(__ret_577, __ret_577, 1, 0); \
+ __ret_577; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint32x4_t __ret; \
- __ret = vmull_u16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vmull_laneq_u16(__p0_578, __p1_578, __p2_578) __extension__ ({ \
+ uint16x4_t __s0_578 = __p0_578; \
+ uint16x8_t __s1_578 = __p1_578; \
+ uint32x4_t __ret_578; \
+ __ret_578 = vmull_u16(__s0_578, splat_laneq_u16(__s1_578, __p2_578)); \
+ __ret_578; \
})
#else
-#define vmull_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
- uint16x4_t __s0 = __p0; \
- uint16x8_t __s1 = __p1; \
- uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __noswap_vmull_u16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmull_laneq_u16(__p0_579, __p1_579, __p2_579) __extension__ ({ \
+ uint16x4_t __s0_579 = __p0_579; \
+ uint16x8_t __s1_579 = __p1_579; \
+ uint16x4_t __rev0_579; __rev0_579 = __builtin_shufflevector(__s0_579, __s0_579, 3, 2, 1, 0); \
+ uint16x8_t __rev1_579; __rev1_579 = __builtin_shufflevector(__s1_579, __s1_579, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint32x4_t __ret_579; \
+ __ret_579 = __noswap_vmull_u16(__rev0_579, __noswap_splat_laneq_u16(__rev1_579, __p2_579)); \
+ __ret_579 = __builtin_shufflevector(__ret_579, __ret_579, 3, 2, 1, 0); \
+ __ret_579; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int64x2_t __ret; \
- __ret = vmull_s32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vmull_laneq_s32(__p0_580, __p1_580, __p2_580) __extension__ ({ \
+ int32x2_t __s0_580 = __p0_580; \
+ int32x4_t __s1_580 = __p1_580; \
+ int64x2_t __ret_580; \
+ __ret_580 = vmull_s32(__s0_580, splat_laneq_s32(__s1_580, __p2_580)); \
+ __ret_580; \
})
#else
-#define vmull_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vmull_s32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmull_laneq_s32(__p0_581, __p1_581, __p2_581) __extension__ ({ \
+ int32x2_t __s0_581 = __p0_581; \
+ int32x4_t __s1_581 = __p1_581; \
+ int32x2_t __rev0_581; __rev0_581 = __builtin_shufflevector(__s0_581, __s0_581, 1, 0); \
+ int32x4_t __rev1_581; __rev1_581 = __builtin_shufflevector(__s1_581, __s1_581, 3, 2, 1, 0); \
+ int64x2_t __ret_581; \
+ __ret_581 = __noswap_vmull_s32(__rev0_581, __noswap_splat_laneq_s32(__rev1_581, __p2_581)); \
+ __ret_581 = __builtin_shufflevector(__ret_581, __ret_581, 1, 0); \
+ __ret_581; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmull_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int32x4_t __ret; \
- __ret = vmull_s16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vmull_laneq_s16(__p0_582, __p1_582, __p2_582) __extension__ ({ \
+ int16x4_t __s0_582 = __p0_582; \
+ int16x8_t __s1_582 = __p1_582; \
+ int32x4_t __ret_582; \
+ __ret_582 = vmull_s16(__s0_582, splat_laneq_s16(__s1_582, __p2_582)); \
+ __ret_582; \
})
#else
-#define vmull_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vmull_s16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmull_laneq_s16(__p0_583, __p1_583, __p2_583) __extension__ ({ \
+ int16x4_t __s0_583 = __p0_583; \
+ int16x8_t __s1_583 = __p1_583; \
+ int16x4_t __rev0_583; __rev0_583 = __builtin_shufflevector(__s0_583, __s0_583, 3, 2, 1, 0); \
+ int16x8_t __rev1_583; __rev1_583 = __builtin_shufflevector(__s1_583, __s1_583, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_583; \
+ __ret_583 = __noswap_vmull_s16(__rev0_583, __noswap_splat_laneq_s16(__rev1_583, __p2_583)); \
+ __ret_583 = __builtin_shufflevector(__ret_583, __ret_583, 3, 2, 1, 0); \
+ __ret_583; \
})
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai float64x1_t vmulx_f64(float64x1_t __p0, float64x1_t __p1) {
float64x1_t __ret;
__ret = (float64x1_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
return __ret;
}
-#else
-__ai float64x1_t vmulx_f64(float64x1_t __p0, float64x1_t __p1) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float32x2_t vmulx_f32(float32x2_t __p0, float32x2_t __p1) {
float32x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float64_t vmulxd_f64(float64_t __p0, float64_t __p1) {
- float64_t __ret;
- __ret = (float64_t) __builtin_neon_vmulxd_f64(__p0, __p1);
- return __ret;
-}
-#else
__ai float64_t vmulxd_f64(float64_t __p0, float64_t __p1) {
float64_t __ret;
__ret = (float64_t) __builtin_neon_vmulxd_f64(__p0, __p1);
return __ret;
}
-__ai float64_t __noswap_vmulxd_f64(float64_t __p0, float64_t __p1) {
- float64_t __ret;
- __ret = (float64_t) __builtin_neon_vmulxd_f64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
- float32_t __ret;
- __ret = (float32_t) __builtin_neon_vmulxs_f32(__p0, __p1);
- return __ret;
-}
-#else
__ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
float32_t __ret;
__ret = (float32_t) __builtin_neon_vmulxs_f32(__p0, __p1);
return __ret;
}
-__ai float32_t __noswap_vmulxs_f32(float32_t __p0, float32_t __p1) {
- float32_t __ret;
- __ret = (float32_t) __builtin_neon_vmulxs_f32(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vmulxd_lane_f64(__p0_150, __p1_150, __p2_150) __extension__ ({ \
- float64_t __s0_150 = __p0_150; \
- float64x1_t __s1_150 = __p1_150; \
- float64_t __ret_150; \
- __ret_150 = vmulxd_f64(__s0_150, vget_lane_f64(__s1_150, __p2_150)); \
- __ret_150; \
-})
-#else
-#define vmulxd_lane_f64(__p0_151, __p1_151, __p2_151) __extension__ ({ \
- float64_t __s0_151 = __p0_151; \
- float64x1_t __s1_151 = __p1_151; \
- float64_t __ret_151; \
- __ret_151 = __noswap_vmulxd_f64(__s0_151, __noswap_vget_lane_f64(__s1_151, __p2_151)); \
- __ret_151; \
+#define vmulxd_lane_f64(__p0_584, __p1_584, __p2_584) __extension__ ({ \
+ float64_t __s0_584 = __p0_584; \
+ float64x1_t __s1_584 = __p1_584; \
+ float64_t __ret_584; \
+ __ret_584 = vmulxd_f64(__s0_584, vget_lane_f64(__s1_584, __p2_584)); \
+ __ret_584; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vmulxs_lane_f32(__p0_152, __p1_152, __p2_152) __extension__ ({ \
- float32_t __s0_152 = __p0_152; \
- float32x2_t __s1_152 = __p1_152; \
- float32_t __ret_152; \
- __ret_152 = vmulxs_f32(__s0_152, vget_lane_f32(__s1_152, __p2_152)); \
- __ret_152; \
+#define vmulxs_lane_f32(__p0_585, __p1_585, __p2_585) __extension__ ({ \
+ float32_t __s0_585 = __p0_585; \
+ float32x2_t __s1_585 = __p1_585; \
+ float32_t __ret_585; \
+ __ret_585 = vmulxs_f32(__s0_585, vget_lane_f32(__s1_585, __p2_585)); \
+ __ret_585; \
})
#else
-#define vmulxs_lane_f32(__p0_153, __p1_153, __p2_153) __extension__ ({ \
- float32_t __s0_153 = __p0_153; \
- float32x2_t __s1_153 = __p1_153; \
- float32x2_t __rev1_153; __rev1_153 = __builtin_shufflevector(__s1_153, __s1_153, 1, 0); \
- float32_t __ret_153; \
- __ret_153 = __noswap_vmulxs_f32(__s0_153, __noswap_vget_lane_f32(__rev1_153, __p2_153)); \
- __ret_153; \
+#define vmulxs_lane_f32(__p0_586, __p1_586, __p2_586) __extension__ ({ \
+ float32_t __s0_586 = __p0_586; \
+ float32x2_t __s1_586 = __p1_586; \
+ float32x2_t __rev1_586; __rev1_586 = __builtin_shufflevector(__s1_586, __s1_586, 1, 0); \
+ float32_t __ret_586; \
+ __ret_586 = vmulxs_f32(__s0_586, __noswap_vget_lane_f32(__rev1_586, __p2_586)); \
+ __ret_586; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulxq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x2_t __s0 = __p0; \
- float64x1_t __s1 = __p1; \
- float64x2_t __ret; \
- __ret = vmulxq_f64(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vmulxq_lane_f64(__p0_587, __p1_587, __p2_587) __extension__ ({ \
+ float64x2_t __s0_587 = __p0_587; \
+ float64x1_t __s1_587 = __p1_587; \
+ float64x2_t __ret_587; \
+ __ret_587 = vmulxq_f64(__s0_587, splatq_lane_f64(__s1_587, __p2_587)); \
+ __ret_587; \
})
#else
-#define vmulxq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x2_t __s0 = __p0; \
- float64x1_t __s1 = __p1; \
- float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float64x2_t __ret; \
- __ret = __noswap_vmulxq_f64(__rev0, __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmulxq_lane_f64(__p0_588, __p1_588, __p2_588) __extension__ ({ \
+ float64x2_t __s0_588 = __p0_588; \
+ float64x1_t __s1_588 = __p1_588; \
+ float64x2_t __rev0_588; __rev0_588 = __builtin_shufflevector(__s0_588, __s0_588, 1, 0); \
+ float64x2_t __ret_588; \
+ __ret_588 = __noswap_vmulxq_f64(__rev0_588, __noswap_splatq_lane_f64(__s1_588, __p2_588)); \
+ __ret_588 = __builtin_shufflevector(__ret_588, __ret_588, 1, 0); \
+ __ret_588; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulxq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x4_t __ret; \
- __ret = vmulxq_f32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vmulxq_lane_f32(__p0_589, __p1_589, __p2_589) __extension__ ({ \
+ float32x4_t __s0_589 = __p0_589; \
+ float32x2_t __s1_589 = __p1_589; \
+ float32x4_t __ret_589; \
+ __ret_589 = vmulxq_f32(__s0_589, splatq_lane_f32(__s1_589, __p2_589)); \
+ __ret_589; \
})
#else
-#define vmulxq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- float32x4_t __ret; \
- __ret = __noswap_vmulxq_f32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmulxq_lane_f32(__p0_590, __p1_590, __p2_590) __extension__ ({ \
+ float32x4_t __s0_590 = __p0_590; \
+ float32x2_t __s1_590 = __p1_590; \
+ float32x4_t __rev0_590; __rev0_590 = __builtin_shufflevector(__s0_590, __s0_590, 3, 2, 1, 0); \
+ float32x2_t __rev1_590; __rev1_590 = __builtin_shufflevector(__s1_590, __s1_590, 1, 0); \
+ float32x4_t __ret_590; \
+ __ret_590 = __noswap_vmulxq_f32(__rev0_590, __noswap_splatq_lane_f32(__rev1_590, __p2_590)); \
+ __ret_590 = __builtin_shufflevector(__ret_590, __ret_590, 3, 2, 1, 0); \
+ __ret_590; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulx_lane_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x2_t __ret; \
- __ret = vmulx_f32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vmulx_lane_f32(__p0_591, __p1_591, __p2_591) __extension__ ({ \
+ float32x2_t __s0_591 = __p0_591; \
+ float32x2_t __s1_591 = __p1_591; \
+ float32x2_t __ret_591; \
+ __ret_591 = vmulx_f32(__s0_591, splat_lane_f32(__s1_591, __p2_591)); \
+ __ret_591; \
})
#else
-#define vmulx_lane_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x2_t __s1 = __p1; \
- float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- float32x2_t __ret; \
- __ret = __noswap_vmulx_f32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmulx_lane_f32(__p0_592, __p1_592, __p2_592) __extension__ ({ \
+ float32x2_t __s0_592 = __p0_592; \
+ float32x2_t __s1_592 = __p1_592; \
+ float32x2_t __rev0_592; __rev0_592 = __builtin_shufflevector(__s0_592, __s0_592, 1, 0); \
+ float32x2_t __rev1_592; __rev1_592 = __builtin_shufflevector(__s1_592, __s1_592, 1, 0); \
+ float32x2_t __ret_592; \
+ __ret_592 = __noswap_vmulx_f32(__rev0_592, __noswap_splat_lane_f32(__rev1_592, __p2_592)); \
+ __ret_592 = __builtin_shufflevector(__ret_592, __ret_592, 1, 0); \
+ __ret_592; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulxd_laneq_f64(__p0_154, __p1_154, __p2_154) __extension__ ({ \
- float64_t __s0_154 = __p0_154; \
- float64x2_t __s1_154 = __p1_154; \
- float64_t __ret_154; \
- __ret_154 = vmulxd_f64(__s0_154, vgetq_lane_f64(__s1_154, __p2_154)); \
- __ret_154; \
+#define vmulxd_laneq_f64(__p0_593, __p1_593, __p2_593) __extension__ ({ \
+ float64_t __s0_593 = __p0_593; \
+ float64x2_t __s1_593 = __p1_593; \
+ float64_t __ret_593; \
+ __ret_593 = vmulxd_f64(__s0_593, vgetq_lane_f64(__s1_593, __p2_593)); \
+ __ret_593; \
})
#else
-#define vmulxd_laneq_f64(__p0_155, __p1_155, __p2_155) __extension__ ({ \
- float64_t __s0_155 = __p0_155; \
- float64x2_t __s1_155 = __p1_155; \
- float64x2_t __rev1_155; __rev1_155 = __builtin_shufflevector(__s1_155, __s1_155, 1, 0); \
- float64_t __ret_155; \
- __ret_155 = __noswap_vmulxd_f64(__s0_155, __noswap_vgetq_lane_f64(__rev1_155, __p2_155)); \
- __ret_155; \
+#define vmulxd_laneq_f64(__p0_594, __p1_594, __p2_594) __extension__ ({ \
+ float64_t __s0_594 = __p0_594; \
+ float64x2_t __s1_594 = __p1_594; \
+ float64x2_t __rev1_594; __rev1_594 = __builtin_shufflevector(__s1_594, __s1_594, 1, 0); \
+ float64_t __ret_594; \
+ __ret_594 = vmulxd_f64(__s0_594, __noswap_vgetq_lane_f64(__rev1_594, __p2_594)); \
+ __ret_594; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulxs_laneq_f32(__p0_156, __p1_156, __p2_156) __extension__ ({ \
- float32_t __s0_156 = __p0_156; \
- float32x4_t __s1_156 = __p1_156; \
- float32_t __ret_156; \
- __ret_156 = vmulxs_f32(__s0_156, vgetq_lane_f32(__s1_156, __p2_156)); \
- __ret_156; \
+#define vmulxs_laneq_f32(__p0_595, __p1_595, __p2_595) __extension__ ({ \
+ float32_t __s0_595 = __p0_595; \
+ float32x4_t __s1_595 = __p1_595; \
+ float32_t __ret_595; \
+ __ret_595 = vmulxs_f32(__s0_595, vgetq_lane_f32(__s1_595, __p2_595)); \
+ __ret_595; \
})
#else
-#define vmulxs_laneq_f32(__p0_157, __p1_157, __p2_157) __extension__ ({ \
- float32_t __s0_157 = __p0_157; \
- float32x4_t __s1_157 = __p1_157; \
- float32x4_t __rev1_157; __rev1_157 = __builtin_shufflevector(__s1_157, __s1_157, 3, 2, 1, 0); \
- float32_t __ret_157; \
- __ret_157 = __noswap_vmulxs_f32(__s0_157, __noswap_vgetq_lane_f32(__rev1_157, __p2_157)); \
- __ret_157; \
+#define vmulxs_laneq_f32(__p0_596, __p1_596, __p2_596) __extension__ ({ \
+ float32_t __s0_596 = __p0_596; \
+ float32x4_t __s1_596 = __p1_596; \
+ float32x4_t __rev1_596; __rev1_596 = __builtin_shufflevector(__s1_596, __s1_596, 3, 2, 1, 0); \
+ float32_t __ret_596; \
+ __ret_596 = vmulxs_f32(__s0_596, __noswap_vgetq_lane_f32(__rev1_596, __p2_596)); \
+ __ret_596; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulxq_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x2_t __s0 = __p0; \
- float64x2_t __s1 = __p1; \
- float64x2_t __ret; \
- __ret = vmulxq_f64(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vmulxq_laneq_f64(__p0_597, __p1_597, __p2_597) __extension__ ({ \
+ float64x2_t __s0_597 = __p0_597; \
+ float64x2_t __s1_597 = __p1_597; \
+ float64x2_t __ret_597; \
+ __ret_597 = vmulxq_f64(__s0_597, splatq_laneq_f64(__s1_597, __p2_597)); \
+ __ret_597; \
})
#else
-#define vmulxq_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x2_t __s0 = __p0; \
- float64x2_t __s1 = __p1; \
- float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- float64x2_t __ret; \
- __ret = __noswap_vmulxq_f64(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmulxq_laneq_f64(__p0_598, __p1_598, __p2_598) __extension__ ({ \
+ float64x2_t __s0_598 = __p0_598; \
+ float64x2_t __s1_598 = __p1_598; \
+ float64x2_t __rev0_598; __rev0_598 = __builtin_shufflevector(__s0_598, __s0_598, 1, 0); \
+ float64x2_t __rev1_598; __rev1_598 = __builtin_shufflevector(__s1_598, __s1_598, 1, 0); \
+ float64x2_t __ret_598; \
+ __ret_598 = __noswap_vmulxq_f64(__rev0_598, __noswap_splatq_laneq_f64(__rev1_598, __p2_598)); \
+ __ret_598 = __builtin_shufflevector(__ret_598, __ret_598, 1, 0); \
+ __ret_598; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulxq_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x4_t __ret; \
- __ret = vmulxq_f32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vmulxq_laneq_f32(__p0_599, __p1_599, __p2_599) __extension__ ({ \
+ float32x4_t __s0_599 = __p0_599; \
+ float32x4_t __s1_599 = __p1_599; \
+ float32x4_t __ret_599; \
+ __ret_599 = vmulxq_f32(__s0_599, splatq_laneq_f32(__s1_599, __p2_599)); \
+ __ret_599; \
})
#else
-#define vmulxq_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x4_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- float32x4_t __ret; \
- __ret = __noswap_vmulxq_f32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmulxq_laneq_f32(__p0_600, __p1_600, __p2_600) __extension__ ({ \
+ float32x4_t __s0_600 = __p0_600; \
+ float32x4_t __s1_600 = __p1_600; \
+ float32x4_t __rev0_600; __rev0_600 = __builtin_shufflevector(__s0_600, __s0_600, 3, 2, 1, 0); \
+ float32x4_t __rev1_600; __rev1_600 = __builtin_shufflevector(__s1_600, __s1_600, 3, 2, 1, 0); \
+ float32x4_t __ret_600; \
+ __ret_600 = __noswap_vmulxq_f32(__rev0_600, __noswap_splatq_laneq_f32(__rev1_600, __p2_600)); \
+ __ret_600 = __builtin_shufflevector(__ret_600, __ret_600, 3, 2, 1, 0); \
+ __ret_600; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulx_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x2_t __ret; \
- __ret = vmulx_f32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vmulx_laneq_f32(__p0_601, __p1_601, __p2_601) __extension__ ({ \
+ float32x2_t __s0_601 = __p0_601; \
+ float32x4_t __s1_601 = __p1_601; \
+ float32x2_t __ret_601; \
+ __ret_601 = vmulx_f32(__s0_601, splat_laneq_f32(__s1_601, __p2_601)); \
+ __ret_601; \
})
#else
-#define vmulx_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
- float32x2_t __s0 = __p0; \
- float32x4_t __s1 = __p1; \
- float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- float32x2_t __ret; \
- __ret = __noswap_vmulx_f32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmulx_laneq_f32(__p0_602, __p1_602, __p2_602) __extension__ ({ \
+ float32x2_t __s0_602 = __p0_602; \
+ float32x4_t __s1_602 = __p1_602; \
+ float32x2_t __rev0_602; __rev0_602 = __builtin_shufflevector(__s0_602, __s0_602, 1, 0); \
+ float32x4_t __rev1_602; __rev1_602 = __builtin_shufflevector(__s1_602, __s1_602, 3, 2, 1, 0); \
+ float32x2_t __ret_602; \
+ __ret_602 = __noswap_vmulx_f32(__rev0_602, __noswap_splat_laneq_f32(__rev1_602, __p2_602)); \
+ __ret_602 = __builtin_shufflevector(__ret_602, __ret_602, 1, 0); \
+ __ret_602; \
})
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vneg_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = -__p0;
- return __ret;
-}
-#else
__ai float64x1_t vneg_f64(float64x1_t __p0) {
float64x1_t __ret;
__ret = -__p0;
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vneg_s64(int64x1_t __p0) {
- int64x1_t __ret;
- __ret = -__p0;
- return __ret;
-}
-#else
__ai int64x1_t vneg_s64(int64x1_t __p0) {
int64x1_t __ret;
__ret = -__p0;
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vnegd_s64(int64_t __p0) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vnegd_s64(__p0);
- return __ret;
-}
-#else
__ai int64_t vnegd_s64(int64_t __p0) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vnegd_s64(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint8x16_t vpaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
uint8x16_t __ret;
#ifdef __LITTLE_ENDIAN__
__ai uint64_t vpaddd_u64(uint64x2_t __p0) {
uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vpaddd_u64((int8x16_t)__p0);
+ __ret = (uint64_t) __builtin_neon_vpaddd_u64(__p0);
return __ret;
}
#else
__ai uint64_t vpaddd_u64(uint64x2_t __p0) {
uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vpaddd_u64((int8x16_t)__rev0);
+ __ret = (uint64_t) __builtin_neon_vpaddd_u64(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float64_t vpaddd_f64(float64x2_t __p0) {
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vpaddd_f64((int8x16_t)__p0);
+ __ret = (float64_t) __builtin_neon_vpaddd_f64(__p0);
return __ret;
}
#else
__ai float64_t vpaddd_f64(float64x2_t __p0) {
float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vpaddd_f64((int8x16_t)__rev0);
+ __ret = (float64_t) __builtin_neon_vpaddd_f64(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai int64_t vpaddd_s64(int64x2_t __p0) {
int64_t __ret;
- __ret = (int64_t) __builtin_neon_vpaddd_s64((int8x16_t)__p0);
+ __ret = (int64_t) __builtin_neon_vpaddd_s64(__p0);
return __ret;
}
#else
__ai int64_t vpaddd_s64(int64x2_t __p0) {
int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
int64_t __ret;
- __ret = (int64_t) __builtin_neon_vpaddd_s64((int8x16_t)__rev0);
+ __ret = (int64_t) __builtin_neon_vpaddd_s64(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float32_t vpadds_f32(float32x2_t __p0) {
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vpadds_f32((int8x8_t)__p0);
+ __ret = (float32_t) __builtin_neon_vpadds_f32(__p0);
return __ret;
}
#else
__ai float32_t vpadds_f32(float32x2_t __p0) {
float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vpadds_f32((int8x8_t)__rev0);
+ __ret = (float32_t) __builtin_neon_vpadds_f32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float64_t vpmaxqd_f64(float64x2_t __p0) {
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vpmaxqd_f64((int8x16_t)__p0);
+ __ret = (float64_t) __builtin_neon_vpmaxqd_f64(__p0);
return __ret;
}
#else
__ai float64_t vpmaxqd_f64(float64x2_t __p0) {
float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vpmaxqd_f64((int8x16_t)__rev0);
+ __ret = (float64_t) __builtin_neon_vpmaxqd_f64(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float32_t vpmaxs_f32(float32x2_t __p0) {
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vpmaxs_f32((int8x8_t)__p0);
+ __ret = (float32_t) __builtin_neon_vpmaxs_f32(__p0);
return __ret;
}
#else
__ai float32_t vpmaxs_f32(float32x2_t __p0) {
float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vpmaxs_f32((int8x8_t)__rev0);
+ __ret = (float32_t) __builtin_neon_vpmaxs_f32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float64_t vpmaxnmqd_f64(float64x2_t __p0) {
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vpmaxnmqd_f64((int8x16_t)__p0);
+ __ret = (float64_t) __builtin_neon_vpmaxnmqd_f64(__p0);
return __ret;
}
#else
__ai float64_t vpmaxnmqd_f64(float64x2_t __p0) {
float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vpmaxnmqd_f64((int8x16_t)__rev0);
+ __ret = (float64_t) __builtin_neon_vpmaxnmqd_f64(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float32_t vpmaxnms_f32(float32x2_t __p0) {
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vpmaxnms_f32((int8x8_t)__p0);
+ __ret = (float32_t) __builtin_neon_vpmaxnms_f32(__p0);
return __ret;
}
#else
__ai float32_t vpmaxnms_f32(float32x2_t __p0) {
float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vpmaxnms_f32((int8x8_t)__rev0);
+ __ret = (float32_t) __builtin_neon_vpmaxnms_f32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float64_t vpminqd_f64(float64x2_t __p0) {
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vpminqd_f64((int8x16_t)__p0);
+ __ret = (float64_t) __builtin_neon_vpminqd_f64(__p0);
return __ret;
}
#else
__ai float64_t vpminqd_f64(float64x2_t __p0) {
float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vpminqd_f64((int8x16_t)__rev0);
+ __ret = (float64_t) __builtin_neon_vpminqd_f64(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float32_t vpmins_f32(float32x2_t __p0) {
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vpmins_f32((int8x8_t)__p0);
+ __ret = (float32_t) __builtin_neon_vpmins_f32(__p0);
return __ret;
}
#else
__ai float32_t vpmins_f32(float32x2_t __p0) {
float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vpmins_f32((int8x8_t)__rev0);
+ __ret = (float32_t) __builtin_neon_vpmins_f32(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float64_t vpminnmqd_f64(float64x2_t __p0) {
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vpminnmqd_f64((int8x16_t)__p0);
+ __ret = (float64_t) __builtin_neon_vpminnmqd_f64(__p0);
return __ret;
}
#else
__ai float64_t vpminnmqd_f64(float64x2_t __p0) {
float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float64_t __ret;
- __ret = (float64_t) __builtin_neon_vpminnmqd_f64((int8x16_t)__rev0);
+ __ret = (float64_t) __builtin_neon_vpminnmqd_f64(__rev0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
__ai float32_t vpminnms_f32(float32x2_t __p0) {
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vpminnms_f32((int8x8_t)__p0);
+ __ret = (float32_t) __builtin_neon_vpminnms_f32(__p0);
return __ret;
}
#else
__ai float32_t vpminnms_f32(float32x2_t __p0) {
float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
float32_t __ret;
- __ret = (float32_t) __builtin_neon_vpminnms_f32((int8x8_t)__rev0);
+ __ret = (float32_t) __builtin_neon_vpminnms_f32(__rev0);
return __ret;
}
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vqabs_s64(int64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vqabs_v((int8x8_t)__p0, 3);
- return __ret;
-}
-#else
__ai int64x1_t vqabs_s64(int64x1_t __p0) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vqabs_v((int8x8_t)__p0, 3);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai int8_t vqabsb_s8(int8_t __p0) {
int8_t __ret;
__ret = (int8_t) __builtin_neon_vqabsb_s8(__p0);
return __ret;
}
-#else
-__ai int8_t vqabsb_s8(int8_t __p0) {
- int8_t __ret;
- __ret = (int8_t) __builtin_neon_vqabsb_s8(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai int32_t vqabss_s32(int32_t __p0) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vqabss_s32(__p0);
return __ret;
}
-#else
-__ai int32_t vqabss_s32(int32_t __p0) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqabss_s32(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vqabsd_s64(int64_t __p0) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vqabsd_s64(__p0);
- return __ret;
-}
-#else
__ai int64_t vqabsd_s64(int64_t __p0) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vqabsd_s64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai int16_t vqabsh_s16(int16_t __p0) {
int16_t __ret;
__ret = (int16_t) __builtin_neon_vqabsh_s16(__p0);
return __ret;
}
-#else
-__ai int16_t vqabsh_s16(int16_t __p0) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vqabsh_s16(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint8_t vqaddb_u8(uint8_t __p0, uint8_t __p1) {
uint8_t __ret;
__ret = (uint8_t) __builtin_neon_vqaddb_u8(__p0, __p1);
return __ret;
}
-#else
-__ai uint8_t vqaddb_u8(uint8_t __p0, uint8_t __p1) {
- uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vqaddb_u8(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vqadds_u32(uint32_t __p0, uint32_t __p1) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vqadds_u32(__p0, __p1);
- return __ret;
-}
-#else
__ai uint32_t vqadds_u32(uint32_t __p0, uint32_t __p1) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vqadds_u32(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vqaddd_u64(uint64_t __p0, uint64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vqaddd_u64(__p0, __p1);
- return __ret;
-}
-#else
__ai uint64_t vqaddd_u64(uint64_t __p0, uint64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vqaddd_u64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16_t vqaddh_u16(uint16_t __p0, uint16_t __p1) {
- uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vqaddh_u16(__p0, __p1);
- return __ret;
-}
-#else
__ai uint16_t vqaddh_u16(uint16_t __p0, uint16_t __p1) {
uint16_t __ret;
__ret = (uint16_t) __builtin_neon_vqaddh_u16(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai int8_t vqaddb_s8(int8_t __p0, int8_t __p1) {
int8_t __ret;
__ret = (int8_t) __builtin_neon_vqaddb_s8(__p0, __p1);
return __ret;
}
-#else
-__ai int8_t vqaddb_s8(int8_t __p0, int8_t __p1) {
- int8_t __ret;
- __ret = (int8_t) __builtin_neon_vqaddb_s8(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai int32_t vqadds_s32(int32_t __p0, int32_t __p1) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vqadds_s32(__p0, __p1);
return __ret;
}
-#else
-__ai int32_t vqadds_s32(int32_t __p0, int32_t __p1) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqadds_s32(__p0, __p1);
- return __ret;
-}
-__ai int32_t __noswap_vqadds_s32(int32_t __p0, int32_t __p1) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqadds_s32(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai int64_t vqaddd_s64(int64_t __p0, int64_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vqaddd_s64(__p0, __p1);
return __ret;
}
-#else
-__ai int64_t vqaddd_s64(int64_t __p0, int64_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vqaddd_s64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16_t vqaddh_s16(int16_t __p0, int16_t __p1) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vqaddh_s16(__p0, __p1);
- return __ret;
-}
-#else
__ai int16_t vqaddh_s16(int16_t __p0, int16_t __p1) {
int16_t __ret;
__ret = (int16_t) __builtin_neon_vqaddh_s16(__p0, __p1);
return __ret;
}
-__ai int16_t __noswap_vqaddh_s16(int16_t __p0, int16_t __p1) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vqaddh_s16(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai int64_t vqdmlals_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vqdmlals_s32(__p0, __p1, __p2);
return __ret;
}
-#else
-__ai int64_t vqdmlals_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vqdmlals_s32(__p0, __p1, __p2);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vqdmlalh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqdmlalh_s16(__p0, __p1, __p2);
- return __ret;
-}
-#else
__ai int32_t vqdmlalh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vqdmlalh_s16(__p0, __p1, __p2);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int64x2_t vqdmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
int64x2_t __ret;
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlal_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = vqdmlal_s32(__s0, vget_high_s32(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vqdmlal_high_lane_s32(__p0_603, __p1_603, __p2_603, __p3_603) __extension__ ({ \
+ int64x2_t __s0_603 = __p0_603; \
+ int32x4_t __s1_603 = __p1_603; \
+ int32x2_t __s2_603 = __p2_603; \
+ int64x2_t __ret_603; \
+ __ret_603 = vqdmlal_s32(__s0_603, vget_high_s32(__s1_603), splat_lane_s32(__s2_603, __p3_603)); \
+ __ret_603; \
})
#else
-#define vqdmlal_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vqdmlal_s32(__rev0, __noswap_vget_high_s32(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vqdmlal_high_lane_s32(__p0_604, __p1_604, __p2_604, __p3_604) __extension__ ({ \
+ int64x2_t __s0_604 = __p0_604; \
+ int32x4_t __s1_604 = __p1_604; \
+ int32x2_t __s2_604 = __p2_604; \
+ int64x2_t __rev0_604; __rev0_604 = __builtin_shufflevector(__s0_604, __s0_604, 1, 0); \
+ int32x4_t __rev1_604; __rev1_604 = __builtin_shufflevector(__s1_604, __s1_604, 3, 2, 1, 0); \
+ int32x2_t __rev2_604; __rev2_604 = __builtin_shufflevector(__s2_604, __s2_604, 1, 0); \
+ int64x2_t __ret_604; \
+ __ret_604 = __noswap_vqdmlal_s32(__rev0_604, __noswap_vget_high_s32(__rev1_604), __noswap_splat_lane_s32(__rev2_604, __p3_604)); \
+ __ret_604 = __builtin_shufflevector(__ret_604, __ret_604, 1, 0); \
+ __ret_604; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlal_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = vqdmlal_s16(__s0, vget_high_s16(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vqdmlal_high_lane_s16(__p0_605, __p1_605, __p2_605, __p3_605) __extension__ ({ \
+ int32x4_t __s0_605 = __p0_605; \
+ int16x8_t __s1_605 = __p1_605; \
+ int16x4_t __s2_605 = __p2_605; \
+ int32x4_t __ret_605; \
+ __ret_605 = vqdmlal_s16(__s0_605, vget_high_s16(__s1_605), splat_lane_s16(__s2_605, __p3_605)); \
+ __ret_605; \
})
#else
-#define vqdmlal_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqdmlal_s16(__rev0, __noswap_vget_high_s16(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqdmlal_high_lane_s16(__p0_606, __p1_606, __p2_606, __p3_606) __extension__ ({ \
+ int32x4_t __s0_606 = __p0_606; \
+ int16x8_t __s1_606 = __p1_606; \
+ int16x4_t __s2_606 = __p2_606; \
+ int32x4_t __rev0_606; __rev0_606 = __builtin_shufflevector(__s0_606, __s0_606, 3, 2, 1, 0); \
+ int16x8_t __rev1_606; __rev1_606 = __builtin_shufflevector(__s1_606, __s1_606, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev2_606; __rev2_606 = __builtin_shufflevector(__s2_606, __s2_606, 3, 2, 1, 0); \
+ int32x4_t __ret_606; \
+ __ret_606 = __noswap_vqdmlal_s16(__rev0_606, __noswap_vget_high_s16(__rev1_606), __noswap_splat_lane_s16(__rev2_606, __p3_606)); \
+ __ret_606 = __builtin_shufflevector(__ret_606, __ret_606, 3, 2, 1, 0); \
+ __ret_606; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlal_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = vqdmlal_s32(__s0, vget_high_s32(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vqdmlal_high_laneq_s32(__p0_607, __p1_607, __p2_607, __p3_607) __extension__ ({ \
+ int64x2_t __s0_607 = __p0_607; \
+ int32x4_t __s1_607 = __p1_607; \
+ int32x4_t __s2_607 = __p2_607; \
+ int64x2_t __ret_607; \
+ __ret_607 = vqdmlal_s32(__s0_607, vget_high_s32(__s1_607), splat_laneq_s32(__s2_607, __p3_607)); \
+ __ret_607; \
})
#else
-#define vqdmlal_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vqdmlal_s32(__rev0, __noswap_vget_high_s32(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vqdmlal_high_laneq_s32(__p0_608, __p1_608, __p2_608, __p3_608) __extension__ ({ \
+ int64x2_t __s0_608 = __p0_608; \
+ int32x4_t __s1_608 = __p1_608; \
+ int32x4_t __s2_608 = __p2_608; \
+ int64x2_t __rev0_608; __rev0_608 = __builtin_shufflevector(__s0_608, __s0_608, 1, 0); \
+ int32x4_t __rev1_608; __rev1_608 = __builtin_shufflevector(__s1_608, __s1_608, 3, 2, 1, 0); \
+ int32x4_t __rev2_608; __rev2_608 = __builtin_shufflevector(__s2_608, __s2_608, 3, 2, 1, 0); \
+ int64x2_t __ret_608; \
+ __ret_608 = __noswap_vqdmlal_s32(__rev0_608, __noswap_vget_high_s32(__rev1_608), __noswap_splat_laneq_s32(__rev2_608, __p3_608)); \
+ __ret_608 = __builtin_shufflevector(__ret_608, __ret_608, 1, 0); \
+ __ret_608; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlal_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = vqdmlal_s16(__s0, vget_high_s16(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vqdmlal_high_laneq_s16(__p0_609, __p1_609, __p2_609, __p3_609) __extension__ ({ \
+ int32x4_t __s0_609 = __p0_609; \
+ int16x8_t __s1_609 = __p1_609; \
+ int16x8_t __s2_609 = __p2_609; \
+ int32x4_t __ret_609; \
+ __ret_609 = vqdmlal_s16(__s0_609, vget_high_s16(__s1_609), splat_laneq_s16(__s2_609, __p3_609)); \
+ __ret_609; \
})
#else
-#define vqdmlal_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqdmlal_s16(__rev0, __noswap_vget_high_s16(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqdmlal_high_laneq_s16(__p0_610, __p1_610, __p2_610, __p3_610) __extension__ ({ \
+ int32x4_t __s0_610 = __p0_610; \
+ int16x8_t __s1_610 = __p1_610; \
+ int16x8_t __s2_610 = __p2_610; \
+ int32x4_t __rev0_610; __rev0_610 = __builtin_shufflevector(__s0_610, __s0_610, 3, 2, 1, 0); \
+ int16x8_t __rev1_610; __rev1_610 = __builtin_shufflevector(__s1_610, __s1_610, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev2_610; __rev2_610 = __builtin_shufflevector(__s2_610, __s2_610, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_610; \
+ __ret_610 = __noswap_vqdmlal_s16(__rev0_610, __noswap_vget_high_s16(__rev1_610), __noswap_splat_laneq_s16(__rev2_610, __p3_610)); \
+ __ret_610 = __builtin_shufflevector(__ret_610, __ret_610, 3, 2, 1, 0); \
+ __ret_610; \
})
#endif
int32_t __s1 = __p1; \
int32x2_t __s2 = __p2; \
int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vqdmlals_lane_s32(__s0, __s1, (int8x8_t)__s2, __p3); \
+ __ret = (int64_t) __builtin_neon_vqdmlals_lane_s32(__s0, __s1, __s2, __p3); \
__ret; \
})
#else
int32x2_t __s2 = __p2; \
int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vqdmlals_lane_s32(__s0, __s1, (int8x8_t)__rev2, __p3); \
+ __ret = (int64_t) __builtin_neon_vqdmlals_lane_s32(__s0, __s1, __rev2, __p3); \
__ret; \
})
#endif
int16_t __s1 = __p1; \
int16x4_t __s2 = __p2; \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vqdmlalh_lane_s16(__s0, __s1, (int8x8_t)__s2, __p3); \
+ __ret = (int32_t) __builtin_neon_vqdmlalh_lane_s16(__s0, __s1, __s2, __p3); \
__ret; \
})
#else
int16x4_t __s2 = __p2; \
int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vqdmlalh_lane_s16(__s0, __s1, (int8x8_t)__rev2, __p3); \
+ __ret = (int32_t) __builtin_neon_vqdmlalh_lane_s16(__s0, __s1, __rev2, __p3); \
__ret; \
})
#endif
int32_t __s1 = __p1; \
int32x4_t __s2 = __p2; \
int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vqdmlals_laneq_s32(__s0, __s1, (int8x16_t)__s2, __p3); \
+ __ret = (int64_t) __builtin_neon_vqdmlals_laneq_s32(__s0, __s1, __s2, __p3); \
__ret; \
})
#else
int32x4_t __s2 = __p2; \
int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vqdmlals_laneq_s32(__s0, __s1, (int8x16_t)__rev2, __p3); \
+ __ret = (int64_t) __builtin_neon_vqdmlals_laneq_s32(__s0, __s1, __rev2, __p3); \
__ret; \
})
#endif
int16_t __s1 = __p1; \
int16x8_t __s2 = __p2; \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vqdmlalh_laneq_s16(__s0, __s1, (int8x16_t)__s2, __p3); \
+ __ret = (int32_t) __builtin_neon_vqdmlalh_laneq_s16(__s0, __s1, __s2, __p3); \
__ret; \
})
#else
int16x8_t __s2 = __p2; \
int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vqdmlalh_laneq_s16(__s0, __s1, (int8x16_t)__rev2, __p3); \
+ __ret = (int32_t) __builtin_neon_vqdmlalh_laneq_s16(__s0, __s1, __rev2, __p3); \
__ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlal_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = vqdmlal_s32(__s0, __s1, __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vqdmlal_laneq_s32(__p0_611, __p1_611, __p2_611, __p3_611) __extension__ ({ \
+ int64x2_t __s0_611 = __p0_611; \
+ int32x2_t __s1_611 = __p1_611; \
+ int32x4_t __s2_611 = __p2_611; \
+ int64x2_t __ret_611; \
+ __ret_611 = vqdmlal_s32(__s0_611, __s1_611, splat_laneq_s32(__s2_611, __p3_611)); \
+ __ret_611; \
})
#else
-#define vqdmlal_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vqdmlal_s32(__rev0, __rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vqdmlal_laneq_s32(__p0_612, __p1_612, __p2_612, __p3_612) __extension__ ({ \
+ int64x2_t __s0_612 = __p0_612; \
+ int32x2_t __s1_612 = __p1_612; \
+ int32x4_t __s2_612 = __p2_612; \
+ int64x2_t __rev0_612; __rev0_612 = __builtin_shufflevector(__s0_612, __s0_612, 1, 0); \
+ int32x2_t __rev1_612; __rev1_612 = __builtin_shufflevector(__s1_612, __s1_612, 1, 0); \
+ int32x4_t __rev2_612; __rev2_612 = __builtin_shufflevector(__s2_612, __s2_612, 3, 2, 1, 0); \
+ int64x2_t __ret_612; \
+ __ret_612 = __noswap_vqdmlal_s32(__rev0_612, __rev1_612, __noswap_splat_laneq_s32(__rev2_612, __p3_612)); \
+ __ret_612 = __builtin_shufflevector(__ret_612, __ret_612, 1, 0); \
+ __ret_612; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlal_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = vqdmlal_s16(__s0, __s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vqdmlal_laneq_s16(__p0_613, __p1_613, __p2_613, __p3_613) __extension__ ({ \
+ int32x4_t __s0_613 = __p0_613; \
+ int16x4_t __s1_613 = __p1_613; \
+ int16x8_t __s2_613 = __p2_613; \
+ int32x4_t __ret_613; \
+ __ret_613 = vqdmlal_s16(__s0_613, __s1_613, splat_laneq_s16(__s2_613, __p3_613)); \
+ __ret_613; \
})
#else
-#define vqdmlal_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqdmlal_s16(__rev0, __rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqdmlal_laneq_s16(__p0_614, __p1_614, __p2_614, __p3_614) __extension__ ({ \
+ int32x4_t __s0_614 = __p0_614; \
+ int16x4_t __s1_614 = __p1_614; \
+ int16x8_t __s2_614 = __p2_614; \
+ int32x4_t __rev0_614; __rev0_614 = __builtin_shufflevector(__s0_614, __s0_614, 3, 2, 1, 0); \
+ int16x4_t __rev1_614; __rev1_614 = __builtin_shufflevector(__s1_614, __s1_614, 3, 2, 1, 0); \
+ int16x8_t __rev2_614; __rev2_614 = __builtin_shufflevector(__s2_614, __s2_614, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_614; \
+ __ret_614 = __noswap_vqdmlal_s16(__rev0_614, __rev1_614, __noswap_splat_laneq_s16(__rev2_614, __p3_614)); \
+ __ret_614 = __builtin_shufflevector(__ret_614, __ret_614, 3, 2, 1, 0); \
+ __ret_614; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vqdmlsls_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vqdmlsls_s32(__p0, __p1, __p2);
- return __ret;
-}
-#else
__ai int64_t vqdmlsls_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vqdmlsls_s32(__p0, __p1, __p2);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vqdmlslh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqdmlslh_s16(__p0, __p1, __p2);
- return __ret;
-}
-#else
__ai int32_t vqdmlslh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vqdmlslh_s16(__p0, __p1, __p2);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int64x2_t vqdmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
int64x2_t __ret;
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlsl_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = vqdmlsl_s32(__s0, vget_high_s32(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vqdmlsl_high_lane_s32(__p0_615, __p1_615, __p2_615, __p3_615) __extension__ ({ \
+ int64x2_t __s0_615 = __p0_615; \
+ int32x4_t __s1_615 = __p1_615; \
+ int32x2_t __s2_615 = __p2_615; \
+ int64x2_t __ret_615; \
+ __ret_615 = vqdmlsl_s32(__s0_615, vget_high_s32(__s1_615), splat_lane_s32(__s2_615, __p3_615)); \
+ __ret_615; \
})
#else
-#define vqdmlsl_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vqdmlsl_s32(__rev0, __noswap_vget_high_s32(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vqdmlsl_high_lane_s32(__p0_616, __p1_616, __p2_616, __p3_616) __extension__ ({ \
+ int64x2_t __s0_616 = __p0_616; \
+ int32x4_t __s1_616 = __p1_616; \
+ int32x2_t __s2_616 = __p2_616; \
+ int64x2_t __rev0_616; __rev0_616 = __builtin_shufflevector(__s0_616, __s0_616, 1, 0); \
+ int32x4_t __rev1_616; __rev1_616 = __builtin_shufflevector(__s1_616, __s1_616, 3, 2, 1, 0); \
+ int32x2_t __rev2_616; __rev2_616 = __builtin_shufflevector(__s2_616, __s2_616, 1, 0); \
+ int64x2_t __ret_616; \
+ __ret_616 = __noswap_vqdmlsl_s32(__rev0_616, __noswap_vget_high_s32(__rev1_616), __noswap_splat_lane_s32(__rev2_616, __p3_616)); \
+ __ret_616 = __builtin_shufflevector(__ret_616, __ret_616, 1, 0); \
+ __ret_616; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlsl_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = vqdmlsl_s16(__s0, vget_high_s16(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vqdmlsl_high_lane_s16(__p0_617, __p1_617, __p2_617, __p3_617) __extension__ ({ \
+ int32x4_t __s0_617 = __p0_617; \
+ int16x8_t __s1_617 = __p1_617; \
+ int16x4_t __s2_617 = __p2_617; \
+ int32x4_t __ret_617; \
+ __ret_617 = vqdmlsl_s16(__s0_617, vget_high_s16(__s1_617), splat_lane_s16(__s2_617, __p3_617)); \
+ __ret_617; \
})
#else
-#define vqdmlsl_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqdmlsl_s16(__rev0, __noswap_vget_high_s16(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqdmlsl_high_lane_s16(__p0_618, __p1_618, __p2_618, __p3_618) __extension__ ({ \
+ int32x4_t __s0_618 = __p0_618; \
+ int16x8_t __s1_618 = __p1_618; \
+ int16x4_t __s2_618 = __p2_618; \
+ int32x4_t __rev0_618; __rev0_618 = __builtin_shufflevector(__s0_618, __s0_618, 3, 2, 1, 0); \
+ int16x8_t __rev1_618; __rev1_618 = __builtin_shufflevector(__s1_618, __s1_618, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev2_618; __rev2_618 = __builtin_shufflevector(__s2_618, __s2_618, 3, 2, 1, 0); \
+ int32x4_t __ret_618; \
+ __ret_618 = __noswap_vqdmlsl_s16(__rev0_618, __noswap_vget_high_s16(__rev1_618), __noswap_splat_lane_s16(__rev2_618, __p3_618)); \
+ __ret_618 = __builtin_shufflevector(__ret_618, __ret_618, 3, 2, 1, 0); \
+ __ret_618; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlsl_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = vqdmlsl_s32(__s0, vget_high_s32(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vqdmlsl_high_laneq_s32(__p0_619, __p1_619, __p2_619, __p3_619) __extension__ ({ \
+ int64x2_t __s0_619 = __p0_619; \
+ int32x4_t __s1_619 = __p1_619; \
+ int32x4_t __s2_619 = __p2_619; \
+ int64x2_t __ret_619; \
+ __ret_619 = vqdmlsl_s32(__s0_619, vget_high_s32(__s1_619), splat_laneq_s32(__s2_619, __p3_619)); \
+ __ret_619; \
})
#else
-#define vqdmlsl_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vqdmlsl_s32(__rev0, __noswap_vget_high_s32(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vqdmlsl_high_laneq_s32(__p0_620, __p1_620, __p2_620, __p3_620) __extension__ ({ \
+ int64x2_t __s0_620 = __p0_620; \
+ int32x4_t __s1_620 = __p1_620; \
+ int32x4_t __s2_620 = __p2_620; \
+ int64x2_t __rev0_620; __rev0_620 = __builtin_shufflevector(__s0_620, __s0_620, 1, 0); \
+ int32x4_t __rev1_620; __rev1_620 = __builtin_shufflevector(__s1_620, __s1_620, 3, 2, 1, 0); \
+ int32x4_t __rev2_620; __rev2_620 = __builtin_shufflevector(__s2_620, __s2_620, 3, 2, 1, 0); \
+ int64x2_t __ret_620; \
+ __ret_620 = __noswap_vqdmlsl_s32(__rev0_620, __noswap_vget_high_s32(__rev1_620), __noswap_splat_laneq_s32(__rev2_620, __p3_620)); \
+ __ret_620 = __builtin_shufflevector(__ret_620, __ret_620, 1, 0); \
+ __ret_620; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlsl_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = vqdmlsl_s16(__s0, vget_high_s16(__s1), __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vqdmlsl_high_laneq_s16(__p0_621, __p1_621, __p2_621, __p3_621) __extension__ ({ \
+ int32x4_t __s0_621 = __p0_621; \
+ int16x8_t __s1_621 = __p1_621; \
+ int16x8_t __s2_621 = __p2_621; \
+ int32x4_t __ret_621; \
+ __ret_621 = vqdmlsl_s16(__s0_621, vget_high_s16(__s1_621), splat_laneq_s16(__s2_621, __p3_621)); \
+ __ret_621; \
})
#else
-#define vqdmlsl_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqdmlsl_s16(__rev0, __noswap_vget_high_s16(__rev1), __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqdmlsl_high_laneq_s16(__p0_622, __p1_622, __p2_622, __p3_622) __extension__ ({ \
+ int32x4_t __s0_622 = __p0_622; \
+ int16x8_t __s1_622 = __p1_622; \
+ int16x8_t __s2_622 = __p2_622; \
+ int32x4_t __rev0_622; __rev0_622 = __builtin_shufflevector(__s0_622, __s0_622, 3, 2, 1, 0); \
+ int16x8_t __rev1_622; __rev1_622 = __builtin_shufflevector(__s1_622, __s1_622, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev2_622; __rev2_622 = __builtin_shufflevector(__s2_622, __s2_622, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_622; \
+ __ret_622 = __noswap_vqdmlsl_s16(__rev0_622, __noswap_vget_high_s16(__rev1_622), __noswap_splat_laneq_s16(__rev2_622, __p3_622)); \
+ __ret_622 = __builtin_shufflevector(__ret_622, __ret_622, 3, 2, 1, 0); \
+ __ret_622; \
})
#endif
int32_t __s1 = __p1; \
int32x2_t __s2 = __p2; \
int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vqdmlsls_lane_s32(__s0, __s1, (int8x8_t)__s2, __p3); \
+ __ret = (int64_t) __builtin_neon_vqdmlsls_lane_s32(__s0, __s1, __s2, __p3); \
__ret; \
})
#else
int32x2_t __s2 = __p2; \
int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vqdmlsls_lane_s32(__s0, __s1, (int8x8_t)__rev2, __p3); \
+ __ret = (int64_t) __builtin_neon_vqdmlsls_lane_s32(__s0, __s1, __rev2, __p3); \
__ret; \
})
#endif
int16_t __s1 = __p1; \
int16x4_t __s2 = __p2; \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vqdmlslh_lane_s16(__s0, __s1, (int8x8_t)__s2, __p3); \
+ __ret = (int32_t) __builtin_neon_vqdmlslh_lane_s16(__s0, __s1, __s2, __p3); \
__ret; \
})
#else
int16x4_t __s2 = __p2; \
int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vqdmlslh_lane_s16(__s0, __s1, (int8x8_t)__rev2, __p3); \
+ __ret = (int32_t) __builtin_neon_vqdmlslh_lane_s16(__s0, __s1, __rev2, __p3); \
__ret; \
})
#endif
int32_t __s1 = __p1; \
int32x4_t __s2 = __p2; \
int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vqdmlsls_laneq_s32(__s0, __s1, (int8x16_t)__s2, __p3); \
+ __ret = (int64_t) __builtin_neon_vqdmlsls_laneq_s32(__s0, __s1, __s2, __p3); \
__ret; \
})
#else
int32x4_t __s2 = __p2; \
int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vqdmlsls_laneq_s32(__s0, __s1, (int8x16_t)__rev2, __p3); \
+ __ret = (int64_t) __builtin_neon_vqdmlsls_laneq_s32(__s0, __s1, __rev2, __p3); \
__ret; \
})
#endif
int16_t __s1 = __p1; \
int16x8_t __s2 = __p2; \
int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vqdmlslh_laneq_s16(__s0, __s1, (int8x16_t)__s2, __p3); \
+ __ret = (int32_t) __builtin_neon_vqdmlslh_laneq_s16(__s0, __s1, __s2, __p3); \
+ __ret; \
+})
+#else
+#define vqdmlslh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
+ int32_t __s0 = __p0; \
+ int16_t __s1 = __p1; \
+ int16x8_t __s2 = __p2; \
+ int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32_t __ret; \
+ __ret = (int32_t) __builtin_neon_vqdmlslh_laneq_s16(__s0, __s1, __rev2, __p3); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vqdmlsl_laneq_s32(__p0_623, __p1_623, __p2_623, __p3_623) __extension__ ({ \
+ int64x2_t __s0_623 = __p0_623; \
+ int32x2_t __s1_623 = __p1_623; \
+ int32x4_t __s2_623 = __p2_623; \
+ int64x2_t __ret_623; \
+ __ret_623 = vqdmlsl_s32(__s0_623, __s1_623, splat_laneq_s32(__s2_623, __p3_623)); \
+ __ret_623; \
+})
+#else
+#define vqdmlsl_laneq_s32(__p0_624, __p1_624, __p2_624, __p3_624) __extension__ ({ \
+ int64x2_t __s0_624 = __p0_624; \
+ int32x2_t __s1_624 = __p1_624; \
+ int32x4_t __s2_624 = __p2_624; \
+ int64x2_t __rev0_624; __rev0_624 = __builtin_shufflevector(__s0_624, __s0_624, 1, 0); \
+ int32x2_t __rev1_624; __rev1_624 = __builtin_shufflevector(__s1_624, __s1_624, 1, 0); \
+ int32x4_t __rev2_624; __rev2_624 = __builtin_shufflevector(__s2_624, __s2_624, 3, 2, 1, 0); \
+ int64x2_t __ret_624; \
+ __ret_624 = __noswap_vqdmlsl_s32(__rev0_624, __rev1_624, __noswap_splat_laneq_s32(__rev2_624, __p3_624)); \
+ __ret_624 = __builtin_shufflevector(__ret_624, __ret_624, 1, 0); \
+ __ret_624; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vqdmlsl_laneq_s16(__p0_625, __p1_625, __p2_625, __p3_625) __extension__ ({ \
+ int32x4_t __s0_625 = __p0_625; \
+ int16x4_t __s1_625 = __p1_625; \
+ int16x8_t __s2_625 = __p2_625; \
+ int32x4_t __ret_625; \
+ __ret_625 = vqdmlsl_s16(__s0_625, __s1_625, splat_laneq_s16(__s2_625, __p3_625)); \
+ __ret_625; \
+})
+#else
+#define vqdmlsl_laneq_s16(__p0_626, __p1_626, __p2_626, __p3_626) __extension__ ({ \
+ int32x4_t __s0_626 = __p0_626; \
+ int16x4_t __s1_626 = __p1_626; \
+ int16x8_t __s2_626 = __p2_626; \
+ int32x4_t __rev0_626; __rev0_626 = __builtin_shufflevector(__s0_626, __s0_626, 3, 2, 1, 0); \
+ int16x4_t __rev1_626; __rev1_626 = __builtin_shufflevector(__s1_626, __s1_626, 3, 2, 1, 0); \
+ int16x8_t __rev2_626; __rev2_626 = __builtin_shufflevector(__s2_626, __s2_626, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_626; \
+ __ret_626 = __noswap_vqdmlsl_s16(__rev0_626, __rev1_626, __noswap_splat_laneq_s16(__rev2_626, __p3_626)); \
+ __ret_626 = __builtin_shufflevector(__ret_626, __ret_626, 3, 2, 1, 0); \
+ __ret_626; \
+})
+#endif
+
+__ai int32_t vqdmulhs_s32(int32_t __p0, int32_t __p1) {
+ int32_t __ret;
+ __ret = (int32_t) __builtin_neon_vqdmulhs_s32(__p0, __p1);
+ return __ret;
+}
+__ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
+ int16_t __ret;
+ __ret = (int16_t) __builtin_neon_vqdmulhh_s16(__p0, __p1);
+ return __ret;
+}
+#ifdef __LITTLE_ENDIAN__
+#define vqdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
+ int32x4_t __s0 = __p0; \
+ int32x2_t __s1 = __p1; \
+ int32x4_t __ret; \
+ __ret = (int32x4_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 2); \
+ __ret; \
+})
+#else
+#define vqdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
+ int32x4_t __s0 = __p0; \
+ int32x2_t __s1 = __p1; \
+ int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
+ int32x4_t __ret; \
+ __ret = (int32x4_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vqdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
+ int16x8_t __s0 = __p0; \
+ int16x4_t __s1 = __p1; \
+ int16x8_t __ret; \
+ __ret = (int16x8_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 1); \
__ret; \
})
#else
-#define vqdmlslh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32_t __s0 = __p0; \
- int16_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vqdmlslh_laneq_s16(__s0, __s1, (int8x16_t)__rev2, __p3); \
+#define vqdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
+ int16x8_t __s0 = __p0; \
+ int16x4_t __s1 = __p1; \
+ int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
+ int16x8_t __ret; \
+ __ret = (int16x8_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
__ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlsl_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
+#define vqdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
+ int32x2_t __s0 = __p0; \
int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = vqdmlsl_s32(__s0, __s1, __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
+ int32x2_t __ret; \
+ __ret = (int32x2_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 2); \
__ret; \
})
#else
-#define vqdmlsl_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
+#define vqdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
+ int32x2_t __s0 = __p0; \
int32x2_t __s1 = __p1; \
- int32x4_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vqdmlsl_s32(__rev0, __rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
+ int32x2_t __ret; \
+ __ret = (int32x2_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
__ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmlsl_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
+#define vqdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
+ int16x4_t __s0 = __p0; \
int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = vqdmlsl_s16(__s0, __s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
+ int16x4_t __ret; \
+ __ret = (int16x4_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 1); \
__ret; \
})
#else
-#define vqdmlsl_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
+#define vqdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
+ int16x4_t __s0 = __p0; \
int16x4_t __s1 = __p1; \
- int16x8_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqdmlsl_s16(__rev0, __rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
+ int16x4_t __ret; \
+ __ret = (int16x4_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
__ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32_t vqdmulhs_s32(int32_t __p0, int32_t __p1) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqdmulhs_s32(__p0, __p1);
- return __ret;
-}
-#else
-__ai int32_t vqdmulhs_s32(int32_t __p0, int32_t __p1) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqdmulhs_s32(__p0, __p1);
- return __ret;
-}
-__ai int32_t __noswap_vqdmulhs_s32(int32_t __p0, int32_t __p1) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqdmulhs_s32(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vqdmulhh_s16(__p0, __p1);
- return __ret;
-}
-#else
-__ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vqdmulhh_s16(__p0, __p1);
- return __ret;
-}
-__ai int16_t __noswap_vqdmulhh_s16(int16_t __p0, int16_t __p1) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vqdmulhh_s16(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqdmulhs_lane_s32(__p0_158, __p1_158, __p2_158) __extension__ ({ \
- int32_t __s0_158 = __p0_158; \
- int32x2_t __s1_158 = __p1_158; \
- int32_t __ret_158; \
- __ret_158 = vqdmulhs_s32(__s0_158, vget_lane_s32(__s1_158, __p2_158)); \
- __ret_158; \
+#define vqdmulhs_lane_s32(__p0_627, __p1_627, __p2_627) __extension__ ({ \
+ int32_t __s0_627 = __p0_627; \
+ int32x2_t __s1_627 = __p1_627; \
+ int32_t __ret_627; \
+ __ret_627 = vqdmulhs_s32(__s0_627, vget_lane_s32(__s1_627, __p2_627)); \
+ __ret_627; \
})
#else
-#define vqdmulhs_lane_s32(__p0_159, __p1_159, __p2_159) __extension__ ({ \
- int32_t __s0_159 = __p0_159; \
- int32x2_t __s1_159 = __p1_159; \
- int32x2_t __rev1_159; __rev1_159 = __builtin_shufflevector(__s1_159, __s1_159, 1, 0); \
- int32_t __ret_159; \
- __ret_159 = __noswap_vqdmulhs_s32(__s0_159, __noswap_vget_lane_s32(__rev1_159, __p2_159)); \
- __ret_159; \
+#define vqdmulhs_lane_s32(__p0_628, __p1_628, __p2_628) __extension__ ({ \
+ int32_t __s0_628 = __p0_628; \
+ int32x2_t __s1_628 = __p1_628; \
+ int32x2_t __rev1_628; __rev1_628 = __builtin_shufflevector(__s1_628, __s1_628, 1, 0); \
+ int32_t __ret_628; \
+ __ret_628 = vqdmulhs_s32(__s0_628, __noswap_vget_lane_s32(__rev1_628, __p2_628)); \
+ __ret_628; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmulhh_lane_s16(__p0_160, __p1_160, __p2_160) __extension__ ({ \
- int16_t __s0_160 = __p0_160; \
- int16x4_t __s1_160 = __p1_160; \
- int16_t __ret_160; \
- __ret_160 = vqdmulhh_s16(__s0_160, vget_lane_s16(__s1_160, __p2_160)); \
- __ret_160; \
+#define vqdmulhh_lane_s16(__p0_629, __p1_629, __p2_629) __extension__ ({ \
+ int16_t __s0_629 = __p0_629; \
+ int16x4_t __s1_629 = __p1_629; \
+ int16_t __ret_629; \
+ __ret_629 = vqdmulhh_s16(__s0_629, vget_lane_s16(__s1_629, __p2_629)); \
+ __ret_629; \
})
#else
-#define vqdmulhh_lane_s16(__p0_161, __p1_161, __p2_161) __extension__ ({ \
- int16_t __s0_161 = __p0_161; \
- int16x4_t __s1_161 = __p1_161; \
- int16x4_t __rev1_161; __rev1_161 = __builtin_shufflevector(__s1_161, __s1_161, 3, 2, 1, 0); \
- int16_t __ret_161; \
- __ret_161 = __noswap_vqdmulhh_s16(__s0_161, __noswap_vget_lane_s16(__rev1_161, __p2_161)); \
- __ret_161; \
+#define vqdmulhh_lane_s16(__p0_630, __p1_630, __p2_630) __extension__ ({ \
+ int16_t __s0_630 = __p0_630; \
+ int16x4_t __s1_630 = __p1_630; \
+ int16x4_t __rev1_630; __rev1_630 = __builtin_shufflevector(__s1_630, __s1_630, 3, 2, 1, 0); \
+ int16_t __ret_630; \
+ __ret_630 = vqdmulhh_s16(__s0_630, __noswap_vget_lane_s16(__rev1_630, __p2_630)); \
+ __ret_630; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmulhs_laneq_s32(__p0_162, __p1_162, __p2_162) __extension__ ({ \
- int32_t __s0_162 = __p0_162; \
- int32x4_t __s1_162 = __p1_162; \
- int32_t __ret_162; \
- __ret_162 = vqdmulhs_s32(__s0_162, vgetq_lane_s32(__s1_162, __p2_162)); \
- __ret_162; \
+#define vqdmulhs_laneq_s32(__p0_631, __p1_631, __p2_631) __extension__ ({ \
+ int32_t __s0_631 = __p0_631; \
+ int32x4_t __s1_631 = __p1_631; \
+ int32_t __ret_631; \
+ __ret_631 = vqdmulhs_s32(__s0_631, vgetq_lane_s32(__s1_631, __p2_631)); \
+ __ret_631; \
})
#else
-#define vqdmulhs_laneq_s32(__p0_163, __p1_163, __p2_163) __extension__ ({ \
- int32_t __s0_163 = __p0_163; \
- int32x4_t __s1_163 = __p1_163; \
- int32x4_t __rev1_163; __rev1_163 = __builtin_shufflevector(__s1_163, __s1_163, 3, 2, 1, 0); \
- int32_t __ret_163; \
- __ret_163 = __noswap_vqdmulhs_s32(__s0_163, __noswap_vgetq_lane_s32(__rev1_163, __p2_163)); \
- __ret_163; \
+#define vqdmulhs_laneq_s32(__p0_632, __p1_632, __p2_632) __extension__ ({ \
+ int32_t __s0_632 = __p0_632; \
+ int32x4_t __s1_632 = __p1_632; \
+ int32x4_t __rev1_632; __rev1_632 = __builtin_shufflevector(__s1_632, __s1_632, 3, 2, 1, 0); \
+ int32_t __ret_632; \
+ __ret_632 = vqdmulhs_s32(__s0_632, __noswap_vgetq_lane_s32(__rev1_632, __p2_632)); \
+ __ret_632; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmulhh_laneq_s16(__p0_164, __p1_164, __p2_164) __extension__ ({ \
- int16_t __s0_164 = __p0_164; \
- int16x8_t __s1_164 = __p1_164; \
- int16_t __ret_164; \
- __ret_164 = vqdmulhh_s16(__s0_164, vgetq_lane_s16(__s1_164, __p2_164)); \
- __ret_164; \
+#define vqdmulhh_laneq_s16(__p0_633, __p1_633, __p2_633) __extension__ ({ \
+ int16_t __s0_633 = __p0_633; \
+ int16x8_t __s1_633 = __p1_633; \
+ int16_t __ret_633; \
+ __ret_633 = vqdmulhh_s16(__s0_633, vgetq_lane_s16(__s1_633, __p2_633)); \
+ __ret_633; \
})
#else
-#define vqdmulhh_laneq_s16(__p0_165, __p1_165, __p2_165) __extension__ ({ \
- int16_t __s0_165 = __p0_165; \
- int16x8_t __s1_165 = __p1_165; \
- int16x8_t __rev1_165; __rev1_165 = __builtin_shufflevector(__s1_165, __s1_165, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16_t __ret_165; \
- __ret_165 = __noswap_vqdmulhh_s16(__s0_165, __noswap_vgetq_lane_s16(__rev1_165, __p2_165)); \
- __ret_165; \
+#define vqdmulhh_laneq_s16(__p0_634, __p1_634, __p2_634) __extension__ ({ \
+ int16_t __s0_634 = __p0_634; \
+ int16x8_t __s1_634 = __p1_634; \
+ int16x8_t __rev1_634; __rev1_634 = __builtin_shufflevector(__s1_634, __s1_634, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16_t __ret_634; \
+ __ret_634 = vqdmulhh_s16(__s0_634, __noswap_vgetq_lane_s16(__rev1_634, __p2_634)); \
+ __ret_634; \
})
#endif
int32x4_t __s0 = __p0; \
int32x4_t __s1 = __p1; \
int32x4_t __ret; \
- __ret = vqdmulhq_s32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
+ __ret = (int32x4_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 34); \
__ret; \
})
#else
int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
int32x4_t __ret; \
- __ret = __noswap_vqdmulhq_s32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
+ __ret = (int32x4_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 34); \
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
__ret; \
})
int16x8_t __s0 = __p0; \
int16x8_t __s1 = __p1; \
int16x8_t __ret; \
- __ret = vqdmulhq_s16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2)); \
+ __ret = (int16x8_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 33); \
__ret; \
})
#else
int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
int16x8_t __ret; \
- __ret = __noswap_vqdmulhq_s16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2)); \
+ __ret = (int16x8_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 33); \
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
__ret; \
})
int32x2_t __s0 = __p0; \
int32x4_t __s1 = __p1; \
int32x2_t __ret; \
- __ret = vqdmulh_s32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
+ __ret = (int32x2_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 2); \
__ret; \
})
#else
int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
int32x2_t __ret; \
- __ret = __noswap_vqdmulh_s32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
+ __ret = (int32x2_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 2); \
__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
__ret; \
})
int16x4_t __s0 = __p0; \
int16x8_t __s1 = __p1; \
int16x4_t __ret; \
- __ret = vqdmulh_s16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
+ __ret = (int16x4_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 1); \
__ret; \
})
#else
int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
int16x4_t __ret; \
- __ret = __noswap_vqdmulh_s16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
+ __ret = (int16x4_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 1); \
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
__ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vqdmulls_s32(int32_t __p0, int32_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vqdmulls_s32(__p0, __p1);
- return __ret;
-}
-#else
__ai int64_t vqdmulls_s32(int32_t __p0, int32_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vqdmulls_s32(__p0, __p1);
return __ret;
}
-__ai int64_t __noswap_vqdmulls_s32(int32_t __p0, int32_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vqdmulls_s32(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vqdmullh_s16(int16_t __p0, int16_t __p1) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqdmullh_s16(__p0, __p1);
- return __ret;
-}
-#else
__ai int32_t vqdmullh_s16(int16_t __p0, int16_t __p1) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vqdmullh_s16(__p0, __p1);
return __ret;
}
-__ai int32_t __noswap_vqdmullh_s16(int16_t __p0, int16_t __p1) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqdmullh_s16(__p0, __p1);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai int64x2_t vqdmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
int64x2_t __ret;
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmull_high_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int64x2_t __ret; \
- __ret = vqdmull_s32(vget_high_s32(__s0), __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vqdmull_high_lane_s32(__p0_635, __p1_635, __p2_635) __extension__ ({ \
+ int32x4_t __s0_635 = __p0_635; \
+ int32x2_t __s1_635 = __p1_635; \
+ int64x2_t __ret_635; \
+ __ret_635 = vqdmull_s32(vget_high_s32(__s0_635), splat_lane_s32(__s1_635, __p2_635)); \
+ __ret_635; \
})
#else
-#define vqdmull_high_lane_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vqdmull_s32(__noswap_vget_high_s32(__rev0), __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vqdmull_high_lane_s32(__p0_636, __p1_636, __p2_636) __extension__ ({ \
+ int32x4_t __s0_636 = __p0_636; \
+ int32x2_t __s1_636 = __p1_636; \
+ int32x4_t __rev0_636; __rev0_636 = __builtin_shufflevector(__s0_636, __s0_636, 3, 2, 1, 0); \
+ int32x2_t __rev1_636; __rev1_636 = __builtin_shufflevector(__s1_636, __s1_636, 1, 0); \
+ int64x2_t __ret_636; \
+ __ret_636 = __noswap_vqdmull_s32(__noswap_vget_high_s32(__rev0_636), __noswap_splat_lane_s32(__rev1_636, __p2_636)); \
+ __ret_636 = __builtin_shufflevector(__ret_636, __ret_636, 1, 0); \
+ __ret_636; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmull_high_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int32x4_t __ret; \
- __ret = vqdmull_s16(vget_high_s16(__s0), __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vqdmull_high_lane_s16(__p0_637, __p1_637, __p2_637) __extension__ ({ \
+ int16x8_t __s0_637 = __p0_637; \
+ int16x4_t __s1_637 = __p1_637; \
+ int32x4_t __ret_637; \
+ __ret_637 = vqdmull_s16(vget_high_s16(__s0_637), splat_lane_s16(__s1_637, __p2_637)); \
+ __ret_637; \
})
#else
-#define vqdmull_high_lane_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqdmull_s16(__noswap_vget_high_s16(__rev0), __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqdmull_high_lane_s16(__p0_638, __p1_638, __p2_638) __extension__ ({ \
+ int16x8_t __s0_638 = __p0_638; \
+ int16x4_t __s1_638 = __p1_638; \
+ int16x8_t __rev0_638; __rev0_638 = __builtin_shufflevector(__s0_638, __s0_638, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev1_638; __rev1_638 = __builtin_shufflevector(__s1_638, __s1_638, 3, 2, 1, 0); \
+ int32x4_t __ret_638; \
+ __ret_638 = __noswap_vqdmull_s16(__noswap_vget_high_s16(__rev0_638), __noswap_splat_lane_s16(__rev1_638, __p2_638)); \
+ __ret_638 = __builtin_shufflevector(__ret_638, __ret_638, 3, 2, 1, 0); \
+ __ret_638; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmull_high_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int64x2_t __ret; \
- __ret = vqdmull_s32(vget_high_s32(__s0), __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vqdmull_high_laneq_s32(__p0_639, __p1_639, __p2_639) __extension__ ({ \
+ int32x4_t __s0_639 = __p0_639; \
+ int32x4_t __s1_639 = __p1_639; \
+ int64x2_t __ret_639; \
+ __ret_639 = vqdmull_s32(vget_high_s32(__s0_639), splat_laneq_s32(__s1_639, __p2_639)); \
+ __ret_639; \
})
#else
-#define vqdmull_high_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vqdmull_s32(__noswap_vget_high_s32(__rev0), __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vqdmull_high_laneq_s32(__p0_640, __p1_640, __p2_640) __extension__ ({ \
+ int32x4_t __s0_640 = __p0_640; \
+ int32x4_t __s1_640 = __p1_640; \
+ int32x4_t __rev0_640; __rev0_640 = __builtin_shufflevector(__s0_640, __s0_640, 3, 2, 1, 0); \
+ int32x4_t __rev1_640; __rev1_640 = __builtin_shufflevector(__s1_640, __s1_640, 3, 2, 1, 0); \
+ int64x2_t __ret_640; \
+ __ret_640 = __noswap_vqdmull_s32(__noswap_vget_high_s32(__rev0_640), __noswap_splat_laneq_s32(__rev1_640, __p2_640)); \
+ __ret_640 = __builtin_shufflevector(__ret_640, __ret_640, 1, 0); \
+ __ret_640; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmull_high_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int32x4_t __ret; \
- __ret = vqdmull_s16(vget_high_s16(__s0), __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vqdmull_high_laneq_s16(__p0_641, __p1_641, __p2_641) __extension__ ({ \
+ int16x8_t __s0_641 = __p0_641; \
+ int16x8_t __s1_641 = __p1_641; \
+ int32x4_t __ret_641; \
+ __ret_641 = vqdmull_s16(vget_high_s16(__s0_641), splat_laneq_s16(__s1_641, __p2_641)); \
+ __ret_641; \
})
#else
-#define vqdmull_high_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x8_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqdmull_s16(__noswap_vget_high_s16(__rev0), __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqdmull_high_laneq_s16(__p0_642, __p1_642, __p2_642) __extension__ ({ \
+ int16x8_t __s0_642 = __p0_642; \
+ int16x8_t __s1_642 = __p1_642; \
+ int16x8_t __rev0_642; __rev0_642 = __builtin_shufflevector(__s0_642, __s0_642, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_642; __rev1_642 = __builtin_shufflevector(__s1_642, __s1_642, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_642; \
+ __ret_642 = __noswap_vqdmull_s16(__noswap_vget_high_s16(__rev0_642), __noswap_splat_laneq_s16(__rev1_642, __p2_642)); \
+ __ret_642 = __builtin_shufflevector(__ret_642, __ret_642, 3, 2, 1, 0); \
+ __ret_642; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmulls_lane_s32(__p0_166, __p1_166, __p2_166) __extension__ ({ \
- int32_t __s0_166 = __p0_166; \
- int32x2_t __s1_166 = __p1_166; \
- int64_t __ret_166; \
- __ret_166 = vqdmulls_s32(__s0_166, vget_lane_s32(__s1_166, __p2_166)); \
- __ret_166; \
+#define vqdmulls_lane_s32(__p0_643, __p1_643, __p2_643) __extension__ ({ \
+ int32_t __s0_643 = __p0_643; \
+ int32x2_t __s1_643 = __p1_643; \
+ int64_t __ret_643; \
+ __ret_643 = vqdmulls_s32(__s0_643, vget_lane_s32(__s1_643, __p2_643)); \
+ __ret_643; \
})
#else
-#define vqdmulls_lane_s32(__p0_167, __p1_167, __p2_167) __extension__ ({ \
- int32_t __s0_167 = __p0_167; \
- int32x2_t __s1_167 = __p1_167; \
- int32x2_t __rev1_167; __rev1_167 = __builtin_shufflevector(__s1_167, __s1_167, 1, 0); \
- int64_t __ret_167; \
- __ret_167 = __noswap_vqdmulls_s32(__s0_167, __noswap_vget_lane_s32(__rev1_167, __p2_167)); \
- __ret_167; \
+#define vqdmulls_lane_s32(__p0_644, __p1_644, __p2_644) __extension__ ({ \
+ int32_t __s0_644 = __p0_644; \
+ int32x2_t __s1_644 = __p1_644; \
+ int32x2_t __rev1_644; __rev1_644 = __builtin_shufflevector(__s1_644, __s1_644, 1, 0); \
+ int64_t __ret_644; \
+ __ret_644 = vqdmulls_s32(__s0_644, __noswap_vget_lane_s32(__rev1_644, __p2_644)); \
+ __ret_644; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmullh_lane_s16(__p0_168, __p1_168, __p2_168) __extension__ ({ \
- int16_t __s0_168 = __p0_168; \
- int16x4_t __s1_168 = __p1_168; \
- int32_t __ret_168; \
- __ret_168 = vqdmullh_s16(__s0_168, vget_lane_s16(__s1_168, __p2_168)); \
- __ret_168; \
+#define vqdmullh_lane_s16(__p0_645, __p1_645, __p2_645) __extension__ ({ \
+ int16_t __s0_645 = __p0_645; \
+ int16x4_t __s1_645 = __p1_645; \
+ int32_t __ret_645; \
+ __ret_645 = vqdmullh_s16(__s0_645, vget_lane_s16(__s1_645, __p2_645)); \
+ __ret_645; \
})
#else
-#define vqdmullh_lane_s16(__p0_169, __p1_169, __p2_169) __extension__ ({ \
- int16_t __s0_169 = __p0_169; \
- int16x4_t __s1_169 = __p1_169; \
- int16x4_t __rev1_169; __rev1_169 = __builtin_shufflevector(__s1_169, __s1_169, 3, 2, 1, 0); \
- int32_t __ret_169; \
- __ret_169 = __noswap_vqdmullh_s16(__s0_169, __noswap_vget_lane_s16(__rev1_169, __p2_169)); \
- __ret_169; \
+#define vqdmullh_lane_s16(__p0_646, __p1_646, __p2_646) __extension__ ({ \
+ int16_t __s0_646 = __p0_646; \
+ int16x4_t __s1_646 = __p1_646; \
+ int16x4_t __rev1_646; __rev1_646 = __builtin_shufflevector(__s1_646, __s1_646, 3, 2, 1, 0); \
+ int32_t __ret_646; \
+ __ret_646 = vqdmullh_s16(__s0_646, __noswap_vget_lane_s16(__rev1_646, __p2_646)); \
+ __ret_646; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmulls_laneq_s32(__p0_170, __p1_170, __p2_170) __extension__ ({ \
- int32_t __s0_170 = __p0_170; \
- int32x4_t __s1_170 = __p1_170; \
- int64_t __ret_170; \
- __ret_170 = vqdmulls_s32(__s0_170, vgetq_lane_s32(__s1_170, __p2_170)); \
- __ret_170; \
+#define vqdmulls_laneq_s32(__p0_647, __p1_647, __p2_647) __extension__ ({ \
+ int32_t __s0_647 = __p0_647; \
+ int32x4_t __s1_647 = __p1_647; \
+ int64_t __ret_647; \
+ __ret_647 = vqdmulls_s32(__s0_647, vgetq_lane_s32(__s1_647, __p2_647)); \
+ __ret_647; \
})
#else
-#define vqdmulls_laneq_s32(__p0_171, __p1_171, __p2_171) __extension__ ({ \
- int32_t __s0_171 = __p0_171; \
- int32x4_t __s1_171 = __p1_171; \
- int32x4_t __rev1_171; __rev1_171 = __builtin_shufflevector(__s1_171, __s1_171, 3, 2, 1, 0); \
- int64_t __ret_171; \
- __ret_171 = __noswap_vqdmulls_s32(__s0_171, __noswap_vgetq_lane_s32(__rev1_171, __p2_171)); \
- __ret_171; \
+#define vqdmulls_laneq_s32(__p0_648, __p1_648, __p2_648) __extension__ ({ \
+ int32_t __s0_648 = __p0_648; \
+ int32x4_t __s1_648 = __p1_648; \
+ int32x4_t __rev1_648; __rev1_648 = __builtin_shufflevector(__s1_648, __s1_648, 3, 2, 1, 0); \
+ int64_t __ret_648; \
+ __ret_648 = vqdmulls_s32(__s0_648, __noswap_vgetq_lane_s32(__rev1_648, __p2_648)); \
+ __ret_648; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmullh_laneq_s16(__p0_172, __p1_172, __p2_172) __extension__ ({ \
- int16_t __s0_172 = __p0_172; \
- int16x8_t __s1_172 = __p1_172; \
- int32_t __ret_172; \
- __ret_172 = vqdmullh_s16(__s0_172, vgetq_lane_s16(__s1_172, __p2_172)); \
- __ret_172; \
+#define vqdmullh_laneq_s16(__p0_649, __p1_649, __p2_649) __extension__ ({ \
+ int16_t __s0_649 = __p0_649; \
+ int16x8_t __s1_649 = __p1_649; \
+ int32_t __ret_649; \
+ __ret_649 = vqdmullh_s16(__s0_649, vgetq_lane_s16(__s1_649, __p2_649)); \
+ __ret_649; \
})
#else
-#define vqdmullh_laneq_s16(__p0_173, __p1_173, __p2_173) __extension__ ({ \
- int16_t __s0_173 = __p0_173; \
- int16x8_t __s1_173 = __p1_173; \
- int16x8_t __rev1_173; __rev1_173 = __builtin_shufflevector(__s1_173, __s1_173, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32_t __ret_173; \
- __ret_173 = __noswap_vqdmullh_s16(__s0_173, __noswap_vgetq_lane_s16(__rev1_173, __p2_173)); \
- __ret_173; \
+#define vqdmullh_laneq_s16(__p0_650, __p1_650, __p2_650) __extension__ ({ \
+ int16_t __s0_650 = __p0_650; \
+ int16x8_t __s1_650 = __p1_650; \
+ int16x8_t __rev1_650; __rev1_650 = __builtin_shufflevector(__s1_650, __s1_650, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32_t __ret_650; \
+ __ret_650 = vqdmullh_s16(__s0_650, __noswap_vgetq_lane_s16(__rev1_650, __p2_650)); \
+ __ret_650; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmull_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int64x2_t __ret; \
- __ret = vqdmull_s32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
- __ret; \
+#define vqdmull_laneq_s32(__p0_651, __p1_651, __p2_651) __extension__ ({ \
+ int32x2_t __s0_651 = __p0_651; \
+ int32x4_t __s1_651 = __p1_651; \
+ int64x2_t __ret_651; \
+ __ret_651 = vqdmull_s32(__s0_651, splat_laneq_s32(__s1_651, __p2_651)); \
+ __ret_651; \
})
#else
-#define vqdmull_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
- int32x2_t __s0 = __p0; \
- int32x4_t __s1 = __p1; \
- int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int64x2_t __ret; \
- __ret = __noswap_vqdmull_s32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vqdmull_laneq_s32(__p0_652, __p1_652, __p2_652) __extension__ ({ \
+ int32x2_t __s0_652 = __p0_652; \
+ int32x4_t __s1_652 = __p1_652; \
+ int32x2_t __rev0_652; __rev0_652 = __builtin_shufflevector(__s0_652, __s0_652, 1, 0); \
+ int32x4_t __rev1_652; __rev1_652 = __builtin_shufflevector(__s1_652, __s1_652, 3, 2, 1, 0); \
+ int64x2_t __ret_652; \
+ __ret_652 = __noswap_vqdmull_s32(__rev0_652, __noswap_splat_laneq_s32(__rev1_652, __p2_652)); \
+ __ret_652 = __builtin_shufflevector(__ret_652, __ret_652, 1, 0); \
+ __ret_652; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqdmull_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int32x4_t __ret; \
- __ret = vqdmull_s16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
- __ret; \
+#define vqdmull_laneq_s16(__p0_653, __p1_653, __p2_653) __extension__ ({ \
+ int16x4_t __s0_653 = __p0_653; \
+ int16x8_t __s1_653 = __p1_653; \
+ int32x4_t __ret_653; \
+ __ret_653 = vqdmull_s16(__s0_653, splat_laneq_s16(__s1_653, __p2_653)); \
+ __ret_653; \
})
#else
-#define vqdmull_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
- int16x4_t __s0 = __p0; \
- int16x8_t __s1 = __p1; \
- int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __noswap_vqdmull_s16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vqdmull_laneq_s16(__p0_654, __p1_654, __p2_654) __extension__ ({ \
+ int16x4_t __s0_654 = __p0_654; \
+ int16x8_t __s1_654 = __p1_654; \
+ int16x4_t __rev0_654; __rev0_654 = __builtin_shufflevector(__s0_654, __s0_654, 3, 2, 1, 0); \
+ int16x8_t __rev1_654; __rev1_654 = __builtin_shufflevector(__s1_654, __s1_654, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_654; \
+ __ret_654 = __noswap_vqdmull_s16(__rev0_654, __noswap_splat_laneq_s16(__rev1_654, __p2_654)); \
+ __ret_654 = __builtin_shufflevector(__ret_654, __ret_654, 3, 2, 1, 0); \
+ __ret_654; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int16_t vqmovns_s32(int32_t __p0) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vqmovns_s32(__p0);
- return __ret;
-}
-#else
__ai int16_t vqmovns_s32(int32_t __p0) {
int16_t __ret;
__ret = (int16_t) __builtin_neon_vqmovns_s32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vqmovnd_s64(int64_t __p0) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqmovnd_s64(__p0);
- return __ret;
-}
-#else
__ai int32_t vqmovnd_s64(int64_t __p0) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vqmovnd_s64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8_t vqmovnh_s16(int16_t __p0) {
- int8_t __ret;
- __ret = (int8_t) __builtin_neon_vqmovnh_s16(__p0);
- return __ret;
-}
-#else
__ai int8_t vqmovnh_s16(int16_t __p0) {
int8_t __ret;
__ret = (int8_t) __builtin_neon_vqmovnh_s16(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16_t vqmovns_u32(uint32_t __p0) {
- uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vqmovns_u32(__p0);
- return __ret;
-}
-#else
__ai uint16_t vqmovns_u32(uint32_t __p0) {
uint16_t __ret;
__ret = (uint16_t) __builtin_neon_vqmovns_u32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vqmovnd_u64(uint64_t __p0) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vqmovnd_u64(__p0);
- return __ret;
-}
-#else
__ai uint32_t vqmovnd_u64(uint64_t __p0) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vqmovnd_u64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8_t vqmovnh_u16(uint16_t __p0) {
- uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vqmovnh_u16(__p0);
- return __ret;
-}
-#else
__ai uint8_t vqmovnh_u16(uint16_t __p0) {
uint8_t __ret;
__ret = (uint8_t) __builtin_neon_vqmovnh_u16(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x8_t vqmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
uint16x8_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int16_t vqmovuns_s32(int32_t __p0) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vqmovuns_s32(__p0);
- return __ret;
-}
-#else
__ai int16_t vqmovuns_s32(int32_t __p0) {
int16_t __ret;
__ret = (int16_t) __builtin_neon_vqmovuns_s32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vqmovund_s64(int64_t __p0) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqmovund_s64(__p0);
- return __ret;
-}
-#else
__ai int32_t vqmovund_s64(int64_t __p0) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vqmovund_s64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8_t vqmovunh_s16(int16_t __p0) {
- int8_t __ret;
- __ret = (int8_t) __builtin_neon_vqmovunh_s16(__p0);
- return __ret;
-}
-#else
__ai int8_t vqmovunh_s16(int16_t __p0) {
int8_t __ret;
__ret = (int8_t) __builtin_neon_vqmovunh_s16(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x8_t vqmovun_high_s32(int16x4_t __p0, int32x4_t __p1) {
uint16x8_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vqneg_s64(int64x1_t __p0) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vqneg_v((int8x8_t)__p0, 3);
- return __ret;
-}
-#else
__ai int64x1_t vqneg_s64(int64x1_t __p0) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vqneg_v((int8x8_t)__p0, 3);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8_t vqnegb_s8(int8_t __p0) {
- int8_t __ret;
- __ret = (int8_t) __builtin_neon_vqnegb_s8(__p0);
- return __ret;
-}
-#else
__ai int8_t vqnegb_s8(int8_t __p0) {
int8_t __ret;
__ret = (int8_t) __builtin_neon_vqnegb_s8(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vqnegs_s32(int32_t __p0) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqnegs_s32(__p0);
- return __ret;
-}
-#else
__ai int32_t vqnegs_s32(int32_t __p0) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vqnegs_s32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vqnegd_s64(int64_t __p0) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vqnegd_s64(__p0);
- return __ret;
-}
-#else
__ai int64_t vqnegd_s64(int64_t __p0) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vqnegd_s64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16_t vqnegh_s16(int16_t __p0) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vqnegh_s16(__p0);
- return __ret;
-}
-#else
__ai int16_t vqnegh_s16(int16_t __p0) {
int16_t __ret;
__ret = (int16_t) __builtin_neon_vqnegh_s16(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vqrdmulhs_s32(int32_t __p0, int32_t __p1) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqrdmulhs_s32(__p0, __p1);
- return __ret;
-}
-#else
__ai int32_t vqrdmulhs_s32(int32_t __p0, int32_t __p1) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vqrdmulhs_s32(__p0, __p1);
return __ret;
}
-__ai int32_t __noswap_vqrdmulhs_s32(int32_t __p0, int32_t __p1) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqrdmulhs_s32(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
int16_t __ret;
__ret = (int16_t) __builtin_neon_vqrdmulhh_s16(__p0, __p1);
return __ret;
}
+#ifdef __LITTLE_ENDIAN__
+#define vqrdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
+ int32x4_t __s0 = __p0; \
+ int32x2_t __s1 = __p1; \
+ int32x4_t __ret; \
+ __ret = (int32x4_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 2); \
+ __ret; \
+})
#else
-__ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vqrdmulhh_s16(__p0, __p1);
- return __ret;
-}
-__ai int16_t __noswap_vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vqrdmulhh_s16(__p0, __p1);
- return __ret;
-}
+#define vqrdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
+ int32x4_t __s0 = __p0; \
+ int32x2_t __s1 = __p1; \
+ int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
+ int32x4_t __ret; \
+ __ret = (int32x4_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmulhs_lane_s32(__p0_174, __p1_174, __p2_174) __extension__ ({ \
- int32_t __s0_174 = __p0_174; \
- int32x2_t __s1_174 = __p1_174; \
- int32_t __ret_174; \
- __ret_174 = vqrdmulhs_s32(__s0_174, vget_lane_s32(__s1_174, __p2_174)); \
- __ret_174; \
+#define vqrdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
+ int16x8_t __s0 = __p0; \
+ int16x4_t __s1 = __p1; \
+ int16x8_t __ret; \
+ __ret = (int16x8_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 1); \
+ __ret; \
})
#else
-#define vqrdmulhs_lane_s32(__p0_175, __p1_175, __p2_175) __extension__ ({ \
- int32_t __s0_175 = __p0_175; \
- int32x2_t __s1_175 = __p1_175; \
- int32x2_t __rev1_175; __rev1_175 = __builtin_shufflevector(__s1_175, __s1_175, 1, 0); \
- int32_t __ret_175; \
- __ret_175 = __noswap_vqrdmulhs_s32(__s0_175, __noswap_vget_lane_s32(__rev1_175, __p2_175)); \
- __ret_175; \
+#define vqrdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
+ int16x8_t __s0 = __p0; \
+ int16x4_t __s1 = __p1; \
+ int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
+ int16x8_t __ret; \
+ __ret = (int16x8_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
+ __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmulhh_lane_s16(__p0_176, __p1_176, __p2_176) __extension__ ({ \
- int16_t __s0_176 = __p0_176; \
- int16x4_t __s1_176 = __p1_176; \
- int16_t __ret_176; \
- __ret_176 = vqrdmulhh_s16(__s0_176, vget_lane_s16(__s1_176, __p2_176)); \
- __ret_176; \
+#define vqrdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
+ int32x2_t __s0 = __p0; \
+ int32x2_t __s1 = __p1; \
+ int32x2_t __ret; \
+ __ret = (int32x2_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 2); \
+ __ret; \
})
#else
-#define vqrdmulhh_lane_s16(__p0_177, __p1_177, __p2_177) __extension__ ({ \
- int16_t __s0_177 = __p0_177; \
- int16x4_t __s1_177 = __p1_177; \
- int16x4_t __rev1_177; __rev1_177 = __builtin_shufflevector(__s1_177, __s1_177, 3, 2, 1, 0); \
- int16_t __ret_177; \
- __ret_177 = __noswap_vqrdmulhh_s16(__s0_177, __noswap_vget_lane_s16(__rev1_177, __p2_177)); \
- __ret_177; \
+#define vqrdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
+ int32x2_t __s0 = __p0; \
+ int32x2_t __s1 = __p1; \
+ int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
+ int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
+ int32x2_t __ret; \
+ __ret = (int32x2_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
+ __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
+ __ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmulhs_laneq_s32(__p0_178, __p1_178, __p2_178) __extension__ ({ \
- int32_t __s0_178 = __p0_178; \
- int32x4_t __s1_178 = __p1_178; \
- int32_t __ret_178; \
- __ret_178 = vqrdmulhs_s32(__s0_178, vgetq_lane_s32(__s1_178, __p2_178)); \
- __ret_178; \
+#define vqrdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
+ int16x4_t __s0 = __p0; \
+ int16x4_t __s1 = __p1; \
+ int16x4_t __ret; \
+ __ret = (int16x4_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 1); \
+ __ret; \
})
#else
-#define vqrdmulhs_laneq_s32(__p0_179, __p1_179, __p2_179) __extension__ ({ \
- int32_t __s0_179 = __p0_179; \
- int32x4_t __s1_179 = __p1_179; \
- int32x4_t __rev1_179; __rev1_179 = __builtin_shufflevector(__s1_179, __s1_179, 3, 2, 1, 0); \
- int32_t __ret_179; \
- __ret_179 = __noswap_vqrdmulhs_s32(__s0_179, __noswap_vgetq_lane_s32(__rev1_179, __p2_179)); \
- __ret_179; \
+#define vqrdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
+ int16x4_t __s0 = __p0; \
+ int16x4_t __s1 = __p1; \
+ int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
+ int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
+ int16x4_t __ret; \
+ __ret = (int16x4_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
+ __ret; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmulhh_laneq_s16(__p0_180, __p1_180, __p2_180) __extension__ ({ \
- int16_t __s0_180 = __p0_180; \
- int16x8_t __s1_180 = __p1_180; \
- int16_t __ret_180; \
- __ret_180 = vqrdmulhh_s16(__s0_180, vgetq_lane_s16(__s1_180, __p2_180)); \
- __ret_180; \
+#define vqrdmulhs_lane_s32(__p0_655, __p1_655, __p2_655) __extension__ ({ \
+ int32_t __s0_655 = __p0_655; \
+ int32x2_t __s1_655 = __p1_655; \
+ int32_t __ret_655; \
+ __ret_655 = vqrdmulhs_s32(__s0_655, vget_lane_s32(__s1_655, __p2_655)); \
+ __ret_655; \
})
#else
-#define vqrdmulhh_laneq_s16(__p0_181, __p1_181, __p2_181) __extension__ ({ \
- int16_t __s0_181 = __p0_181; \
- int16x8_t __s1_181 = __p1_181; \
- int16x8_t __rev1_181; __rev1_181 = __builtin_shufflevector(__s1_181, __s1_181, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16_t __ret_181; \
- __ret_181 = __noswap_vqrdmulhh_s16(__s0_181, __noswap_vgetq_lane_s16(__rev1_181, __p2_181)); \
- __ret_181; \
+#define vqrdmulhs_lane_s32(__p0_656, __p1_656, __p2_656) __extension__ ({ \
+ int32_t __s0_656 = __p0_656; \
+ int32x2_t __s1_656 = __p1_656; \
+ int32x2_t __rev1_656; __rev1_656 = __builtin_shufflevector(__s1_656, __s1_656, 1, 0); \
+ int32_t __ret_656; \
+ __ret_656 = vqrdmulhs_s32(__s0_656, __noswap_vget_lane_s32(__rev1_656, __p2_656)); \
+ __ret_656; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vqrdmulhh_lane_s16(__p0_657, __p1_657, __p2_657) __extension__ ({ \
+ int16_t __s0_657 = __p0_657; \
+ int16x4_t __s1_657 = __p1_657; \
+ int16_t __ret_657; \
+ __ret_657 = vqrdmulhh_s16(__s0_657, vget_lane_s16(__s1_657, __p2_657)); \
+ __ret_657; \
+})
+#else
+#define vqrdmulhh_lane_s16(__p0_658, __p1_658, __p2_658) __extension__ ({ \
+ int16_t __s0_658 = __p0_658; \
+ int16x4_t __s1_658 = __p1_658; \
+ int16x4_t __rev1_658; __rev1_658 = __builtin_shufflevector(__s1_658, __s1_658, 3, 2, 1, 0); \
+ int16_t __ret_658; \
+ __ret_658 = vqrdmulhh_s16(__s0_658, __noswap_vget_lane_s16(__rev1_658, __p2_658)); \
+ __ret_658; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vqrdmulhs_laneq_s32(__p0_659, __p1_659, __p2_659) __extension__ ({ \
+ int32_t __s0_659 = __p0_659; \
+ int32x4_t __s1_659 = __p1_659; \
+ int32_t __ret_659; \
+ __ret_659 = vqrdmulhs_s32(__s0_659, vgetq_lane_s32(__s1_659, __p2_659)); \
+ __ret_659; \
+})
+#else
+#define vqrdmulhs_laneq_s32(__p0_660, __p1_660, __p2_660) __extension__ ({ \
+ int32_t __s0_660 = __p0_660; \
+ int32x4_t __s1_660 = __p1_660; \
+ int32x4_t __rev1_660; __rev1_660 = __builtin_shufflevector(__s1_660, __s1_660, 3, 2, 1, 0); \
+ int32_t __ret_660; \
+ __ret_660 = vqrdmulhs_s32(__s0_660, __noswap_vgetq_lane_s32(__rev1_660, __p2_660)); \
+ __ret_660; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vqrdmulhh_laneq_s16(__p0_661, __p1_661, __p2_661) __extension__ ({ \
+ int16_t __s0_661 = __p0_661; \
+ int16x8_t __s1_661 = __p1_661; \
+ int16_t __ret_661; \
+ __ret_661 = vqrdmulhh_s16(__s0_661, vgetq_lane_s16(__s1_661, __p2_661)); \
+ __ret_661; \
+})
+#else
+#define vqrdmulhh_laneq_s16(__p0_662, __p1_662, __p2_662) __extension__ ({ \
+ int16_t __s0_662 = __p0_662; \
+ int16x8_t __s1_662 = __p1_662; \
+ int16x8_t __rev1_662; __rev1_662 = __builtin_shufflevector(__s1_662, __s1_662, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16_t __ret_662; \
+ __ret_662 = vqrdmulhh_s16(__s0_662, __noswap_vgetq_lane_s16(__rev1_662, __p2_662)); \
+ __ret_662; \
})
#endif
int32x4_t __s0 = __p0; \
int32x4_t __s1 = __p1; \
int32x4_t __ret; \
- __ret = vqrdmulhq_s32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
+ __ret = (int32x4_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 34); \
__ret; \
})
#else
int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
int32x4_t __ret; \
- __ret = __noswap_vqrdmulhq_s32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
+ __ret = (int32x4_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 34); \
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
__ret; \
})
int16x8_t __s0 = __p0; \
int16x8_t __s1 = __p1; \
int16x8_t __ret; \
- __ret = vqrdmulhq_s16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2)); \
+ __ret = (int16x8_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 33); \
__ret; \
})
#else
int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
int16x8_t __ret; \
- __ret = __noswap_vqrdmulhq_s16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2)); \
+ __ret = (int16x8_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 33); \
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
__ret; \
})
int32x2_t __s0 = __p0; \
int32x4_t __s1 = __p1; \
int32x2_t __ret; \
- __ret = vqrdmulh_s32(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2)); \
+ __ret = (int32x2_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 2); \
__ret; \
})
#else
int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
int32x2_t __ret; \
- __ret = __noswap_vqrdmulh_s32(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2)); \
+ __ret = (int32x2_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 2); \
__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
__ret; \
})
int16x4_t __s0 = __p0; \
int16x8_t __s1 = __p1; \
int16x4_t __ret; \
- __ret = vqrdmulh_s16(__s0, __builtin_shufflevector(__s1, __s1, __p2, __p2, __p2, __p2)); \
+ __ret = (int16x4_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 1); \
__ret; \
})
#else
int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
int16x4_t __ret; \
- __ret = __noswap_vqrdmulh_s16(__rev0, __builtin_shufflevector(__rev1, __rev1, __p2, __p2, __p2, __p2)); \
+ __ret = (int16x4_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 1); \
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
__ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint8_t vqrshlb_u8(uint8_t __p0, uint8_t __p1) {
- uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vqrshlb_u8(__p0, __p1);
- return __ret;
-}
-#else
__ai uint8_t vqrshlb_u8(uint8_t __p0, uint8_t __p1) {
uint8_t __ret;
__ret = (uint8_t) __builtin_neon_vqrshlb_u8(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vqrshls_u32(uint32_t __p0, uint32_t __p1) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vqrshls_u32(__p0, __p1);
- return __ret;
-}
-#else
__ai uint32_t vqrshls_u32(uint32_t __p0, uint32_t __p1) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vqrshls_u32(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vqrshld_u64(uint64_t __p0, uint64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vqrshld_u64(__p0, __p1);
- return __ret;
-}
-#else
__ai uint64_t vqrshld_u64(uint64_t __p0, uint64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vqrshld_u64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16_t vqrshlh_u16(uint16_t __p0, uint16_t __p1) {
- uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vqrshlh_u16(__p0, __p1);
- return __ret;
-}
-#else
__ai uint16_t vqrshlh_u16(uint16_t __p0, uint16_t __p1) {
uint16_t __ret;
__ret = (uint16_t) __builtin_neon_vqrshlh_u16(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8_t vqrshlb_s8(int8_t __p0, int8_t __p1) {
- int8_t __ret;
- __ret = (int8_t) __builtin_neon_vqrshlb_s8(__p0, __p1);
- return __ret;
-}
-#else
__ai int8_t vqrshlb_s8(int8_t __p0, int8_t __p1) {
int8_t __ret;
__ret = (int8_t) __builtin_neon_vqrshlb_s8(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vqrshls_s32(int32_t __p0, int32_t __p1) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqrshls_s32(__p0, __p1);
- return __ret;
-}
-#else
__ai int32_t vqrshls_s32(int32_t __p0, int32_t __p1) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vqrshls_s32(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vqrshld_s64(int64_t __p0, int64_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vqrshld_s64(__p0, __p1);
- return __ret;
-}
-#else
__ai int64_t vqrshld_s64(int64_t __p0, int64_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vqrshld_s64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vqrshlh_s16(__p0, __p1);
- return __ret;
-}
-#else
__ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
int16_t __ret;
__ret = (int16_t) __builtin_neon_vqrshlh_s16(__p0, __p1);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vqrshrn_high_n_u32(__p0_182, __p1_182, __p2_182) __extension__ ({ \
- uint16x4_t __s0_182 = __p0_182; \
- uint32x4_t __s1_182 = __p1_182; \
- uint16x8_t __ret_182; \
- __ret_182 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_182), (uint16x4_t)(vqrshrn_n_u32(__s1_182, __p2_182)))); \
- __ret_182; \
-})
-#else
-#define vqrshrn_high_n_u32(__p0_183, __p1_183, __p2_183) __extension__ ({ \
- uint16x4_t __s0_183 = __p0_183; \
- uint32x4_t __s1_183 = __p1_183; \
- uint16x4_t __rev0_183; __rev0_183 = __builtin_shufflevector(__s0_183, __s0_183, 3, 2, 1, 0); \
- uint32x4_t __rev1_183; __rev1_183 = __builtin_shufflevector(__s1_183, __s1_183, 3, 2, 1, 0); \
- uint16x8_t __ret_183; \
- __ret_183 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_183), (uint16x4_t)(__noswap_vqrshrn_n_u32(__rev1_183, __p2_183)))); \
- __ret_183 = __builtin_shufflevector(__ret_183, __ret_183, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_183; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqrshrn_high_n_u64(__p0_184, __p1_184, __p2_184) __extension__ ({ \
- uint32x2_t __s0_184 = __p0_184; \
- uint64x2_t __s1_184 = __p1_184; \
- uint32x4_t __ret_184; \
- __ret_184 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_184), (uint32x2_t)(vqrshrn_n_u64(__s1_184, __p2_184)))); \
- __ret_184; \
+#define vqrshrn_high_n_u32(__p0_663, __p1_663, __p2_663) __extension__ ({ \
+ uint16x4_t __s0_663 = __p0_663; \
+ uint32x4_t __s1_663 = __p1_663; \
+ uint16x8_t __ret_663; \
+ __ret_663 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_663), (uint16x4_t)(vqrshrn_n_u32(__s1_663, __p2_663)))); \
+ __ret_663; \
})
#else
-#define vqrshrn_high_n_u64(__p0_185, __p1_185, __p2_185) __extension__ ({ \
- uint32x2_t __s0_185 = __p0_185; \
- uint64x2_t __s1_185 = __p1_185; \
- uint32x2_t __rev0_185; __rev0_185 = __builtin_shufflevector(__s0_185, __s0_185, 1, 0); \
- uint64x2_t __rev1_185; __rev1_185 = __builtin_shufflevector(__s1_185, __s1_185, 1, 0); \
- uint32x4_t __ret_185; \
- __ret_185 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_185), (uint32x2_t)(__noswap_vqrshrn_n_u64(__rev1_185, __p2_185)))); \
- __ret_185 = __builtin_shufflevector(__ret_185, __ret_185, 3, 2, 1, 0); \
- __ret_185; \
+#define vqrshrn_high_n_u32(__p0_664, __p1_664, __p2_664) __extension__ ({ \
+ uint16x4_t __s0_664 = __p0_664; \
+ uint32x4_t __s1_664 = __p1_664; \
+ uint16x4_t __rev0_664; __rev0_664 = __builtin_shufflevector(__s0_664, __s0_664, 3, 2, 1, 0); \
+ uint32x4_t __rev1_664; __rev1_664 = __builtin_shufflevector(__s1_664, __s1_664, 3, 2, 1, 0); \
+ uint16x8_t __ret_664; \
+ __ret_664 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_664), (uint16x4_t)(__noswap_vqrshrn_n_u32(__rev1_664, __p2_664)))); \
+ __ret_664 = __builtin_shufflevector(__ret_664, __ret_664, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_664; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrshrn_high_n_u16(__p0_186, __p1_186, __p2_186) __extension__ ({ \
- uint8x8_t __s0_186 = __p0_186; \
- uint16x8_t __s1_186 = __p1_186; \
- uint8x16_t __ret_186; \
- __ret_186 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_186), (uint8x8_t)(vqrshrn_n_u16(__s1_186, __p2_186)))); \
- __ret_186; \
+#define vqrshrn_high_n_u64(__p0_665, __p1_665, __p2_665) __extension__ ({ \
+ uint32x2_t __s0_665 = __p0_665; \
+ uint64x2_t __s1_665 = __p1_665; \
+ uint32x4_t __ret_665; \
+ __ret_665 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_665), (uint32x2_t)(vqrshrn_n_u64(__s1_665, __p2_665)))); \
+ __ret_665; \
})
#else
-#define vqrshrn_high_n_u16(__p0_187, __p1_187, __p2_187) __extension__ ({ \
- uint8x8_t __s0_187 = __p0_187; \
- uint16x8_t __s1_187 = __p1_187; \
- uint8x8_t __rev0_187; __rev0_187 = __builtin_shufflevector(__s0_187, __s0_187, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __rev1_187; __rev1_187 = __builtin_shufflevector(__s1_187, __s1_187, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x16_t __ret_187; \
- __ret_187 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_187), (uint8x8_t)(__noswap_vqrshrn_n_u16(__rev1_187, __p2_187)))); \
- __ret_187 = __builtin_shufflevector(__ret_187, __ret_187, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_187; \
+#define vqrshrn_high_n_u64(__p0_666, __p1_666, __p2_666) __extension__ ({ \
+ uint32x2_t __s0_666 = __p0_666; \
+ uint64x2_t __s1_666 = __p1_666; \
+ uint32x2_t __rev0_666; __rev0_666 = __builtin_shufflevector(__s0_666, __s0_666, 1, 0); \
+ uint64x2_t __rev1_666; __rev1_666 = __builtin_shufflevector(__s1_666, __s1_666, 1, 0); \
+ uint32x4_t __ret_666; \
+ __ret_666 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_666), (uint32x2_t)(__noswap_vqrshrn_n_u64(__rev1_666, __p2_666)))); \
+ __ret_666 = __builtin_shufflevector(__ret_666, __ret_666, 3, 2, 1, 0); \
+ __ret_666; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrshrn_high_n_s32(__p0_188, __p1_188, __p2_188) __extension__ ({ \
- int16x4_t __s0_188 = __p0_188; \
- int32x4_t __s1_188 = __p1_188; \
- int16x8_t __ret_188; \
- __ret_188 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_188), (int16x4_t)(vqrshrn_n_s32(__s1_188, __p2_188)))); \
- __ret_188; \
+#define vqrshrn_high_n_u16(__p0_667, __p1_667, __p2_667) __extension__ ({ \
+ uint8x8_t __s0_667 = __p0_667; \
+ uint16x8_t __s1_667 = __p1_667; \
+ uint8x16_t __ret_667; \
+ __ret_667 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_667), (uint8x8_t)(vqrshrn_n_u16(__s1_667, __p2_667)))); \
+ __ret_667; \
})
#else
-#define vqrshrn_high_n_s32(__p0_189, __p1_189, __p2_189) __extension__ ({ \
- int16x4_t __s0_189 = __p0_189; \
- int32x4_t __s1_189 = __p1_189; \
- int16x4_t __rev0_189; __rev0_189 = __builtin_shufflevector(__s0_189, __s0_189, 3, 2, 1, 0); \
- int32x4_t __rev1_189; __rev1_189 = __builtin_shufflevector(__s1_189, __s1_189, 3, 2, 1, 0); \
- int16x8_t __ret_189; \
- __ret_189 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_189), (int16x4_t)(__noswap_vqrshrn_n_s32(__rev1_189, __p2_189)))); \
- __ret_189 = __builtin_shufflevector(__ret_189, __ret_189, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_189; \
+#define vqrshrn_high_n_u16(__p0_668, __p1_668, __p2_668) __extension__ ({ \
+ uint8x8_t __s0_668 = __p0_668; \
+ uint16x8_t __s1_668 = __p1_668; \
+ uint8x8_t __rev0_668; __rev0_668 = __builtin_shufflevector(__s0_668, __s0_668, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __rev1_668; __rev1_668 = __builtin_shufflevector(__s1_668, __s1_668, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __ret_668; \
+ __ret_668 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_668), (uint8x8_t)(__noswap_vqrshrn_n_u16(__rev1_668, __p2_668)))); \
+ __ret_668 = __builtin_shufflevector(__ret_668, __ret_668, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_668; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrshrn_high_n_s64(__p0_190, __p1_190, __p2_190) __extension__ ({ \
- int32x2_t __s0_190 = __p0_190; \
- int64x2_t __s1_190 = __p1_190; \
- int32x4_t __ret_190; \
- __ret_190 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_190), (int32x2_t)(vqrshrn_n_s64(__s1_190, __p2_190)))); \
- __ret_190; \
+#define vqrshrn_high_n_s32(__p0_669, __p1_669, __p2_669) __extension__ ({ \
+ int16x4_t __s0_669 = __p0_669; \
+ int32x4_t __s1_669 = __p1_669; \
+ int16x8_t __ret_669; \
+ __ret_669 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_669), (int16x4_t)(vqrshrn_n_s32(__s1_669, __p2_669)))); \
+ __ret_669; \
})
#else
-#define vqrshrn_high_n_s64(__p0_191, __p1_191, __p2_191) __extension__ ({ \
- int32x2_t __s0_191 = __p0_191; \
- int64x2_t __s1_191 = __p1_191; \
- int32x2_t __rev0_191; __rev0_191 = __builtin_shufflevector(__s0_191, __s0_191, 1, 0); \
- int64x2_t __rev1_191; __rev1_191 = __builtin_shufflevector(__s1_191, __s1_191, 1, 0); \
- int32x4_t __ret_191; \
- __ret_191 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_191), (int32x2_t)(__noswap_vqrshrn_n_s64(__rev1_191, __p2_191)))); \
- __ret_191 = __builtin_shufflevector(__ret_191, __ret_191, 3, 2, 1, 0); \
- __ret_191; \
+#define vqrshrn_high_n_s32(__p0_670, __p1_670, __p2_670) __extension__ ({ \
+ int16x4_t __s0_670 = __p0_670; \
+ int32x4_t __s1_670 = __p1_670; \
+ int16x4_t __rev0_670; __rev0_670 = __builtin_shufflevector(__s0_670, __s0_670, 3, 2, 1, 0); \
+ int32x4_t __rev1_670; __rev1_670 = __builtin_shufflevector(__s1_670, __s1_670, 3, 2, 1, 0); \
+ int16x8_t __ret_670; \
+ __ret_670 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_670), (int16x4_t)(__noswap_vqrshrn_n_s32(__rev1_670, __p2_670)))); \
+ __ret_670 = __builtin_shufflevector(__ret_670, __ret_670, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_670; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrshrn_high_n_s16(__p0_192, __p1_192, __p2_192) __extension__ ({ \
- int8x8_t __s0_192 = __p0_192; \
- int16x8_t __s1_192 = __p1_192; \
- int8x16_t __ret_192; \
- __ret_192 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_192), (int8x8_t)(vqrshrn_n_s16(__s1_192, __p2_192)))); \
- __ret_192; \
+#define vqrshrn_high_n_s64(__p0_671, __p1_671, __p2_671) __extension__ ({ \
+ int32x2_t __s0_671 = __p0_671; \
+ int64x2_t __s1_671 = __p1_671; \
+ int32x4_t __ret_671; \
+ __ret_671 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_671), (int32x2_t)(vqrshrn_n_s64(__s1_671, __p2_671)))); \
+ __ret_671; \
})
#else
-#define vqrshrn_high_n_s16(__p0_193, __p1_193, __p2_193) __extension__ ({ \
- int8x8_t __s0_193 = __p0_193; \
- int16x8_t __s1_193 = __p1_193; \
- int8x8_t __rev0_193; __rev0_193 = __builtin_shufflevector(__s0_193, __s0_193, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1_193; __rev1_193 = __builtin_shufflevector(__s1_193, __s1_193, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x16_t __ret_193; \
- __ret_193 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_193), (int8x8_t)(__noswap_vqrshrn_n_s16(__rev1_193, __p2_193)))); \
- __ret_193 = __builtin_shufflevector(__ret_193, __ret_193, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_193; \
+#define vqrshrn_high_n_s64(__p0_672, __p1_672, __p2_672) __extension__ ({ \
+ int32x2_t __s0_672 = __p0_672; \
+ int64x2_t __s1_672 = __p1_672; \
+ int32x2_t __rev0_672; __rev0_672 = __builtin_shufflevector(__s0_672, __s0_672, 1, 0); \
+ int64x2_t __rev1_672; __rev1_672 = __builtin_shufflevector(__s1_672, __s1_672, 1, 0); \
+ int32x4_t __ret_672; \
+ __ret_672 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_672), (int32x2_t)(__noswap_vqrshrn_n_s64(__rev1_672, __p2_672)))); \
+ __ret_672 = __builtin_shufflevector(__ret_672, __ret_672, 3, 2, 1, 0); \
+ __ret_672; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrshrns_n_u32(__p0, __p1) __extension__ ({ \
- uint32_t __s0 = __p0; \
- uint16_t __ret; \
- __ret = (uint16_t) __builtin_neon_vqrshrns_n_u32(__s0, __p1); \
- __ret; \
+#define vqrshrn_high_n_s16(__p0_673, __p1_673, __p2_673) __extension__ ({ \
+ int8x8_t __s0_673 = __p0_673; \
+ int16x8_t __s1_673 = __p1_673; \
+ int8x16_t __ret_673; \
+ __ret_673 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_673), (int8x8_t)(vqrshrn_n_s16(__s1_673, __p2_673)))); \
+ __ret_673; \
})
#else
+#define vqrshrn_high_n_s16(__p0_674, __p1_674, __p2_674) __extension__ ({ \
+ int8x8_t __s0_674 = __p0_674; \
+ int16x8_t __s1_674 = __p1_674; \
+ int8x8_t __rev0_674; __rev0_674 = __builtin_shufflevector(__s0_674, __s0_674, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_674; __rev1_674 = __builtin_shufflevector(__s1_674, __s1_674, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __ret_674; \
+ __ret_674 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_674), (int8x8_t)(__noswap_vqrshrn_n_s16(__rev1_674, __p2_674)))); \
+ __ret_674 = __builtin_shufflevector(__ret_674, __ret_674, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_674; \
+})
+#endif
+
#define vqrshrns_n_u32(__p0, __p1) __extension__ ({ \
uint32_t __s0 = __p0; \
uint16_t __ret; \
__ret = (uint16_t) __builtin_neon_vqrshrns_n_u32(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqrshrnd_n_u64(__p0, __p1) __extension__ ({ \
- uint64_t __s0 = __p0; \
- uint32_t __ret; \
- __ret = (uint32_t) __builtin_neon_vqrshrnd_n_u64(__s0, __p1); \
- __ret; \
-})
-#else
#define vqrshrnd_n_u64(__p0, __p1) __extension__ ({ \
uint64_t __s0 = __p0; \
uint32_t __ret; \
__ret = (uint32_t) __builtin_neon_vqrshrnd_n_u64(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqrshrnh_n_u16(__p0, __p1) __extension__ ({ \
- uint16_t __s0 = __p0; \
- uint8_t __ret; \
- __ret = (uint8_t) __builtin_neon_vqrshrnh_n_u16(__s0, __p1); \
- __ret; \
-})
-#else
#define vqrshrnh_n_u16(__p0, __p1) __extension__ ({ \
uint16_t __s0 = __p0; \
uint8_t __ret; \
__ret = (uint8_t) __builtin_neon_vqrshrnh_n_u16(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqrshrns_n_s32(__p0, __p1) __extension__ ({ \
- int32_t __s0 = __p0; \
- int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vqrshrns_n_s32(__s0, __p1); \
- __ret; \
-})
-#else
#define vqrshrns_n_s32(__p0, __p1) __extension__ ({ \
int32_t __s0 = __p0; \
int16_t __ret; \
__ret = (int16_t) __builtin_neon_vqrshrns_n_s32(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqrshrnd_n_s64(__p0, __p1) __extension__ ({ \
- int64_t __s0 = __p0; \
- int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vqrshrnd_n_s64(__s0, __p1); \
- __ret; \
-})
-#else
#define vqrshrnd_n_s64(__p0, __p1) __extension__ ({ \
int64_t __s0 = __p0; \
int32_t __ret; \
__ret = (int32_t) __builtin_neon_vqrshrnd_n_s64(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqrshrnh_n_s16(__p0, __p1) __extension__ ({ \
- int16_t __s0 = __p0; \
- int8_t __ret; \
- __ret = (int8_t) __builtin_neon_vqrshrnh_n_s16(__s0, __p1); \
- __ret; \
-})
-#else
#define vqrshrnh_n_s16(__p0, __p1) __extension__ ({ \
int16_t __s0 = __p0; \
int8_t __ret; \
__ret = (int8_t) __builtin_neon_vqrshrnh_n_s16(__s0, __p1); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vqrshrun_high_n_s32(__p0_194, __p1_194, __p2_194) __extension__ ({ \
- int16x4_t __s0_194 = __p0_194; \
- int32x4_t __s1_194 = __p1_194; \
- int16x8_t __ret_194; \
- __ret_194 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_194), (int16x4_t)(vqrshrun_n_s32(__s1_194, __p2_194)))); \
- __ret_194; \
+#define vqrshrun_high_n_s32(__p0_675, __p1_675, __p2_675) __extension__ ({ \
+ int16x4_t __s0_675 = __p0_675; \
+ int32x4_t __s1_675 = __p1_675; \
+ int16x8_t __ret_675; \
+ __ret_675 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_675), (int16x4_t)(vqrshrun_n_s32(__s1_675, __p2_675)))); \
+ __ret_675; \
})
#else
-#define vqrshrun_high_n_s32(__p0_195, __p1_195, __p2_195) __extension__ ({ \
- int16x4_t __s0_195 = __p0_195; \
- int32x4_t __s1_195 = __p1_195; \
- int16x4_t __rev0_195; __rev0_195 = __builtin_shufflevector(__s0_195, __s0_195, 3, 2, 1, 0); \
- int32x4_t __rev1_195; __rev1_195 = __builtin_shufflevector(__s1_195, __s1_195, 3, 2, 1, 0); \
- int16x8_t __ret_195; \
- __ret_195 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_195), (int16x4_t)(__noswap_vqrshrun_n_s32(__rev1_195, __p2_195)))); \
- __ret_195 = __builtin_shufflevector(__ret_195, __ret_195, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_195; \
+#define vqrshrun_high_n_s32(__p0_676, __p1_676, __p2_676) __extension__ ({ \
+ int16x4_t __s0_676 = __p0_676; \
+ int32x4_t __s1_676 = __p1_676; \
+ int16x4_t __rev0_676; __rev0_676 = __builtin_shufflevector(__s0_676, __s0_676, 3, 2, 1, 0); \
+ int32x4_t __rev1_676; __rev1_676 = __builtin_shufflevector(__s1_676, __s1_676, 3, 2, 1, 0); \
+ int16x8_t __ret_676; \
+ __ret_676 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_676), (int16x4_t)(__noswap_vqrshrun_n_s32(__rev1_676, __p2_676)))); \
+ __ret_676 = __builtin_shufflevector(__ret_676, __ret_676, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_676; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrshrun_high_n_s64(__p0_196, __p1_196, __p2_196) __extension__ ({ \
- int32x2_t __s0_196 = __p0_196; \
- int64x2_t __s1_196 = __p1_196; \
- int32x4_t __ret_196; \
- __ret_196 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_196), (int32x2_t)(vqrshrun_n_s64(__s1_196, __p2_196)))); \
- __ret_196; \
+#define vqrshrun_high_n_s64(__p0_677, __p1_677, __p2_677) __extension__ ({ \
+ int32x2_t __s0_677 = __p0_677; \
+ int64x2_t __s1_677 = __p1_677; \
+ int32x4_t __ret_677; \
+ __ret_677 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_677), (int32x2_t)(vqrshrun_n_s64(__s1_677, __p2_677)))); \
+ __ret_677; \
})
#else
-#define vqrshrun_high_n_s64(__p0_197, __p1_197, __p2_197) __extension__ ({ \
- int32x2_t __s0_197 = __p0_197; \
- int64x2_t __s1_197 = __p1_197; \
- int32x2_t __rev0_197; __rev0_197 = __builtin_shufflevector(__s0_197, __s0_197, 1, 0); \
- int64x2_t __rev1_197; __rev1_197 = __builtin_shufflevector(__s1_197, __s1_197, 1, 0); \
- int32x4_t __ret_197; \
- __ret_197 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_197), (int32x2_t)(__noswap_vqrshrun_n_s64(__rev1_197, __p2_197)))); \
- __ret_197 = __builtin_shufflevector(__ret_197, __ret_197, 3, 2, 1, 0); \
- __ret_197; \
+#define vqrshrun_high_n_s64(__p0_678, __p1_678, __p2_678) __extension__ ({ \
+ int32x2_t __s0_678 = __p0_678; \
+ int64x2_t __s1_678 = __p1_678; \
+ int32x2_t __rev0_678; __rev0_678 = __builtin_shufflevector(__s0_678, __s0_678, 1, 0); \
+ int64x2_t __rev1_678; __rev1_678 = __builtin_shufflevector(__s1_678, __s1_678, 1, 0); \
+ int32x4_t __ret_678; \
+ __ret_678 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_678), (int32x2_t)(__noswap_vqrshrun_n_s64(__rev1_678, __p2_678)))); \
+ __ret_678 = __builtin_shufflevector(__ret_678, __ret_678, 3, 2, 1, 0); \
+ __ret_678; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrshrun_high_n_s16(__p0_198, __p1_198, __p2_198) __extension__ ({ \
- int8x8_t __s0_198 = __p0_198; \
- int16x8_t __s1_198 = __p1_198; \
- int8x16_t __ret_198; \
- __ret_198 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_198), (int8x8_t)(vqrshrun_n_s16(__s1_198, __p2_198)))); \
- __ret_198; \
+#define vqrshrun_high_n_s16(__p0_679, __p1_679, __p2_679) __extension__ ({ \
+ int8x8_t __s0_679 = __p0_679; \
+ int16x8_t __s1_679 = __p1_679; \
+ int8x16_t __ret_679; \
+ __ret_679 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_679), (int8x8_t)(vqrshrun_n_s16(__s1_679, __p2_679)))); \
+ __ret_679; \
})
#else
-#define vqrshrun_high_n_s16(__p0_199, __p1_199, __p2_199) __extension__ ({ \
- int8x8_t __s0_199 = __p0_199; \
- int16x8_t __s1_199 = __p1_199; \
- int8x8_t __rev0_199; __rev0_199 = __builtin_shufflevector(__s0_199, __s0_199, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1_199; __rev1_199 = __builtin_shufflevector(__s1_199, __s1_199, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x16_t __ret_199; \
- __ret_199 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_199), (int8x8_t)(__noswap_vqrshrun_n_s16(__rev1_199, __p2_199)))); \
- __ret_199 = __builtin_shufflevector(__ret_199, __ret_199, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_199; \
+#define vqrshrun_high_n_s16(__p0_680, __p1_680, __p2_680) __extension__ ({ \
+ int8x8_t __s0_680 = __p0_680; \
+ int16x8_t __s1_680 = __p1_680; \
+ int8x8_t __rev0_680; __rev0_680 = __builtin_shufflevector(__s0_680, __s0_680, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_680; __rev1_680 = __builtin_shufflevector(__s1_680, __s1_680, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __ret_680; \
+ __ret_680 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_680), (int8x8_t)(__noswap_vqrshrun_n_s16(__rev1_680, __p2_680)))); \
+ __ret_680 = __builtin_shufflevector(__ret_680, __ret_680, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_680; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vqrshruns_n_s32(__p0, __p1) __extension__ ({ \
- int32_t __s0 = __p0; \
- int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vqrshruns_n_s32(__s0, __p1); \
- __ret; \
-})
-#else
#define vqrshruns_n_s32(__p0, __p1) __extension__ ({ \
int32_t __s0 = __p0; \
int16_t __ret; \
__ret = (int16_t) __builtin_neon_vqrshruns_n_s32(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqrshrund_n_s64(__p0, __p1) __extension__ ({ \
- int64_t __s0 = __p0; \
- int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vqrshrund_n_s64(__s0, __p1); \
- __ret; \
-})
-#else
#define vqrshrund_n_s64(__p0, __p1) __extension__ ({ \
int64_t __s0 = __p0; \
int32_t __ret; \
__ret = (int32_t) __builtin_neon_vqrshrund_n_s64(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqrshrunh_n_s16(__p0, __p1) __extension__ ({ \
- int16_t __s0 = __p0; \
- int8_t __ret; \
- __ret = (int8_t) __builtin_neon_vqrshrunh_n_s16(__s0, __p1); \
- __ret; \
-})
-#else
#define vqrshrunh_n_s16(__p0, __p1) __extension__ ({ \
int16_t __s0 = __p0; \
int8_t __ret; \
__ret = (int8_t) __builtin_neon_vqrshrunh_n_s16(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8_t vqshlb_u8(uint8_t __p0, uint8_t __p1) {
- uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vqshlb_u8(__p0, __p1);
- return __ret;
-}
-#else
__ai uint8_t vqshlb_u8(uint8_t __p0, uint8_t __p1) {
uint8_t __ret;
__ret = (uint8_t) __builtin_neon_vqshlb_u8(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vqshls_u32(uint32_t __p0, uint32_t __p1) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vqshls_u32(__p0, __p1);
- return __ret;
-}
-#else
__ai uint32_t vqshls_u32(uint32_t __p0, uint32_t __p1) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vqshls_u32(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vqshld_u64(uint64_t __p0, uint64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vqshld_u64(__p0, __p1);
- return __ret;
-}
-#else
__ai uint64_t vqshld_u64(uint64_t __p0, uint64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vqshld_u64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16_t vqshlh_u16(uint16_t __p0, uint16_t __p1) {
- uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vqshlh_u16(__p0, __p1);
- return __ret;
-}
-#else
__ai uint16_t vqshlh_u16(uint16_t __p0, uint16_t __p1) {
uint16_t __ret;
__ret = (uint16_t) __builtin_neon_vqshlh_u16(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8_t vqshlb_s8(int8_t __p0, int8_t __p1) {
- int8_t __ret;
- __ret = (int8_t) __builtin_neon_vqshlb_s8(__p0, __p1);
- return __ret;
-}
-#else
__ai int8_t vqshlb_s8(int8_t __p0, int8_t __p1) {
int8_t __ret;
__ret = (int8_t) __builtin_neon_vqshlb_s8(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vqshls_s32(int32_t __p0, int32_t __p1) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqshls_s32(__p0, __p1);
- return __ret;
-}
-#else
__ai int32_t vqshls_s32(int32_t __p0, int32_t __p1) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vqshls_s32(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vqshld_s64(int64_t __p0, int64_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vqshld_s64(__p0, __p1);
- return __ret;
-}
-#else
__ai int64_t vqshld_s64(int64_t __p0, int64_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vqshld_s64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vqshlh_s16(__p0, __p1);
- return __ret;
-}
-#else
__ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
int16_t __ret;
__ret = (int16_t) __builtin_neon_vqshlh_s16(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshlb_n_u8(__p0, __p1) __extension__ ({ \
- uint8_t __s0 = __p0; \
- uint8_t __ret; \
- __ret = (uint8_t) __builtin_neon_vqshlb_n_u8(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshlb_n_u8(__p0, __p1) __extension__ ({ \
uint8_t __s0 = __p0; \
uint8_t __ret; \
__ret = (uint8_t) __builtin_neon_vqshlb_n_u8(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshls_n_u32(__p0, __p1) __extension__ ({ \
- uint32_t __s0 = __p0; \
- uint32_t __ret; \
- __ret = (uint32_t) __builtin_neon_vqshls_n_u32(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshls_n_u32(__p0, __p1) __extension__ ({ \
uint32_t __s0 = __p0; \
uint32_t __ret; \
__ret = (uint32_t) __builtin_neon_vqshls_n_u32(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshld_n_u64(__p0, __p1) __extension__ ({ \
- uint64_t __s0 = __p0; \
- uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vqshld_n_u64(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshld_n_u64(__p0, __p1) __extension__ ({ \
uint64_t __s0 = __p0; \
uint64_t __ret; \
__ret = (uint64_t) __builtin_neon_vqshld_n_u64(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshlh_n_u16(__p0, __p1) __extension__ ({ \
- uint16_t __s0 = __p0; \
- uint16_t __ret; \
- __ret = (uint16_t) __builtin_neon_vqshlh_n_u16(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshlh_n_u16(__p0, __p1) __extension__ ({ \
uint16_t __s0 = __p0; \
uint16_t __ret; \
__ret = (uint16_t) __builtin_neon_vqshlh_n_u16(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshlb_n_s8(__p0, __p1) __extension__ ({ \
- int8_t __s0 = __p0; \
- int8_t __ret; \
- __ret = (int8_t) __builtin_neon_vqshlb_n_s8(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshlb_n_s8(__p0, __p1) __extension__ ({ \
int8_t __s0 = __p0; \
int8_t __ret; \
__ret = (int8_t) __builtin_neon_vqshlb_n_s8(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshls_n_s32(__p0, __p1) __extension__ ({ \
- int32_t __s0 = __p0; \
- int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vqshls_n_s32(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshls_n_s32(__p0, __p1) __extension__ ({ \
int32_t __s0 = __p0; \
int32_t __ret; \
__ret = (int32_t) __builtin_neon_vqshls_n_s32(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshld_n_s64(__p0, __p1) __extension__ ({ \
- int64_t __s0 = __p0; \
- int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vqshld_n_s64(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshld_n_s64(__p0, __p1) __extension__ ({ \
int64_t __s0 = __p0; \
int64_t __ret; \
__ret = (int64_t) __builtin_neon_vqshld_n_s64(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshlh_n_s16(__p0, __p1) __extension__ ({ \
- int16_t __s0 = __p0; \
- int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vqshlh_n_s16(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshlh_n_s16(__p0, __p1) __extension__ ({ \
int16_t __s0 = __p0; \
int16_t __ret; \
__ret = (int16_t) __builtin_neon_vqshlh_n_s16(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshlub_n_s8(__p0, __p1) __extension__ ({ \
- int8_t __s0 = __p0; \
- int8_t __ret; \
- __ret = (int8_t) __builtin_neon_vqshlub_n_s8(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshlub_n_s8(__p0, __p1) __extension__ ({ \
int8_t __s0 = __p0; \
int8_t __ret; \
__ret = (int8_t) __builtin_neon_vqshlub_n_s8(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshlus_n_s32(__p0, __p1) __extension__ ({ \
- int32_t __s0 = __p0; \
- int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vqshlus_n_s32(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshlus_n_s32(__p0, __p1) __extension__ ({ \
int32_t __s0 = __p0; \
int32_t __ret; \
__ret = (int32_t) __builtin_neon_vqshlus_n_s32(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshlud_n_s64(__p0, __p1) __extension__ ({ \
- int64_t __s0 = __p0; \
- int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vqshlud_n_s64(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshlud_n_s64(__p0, __p1) __extension__ ({ \
int64_t __s0 = __p0; \
int64_t __ret; \
__ret = (int64_t) __builtin_neon_vqshlud_n_s64(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshluh_n_s16(__p0, __p1) __extension__ ({ \
- int16_t __s0 = __p0; \
- int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vqshluh_n_s16(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshluh_n_s16(__p0, __p1) __extension__ ({ \
int16_t __s0 = __p0; \
int16_t __ret; \
__ret = (int16_t) __builtin_neon_vqshluh_n_s16(__s0, __p1); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vqshrn_high_n_u32(__p0_200, __p1_200, __p2_200) __extension__ ({ \
- uint16x4_t __s0_200 = __p0_200; \
- uint32x4_t __s1_200 = __p1_200; \
- uint16x8_t __ret_200; \
- __ret_200 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_200), (uint16x4_t)(vqshrn_n_u32(__s1_200, __p2_200)))); \
- __ret_200; \
+#define vqshrn_high_n_u32(__p0_681, __p1_681, __p2_681) __extension__ ({ \
+ uint16x4_t __s0_681 = __p0_681; \
+ uint32x4_t __s1_681 = __p1_681; \
+ uint16x8_t __ret_681; \
+ __ret_681 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_681), (uint16x4_t)(vqshrn_n_u32(__s1_681, __p2_681)))); \
+ __ret_681; \
})
#else
-#define vqshrn_high_n_u32(__p0_201, __p1_201, __p2_201) __extension__ ({ \
- uint16x4_t __s0_201 = __p0_201; \
- uint32x4_t __s1_201 = __p1_201; \
- uint16x4_t __rev0_201; __rev0_201 = __builtin_shufflevector(__s0_201, __s0_201, 3, 2, 1, 0); \
- uint32x4_t __rev1_201; __rev1_201 = __builtin_shufflevector(__s1_201, __s1_201, 3, 2, 1, 0); \
- uint16x8_t __ret_201; \
- __ret_201 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_201), (uint16x4_t)(__noswap_vqshrn_n_u32(__rev1_201, __p2_201)))); \
- __ret_201 = __builtin_shufflevector(__ret_201, __ret_201, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_201; \
+#define vqshrn_high_n_u32(__p0_682, __p1_682, __p2_682) __extension__ ({ \
+ uint16x4_t __s0_682 = __p0_682; \
+ uint32x4_t __s1_682 = __p1_682; \
+ uint16x4_t __rev0_682; __rev0_682 = __builtin_shufflevector(__s0_682, __s0_682, 3, 2, 1, 0); \
+ uint32x4_t __rev1_682; __rev1_682 = __builtin_shufflevector(__s1_682, __s1_682, 3, 2, 1, 0); \
+ uint16x8_t __ret_682; \
+ __ret_682 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_682), (uint16x4_t)(__noswap_vqshrn_n_u32(__rev1_682, __p2_682)))); \
+ __ret_682 = __builtin_shufflevector(__ret_682, __ret_682, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_682; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqshrn_high_n_u64(__p0_202, __p1_202, __p2_202) __extension__ ({ \
- uint32x2_t __s0_202 = __p0_202; \
- uint64x2_t __s1_202 = __p1_202; \
- uint32x4_t __ret_202; \
- __ret_202 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_202), (uint32x2_t)(vqshrn_n_u64(__s1_202, __p2_202)))); \
- __ret_202; \
+#define vqshrn_high_n_u64(__p0_683, __p1_683, __p2_683) __extension__ ({ \
+ uint32x2_t __s0_683 = __p0_683; \
+ uint64x2_t __s1_683 = __p1_683; \
+ uint32x4_t __ret_683; \
+ __ret_683 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_683), (uint32x2_t)(vqshrn_n_u64(__s1_683, __p2_683)))); \
+ __ret_683; \
})
#else
-#define vqshrn_high_n_u64(__p0_203, __p1_203, __p2_203) __extension__ ({ \
- uint32x2_t __s0_203 = __p0_203; \
- uint64x2_t __s1_203 = __p1_203; \
- uint32x2_t __rev0_203; __rev0_203 = __builtin_shufflevector(__s0_203, __s0_203, 1, 0); \
- uint64x2_t __rev1_203; __rev1_203 = __builtin_shufflevector(__s1_203, __s1_203, 1, 0); \
- uint32x4_t __ret_203; \
- __ret_203 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_203), (uint32x2_t)(__noswap_vqshrn_n_u64(__rev1_203, __p2_203)))); \
- __ret_203 = __builtin_shufflevector(__ret_203, __ret_203, 3, 2, 1, 0); \
- __ret_203; \
+#define vqshrn_high_n_u64(__p0_684, __p1_684, __p2_684) __extension__ ({ \
+ uint32x2_t __s0_684 = __p0_684; \
+ uint64x2_t __s1_684 = __p1_684; \
+ uint32x2_t __rev0_684; __rev0_684 = __builtin_shufflevector(__s0_684, __s0_684, 1, 0); \
+ uint64x2_t __rev1_684; __rev1_684 = __builtin_shufflevector(__s1_684, __s1_684, 1, 0); \
+ uint32x4_t __ret_684; \
+ __ret_684 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_684), (uint32x2_t)(__noswap_vqshrn_n_u64(__rev1_684, __p2_684)))); \
+ __ret_684 = __builtin_shufflevector(__ret_684, __ret_684, 3, 2, 1, 0); \
+ __ret_684; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqshrn_high_n_u16(__p0_204, __p1_204, __p2_204) __extension__ ({ \
- uint8x8_t __s0_204 = __p0_204; \
- uint16x8_t __s1_204 = __p1_204; \
- uint8x16_t __ret_204; \
- __ret_204 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_204), (uint8x8_t)(vqshrn_n_u16(__s1_204, __p2_204)))); \
- __ret_204; \
+#define vqshrn_high_n_u16(__p0_685, __p1_685, __p2_685) __extension__ ({ \
+ uint8x8_t __s0_685 = __p0_685; \
+ uint16x8_t __s1_685 = __p1_685; \
+ uint8x16_t __ret_685; \
+ __ret_685 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_685), (uint8x8_t)(vqshrn_n_u16(__s1_685, __p2_685)))); \
+ __ret_685; \
})
#else
-#define vqshrn_high_n_u16(__p0_205, __p1_205, __p2_205) __extension__ ({ \
- uint8x8_t __s0_205 = __p0_205; \
- uint16x8_t __s1_205 = __p1_205; \
- uint8x8_t __rev0_205; __rev0_205 = __builtin_shufflevector(__s0_205, __s0_205, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __rev1_205; __rev1_205 = __builtin_shufflevector(__s1_205, __s1_205, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x16_t __ret_205; \
- __ret_205 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_205), (uint8x8_t)(__noswap_vqshrn_n_u16(__rev1_205, __p2_205)))); \
- __ret_205 = __builtin_shufflevector(__ret_205, __ret_205, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_205; \
+#define vqshrn_high_n_u16(__p0_686, __p1_686, __p2_686) __extension__ ({ \
+ uint8x8_t __s0_686 = __p0_686; \
+ uint16x8_t __s1_686 = __p1_686; \
+ uint8x8_t __rev0_686; __rev0_686 = __builtin_shufflevector(__s0_686, __s0_686, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __rev1_686; __rev1_686 = __builtin_shufflevector(__s1_686, __s1_686, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __ret_686; \
+ __ret_686 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_686), (uint8x8_t)(__noswap_vqshrn_n_u16(__rev1_686, __p2_686)))); \
+ __ret_686 = __builtin_shufflevector(__ret_686, __ret_686, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_686; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqshrn_high_n_s32(__p0_206, __p1_206, __p2_206) __extension__ ({ \
- int16x4_t __s0_206 = __p0_206; \
- int32x4_t __s1_206 = __p1_206; \
- int16x8_t __ret_206; \
- __ret_206 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_206), (int16x4_t)(vqshrn_n_s32(__s1_206, __p2_206)))); \
- __ret_206; \
+#define vqshrn_high_n_s32(__p0_687, __p1_687, __p2_687) __extension__ ({ \
+ int16x4_t __s0_687 = __p0_687; \
+ int32x4_t __s1_687 = __p1_687; \
+ int16x8_t __ret_687; \
+ __ret_687 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_687), (int16x4_t)(vqshrn_n_s32(__s1_687, __p2_687)))); \
+ __ret_687; \
})
#else
-#define vqshrn_high_n_s32(__p0_207, __p1_207, __p2_207) __extension__ ({ \
- int16x4_t __s0_207 = __p0_207; \
- int32x4_t __s1_207 = __p1_207; \
- int16x4_t __rev0_207; __rev0_207 = __builtin_shufflevector(__s0_207, __s0_207, 3, 2, 1, 0); \
- int32x4_t __rev1_207; __rev1_207 = __builtin_shufflevector(__s1_207, __s1_207, 3, 2, 1, 0); \
- int16x8_t __ret_207; \
- __ret_207 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_207), (int16x4_t)(__noswap_vqshrn_n_s32(__rev1_207, __p2_207)))); \
- __ret_207 = __builtin_shufflevector(__ret_207, __ret_207, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_207; \
+#define vqshrn_high_n_s32(__p0_688, __p1_688, __p2_688) __extension__ ({ \
+ int16x4_t __s0_688 = __p0_688; \
+ int32x4_t __s1_688 = __p1_688; \
+ int16x4_t __rev0_688; __rev0_688 = __builtin_shufflevector(__s0_688, __s0_688, 3, 2, 1, 0); \
+ int32x4_t __rev1_688; __rev1_688 = __builtin_shufflevector(__s1_688, __s1_688, 3, 2, 1, 0); \
+ int16x8_t __ret_688; \
+ __ret_688 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_688), (int16x4_t)(__noswap_vqshrn_n_s32(__rev1_688, __p2_688)))); \
+ __ret_688 = __builtin_shufflevector(__ret_688, __ret_688, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_688; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqshrn_high_n_s64(__p0_208, __p1_208, __p2_208) __extension__ ({ \
- int32x2_t __s0_208 = __p0_208; \
- int64x2_t __s1_208 = __p1_208; \
- int32x4_t __ret_208; \
- __ret_208 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_208), (int32x2_t)(vqshrn_n_s64(__s1_208, __p2_208)))); \
- __ret_208; \
+#define vqshrn_high_n_s64(__p0_689, __p1_689, __p2_689) __extension__ ({ \
+ int32x2_t __s0_689 = __p0_689; \
+ int64x2_t __s1_689 = __p1_689; \
+ int32x4_t __ret_689; \
+ __ret_689 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_689), (int32x2_t)(vqshrn_n_s64(__s1_689, __p2_689)))); \
+ __ret_689; \
})
#else
-#define vqshrn_high_n_s64(__p0_209, __p1_209, __p2_209) __extension__ ({ \
- int32x2_t __s0_209 = __p0_209; \
- int64x2_t __s1_209 = __p1_209; \
- int32x2_t __rev0_209; __rev0_209 = __builtin_shufflevector(__s0_209, __s0_209, 1, 0); \
- int64x2_t __rev1_209; __rev1_209 = __builtin_shufflevector(__s1_209, __s1_209, 1, 0); \
- int32x4_t __ret_209; \
- __ret_209 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_209), (int32x2_t)(__noswap_vqshrn_n_s64(__rev1_209, __p2_209)))); \
- __ret_209 = __builtin_shufflevector(__ret_209, __ret_209, 3, 2, 1, 0); \
- __ret_209; \
+#define vqshrn_high_n_s64(__p0_690, __p1_690, __p2_690) __extension__ ({ \
+ int32x2_t __s0_690 = __p0_690; \
+ int64x2_t __s1_690 = __p1_690; \
+ int32x2_t __rev0_690; __rev0_690 = __builtin_shufflevector(__s0_690, __s0_690, 1, 0); \
+ int64x2_t __rev1_690; __rev1_690 = __builtin_shufflevector(__s1_690, __s1_690, 1, 0); \
+ int32x4_t __ret_690; \
+ __ret_690 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_690), (int32x2_t)(__noswap_vqshrn_n_s64(__rev1_690, __p2_690)))); \
+ __ret_690 = __builtin_shufflevector(__ret_690, __ret_690, 3, 2, 1, 0); \
+ __ret_690; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqshrn_high_n_s16(__p0_210, __p1_210, __p2_210) __extension__ ({ \
- int8x8_t __s0_210 = __p0_210; \
- int16x8_t __s1_210 = __p1_210; \
- int8x16_t __ret_210; \
- __ret_210 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_210), (int8x8_t)(vqshrn_n_s16(__s1_210, __p2_210)))); \
- __ret_210; \
+#define vqshrn_high_n_s16(__p0_691, __p1_691, __p2_691) __extension__ ({ \
+ int8x8_t __s0_691 = __p0_691; \
+ int16x8_t __s1_691 = __p1_691; \
+ int8x16_t __ret_691; \
+ __ret_691 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_691), (int8x8_t)(vqshrn_n_s16(__s1_691, __p2_691)))); \
+ __ret_691; \
})
#else
-#define vqshrn_high_n_s16(__p0_211, __p1_211, __p2_211) __extension__ ({ \
- int8x8_t __s0_211 = __p0_211; \
- int16x8_t __s1_211 = __p1_211; \
- int8x8_t __rev0_211; __rev0_211 = __builtin_shufflevector(__s0_211, __s0_211, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1_211; __rev1_211 = __builtin_shufflevector(__s1_211, __s1_211, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x16_t __ret_211; \
- __ret_211 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_211), (int8x8_t)(__noswap_vqshrn_n_s16(__rev1_211, __p2_211)))); \
- __ret_211 = __builtin_shufflevector(__ret_211, __ret_211, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_211; \
+#define vqshrn_high_n_s16(__p0_692, __p1_692, __p2_692) __extension__ ({ \
+ int8x8_t __s0_692 = __p0_692; \
+ int16x8_t __s1_692 = __p1_692; \
+ int8x8_t __rev0_692; __rev0_692 = __builtin_shufflevector(__s0_692, __s0_692, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_692; __rev1_692 = __builtin_shufflevector(__s1_692, __s1_692, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __ret_692; \
+ __ret_692 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_692), (int8x8_t)(__noswap_vqshrn_n_s16(__rev1_692, __p2_692)))); \
+ __ret_692 = __builtin_shufflevector(__ret_692, __ret_692, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_692; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vqshrns_n_u32(__p0, __p1) __extension__ ({ \
- uint32_t __s0 = __p0; \
- uint16_t __ret; \
- __ret = (uint16_t) __builtin_neon_vqshrns_n_u32(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshrns_n_u32(__p0, __p1) __extension__ ({ \
uint32_t __s0 = __p0; \
uint16_t __ret; \
__ret = (uint16_t) __builtin_neon_vqshrns_n_u32(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshrnd_n_u64(__p0, __p1) __extension__ ({ \
- uint64_t __s0 = __p0; \
- uint32_t __ret; \
- __ret = (uint32_t) __builtin_neon_vqshrnd_n_u64(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshrnd_n_u64(__p0, __p1) __extension__ ({ \
uint64_t __s0 = __p0; \
uint32_t __ret; \
__ret = (uint32_t) __builtin_neon_vqshrnd_n_u64(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshrnh_n_u16(__p0, __p1) __extension__ ({ \
- uint16_t __s0 = __p0; \
- uint8_t __ret; \
- __ret = (uint8_t) __builtin_neon_vqshrnh_n_u16(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshrnh_n_u16(__p0, __p1) __extension__ ({ \
uint16_t __s0 = __p0; \
uint8_t __ret; \
__ret = (uint8_t) __builtin_neon_vqshrnh_n_u16(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshrns_n_s32(__p0, __p1) __extension__ ({ \
- int32_t __s0 = __p0; \
- int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vqshrns_n_s32(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshrns_n_s32(__p0, __p1) __extension__ ({ \
int32_t __s0 = __p0; \
int16_t __ret; \
__ret = (int16_t) __builtin_neon_vqshrns_n_s32(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshrnd_n_s64(__p0, __p1) __extension__ ({ \
- int64_t __s0 = __p0; \
- int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vqshrnd_n_s64(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshrnd_n_s64(__p0, __p1) __extension__ ({ \
int64_t __s0 = __p0; \
int32_t __ret; \
__ret = (int32_t) __builtin_neon_vqshrnd_n_s64(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshrnh_n_s16(__p0, __p1) __extension__ ({ \
- int16_t __s0 = __p0; \
- int8_t __ret; \
- __ret = (int8_t) __builtin_neon_vqshrnh_n_s16(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshrnh_n_s16(__p0, __p1) __extension__ ({ \
int16_t __s0 = __p0; \
int8_t __ret; \
__ret = (int8_t) __builtin_neon_vqshrnh_n_s16(__s0, __p1); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vqshrun_high_n_s32(__p0_212, __p1_212, __p2_212) __extension__ ({ \
- int16x4_t __s0_212 = __p0_212; \
- int32x4_t __s1_212 = __p1_212; \
- int16x8_t __ret_212; \
- __ret_212 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_212), (int16x4_t)(vqshrun_n_s32(__s1_212, __p2_212)))); \
- __ret_212; \
+#define vqshrun_high_n_s32(__p0_693, __p1_693, __p2_693) __extension__ ({ \
+ int16x4_t __s0_693 = __p0_693; \
+ int32x4_t __s1_693 = __p1_693; \
+ int16x8_t __ret_693; \
+ __ret_693 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_693), (int16x4_t)(vqshrun_n_s32(__s1_693, __p2_693)))); \
+ __ret_693; \
})
#else
-#define vqshrun_high_n_s32(__p0_213, __p1_213, __p2_213) __extension__ ({ \
- int16x4_t __s0_213 = __p0_213; \
- int32x4_t __s1_213 = __p1_213; \
- int16x4_t __rev0_213; __rev0_213 = __builtin_shufflevector(__s0_213, __s0_213, 3, 2, 1, 0); \
- int32x4_t __rev1_213; __rev1_213 = __builtin_shufflevector(__s1_213, __s1_213, 3, 2, 1, 0); \
- int16x8_t __ret_213; \
- __ret_213 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_213), (int16x4_t)(__noswap_vqshrun_n_s32(__rev1_213, __p2_213)))); \
- __ret_213 = __builtin_shufflevector(__ret_213, __ret_213, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_213; \
+#define vqshrun_high_n_s32(__p0_694, __p1_694, __p2_694) __extension__ ({ \
+ int16x4_t __s0_694 = __p0_694; \
+ int32x4_t __s1_694 = __p1_694; \
+ int16x4_t __rev0_694; __rev0_694 = __builtin_shufflevector(__s0_694, __s0_694, 3, 2, 1, 0); \
+ int32x4_t __rev1_694; __rev1_694 = __builtin_shufflevector(__s1_694, __s1_694, 3, 2, 1, 0); \
+ int16x8_t __ret_694; \
+ __ret_694 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_694), (int16x4_t)(__noswap_vqshrun_n_s32(__rev1_694, __p2_694)))); \
+ __ret_694 = __builtin_shufflevector(__ret_694, __ret_694, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_694; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqshrun_high_n_s64(__p0_214, __p1_214, __p2_214) __extension__ ({ \
- int32x2_t __s0_214 = __p0_214; \
- int64x2_t __s1_214 = __p1_214; \
- int32x4_t __ret_214; \
- __ret_214 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_214), (int32x2_t)(vqshrun_n_s64(__s1_214, __p2_214)))); \
- __ret_214; \
+#define vqshrun_high_n_s64(__p0_695, __p1_695, __p2_695) __extension__ ({ \
+ int32x2_t __s0_695 = __p0_695; \
+ int64x2_t __s1_695 = __p1_695; \
+ int32x4_t __ret_695; \
+ __ret_695 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_695), (int32x2_t)(vqshrun_n_s64(__s1_695, __p2_695)))); \
+ __ret_695; \
})
#else
-#define vqshrun_high_n_s64(__p0_215, __p1_215, __p2_215) __extension__ ({ \
- int32x2_t __s0_215 = __p0_215; \
- int64x2_t __s1_215 = __p1_215; \
- int32x2_t __rev0_215; __rev0_215 = __builtin_shufflevector(__s0_215, __s0_215, 1, 0); \
- int64x2_t __rev1_215; __rev1_215 = __builtin_shufflevector(__s1_215, __s1_215, 1, 0); \
- int32x4_t __ret_215; \
- __ret_215 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_215), (int32x2_t)(__noswap_vqshrun_n_s64(__rev1_215, __p2_215)))); \
- __ret_215 = __builtin_shufflevector(__ret_215, __ret_215, 3, 2, 1, 0); \
- __ret_215; \
+#define vqshrun_high_n_s64(__p0_696, __p1_696, __p2_696) __extension__ ({ \
+ int32x2_t __s0_696 = __p0_696; \
+ int64x2_t __s1_696 = __p1_696; \
+ int32x2_t __rev0_696; __rev0_696 = __builtin_shufflevector(__s0_696, __s0_696, 1, 0); \
+ int64x2_t __rev1_696; __rev1_696 = __builtin_shufflevector(__s1_696, __s1_696, 1, 0); \
+ int32x4_t __ret_696; \
+ __ret_696 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_696), (int32x2_t)(__noswap_vqshrun_n_s64(__rev1_696, __p2_696)))); \
+ __ret_696 = __builtin_shufflevector(__ret_696, __ret_696, 3, 2, 1, 0); \
+ __ret_696; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqshrun_high_n_s16(__p0_216, __p1_216, __p2_216) __extension__ ({ \
- int8x8_t __s0_216 = __p0_216; \
- int16x8_t __s1_216 = __p1_216; \
- int8x16_t __ret_216; \
- __ret_216 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_216), (int8x8_t)(vqshrun_n_s16(__s1_216, __p2_216)))); \
- __ret_216; \
+#define vqshrun_high_n_s16(__p0_697, __p1_697, __p2_697) __extension__ ({ \
+ int8x8_t __s0_697 = __p0_697; \
+ int16x8_t __s1_697 = __p1_697; \
+ int8x16_t __ret_697; \
+ __ret_697 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_697), (int8x8_t)(vqshrun_n_s16(__s1_697, __p2_697)))); \
+ __ret_697; \
})
#else
-#define vqshrun_high_n_s16(__p0_217, __p1_217, __p2_217) __extension__ ({ \
- int8x8_t __s0_217 = __p0_217; \
- int16x8_t __s1_217 = __p1_217; \
- int8x8_t __rev0_217; __rev0_217 = __builtin_shufflevector(__s0_217, __s0_217, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1_217; __rev1_217 = __builtin_shufflevector(__s1_217, __s1_217, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x16_t __ret_217; \
- __ret_217 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_217), (int8x8_t)(__noswap_vqshrun_n_s16(__rev1_217, __p2_217)))); \
- __ret_217 = __builtin_shufflevector(__ret_217, __ret_217, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_217; \
+#define vqshrun_high_n_s16(__p0_698, __p1_698, __p2_698) __extension__ ({ \
+ int8x8_t __s0_698 = __p0_698; \
+ int16x8_t __s1_698 = __p1_698; \
+ int8x8_t __rev0_698; __rev0_698 = __builtin_shufflevector(__s0_698, __s0_698, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_698; __rev1_698 = __builtin_shufflevector(__s1_698, __s1_698, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __ret_698; \
+ __ret_698 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_698), (int8x8_t)(__noswap_vqshrun_n_s16(__rev1_698, __p2_698)))); \
+ __ret_698 = __builtin_shufflevector(__ret_698, __ret_698, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_698; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vqshruns_n_s32(__p0, __p1) __extension__ ({ \
- int32_t __s0 = __p0; \
- int16_t __ret; \
- __ret = (int16_t) __builtin_neon_vqshruns_n_s32(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshruns_n_s32(__p0, __p1) __extension__ ({ \
int32_t __s0 = __p0; \
int16_t __ret; \
__ret = (int16_t) __builtin_neon_vqshruns_n_s32(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshrund_n_s64(__p0, __p1) __extension__ ({ \
- int64_t __s0 = __p0; \
- int32_t __ret; \
- __ret = (int32_t) __builtin_neon_vqshrund_n_s64(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshrund_n_s64(__p0, __p1) __extension__ ({ \
int64_t __s0 = __p0; \
int32_t __ret; \
__ret = (int32_t) __builtin_neon_vqshrund_n_s64(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vqshrunh_n_s16(__p0, __p1) __extension__ ({ \
- int16_t __s0 = __p0; \
- int8_t __ret; \
- __ret = (int8_t) __builtin_neon_vqshrunh_n_s16(__s0, __p1); \
- __ret; \
-})
-#else
#define vqshrunh_n_s16(__p0, __p1) __extension__ ({ \
int16_t __s0 = __p0; \
int8_t __ret; \
__ret = (int8_t) __builtin_neon_vqshrunh_n_s16(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint8_t vqsubb_u8(uint8_t __p0, uint8_t __p1) {
- uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vqsubb_u8(__p0, __p1);
- return __ret;
-}
-#else
__ai uint8_t vqsubb_u8(uint8_t __p0, uint8_t __p1) {
uint8_t __ret;
__ret = (uint8_t) __builtin_neon_vqsubb_u8(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vqsubs_u32(uint32_t __p0, uint32_t __p1) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vqsubs_u32(__p0, __p1);
- return __ret;
-}
-#else
__ai uint32_t vqsubs_u32(uint32_t __p0, uint32_t __p1) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vqsubs_u32(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vqsubd_u64(uint64_t __p0, uint64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vqsubd_u64(__p0, __p1);
- return __ret;
-}
-#else
__ai uint64_t vqsubd_u64(uint64_t __p0, uint64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vqsubd_u64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16_t vqsubh_u16(uint16_t __p0, uint16_t __p1) {
- uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vqsubh_u16(__p0, __p1);
- return __ret;
-}
-#else
__ai uint16_t vqsubh_u16(uint16_t __p0, uint16_t __p1) {
uint16_t __ret;
__ret = (uint16_t) __builtin_neon_vqsubh_u16(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8_t vqsubb_s8(int8_t __p0, int8_t __p1) {
- int8_t __ret;
- __ret = (int8_t) __builtin_neon_vqsubb_s8(__p0, __p1);
- return __ret;
-}
-#else
__ai int8_t vqsubb_s8(int8_t __p0, int8_t __p1) {
int8_t __ret;
__ret = (int8_t) __builtin_neon_vqsubb_s8(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vqsubs_s32(int32_t __p0, int32_t __p1) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqsubs_s32(__p0, __p1);
- return __ret;
-}
-#else
__ai int32_t vqsubs_s32(int32_t __p0, int32_t __p1) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vqsubs_s32(__p0, __p1);
return __ret;
}
-__ai int32_t __noswap_vqsubs_s32(int32_t __p0, int32_t __p1) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vqsubs_s32(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vqsubd_s64(int64_t __p0, int64_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vqsubd_s64(__p0, __p1);
- return __ret;
-}
-#else
__ai int64_t vqsubd_s64(int64_t __p0, int64_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vqsubd_s64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16_t vqsubh_s16(int16_t __p0, int16_t __p1) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vqsubh_s16(__p0, __p1);
- return __ret;
-}
-#else
__ai int16_t vqsubh_s16(int16_t __p0, int16_t __p1) {
int16_t __ret;
__ret = (int16_t) __builtin_neon_vqsubh_s16(__p0, __p1);
return __ret;
}
-__ai int16_t __noswap_vqsubh_s16(int16_t __p0, int16_t __p1) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vqsubh_s16(__p0, __p1);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai poly8x8_t vqtbl1_p8(poly8x16_t __p0, uint8x8_t __p1) {
poly8x8_t __ret;
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vqtbl1q_s8(int8x16_t __p0, int8x16_t __p1) {
+__ai int8x16_t vqtbl1q_s8(int8x16_t __p0, uint8x16_t __p1) {
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
return __ret;
}
#else
-__ai int8x16_t vqtbl1q_s8(int8x16_t __p0, int8x16_t __p1) {
+__ai int8x16_t vqtbl1q_s8(int8x16_t __p0, uint8x16_t __p1) {
int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vqtbl1_s8(int8x16_t __p0, int8x8_t __p1) {
+__ai int8x8_t vqtbl1_s8(int8x16_t __p0, uint8x8_t __p1) {
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__p0, (int8x8_t)__p1, 0);
return __ret;
}
#else
-__ai int8x8_t vqtbl1_s8(int8x16_t __p0, int8x8_t __p1) {
+__ai int8x8_t vqtbl1_s8(int8x16_t __p0, uint8x8_t __p1) {
int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__rev0, (int8x8_t)__rev1, 0);
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vqtbl2q_s8(int8x16x2_t __p0, int8x16_t __p1) {
+__ai int8x16_t vqtbl2q_s8(int8x16x2_t __p0, uint8x16_t __p1) {
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p1, 32);
return __ret;
}
#else
-__ai int8x16_t vqtbl2q_s8(int8x16x2_t __p0, int8x16_t __p1) {
+__ai int8x16_t vqtbl2q_s8(int8x16x2_t __p0, uint8x16_t __p1) {
int8x16x2_t __rev0;
__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev1, 32);
__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vqtbl2_s8(int8x16x2_t __p0, int8x8_t __p1) {
+__ai int8x8_t vqtbl2_s8(int8x16x2_t __p0, uint8x8_t __p1) {
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x8_t)__p1, 0);
return __ret;
}
#else
-__ai int8x8_t vqtbl2_s8(int8x16x2_t __p0, int8x8_t __p1) {
+__ai int8x8_t vqtbl2_s8(int8x16x2_t __p0, uint8x8_t __p1) {
int8x16x2_t __rev0;
__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x8_t)__rev1, 0);
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vqtbl3q_s8(int8x16x3_t __p0, int8x16_t __p1) {
+__ai int8x16_t vqtbl3q_s8(int8x16x3_t __p0, uint8x16_t __p1) {
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p1, 32);
return __ret;
}
#else
-__ai int8x16_t vqtbl3q_s8(int8x16x3_t __p0, int8x16_t __p1) {
+__ai int8x16_t vqtbl3q_s8(int8x16x3_t __p0, uint8x16_t __p1) {
int8x16x3_t __rev0;
__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev1, 32);
__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vqtbl3_s8(int8x16x3_t __p0, int8x8_t __p1) {
+__ai int8x8_t vqtbl3_s8(int8x16x3_t __p0, uint8x8_t __p1) {
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x8_t)__p1, 0);
return __ret;
}
#else
-__ai int8x8_t vqtbl3_s8(int8x16x3_t __p0, int8x8_t __p1) {
+__ai int8x8_t vqtbl3_s8(int8x16x3_t __p0, uint8x8_t __p1) {
int8x16x3_t __rev0;
__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x8_t)__rev1, 0);
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vqtbl4q_s8(int8x16x4_t __p0, int8x16_t __p1) {
+__ai int8x16_t vqtbl4q_s8(int8x16x4_t __p0, uint8x16_t __p1) {
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x16_t)__p1, 32);
return __ret;
}
#else
-__ai int8x16_t vqtbl4q_s8(int8x16x4_t __p0, int8x16_t __p1) {
+__ai int8x16_t vqtbl4q_s8(int8x16x4_t __p0, uint8x16_t __p1) {
int8x16x4_t __rev0;
__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev0.val[3], (int8x16_t)__rev1, 32);
__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vqtbl4_s8(int8x16x4_t __p0, int8x8_t __p1) {
+__ai int8x8_t vqtbl4_s8(int8x16x4_t __p0, uint8x8_t __p1) {
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x8_t)__p1, 0);
return __ret;
}
#else
-__ai int8x8_t vqtbl4_s8(int8x16x4_t __p0, int8x8_t __p1) {
+__ai int8x8_t vqtbl4_s8(int8x16x4_t __p0, uint8x8_t __p1) {
int8x16x4_t __rev0;
__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev0.val[3], (int8x8_t)__rev1, 0);
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vqtbx1q_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
+__ai int8x16_t vqtbx1q_s8(int8x16_t __p0, int8x16_t __p1, uint8x16_t __p2) {
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 32);
return __ret;
}
#else
-__ai int8x16_t vqtbx1q_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
+__ai int8x16_t vqtbx1q_s8(int8x16_t __p0, int8x16_t __p1, uint8x16_t __p2) {
int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 32);
__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vqtbx1_s8(int8x8_t __p0, int8x16_t __p1, int8x8_t __p2) {
+__ai int8x8_t vqtbx1_s8(int8x8_t __p0, int8x16_t __p1, uint8x8_t __p2) {
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__p0, (int8x16_t)__p1, (int8x8_t)__p2, 0);
return __ret;
}
#else
-__ai int8x8_t vqtbx1_s8(int8x8_t __p0, int8x16_t __p1, int8x8_t __p2) {
+__ai int8x8_t vqtbx1_s8(int8x8_t __p0, int8x16_t __p1, uint8x8_t __p2) {
int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, 0);
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vqtbx2q_s8(int8x16_t __p0, int8x16x2_t __p1, int8x16_t __p2) {
+__ai int8x16_t vqtbx2q_s8(int8x16_t __p0, int8x16x2_t __p1, uint8x16_t __p2) {
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p2, 32);
return __ret;
}
#else
-__ai int8x16_t vqtbx2q_s8(int8x16_t __p0, int8x16x2_t __p1, int8x16_t __p2) {
+__ai int8x16_t vqtbx2q_s8(int8x16_t __p0, int8x16x2_t __p1, uint8x16_t __p2) {
int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16x2_t __rev1;
__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev2, 32);
__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vqtbx2_s8(int8x8_t __p0, int8x16x2_t __p1, int8x8_t __p2) {
+__ai int8x8_t vqtbx2_s8(int8x8_t __p0, int8x16x2_t __p1, uint8x8_t __p2) {
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x8_t)__p2, 0);
return __ret;
}
#else
-__ai int8x8_t vqtbx2_s8(int8x8_t __p0, int8x16x2_t __p1, int8x8_t __p2) {
+__ai int8x8_t vqtbx2_s8(int8x8_t __p0, int8x16x2_t __p1, uint8x8_t __p2) {
int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16x2_t __rev1;
__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x8_t)__rev2, 0);
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vqtbx3q_s8(int8x16_t __p0, int8x16x3_t __p1, int8x16_t __p2) {
+__ai int8x16_t vqtbx3q_s8(int8x16_t __p0, int8x16x3_t __p1, uint8x16_t __p2) {
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p2, 32);
return __ret;
}
#else
-__ai int8x16_t vqtbx3q_s8(int8x16_t __p0, int8x16x3_t __p1, int8x16_t __p2) {
+__ai int8x16_t vqtbx3q_s8(int8x16_t __p0, int8x16x3_t __p1, uint8x16_t __p2) {
int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16x3_t __rev1;
__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev2, 32);
__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vqtbx3_s8(int8x8_t __p0, int8x16x3_t __p1, int8x8_t __p2) {
+__ai int8x8_t vqtbx3_s8(int8x8_t __p0, int8x16x3_t __p1, uint8x8_t __p2) {
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x8_t)__p2, 0);
return __ret;
}
#else
-__ai int8x8_t vqtbx3_s8(int8x8_t __p0, int8x16x3_t __p1, int8x8_t __p2) {
+__ai int8x8_t vqtbx3_s8(int8x8_t __p0, int8x16x3_t __p1, uint8x8_t __p2) {
int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16x3_t __rev1;
__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x8_t)__rev2, 0);
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vqtbx4q_s8(int8x16_t __p0, int8x16x4_t __p1, int8x16_t __p2) {
+__ai int8x16_t vqtbx4q_s8(int8x16_t __p0, int8x16x4_t __p1, uint8x16_t __p2) {
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x16_t)__p2, 32);
return __ret;
}
#else
-__ai int8x16_t vqtbx4q_s8(int8x16_t __p0, int8x16x4_t __p1, int8x16_t __p2) {
+__ai int8x16_t vqtbx4q_s8(int8x16_t __p0, int8x16x4_t __p1, uint8x16_t __p2) {
int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16x4_t __rev1;
__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], (int8x16_t)__rev2, 32);
__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vqtbx4_s8(int8x8_t __p0, int8x16x4_t __p1, int8x8_t __p2) {
+__ai int8x8_t vqtbx4_s8(int8x8_t __p0, int8x16x4_t __p1, uint8x8_t __p2) {
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x8_t)__p2, 0);
return __ret;
}
#else
-__ai int8x8_t vqtbx4_s8(int8x8_t __p0, int8x16x4_t __p1, int8x8_t __p2) {
+__ai int8x8_t vqtbx4_s8(int8x8_t __p0, int8x16x4_t __p1, uint8x8_t __p2) {
int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16x4_t __rev1;
__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
__rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], (int8x8_t)__rev2, 0);
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai float64x1_t vrecpe_f64(float64x1_t __p0) {
float64x1_t __ret;
__ret = (float64x1_t) __builtin_neon_vrecpe_v((int8x8_t)__p0, 10);
return __ret;
}
-#else
-__ai float64x1_t vrecpe_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrecpe_v((int8x8_t)__p0, 10);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64_t vrecped_f64(float64_t __p0) {
- float64_t __ret;
- __ret = (float64_t) __builtin_neon_vrecped_f64(__p0);
- return __ret;
-}
-#else
__ai float64_t vrecped_f64(float64_t __p0) {
float64_t __ret;
__ret = (float64_t) __builtin_neon_vrecped_f64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32_t vrecpes_f32(float32_t __p0) {
- float32_t __ret;
- __ret = (float32_t) __builtin_neon_vrecpes_f32(__p0);
- return __ret;
-}
-#else
__ai float32_t vrecpes_f32(float32_t __p0) {
float32_t __ret;
__ret = (float32_t) __builtin_neon_vrecpes_f32(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float64x2_t vrecpsq_f64(float64x2_t __p0, float64x2_t __p1) {
float64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vrecps_f64(float64x1_t __p0, float64x1_t __p1) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrecps_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
- return __ret;
-}
-#else
__ai float64x1_t vrecps_f64(float64x1_t __p0, float64x1_t __p1) {
float64x1_t __ret;
__ret = (float64x1_t) __builtin_neon_vrecps_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai float64_t vrecpsd_f64(float64_t __p0, float64_t __p1) {
float64_t __ret;
__ret = (float64_t) __builtin_neon_vrecpsd_f64(__p0, __p1);
return __ret;
}
-#else
-__ai float64_t vrecpsd_f64(float64_t __p0, float64_t __p1) {
- float64_t __ret;
- __ret = (float64_t) __builtin_neon_vrecpsd_f64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32_t vrecpss_f32(float32_t __p0, float32_t __p1) {
- float32_t __ret;
- __ret = (float32_t) __builtin_neon_vrecpss_f32(__p0, __p1);
- return __ret;
-}
-#else
__ai float32_t vrecpss_f32(float32_t __p0, float32_t __p1) {
float32_t __ret;
__ret = (float32_t) __builtin_neon_vrecpss_f32(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float64_t vrecpxd_f64(float64_t __p0) {
- float64_t __ret;
- __ret = (float64_t) __builtin_neon_vrecpxd_f64(__p0);
- return __ret;
-}
-#else
__ai float64_t vrecpxd_f64(float64_t __p0) {
float64_t __ret;
__ret = (float64_t) __builtin_neon_vrecpxd_f64(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32_t vrecpxs_f32(float32_t __p0) {
- float32_t __ret;
- __ret = (float32_t) __builtin_neon_vrecpxs_f32(__p0);
- return __ret;
-}
-#else
__ai float32_t vrecpxs_f32(float32_t __p0) {
float32_t __ret;
__ret = (float32_t) __builtin_neon_vrecpxs_f32(__p0);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vrshld_u64(uint64_t __p0, uint64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vrshld_u64(__p0, __p1);
- return __ret;
-}
-#else
__ai uint64_t vrshld_u64(uint64_t __p0, uint64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vrshld_u64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vrshld_s64(int64_t __p0, int64_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vrshld_s64(__p0, __p1);
- return __ret;
-}
-#else
__ai int64_t vrshld_s64(int64_t __p0, int64_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vrshld_s64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vrshrd_n_u64(__p0, __p1) __extension__ ({ \
uint64_t __s0 = __p0; \
uint64_t __ret; \
__ret = (uint64_t) __builtin_neon_vrshrd_n_u64(__s0, __p1); \
__ret; \
})
-#else
-#define vrshrd_n_u64(__p0, __p1) __extension__ ({ \
- uint64_t __s0 = __p0; \
- uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vrshrd_n_u64(__s0, __p1); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vrshrd_n_s64(__p0, __p1) __extension__ ({ \
- int64_t __s0 = __p0; \
- int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vrshrd_n_s64(__s0, __p1); \
- __ret; \
-})
-#else
#define vrshrd_n_s64(__p0, __p1) __extension__ ({ \
int64_t __s0 = __p0; \
int64_t __ret; \
__ret = (int64_t) __builtin_neon_vrshrd_n_s64(__s0, __p1); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vrshrn_high_n_u32(__p0_218, __p1_218, __p2_218) __extension__ ({ \
- uint16x4_t __s0_218 = __p0_218; \
- uint32x4_t __s1_218 = __p1_218; \
- uint16x8_t __ret_218; \
- __ret_218 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_218), (uint16x4_t)(vrshrn_n_u32(__s1_218, __p2_218)))); \
- __ret_218; \
+#define vrshrn_high_n_u32(__p0_699, __p1_699, __p2_699) __extension__ ({ \
+ uint16x4_t __s0_699 = __p0_699; \
+ uint32x4_t __s1_699 = __p1_699; \
+ uint16x8_t __ret_699; \
+ __ret_699 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_699), (uint16x4_t)(vrshrn_n_u32(__s1_699, __p2_699)))); \
+ __ret_699; \
})
#else
-#define vrshrn_high_n_u32(__p0_219, __p1_219, __p2_219) __extension__ ({ \
- uint16x4_t __s0_219 = __p0_219; \
- uint32x4_t __s1_219 = __p1_219; \
- uint16x4_t __rev0_219; __rev0_219 = __builtin_shufflevector(__s0_219, __s0_219, 3, 2, 1, 0); \
- uint32x4_t __rev1_219; __rev1_219 = __builtin_shufflevector(__s1_219, __s1_219, 3, 2, 1, 0); \
- uint16x8_t __ret_219; \
- __ret_219 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_219), (uint16x4_t)(__noswap_vrshrn_n_u32(__rev1_219, __p2_219)))); \
- __ret_219 = __builtin_shufflevector(__ret_219, __ret_219, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_219; \
+#define vrshrn_high_n_u32(__p0_700, __p1_700, __p2_700) __extension__ ({ \
+ uint16x4_t __s0_700 = __p0_700; \
+ uint32x4_t __s1_700 = __p1_700; \
+ uint16x4_t __rev0_700; __rev0_700 = __builtin_shufflevector(__s0_700, __s0_700, 3, 2, 1, 0); \
+ uint32x4_t __rev1_700; __rev1_700 = __builtin_shufflevector(__s1_700, __s1_700, 3, 2, 1, 0); \
+ uint16x8_t __ret_700; \
+ __ret_700 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_700), (uint16x4_t)(__noswap_vrshrn_n_u32(__rev1_700, __p2_700)))); \
+ __ret_700 = __builtin_shufflevector(__ret_700, __ret_700, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_700; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vrshrn_high_n_u64(__p0_220, __p1_220, __p2_220) __extension__ ({ \
- uint32x2_t __s0_220 = __p0_220; \
- uint64x2_t __s1_220 = __p1_220; \
- uint32x4_t __ret_220; \
- __ret_220 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_220), (uint32x2_t)(vrshrn_n_u64(__s1_220, __p2_220)))); \
- __ret_220; \
+#define vrshrn_high_n_u64(__p0_701, __p1_701, __p2_701) __extension__ ({ \
+ uint32x2_t __s0_701 = __p0_701; \
+ uint64x2_t __s1_701 = __p1_701; \
+ uint32x4_t __ret_701; \
+ __ret_701 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_701), (uint32x2_t)(vrshrn_n_u64(__s1_701, __p2_701)))); \
+ __ret_701; \
})
#else
-#define vrshrn_high_n_u64(__p0_221, __p1_221, __p2_221) __extension__ ({ \
- uint32x2_t __s0_221 = __p0_221; \
- uint64x2_t __s1_221 = __p1_221; \
- uint32x2_t __rev0_221; __rev0_221 = __builtin_shufflevector(__s0_221, __s0_221, 1, 0); \
- uint64x2_t __rev1_221; __rev1_221 = __builtin_shufflevector(__s1_221, __s1_221, 1, 0); \
- uint32x4_t __ret_221; \
- __ret_221 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_221), (uint32x2_t)(__noswap_vrshrn_n_u64(__rev1_221, __p2_221)))); \
- __ret_221 = __builtin_shufflevector(__ret_221, __ret_221, 3, 2, 1, 0); \
- __ret_221; \
+#define vrshrn_high_n_u64(__p0_702, __p1_702, __p2_702) __extension__ ({ \
+ uint32x2_t __s0_702 = __p0_702; \
+ uint64x2_t __s1_702 = __p1_702; \
+ uint32x2_t __rev0_702; __rev0_702 = __builtin_shufflevector(__s0_702, __s0_702, 1, 0); \
+ uint64x2_t __rev1_702; __rev1_702 = __builtin_shufflevector(__s1_702, __s1_702, 1, 0); \
+ uint32x4_t __ret_702; \
+ __ret_702 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_702), (uint32x2_t)(__noswap_vrshrn_n_u64(__rev1_702, __p2_702)))); \
+ __ret_702 = __builtin_shufflevector(__ret_702, __ret_702, 3, 2, 1, 0); \
+ __ret_702; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vrshrn_high_n_u16(__p0_222, __p1_222, __p2_222) __extension__ ({ \
- uint8x8_t __s0_222 = __p0_222; \
- uint16x8_t __s1_222 = __p1_222; \
- uint8x16_t __ret_222; \
- __ret_222 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_222), (uint8x8_t)(vrshrn_n_u16(__s1_222, __p2_222)))); \
- __ret_222; \
+#define vrshrn_high_n_u16(__p0_703, __p1_703, __p2_703) __extension__ ({ \
+ uint8x8_t __s0_703 = __p0_703; \
+ uint16x8_t __s1_703 = __p1_703; \
+ uint8x16_t __ret_703; \
+ __ret_703 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_703), (uint8x8_t)(vrshrn_n_u16(__s1_703, __p2_703)))); \
+ __ret_703; \
})
#else
-#define vrshrn_high_n_u16(__p0_223, __p1_223, __p2_223) __extension__ ({ \
- uint8x8_t __s0_223 = __p0_223; \
- uint16x8_t __s1_223 = __p1_223; \
- uint8x8_t __rev0_223; __rev0_223 = __builtin_shufflevector(__s0_223, __s0_223, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __rev1_223; __rev1_223 = __builtin_shufflevector(__s1_223, __s1_223, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x16_t __ret_223; \
- __ret_223 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_223), (uint8x8_t)(__noswap_vrshrn_n_u16(__rev1_223, __p2_223)))); \
- __ret_223 = __builtin_shufflevector(__ret_223, __ret_223, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_223; \
+#define vrshrn_high_n_u16(__p0_704, __p1_704, __p2_704) __extension__ ({ \
+ uint8x8_t __s0_704 = __p0_704; \
+ uint16x8_t __s1_704 = __p1_704; \
+ uint8x8_t __rev0_704; __rev0_704 = __builtin_shufflevector(__s0_704, __s0_704, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __rev1_704; __rev1_704 = __builtin_shufflevector(__s1_704, __s1_704, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __ret_704; \
+ __ret_704 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_704), (uint8x8_t)(__noswap_vrshrn_n_u16(__rev1_704, __p2_704)))); \
+ __ret_704 = __builtin_shufflevector(__ret_704, __ret_704, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_704; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vrshrn_high_n_s32(__p0_224, __p1_224, __p2_224) __extension__ ({ \
- int16x4_t __s0_224 = __p0_224; \
- int32x4_t __s1_224 = __p1_224; \
- int16x8_t __ret_224; \
- __ret_224 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_224), (int16x4_t)(vrshrn_n_s32(__s1_224, __p2_224)))); \
- __ret_224; \
+#define vrshrn_high_n_s32(__p0_705, __p1_705, __p2_705) __extension__ ({ \
+ int16x4_t __s0_705 = __p0_705; \
+ int32x4_t __s1_705 = __p1_705; \
+ int16x8_t __ret_705; \
+ __ret_705 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_705), (int16x4_t)(vrshrn_n_s32(__s1_705, __p2_705)))); \
+ __ret_705; \
})
#else
-#define vrshrn_high_n_s32(__p0_225, __p1_225, __p2_225) __extension__ ({ \
- int16x4_t __s0_225 = __p0_225; \
- int32x4_t __s1_225 = __p1_225; \
- int16x4_t __rev0_225; __rev0_225 = __builtin_shufflevector(__s0_225, __s0_225, 3, 2, 1, 0); \
- int32x4_t __rev1_225; __rev1_225 = __builtin_shufflevector(__s1_225, __s1_225, 3, 2, 1, 0); \
- int16x8_t __ret_225; \
- __ret_225 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_225), (int16x4_t)(__noswap_vrshrn_n_s32(__rev1_225, __p2_225)))); \
- __ret_225 = __builtin_shufflevector(__ret_225, __ret_225, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_225; \
+#define vrshrn_high_n_s32(__p0_706, __p1_706, __p2_706) __extension__ ({ \
+ int16x4_t __s0_706 = __p0_706; \
+ int32x4_t __s1_706 = __p1_706; \
+ int16x4_t __rev0_706; __rev0_706 = __builtin_shufflevector(__s0_706, __s0_706, 3, 2, 1, 0); \
+ int32x4_t __rev1_706; __rev1_706 = __builtin_shufflevector(__s1_706, __s1_706, 3, 2, 1, 0); \
+ int16x8_t __ret_706; \
+ __ret_706 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_706), (int16x4_t)(__noswap_vrshrn_n_s32(__rev1_706, __p2_706)))); \
+ __ret_706 = __builtin_shufflevector(__ret_706, __ret_706, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_706; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vrshrn_high_n_s64(__p0_226, __p1_226, __p2_226) __extension__ ({ \
- int32x2_t __s0_226 = __p0_226; \
- int64x2_t __s1_226 = __p1_226; \
- int32x4_t __ret_226; \
- __ret_226 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_226), (int32x2_t)(vrshrn_n_s64(__s1_226, __p2_226)))); \
- __ret_226; \
+#define vrshrn_high_n_s64(__p0_707, __p1_707, __p2_707) __extension__ ({ \
+ int32x2_t __s0_707 = __p0_707; \
+ int64x2_t __s1_707 = __p1_707; \
+ int32x4_t __ret_707; \
+ __ret_707 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_707), (int32x2_t)(vrshrn_n_s64(__s1_707, __p2_707)))); \
+ __ret_707; \
})
#else
-#define vrshrn_high_n_s64(__p0_227, __p1_227, __p2_227) __extension__ ({ \
- int32x2_t __s0_227 = __p0_227; \
- int64x2_t __s1_227 = __p1_227; \
- int32x2_t __rev0_227; __rev0_227 = __builtin_shufflevector(__s0_227, __s0_227, 1, 0); \
- int64x2_t __rev1_227; __rev1_227 = __builtin_shufflevector(__s1_227, __s1_227, 1, 0); \
- int32x4_t __ret_227; \
- __ret_227 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_227), (int32x2_t)(__noswap_vrshrn_n_s64(__rev1_227, __p2_227)))); \
- __ret_227 = __builtin_shufflevector(__ret_227, __ret_227, 3, 2, 1, 0); \
- __ret_227; \
+#define vrshrn_high_n_s64(__p0_708, __p1_708, __p2_708) __extension__ ({ \
+ int32x2_t __s0_708 = __p0_708; \
+ int64x2_t __s1_708 = __p1_708; \
+ int32x2_t __rev0_708; __rev0_708 = __builtin_shufflevector(__s0_708, __s0_708, 1, 0); \
+ int64x2_t __rev1_708; __rev1_708 = __builtin_shufflevector(__s1_708, __s1_708, 1, 0); \
+ int32x4_t __ret_708; \
+ __ret_708 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_708), (int32x2_t)(__noswap_vrshrn_n_s64(__rev1_708, __p2_708)))); \
+ __ret_708 = __builtin_shufflevector(__ret_708, __ret_708, 3, 2, 1, 0); \
+ __ret_708; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vrshrn_high_n_s16(__p0_228, __p1_228, __p2_228) __extension__ ({ \
- int8x8_t __s0_228 = __p0_228; \
- int16x8_t __s1_228 = __p1_228; \
- int8x16_t __ret_228; \
- __ret_228 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_228), (int8x8_t)(vrshrn_n_s16(__s1_228, __p2_228)))); \
- __ret_228; \
+#define vrshrn_high_n_s16(__p0_709, __p1_709, __p2_709) __extension__ ({ \
+ int8x8_t __s0_709 = __p0_709; \
+ int16x8_t __s1_709 = __p1_709; \
+ int8x16_t __ret_709; \
+ __ret_709 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_709), (int8x8_t)(vrshrn_n_s16(__s1_709, __p2_709)))); \
+ __ret_709; \
})
#else
-#define vrshrn_high_n_s16(__p0_229, __p1_229, __p2_229) __extension__ ({ \
- int8x8_t __s0_229 = __p0_229; \
- int16x8_t __s1_229 = __p1_229; \
- int8x8_t __rev0_229; __rev0_229 = __builtin_shufflevector(__s0_229, __s0_229, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1_229; __rev1_229 = __builtin_shufflevector(__s1_229, __s1_229, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x16_t __ret_229; \
- __ret_229 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_229), (int8x8_t)(__noswap_vrshrn_n_s16(__rev1_229, __p2_229)))); \
- __ret_229 = __builtin_shufflevector(__ret_229, __ret_229, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_229; \
+#define vrshrn_high_n_s16(__p0_710, __p1_710, __p2_710) __extension__ ({ \
+ int8x8_t __s0_710 = __p0_710; \
+ int16x8_t __s1_710 = __p1_710; \
+ int8x8_t __rev0_710; __rev0_710 = __builtin_shufflevector(__s0_710, __s0_710, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_710; __rev1_710 = __builtin_shufflevector(__s1_710, __s1_710, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __ret_710; \
+ __ret_710 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_710), (int8x8_t)(__noswap_vrshrn_n_s16(__rev1_710, __p2_710)))); \
+ __ret_710 = __builtin_shufflevector(__ret_710, __ret_710, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_710; \
})
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai float64x1_t vrsqrte_f64(float64x1_t __p0) {
float64x1_t __ret;
__ret = (float64x1_t) __builtin_neon_vrsqrte_v((int8x8_t)__p0, 10);
return __ret;
}
-#else
-__ai float64x1_t vrsqrte_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrsqrte_v((int8x8_t)__p0, 10);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai float64_t vrsqrted_f64(float64_t __p0) {
float64_t __ret;
__ret = (float64_t) __builtin_neon_vrsqrted_f64(__p0);
return __ret;
}
-#else
-__ai float64_t vrsqrted_f64(float64_t __p0) {
- float64_t __ret;
- __ret = (float64_t) __builtin_neon_vrsqrted_f64(__p0);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32_t vrsqrtes_f32(float32_t __p0) {
- float32_t __ret;
- __ret = (float32_t) __builtin_neon_vrsqrtes_f32(__p0);
- return __ret;
-}
-#else
__ai float32_t vrsqrtes_f32(float32_t __p0) {
float32_t __ret;
__ret = (float32_t) __builtin_neon_vrsqrtes_f32(__p0);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float64x2_t vrsqrtsq_f64(float64x2_t __p0, float64x2_t __p1) {
float64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai float64x1_t vrsqrts_f64(float64x1_t __p0, float64x1_t __p1) {
float64x1_t __ret;
__ret = (float64x1_t) __builtin_neon_vrsqrts_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
return __ret;
}
-#else
-__ai float64x1_t vrsqrts_f64(float64x1_t __p0, float64x1_t __p1) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vrsqrts_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai float64_t vrsqrtsd_f64(float64_t __p0, float64_t __p1) {
float64_t __ret;
__ret = (float64_t) __builtin_neon_vrsqrtsd_f64(__p0, __p1);
return __ret;
}
-#else
-__ai float64_t vrsqrtsd_f64(float64_t __p0, float64_t __p1) {
- float64_t __ret;
- __ret = (float64_t) __builtin_neon_vrsqrtsd_f64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai float32_t vrsqrtss_f32(float32_t __p0, float32_t __p1) {
- float32_t __ret;
- __ret = (float32_t) __builtin_neon_vrsqrtss_f32(__p0, __p1);
- return __ret;
-}
-#else
__ai float32_t vrsqrtss_f32(float32_t __p0, float32_t __p1) {
float32_t __ret;
__ret = (float32_t) __builtin_neon_vrsqrtss_f32(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vrsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64_t __s0 = __p0; \
- uint64_t __s1 = __p1; \
- uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vrsrad_n_u64(__s0, __s1, __p2); \
- __ret; \
-})
-#else
#define vrsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
uint64_t __s0 = __p0; \
uint64_t __s1 = __p1; \
__ret = (uint64_t) __builtin_neon_vrsrad_n_u64(__s0, __s1, __p2); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vrsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
- int64_t __s0 = __p0; \
- int64_t __s1 = __p1; \
- int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vrsrad_n_s64(__s0, __s1, __p2); \
- __ret; \
-})
-#else
#define vrsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
int64_t __s0 = __p0; \
int64_t __s1 = __p1; \
__ret = (int64_t) __builtin_neon_vrsrad_n_s64(__s0, __s1, __p2); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x8_t vrsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
uint16x8_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vset_lane_p64(__p0, __p1, __p2) __extension__ ({ \
- poly64_t __s0 = __p0; \
- poly64x1_t __s1 = __p1; \
- poly64x1_t __ret; \
- __ret = (poly64x1_t) __builtin_neon_vset_lane_i64(__s0, (int8x8_t)__s1, __p2); \
- __ret; \
-})
-#else
#define vset_lane_p64(__p0, __p1, __p2) __extension__ ({ \
poly64_t __s0 = __p0; \
poly64x1_t __s1 = __p1; \
poly64x1_t __ret; \
- __ret = (poly64x1_t) __builtin_neon_vset_lane_i64(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (poly64x1_t) __builtin_neon_vset_lane_i64(__s0, (poly64x1_t)__s1, __p2); \
__ret; \
})
-#define __noswap_vset_lane_p64(__p0, __p1, __p2) __extension__ ({ \
- poly64_t __s0 = __p0; \
- poly64x1_t __s1 = __p1; \
- poly64x1_t __ret; \
- __ret = (poly64x1_t) __builtin_neon_vset_lane_i64(__s0, (int8x8_t)__s1, __p2); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
poly64_t __s0 = __p0; \
poly64x2_t __s1 = __p1; \
poly64x2_t __ret; \
- __ret = (poly64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (poly64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (poly64x2_t)__s1, __p2); \
__ret; \
})
#else
poly64x2_t __s1 = __p1; \
poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
poly64x2_t __ret; \
- __ret = (poly64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int8x16_t)__rev1, __p2); \
+ __ret = (poly64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (poly64x2_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
__ret; \
})
poly64_t __s0 = __p0; \
poly64x2_t __s1 = __p1; \
poly64x2_t __ret; \
- __ret = (poly64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (poly64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (poly64x2_t)__s1, __p2); \
__ret; \
})
#endif
float64_t __s0 = __p0; \
float64x2_t __s1 = __p1; \
float64x2_t __ret; \
- __ret = (float64x2_t) __builtin_neon_vsetq_lane_f64(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (float64x2_t) __builtin_neon_vsetq_lane_f64(__s0, (float64x2_t)__s1, __p2); \
__ret; \
})
#else
float64x2_t __s1 = __p1; \
float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
float64x2_t __ret; \
- __ret = (float64x2_t) __builtin_neon_vsetq_lane_f64(__s0, (int8x16_t)__rev1, __p2); \
+ __ret = (float64x2_t) __builtin_neon_vsetq_lane_f64(__s0, (float64x2_t)__rev1, __p2); \
__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
__ret; \
})
float64_t __s0 = __p0; \
float64x2_t __s1 = __p1; \
float64x2_t __ret; \
- __ret = (float64x2_t) __builtin_neon_vsetq_lane_f64(__s0, (int8x16_t)__s1, __p2); \
+ __ret = (float64x2_t) __builtin_neon_vsetq_lane_f64(__s0, (float64x2_t)__s1, __p2); \
__ret; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vset_lane_f64(__p0, __p1, __p2) __extension__ ({ \
- float64_t __s0 = __p0; \
- float64x1_t __s1 = __p1; \
- float64x1_t __ret; \
- __ret = (float64x1_t) __builtin_neon_vset_lane_f64(__s0, (int8x8_t)__s1, __p2); \
- __ret; \
-})
-#else
#define vset_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64_t __s0 = __p0; \
float64x1_t __s1 = __p1; \
float64x1_t __ret; \
- __ret = (float64x1_t) __builtin_neon_vset_lane_f64(__s0, (int8x8_t)__s1, __p2); \
- __ret; \
-})
-#define __noswap_vset_lane_f64(__p0, __p1, __p2) __extension__ ({ \
- float64_t __s0 = __p0; \
- float64x1_t __s1 = __p1; \
- float64x1_t __ret; \
- __ret = (float64x1_t) __builtin_neon_vset_lane_f64(__s0, (int8x8_t)__s1, __p2); \
+ __ret = (float64x1_t) __builtin_neon_vset_lane_f64(__s0, (float64x1_t)__s1, __p2); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint64_t vshld_u64(uint64_t __p0, uint64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vshld_u64(__p0, __p1);
return __ret;
}
-#else
-__ai uint64_t vshld_u64(uint64_t __p0, uint64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vshld_u64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vshld_s64(__p0, __p1);
- return __ret;
-}
-#else
__ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vshld_s64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vshld_n_u64(__p0, __p1) __extension__ ({ \
- uint64_t __s0 = __p0; \
- uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vshld_n_u64(__s0, __p1); \
- __ret; \
-})
-#else
#define vshld_n_u64(__p0, __p1) __extension__ ({ \
uint64_t __s0 = __p0; \
uint64_t __ret; \
__ret = (uint64_t) __builtin_neon_vshld_n_u64(__s0, __p1); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vshld_n_s64(__p0, __p1) __extension__ ({ \
int64_t __s0 = __p0; \
int64_t __ret; \
__ret = (int64_t) __builtin_neon_vshld_n_s64(__s0, __p1); \
__ret; \
})
-#else
-#define vshld_n_s64(__p0, __p1) __extension__ ({ \
- int64_t __s0 = __p0; \
- int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vshld_n_s64(__s0, __p1); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vshll_high_n_u8(__p0_230, __p1_230) __extension__ ({ \
- uint8x16_t __s0_230 = __p0_230; \
- uint16x8_t __ret_230; \
- __ret_230 = (uint16x8_t)(vshll_n_u8(vget_high_u8(__s0_230), __p1_230)); \
- __ret_230; \
+#define vshll_high_n_u8(__p0_711, __p1_711) __extension__ ({ \
+ uint8x16_t __s0_711 = __p0_711; \
+ uint16x8_t __ret_711; \
+ __ret_711 = (uint16x8_t)(vshll_n_u8(vget_high_u8(__s0_711), __p1_711)); \
+ __ret_711; \
})
#else
-#define vshll_high_n_u8(__p0_231, __p1_231) __extension__ ({ \
- uint8x16_t __s0_231 = __p0_231; \
- uint8x16_t __rev0_231; __rev0_231 = __builtin_shufflevector(__s0_231, __s0_231, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __ret_231; \
- __ret_231 = (uint16x8_t)(__noswap_vshll_n_u8(__noswap_vget_high_u8(__rev0_231), __p1_231)); \
- __ret_231 = __builtin_shufflevector(__ret_231, __ret_231, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_231; \
+#define vshll_high_n_u8(__p0_712, __p1_712) __extension__ ({ \
+ uint8x16_t __s0_712 = __p0_712; \
+ uint8x16_t __rev0_712; __rev0_712 = __builtin_shufflevector(__s0_712, __s0_712, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __ret_712; \
+ __ret_712 = (uint16x8_t)(__noswap_vshll_n_u8(__noswap_vget_high_u8(__rev0_712), __p1_712)); \
+ __ret_712 = __builtin_shufflevector(__ret_712, __ret_712, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_712; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vshll_high_n_u32(__p0_232, __p1_232) __extension__ ({ \
- uint32x4_t __s0_232 = __p0_232; \
- uint64x2_t __ret_232; \
- __ret_232 = (uint64x2_t)(vshll_n_u32(vget_high_u32(__s0_232), __p1_232)); \
- __ret_232; \
+#define vshll_high_n_u32(__p0_713, __p1_713) __extension__ ({ \
+ uint32x4_t __s0_713 = __p0_713; \
+ uint64x2_t __ret_713; \
+ __ret_713 = (uint64x2_t)(vshll_n_u32(vget_high_u32(__s0_713), __p1_713)); \
+ __ret_713; \
})
#else
-#define vshll_high_n_u32(__p0_233, __p1_233) __extension__ ({ \
- uint32x4_t __s0_233 = __p0_233; \
- uint32x4_t __rev0_233; __rev0_233 = __builtin_shufflevector(__s0_233, __s0_233, 3, 2, 1, 0); \
- uint64x2_t __ret_233; \
- __ret_233 = (uint64x2_t)(__noswap_vshll_n_u32(__noswap_vget_high_u32(__rev0_233), __p1_233)); \
- __ret_233 = __builtin_shufflevector(__ret_233, __ret_233, 1, 0); \
- __ret_233; \
+#define vshll_high_n_u32(__p0_714, __p1_714) __extension__ ({ \
+ uint32x4_t __s0_714 = __p0_714; \
+ uint32x4_t __rev0_714; __rev0_714 = __builtin_shufflevector(__s0_714, __s0_714, 3, 2, 1, 0); \
+ uint64x2_t __ret_714; \
+ __ret_714 = (uint64x2_t)(__noswap_vshll_n_u32(__noswap_vget_high_u32(__rev0_714), __p1_714)); \
+ __ret_714 = __builtin_shufflevector(__ret_714, __ret_714, 1, 0); \
+ __ret_714; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vshll_high_n_u16(__p0_234, __p1_234) __extension__ ({ \
- uint16x8_t __s0_234 = __p0_234; \
- uint32x4_t __ret_234; \
- __ret_234 = (uint32x4_t)(vshll_n_u16(vget_high_u16(__s0_234), __p1_234)); \
- __ret_234; \
+#define vshll_high_n_u16(__p0_715, __p1_715) __extension__ ({ \
+ uint16x8_t __s0_715 = __p0_715; \
+ uint32x4_t __ret_715; \
+ __ret_715 = (uint32x4_t)(vshll_n_u16(vget_high_u16(__s0_715), __p1_715)); \
+ __ret_715; \
})
#else
-#define vshll_high_n_u16(__p0_235, __p1_235) __extension__ ({ \
- uint16x8_t __s0_235 = __p0_235; \
- uint16x8_t __rev0_235; __rev0_235 = __builtin_shufflevector(__s0_235, __s0_235, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint32x4_t __ret_235; \
- __ret_235 = (uint32x4_t)(__noswap_vshll_n_u16(__noswap_vget_high_u16(__rev0_235), __p1_235)); \
- __ret_235 = __builtin_shufflevector(__ret_235, __ret_235, 3, 2, 1, 0); \
- __ret_235; \
+#define vshll_high_n_u16(__p0_716, __p1_716) __extension__ ({ \
+ uint16x8_t __s0_716 = __p0_716; \
+ uint16x8_t __rev0_716; __rev0_716 = __builtin_shufflevector(__s0_716, __s0_716, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint32x4_t __ret_716; \
+ __ret_716 = (uint32x4_t)(__noswap_vshll_n_u16(__noswap_vget_high_u16(__rev0_716), __p1_716)); \
+ __ret_716 = __builtin_shufflevector(__ret_716, __ret_716, 3, 2, 1, 0); \
+ __ret_716; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vshll_high_n_s8(__p0_236, __p1_236) __extension__ ({ \
- int8x16_t __s0_236 = __p0_236; \
- int16x8_t __ret_236; \
- __ret_236 = (int16x8_t)(vshll_n_s8(vget_high_s8(__s0_236), __p1_236)); \
- __ret_236; \
+#define vshll_high_n_s8(__p0_717, __p1_717) __extension__ ({ \
+ int8x16_t __s0_717 = __p0_717; \
+ int16x8_t __ret_717; \
+ __ret_717 = (int16x8_t)(vshll_n_s8(vget_high_s8(__s0_717), __p1_717)); \
+ __ret_717; \
})
#else
-#define vshll_high_n_s8(__p0_237, __p1_237) __extension__ ({ \
- int8x16_t __s0_237 = __p0_237; \
- int8x16_t __rev0_237; __rev0_237 = __builtin_shufflevector(__s0_237, __s0_237, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __ret_237; \
- __ret_237 = (int16x8_t)(__noswap_vshll_n_s8(__noswap_vget_high_s8(__rev0_237), __p1_237)); \
- __ret_237 = __builtin_shufflevector(__ret_237, __ret_237, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_237; \
+#define vshll_high_n_s8(__p0_718, __p1_718) __extension__ ({ \
+ int8x16_t __s0_718 = __p0_718; \
+ int8x16_t __rev0_718; __rev0_718 = __builtin_shufflevector(__s0_718, __s0_718, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __ret_718; \
+ __ret_718 = (int16x8_t)(__noswap_vshll_n_s8(__noswap_vget_high_s8(__rev0_718), __p1_718)); \
+ __ret_718 = __builtin_shufflevector(__ret_718, __ret_718, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_718; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vshll_high_n_s32(__p0_238, __p1_238) __extension__ ({ \
- int32x4_t __s0_238 = __p0_238; \
- int64x2_t __ret_238; \
- __ret_238 = (int64x2_t)(vshll_n_s32(vget_high_s32(__s0_238), __p1_238)); \
- __ret_238; \
+#define vshll_high_n_s32(__p0_719, __p1_719) __extension__ ({ \
+ int32x4_t __s0_719 = __p0_719; \
+ int64x2_t __ret_719; \
+ __ret_719 = (int64x2_t)(vshll_n_s32(vget_high_s32(__s0_719), __p1_719)); \
+ __ret_719; \
})
#else
-#define vshll_high_n_s32(__p0_239, __p1_239) __extension__ ({ \
- int32x4_t __s0_239 = __p0_239; \
- int32x4_t __rev0_239; __rev0_239 = __builtin_shufflevector(__s0_239, __s0_239, 3, 2, 1, 0); \
- int64x2_t __ret_239; \
- __ret_239 = (int64x2_t)(__noswap_vshll_n_s32(__noswap_vget_high_s32(__rev0_239), __p1_239)); \
- __ret_239 = __builtin_shufflevector(__ret_239, __ret_239, 1, 0); \
- __ret_239; \
+#define vshll_high_n_s32(__p0_720, __p1_720) __extension__ ({ \
+ int32x4_t __s0_720 = __p0_720; \
+ int32x4_t __rev0_720; __rev0_720 = __builtin_shufflevector(__s0_720, __s0_720, 3, 2, 1, 0); \
+ int64x2_t __ret_720; \
+ __ret_720 = (int64x2_t)(__noswap_vshll_n_s32(__noswap_vget_high_s32(__rev0_720), __p1_720)); \
+ __ret_720 = __builtin_shufflevector(__ret_720, __ret_720, 1, 0); \
+ __ret_720; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vshll_high_n_s16(__p0_240, __p1_240) __extension__ ({ \
- int16x8_t __s0_240 = __p0_240; \
- int32x4_t __ret_240; \
- __ret_240 = (int32x4_t)(vshll_n_s16(vget_high_s16(__s0_240), __p1_240)); \
- __ret_240; \
+#define vshll_high_n_s16(__p0_721, __p1_721) __extension__ ({ \
+ int16x8_t __s0_721 = __p0_721; \
+ int32x4_t __ret_721; \
+ __ret_721 = (int32x4_t)(vshll_n_s16(vget_high_s16(__s0_721), __p1_721)); \
+ __ret_721; \
})
#else
-#define vshll_high_n_s16(__p0_241, __p1_241) __extension__ ({ \
- int16x8_t __s0_241 = __p0_241; \
- int16x8_t __rev0_241; __rev0_241 = __builtin_shufflevector(__s0_241, __s0_241, 7, 6, 5, 4, 3, 2, 1, 0); \
- int32x4_t __ret_241; \
- __ret_241 = (int32x4_t)(__noswap_vshll_n_s16(__noswap_vget_high_s16(__rev0_241), __p1_241)); \
- __ret_241 = __builtin_shufflevector(__ret_241, __ret_241, 3, 2, 1, 0); \
- __ret_241; \
+#define vshll_high_n_s16(__p0_722, __p1_722) __extension__ ({ \
+ int16x8_t __s0_722 = __p0_722; \
+ int16x8_t __rev0_722; __rev0_722 = __builtin_shufflevector(__s0_722, __s0_722, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_722; \
+ __ret_722 = (int32x4_t)(__noswap_vshll_n_s16(__noswap_vget_high_s16(__rev0_722), __p1_722)); \
+ __ret_722 = __builtin_shufflevector(__ret_722, __ret_722, 3, 2, 1, 0); \
+ __ret_722; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vshrd_n_u64(__p0, __p1) __extension__ ({ \
uint64_t __s0 = __p0; \
uint64_t __ret; \
__ret = (uint64_t) __builtin_neon_vshrd_n_u64(__s0, __p1); \
__ret; \
})
-#else
-#define vshrd_n_u64(__p0, __p1) __extension__ ({ \
- uint64_t __s0 = __p0; \
- uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vshrd_n_u64(__s0, __p1); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vshrd_n_s64(__p0, __p1) __extension__ ({ \
- int64_t __s0 = __p0; \
- int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vshrd_n_s64(__s0, __p1); \
- __ret; \
-})
-#else
#define vshrd_n_s64(__p0, __p1) __extension__ ({ \
int64_t __s0 = __p0; \
int64_t __ret; \
__ret = (int64_t) __builtin_neon_vshrd_n_s64(__s0, __p1); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vshrn_high_n_u32(__p0_242, __p1_242, __p2_242) __extension__ ({ \
- uint16x4_t __s0_242 = __p0_242; \
- uint32x4_t __s1_242 = __p1_242; \
- uint16x8_t __ret_242; \
- __ret_242 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_242), (uint16x4_t)(vshrn_n_u32(__s1_242, __p2_242)))); \
- __ret_242; \
+#define vshrn_high_n_u32(__p0_723, __p1_723, __p2_723) __extension__ ({ \
+ uint16x4_t __s0_723 = __p0_723; \
+ uint32x4_t __s1_723 = __p1_723; \
+ uint16x8_t __ret_723; \
+ __ret_723 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_723), (uint16x4_t)(vshrn_n_u32(__s1_723, __p2_723)))); \
+ __ret_723; \
})
#else
-#define vshrn_high_n_u32(__p0_243, __p1_243, __p2_243) __extension__ ({ \
- uint16x4_t __s0_243 = __p0_243; \
- uint32x4_t __s1_243 = __p1_243; \
- uint16x4_t __rev0_243; __rev0_243 = __builtin_shufflevector(__s0_243, __s0_243, 3, 2, 1, 0); \
- uint32x4_t __rev1_243; __rev1_243 = __builtin_shufflevector(__s1_243, __s1_243, 3, 2, 1, 0); \
- uint16x8_t __ret_243; \
- __ret_243 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_243), (uint16x4_t)(__noswap_vshrn_n_u32(__rev1_243, __p2_243)))); \
- __ret_243 = __builtin_shufflevector(__ret_243, __ret_243, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_243; \
+#define vshrn_high_n_u32(__p0_724, __p1_724, __p2_724) __extension__ ({ \
+ uint16x4_t __s0_724 = __p0_724; \
+ uint32x4_t __s1_724 = __p1_724; \
+ uint16x4_t __rev0_724; __rev0_724 = __builtin_shufflevector(__s0_724, __s0_724, 3, 2, 1, 0); \
+ uint32x4_t __rev1_724; __rev1_724 = __builtin_shufflevector(__s1_724, __s1_724, 3, 2, 1, 0); \
+ uint16x8_t __ret_724; \
+ __ret_724 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_724), (uint16x4_t)(__noswap_vshrn_n_u32(__rev1_724, __p2_724)))); \
+ __ret_724 = __builtin_shufflevector(__ret_724, __ret_724, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_724; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vshrn_high_n_u64(__p0_244, __p1_244, __p2_244) __extension__ ({ \
- uint32x2_t __s0_244 = __p0_244; \
- uint64x2_t __s1_244 = __p1_244; \
- uint32x4_t __ret_244; \
- __ret_244 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_244), (uint32x2_t)(vshrn_n_u64(__s1_244, __p2_244)))); \
- __ret_244; \
+#define vshrn_high_n_u64(__p0_725, __p1_725, __p2_725) __extension__ ({ \
+ uint32x2_t __s0_725 = __p0_725; \
+ uint64x2_t __s1_725 = __p1_725; \
+ uint32x4_t __ret_725; \
+ __ret_725 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_725), (uint32x2_t)(vshrn_n_u64(__s1_725, __p2_725)))); \
+ __ret_725; \
})
#else
-#define vshrn_high_n_u64(__p0_245, __p1_245, __p2_245) __extension__ ({ \
- uint32x2_t __s0_245 = __p0_245; \
- uint64x2_t __s1_245 = __p1_245; \
- uint32x2_t __rev0_245; __rev0_245 = __builtin_shufflevector(__s0_245, __s0_245, 1, 0); \
- uint64x2_t __rev1_245; __rev1_245 = __builtin_shufflevector(__s1_245, __s1_245, 1, 0); \
- uint32x4_t __ret_245; \
- __ret_245 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_245), (uint32x2_t)(__noswap_vshrn_n_u64(__rev1_245, __p2_245)))); \
- __ret_245 = __builtin_shufflevector(__ret_245, __ret_245, 3, 2, 1, 0); \
- __ret_245; \
+#define vshrn_high_n_u64(__p0_726, __p1_726, __p2_726) __extension__ ({ \
+ uint32x2_t __s0_726 = __p0_726; \
+ uint64x2_t __s1_726 = __p1_726; \
+ uint32x2_t __rev0_726; __rev0_726 = __builtin_shufflevector(__s0_726, __s0_726, 1, 0); \
+ uint64x2_t __rev1_726; __rev1_726 = __builtin_shufflevector(__s1_726, __s1_726, 1, 0); \
+ uint32x4_t __ret_726; \
+ __ret_726 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_726), (uint32x2_t)(__noswap_vshrn_n_u64(__rev1_726, __p2_726)))); \
+ __ret_726 = __builtin_shufflevector(__ret_726, __ret_726, 3, 2, 1, 0); \
+ __ret_726; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vshrn_high_n_u16(__p0_246, __p1_246, __p2_246) __extension__ ({ \
- uint8x8_t __s0_246 = __p0_246; \
- uint16x8_t __s1_246 = __p1_246; \
- uint8x16_t __ret_246; \
- __ret_246 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_246), (uint8x8_t)(vshrn_n_u16(__s1_246, __p2_246)))); \
- __ret_246; \
+#define vshrn_high_n_u16(__p0_727, __p1_727, __p2_727) __extension__ ({ \
+ uint8x8_t __s0_727 = __p0_727; \
+ uint16x8_t __s1_727 = __p1_727; \
+ uint8x16_t __ret_727; \
+ __ret_727 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_727), (uint8x8_t)(vshrn_n_u16(__s1_727, __p2_727)))); \
+ __ret_727; \
})
#else
-#define vshrn_high_n_u16(__p0_247, __p1_247, __p2_247) __extension__ ({ \
- uint8x8_t __s0_247 = __p0_247; \
- uint16x8_t __s1_247 = __p1_247; \
- uint8x8_t __rev0_247; __rev0_247 = __builtin_shufflevector(__s0_247, __s0_247, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint16x8_t __rev1_247; __rev1_247 = __builtin_shufflevector(__s1_247, __s1_247, 7, 6, 5, 4, 3, 2, 1, 0); \
- uint8x16_t __ret_247; \
- __ret_247 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_247), (uint8x8_t)(__noswap_vshrn_n_u16(__rev1_247, __p2_247)))); \
- __ret_247 = __builtin_shufflevector(__ret_247, __ret_247, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_247; \
+#define vshrn_high_n_u16(__p0_728, __p1_728, __p2_728) __extension__ ({ \
+ uint8x8_t __s0_728 = __p0_728; \
+ uint16x8_t __s1_728 = __p1_728; \
+ uint8x8_t __rev0_728; __rev0_728 = __builtin_shufflevector(__s0_728, __s0_728, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint16x8_t __rev1_728; __rev1_728 = __builtin_shufflevector(__s1_728, __s1_728, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __ret_728; \
+ __ret_728 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_728), (uint8x8_t)(__noswap_vshrn_n_u16(__rev1_728, __p2_728)))); \
+ __ret_728 = __builtin_shufflevector(__ret_728, __ret_728, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_728; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vshrn_high_n_s32(__p0_248, __p1_248, __p2_248) __extension__ ({ \
- int16x4_t __s0_248 = __p0_248; \
- int32x4_t __s1_248 = __p1_248; \
- int16x8_t __ret_248; \
- __ret_248 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_248), (int16x4_t)(vshrn_n_s32(__s1_248, __p2_248)))); \
- __ret_248; \
+#define vshrn_high_n_s32(__p0_729, __p1_729, __p2_729) __extension__ ({ \
+ int16x4_t __s0_729 = __p0_729; \
+ int32x4_t __s1_729 = __p1_729; \
+ int16x8_t __ret_729; \
+ __ret_729 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_729), (int16x4_t)(vshrn_n_s32(__s1_729, __p2_729)))); \
+ __ret_729; \
})
#else
-#define vshrn_high_n_s32(__p0_249, __p1_249, __p2_249) __extension__ ({ \
- int16x4_t __s0_249 = __p0_249; \
- int32x4_t __s1_249 = __p1_249; \
- int16x4_t __rev0_249; __rev0_249 = __builtin_shufflevector(__s0_249, __s0_249, 3, 2, 1, 0); \
- int32x4_t __rev1_249; __rev1_249 = __builtin_shufflevector(__s1_249, __s1_249, 3, 2, 1, 0); \
- int16x8_t __ret_249; \
- __ret_249 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_249), (int16x4_t)(__noswap_vshrn_n_s32(__rev1_249, __p2_249)))); \
- __ret_249 = __builtin_shufflevector(__ret_249, __ret_249, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_249; \
+#define vshrn_high_n_s32(__p0_730, __p1_730, __p2_730) __extension__ ({ \
+ int16x4_t __s0_730 = __p0_730; \
+ int32x4_t __s1_730 = __p1_730; \
+ int16x4_t __rev0_730; __rev0_730 = __builtin_shufflevector(__s0_730, __s0_730, 3, 2, 1, 0); \
+ int32x4_t __rev1_730; __rev1_730 = __builtin_shufflevector(__s1_730, __s1_730, 3, 2, 1, 0); \
+ int16x8_t __ret_730; \
+ __ret_730 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_730), (int16x4_t)(__noswap_vshrn_n_s32(__rev1_730, __p2_730)))); \
+ __ret_730 = __builtin_shufflevector(__ret_730, __ret_730, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_730; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vshrn_high_n_s64(__p0_250, __p1_250, __p2_250) __extension__ ({ \
- int32x2_t __s0_250 = __p0_250; \
- int64x2_t __s1_250 = __p1_250; \
- int32x4_t __ret_250; \
- __ret_250 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_250), (int32x2_t)(vshrn_n_s64(__s1_250, __p2_250)))); \
- __ret_250; \
+#define vshrn_high_n_s64(__p0_731, __p1_731, __p2_731) __extension__ ({ \
+ int32x2_t __s0_731 = __p0_731; \
+ int64x2_t __s1_731 = __p1_731; \
+ int32x4_t __ret_731; \
+ __ret_731 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_731), (int32x2_t)(vshrn_n_s64(__s1_731, __p2_731)))); \
+ __ret_731; \
})
#else
-#define vshrn_high_n_s64(__p0_251, __p1_251, __p2_251) __extension__ ({ \
- int32x2_t __s0_251 = __p0_251; \
- int64x2_t __s1_251 = __p1_251; \
- int32x2_t __rev0_251; __rev0_251 = __builtin_shufflevector(__s0_251, __s0_251, 1, 0); \
- int64x2_t __rev1_251; __rev1_251 = __builtin_shufflevector(__s1_251, __s1_251, 1, 0); \
- int32x4_t __ret_251; \
- __ret_251 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_251), (int32x2_t)(__noswap_vshrn_n_s64(__rev1_251, __p2_251)))); \
- __ret_251 = __builtin_shufflevector(__ret_251, __ret_251, 3, 2, 1, 0); \
- __ret_251; \
+#define vshrn_high_n_s64(__p0_732, __p1_732, __p2_732) __extension__ ({ \
+ int32x2_t __s0_732 = __p0_732; \
+ int64x2_t __s1_732 = __p1_732; \
+ int32x2_t __rev0_732; __rev0_732 = __builtin_shufflevector(__s0_732, __s0_732, 1, 0); \
+ int64x2_t __rev1_732; __rev1_732 = __builtin_shufflevector(__s1_732, __s1_732, 1, 0); \
+ int32x4_t __ret_732; \
+ __ret_732 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_732), (int32x2_t)(__noswap_vshrn_n_s64(__rev1_732, __p2_732)))); \
+ __ret_732 = __builtin_shufflevector(__ret_732, __ret_732, 3, 2, 1, 0); \
+ __ret_732; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vshrn_high_n_s16(__p0_252, __p1_252, __p2_252) __extension__ ({ \
- int8x8_t __s0_252 = __p0_252; \
- int16x8_t __s1_252 = __p1_252; \
- int8x16_t __ret_252; \
- __ret_252 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_252), (int8x8_t)(vshrn_n_s16(__s1_252, __p2_252)))); \
- __ret_252; \
+#define vshrn_high_n_s16(__p0_733, __p1_733, __p2_733) __extension__ ({ \
+ int8x8_t __s0_733 = __p0_733; \
+ int16x8_t __s1_733 = __p1_733; \
+ int8x16_t __ret_733; \
+ __ret_733 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_733), (int8x8_t)(vshrn_n_s16(__s1_733, __p2_733)))); \
+ __ret_733; \
})
#else
-#define vshrn_high_n_s16(__p0_253, __p1_253, __p2_253) __extension__ ({ \
- int8x8_t __s0_253 = __p0_253; \
- int16x8_t __s1_253 = __p1_253; \
- int8x8_t __rev0_253; __rev0_253 = __builtin_shufflevector(__s0_253, __s0_253, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16x8_t __rev1_253; __rev1_253 = __builtin_shufflevector(__s1_253, __s1_253, 7, 6, 5, 4, 3, 2, 1, 0); \
- int8x16_t __ret_253; \
- __ret_253 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_253), (int8x8_t)(__noswap_vshrn_n_s16(__rev1_253, __p2_253)))); \
- __ret_253 = __builtin_shufflevector(__ret_253, __ret_253, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_253; \
+#define vshrn_high_n_s16(__p0_734, __p1_734, __p2_734) __extension__ ({ \
+ int8x8_t __s0_734 = __p0_734; \
+ int16x8_t __s1_734 = __p1_734; \
+ int8x8_t __rev0_734; __rev0_734 = __builtin_shufflevector(__s0_734, __s0_734, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16x8_t __rev1_734; __rev1_734 = __builtin_shufflevector(__s1_734, __s1_734, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __ret_734; \
+ __ret_734 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_734), (int8x8_t)(__noswap_vshrn_n_s16(__rev1_734, __p2_734)))); \
+ __ret_734 = __builtin_shufflevector(__ret_734, __ret_734, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_734; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vslid_n_u64(__p0, __p1, __p2) __extension__ ({ \
uint64_t __s0 = __p0; \
uint64_t __s1 = __p1; \
__ret = (uint64_t) __builtin_neon_vslid_n_u64(__s0, __s1, __p2); \
__ret; \
})
-#else
-#define vslid_n_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64_t __s0 = __p0; \
- uint64_t __s1 = __p1; \
- uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vslid_n_u64(__s0, __s1, __p2); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vslid_n_s64(__p0, __p1, __p2) __extension__ ({ \
int64_t __s0 = __p0; \
int64_t __s1 = __p1; \
__ret = (int64_t) __builtin_neon_vslid_n_s64(__s0, __s1, __p2); \
__ret; \
})
-#else
-#define vslid_n_s64(__p0, __p1, __p2) __extension__ ({ \
- int64_t __s0 = __p0; \
- int64_t __s1 = __p1; \
- int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vslid_n_s64(__s0, __s1, __p2); \
- __ret; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vsli_n_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x1_t __s0 = __p0; \
poly64x1_t __s1 = __p1; \
__ret = (poly64x1_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 6); \
__ret; \
})
-#else
-#define vsli_n_p64(__p0, __p1, __p2) __extension__ ({ \
- poly64x1_t __s0 = __p0; \
- poly64x1_t __s1 = __p1; \
- poly64x1_t __ret; \
- __ret = (poly64x1_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 6); \
- __ret; \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vsliq_n_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x2_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint8_t vsqaddb_u8(uint8_t __p0, uint8_t __p1) {
+__ai uint8_t vsqaddb_u8(uint8_t __p0, int8_t __p1) {
uint8_t __ret;
__ret = (uint8_t) __builtin_neon_vsqaddb_u8(__p0, __p1);
return __ret;
}
-#else
-__ai uint8_t vsqaddb_u8(uint8_t __p0, uint8_t __p1) {
- uint8_t __ret;
- __ret = (uint8_t) __builtin_neon_vsqaddb_u8(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint32_t vsqadds_u32(uint32_t __p0, uint32_t __p1) {
+__ai uint32_t vsqadds_u32(uint32_t __p0, int32_t __p1) {
uint32_t __ret;
__ret = (uint32_t) __builtin_neon_vsqadds_u32(__p0, __p1);
return __ret;
}
-#else
-__ai uint32_t vsqadds_u32(uint32_t __p0, uint32_t __p1) {
- uint32_t __ret;
- __ret = (uint32_t) __builtin_neon_vsqadds_u32(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vsqaddd_u64(uint64_t __p0, uint64_t __p1) {
+__ai uint64_t vsqaddd_u64(uint64_t __p0, int64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vsqaddd_u64(__p0, __p1);
return __ret;
}
-#else
-__ai uint64_t vsqaddd_u64(uint64_t __p0, uint64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vsqaddd_u64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint16_t vsqaddh_u16(uint16_t __p0, uint16_t __p1) {
+__ai uint16_t vsqaddh_u16(uint16_t __p0, int16_t __p1) {
uint16_t __ret;
__ret = (uint16_t) __builtin_neon_vsqaddh_u16(__p0, __p1);
return __ret;
}
-#else
-__ai uint16_t vsqaddh_u16(uint16_t __p0, uint16_t __p1) {
- uint16_t __ret;
- __ret = (uint16_t) __builtin_neon_vsqaddh_u16(__p0, __p1);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
-__ai uint8x16_t vsqaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
+__ai uint8x16_t vsqaddq_u8(uint8x16_t __p0, int8x16_t __p1) {
uint8x16_t __ret;
__ret = (uint8x16_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
return __ret;
}
#else
-__ai uint8x16_t vsqaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
+__ai uint8x16_t vsqaddq_u8(uint8x16_t __p0, int8x16_t __p1) {
uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
uint8x16_t __ret;
__ret = (uint8x16_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai uint32x4_t vsqaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
+__ai uint32x4_t vsqaddq_u32(uint32x4_t __p0, int32x4_t __p1) {
uint32x4_t __ret;
__ret = (uint32x4_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
return __ret;
}
#else
-__ai uint32x4_t vsqaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
+__ai uint32x4_t vsqaddq_u32(uint32x4_t __p0, int32x4_t __p1) {
uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
uint32x4_t __ret;
__ret = (uint32x4_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai uint64x2_t vsqaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
+__ai uint64x2_t vsqaddq_u64(uint64x2_t __p0, int64x2_t __p1) {
uint64x2_t __ret;
__ret = (uint64x2_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
return __ret;
}
#else
-__ai uint64x2_t vsqaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
+__ai uint64x2_t vsqaddq_u64(uint64x2_t __p0, int64x2_t __p1) {
uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
- uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
+ int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
uint64x2_t __ret;
__ret = (uint64x2_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai uint16x8_t vsqaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
+__ai uint16x8_t vsqaddq_u16(uint16x8_t __p0, int16x8_t __p1) {
uint16x8_t __ret;
__ret = (uint16x8_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
return __ret;
}
#else
-__ai uint16x8_t vsqaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
+__ai uint16x8_t vsqaddq_u16(uint16x8_t __p0, int16x8_t __p1) {
uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
uint16x8_t __ret;
__ret = (uint16x8_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai uint8x8_t vsqadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
+__ai uint8x8_t vsqadd_u8(uint8x8_t __p0, int8x8_t __p1) {
uint8x8_t __ret;
__ret = (uint8x8_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
return __ret;
}
#else
-__ai uint8x8_t vsqadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
+__ai uint8x8_t vsqadd_u8(uint8x8_t __p0, int8x8_t __p1) {
uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
uint8x8_t __ret;
__ret = (uint8x8_t) __builtin_neon_vsqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai uint32x2_t vsqadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
+__ai uint32x2_t vsqadd_u32(uint32x2_t __p0, int32x2_t __p1) {
uint32x2_t __ret;
__ret = (uint32x2_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
return __ret;
}
#else
-__ai uint32x2_t vsqadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
+__ai uint32x2_t vsqadd_u32(uint32x2_t __p0, int32x2_t __p1) {
uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
- uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
+ int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
uint32x2_t __ret;
__ret = (uint32x2_t) __builtin_neon_vsqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai uint64x1_t vsqadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
- return __ret;
-}
-#else
-__ai uint64x1_t vsqadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
+__ai uint64x1_t vsqadd_u64(uint64x1_t __p0, int64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
-__ai uint16x4_t vsqadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
+__ai uint16x4_t vsqadd_u16(uint16x4_t __p0, int16x4_t __p1) {
uint16x4_t __ret;
__ret = (uint16x4_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
return __ret;
}
#else
-__ai uint16x4_t vsqadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
+__ai uint16x4_t vsqadd_u16(uint16x4_t __p0, int16x4_t __p1) {
uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
uint16x4_t __ret;
__ret = (uint16x4_t) __builtin_neon_vsqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai float64x1_t vsqrt_f64(float64x1_t __p0) {
- float64x1_t __ret;
- __ret = (float64x1_t) __builtin_neon_vsqrt_v((int8x8_t)__p0, 10);
- return __ret;
-}
-#else
__ai float64x1_t vsqrt_f64(float64x1_t __p0) {
float64x1_t __ret;
__ret = (float64x1_t) __builtin_neon_vsqrt_v((int8x8_t)__p0, 10);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float32x2_t vsqrt_f32(float32x2_t __p0) {
float32x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64_t __s0 = __p0; \
- uint64_t __s1 = __p1; \
- uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vsrad_n_u64(__s0, __s1, __p2); \
- __ret; \
-})
-#else
#define vsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
uint64_t __s0 = __p0; \
uint64_t __s1 = __p1; \
__ret = (uint64_t) __builtin_neon_vsrad_n_u64(__s0, __s1, __p2); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
- int64_t __s0 = __p0; \
- int64_t __s1 = __p1; \
- int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vsrad_n_s64(__s0, __s1, __p2); \
- __ret; \
-})
-#else
#define vsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
int64_t __s0 = __p0; \
int64_t __s1 = __p1; \
__ret = (int64_t) __builtin_neon_vsrad_n_s64(__s0, __s1, __p2); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vsrid_n_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64_t __s0 = __p0; \
- uint64_t __s1 = __p1; \
- uint64_t __ret; \
- __ret = (uint64_t) __builtin_neon_vsrid_n_u64(__s0, __s1, __p2); \
- __ret; \
-})
-#else
#define vsrid_n_u64(__p0, __p1, __p2) __extension__ ({ \
uint64_t __s0 = __p0; \
uint64_t __s1 = __p1; \
__ret = (uint64_t) __builtin_neon_vsrid_n_u64(__s0, __s1, __p2); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vsrid_n_s64(__p0, __p1, __p2) __extension__ ({ \
- int64_t __s0 = __p0; \
- int64_t __s1 = __p1; \
- int64_t __ret; \
- __ret = (int64_t) __builtin_neon_vsrid_n_s64(__s0, __s1, __p2); \
- __ret; \
-})
-#else
#define vsrid_n_s64(__p0, __p1, __p2) __extension__ ({ \
int64_t __s0 = __p0; \
int64_t __s1 = __p1; \
__ret = (int64_t) __builtin_neon_vsrid_n_s64(__s0, __s1, __p2); \
__ret; \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vsri_n_p64(__p0, __p1, __p2) __extension__ ({ \
- poly64x1_t __s0 = __p0; \
- poly64x1_t __s1 = __p1; \
- poly64x1_t __ret; \
- __ret = (poly64x1_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 6); \
- __ret; \
-})
-#else
#define vsri_n_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x1_t __s0 = __p0; \
poly64x1_t __s1 = __p1; \
__ret = (poly64x1_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 6); \
__ret; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vsriq_n_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x2_t __s0 = __p0; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst1_p64(__p0, __p1) __extension__ ({ \
- poly64x1_t __s1 = __p1; \
- __builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 6); \
-})
-#else
#define vst1_p64(__p0, __p1) __extension__ ({ \
poly64x1_t __s1 = __p1; \
__builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 6); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst1q_p64(__p0, __p1) __extension__ ({ \
poly64x2_t __s1 = __p1; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst1_f64(__p0, __p1) __extension__ ({ \
- float64x1_t __s1 = __p1; \
- __builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 10); \
-})
-#else
#define vst1_f64(__p0, __p1) __extension__ ({ \
float64x1_t __s1 = __p1; \
__builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 10); \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vst1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x1_t __s1 = __p1; \
__builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 6); \
})
-#else
-#define vst1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
- poly64x1_t __s1 = __p1; \
- __builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 6); \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x2_t __s1 = __p1; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x1_t __s1 = __p1; \
- __builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 10); \
-})
-#else
#define vst1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x1_t __s1 = __p1; \
__builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 10); \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst1_p64_x2(__p0, __p1) __extension__ ({ \
- poly64x1x2_t __s1 = __p1; \
- __builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 6); \
-})
-#else
#define vst1_p64_x2(__p0, __p1) __extension__ ({ \
poly64x1x2_t __s1 = __p1; \
__builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 6); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst1q_p64_x2(__p0, __p1) __extension__ ({ \
poly64x2x2_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst1q_f64_x2(__p0, __p1) __extension__ ({ \
float64x2x2_t __s1 = __p1; \
- __builtin_neon_vst1q_x2_v(__p0, __s1.val[0], __s1.val[1], 42); \
+ __builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 42); \
})
#else
#define vst1q_f64_x2(__p0, __p1) __extension__ ({ \
float64x2x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
- __builtin_neon_vst1q_x2_v(__p0, __rev1.val[0], __rev1.val[1], 42); \
+ __builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 42); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst1_f64_x2(__p0, __p1) __extension__ ({ \
- float64x1x2_t __s1 = __p1; \
- __builtin_neon_vst1_x2_v(__p0, __s1.val[0], __s1.val[1], 10); \
-})
-#else
#define vst1_f64_x2(__p0, __p1) __extension__ ({ \
float64x1x2_t __s1 = __p1; \
- __builtin_neon_vst1_x2_v(__p0, __s1.val[0], __s1.val[1], 10); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst1_p64_x3(__p0, __p1) __extension__ ({ \
- poly64x1x3_t __s1 = __p1; \
- __builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 6); \
+ __builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 10); \
})
-#else
#define vst1_p64_x3(__p0, __p1) __extension__ ({ \
poly64x1x3_t __s1 = __p1; \
__builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 6); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst1q_p64_x3(__p0, __p1) __extension__ ({ \
poly64x2x3_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst1q_f64_x3(__p0, __p1) __extension__ ({ \
float64x2x3_t __s1 = __p1; \
- __builtin_neon_vst1q_x3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 42); \
+ __builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 42); \
})
#else
#define vst1q_f64_x3(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
- __builtin_neon_vst1q_x3_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 42); \
+ __builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 42); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vst1_f64_x3(__p0, __p1) __extension__ ({ \
float64x1x3_t __s1 = __p1; \
- __builtin_neon_vst1_x3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 10); \
+ __builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 10); \
})
-#else
-#define vst1_f64_x3(__p0, __p1) __extension__ ({ \
- float64x1x3_t __s1 = __p1; \
- __builtin_neon_vst1_x3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 10); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vst1_p64_x4(__p0, __p1) __extension__ ({ \
poly64x1x4_t __s1 = __p1; \
__builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 6); \
})
-#else
-#define vst1_p64_x4(__p0, __p1) __extension__ ({ \
- poly64x1x4_t __s1 = __p1; \
- __builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 6); \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst1q_p64_x4(__p0, __p1) __extension__ ({ \
poly64x2x4_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst1q_f64_x4(__p0, __p1) __extension__ ({ \
float64x2x4_t __s1 = __p1; \
- __builtin_neon_vst1q_x4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 42); \
+ __builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 42); \
})
#else
#define vst1q_f64_x4(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
- __builtin_neon_vst1q_x4_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 42); \
+ __builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 42); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst1_f64_x4(__p0, __p1) __extension__ ({ \
- float64x1x4_t __s1 = __p1; \
- __builtin_neon_vst1_x4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 10); \
-})
-#else
#define vst1_f64_x4(__p0, __p1) __extension__ ({ \
float64x1x4_t __s1 = __p1; \
- __builtin_neon_vst1_x4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 10); \
+ __builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 10); \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vst2_p64(__p0, __p1) __extension__ ({ \
poly64x1x2_t __s1 = __p1; \
__builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 6); \
})
-#else
-#define vst2_p64(__p0, __p1) __extension__ ({ \
- poly64x1x2_t __s1 = __p1; \
- __builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 6); \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst2q_p64(__p0, __p1) __extension__ ({ \
poly64x2x2_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst2q_f64(__p0, __p1) __extension__ ({ \
float64x2x2_t __s1 = __p1; \
- __builtin_neon_vst2q_v(__p0, __s1.val[0], __s1.val[1], 42); \
+ __builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 42); \
})
#else
#define vst2q_f64(__p0, __p1) __extension__ ({ \
float64x2x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
- __builtin_neon_vst2q_v(__p0, __rev1.val[0], __rev1.val[1], 42); \
+ __builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 42); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst2q_s64(__p0, __p1) __extension__ ({ \
int64x2x2_t __s1 = __p1; \
- __builtin_neon_vst2q_v(__p0, __s1.val[0], __s1.val[1], 35); \
+ __builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 35); \
})
#else
#define vst2q_s64(__p0, __p1) __extension__ ({ \
int64x2x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
- __builtin_neon_vst2q_v(__p0, __rev1.val[0], __rev1.val[1], 35); \
+ __builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 35); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vst2_f64(__p0, __p1) __extension__ ({ \
float64x1x2_t __s1 = __p1; \
- __builtin_neon_vst2_v(__p0, __s1.val[0], __s1.val[1], 10); \
+ __builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 10); \
})
-#else
-#define vst2_f64(__p0, __p1) __extension__ ({ \
- float64x1x2_t __s1 = __p1; \
- __builtin_neon_vst2_v(__p0, __s1.val[0], __s1.val[1], 10); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vst2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x1x2_t __s1 = __p1; \
__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 6); \
})
-#else
-#define vst2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
- poly64x1x2_t __s1 = __p1; \
- __builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 6); \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
poly8x16x2_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x2x2_t __s1 = __p1; \
- __builtin_neon_vst2q_lane_v(__p0, __s1.val[0], __s1.val[1], __p2, 42); \
+ __builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 42); \
})
#else
#define vst2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x2x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
- __builtin_neon_vst2q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __p2, 42); \
+ __builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 42); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64x2x2_t __s1 = __p1; \
- __builtin_neon_vst2q_lane_v(__p0, __s1.val[0], __s1.val[1], __p2, 35); \
+ __builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 35); \
})
#else
#define vst2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64x2x2_t __rev1; \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
- __builtin_neon_vst2q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __p2, 35); \
+ __builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 35); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64x1x2_t __s1 = __p1; \
- __builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 19); \
-})
-#else
#define vst2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
uint64x1x2_t __s1 = __p1; \
__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 19); \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x1x2_t __s1 = __p1; \
- __builtin_neon_vst2_lane_v(__p0, __s1.val[0], __s1.val[1], __p2, 10); \
-})
-#else
#define vst2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x1x2_t __s1 = __p1; \
- __builtin_neon_vst2_lane_v(__p0, __s1.val[0], __s1.val[1], __p2, 10); \
+ __builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 10); \
})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
- int64x1x2_t __s1 = __p1; \
- __builtin_neon_vst2_lane_v(__p0, __s1.val[0], __s1.val[1], __p2, 3); \
-})
-#else
#define vst2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64x1x2_t __s1 = __p1; \
- __builtin_neon_vst2_lane_v(__p0, __s1.val[0], __s1.val[1], __p2, 3); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst3_p64(__p0, __p1) __extension__ ({ \
- poly64x1x3_t __s1 = __p1; \
- __builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 6); \
+ __builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 3); \
})
-#else
#define vst3_p64(__p0, __p1) __extension__ ({ \
poly64x1x3_t __s1 = __p1; \
__builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 6); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst3q_p64(__p0, __p1) __extension__ ({ \
poly64x2x3_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst3q_f64(__p0, __p1) __extension__ ({ \
float64x2x3_t __s1 = __p1; \
- __builtin_neon_vst3q_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 42); \
+ __builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 42); \
})
#else
#define vst3q_f64(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
- __builtin_neon_vst3q_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 42); \
+ __builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 42); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst3q_s64(__p0, __p1) __extension__ ({ \
int64x2x3_t __s1 = __p1; \
- __builtin_neon_vst3q_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 35); \
+ __builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 35); \
})
#else
#define vst3q_s64(__p0, __p1) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
- __builtin_neon_vst3q_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], 35); \
+ __builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 35); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vst3_f64(__p0, __p1) __extension__ ({ \
- float64x1x3_t __s1 = __p1; \
- __builtin_neon_vst3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 10); \
-})
-#else
#define vst3_f64(__p0, __p1) __extension__ ({ \
float64x1x3_t __s1 = __p1; \
- __builtin_neon_vst3_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], 10); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
- poly64x1x3_t __s1 = __p1; \
- __builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 6); \
+ __builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 10); \
})
-#else
#define vst3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x1x3_t __s1 = __p1; \
__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 6); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
poly8x16x3_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x2x3_t __s1 = __p1; \
- __builtin_neon_vst3q_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 42); \
+ __builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 42); \
})
#else
#define vst3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
- __builtin_neon_vst3q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 42); \
+ __builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 42); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64x2x3_t __s1 = __p1; \
- __builtin_neon_vst3q_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 35); \
+ __builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 35); \
})
#else
#define vst3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
- __builtin_neon_vst3q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __p2, 35); \
+ __builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 35); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vst3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
uint64x1x3_t __s1 = __p1; \
__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 19); \
})
-#else
-#define vst3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64x1x3_t __s1 = __p1; \
- __builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 19); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x1x3_t __s1 = __p1; \
- __builtin_neon_vst3_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 10); \
-})
-#else
#define vst3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x1x3_t __s1 = __p1; \
- __builtin_neon_vst3_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 10); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
- int64x1x3_t __s1 = __p1; \
- __builtin_neon_vst3_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 3); \
+ __builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 10); \
})
-#else
#define vst3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64x1x3_t __s1 = __p1; \
- __builtin_neon_vst3_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __p2, 3); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vst4_p64(__p0, __p1) __extension__ ({ \
- poly64x1x4_t __s1 = __p1; \
- __builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 6); \
+ __builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 3); \
})
-#else
#define vst4_p64(__p0, __p1) __extension__ ({ \
poly64x1x4_t __s1 = __p1; \
__builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 6); \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst4q_p64(__p0, __p1) __extension__ ({ \
poly64x2x4_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst4q_f64(__p0, __p1) __extension__ ({ \
float64x2x4_t __s1 = __p1; \
- __builtin_neon_vst4q_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 42); \
+ __builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 42); \
})
#else
#define vst4q_f64(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
- __builtin_neon_vst4q_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 42); \
+ __builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 42); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst4q_s64(__p0, __p1) __extension__ ({ \
int64x2x4_t __s1 = __p1; \
- __builtin_neon_vst4q_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 35); \
+ __builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 35); \
})
#else
#define vst4q_s64(__p0, __p1) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
- __builtin_neon_vst4q_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], 35); \
+ __builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 35); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vst4_f64(__p0, __p1) __extension__ ({ \
float64x1x4_t __s1 = __p1; \
- __builtin_neon_vst4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 10); \
+ __builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 10); \
})
-#else
-#define vst4_f64(__p0, __p1) __extension__ ({ \
- float64x1x4_t __s1 = __p1; \
- __builtin_neon_vst4_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], 10); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vst4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
poly64x1x4_t __s1 = __p1; \
__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 6); \
})
-#else
-#define vst4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
- poly64x1x4_t __s1 = __p1; \
- __builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 6); \
-})
-#endif
-
#ifdef __LITTLE_ENDIAN__
#define vst4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
poly8x16x4_t __s1 = __p1; \
#ifdef __LITTLE_ENDIAN__
#define vst4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x2x4_t __s1 = __p1; \
- __builtin_neon_vst4q_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 42); \
+ __builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 42); \
})
#else
#define vst4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
- __builtin_neon_vst4q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 42); \
+ __builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 42); \
})
#endif
#ifdef __LITTLE_ENDIAN__
#define vst4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64x2x4_t __s1 = __p1; \
- __builtin_neon_vst4q_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 35); \
+ __builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 35); \
})
#else
#define vst4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
- __builtin_neon_vst4q_lane_v(__p0, __rev1.val[0], __rev1.val[1], __rev1.val[2], __rev1.val[3], __p2, 35); \
+ __builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 35); \
})
#endif
-#ifdef __LITTLE_ENDIAN__
#define vst4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
uint64x1x4_t __s1 = __p1; \
__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 19); \
})
-#else
-#define vst4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
- uint64x1x4_t __s1 = __p1; \
- __builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 19); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vst4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
float64x1x4_t __s1 = __p1; \
- __builtin_neon_vst4_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 10); \
+ __builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 10); \
})
-#else
-#define vst4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
- float64x1x4_t __s1 = __p1; \
- __builtin_neon_vst4_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 10); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vst4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
int64x1x4_t __s1 = __p1; \
- __builtin_neon_vst4_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 3); \
+ __builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 3); \
})
-#else
-#define vst4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
- int64x1x4_t __s1 = __p1; \
- __builtin_neon_vst4_lane_v(__p0, __s1.val[0], __s1.val[1], __s1.val[2], __s1.val[3], __p2, 3); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
#define vstrq_p128(__p0, __p1) __extension__ ({ \
poly128_t __s1 = __p1; \
__builtin_neon_vstrq_p128(__p0, __s1); \
})
-#else
-#define vstrq_p128(__p0, __p1) __extension__ ({ \
- poly128_t __s1 = __p1; \
- __builtin_neon_vstrq_p128(__p0, __s1); \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai uint64_t vsubd_u64(uint64_t __p0, uint64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vsubd_u64(__p0, __p1);
- return __ret;
-}
-#else
__ai uint64_t vsubd_u64(uint64_t __p0, uint64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vsubd_u64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai int64_t vsubd_s64(int64_t __p0, int64_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vsubd_s64(__p0, __p1);
return __ret;
}
-#else
-__ai int64_t vsubd_s64(int64_t __p0, int64_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vsubd_s64(__p0, __p1);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai float64x2_t vsubq_f64(float64x2_t __p0, float64x2_t __p1) {
float64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai float64x1_t vsub_f64(float64x1_t __p0, float64x1_t __p1) {
float64x1_t __ret;
__ret = __p0 - __p1;
return __ret;
}
-#else
-__ai float64x1_t vsub_f64(float64x1_t __p0, float64x1_t __p1) {
- float64x1_t __ret;
- __ret = __p0 - __p1;
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint16x8_t vsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
uint16x8_t __ret;
#endif
#ifdef __LITTLE_ENDIAN__
+#define vsudotq_laneq_s32(__p0_735, __p1_735, __p2_735, __p3_735) __extension__ ({ \
+ int32x4_t __s0_735 = __p0_735; \
+ int8x16_t __s1_735 = __p1_735; \
+ uint8x16_t __s2_735 = __p2_735; \
+ int32x4_t __ret_735; \
+uint8x16_t __reint_735 = __s2_735; \
+ __ret_735 = vusdotq_s32(__s0_735, (uint8x16_t)(splatq_laneq_s32(*(int32x4_t *) &__reint_735, __p3_735)), __s1_735); \
+ __ret_735; \
+})
+#else
+#define vsudotq_laneq_s32(__p0_736, __p1_736, __p2_736, __p3_736) __extension__ ({ \
+ int32x4_t __s0_736 = __p0_736; \
+ int8x16_t __s1_736 = __p1_736; \
+ uint8x16_t __s2_736 = __p2_736; \
+ int32x4_t __rev0_736; __rev0_736 = __builtin_shufflevector(__s0_736, __s0_736, 3, 2, 1, 0); \
+ int8x16_t __rev1_736; __rev1_736 = __builtin_shufflevector(__s1_736, __s1_736, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __rev2_736; __rev2_736 = __builtin_shufflevector(__s2_736, __s2_736, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_736; \
+uint8x16_t __reint_736 = __rev2_736; \
+ __ret_736 = __noswap_vusdotq_s32(__rev0_736, (uint8x16_t)(__noswap_splatq_laneq_s32(*(int32x4_t *) &__reint_736, __p3_736)), __rev1_736); \
+ __ret_736 = __builtin_shufflevector(__ret_736, __ret_736, 3, 2, 1, 0); \
+ __ret_736; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vsudot_laneq_s32(__p0_737, __p1_737, __p2_737, __p3_737) __extension__ ({ \
+ int32x2_t __s0_737 = __p0_737; \
+ int8x8_t __s1_737 = __p1_737; \
+ uint8x16_t __s2_737 = __p2_737; \
+ int32x2_t __ret_737; \
+uint8x16_t __reint_737 = __s2_737; \
+ __ret_737 = vusdot_s32(__s0_737, (uint8x8_t)(splat_laneq_s32(*(int32x4_t *) &__reint_737, __p3_737)), __s1_737); \
+ __ret_737; \
+})
+#else
+#define vsudot_laneq_s32(__p0_738, __p1_738, __p2_738, __p3_738) __extension__ ({ \
+ int32x2_t __s0_738 = __p0_738; \
+ int8x8_t __s1_738 = __p1_738; \
+ uint8x16_t __s2_738 = __p2_738; \
+ int32x2_t __rev0_738; __rev0_738 = __builtin_shufflevector(__s0_738, __s0_738, 1, 0); \
+ int8x8_t __rev1_738; __rev1_738 = __builtin_shufflevector(__s1_738, __s1_738, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x16_t __rev2_738; __rev2_738 = __builtin_shufflevector(__s2_738, __s2_738, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x2_t __ret_738; \
+uint8x16_t __reint_738 = __rev2_738; \
+ __ret_738 = __noswap_vusdot_s32(__rev0_738, (uint8x8_t)(__noswap_splat_laneq_s32(*(int32x4_t *) &__reint_738, __p3_738)), __rev1_738); \
+ __ret_738 = __builtin_shufflevector(__ret_738, __ret_738, 1, 0); \
+ __ret_738; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
__ai poly8x8_t vtrn1_p8(poly8x8_t __p0, poly8x8_t __p1) {
poly8x8_t __ret;
__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vtst_p64(poly64x1_t __p0, poly64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
return __ret;
}
-#else
-__ai uint64x1_t vtst_p64(poly64x1_t __p0, poly64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
- return __ret;
-}
-#endif
-
#ifdef __LITTLE_ENDIAN__
__ai uint64x2_t vtstq_p64(poly64x2_t __p0, poly64x2_t __p1) {
uint64x2_t __ret;
}
#endif
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vtst_u64(uint64x1_t __p0, uint64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
return __ret;
}
-#else
-__ai uint64x1_t vtst_u64(uint64x1_t __p0, uint64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint64x1_t vtst_s64(int64x1_t __p0, int64x1_t __p1) {
uint64x1_t __ret;
__ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
return __ret;
}
-#else
-__ai uint64x1_t vtst_s64(int64x1_t __p0, int64x1_t __p1) {
- uint64x1_t __ret;
- __ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
__ai uint64_t vtstd_u64(uint64_t __p0, uint64_t __p1) {
uint64_t __ret;
__ret = (uint64_t) __builtin_neon_vtstd_u64(__p0, __p1);
return __ret;
}
-#else
-__ai uint64_t vtstd_u64(uint64_t __p0, uint64_t __p1) {
- uint64_t __ret;
- __ret = (uint64_t) __builtin_neon_vtstd_u64(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vtstd_s64(int64_t __p0, int64_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vtstd_s64(__p0, __p1);
- return __ret;
-}
-#else
__ai int64_t vtstd_s64(int64_t __p0, int64_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vtstd_s64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int8_t vuqaddb_s8(int8_t __p0, int8_t __p1) {
+__ai int8_t vuqaddb_s8(int8_t __p0, uint8_t __p1) {
int8_t __ret;
__ret = (int8_t) __builtin_neon_vuqaddb_s8(__p0, __p1);
return __ret;
}
-#else
-__ai int8_t vuqaddb_s8(int8_t __p0, int8_t __p1) {
- int8_t __ret;
- __ret = (int8_t) __builtin_neon_vuqaddb_s8(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int32_t vuqadds_s32(int32_t __p0, int32_t __p1) {
+__ai int32_t vuqadds_s32(int32_t __p0, uint32_t __p1) {
int32_t __ret;
__ret = (int32_t) __builtin_neon_vuqadds_s32(__p0, __p1);
return __ret;
}
-#else
-__ai int32_t vuqadds_s32(int32_t __p0, int32_t __p1) {
- int32_t __ret;
- __ret = (int32_t) __builtin_neon_vuqadds_s32(__p0, __p1);
- return __ret;
-}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int64_t vuqaddd_s64(int64_t __p0, int64_t __p1) {
- int64_t __ret;
- __ret = (int64_t) __builtin_neon_vuqaddd_s64(__p0, __p1);
- return __ret;
-}
-#else
-__ai int64_t vuqaddd_s64(int64_t __p0, int64_t __p1) {
+__ai int64_t vuqaddd_s64(int64_t __p0, uint64_t __p1) {
int64_t __ret;
__ret = (int64_t) __builtin_neon_vuqaddd_s64(__p0, __p1);
return __ret;
}
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-__ai int16_t vuqaddh_s16(int16_t __p0, int16_t __p1) {
- int16_t __ret;
- __ret = (int16_t) __builtin_neon_vuqaddh_s16(__p0, __p1);
- return __ret;
-}
-#else
-__ai int16_t vuqaddh_s16(int16_t __p0, int16_t __p1) {
+__ai int16_t vuqaddh_s16(int16_t __p0, uint16_t __p1) {
int16_t __ret;
__ret = (int16_t) __builtin_neon_vuqaddh_s16(__p0, __p1);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
-__ai int8x16_t vuqaddq_s8(int8x16_t __p0, int8x16_t __p1) {
+__ai int8x16_t vuqaddq_s8(int8x16_t __p0, uint8x16_t __p1) {
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
return __ret;
}
#else
-__ai int8x16_t vuqaddq_s8(int8x16_t __p0, int8x16_t __p1) {
+__ai int8x16_t vuqaddq_s8(int8x16_t __p0, uint8x16_t __p1) {
int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
int8x16_t __ret;
__ret = (int8x16_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x4_t vuqaddq_s32(int32x4_t __p0, int32x4_t __p1) {
+__ai int32x4_t vuqaddq_s32(int32x4_t __p0, uint32x4_t __p1) {
int32x4_t __ret;
__ret = (int32x4_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
return __ret;
}
#else
-__ai int32x4_t vuqaddq_s32(int32x4_t __p0, int32x4_t __p1) {
+__ai int32x4_t vuqaddq_s32(int32x4_t __p0, uint32x4_t __p1) {
int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
int32x4_t __ret;
__ret = (int32x4_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int64x2_t vuqaddq_s64(int64x2_t __p0, int64x2_t __p1) {
+__ai int64x2_t vuqaddq_s64(int64x2_t __p0, uint64x2_t __p1) {
int64x2_t __ret;
__ret = (int64x2_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
return __ret;
}
#else
-__ai int64x2_t vuqaddq_s64(int64x2_t __p0, int64x2_t __p1) {
+__ai int64x2_t vuqaddq_s64(int64x2_t __p0, uint64x2_t __p1) {
int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
- int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
+ uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
int64x2_t __ret;
__ret = (int64x2_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 35);
__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16x8_t vuqaddq_s16(int16x8_t __p0, int16x8_t __p1) {
+__ai int16x8_t vuqaddq_s16(int16x8_t __p0, uint16x8_t __p1) {
int16x8_t __ret;
__ret = (int16x8_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
return __ret;
}
#else
-__ai int16x8_t vuqaddq_s16(int16x8_t __p0, int16x8_t __p1) {
+__ai int16x8_t vuqaddq_s16(int16x8_t __p0, uint16x8_t __p1) {
int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
int16x8_t __ret;
__ret = (int16x8_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int8x8_t vuqadd_s8(int8x8_t __p0, int8x8_t __p1) {
+__ai int8x8_t vuqadd_s8(int8x8_t __p0, uint8x8_t __p1) {
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
return __ret;
}
#else
-__ai int8x8_t vuqadd_s8(int8x8_t __p0, int8x8_t __p1) {
+__ai int8x8_t vuqadd_s8(int8x8_t __p0, uint8x8_t __p1) {
int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
- int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
+ uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
int8x8_t __ret;
__ret = (int8x8_t) __builtin_neon_vuqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32x2_t vuqadd_s32(int32x2_t __p0, int32x2_t __p1) {
+__ai int32x2_t vuqadd_s32(int32x2_t __p0, uint32x2_t __p1) {
int32x2_t __ret;
__ret = (int32x2_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
return __ret;
}
#else
-__ai int32x2_t vuqadd_s32(int32x2_t __p0, int32x2_t __p1) {
+__ai int32x2_t vuqadd_s32(int32x2_t __p0, uint32x2_t __p1) {
int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
+ uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
int32x2_t __ret;
__ret = (int32x2_t) __builtin_neon_vuqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
}
#endif
-#ifdef __LITTLE_ENDIAN__
-__ai int64x1_t vuqadd_s64(int64x1_t __p0, int64x1_t __p1) {
- int64x1_t __ret;
- __ret = (int64x1_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
- return __ret;
-}
-#else
-__ai int64x1_t vuqadd_s64(int64x1_t __p0, int64x1_t __p1) {
+__ai int64x1_t vuqadd_s64(int64x1_t __p0, uint64x1_t __p1) {
int64x1_t __ret;
__ret = (int64x1_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
-__ai int16x4_t vuqadd_s16(int16x4_t __p0, int16x4_t __p1) {
+__ai int16x4_t vuqadd_s16(int16x4_t __p0, uint16x4_t __p1) {
int16x4_t __ret;
__ret = (int16x4_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
return __ret;
}
#else
-__ai int16x4_t vuqadd_s16(int16x4_t __p0, int16x4_t __p1) {
+__ai int16x4_t vuqadd_s16(int16x4_t __p0, uint16x4_t __p1) {
int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
+ uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
int16x4_t __ret;
__ret = (int16x4_t) __builtin_neon_vuqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
#endif
#ifdef __LITTLE_ENDIAN__
+#define vusdotq_laneq_s32(__p0_739, __p1_739, __p2_739, __p3_739) __extension__ ({ \
+ int32x4_t __s0_739 = __p0_739; \
+ uint8x16_t __s1_739 = __p1_739; \
+ int8x16_t __s2_739 = __p2_739; \
+ int32x4_t __ret_739; \
+int8x16_t __reint_739 = __s2_739; \
+ __ret_739 = vusdotq_s32(__s0_739, __s1_739, (int8x16_t)(splatq_laneq_s32(*(int32x4_t *) &__reint_739, __p3_739))); \
+ __ret_739; \
+})
+#else
+#define vusdotq_laneq_s32(__p0_740, __p1_740, __p2_740, __p3_740) __extension__ ({ \
+ int32x4_t __s0_740 = __p0_740; \
+ uint8x16_t __s1_740 = __p1_740; \
+ int8x16_t __s2_740 = __p2_740; \
+ int32x4_t __rev0_740; __rev0_740 = __builtin_shufflevector(__s0_740, __s0_740, 3, 2, 1, 0); \
+ uint8x16_t __rev1_740; __rev1_740 = __builtin_shufflevector(__s1_740, __s1_740, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __rev2_740; __rev2_740 = __builtin_shufflevector(__s2_740, __s2_740, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_740; \
+int8x16_t __reint_740 = __rev2_740; \
+ __ret_740 = __noswap_vusdotq_s32(__rev0_740, __rev1_740, (int8x16_t)(__noswap_splatq_laneq_s32(*(int32x4_t *) &__reint_740, __p3_740))); \
+ __ret_740 = __builtin_shufflevector(__ret_740, __ret_740, 3, 2, 1, 0); \
+ __ret_740; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vusdot_laneq_s32(__p0_741, __p1_741, __p2_741, __p3_741) __extension__ ({ \
+ int32x2_t __s0_741 = __p0_741; \
+ uint8x8_t __s1_741 = __p1_741; \
+ int8x16_t __s2_741 = __p2_741; \
+ int32x2_t __ret_741; \
+int8x16_t __reint_741 = __s2_741; \
+ __ret_741 = vusdot_s32(__s0_741, __s1_741, (int8x8_t)(splat_laneq_s32(*(int32x4_t *) &__reint_741, __p3_741))); \
+ __ret_741; \
+})
+#else
+#define vusdot_laneq_s32(__p0_742, __p1_742, __p2_742, __p3_742) __extension__ ({ \
+ int32x2_t __s0_742 = __p0_742; \
+ uint8x8_t __s1_742 = __p1_742; \
+ int8x16_t __s2_742 = __p2_742; \
+ int32x2_t __rev0_742; __rev0_742 = __builtin_shufflevector(__s0_742, __s0_742, 1, 0); \
+ uint8x8_t __rev1_742; __rev1_742 = __builtin_shufflevector(__s1_742, __s1_742, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int8x16_t __rev2_742; __rev2_742 = __builtin_shufflevector(__s2_742, __s2_742, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x2_t __ret_742; \
+int8x16_t __reint_742 = __rev2_742; \
+ __ret_742 = __noswap_vusdot_s32(__rev0_742, __rev1_742, (int8x8_t)(__noswap_splat_laneq_s32(*(int32x4_t *) &__reint_742, __p3_742))); \
+ __ret_742 = __builtin_shufflevector(__ret_742, __ret_742, 1, 0); \
+ __ret_742; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
__ai poly8x8_t vuzp1_p8(poly8x8_t __p0, poly8x8_t __p1) {
poly8x8_t __ret;
__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
#endif
#ifdef __LITTLE_ENDIAN__
-#define vget_lane_f16(__p0_254, __p1_254) __extension__ ({ \
- float16x4_t __s0_254 = __p0_254; \
- float16_t __ret_254; \
-float16x4_t __reint_254 = __s0_254; \
-int16_t __reint1_254 = vget_lane_s16(*(int16x4_t *) &__reint_254, __p1_254); \
- __ret_254 = *(float16_t *) &__reint1_254; \
- __ret_254; \
+#define vget_lane_f16(__p0_743, __p1_743) __extension__ ({ \
+ float16x4_t __s0_743 = __p0_743; \
+ float16_t __ret_743; \
+float16x4_t __reint_743 = __s0_743; \
+int16_t __reint1_743 = vget_lane_s16(*(int16x4_t *) &__reint_743, __p1_743); \
+ __ret_743 = *(float16_t *) &__reint1_743; \
+ __ret_743; \
})
#else
-#define vget_lane_f16(__p0_255, __p1_255) __extension__ ({ \
- float16x4_t __s0_255 = __p0_255; \
- float16x4_t __rev0_255; __rev0_255 = __builtin_shufflevector(__s0_255, __s0_255, 3, 2, 1, 0); \
- float16_t __ret_255; \
-float16x4_t __reint_255 = __rev0_255; \
-int16_t __reint1_255 = __noswap_vget_lane_s16(*(int16x4_t *) &__reint_255, __p1_255); \
- __ret_255 = *(float16_t *) &__reint1_255; \
- __ret_255; \
+#define vget_lane_f16(__p0_744, __p1_744) __extension__ ({ \
+ float16x4_t __s0_744 = __p0_744; \
+ float16x4_t __rev0_744; __rev0_744 = __builtin_shufflevector(__s0_744, __s0_744, 3, 2, 1, 0); \
+ float16_t __ret_744; \
+float16x4_t __reint_744 = __rev0_744; \
+int16_t __reint1_744 = __noswap_vget_lane_s16(*(int16x4_t *) &__reint_744, __p1_744); \
+ __ret_744 = *(float16_t *) &__reint1_744; \
+ __ret_744; \
})
-#define __noswap_vget_lane_f16(__p0_256, __p1_256) __extension__ ({ \
- float16x4_t __s0_256 = __p0_256; \
- float16_t __ret_256; \
-float16x4_t __reint_256 = __s0_256; \
-int16_t __reint1_256 = __noswap_vget_lane_s16(*(int16x4_t *) &__reint_256, __p1_256); \
- __ret_256 = *(float16_t *) &__reint1_256; \
- __ret_256; \
+#define __noswap_vget_lane_f16(__p0_745, __p1_745) __extension__ ({ \
+ float16x4_t __s0_745 = __p0_745; \
+ float16_t __ret_745; \
+float16x4_t __reint_745 = __s0_745; \
+int16_t __reint1_745 = __noswap_vget_lane_s16(*(int16x4_t *) &__reint_745, __p1_745); \
+ __ret_745 = *(float16_t *) &__reint1_745; \
+ __ret_745; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vgetq_lane_f16(__p0_257, __p1_257) __extension__ ({ \
- float16x8_t __s0_257 = __p0_257; \
- float16_t __ret_257; \
-float16x8_t __reint_257 = __s0_257; \
-int16_t __reint1_257 = vgetq_lane_s16(*(int16x8_t *) &__reint_257, __p1_257); \
- __ret_257 = *(float16_t *) &__reint1_257; \
- __ret_257; \
+#define vgetq_lane_f16(__p0_746, __p1_746) __extension__ ({ \
+ float16x8_t __s0_746 = __p0_746; \
+ float16_t __ret_746; \
+float16x8_t __reint_746 = __s0_746; \
+int16_t __reint1_746 = vgetq_lane_s16(*(int16x8_t *) &__reint_746, __p1_746); \
+ __ret_746 = *(float16_t *) &__reint1_746; \
+ __ret_746; \
})
#else
-#define vgetq_lane_f16(__p0_258, __p1_258) __extension__ ({ \
- float16x8_t __s0_258 = __p0_258; \
- float16x8_t __rev0_258; __rev0_258 = __builtin_shufflevector(__s0_258, __s0_258, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16_t __ret_258; \
-float16x8_t __reint_258 = __rev0_258; \
-int16_t __reint1_258 = __noswap_vgetq_lane_s16(*(int16x8_t *) &__reint_258, __p1_258); \
- __ret_258 = *(float16_t *) &__reint1_258; \
- __ret_258; \
+#define vgetq_lane_f16(__p0_747, __p1_747) __extension__ ({ \
+ float16x8_t __s0_747 = __p0_747; \
+ float16x8_t __rev0_747; __rev0_747 = __builtin_shufflevector(__s0_747, __s0_747, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16_t __ret_747; \
+float16x8_t __reint_747 = __rev0_747; \
+int16_t __reint1_747 = __noswap_vgetq_lane_s16(*(int16x8_t *) &__reint_747, __p1_747); \
+ __ret_747 = *(float16_t *) &__reint1_747; \
+ __ret_747; \
})
-#define __noswap_vgetq_lane_f16(__p0_259, __p1_259) __extension__ ({ \
- float16x8_t __s0_259 = __p0_259; \
- float16_t __ret_259; \
-float16x8_t __reint_259 = __s0_259; \
-int16_t __reint1_259 = __noswap_vgetq_lane_s16(*(int16x8_t *) &__reint_259, __p1_259); \
- __ret_259 = *(float16_t *) &__reint1_259; \
- __ret_259; \
+#define __noswap_vgetq_lane_f16(__p0_748, __p1_748) __extension__ ({ \
+ float16x8_t __s0_748 = __p0_748; \
+ float16_t __ret_748; \
+float16x8_t __reint_748 = __s0_748; \
+int16_t __reint1_748 = __noswap_vgetq_lane_s16(*(int16x8_t *) &__reint_748, __p1_748); \
+ __ret_748 = *(float16_t *) &__reint1_748; \
+ __ret_748; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint64x2_t __ret; \
- __ret = __s0 + vmull_u32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlal_lane_u32(__p0_749, __p1_749, __p2_749, __p3_749) __extension__ ({ \
+ uint64x2_t __s0_749 = __p0_749; \
+ uint32x2_t __s1_749 = __p1_749; \
+ uint32x2_t __s2_749 = __p2_749; \
+ uint64x2_t __ret_749; \
+ __ret_749 = __s0_749 + vmull_u32(__s1_749, splat_lane_u32(__s2_749, __p3_749)); \
+ __ret_749; \
})
#else
-#define vmlal_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- uint32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- uint64x2_t __ret; \
- __ret = __rev0 + __noswap_vmull_u32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmlal_lane_u32(__p0_750, __p1_750, __p2_750, __p3_750) __extension__ ({ \
+ uint64x2_t __s0_750 = __p0_750; \
+ uint32x2_t __s1_750 = __p1_750; \
+ uint32x2_t __s2_750 = __p2_750; \
+ uint64x2_t __rev0_750; __rev0_750 = __builtin_shufflevector(__s0_750, __s0_750, 1, 0); \
+ uint32x2_t __rev1_750; __rev1_750 = __builtin_shufflevector(__s1_750, __s1_750, 1, 0); \
+ uint32x2_t __rev2_750; __rev2_750 = __builtin_shufflevector(__s2_750, __s2_750, 1, 0); \
+ uint64x2_t __ret_750; \
+ __ret_750 = __rev0_750 + __noswap_vmull_u32(__rev1_750, __noswap_splat_lane_u32(__rev2_750, __p3_750)); \
+ __ret_750 = __builtin_shufflevector(__ret_750, __ret_750, 1, 0); \
+ __ret_750; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint32x4_t __ret; \
- __ret = __s0 + vmull_u16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlal_lane_u16(__p0_751, __p1_751, __p2_751, __p3_751) __extension__ ({ \
+ uint32x4_t __s0_751 = __p0_751; \
+ uint16x4_t __s1_751 = __p1_751; \
+ uint16x4_t __s2_751 = __p2_751; \
+ uint32x4_t __ret_751; \
+ __ret_751 = __s0_751 + vmull_u16(__s1_751, splat_lane_u16(__s2_751, __p3_751)); \
+ __ret_751; \
})
#else
-#define vmlal_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __rev0 + __noswap_vmull_u16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlal_lane_u16(__p0_752, __p1_752, __p2_752, __p3_752) __extension__ ({ \
+ uint32x4_t __s0_752 = __p0_752; \
+ uint16x4_t __s1_752 = __p1_752; \
+ uint16x4_t __s2_752 = __p2_752; \
+ uint32x4_t __rev0_752; __rev0_752 = __builtin_shufflevector(__s0_752, __s0_752, 3, 2, 1, 0); \
+ uint16x4_t __rev1_752; __rev1_752 = __builtin_shufflevector(__s1_752, __s1_752, 3, 2, 1, 0); \
+ uint16x4_t __rev2_752; __rev2_752 = __builtin_shufflevector(__s2_752, __s2_752, 3, 2, 1, 0); \
+ uint32x4_t __ret_752; \
+ __ret_752 = __rev0_752 + __noswap_vmull_u16(__rev1_752, __noswap_splat_lane_u16(__rev2_752, __p3_752)); \
+ __ret_752 = __builtin_shufflevector(__ret_752, __ret_752, 3, 2, 1, 0); \
+ __ret_752; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = __s0 + vmull_s32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlal_lane_s32(__p0_753, __p1_753, __p2_753, __p3_753) __extension__ ({ \
+ int64x2_t __s0_753 = __p0_753; \
+ int32x2_t __s1_753 = __p1_753; \
+ int32x2_t __s2_753 = __p2_753; \
+ int64x2_t __ret_753; \
+ __ret_753 = __s0_753 + vmull_s32(__s1_753, splat_lane_s32(__s2_753, __p3_753)); \
+ __ret_753; \
})
#else
-#define vmlal_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int64x2_t __ret; \
- __ret = __rev0 + __noswap_vmull_s32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmlal_lane_s32(__p0_754, __p1_754, __p2_754, __p3_754) __extension__ ({ \
+ int64x2_t __s0_754 = __p0_754; \
+ int32x2_t __s1_754 = __p1_754; \
+ int32x2_t __s2_754 = __p2_754; \
+ int64x2_t __rev0_754; __rev0_754 = __builtin_shufflevector(__s0_754, __s0_754, 1, 0); \
+ int32x2_t __rev1_754; __rev1_754 = __builtin_shufflevector(__s1_754, __s1_754, 1, 0); \
+ int32x2_t __rev2_754; __rev2_754 = __builtin_shufflevector(__s2_754, __s2_754, 1, 0); \
+ int64x2_t __ret_754; \
+ __ret_754 = __rev0_754 + __noswap_vmull_s32(__rev1_754, __noswap_splat_lane_s32(__rev2_754, __p3_754)); \
+ __ret_754 = __builtin_shufflevector(__ret_754, __ret_754, 1, 0); \
+ __ret_754; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlal_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = __s0 + vmull_s16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlal_lane_s16(__p0_755, __p1_755, __p2_755, __p3_755) __extension__ ({ \
+ int32x4_t __s0_755 = __p0_755; \
+ int16x4_t __s1_755 = __p1_755; \
+ int16x4_t __s2_755 = __p2_755; \
+ int32x4_t __ret_755; \
+ __ret_755 = __s0_755 + vmull_s16(__s1_755, splat_lane_s16(__s2_755, __p3_755)); \
+ __ret_755; \
})
#else
-#define vmlal_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __rev0 + __noswap_vmull_s16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlal_lane_s16(__p0_756, __p1_756, __p2_756, __p3_756) __extension__ ({ \
+ int32x4_t __s0_756 = __p0_756; \
+ int16x4_t __s1_756 = __p1_756; \
+ int16x4_t __s2_756 = __p2_756; \
+ int32x4_t __rev0_756; __rev0_756 = __builtin_shufflevector(__s0_756, __s0_756, 3, 2, 1, 0); \
+ int16x4_t __rev1_756; __rev1_756 = __builtin_shufflevector(__s1_756, __s1_756, 3, 2, 1, 0); \
+ int16x4_t __rev2_756; __rev2_756 = __builtin_shufflevector(__s2_756, __s2_756, 3, 2, 1, 0); \
+ int32x4_t __ret_756; \
+ __ret_756 = __rev0_756 + __noswap_vmull_s16(__rev1_756, __noswap_splat_lane_s16(__rev2_756, __p3_756)); \
+ __ret_756 = __builtin_shufflevector(__ret_756, __ret_756, 3, 2, 1, 0); \
+ __ret_756; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint64x2_t __ret; \
- __ret = __s0 - vmull_u32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlsl_lane_u32(__p0_757, __p1_757, __p2_757, __p3_757) __extension__ ({ \
+ uint64x2_t __s0_757 = __p0_757; \
+ uint32x2_t __s1_757 = __p1_757; \
+ uint32x2_t __s2_757 = __p2_757; \
+ uint64x2_t __ret_757; \
+ __ret_757 = __s0_757 - vmull_u32(__s1_757, splat_lane_u32(__s2_757, __p3_757)); \
+ __ret_757; \
})
#else
-#define vmlsl_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint64x2_t __s0 = __p0; \
- uint32x2_t __s1 = __p1; \
- uint32x2_t __s2 = __p2; \
- uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- uint32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- uint64x2_t __ret; \
- __ret = __rev0 - __noswap_vmull_u32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmlsl_lane_u32(__p0_758, __p1_758, __p2_758, __p3_758) __extension__ ({ \
+ uint64x2_t __s0_758 = __p0_758; \
+ uint32x2_t __s1_758 = __p1_758; \
+ uint32x2_t __s2_758 = __p2_758; \
+ uint64x2_t __rev0_758; __rev0_758 = __builtin_shufflevector(__s0_758, __s0_758, 1, 0); \
+ uint32x2_t __rev1_758; __rev1_758 = __builtin_shufflevector(__s1_758, __s1_758, 1, 0); \
+ uint32x2_t __rev2_758; __rev2_758 = __builtin_shufflevector(__s2_758, __s2_758, 1, 0); \
+ uint64x2_t __ret_758; \
+ __ret_758 = __rev0_758 - __noswap_vmull_u32(__rev1_758, __noswap_splat_lane_u32(__rev2_758, __p3_758)); \
+ __ret_758 = __builtin_shufflevector(__ret_758, __ret_758, 1, 0); \
+ __ret_758; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint32x4_t __ret; \
- __ret = __s0 - vmull_u16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlsl_lane_u16(__p0_759, __p1_759, __p2_759, __p3_759) __extension__ ({ \
+ uint32x4_t __s0_759 = __p0_759; \
+ uint16x4_t __s1_759 = __p1_759; \
+ uint16x4_t __s2_759 = __p2_759; \
+ uint32x4_t __ret_759; \
+ __ret_759 = __s0_759 - vmull_u16(__s1_759, splat_lane_u16(__s2_759, __p3_759)); \
+ __ret_759; \
})
#else
-#define vmlsl_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
- uint32x4_t __s0 = __p0; \
- uint16x4_t __s1 = __p1; \
- uint16x4_t __s2 = __p2; \
- uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- uint16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- uint32x4_t __ret; \
- __ret = __rev0 - __noswap_vmull_u16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlsl_lane_u16(__p0_760, __p1_760, __p2_760, __p3_760) __extension__ ({ \
+ uint32x4_t __s0_760 = __p0_760; \
+ uint16x4_t __s1_760 = __p1_760; \
+ uint16x4_t __s2_760 = __p2_760; \
+ uint32x4_t __rev0_760; __rev0_760 = __builtin_shufflevector(__s0_760, __s0_760, 3, 2, 1, 0); \
+ uint16x4_t __rev1_760; __rev1_760 = __builtin_shufflevector(__s1_760, __s1_760, 3, 2, 1, 0); \
+ uint16x4_t __rev2_760; __rev2_760 = __builtin_shufflevector(__s2_760, __s2_760, 3, 2, 1, 0); \
+ uint32x4_t __ret_760; \
+ __ret_760 = __rev0_760 - __noswap_vmull_u16(__rev1_760, __noswap_splat_lane_u16(__rev2_760, __p3_760)); \
+ __ret_760 = __builtin_shufflevector(__ret_760, __ret_760, 3, 2, 1, 0); \
+ __ret_760; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __ret; \
- __ret = __s0 - vmull_s32(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3)); \
- __ret; \
+#define vmlsl_lane_s32(__p0_761, __p1_761, __p2_761, __p3_761) __extension__ ({ \
+ int64x2_t __s0_761 = __p0_761; \
+ int32x2_t __s1_761 = __p1_761; \
+ int32x2_t __s2_761 = __p2_761; \
+ int64x2_t __ret_761; \
+ __ret_761 = __s0_761 - vmull_s32(__s1_761, splat_lane_s32(__s2_761, __p3_761)); \
+ __ret_761; \
})
#else
-#define vmlsl_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
- int64x2_t __s0 = __p0; \
- int32x2_t __s1 = __p1; \
- int32x2_t __s2 = __p2; \
- int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
- int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
- int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
- int64x2_t __ret; \
- __ret = __rev0 - __noswap_vmull_s32(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
- __ret; \
+#define vmlsl_lane_s32(__p0_762, __p1_762, __p2_762, __p3_762) __extension__ ({ \
+ int64x2_t __s0_762 = __p0_762; \
+ int32x2_t __s1_762 = __p1_762; \
+ int32x2_t __s2_762 = __p2_762; \
+ int64x2_t __rev0_762; __rev0_762 = __builtin_shufflevector(__s0_762, __s0_762, 1, 0); \
+ int32x2_t __rev1_762; __rev1_762 = __builtin_shufflevector(__s1_762, __s1_762, 1, 0); \
+ int32x2_t __rev2_762; __rev2_762 = __builtin_shufflevector(__s2_762, __s2_762, 1, 0); \
+ int64x2_t __ret_762; \
+ __ret_762 = __rev0_762 - __noswap_vmull_s32(__rev1_762, __noswap_splat_lane_s32(__rev2_762, __p3_762)); \
+ __ret_762 = __builtin_shufflevector(__ret_762, __ret_762, 1, 0); \
+ __ret_762; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmlsl_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __ret; \
- __ret = __s0 - vmull_s16(__s1, __builtin_shufflevector(__s2, __s2, __p3, __p3, __p3, __p3)); \
- __ret; \
+#define vmlsl_lane_s16(__p0_763, __p1_763, __p2_763, __p3_763) __extension__ ({ \
+ int32x4_t __s0_763 = __p0_763; \
+ int16x4_t __s1_763 = __p1_763; \
+ int16x4_t __s2_763 = __p2_763; \
+ int32x4_t __ret_763; \
+ __ret_763 = __s0_763 - vmull_s16(__s1_763, splat_lane_s16(__s2_763, __p3_763)); \
+ __ret_763; \
})
#else
-#define vmlsl_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
- int32x4_t __s0 = __p0; \
- int16x4_t __s1 = __p1; \
- int16x4_t __s2 = __p2; \
- int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
- int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
- int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
- int32x4_t __ret; \
- __ret = __rev0 - __noswap_vmull_s16(__rev1, __builtin_shufflevector(__rev2, __rev2, __p3, __p3, __p3, __p3)); \
- __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
- __ret; \
+#define vmlsl_lane_s16(__p0_764, __p1_764, __p2_764, __p3_764) __extension__ ({ \
+ int32x4_t __s0_764 = __p0_764; \
+ int16x4_t __s1_764 = __p1_764; \
+ int16x4_t __s2_764 = __p2_764; \
+ int32x4_t __rev0_764; __rev0_764 = __builtin_shufflevector(__s0_764, __s0_764, 3, 2, 1, 0); \
+ int16x4_t __rev1_764; __rev1_764 = __builtin_shufflevector(__s1_764, __s1_764, 3, 2, 1, 0); \
+ int16x4_t __rev2_764; __rev2_764 = __builtin_shufflevector(__s2_764, __s2_764, 3, 2, 1, 0); \
+ int32x4_t __ret_764; \
+ __ret_764 = __rev0_764 - __noswap_vmull_s16(__rev1_764, __noswap_splat_lane_s16(__rev2_764, __p3_764)); \
+ __ret_764 = __builtin_shufflevector(__ret_764, __ret_764, 3, 2, 1, 0); \
+ __ret_764; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vset_lane_f16(__p0_260, __p1_260, __p2_260) __extension__ ({ \
- float16_t __s0_260 = __p0_260; \
- float16x4_t __s1_260 = __p1_260; \
- float16x4_t __ret_260; \
-float16_t __reint_260 = __s0_260; \
-float16x4_t __reint1_260 = __s1_260; \
-int16x4_t __reint2_260 = vset_lane_s16(*(int16_t *) &__reint_260, *(int16x4_t *) &__reint1_260, __p2_260); \
- __ret_260 = *(float16x4_t *) &__reint2_260; \
- __ret_260; \
+#define vset_lane_f16(__p0_765, __p1_765, __p2_765) __extension__ ({ \
+ float16_t __s0_765 = __p0_765; \
+ float16x4_t __s1_765 = __p1_765; \
+ float16x4_t __ret_765; \
+float16_t __reint_765 = __s0_765; \
+float16x4_t __reint1_765 = __s1_765; \
+int16x4_t __reint2_765 = vset_lane_s16(*(int16_t *) &__reint_765, *(int16x4_t *) &__reint1_765, __p2_765); \
+ __ret_765 = *(float16x4_t *) &__reint2_765; \
+ __ret_765; \
})
#else
-#define vset_lane_f16(__p0_261, __p1_261, __p2_261) __extension__ ({ \
- float16_t __s0_261 = __p0_261; \
- float16x4_t __s1_261 = __p1_261; \
- float16x4_t __rev1_261; __rev1_261 = __builtin_shufflevector(__s1_261, __s1_261, 3, 2, 1, 0); \
- float16x4_t __ret_261; \
-float16_t __reint_261 = __s0_261; \
-float16x4_t __reint1_261 = __rev1_261; \
-int16x4_t __reint2_261 = __noswap_vset_lane_s16(*(int16_t *) &__reint_261, *(int16x4_t *) &__reint1_261, __p2_261); \
- __ret_261 = *(float16x4_t *) &__reint2_261; \
- __ret_261 = __builtin_shufflevector(__ret_261, __ret_261, 3, 2, 1, 0); \
- __ret_261; \
+#define vset_lane_f16(__p0_766, __p1_766, __p2_766) __extension__ ({ \
+ float16_t __s0_766 = __p0_766; \
+ float16x4_t __s1_766 = __p1_766; \
+ float16x4_t __rev1_766; __rev1_766 = __builtin_shufflevector(__s1_766, __s1_766, 3, 2, 1, 0); \
+ float16x4_t __ret_766; \
+float16_t __reint_766 = __s0_766; \
+float16x4_t __reint1_766 = __rev1_766; \
+int16x4_t __reint2_766 = __noswap_vset_lane_s16(*(int16_t *) &__reint_766, *(int16x4_t *) &__reint1_766, __p2_766); \
+ __ret_766 = *(float16x4_t *) &__reint2_766; \
+ __ret_766 = __builtin_shufflevector(__ret_766, __ret_766, 3, 2, 1, 0); \
+ __ret_766; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vsetq_lane_f16(__p0_262, __p1_262, __p2_262) __extension__ ({ \
- float16_t __s0_262 = __p0_262; \
- float16x8_t __s1_262 = __p1_262; \
- float16x8_t __ret_262; \
-float16_t __reint_262 = __s0_262; \
-float16x8_t __reint1_262 = __s1_262; \
-int16x8_t __reint2_262 = vsetq_lane_s16(*(int16_t *) &__reint_262, *(int16x8_t *) &__reint1_262, __p2_262); \
- __ret_262 = *(float16x8_t *) &__reint2_262; \
- __ret_262; \
+#define vsetq_lane_f16(__p0_767, __p1_767, __p2_767) __extension__ ({ \
+ float16_t __s0_767 = __p0_767; \
+ float16x8_t __s1_767 = __p1_767; \
+ float16x8_t __ret_767; \
+float16_t __reint_767 = __s0_767; \
+float16x8_t __reint1_767 = __s1_767; \
+int16x8_t __reint2_767 = vsetq_lane_s16(*(int16_t *) &__reint_767, *(int16x8_t *) &__reint1_767, __p2_767); \
+ __ret_767 = *(float16x8_t *) &__reint2_767; \
+ __ret_767; \
})
#else
-#define vsetq_lane_f16(__p0_263, __p1_263, __p2_263) __extension__ ({ \
- float16_t __s0_263 = __p0_263; \
- float16x8_t __s1_263 = __p1_263; \
- float16x8_t __rev1_263; __rev1_263 = __builtin_shufflevector(__s1_263, __s1_263, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16x8_t __ret_263; \
-float16_t __reint_263 = __s0_263; \
-float16x8_t __reint1_263 = __rev1_263; \
-int16x8_t __reint2_263 = __noswap_vsetq_lane_s16(*(int16_t *) &__reint_263, *(int16x8_t *) &__reint1_263, __p2_263); \
- __ret_263 = *(float16x8_t *) &__reint2_263; \
- __ret_263 = __builtin_shufflevector(__ret_263, __ret_263, 7, 6, 5, 4, 3, 2, 1, 0); \
- __ret_263; \
+#define vsetq_lane_f16(__p0_768, __p1_768, __p2_768) __extension__ ({ \
+ float16_t __s0_768 = __p0_768; \
+ float16x8_t __s1_768 = __p1_768; \
+ float16x8_t __rev1_768; __rev1_768 = __builtin_shufflevector(__s1_768, __s1_768, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __ret_768; \
+float16_t __reint_768 = __s0_768; \
+float16x8_t __reint1_768 = __rev1_768; \
+int16x8_t __reint2_768 = __noswap_vsetq_lane_s16(*(int16_t *) &__reint_768, *(int16x8_t *) &__reint1_768, __p2_768); \
+ __ret_768 = *(float16x8_t *) &__reint2_768; \
+ __ret_768 = __builtin_shufflevector(__ret_768, __ret_768, 7, 6, 5, 4, 3, 2, 1, 0); \
+ __ret_768; \
})
#endif
-#if defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(__aarch64__)
+#if defined(__ARM_FEATURE_BF16_VECTOR_ARITHMETIC)
#ifdef __LITTLE_ENDIAN__
-#define vmulh_lane_f16(__p0_264, __p1_264, __p2_264) __extension__ ({ \
- float16_t __s0_264 = __p0_264; \
- float16x4_t __s1_264 = __p1_264; \
- float16_t __ret_264; \
- __ret_264 = __s0_264 * vget_lane_f16(__s1_264, __p2_264); \
- __ret_264; \
+#define vbfmlalbq_lane_f32(__p0_769, __p1_769, __p2_769, __p3_769) __extension__ ({ \
+ float32x4_t __s0_769 = __p0_769; \
+ bfloat16x8_t __s1_769 = __p1_769; \
+ bfloat16x4_t __s2_769 = __p2_769; \
+ float32x4_t __ret_769; \
+ __ret_769 = vbfmlalbq_f32(__s0_769, __s1_769, (bfloat16x8_t) {vget_lane_bf16(__s2_769, __p3_769), vget_lane_bf16(__s2_769, __p3_769), vget_lane_bf16(__s2_769, __p3_769), vget_lane_bf16(__s2_769, __p3_769), vget_lane_bf16(__s2_769, __p3_769), vget_lane_bf16(__s2_769, __p3_769), vget_lane_bf16(__s2_769, __p3_769), vget_lane_bf16(__s2_769, __p3_769)}); \
+ __ret_769; \
})
#else
-#define vmulh_lane_f16(__p0_265, __p1_265, __p2_265) __extension__ ({ \
- float16_t __s0_265 = __p0_265; \
- float16x4_t __s1_265 = __p1_265; \
- float16x4_t __rev1_265; __rev1_265 = __builtin_shufflevector(__s1_265, __s1_265, 3, 2, 1, 0); \
- float16_t __ret_265; \
- __ret_265 = __s0_265 * __noswap_vget_lane_f16(__rev1_265, __p2_265); \
- __ret_265; \
+#define vbfmlalbq_lane_f32(__p0_770, __p1_770, __p2_770, __p3_770) __extension__ ({ \
+ float32x4_t __s0_770 = __p0_770; \
+ bfloat16x8_t __s1_770 = __p1_770; \
+ bfloat16x4_t __s2_770 = __p2_770; \
+ float32x4_t __rev0_770; __rev0_770 = __builtin_shufflevector(__s0_770, __s0_770, 3, 2, 1, 0); \
+ bfloat16x8_t __rev1_770; __rev1_770 = __builtin_shufflevector(__s1_770, __s1_770, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x4_t __rev2_770; __rev2_770 = __builtin_shufflevector(__s2_770, __s2_770, 3, 2, 1, 0); \
+ float32x4_t __ret_770; \
+ __ret_770 = __noswap_vbfmlalbq_f32(__rev0_770, __rev1_770, (bfloat16x8_t) {__noswap_vget_lane_bf16(__rev2_770, __p3_770), __noswap_vget_lane_bf16(__rev2_770, __p3_770), __noswap_vget_lane_bf16(__rev2_770, __p3_770), __noswap_vget_lane_bf16(__rev2_770, __p3_770), __noswap_vget_lane_bf16(__rev2_770, __p3_770), __noswap_vget_lane_bf16(__rev2_770, __p3_770), __noswap_vget_lane_bf16(__rev2_770, __p3_770), __noswap_vget_lane_bf16(__rev2_770, __p3_770)}); \
+ __ret_770 = __builtin_shufflevector(__ret_770, __ret_770, 3, 2, 1, 0); \
+ __ret_770; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vmulh_laneq_f16(__p0_266, __p1_266, __p2_266) __extension__ ({ \
- float16_t __s0_266 = __p0_266; \
- float16x8_t __s1_266 = __p1_266; \
- float16_t __ret_266; \
- __ret_266 = __s0_266 * vgetq_lane_f16(__s1_266, __p2_266); \
- __ret_266; \
+#define vbfmlalbq_laneq_f32(__p0_771, __p1_771, __p2_771, __p3_771) __extension__ ({ \
+ float32x4_t __s0_771 = __p0_771; \
+ bfloat16x8_t __s1_771 = __p1_771; \
+ bfloat16x8_t __s2_771 = __p2_771; \
+ float32x4_t __ret_771; \
+ __ret_771 = vbfmlalbq_f32(__s0_771, __s1_771, (bfloat16x8_t) {vgetq_lane_bf16(__s2_771, __p3_771), vgetq_lane_bf16(__s2_771, __p3_771), vgetq_lane_bf16(__s2_771, __p3_771), vgetq_lane_bf16(__s2_771, __p3_771), vgetq_lane_bf16(__s2_771, __p3_771), vgetq_lane_bf16(__s2_771, __p3_771), vgetq_lane_bf16(__s2_771, __p3_771), vgetq_lane_bf16(__s2_771, __p3_771)}); \
+ __ret_771; \
})
#else
-#define vmulh_laneq_f16(__p0_267, __p1_267, __p2_267) __extension__ ({ \
- float16_t __s0_267 = __p0_267; \
- float16x8_t __s1_267 = __p1_267; \
- float16x8_t __rev1_267; __rev1_267 = __builtin_shufflevector(__s1_267, __s1_267, 7, 6, 5, 4, 3, 2, 1, 0); \
- float16_t __ret_267; \
- __ret_267 = __s0_267 * __noswap_vgetq_lane_f16(__rev1_267, __p2_267); \
- __ret_267; \
+#define vbfmlalbq_laneq_f32(__p0_772, __p1_772, __p2_772, __p3_772) __extension__ ({ \
+ float32x4_t __s0_772 = __p0_772; \
+ bfloat16x8_t __s1_772 = __p1_772; \
+ bfloat16x8_t __s2_772 = __p2_772; \
+ float32x4_t __rev0_772; __rev0_772 = __builtin_shufflevector(__s0_772, __s0_772, 3, 2, 1, 0); \
+ bfloat16x8_t __rev1_772; __rev1_772 = __builtin_shufflevector(__s1_772, __s1_772, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x8_t __rev2_772; __rev2_772 = __builtin_shufflevector(__s2_772, __s2_772, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float32x4_t __ret_772; \
+ __ret_772 = __noswap_vbfmlalbq_f32(__rev0_772, __rev1_772, (bfloat16x8_t) {__noswap_vgetq_lane_bf16(__rev2_772, __p3_772), __noswap_vgetq_lane_bf16(__rev2_772, __p3_772), __noswap_vgetq_lane_bf16(__rev2_772, __p3_772), __noswap_vgetq_lane_bf16(__rev2_772, __p3_772), __noswap_vgetq_lane_bf16(__rev2_772, __p3_772), __noswap_vgetq_lane_bf16(__rev2_772, __p3_772), __noswap_vgetq_lane_bf16(__rev2_772, __p3_772), __noswap_vgetq_lane_bf16(__rev2_772, __p3_772)}); \
+ __ret_772 = __builtin_shufflevector(__ret_772, __ret_772, 3, 2, 1, 0); \
+ __ret_772; \
})
#endif
+#ifdef __LITTLE_ENDIAN__
+#define vbfmlaltq_lane_f32(__p0_773, __p1_773, __p2_773, __p3_773) __extension__ ({ \
+ float32x4_t __s0_773 = __p0_773; \
+ bfloat16x8_t __s1_773 = __p1_773; \
+ bfloat16x4_t __s2_773 = __p2_773; \
+ float32x4_t __ret_773; \
+ __ret_773 = vbfmlaltq_f32(__s0_773, __s1_773, (bfloat16x8_t) {vget_lane_bf16(__s2_773, __p3_773), vget_lane_bf16(__s2_773, __p3_773), vget_lane_bf16(__s2_773, __p3_773), vget_lane_bf16(__s2_773, __p3_773), vget_lane_bf16(__s2_773, __p3_773), vget_lane_bf16(__s2_773, __p3_773), vget_lane_bf16(__s2_773, __p3_773), vget_lane_bf16(__s2_773, __p3_773)}); \
+ __ret_773; \
+})
+#else
+#define vbfmlaltq_lane_f32(__p0_774, __p1_774, __p2_774, __p3_774) __extension__ ({ \
+ float32x4_t __s0_774 = __p0_774; \
+ bfloat16x8_t __s1_774 = __p1_774; \
+ bfloat16x4_t __s2_774 = __p2_774; \
+ float32x4_t __rev0_774; __rev0_774 = __builtin_shufflevector(__s0_774, __s0_774, 3, 2, 1, 0); \
+ bfloat16x8_t __rev1_774; __rev1_774 = __builtin_shufflevector(__s1_774, __s1_774, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x4_t __rev2_774; __rev2_774 = __builtin_shufflevector(__s2_774, __s2_774, 3, 2, 1, 0); \
+ float32x4_t __ret_774; \
+ __ret_774 = __noswap_vbfmlaltq_f32(__rev0_774, __rev1_774, (bfloat16x8_t) {__noswap_vget_lane_bf16(__rev2_774, __p3_774), __noswap_vget_lane_bf16(__rev2_774, __p3_774), __noswap_vget_lane_bf16(__rev2_774, __p3_774), __noswap_vget_lane_bf16(__rev2_774, __p3_774), __noswap_vget_lane_bf16(__rev2_774, __p3_774), __noswap_vget_lane_bf16(__rev2_774, __p3_774), __noswap_vget_lane_bf16(__rev2_774, __p3_774), __noswap_vget_lane_bf16(__rev2_774, __p3_774)}); \
+ __ret_774 = __builtin_shufflevector(__ret_774, __ret_774, 3, 2, 1, 0); \
+ __ret_774; \
+})
#endif
-#if defined(__ARM_FEATURE_QRDMX) && defined(__aarch64__)
+
#ifdef __LITTLE_ENDIAN__
-__ai int32_t vqrdmlahs_s32(int32_t __p0, int32_t __p1, int32_t __p2) {
- int32_t __ret;
- __ret = vqadds_s32(__p0, vqrdmulhs_s32(__p1, __p2));
+#define vbfmlaltq_laneq_f32(__p0_775, __p1_775, __p2_775, __p3_775) __extension__ ({ \
+ float32x4_t __s0_775 = __p0_775; \
+ bfloat16x8_t __s1_775 = __p1_775; \
+ bfloat16x8_t __s2_775 = __p2_775; \
+ float32x4_t __ret_775; \
+ __ret_775 = vbfmlaltq_f32(__s0_775, __s1_775, (bfloat16x8_t) {vgetq_lane_bf16(__s2_775, __p3_775), vgetq_lane_bf16(__s2_775, __p3_775), vgetq_lane_bf16(__s2_775, __p3_775), vgetq_lane_bf16(__s2_775, __p3_775), vgetq_lane_bf16(__s2_775, __p3_775), vgetq_lane_bf16(__s2_775, __p3_775), vgetq_lane_bf16(__s2_775, __p3_775), vgetq_lane_bf16(__s2_775, __p3_775)}); \
+ __ret_775; \
+})
+#else
+#define vbfmlaltq_laneq_f32(__p0_776, __p1_776, __p2_776, __p3_776) __extension__ ({ \
+ float32x4_t __s0_776 = __p0_776; \
+ bfloat16x8_t __s1_776 = __p1_776; \
+ bfloat16x8_t __s2_776 = __p2_776; \
+ float32x4_t __rev0_776; __rev0_776 = __builtin_shufflevector(__s0_776, __s0_776, 3, 2, 1, 0); \
+ bfloat16x8_t __rev1_776; __rev1_776 = __builtin_shufflevector(__s1_776, __s1_776, 7, 6, 5, 4, 3, 2, 1, 0); \
+ bfloat16x8_t __rev2_776; __rev2_776 = __builtin_shufflevector(__s2_776, __s2_776, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float32x4_t __ret_776; \
+ __ret_776 = __noswap_vbfmlaltq_f32(__rev0_776, __rev1_776, (bfloat16x8_t) {__noswap_vgetq_lane_bf16(__rev2_776, __p3_776), __noswap_vgetq_lane_bf16(__rev2_776, __p3_776), __noswap_vgetq_lane_bf16(__rev2_776, __p3_776), __noswap_vgetq_lane_bf16(__rev2_776, __p3_776), __noswap_vgetq_lane_bf16(__rev2_776, __p3_776), __noswap_vgetq_lane_bf16(__rev2_776, __p3_776), __noswap_vgetq_lane_bf16(__rev2_776, __p3_776), __noswap_vgetq_lane_bf16(__rev2_776, __p3_776)}); \
+ __ret_776 = __builtin_shufflevector(__ret_776, __ret_776, 3, 2, 1, 0); \
+ __ret_776; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+__ai float32x4_t vcvtq_high_f32_bf16(bfloat16x8_t __p0) {
+ float32x4_t __ret;
+ __ret = vcvt_f32_bf16(vget_high_bf16(__p0));
return __ret;
}
#else
-__ai int32_t vqrdmlahs_s32(int32_t __p0, int32_t __p1, int32_t __p2) {
- int32_t __ret;
- __ret = __noswap_vqadds_s32(__p0, __noswap_vqrdmulhs_s32(__p1, __p2));
+__ai float32x4_t vcvtq_high_f32_bf16(bfloat16x8_t __p0) {
+ bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = __noswap_vcvt_f32_bf16(__noswap_vget_high_bf16(__rev0));
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16_t vqrdmlahh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
- int16_t __ret;
- __ret = vqaddh_s16(__p0, vqrdmulhh_s16(__p1, __p2));
+__ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
+ float32x4_t __ret;
+ __ret = vcvt_f32_bf16(vget_low_bf16(__p0));
return __ret;
}
#else
-__ai int16_t vqrdmlahh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
- int16_t __ret;
- __ret = __noswap_vqaddh_s16(__p0, __noswap_vqrdmulhh_s16(__p1, __p2));
+__ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
+ bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
+ float32x4_t __ret;
+ __ret = __noswap_vcvt_f32_bf16(__noswap_vget_low_bf16(__rev0));
+ __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
return __ret;
}
#endif
+#endif
+#if defined(__ARM_FEATURE_FP16FML) && defined(__aarch64__)
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlahs_lane_s32(__p0_268, __p1_268, __p2_268, __p3_268) __extension__ ({ \
- int32_t __s0_268 = __p0_268; \
- int32_t __s1_268 = __p1_268; \
- int32x2_t __s2_268 = __p2_268; \
- int32_t __ret_268; \
- __ret_268 = vqadds_s32(__s0_268, vqrdmulhs_s32(__s1_268, vget_lane_s32(__s2_268, __p3_268))); \
- __ret_268; \
+#define vfmlalq_lane_high_f16(__p0_777, __p1_777, __p2_777, __p3_777) __extension__ ({ \
+ float32x4_t __s0_777 = __p0_777; \
+ float16x8_t __s1_777 = __p1_777; \
+ float16x4_t __s2_777 = __p2_777; \
+ float32x4_t __ret_777; \
+ __ret_777 = vfmlalq_high_f16(__s0_777, __s1_777, (float16x8_t) {vget_lane_f16(__s2_777, __p3_777), vget_lane_f16(__s2_777, __p3_777), vget_lane_f16(__s2_777, __p3_777), vget_lane_f16(__s2_777, __p3_777), vget_lane_f16(__s2_777, __p3_777), vget_lane_f16(__s2_777, __p3_777), vget_lane_f16(__s2_777, __p3_777), vget_lane_f16(__s2_777, __p3_777)}); \
+ __ret_777; \
})
#else
-#define vqrdmlahs_lane_s32(__p0_269, __p1_269, __p2_269, __p3_269) __extension__ ({ \
- int32_t __s0_269 = __p0_269; \
- int32_t __s1_269 = __p1_269; \
- int32x2_t __s2_269 = __p2_269; \
- int32x2_t __rev2_269; __rev2_269 = __builtin_shufflevector(__s2_269, __s2_269, 1, 0); \
- int32_t __ret_269; \
- __ret_269 = __noswap_vqadds_s32(__s0_269, __noswap_vqrdmulhs_s32(__s1_269, __noswap_vget_lane_s32(__rev2_269, __p3_269))); \
- __ret_269; \
+#define vfmlalq_lane_high_f16(__p0_778, __p1_778, __p2_778, __p3_778) __extension__ ({ \
+ float32x4_t __s0_778 = __p0_778; \
+ float16x8_t __s1_778 = __p1_778; \
+ float16x4_t __s2_778 = __p2_778; \
+ float32x4_t __rev0_778; __rev0_778 = __builtin_shufflevector(__s0_778, __s0_778, 3, 2, 1, 0); \
+ float16x8_t __rev1_778; __rev1_778 = __builtin_shufflevector(__s1_778, __s1_778, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x4_t __rev2_778; __rev2_778 = __builtin_shufflevector(__s2_778, __s2_778, 3, 2, 1, 0); \
+ float32x4_t __ret_778; \
+ __ret_778 = __noswap_vfmlalq_high_f16(__rev0_778, __rev1_778, (float16x8_t) {__noswap_vget_lane_f16(__rev2_778, __p3_778), __noswap_vget_lane_f16(__rev2_778, __p3_778), __noswap_vget_lane_f16(__rev2_778, __p3_778), __noswap_vget_lane_f16(__rev2_778, __p3_778), __noswap_vget_lane_f16(__rev2_778, __p3_778), __noswap_vget_lane_f16(__rev2_778, __p3_778), __noswap_vget_lane_f16(__rev2_778, __p3_778), __noswap_vget_lane_f16(__rev2_778, __p3_778)}); \
+ __ret_778 = __builtin_shufflevector(__ret_778, __ret_778, 3, 2, 1, 0); \
+ __ret_778; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlahh_lane_s16(__p0_270, __p1_270, __p2_270, __p3_270) __extension__ ({ \
- int16_t __s0_270 = __p0_270; \
- int16_t __s1_270 = __p1_270; \
- int16x4_t __s2_270 = __p2_270; \
- int16_t __ret_270; \
- __ret_270 = vqaddh_s16(__s0_270, vqrdmulhh_s16(__s1_270, vget_lane_s16(__s2_270, __p3_270))); \
- __ret_270; \
+#define vfmlal_lane_high_f16(__p0_779, __p1_779, __p2_779, __p3_779) __extension__ ({ \
+ float32x2_t __s0_779 = __p0_779; \
+ float16x4_t __s1_779 = __p1_779; \
+ float16x4_t __s2_779 = __p2_779; \
+ float32x2_t __ret_779; \
+ __ret_779 = vfmlal_high_f16(__s0_779, __s1_779, (float16x4_t) {vget_lane_f16(__s2_779, __p3_779), vget_lane_f16(__s2_779, __p3_779), vget_lane_f16(__s2_779, __p3_779), vget_lane_f16(__s2_779, __p3_779)}); \
+ __ret_779; \
})
#else
-#define vqrdmlahh_lane_s16(__p0_271, __p1_271, __p2_271, __p3_271) __extension__ ({ \
- int16_t __s0_271 = __p0_271; \
- int16_t __s1_271 = __p1_271; \
- int16x4_t __s2_271 = __p2_271; \
- int16x4_t __rev2_271; __rev2_271 = __builtin_shufflevector(__s2_271, __s2_271, 3, 2, 1, 0); \
- int16_t __ret_271; \
- __ret_271 = __noswap_vqaddh_s16(__s0_271, __noswap_vqrdmulhh_s16(__s1_271, __noswap_vget_lane_s16(__rev2_271, __p3_271))); \
- __ret_271; \
+#define vfmlal_lane_high_f16(__p0_780, __p1_780, __p2_780, __p3_780) __extension__ ({ \
+ float32x2_t __s0_780 = __p0_780; \
+ float16x4_t __s1_780 = __p1_780; \
+ float16x4_t __s2_780 = __p2_780; \
+ float32x2_t __rev0_780; __rev0_780 = __builtin_shufflevector(__s0_780, __s0_780, 1, 0); \
+ float16x4_t __rev1_780; __rev1_780 = __builtin_shufflevector(__s1_780, __s1_780, 3, 2, 1, 0); \
+ float16x4_t __rev2_780; __rev2_780 = __builtin_shufflevector(__s2_780, __s2_780, 3, 2, 1, 0); \
+ float32x2_t __ret_780; \
+ __ret_780 = __noswap_vfmlal_high_f16(__rev0_780, __rev1_780, (float16x4_t) {__noswap_vget_lane_f16(__rev2_780, __p3_780), __noswap_vget_lane_f16(__rev2_780, __p3_780), __noswap_vget_lane_f16(__rev2_780, __p3_780), __noswap_vget_lane_f16(__rev2_780, __p3_780)}); \
+ __ret_780 = __builtin_shufflevector(__ret_780, __ret_780, 1, 0); \
+ __ret_780; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlahs_laneq_s32(__p0_272, __p1_272, __p2_272, __p3_272) __extension__ ({ \
- int32_t __s0_272 = __p0_272; \
- int32_t __s1_272 = __p1_272; \
- int32x4_t __s2_272 = __p2_272; \
- int32_t __ret_272; \
- __ret_272 = vqadds_s32(__s0_272, vqrdmulhs_s32(__s1_272, vgetq_lane_s32(__s2_272, __p3_272))); \
- __ret_272; \
+#define vfmlalq_lane_low_f16(__p0_781, __p1_781, __p2_781, __p3_781) __extension__ ({ \
+ float32x4_t __s0_781 = __p0_781; \
+ float16x8_t __s1_781 = __p1_781; \
+ float16x4_t __s2_781 = __p2_781; \
+ float32x4_t __ret_781; \
+ __ret_781 = vfmlalq_low_f16(__s0_781, __s1_781, (float16x8_t) {vget_lane_f16(__s2_781, __p3_781), vget_lane_f16(__s2_781, __p3_781), vget_lane_f16(__s2_781, __p3_781), vget_lane_f16(__s2_781, __p3_781), vget_lane_f16(__s2_781, __p3_781), vget_lane_f16(__s2_781, __p3_781), vget_lane_f16(__s2_781, __p3_781), vget_lane_f16(__s2_781, __p3_781)}); \
+ __ret_781; \
})
#else
-#define vqrdmlahs_laneq_s32(__p0_273, __p1_273, __p2_273, __p3_273) __extension__ ({ \
- int32_t __s0_273 = __p0_273; \
- int32_t __s1_273 = __p1_273; \
- int32x4_t __s2_273 = __p2_273; \
- int32x4_t __rev2_273; __rev2_273 = __builtin_shufflevector(__s2_273, __s2_273, 3, 2, 1, 0); \
- int32_t __ret_273; \
- __ret_273 = __noswap_vqadds_s32(__s0_273, __noswap_vqrdmulhs_s32(__s1_273, __noswap_vgetq_lane_s32(__rev2_273, __p3_273))); \
- __ret_273; \
+#define vfmlalq_lane_low_f16(__p0_782, __p1_782, __p2_782, __p3_782) __extension__ ({ \
+ float32x4_t __s0_782 = __p0_782; \
+ float16x8_t __s1_782 = __p1_782; \
+ float16x4_t __s2_782 = __p2_782; \
+ float32x4_t __rev0_782; __rev0_782 = __builtin_shufflevector(__s0_782, __s0_782, 3, 2, 1, 0); \
+ float16x8_t __rev1_782; __rev1_782 = __builtin_shufflevector(__s1_782, __s1_782, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x4_t __rev2_782; __rev2_782 = __builtin_shufflevector(__s2_782, __s2_782, 3, 2, 1, 0); \
+ float32x4_t __ret_782; \
+ __ret_782 = __noswap_vfmlalq_low_f16(__rev0_782, __rev1_782, (float16x8_t) {__noswap_vget_lane_f16(__rev2_782, __p3_782), __noswap_vget_lane_f16(__rev2_782, __p3_782), __noswap_vget_lane_f16(__rev2_782, __p3_782), __noswap_vget_lane_f16(__rev2_782, __p3_782), __noswap_vget_lane_f16(__rev2_782, __p3_782), __noswap_vget_lane_f16(__rev2_782, __p3_782), __noswap_vget_lane_f16(__rev2_782, __p3_782), __noswap_vget_lane_f16(__rev2_782, __p3_782)}); \
+ __ret_782 = __builtin_shufflevector(__ret_782, __ret_782, 3, 2, 1, 0); \
+ __ret_782; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlahh_laneq_s16(__p0_274, __p1_274, __p2_274, __p3_274) __extension__ ({ \
- int16_t __s0_274 = __p0_274; \
- int16_t __s1_274 = __p1_274; \
- int16x8_t __s2_274 = __p2_274; \
- int16_t __ret_274; \
- __ret_274 = vqaddh_s16(__s0_274, vqrdmulhh_s16(__s1_274, vgetq_lane_s16(__s2_274, __p3_274))); \
- __ret_274; \
+#define vfmlal_lane_low_f16(__p0_783, __p1_783, __p2_783, __p3_783) __extension__ ({ \
+ float32x2_t __s0_783 = __p0_783; \
+ float16x4_t __s1_783 = __p1_783; \
+ float16x4_t __s2_783 = __p2_783; \
+ float32x2_t __ret_783; \
+ __ret_783 = vfmlal_low_f16(__s0_783, __s1_783, (float16x4_t) {vget_lane_f16(__s2_783, __p3_783), vget_lane_f16(__s2_783, __p3_783), vget_lane_f16(__s2_783, __p3_783), vget_lane_f16(__s2_783, __p3_783)}); \
+ __ret_783; \
})
#else
-#define vqrdmlahh_laneq_s16(__p0_275, __p1_275, __p2_275, __p3_275) __extension__ ({ \
- int16_t __s0_275 = __p0_275; \
- int16_t __s1_275 = __p1_275; \
- int16x8_t __s2_275 = __p2_275; \
- int16x8_t __rev2_275; __rev2_275 = __builtin_shufflevector(__s2_275, __s2_275, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16_t __ret_275; \
- __ret_275 = __noswap_vqaddh_s16(__s0_275, __noswap_vqrdmulhh_s16(__s1_275, __noswap_vgetq_lane_s16(__rev2_275, __p3_275))); \
- __ret_275; \
+#define vfmlal_lane_low_f16(__p0_784, __p1_784, __p2_784, __p3_784) __extension__ ({ \
+ float32x2_t __s0_784 = __p0_784; \
+ float16x4_t __s1_784 = __p1_784; \
+ float16x4_t __s2_784 = __p2_784; \
+ float32x2_t __rev0_784; __rev0_784 = __builtin_shufflevector(__s0_784, __s0_784, 1, 0); \
+ float16x4_t __rev1_784; __rev1_784 = __builtin_shufflevector(__s1_784, __s1_784, 3, 2, 1, 0); \
+ float16x4_t __rev2_784; __rev2_784 = __builtin_shufflevector(__s2_784, __s2_784, 3, 2, 1, 0); \
+ float32x2_t __ret_784; \
+ __ret_784 = __noswap_vfmlal_low_f16(__rev0_784, __rev1_784, (float16x4_t) {__noswap_vget_lane_f16(__rev2_784, __p3_784), __noswap_vget_lane_f16(__rev2_784, __p3_784), __noswap_vget_lane_f16(__rev2_784, __p3_784), __noswap_vget_lane_f16(__rev2_784, __p3_784)}); \
+ __ret_784 = __builtin_shufflevector(__ret_784, __ret_784, 1, 0); \
+ __ret_784; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int32_t vqrdmlshs_s32(int32_t __p0, int32_t __p1, int32_t __p2) {
+#define vfmlalq_laneq_high_f16(__p0_785, __p1_785, __p2_785, __p3_785) __extension__ ({ \
+ float32x4_t __s0_785 = __p0_785; \
+ float16x8_t __s1_785 = __p1_785; \
+ float16x8_t __s2_785 = __p2_785; \
+ float32x4_t __ret_785; \
+ __ret_785 = vfmlalq_high_f16(__s0_785, __s1_785, (float16x8_t) {vgetq_lane_f16(__s2_785, __p3_785), vgetq_lane_f16(__s2_785, __p3_785), vgetq_lane_f16(__s2_785, __p3_785), vgetq_lane_f16(__s2_785, __p3_785), vgetq_lane_f16(__s2_785, __p3_785), vgetq_lane_f16(__s2_785, __p3_785), vgetq_lane_f16(__s2_785, __p3_785), vgetq_lane_f16(__s2_785, __p3_785)}); \
+ __ret_785; \
+})
+#else
+#define vfmlalq_laneq_high_f16(__p0_786, __p1_786, __p2_786, __p3_786) __extension__ ({ \
+ float32x4_t __s0_786 = __p0_786; \
+ float16x8_t __s1_786 = __p1_786; \
+ float16x8_t __s2_786 = __p2_786; \
+ float32x4_t __rev0_786; __rev0_786 = __builtin_shufflevector(__s0_786, __s0_786, 3, 2, 1, 0); \
+ float16x8_t __rev1_786; __rev1_786 = __builtin_shufflevector(__s1_786, __s1_786, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __rev2_786; __rev2_786 = __builtin_shufflevector(__s2_786, __s2_786, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float32x4_t __ret_786; \
+ __ret_786 = __noswap_vfmlalq_high_f16(__rev0_786, __rev1_786, (float16x8_t) {__noswap_vgetq_lane_f16(__rev2_786, __p3_786), __noswap_vgetq_lane_f16(__rev2_786, __p3_786), __noswap_vgetq_lane_f16(__rev2_786, __p3_786), __noswap_vgetq_lane_f16(__rev2_786, __p3_786), __noswap_vgetq_lane_f16(__rev2_786, __p3_786), __noswap_vgetq_lane_f16(__rev2_786, __p3_786), __noswap_vgetq_lane_f16(__rev2_786, __p3_786), __noswap_vgetq_lane_f16(__rev2_786, __p3_786)}); \
+ __ret_786 = __builtin_shufflevector(__ret_786, __ret_786, 3, 2, 1, 0); \
+ __ret_786; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vfmlal_laneq_high_f16(__p0_787, __p1_787, __p2_787, __p3_787) __extension__ ({ \
+ float32x2_t __s0_787 = __p0_787; \
+ float16x4_t __s1_787 = __p1_787; \
+ float16x8_t __s2_787 = __p2_787; \
+ float32x2_t __ret_787; \
+ __ret_787 = vfmlal_high_f16(__s0_787, __s1_787, (float16x4_t) {vgetq_lane_f16(__s2_787, __p3_787), vgetq_lane_f16(__s2_787, __p3_787), vgetq_lane_f16(__s2_787, __p3_787), vgetq_lane_f16(__s2_787, __p3_787)}); \
+ __ret_787; \
+})
+#else
+#define vfmlal_laneq_high_f16(__p0_788, __p1_788, __p2_788, __p3_788) __extension__ ({ \
+ float32x2_t __s0_788 = __p0_788; \
+ float16x4_t __s1_788 = __p1_788; \
+ float16x8_t __s2_788 = __p2_788; \
+ float32x2_t __rev0_788; __rev0_788 = __builtin_shufflevector(__s0_788, __s0_788, 1, 0); \
+ float16x4_t __rev1_788; __rev1_788 = __builtin_shufflevector(__s1_788, __s1_788, 3, 2, 1, 0); \
+ float16x8_t __rev2_788; __rev2_788 = __builtin_shufflevector(__s2_788, __s2_788, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float32x2_t __ret_788; \
+ __ret_788 = __noswap_vfmlal_high_f16(__rev0_788, __rev1_788, (float16x4_t) {__noswap_vgetq_lane_f16(__rev2_788, __p3_788), __noswap_vgetq_lane_f16(__rev2_788, __p3_788), __noswap_vgetq_lane_f16(__rev2_788, __p3_788), __noswap_vgetq_lane_f16(__rev2_788, __p3_788)}); \
+ __ret_788 = __builtin_shufflevector(__ret_788, __ret_788, 1, 0); \
+ __ret_788; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vfmlalq_laneq_low_f16(__p0_789, __p1_789, __p2_789, __p3_789) __extension__ ({ \
+ float32x4_t __s0_789 = __p0_789; \
+ float16x8_t __s1_789 = __p1_789; \
+ float16x8_t __s2_789 = __p2_789; \
+ float32x4_t __ret_789; \
+ __ret_789 = vfmlalq_low_f16(__s0_789, __s1_789, (float16x8_t) {vgetq_lane_f16(__s2_789, __p3_789), vgetq_lane_f16(__s2_789, __p3_789), vgetq_lane_f16(__s2_789, __p3_789), vgetq_lane_f16(__s2_789, __p3_789), vgetq_lane_f16(__s2_789, __p3_789), vgetq_lane_f16(__s2_789, __p3_789), vgetq_lane_f16(__s2_789, __p3_789), vgetq_lane_f16(__s2_789, __p3_789)}); \
+ __ret_789; \
+})
+#else
+#define vfmlalq_laneq_low_f16(__p0_790, __p1_790, __p2_790, __p3_790) __extension__ ({ \
+ float32x4_t __s0_790 = __p0_790; \
+ float16x8_t __s1_790 = __p1_790; \
+ float16x8_t __s2_790 = __p2_790; \
+ float32x4_t __rev0_790; __rev0_790 = __builtin_shufflevector(__s0_790, __s0_790, 3, 2, 1, 0); \
+ float16x8_t __rev1_790; __rev1_790 = __builtin_shufflevector(__s1_790, __s1_790, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __rev2_790; __rev2_790 = __builtin_shufflevector(__s2_790, __s2_790, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float32x4_t __ret_790; \
+ __ret_790 = __noswap_vfmlalq_low_f16(__rev0_790, __rev1_790, (float16x8_t) {__noswap_vgetq_lane_f16(__rev2_790, __p3_790), __noswap_vgetq_lane_f16(__rev2_790, __p3_790), __noswap_vgetq_lane_f16(__rev2_790, __p3_790), __noswap_vgetq_lane_f16(__rev2_790, __p3_790), __noswap_vgetq_lane_f16(__rev2_790, __p3_790), __noswap_vgetq_lane_f16(__rev2_790, __p3_790), __noswap_vgetq_lane_f16(__rev2_790, __p3_790), __noswap_vgetq_lane_f16(__rev2_790, __p3_790)}); \
+ __ret_790 = __builtin_shufflevector(__ret_790, __ret_790, 3, 2, 1, 0); \
+ __ret_790; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vfmlal_laneq_low_f16(__p0_791, __p1_791, __p2_791, __p3_791) __extension__ ({ \
+ float32x2_t __s0_791 = __p0_791; \
+ float16x4_t __s1_791 = __p1_791; \
+ float16x8_t __s2_791 = __p2_791; \
+ float32x2_t __ret_791; \
+ __ret_791 = vfmlal_low_f16(__s0_791, __s1_791, (float16x4_t) {vgetq_lane_f16(__s2_791, __p3_791), vgetq_lane_f16(__s2_791, __p3_791), vgetq_lane_f16(__s2_791, __p3_791), vgetq_lane_f16(__s2_791, __p3_791)}); \
+ __ret_791; \
+})
+#else
+#define vfmlal_laneq_low_f16(__p0_792, __p1_792, __p2_792, __p3_792) __extension__ ({ \
+ float32x2_t __s0_792 = __p0_792; \
+ float16x4_t __s1_792 = __p1_792; \
+ float16x8_t __s2_792 = __p2_792; \
+ float32x2_t __rev0_792; __rev0_792 = __builtin_shufflevector(__s0_792, __s0_792, 1, 0); \
+ float16x4_t __rev1_792; __rev1_792 = __builtin_shufflevector(__s1_792, __s1_792, 3, 2, 1, 0); \
+ float16x8_t __rev2_792; __rev2_792 = __builtin_shufflevector(__s2_792, __s2_792, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float32x2_t __ret_792; \
+ __ret_792 = __noswap_vfmlal_low_f16(__rev0_792, __rev1_792, (float16x4_t) {__noswap_vgetq_lane_f16(__rev2_792, __p3_792), __noswap_vgetq_lane_f16(__rev2_792, __p3_792), __noswap_vgetq_lane_f16(__rev2_792, __p3_792), __noswap_vgetq_lane_f16(__rev2_792, __p3_792)}); \
+ __ret_792 = __builtin_shufflevector(__ret_792, __ret_792, 1, 0); \
+ __ret_792; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vfmlslq_lane_high_f16(__p0_793, __p1_793, __p2_793, __p3_793) __extension__ ({ \
+ float32x4_t __s0_793 = __p0_793; \
+ float16x8_t __s1_793 = __p1_793; \
+ float16x4_t __s2_793 = __p2_793; \
+ float32x4_t __ret_793; \
+ __ret_793 = vfmlslq_high_f16(__s0_793, __s1_793, (float16x8_t) {vget_lane_f16(__s2_793, __p3_793), vget_lane_f16(__s2_793, __p3_793), vget_lane_f16(__s2_793, __p3_793), vget_lane_f16(__s2_793, __p3_793), vget_lane_f16(__s2_793, __p3_793), vget_lane_f16(__s2_793, __p3_793), vget_lane_f16(__s2_793, __p3_793), vget_lane_f16(__s2_793, __p3_793)}); \
+ __ret_793; \
+})
+#else
+#define vfmlslq_lane_high_f16(__p0_794, __p1_794, __p2_794, __p3_794) __extension__ ({ \
+ float32x4_t __s0_794 = __p0_794; \
+ float16x8_t __s1_794 = __p1_794; \
+ float16x4_t __s2_794 = __p2_794; \
+ float32x4_t __rev0_794; __rev0_794 = __builtin_shufflevector(__s0_794, __s0_794, 3, 2, 1, 0); \
+ float16x8_t __rev1_794; __rev1_794 = __builtin_shufflevector(__s1_794, __s1_794, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x4_t __rev2_794; __rev2_794 = __builtin_shufflevector(__s2_794, __s2_794, 3, 2, 1, 0); \
+ float32x4_t __ret_794; \
+ __ret_794 = __noswap_vfmlslq_high_f16(__rev0_794, __rev1_794, (float16x8_t) {__noswap_vget_lane_f16(__rev2_794, __p3_794), __noswap_vget_lane_f16(__rev2_794, __p3_794), __noswap_vget_lane_f16(__rev2_794, __p3_794), __noswap_vget_lane_f16(__rev2_794, __p3_794), __noswap_vget_lane_f16(__rev2_794, __p3_794), __noswap_vget_lane_f16(__rev2_794, __p3_794), __noswap_vget_lane_f16(__rev2_794, __p3_794), __noswap_vget_lane_f16(__rev2_794, __p3_794)}); \
+ __ret_794 = __builtin_shufflevector(__ret_794, __ret_794, 3, 2, 1, 0); \
+ __ret_794; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vfmlsl_lane_high_f16(__p0_795, __p1_795, __p2_795, __p3_795) __extension__ ({ \
+ float32x2_t __s0_795 = __p0_795; \
+ float16x4_t __s1_795 = __p1_795; \
+ float16x4_t __s2_795 = __p2_795; \
+ float32x2_t __ret_795; \
+ __ret_795 = vfmlsl_high_f16(__s0_795, __s1_795, (float16x4_t) {vget_lane_f16(__s2_795, __p3_795), vget_lane_f16(__s2_795, __p3_795), vget_lane_f16(__s2_795, __p3_795), vget_lane_f16(__s2_795, __p3_795)}); \
+ __ret_795; \
+})
+#else
+#define vfmlsl_lane_high_f16(__p0_796, __p1_796, __p2_796, __p3_796) __extension__ ({ \
+ float32x2_t __s0_796 = __p0_796; \
+ float16x4_t __s1_796 = __p1_796; \
+ float16x4_t __s2_796 = __p2_796; \
+ float32x2_t __rev0_796; __rev0_796 = __builtin_shufflevector(__s0_796, __s0_796, 1, 0); \
+ float16x4_t __rev1_796; __rev1_796 = __builtin_shufflevector(__s1_796, __s1_796, 3, 2, 1, 0); \
+ float16x4_t __rev2_796; __rev2_796 = __builtin_shufflevector(__s2_796, __s2_796, 3, 2, 1, 0); \
+ float32x2_t __ret_796; \
+ __ret_796 = __noswap_vfmlsl_high_f16(__rev0_796, __rev1_796, (float16x4_t) {__noswap_vget_lane_f16(__rev2_796, __p3_796), __noswap_vget_lane_f16(__rev2_796, __p3_796), __noswap_vget_lane_f16(__rev2_796, __p3_796), __noswap_vget_lane_f16(__rev2_796, __p3_796)}); \
+ __ret_796 = __builtin_shufflevector(__ret_796, __ret_796, 1, 0); \
+ __ret_796; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vfmlslq_lane_low_f16(__p0_797, __p1_797, __p2_797, __p3_797) __extension__ ({ \
+ float32x4_t __s0_797 = __p0_797; \
+ float16x8_t __s1_797 = __p1_797; \
+ float16x4_t __s2_797 = __p2_797; \
+ float32x4_t __ret_797; \
+ __ret_797 = vfmlslq_low_f16(__s0_797, __s1_797, (float16x8_t) {vget_lane_f16(__s2_797, __p3_797), vget_lane_f16(__s2_797, __p3_797), vget_lane_f16(__s2_797, __p3_797), vget_lane_f16(__s2_797, __p3_797), vget_lane_f16(__s2_797, __p3_797), vget_lane_f16(__s2_797, __p3_797), vget_lane_f16(__s2_797, __p3_797), vget_lane_f16(__s2_797, __p3_797)}); \
+ __ret_797; \
+})
+#else
+#define vfmlslq_lane_low_f16(__p0_798, __p1_798, __p2_798, __p3_798) __extension__ ({ \
+ float32x4_t __s0_798 = __p0_798; \
+ float16x8_t __s1_798 = __p1_798; \
+ float16x4_t __s2_798 = __p2_798; \
+ float32x4_t __rev0_798; __rev0_798 = __builtin_shufflevector(__s0_798, __s0_798, 3, 2, 1, 0); \
+ float16x8_t __rev1_798; __rev1_798 = __builtin_shufflevector(__s1_798, __s1_798, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x4_t __rev2_798; __rev2_798 = __builtin_shufflevector(__s2_798, __s2_798, 3, 2, 1, 0); \
+ float32x4_t __ret_798; \
+ __ret_798 = __noswap_vfmlslq_low_f16(__rev0_798, __rev1_798, (float16x8_t) {__noswap_vget_lane_f16(__rev2_798, __p3_798), __noswap_vget_lane_f16(__rev2_798, __p3_798), __noswap_vget_lane_f16(__rev2_798, __p3_798), __noswap_vget_lane_f16(__rev2_798, __p3_798), __noswap_vget_lane_f16(__rev2_798, __p3_798), __noswap_vget_lane_f16(__rev2_798, __p3_798), __noswap_vget_lane_f16(__rev2_798, __p3_798), __noswap_vget_lane_f16(__rev2_798, __p3_798)}); \
+ __ret_798 = __builtin_shufflevector(__ret_798, __ret_798, 3, 2, 1, 0); \
+ __ret_798; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vfmlsl_lane_low_f16(__p0_799, __p1_799, __p2_799, __p3_799) __extension__ ({ \
+ float32x2_t __s0_799 = __p0_799; \
+ float16x4_t __s1_799 = __p1_799; \
+ float16x4_t __s2_799 = __p2_799; \
+ float32x2_t __ret_799; \
+ __ret_799 = vfmlsl_low_f16(__s0_799, __s1_799, (float16x4_t) {vget_lane_f16(__s2_799, __p3_799), vget_lane_f16(__s2_799, __p3_799), vget_lane_f16(__s2_799, __p3_799), vget_lane_f16(__s2_799, __p3_799)}); \
+ __ret_799; \
+})
+#else
+#define vfmlsl_lane_low_f16(__p0_800, __p1_800, __p2_800, __p3_800) __extension__ ({ \
+ float32x2_t __s0_800 = __p0_800; \
+ float16x4_t __s1_800 = __p1_800; \
+ float16x4_t __s2_800 = __p2_800; \
+ float32x2_t __rev0_800; __rev0_800 = __builtin_shufflevector(__s0_800, __s0_800, 1, 0); \
+ float16x4_t __rev1_800; __rev1_800 = __builtin_shufflevector(__s1_800, __s1_800, 3, 2, 1, 0); \
+ float16x4_t __rev2_800; __rev2_800 = __builtin_shufflevector(__s2_800, __s2_800, 3, 2, 1, 0); \
+ float32x2_t __ret_800; \
+ __ret_800 = __noswap_vfmlsl_low_f16(__rev0_800, __rev1_800, (float16x4_t) {__noswap_vget_lane_f16(__rev2_800, __p3_800), __noswap_vget_lane_f16(__rev2_800, __p3_800), __noswap_vget_lane_f16(__rev2_800, __p3_800), __noswap_vget_lane_f16(__rev2_800, __p3_800)}); \
+ __ret_800 = __builtin_shufflevector(__ret_800, __ret_800, 1, 0); \
+ __ret_800; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vfmlslq_laneq_high_f16(__p0_801, __p1_801, __p2_801, __p3_801) __extension__ ({ \
+ float32x4_t __s0_801 = __p0_801; \
+ float16x8_t __s1_801 = __p1_801; \
+ float16x8_t __s2_801 = __p2_801; \
+ float32x4_t __ret_801; \
+ __ret_801 = vfmlslq_high_f16(__s0_801, __s1_801, (float16x8_t) {vgetq_lane_f16(__s2_801, __p3_801), vgetq_lane_f16(__s2_801, __p3_801), vgetq_lane_f16(__s2_801, __p3_801), vgetq_lane_f16(__s2_801, __p3_801), vgetq_lane_f16(__s2_801, __p3_801), vgetq_lane_f16(__s2_801, __p3_801), vgetq_lane_f16(__s2_801, __p3_801), vgetq_lane_f16(__s2_801, __p3_801)}); \
+ __ret_801; \
+})
+#else
+#define vfmlslq_laneq_high_f16(__p0_802, __p1_802, __p2_802, __p3_802) __extension__ ({ \
+ float32x4_t __s0_802 = __p0_802; \
+ float16x8_t __s1_802 = __p1_802; \
+ float16x8_t __s2_802 = __p2_802; \
+ float32x4_t __rev0_802; __rev0_802 = __builtin_shufflevector(__s0_802, __s0_802, 3, 2, 1, 0); \
+ float16x8_t __rev1_802; __rev1_802 = __builtin_shufflevector(__s1_802, __s1_802, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __rev2_802; __rev2_802 = __builtin_shufflevector(__s2_802, __s2_802, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float32x4_t __ret_802; \
+ __ret_802 = __noswap_vfmlslq_high_f16(__rev0_802, __rev1_802, (float16x8_t) {__noswap_vgetq_lane_f16(__rev2_802, __p3_802), __noswap_vgetq_lane_f16(__rev2_802, __p3_802), __noswap_vgetq_lane_f16(__rev2_802, __p3_802), __noswap_vgetq_lane_f16(__rev2_802, __p3_802), __noswap_vgetq_lane_f16(__rev2_802, __p3_802), __noswap_vgetq_lane_f16(__rev2_802, __p3_802), __noswap_vgetq_lane_f16(__rev2_802, __p3_802), __noswap_vgetq_lane_f16(__rev2_802, __p3_802)}); \
+ __ret_802 = __builtin_shufflevector(__ret_802, __ret_802, 3, 2, 1, 0); \
+ __ret_802; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vfmlsl_laneq_high_f16(__p0_803, __p1_803, __p2_803, __p3_803) __extension__ ({ \
+ float32x2_t __s0_803 = __p0_803; \
+ float16x4_t __s1_803 = __p1_803; \
+ float16x8_t __s2_803 = __p2_803; \
+ float32x2_t __ret_803; \
+ __ret_803 = vfmlsl_high_f16(__s0_803, __s1_803, (float16x4_t) {vgetq_lane_f16(__s2_803, __p3_803), vgetq_lane_f16(__s2_803, __p3_803), vgetq_lane_f16(__s2_803, __p3_803), vgetq_lane_f16(__s2_803, __p3_803)}); \
+ __ret_803; \
+})
+#else
+#define vfmlsl_laneq_high_f16(__p0_804, __p1_804, __p2_804, __p3_804) __extension__ ({ \
+ float32x2_t __s0_804 = __p0_804; \
+ float16x4_t __s1_804 = __p1_804; \
+ float16x8_t __s2_804 = __p2_804; \
+ float32x2_t __rev0_804; __rev0_804 = __builtin_shufflevector(__s0_804, __s0_804, 1, 0); \
+ float16x4_t __rev1_804; __rev1_804 = __builtin_shufflevector(__s1_804, __s1_804, 3, 2, 1, 0); \
+ float16x8_t __rev2_804; __rev2_804 = __builtin_shufflevector(__s2_804, __s2_804, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float32x2_t __ret_804; \
+ __ret_804 = __noswap_vfmlsl_high_f16(__rev0_804, __rev1_804, (float16x4_t) {__noswap_vgetq_lane_f16(__rev2_804, __p3_804), __noswap_vgetq_lane_f16(__rev2_804, __p3_804), __noswap_vgetq_lane_f16(__rev2_804, __p3_804), __noswap_vgetq_lane_f16(__rev2_804, __p3_804)}); \
+ __ret_804 = __builtin_shufflevector(__ret_804, __ret_804, 1, 0); \
+ __ret_804; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vfmlslq_laneq_low_f16(__p0_805, __p1_805, __p2_805, __p3_805) __extension__ ({ \
+ float32x4_t __s0_805 = __p0_805; \
+ float16x8_t __s1_805 = __p1_805; \
+ float16x8_t __s2_805 = __p2_805; \
+ float32x4_t __ret_805; \
+ __ret_805 = vfmlslq_low_f16(__s0_805, __s1_805, (float16x8_t) {vgetq_lane_f16(__s2_805, __p3_805), vgetq_lane_f16(__s2_805, __p3_805), vgetq_lane_f16(__s2_805, __p3_805), vgetq_lane_f16(__s2_805, __p3_805), vgetq_lane_f16(__s2_805, __p3_805), vgetq_lane_f16(__s2_805, __p3_805), vgetq_lane_f16(__s2_805, __p3_805), vgetq_lane_f16(__s2_805, __p3_805)}); \
+ __ret_805; \
+})
+#else
+#define vfmlslq_laneq_low_f16(__p0_806, __p1_806, __p2_806, __p3_806) __extension__ ({ \
+ float32x4_t __s0_806 = __p0_806; \
+ float16x8_t __s1_806 = __p1_806; \
+ float16x8_t __s2_806 = __p2_806; \
+ float32x4_t __rev0_806; __rev0_806 = __builtin_shufflevector(__s0_806, __s0_806, 3, 2, 1, 0); \
+ float16x8_t __rev1_806; __rev1_806 = __builtin_shufflevector(__s1_806, __s1_806, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16x8_t __rev2_806; __rev2_806 = __builtin_shufflevector(__s2_806, __s2_806, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float32x4_t __ret_806; \
+ __ret_806 = __noswap_vfmlslq_low_f16(__rev0_806, __rev1_806, (float16x8_t) {__noswap_vgetq_lane_f16(__rev2_806, __p3_806), __noswap_vgetq_lane_f16(__rev2_806, __p3_806), __noswap_vgetq_lane_f16(__rev2_806, __p3_806), __noswap_vgetq_lane_f16(__rev2_806, __p3_806), __noswap_vgetq_lane_f16(__rev2_806, __p3_806), __noswap_vgetq_lane_f16(__rev2_806, __p3_806), __noswap_vgetq_lane_f16(__rev2_806, __p3_806), __noswap_vgetq_lane_f16(__rev2_806, __p3_806)}); \
+ __ret_806 = __builtin_shufflevector(__ret_806, __ret_806, 3, 2, 1, 0); \
+ __ret_806; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vfmlsl_laneq_low_f16(__p0_807, __p1_807, __p2_807, __p3_807) __extension__ ({ \
+ float32x2_t __s0_807 = __p0_807; \
+ float16x4_t __s1_807 = __p1_807; \
+ float16x8_t __s2_807 = __p2_807; \
+ float32x2_t __ret_807; \
+ __ret_807 = vfmlsl_low_f16(__s0_807, __s1_807, (float16x4_t) {vgetq_lane_f16(__s2_807, __p3_807), vgetq_lane_f16(__s2_807, __p3_807), vgetq_lane_f16(__s2_807, __p3_807), vgetq_lane_f16(__s2_807, __p3_807)}); \
+ __ret_807; \
+})
+#else
+#define vfmlsl_laneq_low_f16(__p0_808, __p1_808, __p2_808, __p3_808) __extension__ ({ \
+ float32x2_t __s0_808 = __p0_808; \
+ float16x4_t __s1_808 = __p1_808; \
+ float16x8_t __s2_808 = __p2_808; \
+ float32x2_t __rev0_808; __rev0_808 = __builtin_shufflevector(__s0_808, __s0_808, 1, 0); \
+ float16x4_t __rev1_808; __rev1_808 = __builtin_shufflevector(__s1_808, __s1_808, 3, 2, 1, 0); \
+ float16x8_t __rev2_808; __rev2_808 = __builtin_shufflevector(__s2_808, __s2_808, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float32x2_t __ret_808; \
+ __ret_808 = __noswap_vfmlsl_low_f16(__rev0_808, __rev1_808, (float16x4_t) {__noswap_vgetq_lane_f16(__rev2_808, __p3_808), __noswap_vgetq_lane_f16(__rev2_808, __p3_808), __noswap_vgetq_lane_f16(__rev2_808, __p3_808), __noswap_vgetq_lane_f16(__rev2_808, __p3_808)}); \
+ __ret_808 = __builtin_shufflevector(__ret_808, __ret_808, 1, 0); \
+ __ret_808; \
+})
+#endif
+
+#endif
+#if defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(__aarch64__)
+#ifdef __LITTLE_ENDIAN__
+#define vmulh_lane_f16(__p0_809, __p1_809, __p2_809) __extension__ ({ \
+ float16_t __s0_809 = __p0_809; \
+ float16x4_t __s1_809 = __p1_809; \
+ float16_t __ret_809; \
+ __ret_809 = __s0_809 * vget_lane_f16(__s1_809, __p2_809); \
+ __ret_809; \
+})
+#else
+#define vmulh_lane_f16(__p0_810, __p1_810, __p2_810) __extension__ ({ \
+ float16_t __s0_810 = __p0_810; \
+ float16x4_t __s1_810 = __p1_810; \
+ float16x4_t __rev1_810; __rev1_810 = __builtin_shufflevector(__s1_810, __s1_810, 3, 2, 1, 0); \
+ float16_t __ret_810; \
+ __ret_810 = __s0_810 * __noswap_vget_lane_f16(__rev1_810, __p2_810); \
+ __ret_810; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vmulh_laneq_f16(__p0_811, __p1_811, __p2_811) __extension__ ({ \
+ float16_t __s0_811 = __p0_811; \
+ float16x8_t __s1_811 = __p1_811; \
+ float16_t __ret_811; \
+ __ret_811 = __s0_811 * vgetq_lane_f16(__s1_811, __p2_811); \
+ __ret_811; \
+})
+#else
+#define vmulh_laneq_f16(__p0_812, __p1_812, __p2_812) __extension__ ({ \
+ float16_t __s0_812 = __p0_812; \
+ float16x8_t __s1_812 = __p1_812; \
+ float16x8_t __rev1_812; __rev1_812 = __builtin_shufflevector(__s1_812, __s1_812, 7, 6, 5, 4, 3, 2, 1, 0); \
+ float16_t __ret_812; \
+ __ret_812 = __s0_812 * __noswap_vgetq_lane_f16(__rev1_812, __p2_812); \
+ __ret_812; \
+})
+#endif
+
+#endif
+#if defined(__ARM_FEATURE_MATMUL_INT8)
+#ifdef __LITTLE_ENDIAN__
+#define vsudotq_lane_s32(__p0_813, __p1_813, __p2_813, __p3_813) __extension__ ({ \
+ int32x4_t __s0_813 = __p0_813; \
+ int8x16_t __s1_813 = __p1_813; \
+ uint8x8_t __s2_813 = __p2_813; \
+ int32x4_t __ret_813; \
+uint8x8_t __reint_813 = __s2_813; \
+ __ret_813 = vusdotq_s32(__s0_813, (uint8x16_t)(splatq_lane_s32(*(int32x2_t *) &__reint_813, __p3_813)), __s1_813); \
+ __ret_813; \
+})
+#else
+#define vsudotq_lane_s32(__p0_814, __p1_814, __p2_814, __p3_814) __extension__ ({ \
+ int32x4_t __s0_814 = __p0_814; \
+ int8x16_t __s1_814 = __p1_814; \
+ uint8x8_t __s2_814 = __p2_814; \
+ int32x4_t __rev0_814; __rev0_814 = __builtin_shufflevector(__s0_814, __s0_814, 3, 2, 1, 0); \
+ int8x16_t __rev1_814; __rev1_814 = __builtin_shufflevector(__s1_814, __s1_814, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x8_t __rev2_814; __rev2_814 = __builtin_shufflevector(__s2_814, __s2_814, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x4_t __ret_814; \
+uint8x8_t __reint_814 = __rev2_814; \
+ __ret_814 = __noswap_vusdotq_s32(__rev0_814, (uint8x16_t)(__noswap_splatq_lane_s32(*(int32x2_t *) &__reint_814, __p3_814)), __rev1_814); \
+ __ret_814 = __builtin_shufflevector(__ret_814, __ret_814, 3, 2, 1, 0); \
+ __ret_814; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vsudot_lane_s32(__p0_815, __p1_815, __p2_815, __p3_815) __extension__ ({ \
+ int32x2_t __s0_815 = __p0_815; \
+ int8x8_t __s1_815 = __p1_815; \
+ uint8x8_t __s2_815 = __p2_815; \
+ int32x2_t __ret_815; \
+uint8x8_t __reint_815 = __s2_815; \
+ __ret_815 = vusdot_s32(__s0_815, (uint8x8_t)(splat_lane_s32(*(int32x2_t *) &__reint_815, __p3_815)), __s1_815); \
+ __ret_815; \
+})
+#else
+#define vsudot_lane_s32(__p0_816, __p1_816, __p2_816, __p3_816) __extension__ ({ \
+ int32x2_t __s0_816 = __p0_816; \
+ int8x8_t __s1_816 = __p1_816; \
+ uint8x8_t __s2_816 = __p2_816; \
+ int32x2_t __rev0_816; __rev0_816 = __builtin_shufflevector(__s0_816, __s0_816, 1, 0); \
+ int8x8_t __rev1_816; __rev1_816 = __builtin_shufflevector(__s1_816, __s1_816, 7, 6, 5, 4, 3, 2, 1, 0); \
+ uint8x8_t __rev2_816; __rev2_816 = __builtin_shufflevector(__s2_816, __s2_816, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int32x2_t __ret_816; \
+uint8x8_t __reint_816 = __rev2_816; \
+ __ret_816 = __noswap_vusdot_s32(__rev0_816, (uint8x8_t)(__noswap_splat_lane_s32(*(int32x2_t *) &__reint_816, __p3_816)), __rev1_816); \
+ __ret_816 = __builtin_shufflevector(__ret_816, __ret_816, 1, 0); \
+ __ret_816; \
+})
+#endif
+
+#endif
+#if defined(__ARM_FEATURE_QRDMX) && defined(__aarch64__)
+__ai int32_t vqrdmlahs_s32(int32_t __p0, int32_t __p1, int32_t __p2) {
int32_t __ret;
- __ret = vqsubs_s32(__p0, vqrdmulhs_s32(__p1, __p2));
+ __ret = vqadds_s32(__p0, vqrdmulhs_s32(__p1, __p2));
return __ret;
}
-#else
-__ai int32_t vqrdmlshs_s32(int32_t __p0, int32_t __p1, int32_t __p2) {
- int32_t __ret;
- __ret = __noswap_vqsubs_s32(__p0, __noswap_vqrdmulhs_s32(__p1, __p2));
+__ai int16_t vqrdmlahh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
+ int16_t __ret;
+ __ret = vqaddh_s16(__p0, vqrdmulhh_s16(__p1, __p2));
return __ret;
}
+#ifdef __LITTLE_ENDIAN__
+#define vqrdmlahs_lane_s32(__p0_817, __p1_817, __p2_817, __p3_817) __extension__ ({ \
+ int32_t __s0_817 = __p0_817; \
+ int32_t __s1_817 = __p1_817; \
+ int32x2_t __s2_817 = __p2_817; \
+ int32_t __ret_817; \
+ __ret_817 = vqadds_s32(__s0_817, vqrdmulhs_s32(__s1_817, vget_lane_s32(__s2_817, __p3_817))); \
+ __ret_817; \
+})
+#else
+#define vqrdmlahs_lane_s32(__p0_818, __p1_818, __p2_818, __p3_818) __extension__ ({ \
+ int32_t __s0_818 = __p0_818; \
+ int32_t __s1_818 = __p1_818; \
+ int32x2_t __s2_818 = __p2_818; \
+ int32x2_t __rev2_818; __rev2_818 = __builtin_shufflevector(__s2_818, __s2_818, 1, 0); \
+ int32_t __ret_818; \
+ __ret_818 = vqadds_s32(__s0_818, vqrdmulhs_s32(__s1_818, __noswap_vget_lane_s32(__rev2_818, __p3_818))); \
+ __ret_818; \
+})
#endif
#ifdef __LITTLE_ENDIAN__
-__ai int16_t vqrdmlshh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
- int16_t __ret;
- __ret = vqsubh_s16(__p0, vqrdmulhh_s16(__p1, __p2));
+#define vqrdmlahh_lane_s16(__p0_819, __p1_819, __p2_819, __p3_819) __extension__ ({ \
+ int16_t __s0_819 = __p0_819; \
+ int16_t __s1_819 = __p1_819; \
+ int16x4_t __s2_819 = __p2_819; \
+ int16_t __ret_819; \
+ __ret_819 = vqaddh_s16(__s0_819, vqrdmulhh_s16(__s1_819, vget_lane_s16(__s2_819, __p3_819))); \
+ __ret_819; \
+})
+#else
+#define vqrdmlahh_lane_s16(__p0_820, __p1_820, __p2_820, __p3_820) __extension__ ({ \
+ int16_t __s0_820 = __p0_820; \
+ int16_t __s1_820 = __p1_820; \
+ int16x4_t __s2_820 = __p2_820; \
+ int16x4_t __rev2_820; __rev2_820 = __builtin_shufflevector(__s2_820, __s2_820, 3, 2, 1, 0); \
+ int16_t __ret_820; \
+ __ret_820 = vqaddh_s16(__s0_820, vqrdmulhh_s16(__s1_820, __noswap_vget_lane_s16(__rev2_820, __p3_820))); \
+ __ret_820; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vqrdmlahs_laneq_s32(__p0_821, __p1_821, __p2_821, __p3_821) __extension__ ({ \
+ int32_t __s0_821 = __p0_821; \
+ int32_t __s1_821 = __p1_821; \
+ int32x4_t __s2_821 = __p2_821; \
+ int32_t __ret_821; \
+ __ret_821 = vqadds_s32(__s0_821, vqrdmulhs_s32(__s1_821, vgetq_lane_s32(__s2_821, __p3_821))); \
+ __ret_821; \
+})
+#else
+#define vqrdmlahs_laneq_s32(__p0_822, __p1_822, __p2_822, __p3_822) __extension__ ({ \
+ int32_t __s0_822 = __p0_822; \
+ int32_t __s1_822 = __p1_822; \
+ int32x4_t __s2_822 = __p2_822; \
+ int32x4_t __rev2_822; __rev2_822 = __builtin_shufflevector(__s2_822, __s2_822, 3, 2, 1, 0); \
+ int32_t __ret_822; \
+ __ret_822 = vqadds_s32(__s0_822, vqrdmulhs_s32(__s1_822, __noswap_vgetq_lane_s32(__rev2_822, __p3_822))); \
+ __ret_822; \
+})
+#endif
+
+#ifdef __LITTLE_ENDIAN__
+#define vqrdmlahh_laneq_s16(__p0_823, __p1_823, __p2_823, __p3_823) __extension__ ({ \
+ int16_t __s0_823 = __p0_823; \
+ int16_t __s1_823 = __p1_823; \
+ int16x8_t __s2_823 = __p2_823; \
+ int16_t __ret_823; \
+ __ret_823 = vqaddh_s16(__s0_823, vqrdmulhh_s16(__s1_823, vgetq_lane_s16(__s2_823, __p3_823))); \
+ __ret_823; \
+})
+#else
+#define vqrdmlahh_laneq_s16(__p0_824, __p1_824, __p2_824, __p3_824) __extension__ ({ \
+ int16_t __s0_824 = __p0_824; \
+ int16_t __s1_824 = __p1_824; \
+ int16x8_t __s2_824 = __p2_824; \
+ int16x8_t __rev2_824; __rev2_824 = __builtin_shufflevector(__s2_824, __s2_824, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16_t __ret_824; \
+ __ret_824 = vqaddh_s16(__s0_824, vqrdmulhh_s16(__s1_824, __noswap_vgetq_lane_s16(__rev2_824, __p3_824))); \
+ __ret_824; \
+})
+#endif
+
+__ai int32_t vqrdmlshs_s32(int32_t __p0, int32_t __p1, int32_t __p2) {
+ int32_t __ret;
+ __ret = vqsubs_s32(__p0, vqrdmulhs_s32(__p1, __p2));
return __ret;
}
-#else
__ai int16_t vqrdmlshh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
int16_t __ret;
- __ret = __noswap_vqsubh_s16(__p0, __noswap_vqrdmulhh_s16(__p1, __p2));
+ __ret = vqsubh_s16(__p0, vqrdmulhh_s16(__p1, __p2));
return __ret;
}
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlshs_lane_s32(__p0_276, __p1_276, __p2_276, __p3_276) __extension__ ({ \
- int32_t __s0_276 = __p0_276; \
- int32_t __s1_276 = __p1_276; \
- int32x2_t __s2_276 = __p2_276; \
- int32_t __ret_276; \
- __ret_276 = vqsubs_s32(__s0_276, vqrdmulhs_s32(__s1_276, vget_lane_s32(__s2_276, __p3_276))); \
- __ret_276; \
+#define vqrdmlshs_lane_s32(__p0_825, __p1_825, __p2_825, __p3_825) __extension__ ({ \
+ int32_t __s0_825 = __p0_825; \
+ int32_t __s1_825 = __p1_825; \
+ int32x2_t __s2_825 = __p2_825; \
+ int32_t __ret_825; \
+ __ret_825 = vqsubs_s32(__s0_825, vqrdmulhs_s32(__s1_825, vget_lane_s32(__s2_825, __p3_825))); \
+ __ret_825; \
})
#else
-#define vqrdmlshs_lane_s32(__p0_277, __p1_277, __p2_277, __p3_277) __extension__ ({ \
- int32_t __s0_277 = __p0_277; \
- int32_t __s1_277 = __p1_277; \
- int32x2_t __s2_277 = __p2_277; \
- int32x2_t __rev2_277; __rev2_277 = __builtin_shufflevector(__s2_277, __s2_277, 1, 0); \
- int32_t __ret_277; \
- __ret_277 = __noswap_vqsubs_s32(__s0_277, __noswap_vqrdmulhs_s32(__s1_277, __noswap_vget_lane_s32(__rev2_277, __p3_277))); \
- __ret_277; \
+#define vqrdmlshs_lane_s32(__p0_826, __p1_826, __p2_826, __p3_826) __extension__ ({ \
+ int32_t __s0_826 = __p0_826; \
+ int32_t __s1_826 = __p1_826; \
+ int32x2_t __s2_826 = __p2_826; \
+ int32x2_t __rev2_826; __rev2_826 = __builtin_shufflevector(__s2_826, __s2_826, 1, 0); \
+ int32_t __ret_826; \
+ __ret_826 = vqsubs_s32(__s0_826, vqrdmulhs_s32(__s1_826, __noswap_vget_lane_s32(__rev2_826, __p3_826))); \
+ __ret_826; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlshh_lane_s16(__p0_278, __p1_278, __p2_278, __p3_278) __extension__ ({ \
- int16_t __s0_278 = __p0_278; \
- int16_t __s1_278 = __p1_278; \
- int16x4_t __s2_278 = __p2_278; \
- int16_t __ret_278; \
- __ret_278 = vqsubh_s16(__s0_278, vqrdmulhh_s16(__s1_278, vget_lane_s16(__s2_278, __p3_278))); \
- __ret_278; \
+#define vqrdmlshh_lane_s16(__p0_827, __p1_827, __p2_827, __p3_827) __extension__ ({ \
+ int16_t __s0_827 = __p0_827; \
+ int16_t __s1_827 = __p1_827; \
+ int16x4_t __s2_827 = __p2_827; \
+ int16_t __ret_827; \
+ __ret_827 = vqsubh_s16(__s0_827, vqrdmulhh_s16(__s1_827, vget_lane_s16(__s2_827, __p3_827))); \
+ __ret_827; \
})
#else
-#define vqrdmlshh_lane_s16(__p0_279, __p1_279, __p2_279, __p3_279) __extension__ ({ \
- int16_t __s0_279 = __p0_279; \
- int16_t __s1_279 = __p1_279; \
- int16x4_t __s2_279 = __p2_279; \
- int16x4_t __rev2_279; __rev2_279 = __builtin_shufflevector(__s2_279, __s2_279, 3, 2, 1, 0); \
- int16_t __ret_279; \
- __ret_279 = __noswap_vqsubh_s16(__s0_279, __noswap_vqrdmulhh_s16(__s1_279, __noswap_vget_lane_s16(__rev2_279, __p3_279))); \
- __ret_279; \
+#define vqrdmlshh_lane_s16(__p0_828, __p1_828, __p2_828, __p3_828) __extension__ ({ \
+ int16_t __s0_828 = __p0_828; \
+ int16_t __s1_828 = __p1_828; \
+ int16x4_t __s2_828 = __p2_828; \
+ int16x4_t __rev2_828; __rev2_828 = __builtin_shufflevector(__s2_828, __s2_828, 3, 2, 1, 0); \
+ int16_t __ret_828; \
+ __ret_828 = vqsubh_s16(__s0_828, vqrdmulhh_s16(__s1_828, __noswap_vget_lane_s16(__rev2_828, __p3_828))); \
+ __ret_828; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlshs_laneq_s32(__p0_280, __p1_280, __p2_280, __p3_280) __extension__ ({ \
- int32_t __s0_280 = __p0_280; \
- int32_t __s1_280 = __p1_280; \
- int32x4_t __s2_280 = __p2_280; \
- int32_t __ret_280; \
- __ret_280 = vqsubs_s32(__s0_280, vqrdmulhs_s32(__s1_280, vgetq_lane_s32(__s2_280, __p3_280))); \
- __ret_280; \
+#define vqrdmlshs_laneq_s32(__p0_829, __p1_829, __p2_829, __p3_829) __extension__ ({ \
+ int32_t __s0_829 = __p0_829; \
+ int32_t __s1_829 = __p1_829; \
+ int32x4_t __s2_829 = __p2_829; \
+ int32_t __ret_829; \
+ __ret_829 = vqsubs_s32(__s0_829, vqrdmulhs_s32(__s1_829, vgetq_lane_s32(__s2_829, __p3_829))); \
+ __ret_829; \
})
#else
-#define vqrdmlshs_laneq_s32(__p0_281, __p1_281, __p2_281, __p3_281) __extension__ ({ \
- int32_t __s0_281 = __p0_281; \
- int32_t __s1_281 = __p1_281; \
- int32x4_t __s2_281 = __p2_281; \
- int32x4_t __rev2_281; __rev2_281 = __builtin_shufflevector(__s2_281, __s2_281, 3, 2, 1, 0); \
- int32_t __ret_281; \
- __ret_281 = __noswap_vqsubs_s32(__s0_281, __noswap_vqrdmulhs_s32(__s1_281, __noswap_vgetq_lane_s32(__rev2_281, __p3_281))); \
- __ret_281; \
+#define vqrdmlshs_laneq_s32(__p0_830, __p1_830, __p2_830, __p3_830) __extension__ ({ \
+ int32_t __s0_830 = __p0_830; \
+ int32_t __s1_830 = __p1_830; \
+ int32x4_t __s2_830 = __p2_830; \
+ int32x4_t __rev2_830; __rev2_830 = __builtin_shufflevector(__s2_830, __s2_830, 3, 2, 1, 0); \
+ int32_t __ret_830; \
+ __ret_830 = vqsubs_s32(__s0_830, vqrdmulhs_s32(__s1_830, __noswap_vgetq_lane_s32(__rev2_830, __p3_830))); \
+ __ret_830; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vqrdmlshh_laneq_s16(__p0_282, __p1_282, __p2_282, __p3_282) __extension__ ({ \
- int16_t __s0_282 = __p0_282; \
- int16_t __s1_282 = __p1_282; \
- int16x8_t __s2_282 = __p2_282; \
- int16_t __ret_282; \
- __ret_282 = vqsubh_s16(__s0_282, vqrdmulhh_s16(__s1_282, vgetq_lane_s16(__s2_282, __p3_282))); \
- __ret_282; \
+#define vqrdmlshh_laneq_s16(__p0_831, __p1_831, __p2_831, __p3_831) __extension__ ({ \
+ int16_t __s0_831 = __p0_831; \
+ int16_t __s1_831 = __p1_831; \
+ int16x8_t __s2_831 = __p2_831; \
+ int16_t __ret_831; \
+ __ret_831 = vqsubh_s16(__s0_831, vqrdmulhh_s16(__s1_831, vgetq_lane_s16(__s2_831, __p3_831))); \
+ __ret_831; \
})
#else
-#define vqrdmlshh_laneq_s16(__p0_283, __p1_283, __p2_283, __p3_283) __extension__ ({ \
- int16_t __s0_283 = __p0_283; \
- int16_t __s1_283 = __p1_283; \
- int16x8_t __s2_283 = __p2_283; \
- int16x8_t __rev2_283; __rev2_283 = __builtin_shufflevector(__s2_283, __s2_283, 7, 6, 5, 4, 3, 2, 1, 0); \
- int16_t __ret_283; \
- __ret_283 = __noswap_vqsubh_s16(__s0_283, __noswap_vqrdmulhh_s16(__s1_283, __noswap_vgetq_lane_s16(__rev2_283, __p3_283))); \
- __ret_283; \
+#define vqrdmlshh_laneq_s16(__p0_832, __p1_832, __p2_832, __p3_832) __extension__ ({ \
+ int16_t __s0_832 = __p0_832; \
+ int16_t __s1_832 = __p1_832; \
+ int16x8_t __s2_832 = __p2_832; \
+ int16x8_t __rev2_832; __rev2_832 = __builtin_shufflevector(__s2_832, __s2_832, 7, 6, 5, 4, 3, 2, 1, 0); \
+ int16_t __ret_832; \
+ __ret_832 = vqsubh_s16(__s0_832, vqrdmulhh_s16(__s1_832, __noswap_vgetq_lane_s16(__rev2_832, __p3_832))); \
+ __ret_832; \
})
#endif
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_lane_p64(__p0_284, __p1_284, __p2_284, __p3_284) __extension__ ({ \
- poly64x2_t __s0_284 = __p0_284; \
- poly64x1_t __s2_284 = __p2_284; \
- poly64x2_t __ret_284; \
- __ret_284 = vsetq_lane_p64(vget_lane_p64(__s2_284, __p3_284), __s0_284, __p1_284); \
- __ret_284; \
-})
-#else
-#define vcopyq_lane_p64(__p0_285, __p1_285, __p2_285, __p3_285) __extension__ ({ \
- poly64x2_t __s0_285 = __p0_285; \
- poly64x1_t __s2_285 = __p2_285; \
- poly64x2_t __rev0_285; __rev0_285 = __builtin_shufflevector(__s0_285, __s0_285, 1, 0); \
- poly64x2_t __ret_285; \
- __ret_285 = __noswap_vsetq_lane_p64(__noswap_vget_lane_p64(__s2_285, __p3_285), __rev0_285, __p1_285); \
- __ret_285 = __builtin_shufflevector(__ret_285, __ret_285, 1, 0); \
- __ret_285; \
-})
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#define vcopyq_lane_f64(__p0_286, __p1_286, __p2_286, __p3_286) __extension__ ({ \
- float64x2_t __s0_286 = __p0_286; \
- float64x1_t __s2_286 = __p2_286; \
- float64x2_t __ret_286; \
- __ret_286 = vsetq_lane_f64(vget_lane_f64(__s2_286, __p3_286), __s0_286, __p1_286); \
- __ret_286; \
+#define vcopyq_lane_p64(__p0_833, __p1_833, __p2_833, __p3_833) __extension__ ({ \
+ poly64x2_t __s0_833 = __p0_833; \
+ poly64x1_t __s2_833 = __p2_833; \
+ poly64x2_t __ret_833; \
+ __ret_833 = vsetq_lane_p64(vget_lane_p64(__s2_833, __p3_833), __s0_833, __p1_833); \
+ __ret_833; \
})
#else
-#define vcopyq_lane_f64(__p0_287, __p1_287, __p2_287, __p3_287) __extension__ ({ \
- float64x2_t __s0_287 = __p0_287; \
- float64x1_t __s2_287 = __p2_287; \
- float64x2_t __rev0_287; __rev0_287 = __builtin_shufflevector(__s0_287, __s0_287, 1, 0); \
- float64x2_t __ret_287; \
- __ret_287 = __noswap_vsetq_lane_f64(__noswap_vget_lane_f64(__s2_287, __p3_287), __rev0_287, __p1_287); \
- __ret_287 = __builtin_shufflevector(__ret_287, __ret_287, 1, 0); \
- __ret_287; \
+#define vcopyq_lane_p64(__p0_834, __p1_834, __p2_834, __p3_834) __extension__ ({ \
+ poly64x2_t __s0_834 = __p0_834; \
+ poly64x1_t __s2_834 = __p2_834; \
+ poly64x2_t __rev0_834; __rev0_834 = __builtin_shufflevector(__s0_834, __s0_834, 1, 0); \
+ poly64x2_t __ret_834; \
+ __ret_834 = __noswap_vsetq_lane_p64(vget_lane_p64(__s2_834, __p3_834), __rev0_834, __p1_834); \
+ __ret_834 = __builtin_shufflevector(__ret_834, __ret_834, 1, 0); \
+ __ret_834; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_lane_p64(__p0_288, __p1_288, __p2_288, __p3_288) __extension__ ({ \
- poly64x1_t __s0_288 = __p0_288; \
- poly64x1_t __s2_288 = __p2_288; \
- poly64x1_t __ret_288; \
- __ret_288 = vset_lane_p64(vget_lane_p64(__s2_288, __p3_288), __s0_288, __p1_288); \
- __ret_288; \
+#define vcopyq_lane_f64(__p0_835, __p1_835, __p2_835, __p3_835) __extension__ ({ \
+ float64x2_t __s0_835 = __p0_835; \
+ float64x1_t __s2_835 = __p2_835; \
+ float64x2_t __ret_835; \
+ __ret_835 = vsetq_lane_f64(vget_lane_f64(__s2_835, __p3_835), __s0_835, __p1_835); \
+ __ret_835; \
})
#else
-#define vcopy_lane_p64(__p0_289, __p1_289, __p2_289, __p3_289) __extension__ ({ \
- poly64x1_t __s0_289 = __p0_289; \
- poly64x1_t __s2_289 = __p2_289; \
- poly64x1_t __ret_289; \
- __ret_289 = __noswap_vset_lane_p64(__noswap_vget_lane_p64(__s2_289, __p3_289), __s0_289, __p1_289); \
- __ret_289; \
+#define vcopyq_lane_f64(__p0_836, __p1_836, __p2_836, __p3_836) __extension__ ({ \
+ float64x2_t __s0_836 = __p0_836; \
+ float64x1_t __s2_836 = __p2_836; \
+ float64x2_t __rev0_836; __rev0_836 = __builtin_shufflevector(__s0_836, __s0_836, 1, 0); \
+ float64x2_t __ret_836; \
+ __ret_836 = __noswap_vsetq_lane_f64(vget_lane_f64(__s2_836, __p3_836), __rev0_836, __p1_836); \
+ __ret_836 = __builtin_shufflevector(__ret_836, __ret_836, 1, 0); \
+ __ret_836; \
})
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vcopy_lane_f64(__p0_290, __p1_290, __p2_290, __p3_290) __extension__ ({ \
- float64x1_t __s0_290 = __p0_290; \
- float64x1_t __s2_290 = __p2_290; \
- float64x1_t __ret_290; \
- __ret_290 = vset_lane_f64(vget_lane_f64(__s2_290, __p3_290), __s0_290, __p1_290); \
- __ret_290; \
+#define vcopy_lane_p64(__p0_837, __p1_837, __p2_837, __p3_837) __extension__ ({ \
+ poly64x1_t __s0_837 = __p0_837; \
+ poly64x1_t __s2_837 = __p2_837; \
+ poly64x1_t __ret_837; \
+ __ret_837 = vset_lane_p64(vget_lane_p64(__s2_837, __p3_837), __s0_837, __p1_837); \
+ __ret_837; \
})
-#else
-#define vcopy_lane_f64(__p0_291, __p1_291, __p2_291, __p3_291) __extension__ ({ \
- float64x1_t __s0_291 = __p0_291; \
- float64x1_t __s2_291 = __p2_291; \
- float64x1_t __ret_291; \
- __ret_291 = __noswap_vset_lane_f64(__noswap_vget_lane_f64(__s2_291, __p3_291), __s0_291, __p1_291); \
- __ret_291; \
+#define vcopy_lane_f64(__p0_838, __p1_838, __p2_838, __p3_838) __extension__ ({ \
+ float64x1_t __s0_838 = __p0_838; \
+ float64x1_t __s2_838 = __p2_838; \
+ float64x1_t __ret_838; \
+ __ret_838 = vset_lane_f64(vget_lane_f64(__s2_838, __p3_838), __s0_838, __p1_838); \
+ __ret_838; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_laneq_p64(__p0_292, __p1_292, __p2_292, __p3_292) __extension__ ({ \
- poly64x2_t __s0_292 = __p0_292; \
- poly64x2_t __s2_292 = __p2_292; \
- poly64x2_t __ret_292; \
- __ret_292 = vsetq_lane_p64(vgetq_lane_p64(__s2_292, __p3_292), __s0_292, __p1_292); \
- __ret_292; \
+#define vcopyq_laneq_p64(__p0_839, __p1_839, __p2_839, __p3_839) __extension__ ({ \
+ poly64x2_t __s0_839 = __p0_839; \
+ poly64x2_t __s2_839 = __p2_839; \
+ poly64x2_t __ret_839; \
+ __ret_839 = vsetq_lane_p64(vgetq_lane_p64(__s2_839, __p3_839), __s0_839, __p1_839); \
+ __ret_839; \
})
#else
-#define vcopyq_laneq_p64(__p0_293, __p1_293, __p2_293, __p3_293) __extension__ ({ \
- poly64x2_t __s0_293 = __p0_293; \
- poly64x2_t __s2_293 = __p2_293; \
- poly64x2_t __rev0_293; __rev0_293 = __builtin_shufflevector(__s0_293, __s0_293, 1, 0); \
- poly64x2_t __rev2_293; __rev2_293 = __builtin_shufflevector(__s2_293, __s2_293, 1, 0); \
- poly64x2_t __ret_293; \
- __ret_293 = __noswap_vsetq_lane_p64(__noswap_vgetq_lane_p64(__rev2_293, __p3_293), __rev0_293, __p1_293); \
- __ret_293 = __builtin_shufflevector(__ret_293, __ret_293, 1, 0); \
- __ret_293; \
+#define vcopyq_laneq_p64(__p0_840, __p1_840, __p2_840, __p3_840) __extension__ ({ \
+ poly64x2_t __s0_840 = __p0_840; \
+ poly64x2_t __s2_840 = __p2_840; \
+ poly64x2_t __rev0_840; __rev0_840 = __builtin_shufflevector(__s0_840, __s0_840, 1, 0); \
+ poly64x2_t __rev2_840; __rev2_840 = __builtin_shufflevector(__s2_840, __s2_840, 1, 0); \
+ poly64x2_t __ret_840; \
+ __ret_840 = __noswap_vsetq_lane_p64(__noswap_vgetq_lane_p64(__rev2_840, __p3_840), __rev0_840, __p1_840); \
+ __ret_840 = __builtin_shufflevector(__ret_840, __ret_840, 1, 0); \
+ __ret_840; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopyq_laneq_f64(__p0_294, __p1_294, __p2_294, __p3_294) __extension__ ({ \
- float64x2_t __s0_294 = __p0_294; \
- float64x2_t __s2_294 = __p2_294; \
- float64x2_t __ret_294; \
- __ret_294 = vsetq_lane_f64(vgetq_lane_f64(__s2_294, __p3_294), __s0_294, __p1_294); \
- __ret_294; \
+#define vcopyq_laneq_f64(__p0_841, __p1_841, __p2_841, __p3_841) __extension__ ({ \
+ float64x2_t __s0_841 = __p0_841; \
+ float64x2_t __s2_841 = __p2_841; \
+ float64x2_t __ret_841; \
+ __ret_841 = vsetq_lane_f64(vgetq_lane_f64(__s2_841, __p3_841), __s0_841, __p1_841); \
+ __ret_841; \
})
#else
-#define vcopyq_laneq_f64(__p0_295, __p1_295, __p2_295, __p3_295) __extension__ ({ \
- float64x2_t __s0_295 = __p0_295; \
- float64x2_t __s2_295 = __p2_295; \
- float64x2_t __rev0_295; __rev0_295 = __builtin_shufflevector(__s0_295, __s0_295, 1, 0); \
- float64x2_t __rev2_295; __rev2_295 = __builtin_shufflevector(__s2_295, __s2_295, 1, 0); \
- float64x2_t __ret_295; \
- __ret_295 = __noswap_vsetq_lane_f64(__noswap_vgetq_lane_f64(__rev2_295, __p3_295), __rev0_295, __p1_295); \
- __ret_295 = __builtin_shufflevector(__ret_295, __ret_295, 1, 0); \
- __ret_295; \
+#define vcopyq_laneq_f64(__p0_842, __p1_842, __p2_842, __p3_842) __extension__ ({ \
+ float64x2_t __s0_842 = __p0_842; \
+ float64x2_t __s2_842 = __p2_842; \
+ float64x2_t __rev0_842; __rev0_842 = __builtin_shufflevector(__s0_842, __s0_842, 1, 0); \
+ float64x2_t __rev2_842; __rev2_842 = __builtin_shufflevector(__s2_842, __s2_842, 1, 0); \
+ float64x2_t __ret_842; \
+ __ret_842 = __noswap_vsetq_lane_f64(__noswap_vgetq_lane_f64(__rev2_842, __p3_842), __rev0_842, __p1_842); \
+ __ret_842 = __builtin_shufflevector(__ret_842, __ret_842, 1, 0); \
+ __ret_842; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_laneq_p64(__p0_296, __p1_296, __p2_296, __p3_296) __extension__ ({ \
- poly64x1_t __s0_296 = __p0_296; \
- poly64x2_t __s2_296 = __p2_296; \
- poly64x1_t __ret_296; \
- __ret_296 = vset_lane_p64(vgetq_lane_p64(__s2_296, __p3_296), __s0_296, __p1_296); \
- __ret_296; \
+#define vcopy_laneq_p64(__p0_843, __p1_843, __p2_843, __p3_843) __extension__ ({ \
+ poly64x1_t __s0_843 = __p0_843; \
+ poly64x2_t __s2_843 = __p2_843; \
+ poly64x1_t __ret_843; \
+ __ret_843 = vset_lane_p64(vgetq_lane_p64(__s2_843, __p3_843), __s0_843, __p1_843); \
+ __ret_843; \
})
#else
-#define vcopy_laneq_p64(__p0_297, __p1_297, __p2_297, __p3_297) __extension__ ({ \
- poly64x1_t __s0_297 = __p0_297; \
- poly64x2_t __s2_297 = __p2_297; \
- poly64x2_t __rev2_297; __rev2_297 = __builtin_shufflevector(__s2_297, __s2_297, 1, 0); \
- poly64x1_t __ret_297; \
- __ret_297 = __noswap_vset_lane_p64(__noswap_vgetq_lane_p64(__rev2_297, __p3_297), __s0_297, __p1_297); \
- __ret_297; \
+#define vcopy_laneq_p64(__p0_844, __p1_844, __p2_844, __p3_844) __extension__ ({ \
+ poly64x1_t __s0_844 = __p0_844; \
+ poly64x2_t __s2_844 = __p2_844; \
+ poly64x2_t __rev2_844; __rev2_844 = __builtin_shufflevector(__s2_844, __s2_844, 1, 0); \
+ poly64x1_t __ret_844; \
+ __ret_844 = vset_lane_p64(__noswap_vgetq_lane_p64(__rev2_844, __p3_844), __s0_844, __p1_844); \
+ __ret_844; \
})
#endif
#ifdef __LITTLE_ENDIAN__
-#define vcopy_laneq_f64(__p0_298, __p1_298, __p2_298, __p3_298) __extension__ ({ \
- float64x1_t __s0_298 = __p0_298; \
- float64x2_t __s2_298 = __p2_298; \
- float64x1_t __ret_298; \
- __ret_298 = vset_lane_f64(vgetq_lane_f64(__s2_298, __p3_298), __s0_298, __p1_298); \
- __ret_298; \
+#define vcopy_laneq_f64(__p0_845, __p1_845, __p2_845, __p3_845) __extension__ ({ \
+ float64x1_t __s0_845 = __p0_845; \
+ float64x2_t __s2_845 = __p2_845; \
+ float64x1_t __ret_845; \
+ __ret_845 = vset_lane_f64(vgetq_lane_f64(__s2_845, __p3_845), __s0_845, __p1_845); \
+ __ret_845; \
})
#else
-#define vcopy_laneq_f64(__p0_299, __p1_299, __p2_299, __p3_299) __extension__ ({ \
- float64x1_t __s0_299 = __p0_299; \
- float64x2_t __s2_299 = __p2_299; \
- float64x2_t __rev2_299; __rev2_299 = __builtin_shufflevector(__s2_299, __s2_299, 1, 0); \
- float64x1_t __ret_299; \
- __ret_299 = __noswap_vset_lane_f64(__noswap_vgetq_lane_f64(__rev2_299, __p3_299), __s0_299, __p1_299); \
- __ret_299; \
+#define vcopy_laneq_f64(__p0_846, __p1_846, __p2_846, __p3_846) __extension__ ({ \
+ float64x1_t __s0_846 = __p0_846; \
+ float64x2_t __s2_846 = __p2_846; \
+ float64x2_t __rev2_846; __rev2_846 = __builtin_shufflevector(__s2_846, __s2_846, 1, 0); \
+ float64x1_t __ret_846; \
+ __ret_846 = vset_lane_f64(__noswap_vgetq_lane_f64(__rev2_846, __p3_846), __s0_846, __p1_846); \
+ __ret_846; \
})
#endif
}
#endif
-#ifdef __LITTLE_ENDIAN__
-#define vmulx_lane_f64(__p0_300, __p1_300, __p2_300) __extension__ ({ \
- float64x1_t __s0_300 = __p0_300; \
- float64x1_t __s1_300 = __p1_300; \
- float64x1_t __ret_300; \
- float64_t __x_300 = vget_lane_f64(__s0_300, 0); \
- float64_t __y_300 = vget_lane_f64(__s1_300, __p2_300); \
- float64_t __z_300 = vmulxd_f64(__x_300, __y_300); \
- __ret_300 = vset_lane_f64(__z_300, __s0_300, __p2_300); \
- __ret_300; \
-})
-#else
-#define vmulx_lane_f64(__p0_301, __p1_301, __p2_301) __extension__ ({ \
- float64x1_t __s0_301 = __p0_301; \
- float64x1_t __s1_301 = __p1_301; \
- float64x1_t __ret_301; \
- float64_t __x_301 = __noswap_vget_lane_f64(__s0_301, 0); \
- float64_t __y_301 = __noswap_vget_lane_f64(__s1_301, __p2_301); \
- float64_t __z_301 = __noswap_vmulxd_f64(__x_301, __y_301); \
- __ret_301 = __noswap_vset_lane_f64(__z_301, __s0_301, __p2_301); \
- __ret_301; \
+#define vmulx_lane_f64(__p0_847, __p1_847, __p2_847) __extension__ ({ \
+ float64x1_t __s0_847 = __p0_847; \
+ float64x1_t __s1_847 = __p1_847; \
+ float64x1_t __ret_847; \
+ float64_t __x_847 = vget_lane_f64(__s0_847, 0); \
+ float64_t __y_847 = vget_lane_f64(__s1_847, __p2_847); \
+ float64_t __z_847 = vmulxd_f64(__x_847, __y_847); \
+ __ret_847 = vset_lane_f64(__z_847, __s0_847, __p2_847); \
+ __ret_847; \
})
-#endif
-
#ifdef __LITTLE_ENDIAN__
-#define vmulx_laneq_f64(__p0_302, __p1_302, __p2_302) __extension__ ({ \
- float64x1_t __s0_302 = __p0_302; \
- float64x2_t __s1_302 = __p1_302; \
- float64x1_t __ret_302; \
- float64_t __x_302 = vget_lane_f64(__s0_302, 0); \
- float64_t __y_302 = vgetq_lane_f64(__s1_302, __p2_302); \
- float64_t __z_302 = vmulxd_f64(__x_302, __y_302); \
- __ret_302 = vset_lane_f64(__z_302, __s0_302, 0); \
- __ret_302; \
+#define vmulx_laneq_f64(__p0_848, __p1_848, __p2_848) __extension__ ({ \
+ float64x1_t __s0_848 = __p0_848; \
+ float64x2_t __s1_848 = __p1_848; \
+ float64x1_t __ret_848; \
+ float64_t __x_848 = vget_lane_f64(__s0_848, 0); \
+ float64_t __y_848 = vgetq_lane_f64(__s1_848, __p2_848); \
+ float64_t __z_848 = vmulxd_f64(__x_848, __y_848); \
+ __ret_848 = vset_lane_f64(__z_848, __s0_848, 0); \
+ __ret_848; \
})
#else
-#define vmulx_laneq_f64(__p0_303, __p1_303, __p2_303) __extension__ ({ \
- float64x1_t __s0_303 = __p0_303; \
- float64x2_t __s1_303 = __p1_303; \
- float64x2_t __rev1_303; __rev1_303 = __builtin_shufflevector(__s1_303, __s1_303, 1, 0); \
- float64x1_t __ret_303; \
- float64_t __x_303 = __noswap_vget_lane_f64(__s0_303, 0); \
- float64_t __y_303 = __noswap_vgetq_lane_f64(__rev1_303, __p2_303); \
- float64_t __z_303 = __noswap_vmulxd_f64(__x_303, __y_303); \
- __ret_303 = __noswap_vset_lane_f64(__z_303, __s0_303, 0); \
- __ret_303; \
+#define vmulx_laneq_f64(__p0_849, __p1_849, __p2_849) __extension__ ({ \
+ float64x1_t __s0_849 = __p0_849; \
+ float64x2_t __s1_849 = __p1_849; \
+ float64x2_t __rev1_849; __rev1_849 = __builtin_shufflevector(__s1_849, __s1_849, 1, 0); \
+ float64x1_t __ret_849; \
+ float64_t __x_849 = vget_lane_f64(__s0_849, 0); \
+ float64_t __y_849 = __noswap_vgetq_lane_f64(__rev1_849, __p2_849); \
+ float64_t __z_849 = vmulxd_f64(__x_849, __y_849); \
+ __ret_849 = vset_lane_f64(__z_849, __s0_849, 0); \
+ __ret_849; \
})
#endif
#undef __ai
+#endif /* if !defined(__ARM_NEON) */
+#endif /* ifndef __ARM_FP */
#endif /* __ARM_NEON_H */
--- /dev/null
+/*===---- arm_sve.h - ARM SVE intrinsics -----------------------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __ARM_SVE_H
+#define __ARM_SVE_H
+
+#if !defined(__ARM_FEATURE_SVE)
+#error "SVE support not enabled"
+#else
+
+#if !defined(__LITTLE_ENDIAN__)
+#error "Big endian is currently not supported for arm_sve.h"
+#endif
+#include <stdint.h>
+
+#ifdef __cplusplus
+extern "C" {
+#else
+#include <stdbool.h>
+#endif
+
+typedef __fp16 float16_t;
+typedef float float32_t;
+typedef double float64_t;
+typedef __SVInt8_t svint8_t;
+typedef __SVInt16_t svint16_t;
+typedef __SVInt32_t svint32_t;
+typedef __SVInt64_t svint64_t;
+typedef __SVUint8_t svuint8_t;
+typedef __SVUint16_t svuint16_t;
+typedef __SVUint32_t svuint32_t;
+typedef __SVUint64_t svuint64_t;
+typedef __SVFloat16_t svfloat16_t;
+
+#if defined(__ARM_FEATURE_SVE_BF16) && !defined(__ARM_FEATURE_BF16_SCALAR_ARITHMETIC)
+#error "__ARM_FEATURE_BF16_SCALAR_ARITHMETIC must be defined when __ARM_FEATURE_SVE_BF16 is defined"
+#endif
+
+#if defined(__ARM_FEATURE_SVE_BF16)
+typedef __SVBFloat16_t svbfloat16_t;
+#endif
+
+#if defined(__ARM_FEATURE_BF16_SCALAR_ARITHMETIC)
+#include <arm_bf16.h>
+typedef __bf16 bfloat16_t;
+#endif
+
+typedef __SVFloat32_t svfloat32_t;
+typedef __SVFloat64_t svfloat64_t;
+typedef __clang_svint8x2_t svint8x2_t;
+typedef __clang_svint16x2_t svint16x2_t;
+typedef __clang_svint32x2_t svint32x2_t;
+typedef __clang_svint64x2_t svint64x2_t;
+typedef __clang_svuint8x2_t svuint8x2_t;
+typedef __clang_svuint16x2_t svuint16x2_t;
+typedef __clang_svuint32x2_t svuint32x2_t;
+typedef __clang_svuint64x2_t svuint64x2_t;
+typedef __clang_svfloat16x2_t svfloat16x2_t;
+typedef __clang_svfloat32x2_t svfloat32x2_t;
+typedef __clang_svfloat64x2_t svfloat64x2_t;
+typedef __clang_svint8x3_t svint8x3_t;
+typedef __clang_svint16x3_t svint16x3_t;
+typedef __clang_svint32x3_t svint32x3_t;
+typedef __clang_svint64x3_t svint64x3_t;
+typedef __clang_svuint8x3_t svuint8x3_t;
+typedef __clang_svuint16x3_t svuint16x3_t;
+typedef __clang_svuint32x3_t svuint32x3_t;
+typedef __clang_svuint64x3_t svuint64x3_t;
+typedef __clang_svfloat16x3_t svfloat16x3_t;
+typedef __clang_svfloat32x3_t svfloat32x3_t;
+typedef __clang_svfloat64x3_t svfloat64x3_t;
+typedef __clang_svint8x4_t svint8x4_t;
+typedef __clang_svint16x4_t svint16x4_t;
+typedef __clang_svint32x4_t svint32x4_t;
+typedef __clang_svint64x4_t svint64x4_t;
+typedef __clang_svuint8x4_t svuint8x4_t;
+typedef __clang_svuint16x4_t svuint16x4_t;
+typedef __clang_svuint32x4_t svuint32x4_t;
+typedef __clang_svuint64x4_t svuint64x4_t;
+typedef __clang_svfloat16x4_t svfloat16x4_t;
+typedef __clang_svfloat32x4_t svfloat32x4_t;
+typedef __clang_svfloat64x4_t svfloat64x4_t;
+typedef __SVBool_t svbool_t;
+
+#ifdef __ARM_FEATURE_SVE_BF16
+typedef __clang_svbfloat16x2_t svbfloat16x2_t;
+typedef __clang_svbfloat16x3_t svbfloat16x3_t;
+typedef __clang_svbfloat16x4_t svbfloat16x4_t;
+#endif
+typedef enum
+{
+ SV_POW2 = 0,
+ SV_VL1 = 1,
+ SV_VL2 = 2,
+ SV_VL3 = 3,
+ SV_VL4 = 4,
+ SV_VL5 = 5,
+ SV_VL6 = 6,
+ SV_VL7 = 7,
+ SV_VL8 = 8,
+ SV_VL16 = 9,
+ SV_VL32 = 10,
+ SV_VL64 = 11,
+ SV_VL128 = 12,
+ SV_VL256 = 13,
+ SV_MUL4 = 29,
+ SV_MUL3 = 30,
+ SV_ALL = 31
+} sv_pattern;
+
+typedef enum
+{
+ SV_PLDL1KEEP = 0,
+ SV_PLDL1STRM = 1,
+ SV_PLDL2KEEP = 2,
+ SV_PLDL2STRM = 3,
+ SV_PLDL3KEEP = 4,
+ SV_PLDL3STRM = 5,
+ SV_PSTL1KEEP = 8,
+ SV_PSTL1STRM = 9,
+ SV_PSTL2KEEP = 10,
+ SV_PSTL2STRM = 11,
+ SV_PSTL3KEEP = 12,
+ SV_PSTL3STRM = 13
+} sv_prfop;
+
+/* Function attributes */
+#define __aio static inline __attribute__((__always_inline__, __nodebug__, __overloadable__))
+
+#define svreinterpret_s8_s8(...) __builtin_sve_reinterpret_s8_s8(__VA_ARGS__)
+#define svreinterpret_s8_s16(...) __builtin_sve_reinterpret_s8_s16(__VA_ARGS__)
+#define svreinterpret_s8_s32(...) __builtin_sve_reinterpret_s8_s32(__VA_ARGS__)
+#define svreinterpret_s8_s64(...) __builtin_sve_reinterpret_s8_s64(__VA_ARGS__)
+#define svreinterpret_s8_u8(...) __builtin_sve_reinterpret_s8_u8(__VA_ARGS__)
+#define svreinterpret_s8_u16(...) __builtin_sve_reinterpret_s8_u16(__VA_ARGS__)
+#define svreinterpret_s8_u32(...) __builtin_sve_reinterpret_s8_u32(__VA_ARGS__)
+#define svreinterpret_s8_u64(...) __builtin_sve_reinterpret_s8_u64(__VA_ARGS__)
+#define svreinterpret_s8_f16(...) __builtin_sve_reinterpret_s8_f16(__VA_ARGS__)
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_s8_bf16(...) __builtin_sve_reinterpret_s8_bf16(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#define svreinterpret_s8_f32(...) __builtin_sve_reinterpret_s8_f32(__VA_ARGS__)
+#define svreinterpret_s8_f64(...) __builtin_sve_reinterpret_s8_f64(__VA_ARGS__)
+#define svreinterpret_s16_s8(...) __builtin_sve_reinterpret_s16_s8(__VA_ARGS__)
+#define svreinterpret_s16_s16(...) __builtin_sve_reinterpret_s16_s16(__VA_ARGS__)
+#define svreinterpret_s16_s32(...) __builtin_sve_reinterpret_s16_s32(__VA_ARGS__)
+#define svreinterpret_s16_s64(...) __builtin_sve_reinterpret_s16_s64(__VA_ARGS__)
+#define svreinterpret_s16_u8(...) __builtin_sve_reinterpret_s16_u8(__VA_ARGS__)
+#define svreinterpret_s16_u16(...) __builtin_sve_reinterpret_s16_u16(__VA_ARGS__)
+#define svreinterpret_s16_u32(...) __builtin_sve_reinterpret_s16_u32(__VA_ARGS__)
+#define svreinterpret_s16_u64(...) __builtin_sve_reinterpret_s16_u64(__VA_ARGS__)
+#define svreinterpret_s16_f16(...) __builtin_sve_reinterpret_s16_f16(__VA_ARGS__)
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_s16_bf16(...) __builtin_sve_reinterpret_s16_bf16(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#define svreinterpret_s16_f32(...) __builtin_sve_reinterpret_s16_f32(__VA_ARGS__)
+#define svreinterpret_s16_f64(...) __builtin_sve_reinterpret_s16_f64(__VA_ARGS__)
+#define svreinterpret_s32_s8(...) __builtin_sve_reinterpret_s32_s8(__VA_ARGS__)
+#define svreinterpret_s32_s16(...) __builtin_sve_reinterpret_s32_s16(__VA_ARGS__)
+#define svreinterpret_s32_s32(...) __builtin_sve_reinterpret_s32_s32(__VA_ARGS__)
+#define svreinterpret_s32_s64(...) __builtin_sve_reinterpret_s32_s64(__VA_ARGS__)
+#define svreinterpret_s32_u8(...) __builtin_sve_reinterpret_s32_u8(__VA_ARGS__)
+#define svreinterpret_s32_u16(...) __builtin_sve_reinterpret_s32_u16(__VA_ARGS__)
+#define svreinterpret_s32_u32(...) __builtin_sve_reinterpret_s32_u32(__VA_ARGS__)
+#define svreinterpret_s32_u64(...) __builtin_sve_reinterpret_s32_u64(__VA_ARGS__)
+#define svreinterpret_s32_f16(...) __builtin_sve_reinterpret_s32_f16(__VA_ARGS__)
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_s32_bf16(...) __builtin_sve_reinterpret_s32_bf16(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#define svreinterpret_s32_f32(...) __builtin_sve_reinterpret_s32_f32(__VA_ARGS__)
+#define svreinterpret_s32_f64(...) __builtin_sve_reinterpret_s32_f64(__VA_ARGS__)
+#define svreinterpret_s64_s8(...) __builtin_sve_reinterpret_s64_s8(__VA_ARGS__)
+#define svreinterpret_s64_s16(...) __builtin_sve_reinterpret_s64_s16(__VA_ARGS__)
+#define svreinterpret_s64_s32(...) __builtin_sve_reinterpret_s64_s32(__VA_ARGS__)
+#define svreinterpret_s64_s64(...) __builtin_sve_reinterpret_s64_s64(__VA_ARGS__)
+#define svreinterpret_s64_u8(...) __builtin_sve_reinterpret_s64_u8(__VA_ARGS__)
+#define svreinterpret_s64_u16(...) __builtin_sve_reinterpret_s64_u16(__VA_ARGS__)
+#define svreinterpret_s64_u32(...) __builtin_sve_reinterpret_s64_u32(__VA_ARGS__)
+#define svreinterpret_s64_u64(...) __builtin_sve_reinterpret_s64_u64(__VA_ARGS__)
+#define svreinterpret_s64_f16(...) __builtin_sve_reinterpret_s64_f16(__VA_ARGS__)
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_s64_bf16(...) __builtin_sve_reinterpret_s64_bf16(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#define svreinterpret_s64_f32(...) __builtin_sve_reinterpret_s64_f32(__VA_ARGS__)
+#define svreinterpret_s64_f64(...) __builtin_sve_reinterpret_s64_f64(__VA_ARGS__)
+#define svreinterpret_u8_s8(...) __builtin_sve_reinterpret_u8_s8(__VA_ARGS__)
+#define svreinterpret_u8_s16(...) __builtin_sve_reinterpret_u8_s16(__VA_ARGS__)
+#define svreinterpret_u8_s32(...) __builtin_sve_reinterpret_u8_s32(__VA_ARGS__)
+#define svreinterpret_u8_s64(...) __builtin_sve_reinterpret_u8_s64(__VA_ARGS__)
+#define svreinterpret_u8_u8(...) __builtin_sve_reinterpret_u8_u8(__VA_ARGS__)
+#define svreinterpret_u8_u16(...) __builtin_sve_reinterpret_u8_u16(__VA_ARGS__)
+#define svreinterpret_u8_u32(...) __builtin_sve_reinterpret_u8_u32(__VA_ARGS__)
+#define svreinterpret_u8_u64(...) __builtin_sve_reinterpret_u8_u64(__VA_ARGS__)
+#define svreinterpret_u8_f16(...) __builtin_sve_reinterpret_u8_f16(__VA_ARGS__)
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_u8_bf16(...) __builtin_sve_reinterpret_u8_bf16(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#define svreinterpret_u8_f32(...) __builtin_sve_reinterpret_u8_f32(__VA_ARGS__)
+#define svreinterpret_u8_f64(...) __builtin_sve_reinterpret_u8_f64(__VA_ARGS__)
+#define svreinterpret_u16_s8(...) __builtin_sve_reinterpret_u16_s8(__VA_ARGS__)
+#define svreinterpret_u16_s16(...) __builtin_sve_reinterpret_u16_s16(__VA_ARGS__)
+#define svreinterpret_u16_s32(...) __builtin_sve_reinterpret_u16_s32(__VA_ARGS__)
+#define svreinterpret_u16_s64(...) __builtin_sve_reinterpret_u16_s64(__VA_ARGS__)
+#define svreinterpret_u16_u8(...) __builtin_sve_reinterpret_u16_u8(__VA_ARGS__)
+#define svreinterpret_u16_u16(...) __builtin_sve_reinterpret_u16_u16(__VA_ARGS__)
+#define svreinterpret_u16_u32(...) __builtin_sve_reinterpret_u16_u32(__VA_ARGS__)
+#define svreinterpret_u16_u64(...) __builtin_sve_reinterpret_u16_u64(__VA_ARGS__)
+#define svreinterpret_u16_f16(...) __builtin_sve_reinterpret_u16_f16(__VA_ARGS__)
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_u16_bf16(...) __builtin_sve_reinterpret_u16_bf16(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#define svreinterpret_u16_f32(...) __builtin_sve_reinterpret_u16_f32(__VA_ARGS__)
+#define svreinterpret_u16_f64(...) __builtin_sve_reinterpret_u16_f64(__VA_ARGS__)
+#define svreinterpret_u32_s8(...) __builtin_sve_reinterpret_u32_s8(__VA_ARGS__)
+#define svreinterpret_u32_s16(...) __builtin_sve_reinterpret_u32_s16(__VA_ARGS__)
+#define svreinterpret_u32_s32(...) __builtin_sve_reinterpret_u32_s32(__VA_ARGS__)
+#define svreinterpret_u32_s64(...) __builtin_sve_reinterpret_u32_s64(__VA_ARGS__)
+#define svreinterpret_u32_u8(...) __builtin_sve_reinterpret_u32_u8(__VA_ARGS__)
+#define svreinterpret_u32_u16(...) __builtin_sve_reinterpret_u32_u16(__VA_ARGS__)
+#define svreinterpret_u32_u32(...) __builtin_sve_reinterpret_u32_u32(__VA_ARGS__)
+#define svreinterpret_u32_u64(...) __builtin_sve_reinterpret_u32_u64(__VA_ARGS__)
+#define svreinterpret_u32_f16(...) __builtin_sve_reinterpret_u32_f16(__VA_ARGS__)
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_u32_bf16(...) __builtin_sve_reinterpret_u32_bf16(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#define svreinterpret_u32_f32(...) __builtin_sve_reinterpret_u32_f32(__VA_ARGS__)
+#define svreinterpret_u32_f64(...) __builtin_sve_reinterpret_u32_f64(__VA_ARGS__)
+#define svreinterpret_u64_s8(...) __builtin_sve_reinterpret_u64_s8(__VA_ARGS__)
+#define svreinterpret_u64_s16(...) __builtin_sve_reinterpret_u64_s16(__VA_ARGS__)
+#define svreinterpret_u64_s32(...) __builtin_sve_reinterpret_u64_s32(__VA_ARGS__)
+#define svreinterpret_u64_s64(...) __builtin_sve_reinterpret_u64_s64(__VA_ARGS__)
+#define svreinterpret_u64_u8(...) __builtin_sve_reinterpret_u64_u8(__VA_ARGS__)
+#define svreinterpret_u64_u16(...) __builtin_sve_reinterpret_u64_u16(__VA_ARGS__)
+#define svreinterpret_u64_u32(...) __builtin_sve_reinterpret_u64_u32(__VA_ARGS__)
+#define svreinterpret_u64_u64(...) __builtin_sve_reinterpret_u64_u64(__VA_ARGS__)
+#define svreinterpret_u64_f16(...) __builtin_sve_reinterpret_u64_f16(__VA_ARGS__)
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_u64_bf16(...) __builtin_sve_reinterpret_u64_bf16(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#define svreinterpret_u64_f32(...) __builtin_sve_reinterpret_u64_f32(__VA_ARGS__)
+#define svreinterpret_u64_f64(...) __builtin_sve_reinterpret_u64_f64(__VA_ARGS__)
+#define svreinterpret_f16_s8(...) __builtin_sve_reinterpret_f16_s8(__VA_ARGS__)
+#define svreinterpret_f16_s16(...) __builtin_sve_reinterpret_f16_s16(__VA_ARGS__)
+#define svreinterpret_f16_s32(...) __builtin_sve_reinterpret_f16_s32(__VA_ARGS__)
+#define svreinterpret_f16_s64(...) __builtin_sve_reinterpret_f16_s64(__VA_ARGS__)
+#define svreinterpret_f16_u8(...) __builtin_sve_reinterpret_f16_u8(__VA_ARGS__)
+#define svreinterpret_f16_u16(...) __builtin_sve_reinterpret_f16_u16(__VA_ARGS__)
+#define svreinterpret_f16_u32(...) __builtin_sve_reinterpret_f16_u32(__VA_ARGS__)
+#define svreinterpret_f16_u64(...) __builtin_sve_reinterpret_f16_u64(__VA_ARGS__)
+#define svreinterpret_f16_f16(...) __builtin_sve_reinterpret_f16_f16(__VA_ARGS__)
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_f16_bf16(...) __builtin_sve_reinterpret_f16_bf16(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#define svreinterpret_f16_f32(...) __builtin_sve_reinterpret_f16_f32(__VA_ARGS__)
+#define svreinterpret_f16_f64(...) __builtin_sve_reinterpret_f16_f64(__VA_ARGS__)
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_bf16_s8(...) __builtin_sve_reinterpret_bf16_s8(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_bf16_s16(...) __builtin_sve_reinterpret_bf16_s16(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_bf16_s32(...) __builtin_sve_reinterpret_bf16_s32(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_bf16_s64(...) __builtin_sve_reinterpret_bf16_s64(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_bf16_u8(...) __builtin_sve_reinterpret_bf16_u8(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_bf16_u16(...) __builtin_sve_reinterpret_bf16_u16(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_bf16_u32(...) __builtin_sve_reinterpret_bf16_u32(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_bf16_u64(...) __builtin_sve_reinterpret_bf16_u64(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_bf16_f16(...) __builtin_sve_reinterpret_bf16_f16(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_bf16_bf16(...) __builtin_sve_reinterpret_bf16_bf16(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_bf16_f32(...) __builtin_sve_reinterpret_bf16_f32(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_bf16_f64(...) __builtin_sve_reinterpret_bf16_f64(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#define svreinterpret_f32_s8(...) __builtin_sve_reinterpret_f32_s8(__VA_ARGS__)
+#define svreinterpret_f32_s16(...) __builtin_sve_reinterpret_f32_s16(__VA_ARGS__)
+#define svreinterpret_f32_s32(...) __builtin_sve_reinterpret_f32_s32(__VA_ARGS__)
+#define svreinterpret_f32_s64(...) __builtin_sve_reinterpret_f32_s64(__VA_ARGS__)
+#define svreinterpret_f32_u8(...) __builtin_sve_reinterpret_f32_u8(__VA_ARGS__)
+#define svreinterpret_f32_u16(...) __builtin_sve_reinterpret_f32_u16(__VA_ARGS__)
+#define svreinterpret_f32_u32(...) __builtin_sve_reinterpret_f32_u32(__VA_ARGS__)
+#define svreinterpret_f32_u64(...) __builtin_sve_reinterpret_f32_u64(__VA_ARGS__)
+#define svreinterpret_f32_f16(...) __builtin_sve_reinterpret_f32_f16(__VA_ARGS__)
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_f32_bf16(...) __builtin_sve_reinterpret_f32_bf16(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#define svreinterpret_f32_f32(...) __builtin_sve_reinterpret_f32_f32(__VA_ARGS__)
+#define svreinterpret_f32_f64(...) __builtin_sve_reinterpret_f32_f64(__VA_ARGS__)
+#define svreinterpret_f64_s8(...) __builtin_sve_reinterpret_f64_s8(__VA_ARGS__)
+#define svreinterpret_f64_s16(...) __builtin_sve_reinterpret_f64_s16(__VA_ARGS__)
+#define svreinterpret_f64_s32(...) __builtin_sve_reinterpret_f64_s32(__VA_ARGS__)
+#define svreinterpret_f64_s64(...) __builtin_sve_reinterpret_f64_s64(__VA_ARGS__)
+#define svreinterpret_f64_u8(...) __builtin_sve_reinterpret_f64_u8(__VA_ARGS__)
+#define svreinterpret_f64_u16(...) __builtin_sve_reinterpret_f64_u16(__VA_ARGS__)
+#define svreinterpret_f64_u32(...) __builtin_sve_reinterpret_f64_u32(__VA_ARGS__)
+#define svreinterpret_f64_u64(...) __builtin_sve_reinterpret_f64_u64(__VA_ARGS__)
+#define svreinterpret_f64_f16(...) __builtin_sve_reinterpret_f64_f16(__VA_ARGS__)
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svreinterpret_f64_bf16(...) __builtin_sve_reinterpret_f64_bf16(__VA_ARGS__)
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#define svreinterpret_f64_f32(...) __builtin_sve_reinterpret_f64_f32(__VA_ARGS__)
+#define svreinterpret_f64_f64(...) __builtin_sve_reinterpret_f64_f64(__VA_ARGS__)
+__aio svint8_t svreinterpret_s8(svint8_t op) {
+ return __builtin_sve_reinterpret_s8_s8(op);
+}
+
+__aio svint8_t svreinterpret_s8(svint16_t op) {
+ return __builtin_sve_reinterpret_s8_s16(op);
+}
+
+__aio svint8_t svreinterpret_s8(svint32_t op) {
+ return __builtin_sve_reinterpret_s8_s32(op);
+}
+
+__aio svint8_t svreinterpret_s8(svint64_t op) {
+ return __builtin_sve_reinterpret_s8_s64(op);
+}
+
+__aio svint8_t svreinterpret_s8(svuint8_t op) {
+ return __builtin_sve_reinterpret_s8_u8(op);
+}
+
+__aio svint8_t svreinterpret_s8(svuint16_t op) {
+ return __builtin_sve_reinterpret_s8_u16(op);
+}
+
+__aio svint8_t svreinterpret_s8(svuint32_t op) {
+ return __builtin_sve_reinterpret_s8_u32(op);
+}
+
+__aio svint8_t svreinterpret_s8(svuint64_t op) {
+ return __builtin_sve_reinterpret_s8_u64(op);
+}
+
+__aio svint8_t svreinterpret_s8(svfloat16_t op) {
+ return __builtin_sve_reinterpret_s8_f16(op);
+}
+
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svint8_t svreinterpret_s8(svbfloat16_t op) {
+ return __builtin_sve_reinterpret_s8_bf16(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+__aio svint8_t svreinterpret_s8(svfloat32_t op) {
+ return __builtin_sve_reinterpret_s8_f32(op);
+}
+
+__aio svint8_t svreinterpret_s8(svfloat64_t op) {
+ return __builtin_sve_reinterpret_s8_f64(op);
+}
+
+__aio svint16_t svreinterpret_s16(svint8_t op) {
+ return __builtin_sve_reinterpret_s16_s8(op);
+}
+
+__aio svint16_t svreinterpret_s16(svint16_t op) {
+ return __builtin_sve_reinterpret_s16_s16(op);
+}
+
+__aio svint16_t svreinterpret_s16(svint32_t op) {
+ return __builtin_sve_reinterpret_s16_s32(op);
+}
+
+__aio svint16_t svreinterpret_s16(svint64_t op) {
+ return __builtin_sve_reinterpret_s16_s64(op);
+}
+
+__aio svint16_t svreinterpret_s16(svuint8_t op) {
+ return __builtin_sve_reinterpret_s16_u8(op);
+}
+
+__aio svint16_t svreinterpret_s16(svuint16_t op) {
+ return __builtin_sve_reinterpret_s16_u16(op);
+}
+
+__aio svint16_t svreinterpret_s16(svuint32_t op) {
+ return __builtin_sve_reinterpret_s16_u32(op);
+}
+
+__aio svint16_t svreinterpret_s16(svuint64_t op) {
+ return __builtin_sve_reinterpret_s16_u64(op);
+}
+
+__aio svint16_t svreinterpret_s16(svfloat16_t op) {
+ return __builtin_sve_reinterpret_s16_f16(op);
+}
+
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svint16_t svreinterpret_s16(svbfloat16_t op) {
+ return __builtin_sve_reinterpret_s16_bf16(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+__aio svint16_t svreinterpret_s16(svfloat32_t op) {
+ return __builtin_sve_reinterpret_s16_f32(op);
+}
+
+__aio svint16_t svreinterpret_s16(svfloat64_t op) {
+ return __builtin_sve_reinterpret_s16_f64(op);
+}
+
+__aio svint32_t svreinterpret_s32(svint8_t op) {
+ return __builtin_sve_reinterpret_s32_s8(op);
+}
+
+__aio svint32_t svreinterpret_s32(svint16_t op) {
+ return __builtin_sve_reinterpret_s32_s16(op);
+}
+
+__aio svint32_t svreinterpret_s32(svint32_t op) {
+ return __builtin_sve_reinterpret_s32_s32(op);
+}
+
+__aio svint32_t svreinterpret_s32(svint64_t op) {
+ return __builtin_sve_reinterpret_s32_s64(op);
+}
+
+__aio svint32_t svreinterpret_s32(svuint8_t op) {
+ return __builtin_sve_reinterpret_s32_u8(op);
+}
+
+__aio svint32_t svreinterpret_s32(svuint16_t op) {
+ return __builtin_sve_reinterpret_s32_u16(op);
+}
+
+__aio svint32_t svreinterpret_s32(svuint32_t op) {
+ return __builtin_sve_reinterpret_s32_u32(op);
+}
+
+__aio svint32_t svreinterpret_s32(svuint64_t op) {
+ return __builtin_sve_reinterpret_s32_u64(op);
+}
+
+__aio svint32_t svreinterpret_s32(svfloat16_t op) {
+ return __builtin_sve_reinterpret_s32_f16(op);
+}
+
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svint32_t svreinterpret_s32(svbfloat16_t op) {
+ return __builtin_sve_reinterpret_s32_bf16(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+__aio svint32_t svreinterpret_s32(svfloat32_t op) {
+ return __builtin_sve_reinterpret_s32_f32(op);
+}
+
+__aio svint32_t svreinterpret_s32(svfloat64_t op) {
+ return __builtin_sve_reinterpret_s32_f64(op);
+}
+
+__aio svint64_t svreinterpret_s64(svint8_t op) {
+ return __builtin_sve_reinterpret_s64_s8(op);
+}
+
+__aio svint64_t svreinterpret_s64(svint16_t op) {
+ return __builtin_sve_reinterpret_s64_s16(op);
+}
+
+__aio svint64_t svreinterpret_s64(svint32_t op) {
+ return __builtin_sve_reinterpret_s64_s32(op);
+}
+
+__aio svint64_t svreinterpret_s64(svint64_t op) {
+ return __builtin_sve_reinterpret_s64_s64(op);
+}
+
+__aio svint64_t svreinterpret_s64(svuint8_t op) {
+ return __builtin_sve_reinterpret_s64_u8(op);
+}
+
+__aio svint64_t svreinterpret_s64(svuint16_t op) {
+ return __builtin_sve_reinterpret_s64_u16(op);
+}
+
+__aio svint64_t svreinterpret_s64(svuint32_t op) {
+ return __builtin_sve_reinterpret_s64_u32(op);
+}
+
+__aio svint64_t svreinterpret_s64(svuint64_t op) {
+ return __builtin_sve_reinterpret_s64_u64(op);
+}
+
+__aio svint64_t svreinterpret_s64(svfloat16_t op) {
+ return __builtin_sve_reinterpret_s64_f16(op);
+}
+
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svint64_t svreinterpret_s64(svbfloat16_t op) {
+ return __builtin_sve_reinterpret_s64_bf16(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+__aio svint64_t svreinterpret_s64(svfloat32_t op) {
+ return __builtin_sve_reinterpret_s64_f32(op);
+}
+
+__aio svint64_t svreinterpret_s64(svfloat64_t op) {
+ return __builtin_sve_reinterpret_s64_f64(op);
+}
+
+__aio svuint8_t svreinterpret_u8(svint8_t op) {
+ return __builtin_sve_reinterpret_u8_s8(op);
+}
+
+__aio svuint8_t svreinterpret_u8(svint16_t op) {
+ return __builtin_sve_reinterpret_u8_s16(op);
+}
+
+__aio svuint8_t svreinterpret_u8(svint32_t op) {
+ return __builtin_sve_reinterpret_u8_s32(op);
+}
+
+__aio svuint8_t svreinterpret_u8(svint64_t op) {
+ return __builtin_sve_reinterpret_u8_s64(op);
+}
+
+__aio svuint8_t svreinterpret_u8(svuint8_t op) {
+ return __builtin_sve_reinterpret_u8_u8(op);
+}
+
+__aio svuint8_t svreinterpret_u8(svuint16_t op) {
+ return __builtin_sve_reinterpret_u8_u16(op);
+}
+
+__aio svuint8_t svreinterpret_u8(svuint32_t op) {
+ return __builtin_sve_reinterpret_u8_u32(op);
+}
+
+__aio svuint8_t svreinterpret_u8(svuint64_t op) {
+ return __builtin_sve_reinterpret_u8_u64(op);
+}
+
+__aio svuint8_t svreinterpret_u8(svfloat16_t op) {
+ return __builtin_sve_reinterpret_u8_f16(op);
+}
+
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svuint8_t svreinterpret_u8(svbfloat16_t op) {
+ return __builtin_sve_reinterpret_u8_bf16(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+__aio svuint8_t svreinterpret_u8(svfloat32_t op) {
+ return __builtin_sve_reinterpret_u8_f32(op);
+}
+
+__aio svuint8_t svreinterpret_u8(svfloat64_t op) {
+ return __builtin_sve_reinterpret_u8_f64(op);
+}
+
+__aio svuint16_t svreinterpret_u16(svint8_t op) {
+ return __builtin_sve_reinterpret_u16_s8(op);
+}
+
+__aio svuint16_t svreinterpret_u16(svint16_t op) {
+ return __builtin_sve_reinterpret_u16_s16(op);
+}
+
+__aio svuint16_t svreinterpret_u16(svint32_t op) {
+ return __builtin_sve_reinterpret_u16_s32(op);
+}
+
+__aio svuint16_t svreinterpret_u16(svint64_t op) {
+ return __builtin_sve_reinterpret_u16_s64(op);
+}
+
+__aio svuint16_t svreinterpret_u16(svuint8_t op) {
+ return __builtin_sve_reinterpret_u16_u8(op);
+}
+
+__aio svuint16_t svreinterpret_u16(svuint16_t op) {
+ return __builtin_sve_reinterpret_u16_u16(op);
+}
+
+__aio svuint16_t svreinterpret_u16(svuint32_t op) {
+ return __builtin_sve_reinterpret_u16_u32(op);
+}
+
+__aio svuint16_t svreinterpret_u16(svuint64_t op) {
+ return __builtin_sve_reinterpret_u16_u64(op);
+}
+
+__aio svuint16_t svreinterpret_u16(svfloat16_t op) {
+ return __builtin_sve_reinterpret_u16_f16(op);
+}
+
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svuint16_t svreinterpret_u16(svbfloat16_t op) {
+ return __builtin_sve_reinterpret_u16_bf16(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+__aio svuint16_t svreinterpret_u16(svfloat32_t op) {
+ return __builtin_sve_reinterpret_u16_f32(op);
+}
+
+__aio svuint16_t svreinterpret_u16(svfloat64_t op) {
+ return __builtin_sve_reinterpret_u16_f64(op);
+}
+
+__aio svuint32_t svreinterpret_u32(svint8_t op) {
+ return __builtin_sve_reinterpret_u32_s8(op);
+}
+
+__aio svuint32_t svreinterpret_u32(svint16_t op) {
+ return __builtin_sve_reinterpret_u32_s16(op);
+}
+
+__aio svuint32_t svreinterpret_u32(svint32_t op) {
+ return __builtin_sve_reinterpret_u32_s32(op);
+}
+
+__aio svuint32_t svreinterpret_u32(svint64_t op) {
+ return __builtin_sve_reinterpret_u32_s64(op);
+}
+
+__aio svuint32_t svreinterpret_u32(svuint8_t op) {
+ return __builtin_sve_reinterpret_u32_u8(op);
+}
+
+__aio svuint32_t svreinterpret_u32(svuint16_t op) {
+ return __builtin_sve_reinterpret_u32_u16(op);
+}
+
+__aio svuint32_t svreinterpret_u32(svuint32_t op) {
+ return __builtin_sve_reinterpret_u32_u32(op);
+}
+
+__aio svuint32_t svreinterpret_u32(svuint64_t op) {
+ return __builtin_sve_reinterpret_u32_u64(op);
+}
+
+__aio svuint32_t svreinterpret_u32(svfloat16_t op) {
+ return __builtin_sve_reinterpret_u32_f16(op);
+}
+
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svuint32_t svreinterpret_u32(svbfloat16_t op) {
+ return __builtin_sve_reinterpret_u32_bf16(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+__aio svuint32_t svreinterpret_u32(svfloat32_t op) {
+ return __builtin_sve_reinterpret_u32_f32(op);
+}
+
+__aio svuint32_t svreinterpret_u32(svfloat64_t op) {
+ return __builtin_sve_reinterpret_u32_f64(op);
+}
+
+__aio svuint64_t svreinterpret_u64(svint8_t op) {
+ return __builtin_sve_reinterpret_u64_s8(op);
+}
+
+__aio svuint64_t svreinterpret_u64(svint16_t op) {
+ return __builtin_sve_reinterpret_u64_s16(op);
+}
+
+__aio svuint64_t svreinterpret_u64(svint32_t op) {
+ return __builtin_sve_reinterpret_u64_s32(op);
+}
+
+__aio svuint64_t svreinterpret_u64(svint64_t op) {
+ return __builtin_sve_reinterpret_u64_s64(op);
+}
+
+__aio svuint64_t svreinterpret_u64(svuint8_t op) {
+ return __builtin_sve_reinterpret_u64_u8(op);
+}
+
+__aio svuint64_t svreinterpret_u64(svuint16_t op) {
+ return __builtin_sve_reinterpret_u64_u16(op);
+}
+
+__aio svuint64_t svreinterpret_u64(svuint32_t op) {
+ return __builtin_sve_reinterpret_u64_u32(op);
+}
+
+__aio svuint64_t svreinterpret_u64(svuint64_t op) {
+ return __builtin_sve_reinterpret_u64_u64(op);
+}
+
+__aio svuint64_t svreinterpret_u64(svfloat16_t op) {
+ return __builtin_sve_reinterpret_u64_f16(op);
+}
+
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svuint64_t svreinterpret_u64(svbfloat16_t op) {
+ return __builtin_sve_reinterpret_u64_bf16(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+__aio svuint64_t svreinterpret_u64(svfloat32_t op) {
+ return __builtin_sve_reinterpret_u64_f32(op);
+}
+
+__aio svuint64_t svreinterpret_u64(svfloat64_t op) {
+ return __builtin_sve_reinterpret_u64_f64(op);
+}
+
+__aio svfloat16_t svreinterpret_f16(svint8_t op) {
+ return __builtin_sve_reinterpret_f16_s8(op);
+}
+
+__aio svfloat16_t svreinterpret_f16(svint16_t op) {
+ return __builtin_sve_reinterpret_f16_s16(op);
+}
+
+__aio svfloat16_t svreinterpret_f16(svint32_t op) {
+ return __builtin_sve_reinterpret_f16_s32(op);
+}
+
+__aio svfloat16_t svreinterpret_f16(svint64_t op) {
+ return __builtin_sve_reinterpret_f16_s64(op);
+}
+
+__aio svfloat16_t svreinterpret_f16(svuint8_t op) {
+ return __builtin_sve_reinterpret_f16_u8(op);
+}
+
+__aio svfloat16_t svreinterpret_f16(svuint16_t op) {
+ return __builtin_sve_reinterpret_f16_u16(op);
+}
+
+__aio svfloat16_t svreinterpret_f16(svuint32_t op) {
+ return __builtin_sve_reinterpret_f16_u32(op);
+}
+
+__aio svfloat16_t svreinterpret_f16(svuint64_t op) {
+ return __builtin_sve_reinterpret_f16_u64(op);
+}
+
+__aio svfloat16_t svreinterpret_f16(svfloat16_t op) {
+ return __builtin_sve_reinterpret_f16_f16(op);
+}
+
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svfloat16_t svreinterpret_f16(svbfloat16_t op) {
+ return __builtin_sve_reinterpret_f16_bf16(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+__aio svfloat16_t svreinterpret_f16(svfloat32_t op) {
+ return __builtin_sve_reinterpret_f16_f32(op);
+}
+
+__aio svfloat16_t svreinterpret_f16(svfloat64_t op) {
+ return __builtin_sve_reinterpret_f16_f64(op);
+}
+
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svbfloat16_t svreinterpret_bf16(svint8_t op) {
+ return __builtin_sve_reinterpret_bf16_s8(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svbfloat16_t svreinterpret_bf16(svint16_t op) {
+ return __builtin_sve_reinterpret_bf16_s16(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svbfloat16_t svreinterpret_bf16(svint32_t op) {
+ return __builtin_sve_reinterpret_bf16_s32(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svbfloat16_t svreinterpret_bf16(svint64_t op) {
+ return __builtin_sve_reinterpret_bf16_s64(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svbfloat16_t svreinterpret_bf16(svuint8_t op) {
+ return __builtin_sve_reinterpret_bf16_u8(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svbfloat16_t svreinterpret_bf16(svuint16_t op) {
+ return __builtin_sve_reinterpret_bf16_u16(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svbfloat16_t svreinterpret_bf16(svuint32_t op) {
+ return __builtin_sve_reinterpret_bf16_u32(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svbfloat16_t svreinterpret_bf16(svuint64_t op) {
+ return __builtin_sve_reinterpret_bf16_u64(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svbfloat16_t svreinterpret_bf16(svfloat16_t op) {
+ return __builtin_sve_reinterpret_bf16_f16(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svbfloat16_t svreinterpret_bf16(svbfloat16_t op) {
+ return __builtin_sve_reinterpret_bf16_bf16(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svbfloat16_t svreinterpret_bf16(svfloat32_t op) {
+ return __builtin_sve_reinterpret_bf16_f32(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svbfloat16_t svreinterpret_bf16(svfloat64_t op) {
+ return __builtin_sve_reinterpret_bf16_f64(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+__aio svfloat32_t svreinterpret_f32(svint8_t op) {
+ return __builtin_sve_reinterpret_f32_s8(op);
+}
+
+__aio svfloat32_t svreinterpret_f32(svint16_t op) {
+ return __builtin_sve_reinterpret_f32_s16(op);
+}
+
+__aio svfloat32_t svreinterpret_f32(svint32_t op) {
+ return __builtin_sve_reinterpret_f32_s32(op);
+}
+
+__aio svfloat32_t svreinterpret_f32(svint64_t op) {
+ return __builtin_sve_reinterpret_f32_s64(op);
+}
+
+__aio svfloat32_t svreinterpret_f32(svuint8_t op) {
+ return __builtin_sve_reinterpret_f32_u8(op);
+}
+
+__aio svfloat32_t svreinterpret_f32(svuint16_t op) {
+ return __builtin_sve_reinterpret_f32_u16(op);
+}
+
+__aio svfloat32_t svreinterpret_f32(svuint32_t op) {
+ return __builtin_sve_reinterpret_f32_u32(op);
+}
+
+__aio svfloat32_t svreinterpret_f32(svuint64_t op) {
+ return __builtin_sve_reinterpret_f32_u64(op);
+}
+
+__aio svfloat32_t svreinterpret_f32(svfloat16_t op) {
+ return __builtin_sve_reinterpret_f32_f16(op);
+}
+
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svfloat32_t svreinterpret_f32(svbfloat16_t op) {
+ return __builtin_sve_reinterpret_f32_bf16(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+__aio svfloat32_t svreinterpret_f32(svfloat32_t op) {
+ return __builtin_sve_reinterpret_f32_f32(op);
+}
+
+__aio svfloat32_t svreinterpret_f32(svfloat64_t op) {
+ return __builtin_sve_reinterpret_f32_f64(op);
+}
+
+__aio svfloat64_t svreinterpret_f64(svint8_t op) {
+ return __builtin_sve_reinterpret_f64_s8(op);
+}
+
+__aio svfloat64_t svreinterpret_f64(svint16_t op) {
+ return __builtin_sve_reinterpret_f64_s16(op);
+}
+
+__aio svfloat64_t svreinterpret_f64(svint32_t op) {
+ return __builtin_sve_reinterpret_f64_s32(op);
+}
+
+__aio svfloat64_t svreinterpret_f64(svint64_t op) {
+ return __builtin_sve_reinterpret_f64_s64(op);
+}
+
+__aio svfloat64_t svreinterpret_f64(svuint8_t op) {
+ return __builtin_sve_reinterpret_f64_u8(op);
+}
+
+__aio svfloat64_t svreinterpret_f64(svuint16_t op) {
+ return __builtin_sve_reinterpret_f64_u16(op);
+}
+
+__aio svfloat64_t svreinterpret_f64(svuint32_t op) {
+ return __builtin_sve_reinterpret_f64_u32(op);
+}
+
+__aio svfloat64_t svreinterpret_f64(svuint64_t op) {
+ return __builtin_sve_reinterpret_f64_u64(op);
+}
+
+__aio svfloat64_t svreinterpret_f64(svfloat16_t op) {
+ return __builtin_sve_reinterpret_f64_f16(op);
+}
+
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio svfloat64_t svreinterpret_f64(svbfloat16_t op) {
+ return __builtin_sve_reinterpret_f64_bf16(op);
+}
+
+#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
+__aio svfloat64_t svreinterpret_f64(svfloat32_t op) {
+ return __builtin_sve_reinterpret_f64_f32(op);
+}
+
+__aio svfloat64_t svreinterpret_f64(svfloat64_t op) {
+ return __builtin_sve_reinterpret_f64_f64(op);
+}
+
+#define svabd_n_f64_m(...) __builtin_sve_svabd_n_f64_m(__VA_ARGS__)
+#define svabd_n_f32_m(...) __builtin_sve_svabd_n_f32_m(__VA_ARGS__)
+#define svabd_n_f16_m(...) __builtin_sve_svabd_n_f16_m(__VA_ARGS__)
+#define svabd_n_f64_x(...) __builtin_sve_svabd_n_f64_x(__VA_ARGS__)
+#define svabd_n_f32_x(...) __builtin_sve_svabd_n_f32_x(__VA_ARGS__)
+#define svabd_n_f16_x(...) __builtin_sve_svabd_n_f16_x(__VA_ARGS__)
+#define svabd_n_f64_z(...) __builtin_sve_svabd_n_f64_z(__VA_ARGS__)
+#define svabd_n_f32_z(...) __builtin_sve_svabd_n_f32_z(__VA_ARGS__)
+#define svabd_n_f16_z(...) __builtin_sve_svabd_n_f16_z(__VA_ARGS__)
+#define svabd_n_s8_m(...) __builtin_sve_svabd_n_s8_m(__VA_ARGS__)
+#define svabd_n_s32_m(...) __builtin_sve_svabd_n_s32_m(__VA_ARGS__)
+#define svabd_n_s64_m(...) __builtin_sve_svabd_n_s64_m(__VA_ARGS__)
+#define svabd_n_s16_m(...) __builtin_sve_svabd_n_s16_m(__VA_ARGS__)
+#define svabd_n_s8_x(...) __builtin_sve_svabd_n_s8_x(__VA_ARGS__)
+#define svabd_n_s32_x(...) __builtin_sve_svabd_n_s32_x(__VA_ARGS__)
+#define svabd_n_s64_x(...) __builtin_sve_svabd_n_s64_x(__VA_ARGS__)
+#define svabd_n_s16_x(...) __builtin_sve_svabd_n_s16_x(__VA_ARGS__)
+#define svabd_n_s8_z(...) __builtin_sve_svabd_n_s8_z(__VA_ARGS__)
+#define svabd_n_s32_z(...) __builtin_sve_svabd_n_s32_z(__VA_ARGS__)
+#define svabd_n_s64_z(...) __builtin_sve_svabd_n_s64_z(__VA_ARGS__)
+#define svabd_n_s16_z(...) __builtin_sve_svabd_n_s16_z(__VA_ARGS__)
+#define svabd_n_u8_m(...) __builtin_sve_svabd_n_u8_m(__VA_ARGS__)
+#define svabd_n_u32_m(...) __builtin_sve_svabd_n_u32_m(__VA_ARGS__)
+#define svabd_n_u64_m(...) __builtin_sve_svabd_n_u64_m(__VA_ARGS__)
+#define svabd_n_u16_m(...) __builtin_sve_svabd_n_u16_m(__VA_ARGS__)
+#define svabd_n_u8_x(...) __builtin_sve_svabd_n_u8_x(__VA_ARGS__)
+#define svabd_n_u32_x(...) __builtin_sve_svabd_n_u32_x(__VA_ARGS__)
+#define svabd_n_u64_x(...) __builtin_sve_svabd_n_u64_x(__VA_ARGS__)
+#define svabd_n_u16_x(...) __builtin_sve_svabd_n_u16_x(__VA_ARGS__)
+#define svabd_n_u8_z(...) __builtin_sve_svabd_n_u8_z(__VA_ARGS__)
+#define svabd_n_u32_z(...) __builtin_sve_svabd_n_u32_z(__VA_ARGS__)
+#define svabd_n_u64_z(...) __builtin_sve_svabd_n_u64_z(__VA_ARGS__)
+#define svabd_n_u16_z(...) __builtin_sve_svabd_n_u16_z(__VA_ARGS__)
+#define svabd_f64_m(...) __builtin_sve_svabd_f64_m(__VA_ARGS__)
+#define svabd_f32_m(...) __builtin_sve_svabd_f32_m(__VA_ARGS__)
+#define svabd_f16_m(...) __builtin_sve_svabd_f16_m(__VA_ARGS__)
+#define svabd_f64_x(...) __builtin_sve_svabd_f64_x(__VA_ARGS__)
+#define svabd_f32_x(...) __builtin_sve_svabd_f32_x(__VA_ARGS__)
+#define svabd_f16_x(...) __builtin_sve_svabd_f16_x(__VA_ARGS__)
+#define svabd_f64_z(...) __builtin_sve_svabd_f64_z(__VA_ARGS__)
+#define svabd_f32_z(...) __builtin_sve_svabd_f32_z(__VA_ARGS__)
+#define svabd_f16_z(...) __builtin_sve_svabd_f16_z(__VA_ARGS__)
+#define svabd_s8_m(...) __builtin_sve_svabd_s8_m(__VA_ARGS__)
+#define svabd_s32_m(...) __builtin_sve_svabd_s32_m(__VA_ARGS__)
+#define svabd_s64_m(...) __builtin_sve_svabd_s64_m(__VA_ARGS__)
+#define svabd_s16_m(...) __builtin_sve_svabd_s16_m(__VA_ARGS__)
+#define svabd_s8_x(...) __builtin_sve_svabd_s8_x(__VA_ARGS__)
+#define svabd_s32_x(...) __builtin_sve_svabd_s32_x(__VA_ARGS__)
+#define svabd_s64_x(...) __builtin_sve_svabd_s64_x(__VA_ARGS__)
+#define svabd_s16_x(...) __builtin_sve_svabd_s16_x(__VA_ARGS__)
+#define svabd_s8_z(...) __builtin_sve_svabd_s8_z(__VA_ARGS__)
+#define svabd_s32_z(...) __builtin_sve_svabd_s32_z(__VA_ARGS__)
+#define svabd_s64_z(...) __builtin_sve_svabd_s64_z(__VA_ARGS__)
+#define svabd_s16_z(...) __builtin_sve_svabd_s16_z(__VA_ARGS__)
+#define svabd_u8_m(...) __builtin_sve_svabd_u8_m(__VA_ARGS__)
+#define svabd_u32_m(...) __builtin_sve_svabd_u32_m(__VA_ARGS__)
+#define svabd_u64_m(...) __builtin_sve_svabd_u64_m(__VA_ARGS__)
+#define svabd_u16_m(...) __builtin_sve_svabd_u16_m(__VA_ARGS__)
+#define svabd_u8_x(...) __builtin_sve_svabd_u8_x(__VA_ARGS__)
+#define svabd_u32_x(...) __builtin_sve_svabd_u32_x(__VA_ARGS__)
+#define svabd_u64_x(...) __builtin_sve_svabd_u64_x(__VA_ARGS__)
+#define svabd_u16_x(...) __builtin_sve_svabd_u16_x(__VA_ARGS__)
+#define svabd_u8_z(...) __builtin_sve_svabd_u8_z(__VA_ARGS__)
+#define svabd_u32_z(...) __builtin_sve_svabd_u32_z(__VA_ARGS__)
+#define svabd_u64_z(...) __builtin_sve_svabd_u64_z(__VA_ARGS__)
+#define svabd_u16_z(...) __builtin_sve_svabd_u16_z(__VA_ARGS__)
+#define svabs_f64_m(...) __builtin_sve_svabs_f64_m(__VA_ARGS__)
+#define svabs_f32_m(...) __builtin_sve_svabs_f32_m(__VA_ARGS__)
+#define svabs_f16_m(...) __builtin_sve_svabs_f16_m(__VA_ARGS__)
+#define svabs_f64_x(...) __builtin_sve_svabs_f64_x(__VA_ARGS__)
+#define svabs_f32_x(...) __builtin_sve_svabs_f32_x(__VA_ARGS__)
+#define svabs_f16_x(...) __builtin_sve_svabs_f16_x(__VA_ARGS__)
+#define svabs_f64_z(...) __builtin_sve_svabs_f64_z(__VA_ARGS__)
+#define svabs_f32_z(...) __builtin_sve_svabs_f32_z(__VA_ARGS__)
+#define svabs_f16_z(...) __builtin_sve_svabs_f16_z(__VA_ARGS__)
+#define svabs_s8_m(...) __builtin_sve_svabs_s8_m(__VA_ARGS__)
+#define svabs_s32_m(...) __builtin_sve_svabs_s32_m(__VA_ARGS__)
+#define svabs_s64_m(...) __builtin_sve_svabs_s64_m(__VA_ARGS__)
+#define svabs_s16_m(...) __builtin_sve_svabs_s16_m(__VA_ARGS__)
+#define svabs_s8_x(...) __builtin_sve_svabs_s8_x(__VA_ARGS__)
+#define svabs_s32_x(...) __builtin_sve_svabs_s32_x(__VA_ARGS__)
+#define svabs_s64_x(...) __builtin_sve_svabs_s64_x(__VA_ARGS__)
+#define svabs_s16_x(...) __builtin_sve_svabs_s16_x(__VA_ARGS__)
+#define svabs_s8_z(...) __builtin_sve_svabs_s8_z(__VA_ARGS__)
+#define svabs_s32_z(...) __builtin_sve_svabs_s32_z(__VA_ARGS__)
+#define svabs_s64_z(...) __builtin_sve_svabs_s64_z(__VA_ARGS__)
+#define svabs_s16_z(...) __builtin_sve_svabs_s16_z(__VA_ARGS__)
+#define svacge_n_f64(...) __builtin_sve_svacge_n_f64(__VA_ARGS__)
+#define svacge_n_f32(...) __builtin_sve_svacge_n_f32(__VA_ARGS__)
+#define svacge_n_f16(...) __builtin_sve_svacge_n_f16(__VA_ARGS__)
+#define svacge_f64(...) __builtin_sve_svacge_f64(__VA_ARGS__)
+#define svacge_f32(...) __builtin_sve_svacge_f32(__VA_ARGS__)
+#define svacge_f16(...) __builtin_sve_svacge_f16(__VA_ARGS__)
+#define svacgt_n_f64(...) __builtin_sve_svacgt_n_f64(__VA_ARGS__)
+#define svacgt_n_f32(...) __builtin_sve_svacgt_n_f32(__VA_ARGS__)
+#define svacgt_n_f16(...) __builtin_sve_svacgt_n_f16(__VA_ARGS__)
+#define svacgt_f64(...) __builtin_sve_svacgt_f64(__VA_ARGS__)
+#define svacgt_f32(...) __builtin_sve_svacgt_f32(__VA_ARGS__)
+#define svacgt_f16(...) __builtin_sve_svacgt_f16(__VA_ARGS__)
+#define svacle_n_f64(...) __builtin_sve_svacle_n_f64(__VA_ARGS__)
+#define svacle_n_f32(...) __builtin_sve_svacle_n_f32(__VA_ARGS__)
+#define svacle_n_f16(...) __builtin_sve_svacle_n_f16(__VA_ARGS__)
+#define svacle_f64(...) __builtin_sve_svacle_f64(__VA_ARGS__)
+#define svacle_f32(...) __builtin_sve_svacle_f32(__VA_ARGS__)
+#define svacle_f16(...) __builtin_sve_svacle_f16(__VA_ARGS__)
+#define svaclt_n_f64(...) __builtin_sve_svaclt_n_f64(__VA_ARGS__)
+#define svaclt_n_f32(...) __builtin_sve_svaclt_n_f32(__VA_ARGS__)
+#define svaclt_n_f16(...) __builtin_sve_svaclt_n_f16(__VA_ARGS__)
+#define svaclt_f64(...) __builtin_sve_svaclt_f64(__VA_ARGS__)
+#define svaclt_f32(...) __builtin_sve_svaclt_f32(__VA_ARGS__)
+#define svaclt_f16(...) __builtin_sve_svaclt_f16(__VA_ARGS__)
+#define svadd_n_f64_m(...) __builtin_sve_svadd_n_f64_m(__VA_ARGS__)
+#define svadd_n_f32_m(...) __builtin_sve_svadd_n_f32_m(__VA_ARGS__)
+#define svadd_n_f16_m(...) __builtin_sve_svadd_n_f16_m(__VA_ARGS__)
+#define svadd_n_f64_x(...) __builtin_sve_svadd_n_f64_x(__VA_ARGS__)
+#define svadd_n_f32_x(...) __builtin_sve_svadd_n_f32_x(__VA_ARGS__)
+#define svadd_n_f16_x(...) __builtin_sve_svadd_n_f16_x(__VA_ARGS__)
+#define svadd_n_f64_z(...) __builtin_sve_svadd_n_f64_z(__VA_ARGS__)
+#define svadd_n_f32_z(...) __builtin_sve_svadd_n_f32_z(__VA_ARGS__)
+#define svadd_n_f16_z(...) __builtin_sve_svadd_n_f16_z(__VA_ARGS__)
+#define svadd_n_u8_m(...) __builtin_sve_svadd_n_u8_m(__VA_ARGS__)
+#define svadd_n_u32_m(...) __builtin_sve_svadd_n_u32_m(__VA_ARGS__)
+#define svadd_n_u64_m(...) __builtin_sve_svadd_n_u64_m(__VA_ARGS__)
+#define svadd_n_u16_m(...) __builtin_sve_svadd_n_u16_m(__VA_ARGS__)
+#define svadd_n_s8_m(...) __builtin_sve_svadd_n_s8_m(__VA_ARGS__)
+#define svadd_n_s32_m(...) __builtin_sve_svadd_n_s32_m(__VA_ARGS__)
+#define svadd_n_s64_m(...) __builtin_sve_svadd_n_s64_m(__VA_ARGS__)
+#define svadd_n_s16_m(...) __builtin_sve_svadd_n_s16_m(__VA_ARGS__)
+#define svadd_n_u8_x(...) __builtin_sve_svadd_n_u8_x(__VA_ARGS__)
+#define svadd_n_u32_x(...) __builtin_sve_svadd_n_u32_x(__VA_ARGS__)
+#define svadd_n_u64_x(...) __builtin_sve_svadd_n_u64_x(__VA_ARGS__)
+#define svadd_n_u16_x(...) __builtin_sve_svadd_n_u16_x(__VA_ARGS__)
+#define svadd_n_s8_x(...) __builtin_sve_svadd_n_s8_x(__VA_ARGS__)
+#define svadd_n_s32_x(...) __builtin_sve_svadd_n_s32_x(__VA_ARGS__)
+#define svadd_n_s64_x(...) __builtin_sve_svadd_n_s64_x(__VA_ARGS__)
+#define svadd_n_s16_x(...) __builtin_sve_svadd_n_s16_x(__VA_ARGS__)
+#define svadd_n_u8_z(...) __builtin_sve_svadd_n_u8_z(__VA_ARGS__)
+#define svadd_n_u32_z(...) __builtin_sve_svadd_n_u32_z(__VA_ARGS__)
+#define svadd_n_u64_z(...) __builtin_sve_svadd_n_u64_z(__VA_ARGS__)
+#define svadd_n_u16_z(...) __builtin_sve_svadd_n_u16_z(__VA_ARGS__)
+#define svadd_n_s8_z(...) __builtin_sve_svadd_n_s8_z(__VA_ARGS__)
+#define svadd_n_s32_z(...) __builtin_sve_svadd_n_s32_z(__VA_ARGS__)
+#define svadd_n_s64_z(...) __builtin_sve_svadd_n_s64_z(__VA_ARGS__)
+#define svadd_n_s16_z(...) __builtin_sve_svadd_n_s16_z(__VA_ARGS__)
+#define svadd_f64_m(...) __builtin_sve_svadd_f64_m(__VA_ARGS__)
+#define svadd_f32_m(...) __builtin_sve_svadd_f32_m(__VA_ARGS__)
+#define svadd_f16_m(...) __builtin_sve_svadd_f16_m(__VA_ARGS__)
+#define svadd_f64_x(...) __builtin_sve_svadd_f64_x(__VA_ARGS__)
+#define svadd_f32_x(...) __builtin_sve_svadd_f32_x(__VA_ARGS__)
+#define svadd_f16_x(...) __builtin_sve_svadd_f16_x(__VA_ARGS__)
+#define svadd_f64_z(...) __builtin_sve_svadd_f64_z(__VA_ARGS__)
+#define svadd_f32_z(...) __builtin_sve_svadd_f32_z(__VA_ARGS__)
+#define svadd_f16_z(...) __builtin_sve_svadd_f16_z(__VA_ARGS__)
+#define svadd_u8_m(...) __builtin_sve_svadd_u8_m(__VA_ARGS__)
+#define svadd_u32_m(...) __builtin_sve_svadd_u32_m(__VA_ARGS__)
+#define svadd_u64_m(...) __builtin_sve_svadd_u64_m(__VA_ARGS__)
+#define svadd_u16_m(...) __builtin_sve_svadd_u16_m(__VA_ARGS__)
+#define svadd_s8_m(...) __builtin_sve_svadd_s8_m(__VA_ARGS__)
+#define svadd_s32_m(...) __builtin_sve_svadd_s32_m(__VA_ARGS__)
+#define svadd_s64_m(...) __builtin_sve_svadd_s64_m(__VA_ARGS__)
+#define svadd_s16_m(...) __builtin_sve_svadd_s16_m(__VA_ARGS__)
+#define svadd_u8_x(...) __builtin_sve_svadd_u8_x(__VA_ARGS__)
+#define svadd_u32_x(...) __builtin_sve_svadd_u32_x(__VA_ARGS__)
+#define svadd_u64_x(...) __builtin_sve_svadd_u64_x(__VA_ARGS__)
+#define svadd_u16_x(...) __builtin_sve_svadd_u16_x(__VA_ARGS__)
+#define svadd_s8_x(...) __builtin_sve_svadd_s8_x(__VA_ARGS__)
+#define svadd_s32_x(...) __builtin_sve_svadd_s32_x(__VA_ARGS__)
+#define svadd_s64_x(...) __builtin_sve_svadd_s64_x(__VA_ARGS__)
+#define svadd_s16_x(...) __builtin_sve_svadd_s16_x(__VA_ARGS__)
+#define svadd_u8_z(...) __builtin_sve_svadd_u8_z(__VA_ARGS__)
+#define svadd_u32_z(...) __builtin_sve_svadd_u32_z(__VA_ARGS__)
+#define svadd_u64_z(...) __builtin_sve_svadd_u64_z(__VA_ARGS__)
+#define svadd_u16_z(...) __builtin_sve_svadd_u16_z(__VA_ARGS__)
+#define svadd_s8_z(...) __builtin_sve_svadd_s8_z(__VA_ARGS__)
+#define svadd_s32_z(...) __builtin_sve_svadd_s32_z(__VA_ARGS__)
+#define svadd_s64_z(...) __builtin_sve_svadd_s64_z(__VA_ARGS__)
+#define svadd_s16_z(...) __builtin_sve_svadd_s16_z(__VA_ARGS__)
+#define svadda_f64(...) __builtin_sve_svadda_f64(__VA_ARGS__)
+#define svadda_f32(...) __builtin_sve_svadda_f32(__VA_ARGS__)
+#define svadda_f16(...) __builtin_sve_svadda_f16(__VA_ARGS__)
+#define svaddv_s8(...) __builtin_sve_svaddv_s8(__VA_ARGS__)
+#define svaddv_s32(...) __builtin_sve_svaddv_s32(__VA_ARGS__)
+#define svaddv_s64(...) __builtin_sve_svaddv_s64(__VA_ARGS__)
+#define svaddv_s16(...) __builtin_sve_svaddv_s16(__VA_ARGS__)
+#define svaddv_u8(...) __builtin_sve_svaddv_u8(__VA_ARGS__)
+#define svaddv_u32(...) __builtin_sve_svaddv_u32(__VA_ARGS__)
+#define svaddv_u64(...) __builtin_sve_svaddv_u64(__VA_ARGS__)
+#define svaddv_u16(...) __builtin_sve_svaddv_u16(__VA_ARGS__)
+#define svaddv_f64(...) __builtin_sve_svaddv_f64(__VA_ARGS__)
+#define svaddv_f32(...) __builtin_sve_svaddv_f32(__VA_ARGS__)
+#define svaddv_f16(...) __builtin_sve_svaddv_f16(__VA_ARGS__)
+#define svadrb_u32base_u32offset(...) __builtin_sve_svadrb_u32base_u32offset(__VA_ARGS__)
+#define svadrb_u64base_u64offset(...) __builtin_sve_svadrb_u64base_u64offset(__VA_ARGS__)
+#define svadrb_u32base_s32offset(...) __builtin_sve_svadrb_u32base_s32offset(__VA_ARGS__)
+#define svadrb_u64base_s64offset(...) __builtin_sve_svadrb_u64base_s64offset(__VA_ARGS__)
+#define svadrd_u32base_u32index(...) __builtin_sve_svadrd_u32base_u32index(__VA_ARGS__)
+#define svadrd_u64base_u64index(...) __builtin_sve_svadrd_u64base_u64index(__VA_ARGS__)
+#define svadrd_u32base_s32index(...) __builtin_sve_svadrd_u32base_s32index(__VA_ARGS__)
+#define svadrd_u64base_s64index(...) __builtin_sve_svadrd_u64base_s64index(__VA_ARGS__)
+#define svadrh_u32base_u32index(...) __builtin_sve_svadrh_u32base_u32index(__VA_ARGS__)
+#define svadrh_u64base_u64index(...) __builtin_sve_svadrh_u64base_u64index(__VA_ARGS__)
+#define svadrh_u32base_s32index(...) __builtin_sve_svadrh_u32base_s32index(__VA_ARGS__)
+#define svadrh_u64base_s64index(...) __builtin_sve_svadrh_u64base_s64index(__VA_ARGS__)
+#define svadrw_u32base_u32index(...) __builtin_sve_svadrw_u32base_u32index(__VA_ARGS__)
+#define svadrw_u64base_u64index(...) __builtin_sve_svadrw_u64base_u64index(__VA_ARGS__)
+#define svadrw_u32base_s32index(...) __builtin_sve_svadrw_u32base_s32index(__VA_ARGS__)
+#define svadrw_u64base_s64index(...) __builtin_sve_svadrw_u64base_s64index(__VA_ARGS__)
+#define svand_b_z(...) __builtin_sve_svand_b_z(__VA_ARGS__)
+#define svand_n_u8_m(...) __builtin_sve_svand_n_u8_m(__VA_ARGS__)
+#define svand_n_u32_m(...) __builtin_sve_svand_n_u32_m(__VA_ARGS__)
+#define svand_n_u64_m(...) __builtin_sve_svand_n_u64_m(__VA_ARGS__)
+#define svand_n_u16_m(...) __builtin_sve_svand_n_u16_m(__VA_ARGS__)
+#define svand_n_s8_m(...) __builtin_sve_svand_n_s8_m(__VA_ARGS__)
+#define svand_n_s32_m(...) __builtin_sve_svand_n_s32_m(__VA_ARGS__)
+#define svand_n_s64_m(...) __builtin_sve_svand_n_s64_m(__VA_ARGS__)
+#define svand_n_s16_m(...) __builtin_sve_svand_n_s16_m(__VA_ARGS__)
+#define svand_n_u8_x(...) __builtin_sve_svand_n_u8_x(__VA_ARGS__)
+#define svand_n_u32_x(...) __builtin_sve_svand_n_u32_x(__VA_ARGS__)
+#define svand_n_u64_x(...) __builtin_sve_svand_n_u64_x(__VA_ARGS__)
+#define svand_n_u16_x(...) __builtin_sve_svand_n_u16_x(__VA_ARGS__)
+#define svand_n_s8_x(...) __builtin_sve_svand_n_s8_x(__VA_ARGS__)
+#define svand_n_s32_x(...) __builtin_sve_svand_n_s32_x(__VA_ARGS__)
+#define svand_n_s64_x(...) __builtin_sve_svand_n_s64_x(__VA_ARGS__)
+#define svand_n_s16_x(...) __builtin_sve_svand_n_s16_x(__VA_ARGS__)
+#define svand_n_u8_z(...) __builtin_sve_svand_n_u8_z(__VA_ARGS__)
+#define svand_n_u32_z(...) __builtin_sve_svand_n_u32_z(__VA_ARGS__)
+#define svand_n_u64_z(...) __builtin_sve_svand_n_u64_z(__VA_ARGS__)
+#define svand_n_u16_z(...) __builtin_sve_svand_n_u16_z(__VA_ARGS__)
+#define svand_n_s8_z(...) __builtin_sve_svand_n_s8_z(__VA_ARGS__)
+#define svand_n_s32_z(...) __builtin_sve_svand_n_s32_z(__VA_ARGS__)
+#define svand_n_s64_z(...) __builtin_sve_svand_n_s64_z(__VA_ARGS__)
+#define svand_n_s16_z(...) __builtin_sve_svand_n_s16_z(__VA_ARGS__)
+#define svand_u8_m(...) __builtin_sve_svand_u8_m(__VA_ARGS__)
+#define svand_u32_m(...) __builtin_sve_svand_u32_m(__VA_ARGS__)
+#define svand_u64_m(...) __builtin_sve_svand_u64_m(__VA_ARGS__)
+#define svand_u16_m(...) __builtin_sve_svand_u16_m(__VA_ARGS__)
+#define svand_s8_m(...) __builtin_sve_svand_s8_m(__VA_ARGS__)
+#define svand_s32_m(...) __builtin_sve_svand_s32_m(__VA_ARGS__)
+#define svand_s64_m(...) __builtin_sve_svand_s64_m(__VA_ARGS__)
+#define svand_s16_m(...) __builtin_sve_svand_s16_m(__VA_ARGS__)
+#define svand_u8_x(...) __builtin_sve_svand_u8_x(__VA_ARGS__)
+#define svand_u32_x(...) __builtin_sve_svand_u32_x(__VA_ARGS__)
+#define svand_u64_x(...) __builtin_sve_svand_u64_x(__VA_ARGS__)
+#define svand_u16_x(...) __builtin_sve_svand_u16_x(__VA_ARGS__)
+#define svand_s8_x(...) __builtin_sve_svand_s8_x(__VA_ARGS__)
+#define svand_s32_x(...) __builtin_sve_svand_s32_x(__VA_ARGS__)
+#define svand_s64_x(...) __builtin_sve_svand_s64_x(__VA_ARGS__)
+#define svand_s16_x(...) __builtin_sve_svand_s16_x(__VA_ARGS__)
+#define svand_u8_z(...) __builtin_sve_svand_u8_z(__VA_ARGS__)
+#define svand_u32_z(...) __builtin_sve_svand_u32_z(__VA_ARGS__)
+#define svand_u64_z(...) __builtin_sve_svand_u64_z(__VA_ARGS__)
+#define svand_u16_z(...) __builtin_sve_svand_u16_z(__VA_ARGS__)
+#define svand_s8_z(...) __builtin_sve_svand_s8_z(__VA_ARGS__)
+#define svand_s32_z(...) __builtin_sve_svand_s32_z(__VA_ARGS__)
+#define svand_s64_z(...) __builtin_sve_svand_s64_z(__VA_ARGS__)
+#define svand_s16_z(...) __builtin_sve_svand_s16_z(__VA_ARGS__)
+#define svandv_u8(...) __builtin_sve_svandv_u8(__VA_ARGS__)
+#define svandv_u32(...) __builtin_sve_svandv_u32(__VA_ARGS__)
+#define svandv_u64(...) __builtin_sve_svandv_u64(__VA_ARGS__)
+#define svandv_u16(...) __builtin_sve_svandv_u16(__VA_ARGS__)
+#define svandv_s8(...) __builtin_sve_svandv_s8(__VA_ARGS__)
+#define svandv_s32(...) __builtin_sve_svandv_s32(__VA_ARGS__)
+#define svandv_s64(...) __builtin_sve_svandv_s64(__VA_ARGS__)
+#define svandv_s16(...) __builtin_sve_svandv_s16(__VA_ARGS__)
+#define svasr_n_s8_m(...) __builtin_sve_svasr_n_s8_m(__VA_ARGS__)
+#define svasr_n_s32_m(...) __builtin_sve_svasr_n_s32_m(__VA_ARGS__)
+#define svasr_n_s64_m(...) __builtin_sve_svasr_n_s64_m(__VA_ARGS__)
+#define svasr_n_s16_m(...) __builtin_sve_svasr_n_s16_m(__VA_ARGS__)
+#define svasr_n_s8_x(...) __builtin_sve_svasr_n_s8_x(__VA_ARGS__)
+#define svasr_n_s32_x(...) __builtin_sve_svasr_n_s32_x(__VA_ARGS__)
+#define svasr_n_s64_x(...) __builtin_sve_svasr_n_s64_x(__VA_ARGS__)
+#define svasr_n_s16_x(...) __builtin_sve_svasr_n_s16_x(__VA_ARGS__)
+#define svasr_n_s8_z(...) __builtin_sve_svasr_n_s8_z(__VA_ARGS__)
+#define svasr_n_s32_z(...) __builtin_sve_svasr_n_s32_z(__VA_ARGS__)
+#define svasr_n_s64_z(...) __builtin_sve_svasr_n_s64_z(__VA_ARGS__)
+#define svasr_n_s16_z(...) __builtin_sve_svasr_n_s16_z(__VA_ARGS__)
+#define svasr_s8_m(...) __builtin_sve_svasr_s8_m(__VA_ARGS__)
+#define svasr_s32_m(...) __builtin_sve_svasr_s32_m(__VA_ARGS__)
+#define svasr_s64_m(...) __builtin_sve_svasr_s64_m(__VA_ARGS__)
+#define svasr_s16_m(...) __builtin_sve_svasr_s16_m(__VA_ARGS__)
+#define svasr_s8_x(...) __builtin_sve_svasr_s8_x(__VA_ARGS__)
+#define svasr_s32_x(...) __builtin_sve_svasr_s32_x(__VA_ARGS__)
+#define svasr_s64_x(...) __builtin_sve_svasr_s64_x(__VA_ARGS__)
+#define svasr_s16_x(...) __builtin_sve_svasr_s16_x(__VA_ARGS__)
+#define svasr_s8_z(...) __builtin_sve_svasr_s8_z(__VA_ARGS__)
+#define svasr_s32_z(...) __builtin_sve_svasr_s32_z(__VA_ARGS__)
+#define svasr_s64_z(...) __builtin_sve_svasr_s64_z(__VA_ARGS__)
+#define svasr_s16_z(...) __builtin_sve_svasr_s16_z(__VA_ARGS__)
+#define svasr_wide_n_s8_m(...) __builtin_sve_svasr_wide_n_s8_m(__VA_ARGS__)
+#define svasr_wide_n_s32_m(...) __builtin_sve_svasr_wide_n_s32_m(__VA_ARGS__)
+#define svasr_wide_n_s16_m(...) __builtin_sve_svasr_wide_n_s16_m(__VA_ARGS__)
+#define svasr_wide_n_s8_x(...) __builtin_sve_svasr_wide_n_s8_x(__VA_ARGS__)
+#define svasr_wide_n_s32_x(...) __builtin_sve_svasr_wide_n_s32_x(__VA_ARGS__)
+#define svasr_wide_n_s16_x(...) __builtin_sve_svasr_wide_n_s16_x(__VA_ARGS__)
+#define svasr_wide_n_s8_z(...) __builtin_sve_svasr_wide_n_s8_z(__VA_ARGS__)
+#define svasr_wide_n_s32_z(...) __builtin_sve_svasr_wide_n_s32_z(__VA_ARGS__)
+#define svasr_wide_n_s16_z(...) __builtin_sve_svasr_wide_n_s16_z(__VA_ARGS__)
+#define svasr_wide_s8_m(...) __builtin_sve_svasr_wide_s8_m(__VA_ARGS__)
+#define svasr_wide_s32_m(...) __builtin_sve_svasr_wide_s32_m(__VA_ARGS__)
+#define svasr_wide_s16_m(...) __builtin_sve_svasr_wide_s16_m(__VA_ARGS__)
+#define svasr_wide_s8_x(...) __builtin_sve_svasr_wide_s8_x(__VA_ARGS__)
+#define svasr_wide_s32_x(...) __builtin_sve_svasr_wide_s32_x(__VA_ARGS__)
+#define svasr_wide_s16_x(...) __builtin_sve_svasr_wide_s16_x(__VA_ARGS__)
+#define svasr_wide_s8_z(...) __builtin_sve_svasr_wide_s8_z(__VA_ARGS__)
+#define svasr_wide_s32_z(...) __builtin_sve_svasr_wide_s32_z(__VA_ARGS__)
+#define svasr_wide_s16_z(...) __builtin_sve_svasr_wide_s16_z(__VA_ARGS__)
+#define svasrd_n_s8_m(...) __builtin_sve_svasrd_n_s8_m(__VA_ARGS__)
+#define svasrd_n_s32_m(...) __builtin_sve_svasrd_n_s32_m(__VA_ARGS__)
+#define svasrd_n_s64_m(...) __builtin_sve_svasrd_n_s64_m(__VA_ARGS__)
+#define svasrd_n_s16_m(...) __builtin_sve_svasrd_n_s16_m(__VA_ARGS__)
+#define svasrd_n_s8_x(...) __builtin_sve_svasrd_n_s8_x(__VA_ARGS__)
+#define svasrd_n_s32_x(...) __builtin_sve_svasrd_n_s32_x(__VA_ARGS__)
+#define svasrd_n_s64_x(...) __builtin_sve_svasrd_n_s64_x(__VA_ARGS__)
+#define svasrd_n_s16_x(...) __builtin_sve_svasrd_n_s16_x(__VA_ARGS__)
+#define svasrd_n_s8_z(...) __builtin_sve_svasrd_n_s8_z(__VA_ARGS__)
+#define svasrd_n_s32_z(...) __builtin_sve_svasrd_n_s32_z(__VA_ARGS__)
+#define svasrd_n_s64_z(...) __builtin_sve_svasrd_n_s64_z(__VA_ARGS__)
+#define svasrd_n_s16_z(...) __builtin_sve_svasrd_n_s16_z(__VA_ARGS__)
+#define svbic_b_z(...) __builtin_sve_svbic_b_z(__VA_ARGS__)
+#define svbic_n_u8_m(...) __builtin_sve_svbic_n_u8_m(__VA_ARGS__)
+#define svbic_n_u32_m(...) __builtin_sve_svbic_n_u32_m(__VA_ARGS__)
+#define svbic_n_u64_m(...) __builtin_sve_svbic_n_u64_m(__VA_ARGS__)
+#define svbic_n_u16_m(...) __builtin_sve_svbic_n_u16_m(__VA_ARGS__)
+#define svbic_n_s8_m(...) __builtin_sve_svbic_n_s8_m(__VA_ARGS__)
+#define svbic_n_s32_m(...) __builtin_sve_svbic_n_s32_m(__VA_ARGS__)
+#define svbic_n_s64_m(...) __builtin_sve_svbic_n_s64_m(__VA_ARGS__)
+#define svbic_n_s16_m(...) __builtin_sve_svbic_n_s16_m(__VA_ARGS__)
+#define svbic_n_u8_x(...) __builtin_sve_svbic_n_u8_x(__VA_ARGS__)
+#define svbic_n_u32_x(...) __builtin_sve_svbic_n_u32_x(__VA_ARGS__)
+#define svbic_n_u64_x(...) __builtin_sve_svbic_n_u64_x(__VA_ARGS__)
+#define svbic_n_u16_x(...) __builtin_sve_svbic_n_u16_x(__VA_ARGS__)
+#define svbic_n_s8_x(...) __builtin_sve_svbic_n_s8_x(__VA_ARGS__)
+#define svbic_n_s32_x(...) __builtin_sve_svbic_n_s32_x(__VA_ARGS__)
+#define svbic_n_s64_x(...) __builtin_sve_svbic_n_s64_x(__VA_ARGS__)
+#define svbic_n_s16_x(...) __builtin_sve_svbic_n_s16_x(__VA_ARGS__)
+#define svbic_n_u8_z(...) __builtin_sve_svbic_n_u8_z(__VA_ARGS__)
+#define svbic_n_u32_z(...) __builtin_sve_svbic_n_u32_z(__VA_ARGS__)
+#define svbic_n_u64_z(...) __builtin_sve_svbic_n_u64_z(__VA_ARGS__)
+#define svbic_n_u16_z(...) __builtin_sve_svbic_n_u16_z(__VA_ARGS__)
+#define svbic_n_s8_z(...) __builtin_sve_svbic_n_s8_z(__VA_ARGS__)
+#define svbic_n_s32_z(...) __builtin_sve_svbic_n_s32_z(__VA_ARGS__)
+#define svbic_n_s64_z(...) __builtin_sve_svbic_n_s64_z(__VA_ARGS__)
+#define svbic_n_s16_z(...) __builtin_sve_svbic_n_s16_z(__VA_ARGS__)
+#define svbic_u8_m(...) __builtin_sve_svbic_u8_m(__VA_ARGS__)
+#define svbic_u32_m(...) __builtin_sve_svbic_u32_m(__VA_ARGS__)
+#define svbic_u64_m(...) __builtin_sve_svbic_u64_m(__VA_ARGS__)
+#define svbic_u16_m(...) __builtin_sve_svbic_u16_m(__VA_ARGS__)
+#define svbic_s8_m(...) __builtin_sve_svbic_s8_m(__VA_ARGS__)
+#define svbic_s32_m(...) __builtin_sve_svbic_s32_m(__VA_ARGS__)
+#define svbic_s64_m(...) __builtin_sve_svbic_s64_m(__VA_ARGS__)
+#define svbic_s16_m(...) __builtin_sve_svbic_s16_m(__VA_ARGS__)
+#define svbic_u8_x(...) __builtin_sve_svbic_u8_x(__VA_ARGS__)
+#define svbic_u32_x(...) __builtin_sve_svbic_u32_x(__VA_ARGS__)
+#define svbic_u64_x(...) __builtin_sve_svbic_u64_x(__VA_ARGS__)
+#define svbic_u16_x(...) __builtin_sve_svbic_u16_x(__VA_ARGS__)
+#define svbic_s8_x(...) __builtin_sve_svbic_s8_x(__VA_ARGS__)
+#define svbic_s32_x(...) __builtin_sve_svbic_s32_x(__VA_ARGS__)
+#define svbic_s64_x(...) __builtin_sve_svbic_s64_x(__VA_ARGS__)
+#define svbic_s16_x(...) __builtin_sve_svbic_s16_x(__VA_ARGS__)
+#define svbic_u8_z(...) __builtin_sve_svbic_u8_z(__VA_ARGS__)
+#define svbic_u32_z(...) __builtin_sve_svbic_u32_z(__VA_ARGS__)
+#define svbic_u64_z(...) __builtin_sve_svbic_u64_z(__VA_ARGS__)
+#define svbic_u16_z(...) __builtin_sve_svbic_u16_z(__VA_ARGS__)
+#define svbic_s8_z(...) __builtin_sve_svbic_s8_z(__VA_ARGS__)
+#define svbic_s32_z(...) __builtin_sve_svbic_s32_z(__VA_ARGS__)
+#define svbic_s64_z(...) __builtin_sve_svbic_s64_z(__VA_ARGS__)
+#define svbic_s16_z(...) __builtin_sve_svbic_s16_z(__VA_ARGS__)
+#define svbrka_b_m(...) __builtin_sve_svbrka_b_m(__VA_ARGS__)
+#define svbrka_b_z(...) __builtin_sve_svbrka_b_z(__VA_ARGS__)
+#define svbrkb_b_m(...) __builtin_sve_svbrkb_b_m(__VA_ARGS__)
+#define svbrkb_b_z(...) __builtin_sve_svbrkb_b_z(__VA_ARGS__)
+#define svbrkn_b_z(...) __builtin_sve_svbrkn_b_z(__VA_ARGS__)
+#define svbrkpa_b_z(...) __builtin_sve_svbrkpa_b_z(__VA_ARGS__)
+#define svbrkpb_b_z(...) __builtin_sve_svbrkpb_b_z(__VA_ARGS__)
+#define svcadd_f64_m(...) __builtin_sve_svcadd_f64_m(__VA_ARGS__)
+#define svcadd_f32_m(...) __builtin_sve_svcadd_f32_m(__VA_ARGS__)
+#define svcadd_f16_m(...) __builtin_sve_svcadd_f16_m(__VA_ARGS__)
+#define svcadd_f64_x(...) __builtin_sve_svcadd_f64_x(__VA_ARGS__)
+#define svcadd_f32_x(...) __builtin_sve_svcadd_f32_x(__VA_ARGS__)
+#define svcadd_f16_x(...) __builtin_sve_svcadd_f16_x(__VA_ARGS__)
+#define svcadd_f64_z(...) __builtin_sve_svcadd_f64_z(__VA_ARGS__)
+#define svcadd_f32_z(...) __builtin_sve_svcadd_f32_z(__VA_ARGS__)
+#define svcadd_f16_z(...) __builtin_sve_svcadd_f16_z(__VA_ARGS__)
+#define svclasta_n_u8(...) __builtin_sve_svclasta_n_u8(__VA_ARGS__)
+#define svclasta_n_u32(...) __builtin_sve_svclasta_n_u32(__VA_ARGS__)
+#define svclasta_n_u64(...) __builtin_sve_svclasta_n_u64(__VA_ARGS__)
+#define svclasta_n_u16(...) __builtin_sve_svclasta_n_u16(__VA_ARGS__)
+#define svclasta_n_s8(...) __builtin_sve_svclasta_n_s8(__VA_ARGS__)
+#define svclasta_n_f64(...) __builtin_sve_svclasta_n_f64(__VA_ARGS__)
+#define svclasta_n_f32(...) __builtin_sve_svclasta_n_f32(__VA_ARGS__)
+#define svclasta_n_f16(...) __builtin_sve_svclasta_n_f16(__VA_ARGS__)
+#define svclasta_n_s32(...) __builtin_sve_svclasta_n_s32(__VA_ARGS__)
+#define svclasta_n_s64(...) __builtin_sve_svclasta_n_s64(__VA_ARGS__)
+#define svclasta_n_s16(...) __builtin_sve_svclasta_n_s16(__VA_ARGS__)
+#define svclasta_u8(...) __builtin_sve_svclasta_u8(__VA_ARGS__)
+#define svclasta_u32(...) __builtin_sve_svclasta_u32(__VA_ARGS__)
+#define svclasta_u64(...) __builtin_sve_svclasta_u64(__VA_ARGS__)
+#define svclasta_u16(...) __builtin_sve_svclasta_u16(__VA_ARGS__)
+#define svclasta_s8(...) __builtin_sve_svclasta_s8(__VA_ARGS__)
+#define svclasta_f64(...) __builtin_sve_svclasta_f64(__VA_ARGS__)
+#define svclasta_f32(...) __builtin_sve_svclasta_f32(__VA_ARGS__)
+#define svclasta_f16(...) __builtin_sve_svclasta_f16(__VA_ARGS__)
+#define svclasta_s32(...) __builtin_sve_svclasta_s32(__VA_ARGS__)
+#define svclasta_s64(...) __builtin_sve_svclasta_s64(__VA_ARGS__)
+#define svclasta_s16(...) __builtin_sve_svclasta_s16(__VA_ARGS__)
+#define svclastb_n_u8(...) __builtin_sve_svclastb_n_u8(__VA_ARGS__)
+#define svclastb_n_u32(...) __builtin_sve_svclastb_n_u32(__VA_ARGS__)
+#define svclastb_n_u64(...) __builtin_sve_svclastb_n_u64(__VA_ARGS__)
+#define svclastb_n_u16(...) __builtin_sve_svclastb_n_u16(__VA_ARGS__)
+#define svclastb_n_s8(...) __builtin_sve_svclastb_n_s8(__VA_ARGS__)
+#define svclastb_n_f64(...) __builtin_sve_svclastb_n_f64(__VA_ARGS__)
+#define svclastb_n_f32(...) __builtin_sve_svclastb_n_f32(__VA_ARGS__)
+#define svclastb_n_f16(...) __builtin_sve_svclastb_n_f16(__VA_ARGS__)
+#define svclastb_n_s32(...) __builtin_sve_svclastb_n_s32(__VA_ARGS__)
+#define svclastb_n_s64(...) __builtin_sve_svclastb_n_s64(__VA_ARGS__)
+#define svclastb_n_s16(...) __builtin_sve_svclastb_n_s16(__VA_ARGS__)
+#define svclastb_u8(...) __builtin_sve_svclastb_u8(__VA_ARGS__)
+#define svclastb_u32(...) __builtin_sve_svclastb_u32(__VA_ARGS__)
+#define svclastb_u64(...) __builtin_sve_svclastb_u64(__VA_ARGS__)
+#define svclastb_u16(...) __builtin_sve_svclastb_u16(__VA_ARGS__)
+#define svclastb_s8(...) __builtin_sve_svclastb_s8(__VA_ARGS__)
+#define svclastb_f64(...) __builtin_sve_svclastb_f64(__VA_ARGS__)
+#define svclastb_f32(...) __builtin_sve_svclastb_f32(__VA_ARGS__)
+#define svclastb_f16(...) __builtin_sve_svclastb_f16(__VA_ARGS__)
+#define svclastb_s32(...) __builtin_sve_svclastb_s32(__VA_ARGS__)
+#define svclastb_s64(...) __builtin_sve_svclastb_s64(__VA_ARGS__)
+#define svclastb_s16(...) __builtin_sve_svclastb_s16(__VA_ARGS__)
+#define svcls_s8_m(...) __builtin_sve_svcls_s8_m(__VA_ARGS__)
+#define svcls_s32_m(...) __builtin_sve_svcls_s32_m(__VA_ARGS__)
+#define svcls_s64_m(...) __builtin_sve_svcls_s64_m(__VA_ARGS__)
+#define svcls_s16_m(...) __builtin_sve_svcls_s16_m(__VA_ARGS__)
+#define svcls_s8_x(...) __builtin_sve_svcls_s8_x(__VA_ARGS__)
+#define svcls_s32_x(...) __builtin_sve_svcls_s32_x(__VA_ARGS__)
+#define svcls_s64_x(...) __builtin_sve_svcls_s64_x(__VA_ARGS__)
+#define svcls_s16_x(...) __builtin_sve_svcls_s16_x(__VA_ARGS__)
+#define svcls_s8_z(...) __builtin_sve_svcls_s8_z(__VA_ARGS__)
+#define svcls_s32_z(...) __builtin_sve_svcls_s32_z(__VA_ARGS__)
+#define svcls_s64_z(...) __builtin_sve_svcls_s64_z(__VA_ARGS__)
+#define svcls_s16_z(...) __builtin_sve_svcls_s16_z(__VA_ARGS__)
+#define svclz_u8_m(...) __builtin_sve_svclz_u8_m(__VA_ARGS__)
+#define svclz_u32_m(...) __builtin_sve_svclz_u32_m(__VA_ARGS__)
+#define svclz_u64_m(...) __builtin_sve_svclz_u64_m(__VA_ARGS__)
+#define svclz_u16_m(...) __builtin_sve_svclz_u16_m(__VA_ARGS__)
+#define svclz_s8_m(...) __builtin_sve_svclz_s8_m(__VA_ARGS__)
+#define svclz_s32_m(...) __builtin_sve_svclz_s32_m(__VA_ARGS__)
+#define svclz_s64_m(...) __builtin_sve_svclz_s64_m(__VA_ARGS__)
+#define svclz_s16_m(...) __builtin_sve_svclz_s16_m(__VA_ARGS__)
+#define svclz_u8_x(...) __builtin_sve_svclz_u8_x(__VA_ARGS__)
+#define svclz_u32_x(...) __builtin_sve_svclz_u32_x(__VA_ARGS__)
+#define svclz_u64_x(...) __builtin_sve_svclz_u64_x(__VA_ARGS__)
+#define svclz_u16_x(...) __builtin_sve_svclz_u16_x(__VA_ARGS__)
+#define svclz_s8_x(...) __builtin_sve_svclz_s8_x(__VA_ARGS__)
+#define svclz_s32_x(...) __builtin_sve_svclz_s32_x(__VA_ARGS__)
+#define svclz_s64_x(...) __builtin_sve_svclz_s64_x(__VA_ARGS__)
+#define svclz_s16_x(...) __builtin_sve_svclz_s16_x(__VA_ARGS__)
+#define svclz_u8_z(...) __builtin_sve_svclz_u8_z(__VA_ARGS__)
+#define svclz_u32_z(...) __builtin_sve_svclz_u32_z(__VA_ARGS__)
+#define svclz_u64_z(...) __builtin_sve_svclz_u64_z(__VA_ARGS__)
+#define svclz_u16_z(...) __builtin_sve_svclz_u16_z(__VA_ARGS__)
+#define svclz_s8_z(...) __builtin_sve_svclz_s8_z(__VA_ARGS__)
+#define svclz_s32_z(...) __builtin_sve_svclz_s32_z(__VA_ARGS__)
+#define svclz_s64_z(...) __builtin_sve_svclz_s64_z(__VA_ARGS__)
+#define svclz_s16_z(...) __builtin_sve_svclz_s16_z(__VA_ARGS__)
+#define svcmla_f64_m(...) __builtin_sve_svcmla_f64_m(__VA_ARGS__)
+#define svcmla_f32_m(...) __builtin_sve_svcmla_f32_m(__VA_ARGS__)
+#define svcmla_f16_m(...) __builtin_sve_svcmla_f16_m(__VA_ARGS__)
+#define svcmla_f64_x(...) __builtin_sve_svcmla_f64_x(__VA_ARGS__)
+#define svcmla_f32_x(...) __builtin_sve_svcmla_f32_x(__VA_ARGS__)
+#define svcmla_f16_x(...) __builtin_sve_svcmla_f16_x(__VA_ARGS__)
+#define svcmla_f64_z(...) __builtin_sve_svcmla_f64_z(__VA_ARGS__)
+#define svcmla_f32_z(...) __builtin_sve_svcmla_f32_z(__VA_ARGS__)
+#define svcmla_f16_z(...) __builtin_sve_svcmla_f16_z(__VA_ARGS__)
+#define svcmla_lane_f32(...) __builtin_sve_svcmla_lane_f32(__VA_ARGS__)
+#define svcmla_lane_f16(...) __builtin_sve_svcmla_lane_f16(__VA_ARGS__)
+#define svcmpeq_n_f64(...) __builtin_sve_svcmpeq_n_f64(__VA_ARGS__)
+#define svcmpeq_n_f32(...) __builtin_sve_svcmpeq_n_f32(__VA_ARGS__)
+#define svcmpeq_n_f16(...) __builtin_sve_svcmpeq_n_f16(__VA_ARGS__)
+#define svcmpeq_n_u8(...) __builtin_sve_svcmpeq_n_u8(__VA_ARGS__)
+#define svcmpeq_n_u32(...) __builtin_sve_svcmpeq_n_u32(__VA_ARGS__)
+#define svcmpeq_n_u64(...) __builtin_sve_svcmpeq_n_u64(__VA_ARGS__)
+#define svcmpeq_n_u16(...) __builtin_sve_svcmpeq_n_u16(__VA_ARGS__)
+#define svcmpeq_n_s8(...) __builtin_sve_svcmpeq_n_s8(__VA_ARGS__)
+#define svcmpeq_n_s32(...) __builtin_sve_svcmpeq_n_s32(__VA_ARGS__)
+#define svcmpeq_n_s64(...) __builtin_sve_svcmpeq_n_s64(__VA_ARGS__)
+#define svcmpeq_n_s16(...) __builtin_sve_svcmpeq_n_s16(__VA_ARGS__)
+#define svcmpeq_u8(...) __builtin_sve_svcmpeq_u8(__VA_ARGS__)
+#define svcmpeq_u32(...) __builtin_sve_svcmpeq_u32(__VA_ARGS__)
+#define svcmpeq_u64(...) __builtin_sve_svcmpeq_u64(__VA_ARGS__)
+#define svcmpeq_u16(...) __builtin_sve_svcmpeq_u16(__VA_ARGS__)
+#define svcmpeq_s8(...) __builtin_sve_svcmpeq_s8(__VA_ARGS__)
+#define svcmpeq_s32(...) __builtin_sve_svcmpeq_s32(__VA_ARGS__)
+#define svcmpeq_s64(...) __builtin_sve_svcmpeq_s64(__VA_ARGS__)
+#define svcmpeq_s16(...) __builtin_sve_svcmpeq_s16(__VA_ARGS__)
+#define svcmpeq_f64(...) __builtin_sve_svcmpeq_f64(__VA_ARGS__)
+#define svcmpeq_f32(...) __builtin_sve_svcmpeq_f32(__VA_ARGS__)
+#define svcmpeq_f16(...) __builtin_sve_svcmpeq_f16(__VA_ARGS__)
+#define svcmpeq_wide_n_s8(...) __builtin_sve_svcmpeq_wide_n_s8(__VA_ARGS__)
+#define svcmpeq_wide_n_s32(...) __builtin_sve_svcmpeq_wide_n_s32(__VA_ARGS__)
+#define svcmpeq_wide_n_s16(...) __builtin_sve_svcmpeq_wide_n_s16(__VA_ARGS__)
+#define svcmpeq_wide_s8(...) __builtin_sve_svcmpeq_wide_s8(__VA_ARGS__)
+#define svcmpeq_wide_s32(...) __builtin_sve_svcmpeq_wide_s32(__VA_ARGS__)
+#define svcmpeq_wide_s16(...) __builtin_sve_svcmpeq_wide_s16(__VA_ARGS__)
+#define svcmpge_n_f64(...) __builtin_sve_svcmpge_n_f64(__VA_ARGS__)
+#define svcmpge_n_f32(...) __builtin_sve_svcmpge_n_f32(__VA_ARGS__)
+#define svcmpge_n_f16(...) __builtin_sve_svcmpge_n_f16(__VA_ARGS__)
+#define svcmpge_n_s8(...) __builtin_sve_svcmpge_n_s8(__VA_ARGS__)
+#define svcmpge_n_s32(...) __builtin_sve_svcmpge_n_s32(__VA_ARGS__)
+#define svcmpge_n_s64(...) __builtin_sve_svcmpge_n_s64(__VA_ARGS__)
+#define svcmpge_n_s16(...) __builtin_sve_svcmpge_n_s16(__VA_ARGS__)
+#define svcmpge_n_u8(...) __builtin_sve_svcmpge_n_u8(__VA_ARGS__)
+#define svcmpge_n_u32(...) __builtin_sve_svcmpge_n_u32(__VA_ARGS__)
+#define svcmpge_n_u64(...) __builtin_sve_svcmpge_n_u64(__VA_ARGS__)
+#define svcmpge_n_u16(...) __builtin_sve_svcmpge_n_u16(__VA_ARGS__)
+#define svcmpge_s8(...) __builtin_sve_svcmpge_s8(__VA_ARGS__)
+#define svcmpge_s32(...) __builtin_sve_svcmpge_s32(__VA_ARGS__)
+#define svcmpge_s64(...) __builtin_sve_svcmpge_s64(__VA_ARGS__)
+#define svcmpge_s16(...) __builtin_sve_svcmpge_s16(__VA_ARGS__)
+#define svcmpge_f64(...) __builtin_sve_svcmpge_f64(__VA_ARGS__)
+#define svcmpge_f32(...) __builtin_sve_svcmpge_f32(__VA_ARGS__)
+#define svcmpge_f16(...) __builtin_sve_svcmpge_f16(__VA_ARGS__)
+#define svcmpge_u8(...) __builtin_sve_svcmpge_u8(__VA_ARGS__)
+#define svcmpge_u32(...) __builtin_sve_svcmpge_u32(__VA_ARGS__)
+#define svcmpge_u64(...) __builtin_sve_svcmpge_u64(__VA_ARGS__)
+#define svcmpge_u16(...) __builtin_sve_svcmpge_u16(__VA_ARGS__)
+#define svcmpge_wide_n_s8(...) __builtin_sve_svcmpge_wide_n_s8(__VA_ARGS__)
+#define svcmpge_wide_n_s32(...) __builtin_sve_svcmpge_wide_n_s32(__VA_ARGS__)
+#define svcmpge_wide_n_s16(...) __builtin_sve_svcmpge_wide_n_s16(__VA_ARGS__)
+#define svcmpge_wide_n_u8(...) __builtin_sve_svcmpge_wide_n_u8(__VA_ARGS__)
+#define svcmpge_wide_n_u32(...) __builtin_sve_svcmpge_wide_n_u32(__VA_ARGS__)
+#define svcmpge_wide_n_u16(...) __builtin_sve_svcmpge_wide_n_u16(__VA_ARGS__)
+#define svcmpge_wide_s8(...) __builtin_sve_svcmpge_wide_s8(__VA_ARGS__)
+#define svcmpge_wide_s32(...) __builtin_sve_svcmpge_wide_s32(__VA_ARGS__)
+#define svcmpge_wide_s16(...) __builtin_sve_svcmpge_wide_s16(__VA_ARGS__)
+#define svcmpge_wide_u8(...) __builtin_sve_svcmpge_wide_u8(__VA_ARGS__)
+#define svcmpge_wide_u32(...) __builtin_sve_svcmpge_wide_u32(__VA_ARGS__)
+#define svcmpge_wide_u16(...) __builtin_sve_svcmpge_wide_u16(__VA_ARGS__)
+#define svcmpgt_n_f64(...) __builtin_sve_svcmpgt_n_f64(__VA_ARGS__)
+#define svcmpgt_n_f32(...) __builtin_sve_svcmpgt_n_f32(__VA_ARGS__)
+#define svcmpgt_n_f16(...) __builtin_sve_svcmpgt_n_f16(__VA_ARGS__)
+#define svcmpgt_n_s8(...) __builtin_sve_svcmpgt_n_s8(__VA_ARGS__)
+#define svcmpgt_n_s32(...) __builtin_sve_svcmpgt_n_s32(__VA_ARGS__)
+#define svcmpgt_n_s64(...) __builtin_sve_svcmpgt_n_s64(__VA_ARGS__)
+#define svcmpgt_n_s16(...) __builtin_sve_svcmpgt_n_s16(__VA_ARGS__)
+#define svcmpgt_n_u8(...) __builtin_sve_svcmpgt_n_u8(__VA_ARGS__)
+#define svcmpgt_n_u32(...) __builtin_sve_svcmpgt_n_u32(__VA_ARGS__)
+#define svcmpgt_n_u64(...) __builtin_sve_svcmpgt_n_u64(__VA_ARGS__)
+#define svcmpgt_n_u16(...) __builtin_sve_svcmpgt_n_u16(__VA_ARGS__)
+#define svcmpgt_s8(...) __builtin_sve_svcmpgt_s8(__VA_ARGS__)
+#define svcmpgt_s32(...) __builtin_sve_svcmpgt_s32(__VA_ARGS__)
+#define svcmpgt_s64(...) __builtin_sve_svcmpgt_s64(__VA_ARGS__)
+#define svcmpgt_s16(...) __builtin_sve_svcmpgt_s16(__VA_ARGS__)
+#define svcmpgt_f64(...) __builtin_sve_svcmpgt_f64(__VA_ARGS__)
+#define svcmpgt_f32(...) __builtin_sve_svcmpgt_f32(__VA_ARGS__)
+#define svcmpgt_f16(...) __builtin_sve_svcmpgt_f16(__VA_ARGS__)
+#define svcmpgt_u8(...) __builtin_sve_svcmpgt_u8(__VA_ARGS__)
+#define svcmpgt_u32(...) __builtin_sve_svcmpgt_u32(__VA_ARGS__)
+#define svcmpgt_u64(...) __builtin_sve_svcmpgt_u64(__VA_ARGS__)
+#define svcmpgt_u16(...) __builtin_sve_svcmpgt_u16(__VA_ARGS__)
+#define svcmpgt_wide_n_s8(...) __builtin_sve_svcmpgt_wide_n_s8(__VA_ARGS__)
+#define svcmpgt_wide_n_s32(...) __builtin_sve_svcmpgt_wide_n_s32(__VA_ARGS__)
+#define svcmpgt_wide_n_s16(...) __builtin_sve_svcmpgt_wide_n_s16(__VA_ARGS__)
+#define svcmpgt_wide_n_u8(...) __builtin_sve_svcmpgt_wide_n_u8(__VA_ARGS__)
+#define svcmpgt_wide_n_u32(...) __builtin_sve_svcmpgt_wide_n_u32(__VA_ARGS__)
+#define svcmpgt_wide_n_u16(...) __builtin_sve_svcmpgt_wide_n_u16(__VA_ARGS__)
+#define svcmpgt_wide_s8(...) __builtin_sve_svcmpgt_wide_s8(__VA_ARGS__)
+#define svcmpgt_wide_s32(...) __builtin_sve_svcmpgt_wide_s32(__VA_ARGS__)
+#define svcmpgt_wide_s16(...) __builtin_sve_svcmpgt_wide_s16(__VA_ARGS__)
+#define svcmpgt_wide_u8(...) __builtin_sve_svcmpgt_wide_u8(__VA_ARGS__)
+#define svcmpgt_wide_u32(...) __builtin_sve_svcmpgt_wide_u32(__VA_ARGS__)
+#define svcmpgt_wide_u16(...) __builtin_sve_svcmpgt_wide_u16(__VA_ARGS__)
+#define svcmple_n_f64(...) __builtin_sve_svcmple_n_f64(__VA_ARGS__)
+#define svcmple_n_f32(...) __builtin_sve_svcmple_n_f32(__VA_ARGS__)
+#define svcmple_n_f16(...) __builtin_sve_svcmple_n_f16(__VA_ARGS__)
+#define svcmple_n_s8(...) __builtin_sve_svcmple_n_s8(__VA_ARGS__)
+#define svcmple_n_s32(...) __builtin_sve_svcmple_n_s32(__VA_ARGS__)
+#define svcmple_n_s64(...) __builtin_sve_svcmple_n_s64(__VA_ARGS__)
+#define svcmple_n_s16(...) __builtin_sve_svcmple_n_s16(__VA_ARGS__)
+#define svcmple_n_u8(...) __builtin_sve_svcmple_n_u8(__VA_ARGS__)
+#define svcmple_n_u32(...) __builtin_sve_svcmple_n_u32(__VA_ARGS__)
+#define svcmple_n_u64(...) __builtin_sve_svcmple_n_u64(__VA_ARGS__)
+#define svcmple_n_u16(...) __builtin_sve_svcmple_n_u16(__VA_ARGS__)
+#define svcmple_s8(...) __builtin_sve_svcmple_s8(__VA_ARGS__)
+#define svcmple_s32(...) __builtin_sve_svcmple_s32(__VA_ARGS__)
+#define svcmple_s64(...) __builtin_sve_svcmple_s64(__VA_ARGS__)
+#define svcmple_s16(...) __builtin_sve_svcmple_s16(__VA_ARGS__)
+#define svcmple_f64(...) __builtin_sve_svcmple_f64(__VA_ARGS__)
+#define svcmple_f32(...) __builtin_sve_svcmple_f32(__VA_ARGS__)
+#define svcmple_f16(...) __builtin_sve_svcmple_f16(__VA_ARGS__)
+#define svcmple_u8(...) __builtin_sve_svcmple_u8(__VA_ARGS__)
+#define svcmple_u32(...) __builtin_sve_svcmple_u32(__VA_ARGS__)
+#define svcmple_u64(...) __builtin_sve_svcmple_u64(__VA_ARGS__)
+#define svcmple_u16(...) __builtin_sve_svcmple_u16(__VA_ARGS__)
+#define svcmple_wide_n_s8(...) __builtin_sve_svcmple_wide_n_s8(__VA_ARGS__)
+#define svcmple_wide_n_s32(...) __builtin_sve_svcmple_wide_n_s32(__VA_ARGS__)
+#define svcmple_wide_n_s16(...) __builtin_sve_svcmple_wide_n_s16(__VA_ARGS__)
+#define svcmple_wide_n_u8(...) __builtin_sve_svcmple_wide_n_u8(__VA_ARGS__)
+#define svcmple_wide_n_u32(...) __builtin_sve_svcmple_wide_n_u32(__VA_ARGS__)
+#define svcmple_wide_n_u16(...) __builtin_sve_svcmple_wide_n_u16(__VA_ARGS__)
+#define svcmple_wide_s8(...) __builtin_sve_svcmple_wide_s8(__VA_ARGS__)
+#define svcmple_wide_s32(...) __builtin_sve_svcmple_wide_s32(__VA_ARGS__)
+#define svcmple_wide_s16(...) __builtin_sve_svcmple_wide_s16(__VA_ARGS__)
+#define svcmple_wide_u8(...) __builtin_sve_svcmple_wide_u8(__VA_ARGS__)
+#define svcmple_wide_u32(...) __builtin_sve_svcmple_wide_u32(__VA_ARGS__)
+#define svcmple_wide_u16(...) __builtin_sve_svcmple_wide_u16(__VA_ARGS__)
+#define svcmplt_n_u8(...) __builtin_sve_svcmplt_n_u8(__VA_ARGS__)
+#define svcmplt_n_u32(...) __builtin_sve_svcmplt_n_u32(__VA_ARGS__)
+#define svcmplt_n_u64(...) __builtin_sve_svcmplt_n_u64(__VA_ARGS__)
+#define svcmplt_n_u16(...) __builtin_sve_svcmplt_n_u16(__VA_ARGS__)
+#define svcmplt_n_f64(...) __builtin_sve_svcmplt_n_f64(__VA_ARGS__)
+#define svcmplt_n_f32(...) __builtin_sve_svcmplt_n_f32(__VA_ARGS__)
+#define svcmplt_n_f16(...) __builtin_sve_svcmplt_n_f16(__VA_ARGS__)
+#define svcmplt_n_s8(...) __builtin_sve_svcmplt_n_s8(__VA_ARGS__)
+#define svcmplt_n_s32(...) __builtin_sve_svcmplt_n_s32(__VA_ARGS__)
+#define svcmplt_n_s64(...) __builtin_sve_svcmplt_n_s64(__VA_ARGS__)
+#define svcmplt_n_s16(...) __builtin_sve_svcmplt_n_s16(__VA_ARGS__)
+#define svcmplt_u8(...) __builtin_sve_svcmplt_u8(__VA_ARGS__)
+#define svcmplt_u32(...) __builtin_sve_svcmplt_u32(__VA_ARGS__)
+#define svcmplt_u64(...) __builtin_sve_svcmplt_u64(__VA_ARGS__)
+#define svcmplt_u16(...) __builtin_sve_svcmplt_u16(__VA_ARGS__)
+#define svcmplt_s8(...) __builtin_sve_svcmplt_s8(__VA_ARGS__)
+#define svcmplt_s32(...) __builtin_sve_svcmplt_s32(__VA_ARGS__)
+#define svcmplt_s64(...) __builtin_sve_svcmplt_s64(__VA_ARGS__)
+#define svcmplt_s16(...) __builtin_sve_svcmplt_s16(__VA_ARGS__)
+#define svcmplt_f64(...) __builtin_sve_svcmplt_f64(__VA_ARGS__)
+#define svcmplt_f32(...) __builtin_sve_svcmplt_f32(__VA_ARGS__)
+#define svcmplt_f16(...) __builtin_sve_svcmplt_f16(__VA_ARGS__)
+#define svcmplt_wide_n_u8(...) __builtin_sve_svcmplt_wide_n_u8(__VA_ARGS__)
+#define svcmplt_wide_n_u32(...) __builtin_sve_svcmplt_wide_n_u32(__VA_ARGS__)
+#define svcmplt_wide_n_u16(...) __builtin_sve_svcmplt_wide_n_u16(__VA_ARGS__)
+#define svcmplt_wide_n_s8(...) __builtin_sve_svcmplt_wide_n_s8(__VA_ARGS__)
+#define svcmplt_wide_n_s32(...) __builtin_sve_svcmplt_wide_n_s32(__VA_ARGS__)
+#define svcmplt_wide_n_s16(...) __builtin_sve_svcmplt_wide_n_s16(__VA_ARGS__)
+#define svcmplt_wide_u8(...) __builtin_sve_svcmplt_wide_u8(__VA_ARGS__)
+#define svcmplt_wide_u32(...) __builtin_sve_svcmplt_wide_u32(__VA_ARGS__)
+#define svcmplt_wide_u16(...) __builtin_sve_svcmplt_wide_u16(__VA_ARGS__)
+#define svcmplt_wide_s8(...) __builtin_sve_svcmplt_wide_s8(__VA_ARGS__)
+#define svcmplt_wide_s32(...) __builtin_sve_svcmplt_wide_s32(__VA_ARGS__)
+#define svcmplt_wide_s16(...) __builtin_sve_svcmplt_wide_s16(__VA_ARGS__)
+#define svcmpne_n_f64(...) __builtin_sve_svcmpne_n_f64(__VA_ARGS__)
+#define svcmpne_n_f32(...) __builtin_sve_svcmpne_n_f32(__VA_ARGS__)
+#define svcmpne_n_f16(...) __builtin_sve_svcmpne_n_f16(__VA_ARGS__)
+#define svcmpne_n_u8(...) __builtin_sve_svcmpne_n_u8(__VA_ARGS__)
+#define svcmpne_n_u32(...) __builtin_sve_svcmpne_n_u32(__VA_ARGS__)
+#define svcmpne_n_u64(...) __builtin_sve_svcmpne_n_u64(__VA_ARGS__)
+#define svcmpne_n_u16(...) __builtin_sve_svcmpne_n_u16(__VA_ARGS__)
+#define svcmpne_n_s8(...) __builtin_sve_svcmpne_n_s8(__VA_ARGS__)
+#define svcmpne_n_s32(...) __builtin_sve_svcmpne_n_s32(__VA_ARGS__)
+#define svcmpne_n_s64(...) __builtin_sve_svcmpne_n_s64(__VA_ARGS__)
+#define svcmpne_n_s16(...) __builtin_sve_svcmpne_n_s16(__VA_ARGS__)
+#define svcmpne_u8(...) __builtin_sve_svcmpne_u8(__VA_ARGS__)
+#define svcmpne_u32(...) __builtin_sve_svcmpne_u32(__VA_ARGS__)
+#define svcmpne_u64(...) __builtin_sve_svcmpne_u64(__VA_ARGS__)
+#define svcmpne_u16(...) __builtin_sve_svcmpne_u16(__VA_ARGS__)
+#define svcmpne_s8(...) __builtin_sve_svcmpne_s8(__VA_ARGS__)
+#define svcmpne_s32(...) __builtin_sve_svcmpne_s32(__VA_ARGS__)
+#define svcmpne_s64(...) __builtin_sve_svcmpne_s64(__VA_ARGS__)
+#define svcmpne_s16(...) __builtin_sve_svcmpne_s16(__VA_ARGS__)
+#define svcmpne_f64(...) __builtin_sve_svcmpne_f64(__VA_ARGS__)
+#define svcmpne_f32(...) __builtin_sve_svcmpne_f32(__VA_ARGS__)
+#define svcmpne_f16(...) __builtin_sve_svcmpne_f16(__VA_ARGS__)
+#define svcmpne_wide_n_s8(...) __builtin_sve_svcmpne_wide_n_s8(__VA_ARGS__)
+#define svcmpne_wide_n_s32(...) __builtin_sve_svcmpne_wide_n_s32(__VA_ARGS__)
+#define svcmpne_wide_n_s16(...) __builtin_sve_svcmpne_wide_n_s16(__VA_ARGS__)
+#define svcmpne_wide_s8(...) __builtin_sve_svcmpne_wide_s8(__VA_ARGS__)
+#define svcmpne_wide_s32(...) __builtin_sve_svcmpne_wide_s32(__VA_ARGS__)
+#define svcmpne_wide_s16(...) __builtin_sve_svcmpne_wide_s16(__VA_ARGS__)
+#define svcmpuo_n_f64(...) __builtin_sve_svcmpuo_n_f64(__VA_ARGS__)
+#define svcmpuo_n_f32(...) __builtin_sve_svcmpuo_n_f32(__VA_ARGS__)
+#define svcmpuo_n_f16(...) __builtin_sve_svcmpuo_n_f16(__VA_ARGS__)
+#define svcmpuo_f64(...) __builtin_sve_svcmpuo_f64(__VA_ARGS__)
+#define svcmpuo_f32(...) __builtin_sve_svcmpuo_f32(__VA_ARGS__)
+#define svcmpuo_f16(...) __builtin_sve_svcmpuo_f16(__VA_ARGS__)
+#define svcnot_u8_m(...) __builtin_sve_svcnot_u8_m(__VA_ARGS__)
+#define svcnot_u32_m(...) __builtin_sve_svcnot_u32_m(__VA_ARGS__)
+#define svcnot_u64_m(...) __builtin_sve_svcnot_u64_m(__VA_ARGS__)
+#define svcnot_u16_m(...) __builtin_sve_svcnot_u16_m(__VA_ARGS__)
+#define svcnot_s8_m(...) __builtin_sve_svcnot_s8_m(__VA_ARGS__)
+#define svcnot_s32_m(...) __builtin_sve_svcnot_s32_m(__VA_ARGS__)
+#define svcnot_s64_m(...) __builtin_sve_svcnot_s64_m(__VA_ARGS__)
+#define svcnot_s16_m(...) __builtin_sve_svcnot_s16_m(__VA_ARGS__)
+#define svcnot_u8_x(...) __builtin_sve_svcnot_u8_x(__VA_ARGS__)
+#define svcnot_u32_x(...) __builtin_sve_svcnot_u32_x(__VA_ARGS__)
+#define svcnot_u64_x(...) __builtin_sve_svcnot_u64_x(__VA_ARGS__)
+#define svcnot_u16_x(...) __builtin_sve_svcnot_u16_x(__VA_ARGS__)
+#define svcnot_s8_x(...) __builtin_sve_svcnot_s8_x(__VA_ARGS__)
+#define svcnot_s32_x(...) __builtin_sve_svcnot_s32_x(__VA_ARGS__)
+#define svcnot_s64_x(...) __builtin_sve_svcnot_s64_x(__VA_ARGS__)
+#define svcnot_s16_x(...) __builtin_sve_svcnot_s16_x(__VA_ARGS__)
+#define svcnot_u8_z(...) __builtin_sve_svcnot_u8_z(__VA_ARGS__)
+#define svcnot_u32_z(...) __builtin_sve_svcnot_u32_z(__VA_ARGS__)
+#define svcnot_u64_z(...) __builtin_sve_svcnot_u64_z(__VA_ARGS__)
+#define svcnot_u16_z(...) __builtin_sve_svcnot_u16_z(__VA_ARGS__)
+#define svcnot_s8_z(...) __builtin_sve_svcnot_s8_z(__VA_ARGS__)
+#define svcnot_s32_z(...) __builtin_sve_svcnot_s32_z(__VA_ARGS__)
+#define svcnot_s64_z(...) __builtin_sve_svcnot_s64_z(__VA_ARGS__)
+#define svcnot_s16_z(...) __builtin_sve_svcnot_s16_z(__VA_ARGS__)
+#define svcnt_u8_m(...) __builtin_sve_svcnt_u8_m(__VA_ARGS__)
+#define svcnt_u32_m(...) __builtin_sve_svcnt_u32_m(__VA_ARGS__)
+#define svcnt_u64_m(...) __builtin_sve_svcnt_u64_m(__VA_ARGS__)
+#define svcnt_u16_m(...) __builtin_sve_svcnt_u16_m(__VA_ARGS__)
+#define svcnt_s8_m(...) __builtin_sve_svcnt_s8_m(__VA_ARGS__)
+#define svcnt_f64_m(...) __builtin_sve_svcnt_f64_m(__VA_ARGS__)
+#define svcnt_f32_m(...) __builtin_sve_svcnt_f32_m(__VA_ARGS__)
+#define svcnt_f16_m(...) __builtin_sve_svcnt_f16_m(__VA_ARGS__)
+#define svcnt_s32_m(...) __builtin_sve_svcnt_s32_m(__VA_ARGS__)
+#define svcnt_s64_m(...) __builtin_sve_svcnt_s64_m(__VA_ARGS__)
+#define svcnt_s16_m(...) __builtin_sve_svcnt_s16_m(__VA_ARGS__)
+#define svcnt_u8_x(...) __builtin_sve_svcnt_u8_x(__VA_ARGS__)
+#define svcnt_u32_x(...) __builtin_sve_svcnt_u32_x(__VA_ARGS__)
+#define svcnt_u64_x(...) __builtin_sve_svcnt_u64_x(__VA_ARGS__)
+#define svcnt_u16_x(...) __builtin_sve_svcnt_u16_x(__VA_ARGS__)
+#define svcnt_s8_x(...) __builtin_sve_svcnt_s8_x(__VA_ARGS__)
+#define svcnt_f64_x(...) __builtin_sve_svcnt_f64_x(__VA_ARGS__)
+#define svcnt_f32_x(...) __builtin_sve_svcnt_f32_x(__VA_ARGS__)
+#define svcnt_f16_x(...) __builtin_sve_svcnt_f16_x(__VA_ARGS__)
+#define svcnt_s32_x(...) __builtin_sve_svcnt_s32_x(__VA_ARGS__)
+#define svcnt_s64_x(...) __builtin_sve_svcnt_s64_x(__VA_ARGS__)
+#define svcnt_s16_x(...) __builtin_sve_svcnt_s16_x(__VA_ARGS__)
+#define svcnt_u8_z(...) __builtin_sve_svcnt_u8_z(__VA_ARGS__)
+#define svcnt_u32_z(...) __builtin_sve_svcnt_u32_z(__VA_ARGS__)
+#define svcnt_u64_z(...) __builtin_sve_svcnt_u64_z(__VA_ARGS__)
+#define svcnt_u16_z(...) __builtin_sve_svcnt_u16_z(__VA_ARGS__)
+#define svcnt_s8_z(...) __builtin_sve_svcnt_s8_z(__VA_ARGS__)
+#define svcnt_f64_z(...) __builtin_sve_svcnt_f64_z(__VA_ARGS__)
+#define svcnt_f32_z(...) __builtin_sve_svcnt_f32_z(__VA_ARGS__)
+#define svcnt_f16_z(...) __builtin_sve_svcnt_f16_z(__VA_ARGS__)
+#define svcnt_s32_z(...) __builtin_sve_svcnt_s32_z(__VA_ARGS__)
+#define svcnt_s64_z(...) __builtin_sve_svcnt_s64_z(__VA_ARGS__)
+#define svcnt_s16_z(...) __builtin_sve_svcnt_s16_z(__VA_ARGS__)
+#define svcntb(...) __builtin_sve_svcntb(__VA_ARGS__)
+#define svcntb_pat(...) __builtin_sve_svcntb_pat(__VA_ARGS__)
+#define svcntd(...) __builtin_sve_svcntd(__VA_ARGS__)
+#define svcntd_pat(...) __builtin_sve_svcntd_pat(__VA_ARGS__)
+#define svcnth(...) __builtin_sve_svcnth(__VA_ARGS__)
+#define svcnth_pat(...) __builtin_sve_svcnth_pat(__VA_ARGS__)
+#define svcntp_b8(...) __builtin_sve_svcntp_b8(__VA_ARGS__)
+#define svcntp_b32(...) __builtin_sve_svcntp_b32(__VA_ARGS__)
+#define svcntp_b64(...) __builtin_sve_svcntp_b64(__VA_ARGS__)
+#define svcntp_b16(...) __builtin_sve_svcntp_b16(__VA_ARGS__)
+#define svcntw(...) __builtin_sve_svcntw(__VA_ARGS__)
+#define svcntw_pat(...) __builtin_sve_svcntw_pat(__VA_ARGS__)
+#define svcompact_u32(...) __builtin_sve_svcompact_u32(__VA_ARGS__)
+#define svcompact_u64(...) __builtin_sve_svcompact_u64(__VA_ARGS__)
+#define svcompact_f64(...) __builtin_sve_svcompact_f64(__VA_ARGS__)
+#define svcompact_f32(...) __builtin_sve_svcompact_f32(__VA_ARGS__)
+#define svcompact_s32(...) __builtin_sve_svcompact_s32(__VA_ARGS__)
+#define svcompact_s64(...) __builtin_sve_svcompact_s64(__VA_ARGS__)
+#define svcreate2_u8(...) __builtin_sve_svcreate2_u8(__VA_ARGS__)
+#define svcreate2_u32(...) __builtin_sve_svcreate2_u32(__VA_ARGS__)
+#define svcreate2_u64(...) __builtin_sve_svcreate2_u64(__VA_ARGS__)
+#define svcreate2_u16(...) __builtin_sve_svcreate2_u16(__VA_ARGS__)
+#define svcreate2_s8(...) __builtin_sve_svcreate2_s8(__VA_ARGS__)
+#define svcreate2_f64(...) __builtin_sve_svcreate2_f64(__VA_ARGS__)
+#define svcreate2_f32(...) __builtin_sve_svcreate2_f32(__VA_ARGS__)
+#define svcreate2_f16(...) __builtin_sve_svcreate2_f16(__VA_ARGS__)
+#define svcreate2_s32(...) __builtin_sve_svcreate2_s32(__VA_ARGS__)
+#define svcreate2_s64(...) __builtin_sve_svcreate2_s64(__VA_ARGS__)
+#define svcreate2_s16(...) __builtin_sve_svcreate2_s16(__VA_ARGS__)
+#define svcreate3_u8(...) __builtin_sve_svcreate3_u8(__VA_ARGS__)
+#define svcreate3_u32(...) __builtin_sve_svcreate3_u32(__VA_ARGS__)
+#define svcreate3_u64(...) __builtin_sve_svcreate3_u64(__VA_ARGS__)
+#define svcreate3_u16(...) __builtin_sve_svcreate3_u16(__VA_ARGS__)
+#define svcreate3_s8(...) __builtin_sve_svcreate3_s8(__VA_ARGS__)
+#define svcreate3_f64(...) __builtin_sve_svcreate3_f64(__VA_ARGS__)
+#define svcreate3_f32(...) __builtin_sve_svcreate3_f32(__VA_ARGS__)
+#define svcreate3_f16(...) __builtin_sve_svcreate3_f16(__VA_ARGS__)
+#define svcreate3_s32(...) __builtin_sve_svcreate3_s32(__VA_ARGS__)
+#define svcreate3_s64(...) __builtin_sve_svcreate3_s64(__VA_ARGS__)
+#define svcreate3_s16(...) __builtin_sve_svcreate3_s16(__VA_ARGS__)
+#define svcreate4_u8(...) __builtin_sve_svcreate4_u8(__VA_ARGS__)
+#define svcreate4_u32(...) __builtin_sve_svcreate4_u32(__VA_ARGS__)
+#define svcreate4_u64(...) __builtin_sve_svcreate4_u64(__VA_ARGS__)
+#define svcreate4_u16(...) __builtin_sve_svcreate4_u16(__VA_ARGS__)
+#define svcreate4_s8(...) __builtin_sve_svcreate4_s8(__VA_ARGS__)
+#define svcreate4_f64(...) __builtin_sve_svcreate4_f64(__VA_ARGS__)
+#define svcreate4_f32(...) __builtin_sve_svcreate4_f32(__VA_ARGS__)
+#define svcreate4_f16(...) __builtin_sve_svcreate4_f16(__VA_ARGS__)
+#define svcreate4_s32(...) __builtin_sve_svcreate4_s32(__VA_ARGS__)
+#define svcreate4_s64(...) __builtin_sve_svcreate4_s64(__VA_ARGS__)
+#define svcreate4_s16(...) __builtin_sve_svcreate4_s16(__VA_ARGS__)
+#define svcvt_f16_f32_m(...) __builtin_sve_svcvt_f16_f32_m(__VA_ARGS__)
+#define svcvt_f16_f32_x(...) __builtin_sve_svcvt_f16_f32_x(__VA_ARGS__)
+#define svcvt_f16_f32_z(...) __builtin_sve_svcvt_f16_f32_z(__VA_ARGS__)
+#define svcvt_f16_f64_m(...) __builtin_sve_svcvt_f16_f64_m(__VA_ARGS__)
+#define svcvt_f16_f64_x(...) __builtin_sve_svcvt_f16_f64_x(__VA_ARGS__)
+#define svcvt_f16_f64_z(...) __builtin_sve_svcvt_f16_f64_z(__VA_ARGS__)
+#define svcvt_f16_s16_m(...) __builtin_sve_svcvt_f16_s16_m(__VA_ARGS__)
+#define svcvt_f16_s16_x(...) __builtin_sve_svcvt_f16_s16_x(__VA_ARGS__)
+#define svcvt_f16_s16_z(...) __builtin_sve_svcvt_f16_s16_z(__VA_ARGS__)
+#define svcvt_f16_s32_m(...) __builtin_sve_svcvt_f16_s32_m(__VA_ARGS__)
+#define svcvt_f16_s32_x(...) __builtin_sve_svcvt_f16_s32_x(__VA_ARGS__)
+#define svcvt_f16_s32_z(...) __builtin_sve_svcvt_f16_s32_z(__VA_ARGS__)
+#define svcvt_f16_s64_m(...) __builtin_sve_svcvt_f16_s64_m(__VA_ARGS__)
+#define svcvt_f16_s64_x(...) __builtin_sve_svcvt_f16_s64_x(__VA_ARGS__)
+#define svcvt_f16_s64_z(...) __builtin_sve_svcvt_f16_s64_z(__VA_ARGS__)
+#define svcvt_f16_u16_m(...) __builtin_sve_svcvt_f16_u16_m(__VA_ARGS__)
+#define svcvt_f16_u16_x(...) __builtin_sve_svcvt_f16_u16_x(__VA_ARGS__)
+#define svcvt_f16_u16_z(...) __builtin_sve_svcvt_f16_u16_z(__VA_ARGS__)
+#define svcvt_f16_u32_m(...) __builtin_sve_svcvt_f16_u32_m(__VA_ARGS__)
+#define svcvt_f16_u32_x(...) __builtin_sve_svcvt_f16_u32_x(__VA_ARGS__)
+#define svcvt_f16_u32_z(...) __builtin_sve_svcvt_f16_u32_z(__VA_ARGS__)
+#define svcvt_f16_u64_m(...) __builtin_sve_svcvt_f16_u64_m(__VA_ARGS__)
+#define svcvt_f16_u64_x(...) __builtin_sve_svcvt_f16_u64_x(__VA_ARGS__)
+#define svcvt_f16_u64_z(...) __builtin_sve_svcvt_f16_u64_z(__VA_ARGS__)
+#define svcvt_f32_f16_m(...) __builtin_sve_svcvt_f32_f16_m(__VA_ARGS__)
+#define svcvt_f32_f16_x(...) __builtin_sve_svcvt_f32_f16_x(__VA_ARGS__)
+#define svcvt_f32_f16_z(...) __builtin_sve_svcvt_f32_f16_z(__VA_ARGS__)
+#define svcvt_f32_f64_m(...) __builtin_sve_svcvt_f32_f64_m(__VA_ARGS__)
+#define svcvt_f32_f64_x(...) __builtin_sve_svcvt_f32_f64_x(__VA_ARGS__)
+#define svcvt_f32_f64_z(...) __builtin_sve_svcvt_f32_f64_z(__VA_ARGS__)
+#define svcvt_f32_s32_m(...) __builtin_sve_svcvt_f32_s32_m(__VA_ARGS__)
+#define svcvt_f32_s32_x(...) __builtin_sve_svcvt_f32_s32_x(__VA_ARGS__)
+#define svcvt_f32_s32_z(...) __builtin_sve_svcvt_f32_s32_z(__VA_ARGS__)
+#define svcvt_f32_s64_m(...) __builtin_sve_svcvt_f32_s64_m(__VA_ARGS__)
+#define svcvt_f32_s64_x(...) __builtin_sve_svcvt_f32_s64_x(__VA_ARGS__)
+#define svcvt_f32_s64_z(...) __builtin_sve_svcvt_f32_s64_z(__VA_ARGS__)
+#define svcvt_f32_u32_m(...) __builtin_sve_svcvt_f32_u32_m(__VA_ARGS__)
+#define svcvt_f32_u32_x(...) __builtin_sve_svcvt_f32_u32_x(__VA_ARGS__)
+#define svcvt_f32_u32_z(...) __builtin_sve_svcvt_f32_u32_z(__VA_ARGS__)
+#define svcvt_f32_u64_m(...) __builtin_sve_svcvt_f32_u64_m(__VA_ARGS__)
+#define svcvt_f32_u64_x(...) __builtin_sve_svcvt_f32_u64_x(__VA_ARGS__)
+#define svcvt_f32_u64_z(...) __builtin_sve_svcvt_f32_u64_z(__VA_ARGS__)
+#define svcvt_f64_f16_m(...) __builtin_sve_svcvt_f64_f16_m(__VA_ARGS__)
+#define svcvt_f64_f16_x(...) __builtin_sve_svcvt_f64_f16_x(__VA_ARGS__)
+#define svcvt_f64_f16_z(...) __builtin_sve_svcvt_f64_f16_z(__VA_ARGS__)
+#define svcvt_f64_f32_m(...) __builtin_sve_svcvt_f64_f32_m(__VA_ARGS__)
+#define svcvt_f64_f32_x(...) __builtin_sve_svcvt_f64_f32_x(__VA_ARGS__)
+#define svcvt_f64_f32_z(...) __builtin_sve_svcvt_f64_f32_z(__VA_ARGS__)
+#define svcvt_f64_s32_m(...) __builtin_sve_svcvt_f64_s32_m(__VA_ARGS__)
+#define svcvt_f64_s32_x(...) __builtin_sve_svcvt_f64_s32_x(__VA_ARGS__)
+#define svcvt_f64_s32_z(...) __builtin_sve_svcvt_f64_s32_z(__VA_ARGS__)
+#define svcvt_f64_s64_m(...) __builtin_sve_svcvt_f64_s64_m(__VA_ARGS__)
+#define svcvt_f64_s64_x(...) __builtin_sve_svcvt_f64_s64_x(__VA_ARGS__)
+#define svcvt_f64_s64_z(...) __builtin_sve_svcvt_f64_s64_z(__VA_ARGS__)
+#define svcvt_f64_u32_m(...) __builtin_sve_svcvt_f64_u32_m(__VA_ARGS__)
+#define svcvt_f64_u32_x(...) __builtin_sve_svcvt_f64_u32_x(__VA_ARGS__)
+#define svcvt_f64_u32_z(...) __builtin_sve_svcvt_f64_u32_z(__VA_ARGS__)
+#define svcvt_f64_u64_m(...) __builtin_sve_svcvt_f64_u64_m(__VA_ARGS__)
+#define svcvt_f64_u64_x(...) __builtin_sve_svcvt_f64_u64_x(__VA_ARGS__)
+#define svcvt_f64_u64_z(...) __builtin_sve_svcvt_f64_u64_z(__VA_ARGS__)
+#define svcvt_s16_f16_m(...) __builtin_sve_svcvt_s16_f16_m(__VA_ARGS__)
+#define svcvt_s16_f16_x(...) __builtin_sve_svcvt_s16_f16_x(__VA_ARGS__)
+#define svcvt_s16_f16_z(...) __builtin_sve_svcvt_s16_f16_z(__VA_ARGS__)
+#define svcvt_s32_f16_m(...) __builtin_sve_svcvt_s32_f16_m(__VA_ARGS__)
+#define svcvt_s32_f16_x(...) __builtin_sve_svcvt_s32_f16_x(__VA_ARGS__)
+#define svcvt_s32_f16_z(...) __builtin_sve_svcvt_s32_f16_z(__VA_ARGS__)
+#define svcvt_s32_f32_m(...) __builtin_sve_svcvt_s32_f32_m(__VA_ARGS__)
+#define svcvt_s32_f32_x(...) __builtin_sve_svcvt_s32_f32_x(__VA_ARGS__)
+#define svcvt_s32_f32_z(...) __builtin_sve_svcvt_s32_f32_z(__VA_ARGS__)
+#define svcvt_s32_f64_m(...) __builtin_sve_svcvt_s32_f64_m(__VA_ARGS__)
+#define svcvt_s32_f64_x(...) __builtin_sve_svcvt_s32_f64_x(__VA_ARGS__)
+#define svcvt_s32_f64_z(...) __builtin_sve_svcvt_s32_f64_z(__VA_ARGS__)
+#define svcvt_s64_f16_m(...) __builtin_sve_svcvt_s64_f16_m(__VA_ARGS__)
+#define svcvt_s64_f16_x(...) __builtin_sve_svcvt_s64_f16_x(__VA_ARGS__)
+#define svcvt_s64_f16_z(...) __builtin_sve_svcvt_s64_f16_z(__VA_ARGS__)
+#define svcvt_s64_f32_m(...) __builtin_sve_svcvt_s64_f32_m(__VA_ARGS__)
+#define svcvt_s64_f32_x(...) __builtin_sve_svcvt_s64_f32_x(__VA_ARGS__)
+#define svcvt_s64_f32_z(...) __builtin_sve_svcvt_s64_f32_z(__VA_ARGS__)
+#define svcvt_s64_f64_m(...) __builtin_sve_svcvt_s64_f64_m(__VA_ARGS__)
+#define svcvt_s64_f64_x(...) __builtin_sve_svcvt_s64_f64_x(__VA_ARGS__)
+#define svcvt_s64_f64_z(...) __builtin_sve_svcvt_s64_f64_z(__VA_ARGS__)
+#define svcvt_u16_f16_m(...) __builtin_sve_svcvt_u16_f16_m(__VA_ARGS__)
+#define svcvt_u16_f16_x(...) __builtin_sve_svcvt_u16_f16_x(__VA_ARGS__)
+#define svcvt_u16_f16_z(...) __builtin_sve_svcvt_u16_f16_z(__VA_ARGS__)
+#define svcvt_u32_f16_m(...) __builtin_sve_svcvt_u32_f16_m(__VA_ARGS__)
+#define svcvt_u32_f16_x(...) __builtin_sve_svcvt_u32_f16_x(__VA_ARGS__)
+#define svcvt_u32_f16_z(...) __builtin_sve_svcvt_u32_f16_z(__VA_ARGS__)
+#define svcvt_u32_f32_m(...) __builtin_sve_svcvt_u32_f32_m(__VA_ARGS__)
+#define svcvt_u32_f32_x(...) __builtin_sve_svcvt_u32_f32_x(__VA_ARGS__)
+#define svcvt_u32_f32_z(...) __builtin_sve_svcvt_u32_f32_z(__VA_ARGS__)
+#define svcvt_u32_f64_m(...) __builtin_sve_svcvt_u32_f64_m(__VA_ARGS__)
+#define svcvt_u32_f64_x(...) __builtin_sve_svcvt_u32_f64_x(__VA_ARGS__)
+#define svcvt_u32_f64_z(...) __builtin_sve_svcvt_u32_f64_z(__VA_ARGS__)
+#define svcvt_u64_f16_m(...) __builtin_sve_svcvt_u64_f16_m(__VA_ARGS__)
+#define svcvt_u64_f16_x(...) __builtin_sve_svcvt_u64_f16_x(__VA_ARGS__)
+#define svcvt_u64_f16_z(...) __builtin_sve_svcvt_u64_f16_z(__VA_ARGS__)
+#define svcvt_u64_f32_m(...) __builtin_sve_svcvt_u64_f32_m(__VA_ARGS__)
+#define svcvt_u64_f32_x(...) __builtin_sve_svcvt_u64_f32_x(__VA_ARGS__)
+#define svcvt_u64_f32_z(...) __builtin_sve_svcvt_u64_f32_z(__VA_ARGS__)
+#define svcvt_u64_f64_m(...) __builtin_sve_svcvt_u64_f64_m(__VA_ARGS__)
+#define svcvt_u64_f64_x(...) __builtin_sve_svcvt_u64_f64_x(__VA_ARGS__)
+#define svcvt_u64_f64_z(...) __builtin_sve_svcvt_u64_f64_z(__VA_ARGS__)
+#define svdiv_n_f64_m(...) __builtin_sve_svdiv_n_f64_m(__VA_ARGS__)
+#define svdiv_n_f32_m(...) __builtin_sve_svdiv_n_f32_m(__VA_ARGS__)
+#define svdiv_n_f16_m(...) __builtin_sve_svdiv_n_f16_m(__VA_ARGS__)
+#define svdiv_n_f64_x(...) __builtin_sve_svdiv_n_f64_x(__VA_ARGS__)
+#define svdiv_n_f32_x(...) __builtin_sve_svdiv_n_f32_x(__VA_ARGS__)
+#define svdiv_n_f16_x(...) __builtin_sve_svdiv_n_f16_x(__VA_ARGS__)
+#define svdiv_n_f64_z(...) __builtin_sve_svdiv_n_f64_z(__VA_ARGS__)
+#define svdiv_n_f32_z(...) __builtin_sve_svdiv_n_f32_z(__VA_ARGS__)
+#define svdiv_n_f16_z(...) __builtin_sve_svdiv_n_f16_z(__VA_ARGS__)
+#define svdiv_n_s32_m(...) __builtin_sve_svdiv_n_s32_m(__VA_ARGS__)
+#define svdiv_n_s64_m(...) __builtin_sve_svdiv_n_s64_m(__VA_ARGS__)
+#define svdiv_n_s32_x(...) __builtin_sve_svdiv_n_s32_x(__VA_ARGS__)
+#define svdiv_n_s64_x(...) __builtin_sve_svdiv_n_s64_x(__VA_ARGS__)
+#define svdiv_n_s32_z(...) __builtin_sve_svdiv_n_s32_z(__VA_ARGS__)
+#define svdiv_n_s64_z(...) __builtin_sve_svdiv_n_s64_z(__VA_ARGS__)
+#define svdiv_n_u32_m(...) __builtin_sve_svdiv_n_u32_m(__VA_ARGS__)
+#define svdiv_n_u64_m(...) __builtin_sve_svdiv_n_u64_m(__VA_ARGS__)
+#define svdiv_n_u32_x(...) __builtin_sve_svdiv_n_u32_x(__VA_ARGS__)
+#define svdiv_n_u64_x(...) __builtin_sve_svdiv_n_u64_x(__VA_ARGS__)
+#define svdiv_n_u32_z(...) __builtin_sve_svdiv_n_u32_z(__VA_ARGS__)
+#define svdiv_n_u64_z(...) __builtin_sve_svdiv_n_u64_z(__VA_ARGS__)
+#define svdiv_f64_m(...) __builtin_sve_svdiv_f64_m(__VA_ARGS__)
+#define svdiv_f32_m(...) __builtin_sve_svdiv_f32_m(__VA_ARGS__)
+#define svdiv_f16_m(...) __builtin_sve_svdiv_f16_m(__VA_ARGS__)
+#define svdiv_f64_x(...) __builtin_sve_svdiv_f64_x(__VA_ARGS__)
+#define svdiv_f32_x(...) __builtin_sve_svdiv_f32_x(__VA_ARGS__)
+#define svdiv_f16_x(...) __builtin_sve_svdiv_f16_x(__VA_ARGS__)
+#define svdiv_f64_z(...) __builtin_sve_svdiv_f64_z(__VA_ARGS__)
+#define svdiv_f32_z(...) __builtin_sve_svdiv_f32_z(__VA_ARGS__)
+#define svdiv_f16_z(...) __builtin_sve_svdiv_f16_z(__VA_ARGS__)
+#define svdiv_s32_m(...) __builtin_sve_svdiv_s32_m(__VA_ARGS__)
+#define svdiv_s64_m(...) __builtin_sve_svdiv_s64_m(__VA_ARGS__)
+#define svdiv_s32_x(...) __builtin_sve_svdiv_s32_x(__VA_ARGS__)
+#define svdiv_s64_x(...) __builtin_sve_svdiv_s64_x(__VA_ARGS__)
+#define svdiv_s32_z(...) __builtin_sve_svdiv_s32_z(__VA_ARGS__)
+#define svdiv_s64_z(...) __builtin_sve_svdiv_s64_z(__VA_ARGS__)
+#define svdiv_u32_m(...) __builtin_sve_svdiv_u32_m(__VA_ARGS__)
+#define svdiv_u64_m(...) __builtin_sve_svdiv_u64_m(__VA_ARGS__)
+#define svdiv_u32_x(...) __builtin_sve_svdiv_u32_x(__VA_ARGS__)
+#define svdiv_u64_x(...) __builtin_sve_svdiv_u64_x(__VA_ARGS__)
+#define svdiv_u32_z(...) __builtin_sve_svdiv_u32_z(__VA_ARGS__)
+#define svdiv_u64_z(...) __builtin_sve_svdiv_u64_z(__VA_ARGS__)
+#define svdivr_n_f64_m(...) __builtin_sve_svdivr_n_f64_m(__VA_ARGS__)
+#define svdivr_n_f32_m(...) __builtin_sve_svdivr_n_f32_m(__VA_ARGS__)
+#define svdivr_n_f16_m(...) __builtin_sve_svdivr_n_f16_m(__VA_ARGS__)
+#define svdivr_n_f64_x(...) __builtin_sve_svdivr_n_f64_x(__VA_ARGS__)
+#define svdivr_n_f32_x(...) __builtin_sve_svdivr_n_f32_x(__VA_ARGS__)
+#define svdivr_n_f16_x(...) __builtin_sve_svdivr_n_f16_x(__VA_ARGS__)
+#define svdivr_n_f64_z(...) __builtin_sve_svdivr_n_f64_z(__VA_ARGS__)
+#define svdivr_n_f32_z(...) __builtin_sve_svdivr_n_f32_z(__VA_ARGS__)
+#define svdivr_n_f16_z(...) __builtin_sve_svdivr_n_f16_z(__VA_ARGS__)
+#define svdivr_n_s32_m(...) __builtin_sve_svdivr_n_s32_m(__VA_ARGS__)
+#define svdivr_n_s64_m(...) __builtin_sve_svdivr_n_s64_m(__VA_ARGS__)
+#define svdivr_n_s32_x(...) __builtin_sve_svdivr_n_s32_x(__VA_ARGS__)
+#define svdivr_n_s64_x(...) __builtin_sve_svdivr_n_s64_x(__VA_ARGS__)
+#define svdivr_n_s32_z(...) __builtin_sve_svdivr_n_s32_z(__VA_ARGS__)
+#define svdivr_n_s64_z(...) __builtin_sve_svdivr_n_s64_z(__VA_ARGS__)
+#define svdivr_n_u32_m(...) __builtin_sve_svdivr_n_u32_m(__VA_ARGS__)
+#define svdivr_n_u64_m(...) __builtin_sve_svdivr_n_u64_m(__VA_ARGS__)
+#define svdivr_n_u32_x(...) __builtin_sve_svdivr_n_u32_x(__VA_ARGS__)
+#define svdivr_n_u64_x(...) __builtin_sve_svdivr_n_u64_x(__VA_ARGS__)
+#define svdivr_n_u32_z(...) __builtin_sve_svdivr_n_u32_z(__VA_ARGS__)
+#define svdivr_n_u64_z(...) __builtin_sve_svdivr_n_u64_z(__VA_ARGS__)
+#define svdivr_f64_m(...) __builtin_sve_svdivr_f64_m(__VA_ARGS__)
+#define svdivr_f32_m(...) __builtin_sve_svdivr_f32_m(__VA_ARGS__)
+#define svdivr_f16_m(...) __builtin_sve_svdivr_f16_m(__VA_ARGS__)
+#define svdivr_f64_x(...) __builtin_sve_svdivr_f64_x(__VA_ARGS__)
+#define svdivr_f32_x(...) __builtin_sve_svdivr_f32_x(__VA_ARGS__)
+#define svdivr_f16_x(...) __builtin_sve_svdivr_f16_x(__VA_ARGS__)
+#define svdivr_f64_z(...) __builtin_sve_svdivr_f64_z(__VA_ARGS__)
+#define svdivr_f32_z(...) __builtin_sve_svdivr_f32_z(__VA_ARGS__)
+#define svdivr_f16_z(...) __builtin_sve_svdivr_f16_z(__VA_ARGS__)
+#define svdivr_s32_m(...) __builtin_sve_svdivr_s32_m(__VA_ARGS__)
+#define svdivr_s64_m(...) __builtin_sve_svdivr_s64_m(__VA_ARGS__)
+#define svdivr_s32_x(...) __builtin_sve_svdivr_s32_x(__VA_ARGS__)
+#define svdivr_s64_x(...) __builtin_sve_svdivr_s64_x(__VA_ARGS__)
+#define svdivr_s32_z(...) __builtin_sve_svdivr_s32_z(__VA_ARGS__)
+#define svdivr_s64_z(...) __builtin_sve_svdivr_s64_z(__VA_ARGS__)
+#define svdivr_u32_m(...) __builtin_sve_svdivr_u32_m(__VA_ARGS__)
+#define svdivr_u64_m(...) __builtin_sve_svdivr_u64_m(__VA_ARGS__)
+#define svdivr_u32_x(...) __builtin_sve_svdivr_u32_x(__VA_ARGS__)
+#define svdivr_u64_x(...) __builtin_sve_svdivr_u64_x(__VA_ARGS__)
+#define svdivr_u32_z(...) __builtin_sve_svdivr_u32_z(__VA_ARGS__)
+#define svdivr_u64_z(...) __builtin_sve_svdivr_u64_z(__VA_ARGS__)
+#define svdot_n_s32(...) __builtin_sve_svdot_n_s32(__VA_ARGS__)
+#define svdot_n_s64(...) __builtin_sve_svdot_n_s64(__VA_ARGS__)
+#define svdot_n_u32(...) __builtin_sve_svdot_n_u32(__VA_ARGS__)
+#define svdot_n_u64(...) __builtin_sve_svdot_n_u64(__VA_ARGS__)
+#define svdot_s32(...) __builtin_sve_svdot_s32(__VA_ARGS__)
+#define svdot_s64(...) __builtin_sve_svdot_s64(__VA_ARGS__)
+#define svdot_u32(...) __builtin_sve_svdot_u32(__VA_ARGS__)
+#define svdot_u64(...) __builtin_sve_svdot_u64(__VA_ARGS__)
+#define svdot_lane_s32(...) __builtin_sve_svdot_lane_s32(__VA_ARGS__)
+#define svdot_lane_s64(...) __builtin_sve_svdot_lane_s64(__VA_ARGS__)
+#define svdot_lane_u32(...) __builtin_sve_svdot_lane_u32(__VA_ARGS__)
+#define svdot_lane_u64(...) __builtin_sve_svdot_lane_u64(__VA_ARGS__)
+#define svdup_n_u8(...) __builtin_sve_svdup_n_u8(__VA_ARGS__)
+#define svdup_n_u32(...) __builtin_sve_svdup_n_u32(__VA_ARGS__)
+#define svdup_n_u64(...) __builtin_sve_svdup_n_u64(__VA_ARGS__)
+#define svdup_n_u16(...) __builtin_sve_svdup_n_u16(__VA_ARGS__)
+#define svdup_n_s8(...) __builtin_sve_svdup_n_s8(__VA_ARGS__)
+#define svdup_n_f64(...) __builtin_sve_svdup_n_f64(__VA_ARGS__)
+#define svdup_n_f32(...) __builtin_sve_svdup_n_f32(__VA_ARGS__)
+#define svdup_n_f16(...) __builtin_sve_svdup_n_f16(__VA_ARGS__)
+#define svdup_n_s32(...) __builtin_sve_svdup_n_s32(__VA_ARGS__)
+#define svdup_n_s64(...) __builtin_sve_svdup_n_s64(__VA_ARGS__)
+#define svdup_n_s16(...) __builtin_sve_svdup_n_s16(__VA_ARGS__)
+#define svdup_n_u8_m(...) __builtin_sve_svdup_n_u8_m(__VA_ARGS__)
+#define svdup_n_u32_m(...) __builtin_sve_svdup_n_u32_m(__VA_ARGS__)
+#define svdup_n_u64_m(...) __builtin_sve_svdup_n_u64_m(__VA_ARGS__)
+#define svdup_n_u16_m(...) __builtin_sve_svdup_n_u16_m(__VA_ARGS__)
+#define svdup_n_s8_m(...) __builtin_sve_svdup_n_s8_m(__VA_ARGS__)
+#define svdup_n_f64_m(...) __builtin_sve_svdup_n_f64_m(__VA_ARGS__)
+#define svdup_n_f32_m(...) __builtin_sve_svdup_n_f32_m(__VA_ARGS__)
+#define svdup_n_f16_m(...) __builtin_sve_svdup_n_f16_m(__VA_ARGS__)
+#define svdup_n_s32_m(...) __builtin_sve_svdup_n_s32_m(__VA_ARGS__)
+#define svdup_n_s64_m(...) __builtin_sve_svdup_n_s64_m(__VA_ARGS__)
+#define svdup_n_s16_m(...) __builtin_sve_svdup_n_s16_m(__VA_ARGS__)
+#define svdup_n_b8(...) __builtin_sve_svdup_n_b8(__VA_ARGS__)
+#define svdup_n_b32(...) __builtin_sve_svdup_n_b32(__VA_ARGS__)
+#define svdup_n_b64(...) __builtin_sve_svdup_n_b64(__VA_ARGS__)
+#define svdup_n_b16(...) __builtin_sve_svdup_n_b16(__VA_ARGS__)
+#define svdup_n_u8_x(...) __builtin_sve_svdup_n_u8_x(__VA_ARGS__)
+#define svdup_n_u32_x(...) __builtin_sve_svdup_n_u32_x(__VA_ARGS__)
+#define svdup_n_u64_x(...) __builtin_sve_svdup_n_u64_x(__VA_ARGS__)
+#define svdup_n_u16_x(...) __builtin_sve_svdup_n_u16_x(__VA_ARGS__)
+#define svdup_n_s8_x(...) __builtin_sve_svdup_n_s8_x(__VA_ARGS__)
+#define svdup_n_f64_x(...) __builtin_sve_svdup_n_f64_x(__VA_ARGS__)
+#define svdup_n_f32_x(...) __builtin_sve_svdup_n_f32_x(__VA_ARGS__)
+#define svdup_n_f16_x(...) __builtin_sve_svdup_n_f16_x(__VA_ARGS__)
+#define svdup_n_s32_x(...) __builtin_sve_svdup_n_s32_x(__VA_ARGS__)
+#define svdup_n_s64_x(...) __builtin_sve_svdup_n_s64_x(__VA_ARGS__)
+#define svdup_n_s16_x(...) __builtin_sve_svdup_n_s16_x(__VA_ARGS__)
+#define svdup_n_u8_z(...) __builtin_sve_svdup_n_u8_z(__VA_ARGS__)
+#define svdup_n_u32_z(...) __builtin_sve_svdup_n_u32_z(__VA_ARGS__)
+#define svdup_n_u64_z(...) __builtin_sve_svdup_n_u64_z(__VA_ARGS__)
+#define svdup_n_u16_z(...) __builtin_sve_svdup_n_u16_z(__VA_ARGS__)
+#define svdup_n_s8_z(...) __builtin_sve_svdup_n_s8_z(__VA_ARGS__)
+#define svdup_n_f64_z(...) __builtin_sve_svdup_n_f64_z(__VA_ARGS__)
+#define svdup_n_f32_z(...) __builtin_sve_svdup_n_f32_z(__VA_ARGS__)
+#define svdup_n_f16_z(...) __builtin_sve_svdup_n_f16_z(__VA_ARGS__)
+#define svdup_n_s32_z(...) __builtin_sve_svdup_n_s32_z(__VA_ARGS__)
+#define svdup_n_s64_z(...) __builtin_sve_svdup_n_s64_z(__VA_ARGS__)
+#define svdup_n_s16_z(...) __builtin_sve_svdup_n_s16_z(__VA_ARGS__)
+#define svdup_lane_u8(...) __builtin_sve_svdup_lane_u8(__VA_ARGS__)
+#define svdup_lane_u32(...) __builtin_sve_svdup_lane_u32(__VA_ARGS__)
+#define svdup_lane_u64(...) __builtin_sve_svdup_lane_u64(__VA_ARGS__)
+#define svdup_lane_u16(...) __builtin_sve_svdup_lane_u16(__VA_ARGS__)
+#define svdup_lane_s8(...) __builtin_sve_svdup_lane_s8(__VA_ARGS__)
+#define svdup_lane_f64(...) __builtin_sve_svdup_lane_f64(__VA_ARGS__)
+#define svdup_lane_f32(...) __builtin_sve_svdup_lane_f32(__VA_ARGS__)
+#define svdup_lane_f16(...) __builtin_sve_svdup_lane_f16(__VA_ARGS__)
+#define svdup_lane_s32(...) __builtin_sve_svdup_lane_s32(__VA_ARGS__)
+#define svdup_lane_s64(...) __builtin_sve_svdup_lane_s64(__VA_ARGS__)
+#define svdup_lane_s16(...) __builtin_sve_svdup_lane_s16(__VA_ARGS__)
+#define svdupq_n_u16(...) __builtin_sve_svdupq_n_u16(__VA_ARGS__)
+#define svdupq_n_f16(...) __builtin_sve_svdupq_n_f16(__VA_ARGS__)
+#define svdupq_n_s16(...) __builtin_sve_svdupq_n_s16(__VA_ARGS__)
+#define svdupq_n_u32(...) __builtin_sve_svdupq_n_u32(__VA_ARGS__)
+#define svdupq_n_f32(...) __builtin_sve_svdupq_n_f32(__VA_ARGS__)
+#define svdupq_n_s32(...) __builtin_sve_svdupq_n_s32(__VA_ARGS__)
+#define svdupq_n_u64(...) __builtin_sve_svdupq_n_u64(__VA_ARGS__)
+#define svdupq_n_f64(...) __builtin_sve_svdupq_n_f64(__VA_ARGS__)
+#define svdupq_n_s64(...) __builtin_sve_svdupq_n_s64(__VA_ARGS__)
+#define svdupq_n_u8(...) __builtin_sve_svdupq_n_u8(__VA_ARGS__)
+#define svdupq_n_s8(...) __builtin_sve_svdupq_n_s8(__VA_ARGS__)
+#define svdupq_n_b16(...) __builtin_sve_svdupq_n_b16(__VA_ARGS__)
+#define svdupq_n_b32(...) __builtin_sve_svdupq_n_b32(__VA_ARGS__)
+#define svdupq_n_b64(...) __builtin_sve_svdupq_n_b64(__VA_ARGS__)
+#define svdupq_n_b8(...) __builtin_sve_svdupq_n_b8(__VA_ARGS__)
+#define svdupq_lane_u8(...) __builtin_sve_svdupq_lane_u8(__VA_ARGS__)
+#define svdupq_lane_u32(...) __builtin_sve_svdupq_lane_u32(__VA_ARGS__)
+#define svdupq_lane_u64(...) __builtin_sve_svdupq_lane_u64(__VA_ARGS__)
+#define svdupq_lane_u16(...) __builtin_sve_svdupq_lane_u16(__VA_ARGS__)
+#define svdupq_lane_s8(...) __builtin_sve_svdupq_lane_s8(__VA_ARGS__)
+#define svdupq_lane_f64(...) __builtin_sve_svdupq_lane_f64(__VA_ARGS__)
+#define svdupq_lane_f32(...) __builtin_sve_svdupq_lane_f32(__VA_ARGS__)
+#define svdupq_lane_f16(...) __builtin_sve_svdupq_lane_f16(__VA_ARGS__)
+#define svdupq_lane_s32(...) __builtin_sve_svdupq_lane_s32(__VA_ARGS__)
+#define svdupq_lane_s64(...) __builtin_sve_svdupq_lane_s64(__VA_ARGS__)
+#define svdupq_lane_s16(...) __builtin_sve_svdupq_lane_s16(__VA_ARGS__)
+#define sveor_b_z(...) __builtin_sve_sveor_b_z(__VA_ARGS__)
+#define sveor_n_u8_m(...) __builtin_sve_sveor_n_u8_m(__VA_ARGS__)
+#define sveor_n_u32_m(...) __builtin_sve_sveor_n_u32_m(__VA_ARGS__)
+#define sveor_n_u64_m(...) __builtin_sve_sveor_n_u64_m(__VA_ARGS__)
+#define sveor_n_u16_m(...) __builtin_sve_sveor_n_u16_m(__VA_ARGS__)
+#define sveor_n_s8_m(...) __builtin_sve_sveor_n_s8_m(__VA_ARGS__)
+#define sveor_n_s32_m(...) __builtin_sve_sveor_n_s32_m(__VA_ARGS__)
+#define sveor_n_s64_m(...) __builtin_sve_sveor_n_s64_m(__VA_ARGS__)
+#define sveor_n_s16_m(...) __builtin_sve_sveor_n_s16_m(__VA_ARGS__)
+#define sveor_n_u8_x(...) __builtin_sve_sveor_n_u8_x(__VA_ARGS__)
+#define sveor_n_u32_x(...) __builtin_sve_sveor_n_u32_x(__VA_ARGS__)
+#define sveor_n_u64_x(...) __builtin_sve_sveor_n_u64_x(__VA_ARGS__)
+#define sveor_n_u16_x(...) __builtin_sve_sveor_n_u16_x(__VA_ARGS__)
+#define sveor_n_s8_x(...) __builtin_sve_sveor_n_s8_x(__VA_ARGS__)
+#define sveor_n_s32_x(...) __builtin_sve_sveor_n_s32_x(__VA_ARGS__)
+#define sveor_n_s64_x(...) __builtin_sve_sveor_n_s64_x(__VA_ARGS__)
+#define sveor_n_s16_x(...) __builtin_sve_sveor_n_s16_x(__VA_ARGS__)
+#define sveor_n_u8_z(...) __builtin_sve_sveor_n_u8_z(__VA_ARGS__)
+#define sveor_n_u32_z(...) __builtin_sve_sveor_n_u32_z(__VA_ARGS__)
+#define sveor_n_u64_z(...) __builtin_sve_sveor_n_u64_z(__VA_ARGS__)
+#define sveor_n_u16_z(...) __builtin_sve_sveor_n_u16_z(__VA_ARGS__)
+#define sveor_n_s8_z(...) __builtin_sve_sveor_n_s8_z(__VA_ARGS__)
+#define sveor_n_s32_z(...) __builtin_sve_sveor_n_s32_z(__VA_ARGS__)
+#define sveor_n_s64_z(...) __builtin_sve_sveor_n_s64_z(__VA_ARGS__)
+#define sveor_n_s16_z(...) __builtin_sve_sveor_n_s16_z(__VA_ARGS__)
+#define sveor_u8_m(...) __builtin_sve_sveor_u8_m(__VA_ARGS__)
+#define sveor_u32_m(...) __builtin_sve_sveor_u32_m(__VA_ARGS__)
+#define sveor_u64_m(...) __builtin_sve_sveor_u64_m(__VA_ARGS__)
+#define sveor_u16_m(...) __builtin_sve_sveor_u16_m(__VA_ARGS__)
+#define sveor_s8_m(...) __builtin_sve_sveor_s8_m(__VA_ARGS__)
+#define sveor_s32_m(...) __builtin_sve_sveor_s32_m(__VA_ARGS__)
+#define sveor_s64_m(...) __builtin_sve_sveor_s64_m(__VA_ARGS__)
+#define sveor_s16_m(...) __builtin_sve_sveor_s16_m(__VA_ARGS__)
+#define sveor_u8_x(...) __builtin_sve_sveor_u8_x(__VA_ARGS__)
+#define sveor_u32_x(...) __builtin_sve_sveor_u32_x(__VA_ARGS__)
+#define sveor_u64_x(...) __builtin_sve_sveor_u64_x(__VA_ARGS__)
+#define sveor_u16_x(...) __builtin_sve_sveor_u16_x(__VA_ARGS__)
+#define sveor_s8_x(...) __builtin_sve_sveor_s8_x(__VA_ARGS__)
+#define sveor_s32_x(...) __builtin_sve_sveor_s32_x(__VA_ARGS__)
+#define sveor_s64_x(...) __builtin_sve_sveor_s64_x(__VA_ARGS__)
+#define sveor_s16_x(...) __builtin_sve_sveor_s16_x(__VA_ARGS__)
+#define sveor_u8_z(...) __builtin_sve_sveor_u8_z(__VA_ARGS__)
+#define sveor_u32_z(...) __builtin_sve_sveor_u32_z(__VA_ARGS__)
+#define sveor_u64_z(...) __builtin_sve_sveor_u64_z(__VA_ARGS__)
+#define sveor_u16_z(...) __builtin_sve_sveor_u16_z(__VA_ARGS__)
+#define sveor_s8_z(...) __builtin_sve_sveor_s8_z(__VA_ARGS__)
+#define sveor_s32_z(...) __builtin_sve_sveor_s32_z(__VA_ARGS__)
+#define sveor_s64_z(...) __builtin_sve_sveor_s64_z(__VA_ARGS__)
+#define sveor_s16_z(...) __builtin_sve_sveor_s16_z(__VA_ARGS__)
+#define sveorv_u8(...) __builtin_sve_sveorv_u8(__VA_ARGS__)
+#define sveorv_u32(...) __builtin_sve_sveorv_u32(__VA_ARGS__)
+#define sveorv_u64(...) __builtin_sve_sveorv_u64(__VA_ARGS__)
+#define sveorv_u16(...) __builtin_sve_sveorv_u16(__VA_ARGS__)
+#define sveorv_s8(...) __builtin_sve_sveorv_s8(__VA_ARGS__)
+#define sveorv_s32(...) __builtin_sve_sveorv_s32(__VA_ARGS__)
+#define sveorv_s64(...) __builtin_sve_sveorv_s64(__VA_ARGS__)
+#define sveorv_s16(...) __builtin_sve_sveorv_s16(__VA_ARGS__)
+#define svexpa_f64(...) __builtin_sve_svexpa_f64(__VA_ARGS__)
+#define svexpa_f32(...) __builtin_sve_svexpa_f32(__VA_ARGS__)
+#define svexpa_f16(...) __builtin_sve_svexpa_f16(__VA_ARGS__)
+#define svext_u8(...) __builtin_sve_svext_u8(__VA_ARGS__)
+#define svext_u32(...) __builtin_sve_svext_u32(__VA_ARGS__)
+#define svext_u64(...) __builtin_sve_svext_u64(__VA_ARGS__)
+#define svext_u16(...) __builtin_sve_svext_u16(__VA_ARGS__)
+#define svext_s8(...) __builtin_sve_svext_s8(__VA_ARGS__)
+#define svext_f64(...) __builtin_sve_svext_f64(__VA_ARGS__)
+#define svext_f32(...) __builtin_sve_svext_f32(__VA_ARGS__)
+#define svext_f16(...) __builtin_sve_svext_f16(__VA_ARGS__)
+#define svext_s32(...) __builtin_sve_svext_s32(__VA_ARGS__)
+#define svext_s64(...) __builtin_sve_svext_s64(__VA_ARGS__)
+#define svext_s16(...) __builtin_sve_svext_s16(__VA_ARGS__)
+#define svextb_s32_m(...) __builtin_sve_svextb_s32_m(__VA_ARGS__)
+#define svextb_s64_m(...) __builtin_sve_svextb_s64_m(__VA_ARGS__)
+#define svextb_s16_m(...) __builtin_sve_svextb_s16_m(__VA_ARGS__)
+#define svextb_s32_x(...) __builtin_sve_svextb_s32_x(__VA_ARGS__)
+#define svextb_s64_x(...) __builtin_sve_svextb_s64_x(__VA_ARGS__)
+#define svextb_s16_x(...) __builtin_sve_svextb_s16_x(__VA_ARGS__)
+#define svextb_s32_z(...) __builtin_sve_svextb_s32_z(__VA_ARGS__)
+#define svextb_s64_z(...) __builtin_sve_svextb_s64_z(__VA_ARGS__)
+#define svextb_s16_z(...) __builtin_sve_svextb_s16_z(__VA_ARGS__)
+#define svextb_u32_m(...) __builtin_sve_svextb_u32_m(__VA_ARGS__)
+#define svextb_u64_m(...) __builtin_sve_svextb_u64_m(__VA_ARGS__)
+#define svextb_u16_m(...) __builtin_sve_svextb_u16_m(__VA_ARGS__)
+#define svextb_u32_x(...) __builtin_sve_svextb_u32_x(__VA_ARGS__)
+#define svextb_u64_x(...) __builtin_sve_svextb_u64_x(__VA_ARGS__)
+#define svextb_u16_x(...) __builtin_sve_svextb_u16_x(__VA_ARGS__)
+#define svextb_u32_z(...) __builtin_sve_svextb_u32_z(__VA_ARGS__)
+#define svextb_u64_z(...) __builtin_sve_svextb_u64_z(__VA_ARGS__)
+#define svextb_u16_z(...) __builtin_sve_svextb_u16_z(__VA_ARGS__)
+#define svexth_s32_m(...) __builtin_sve_svexth_s32_m(__VA_ARGS__)
+#define svexth_s64_m(...) __builtin_sve_svexth_s64_m(__VA_ARGS__)
+#define svexth_s32_x(...) __builtin_sve_svexth_s32_x(__VA_ARGS__)
+#define svexth_s64_x(...) __builtin_sve_svexth_s64_x(__VA_ARGS__)
+#define svexth_s32_z(...) __builtin_sve_svexth_s32_z(__VA_ARGS__)
+#define svexth_s64_z(...) __builtin_sve_svexth_s64_z(__VA_ARGS__)
+#define svexth_u32_m(...) __builtin_sve_svexth_u32_m(__VA_ARGS__)
+#define svexth_u64_m(...) __builtin_sve_svexth_u64_m(__VA_ARGS__)
+#define svexth_u32_x(...) __builtin_sve_svexth_u32_x(__VA_ARGS__)
+#define svexth_u64_x(...) __builtin_sve_svexth_u64_x(__VA_ARGS__)
+#define svexth_u32_z(...) __builtin_sve_svexth_u32_z(__VA_ARGS__)
+#define svexth_u64_z(...) __builtin_sve_svexth_u64_z(__VA_ARGS__)
+#define svextw_s64_m(...) __builtin_sve_svextw_s64_m(__VA_ARGS__)
+#define svextw_s64_x(...) __builtin_sve_svextw_s64_x(__VA_ARGS__)
+#define svextw_s64_z(...) __builtin_sve_svextw_s64_z(__VA_ARGS__)
+#define svextw_u64_m(...) __builtin_sve_svextw_u64_m(__VA_ARGS__)
+#define svextw_u64_x(...) __builtin_sve_svextw_u64_x(__VA_ARGS__)
+#define svextw_u64_z(...) __builtin_sve_svextw_u64_z(__VA_ARGS__)
+#define svget2_u8(...) __builtin_sve_svget2_u8(__VA_ARGS__)
+#define svget2_u32(...) __builtin_sve_svget2_u32(__VA_ARGS__)
+#define svget2_u64(...) __builtin_sve_svget2_u64(__VA_ARGS__)
+#define svget2_u16(...) __builtin_sve_svget2_u16(__VA_ARGS__)
+#define svget2_s8(...) __builtin_sve_svget2_s8(__VA_ARGS__)
+#define svget2_f64(...) __builtin_sve_svget2_f64(__VA_ARGS__)
+#define svget2_f32(...) __builtin_sve_svget2_f32(__VA_ARGS__)
+#define svget2_f16(...) __builtin_sve_svget2_f16(__VA_ARGS__)
+#define svget2_s32(...) __builtin_sve_svget2_s32(__VA_ARGS__)
+#define svget2_s64(...) __builtin_sve_svget2_s64(__VA_ARGS__)
+#define svget2_s16(...) __builtin_sve_svget2_s16(__VA_ARGS__)
+#define svget3_u8(...) __builtin_sve_svget3_u8(__VA_ARGS__)
+#define svget3_u32(...) __builtin_sve_svget3_u32(__VA_ARGS__)
+#define svget3_u64(...) __builtin_sve_svget3_u64(__VA_ARGS__)
+#define svget3_u16(...) __builtin_sve_svget3_u16(__VA_ARGS__)
+#define svget3_s8(...) __builtin_sve_svget3_s8(__VA_ARGS__)
+#define svget3_f64(...) __builtin_sve_svget3_f64(__VA_ARGS__)
+#define svget3_f32(...) __builtin_sve_svget3_f32(__VA_ARGS__)
+#define svget3_f16(...) __builtin_sve_svget3_f16(__VA_ARGS__)
+#define svget3_s32(...) __builtin_sve_svget3_s32(__VA_ARGS__)
+#define svget3_s64(...) __builtin_sve_svget3_s64(__VA_ARGS__)
+#define svget3_s16(...) __builtin_sve_svget3_s16(__VA_ARGS__)
+#define svget4_u8(...) __builtin_sve_svget4_u8(__VA_ARGS__)
+#define svget4_u32(...) __builtin_sve_svget4_u32(__VA_ARGS__)
+#define svget4_u64(...) __builtin_sve_svget4_u64(__VA_ARGS__)
+#define svget4_u16(...) __builtin_sve_svget4_u16(__VA_ARGS__)
+#define svget4_s8(...) __builtin_sve_svget4_s8(__VA_ARGS__)
+#define svget4_f64(...) __builtin_sve_svget4_f64(__VA_ARGS__)
+#define svget4_f32(...) __builtin_sve_svget4_f32(__VA_ARGS__)
+#define svget4_f16(...) __builtin_sve_svget4_f16(__VA_ARGS__)
+#define svget4_s32(...) __builtin_sve_svget4_s32(__VA_ARGS__)
+#define svget4_s64(...) __builtin_sve_svget4_s64(__VA_ARGS__)
+#define svget4_s16(...) __builtin_sve_svget4_s16(__VA_ARGS__)
+#define svindex_u8(...) __builtin_sve_svindex_u8(__VA_ARGS__)
+#define svindex_u32(...) __builtin_sve_svindex_u32(__VA_ARGS__)
+#define svindex_u64(...) __builtin_sve_svindex_u64(__VA_ARGS__)
+#define svindex_u16(...) __builtin_sve_svindex_u16(__VA_ARGS__)
+#define svindex_s8(...) __builtin_sve_svindex_s8(__VA_ARGS__)
+#define svindex_s32(...) __builtin_sve_svindex_s32(__VA_ARGS__)
+#define svindex_s64(...) __builtin_sve_svindex_s64(__VA_ARGS__)
+#define svindex_s16(...) __builtin_sve_svindex_s16(__VA_ARGS__)
+#define svinsr_n_u8(...) __builtin_sve_svinsr_n_u8(__VA_ARGS__)
+#define svinsr_n_u32(...) __builtin_sve_svinsr_n_u32(__VA_ARGS__)
+#define svinsr_n_u64(...) __builtin_sve_svinsr_n_u64(__VA_ARGS__)
+#define svinsr_n_u16(...) __builtin_sve_svinsr_n_u16(__VA_ARGS__)
+#define svinsr_n_s8(...) __builtin_sve_svinsr_n_s8(__VA_ARGS__)
+#define svinsr_n_f64(...) __builtin_sve_svinsr_n_f64(__VA_ARGS__)
+#define svinsr_n_f32(...) __builtin_sve_svinsr_n_f32(__VA_ARGS__)
+#define svinsr_n_f16(...) __builtin_sve_svinsr_n_f16(__VA_ARGS__)
+#define svinsr_n_s32(...) __builtin_sve_svinsr_n_s32(__VA_ARGS__)
+#define svinsr_n_s64(...) __builtin_sve_svinsr_n_s64(__VA_ARGS__)
+#define svinsr_n_s16(...) __builtin_sve_svinsr_n_s16(__VA_ARGS__)
+#define svlasta_u8(...) __builtin_sve_svlasta_u8(__VA_ARGS__)
+#define svlasta_u32(...) __builtin_sve_svlasta_u32(__VA_ARGS__)
+#define svlasta_u64(...) __builtin_sve_svlasta_u64(__VA_ARGS__)
+#define svlasta_u16(...) __builtin_sve_svlasta_u16(__VA_ARGS__)
+#define svlasta_s8(...) __builtin_sve_svlasta_s8(__VA_ARGS__)
+#define svlasta_f64(...) __builtin_sve_svlasta_f64(__VA_ARGS__)
+#define svlasta_f32(...) __builtin_sve_svlasta_f32(__VA_ARGS__)
+#define svlasta_f16(...) __builtin_sve_svlasta_f16(__VA_ARGS__)
+#define svlasta_s32(...) __builtin_sve_svlasta_s32(__VA_ARGS__)
+#define svlasta_s64(...) __builtin_sve_svlasta_s64(__VA_ARGS__)
+#define svlasta_s16(...) __builtin_sve_svlasta_s16(__VA_ARGS__)
+#define svlastb_u8(...) __builtin_sve_svlastb_u8(__VA_ARGS__)
+#define svlastb_u32(...) __builtin_sve_svlastb_u32(__VA_ARGS__)
+#define svlastb_u64(...) __builtin_sve_svlastb_u64(__VA_ARGS__)
+#define svlastb_u16(...) __builtin_sve_svlastb_u16(__VA_ARGS__)
+#define svlastb_s8(...) __builtin_sve_svlastb_s8(__VA_ARGS__)
+#define svlastb_f64(...) __builtin_sve_svlastb_f64(__VA_ARGS__)
+#define svlastb_f32(...) __builtin_sve_svlastb_f32(__VA_ARGS__)
+#define svlastb_f16(...) __builtin_sve_svlastb_f16(__VA_ARGS__)
+#define svlastb_s32(...) __builtin_sve_svlastb_s32(__VA_ARGS__)
+#define svlastb_s64(...) __builtin_sve_svlastb_s64(__VA_ARGS__)
+#define svlastb_s16(...) __builtin_sve_svlastb_s16(__VA_ARGS__)
+#define svld1_u8(...) __builtin_sve_svld1_u8(__VA_ARGS__)
+#define svld1_u32(...) __builtin_sve_svld1_u32(__VA_ARGS__)
+#define svld1_u64(...) __builtin_sve_svld1_u64(__VA_ARGS__)
+#define svld1_u16(...) __builtin_sve_svld1_u16(__VA_ARGS__)
+#define svld1_s8(...) __builtin_sve_svld1_s8(__VA_ARGS__)
+#define svld1_f64(...) __builtin_sve_svld1_f64(__VA_ARGS__)
+#define svld1_f32(...) __builtin_sve_svld1_f32(__VA_ARGS__)
+#define svld1_f16(...) __builtin_sve_svld1_f16(__VA_ARGS__)
+#define svld1_s32(...) __builtin_sve_svld1_s32(__VA_ARGS__)
+#define svld1_s64(...) __builtin_sve_svld1_s64(__VA_ARGS__)
+#define svld1_s16(...) __builtin_sve_svld1_s16(__VA_ARGS__)
+#define svld1_gather_u32base_index_u32(...) __builtin_sve_svld1_gather_u32base_index_u32(__VA_ARGS__)
+#define svld1_gather_u64base_index_u64(...) __builtin_sve_svld1_gather_u64base_index_u64(__VA_ARGS__)
+#define svld1_gather_u64base_index_f64(...) __builtin_sve_svld1_gather_u64base_index_f64(__VA_ARGS__)
+#define svld1_gather_u32base_index_f32(...) __builtin_sve_svld1_gather_u32base_index_f32(__VA_ARGS__)
+#define svld1_gather_u32base_index_s32(...) __builtin_sve_svld1_gather_u32base_index_s32(__VA_ARGS__)
+#define svld1_gather_u64base_index_s64(...) __builtin_sve_svld1_gather_u64base_index_s64(__VA_ARGS__)
+#define svld1_gather_u32base_offset_u32(...) __builtin_sve_svld1_gather_u32base_offset_u32(__VA_ARGS__)
+#define svld1_gather_u64base_offset_u64(...) __builtin_sve_svld1_gather_u64base_offset_u64(__VA_ARGS__)
+#define svld1_gather_u64base_offset_f64(...) __builtin_sve_svld1_gather_u64base_offset_f64(__VA_ARGS__)
+#define svld1_gather_u32base_offset_f32(...) __builtin_sve_svld1_gather_u32base_offset_f32(__VA_ARGS__)
+#define svld1_gather_u32base_offset_s32(...) __builtin_sve_svld1_gather_u32base_offset_s32(__VA_ARGS__)
+#define svld1_gather_u64base_offset_s64(...) __builtin_sve_svld1_gather_u64base_offset_s64(__VA_ARGS__)
+#define svld1_gather_u32base_u32(...) __builtin_sve_svld1_gather_u32base_u32(__VA_ARGS__)
+#define svld1_gather_u64base_u64(...) __builtin_sve_svld1_gather_u64base_u64(__VA_ARGS__)
+#define svld1_gather_u64base_f64(...) __builtin_sve_svld1_gather_u64base_f64(__VA_ARGS__)
+#define svld1_gather_u32base_f32(...) __builtin_sve_svld1_gather_u32base_f32(__VA_ARGS__)
+#define svld1_gather_u32base_s32(...) __builtin_sve_svld1_gather_u32base_s32(__VA_ARGS__)
+#define svld1_gather_u64base_s64(...) __builtin_sve_svld1_gather_u64base_s64(__VA_ARGS__)
+#define svld1_gather_s32index_u32(...) __builtin_sve_svld1_gather_s32index_u32(__VA_ARGS__)
+#define svld1_gather_s32index_f32(...) __builtin_sve_svld1_gather_s32index_f32(__VA_ARGS__)
+#define svld1_gather_s32index_s32(...) __builtin_sve_svld1_gather_s32index_s32(__VA_ARGS__)
+#define svld1_gather_u32index_u32(...) __builtin_sve_svld1_gather_u32index_u32(__VA_ARGS__)
+#define svld1_gather_u32index_f32(...) __builtin_sve_svld1_gather_u32index_f32(__VA_ARGS__)
+#define svld1_gather_u32index_s32(...) __builtin_sve_svld1_gather_u32index_s32(__VA_ARGS__)
+#define svld1_gather_s64index_u64(...) __builtin_sve_svld1_gather_s64index_u64(__VA_ARGS__)
+#define svld1_gather_s64index_f64(...) __builtin_sve_svld1_gather_s64index_f64(__VA_ARGS__)
+#define svld1_gather_s64index_s64(...) __builtin_sve_svld1_gather_s64index_s64(__VA_ARGS__)
+#define svld1_gather_u64index_u64(...) __builtin_sve_svld1_gather_u64index_u64(__VA_ARGS__)
+#define svld1_gather_u64index_f64(...) __builtin_sve_svld1_gather_u64index_f64(__VA_ARGS__)
+#define svld1_gather_u64index_s64(...) __builtin_sve_svld1_gather_u64index_s64(__VA_ARGS__)
+#define svld1_gather_s32offset_u32(...) __builtin_sve_svld1_gather_s32offset_u32(__VA_ARGS__)
+#define svld1_gather_s32offset_f32(...) __builtin_sve_svld1_gather_s32offset_f32(__VA_ARGS__)
+#define svld1_gather_s32offset_s32(...) __builtin_sve_svld1_gather_s32offset_s32(__VA_ARGS__)
+#define svld1_gather_u32offset_u32(...) __builtin_sve_svld1_gather_u32offset_u32(__VA_ARGS__)
+#define svld1_gather_u32offset_f32(...) __builtin_sve_svld1_gather_u32offset_f32(__VA_ARGS__)
+#define svld1_gather_u32offset_s32(...) __builtin_sve_svld1_gather_u32offset_s32(__VA_ARGS__)
+#define svld1_gather_s64offset_u64(...) __builtin_sve_svld1_gather_s64offset_u64(__VA_ARGS__)
+#define svld1_gather_s64offset_f64(...) __builtin_sve_svld1_gather_s64offset_f64(__VA_ARGS__)
+#define svld1_gather_s64offset_s64(...) __builtin_sve_svld1_gather_s64offset_s64(__VA_ARGS__)
+#define svld1_gather_u64offset_u64(...) __builtin_sve_svld1_gather_u64offset_u64(__VA_ARGS__)
+#define svld1_gather_u64offset_f64(...) __builtin_sve_svld1_gather_u64offset_f64(__VA_ARGS__)
+#define svld1_gather_u64offset_s64(...) __builtin_sve_svld1_gather_u64offset_s64(__VA_ARGS__)
+#define svld1_vnum_u8(...) __builtin_sve_svld1_vnum_u8(__VA_ARGS__)
+#define svld1_vnum_u32(...) __builtin_sve_svld1_vnum_u32(__VA_ARGS__)
+#define svld1_vnum_u64(...) __builtin_sve_svld1_vnum_u64(__VA_ARGS__)
+#define svld1_vnum_u16(...) __builtin_sve_svld1_vnum_u16(__VA_ARGS__)
+#define svld1_vnum_s8(...) __builtin_sve_svld1_vnum_s8(__VA_ARGS__)
+#define svld1_vnum_f64(...) __builtin_sve_svld1_vnum_f64(__VA_ARGS__)
+#define svld1_vnum_f32(...) __builtin_sve_svld1_vnum_f32(__VA_ARGS__)
+#define svld1_vnum_f16(...) __builtin_sve_svld1_vnum_f16(__VA_ARGS__)
+#define svld1_vnum_s32(...) __builtin_sve_svld1_vnum_s32(__VA_ARGS__)
+#define svld1_vnum_s64(...) __builtin_sve_svld1_vnum_s64(__VA_ARGS__)
+#define svld1_vnum_s16(...) __builtin_sve_svld1_vnum_s16(__VA_ARGS__)
+#define svld1rq_u8(...) __builtin_sve_svld1rq_u8(__VA_ARGS__)
+#define svld1rq_u32(...) __builtin_sve_svld1rq_u32(__VA_ARGS__)
+#define svld1rq_u64(...) __builtin_sve_svld1rq_u64(__VA_ARGS__)
+#define svld1rq_u16(...) __builtin_sve_svld1rq_u16(__VA_ARGS__)
+#define svld1rq_s8(...) __builtin_sve_svld1rq_s8(__VA_ARGS__)
+#define svld1rq_f64(...) __builtin_sve_svld1rq_f64(__VA_ARGS__)
+#define svld1rq_f32(...) __builtin_sve_svld1rq_f32(__VA_ARGS__)
+#define svld1rq_f16(...) __builtin_sve_svld1rq_f16(__VA_ARGS__)
+#define svld1rq_s32(...) __builtin_sve_svld1rq_s32(__VA_ARGS__)
+#define svld1rq_s64(...) __builtin_sve_svld1rq_s64(__VA_ARGS__)
+#define svld1rq_s16(...) __builtin_sve_svld1rq_s16(__VA_ARGS__)
+#define svld1sb_gather_u32base_offset_u32(...) __builtin_sve_svld1sb_gather_u32base_offset_u32(__VA_ARGS__)
+#define svld1sb_gather_u64base_offset_u64(...) __builtin_sve_svld1sb_gather_u64base_offset_u64(__VA_ARGS__)
+#define svld1sb_gather_u32base_offset_s32(...) __builtin_sve_svld1sb_gather_u32base_offset_s32(__VA_ARGS__)
+#define svld1sb_gather_u64base_offset_s64(...) __builtin_sve_svld1sb_gather_u64base_offset_s64(__VA_ARGS__)
+#define svld1sb_gather_u32base_u32(...) __builtin_sve_svld1sb_gather_u32base_u32(__VA_ARGS__)
+#define svld1sb_gather_u64base_u64(...) __builtin_sve_svld1sb_gather_u64base_u64(__VA_ARGS__)
+#define svld1sb_gather_u32base_s32(...) __builtin_sve_svld1sb_gather_u32base_s32(__VA_ARGS__)
+#define svld1sb_gather_u64base_s64(...) __builtin_sve_svld1sb_gather_u64base_s64(__VA_ARGS__)
+#define svld1sb_gather_s32offset_u32(...) __builtin_sve_svld1sb_gather_s32offset_u32(__VA_ARGS__)
+#define svld1sb_gather_s32offset_s32(...) __builtin_sve_svld1sb_gather_s32offset_s32(__VA_ARGS__)
+#define svld1sb_gather_u32offset_u32(...) __builtin_sve_svld1sb_gather_u32offset_u32(__VA_ARGS__)
+#define svld1sb_gather_u32offset_s32(...) __builtin_sve_svld1sb_gather_u32offset_s32(__VA_ARGS__)
+#define svld1sb_gather_s64offset_u64(...) __builtin_sve_svld1sb_gather_s64offset_u64(__VA_ARGS__)
+#define svld1sb_gather_s64offset_s64(...) __builtin_sve_svld1sb_gather_s64offset_s64(__VA_ARGS__)
+#define svld1sb_gather_u64offset_u64(...) __builtin_sve_svld1sb_gather_u64offset_u64(__VA_ARGS__)
+#define svld1sb_gather_u64offset_s64(...) __builtin_sve_svld1sb_gather_u64offset_s64(__VA_ARGS__)
+#define svld1sb_vnum_u32(...) __builtin_sve_svld1sb_vnum_u32(__VA_ARGS__)
+#define svld1sb_vnum_u64(...) __builtin_sve_svld1sb_vnum_u64(__VA_ARGS__)
+#define svld1sb_vnum_u16(...) __builtin_sve_svld1sb_vnum_u16(__VA_ARGS__)
+#define svld1sb_vnum_s32(...) __builtin_sve_svld1sb_vnum_s32(__VA_ARGS__)
+#define svld1sb_vnum_s64(...) __builtin_sve_svld1sb_vnum_s64(__VA_ARGS__)
+#define svld1sb_vnum_s16(...) __builtin_sve_svld1sb_vnum_s16(__VA_ARGS__)
+#define svld1sb_u32(...) __builtin_sve_svld1sb_u32(__VA_ARGS__)
+#define svld1sb_u64(...) __builtin_sve_svld1sb_u64(__VA_ARGS__)
+#define svld1sb_u16(...) __builtin_sve_svld1sb_u16(__VA_ARGS__)
+#define svld1sb_s32(...) __builtin_sve_svld1sb_s32(__VA_ARGS__)
+#define svld1sb_s64(...) __builtin_sve_svld1sb_s64(__VA_ARGS__)
+#define svld1sb_s16(...) __builtin_sve_svld1sb_s16(__VA_ARGS__)
+#define svld1sh_gather_u32base_index_u32(...) __builtin_sve_svld1sh_gather_u32base_index_u32(__VA_ARGS__)
+#define svld1sh_gather_u64base_index_u64(...) __builtin_sve_svld1sh_gather_u64base_index_u64(__VA_ARGS__)
+#define svld1sh_gather_u32base_index_s32(...) __builtin_sve_svld1sh_gather_u32base_index_s32(__VA_ARGS__)
+#define svld1sh_gather_u64base_index_s64(...) __builtin_sve_svld1sh_gather_u64base_index_s64(__VA_ARGS__)
+#define svld1sh_gather_u32base_offset_u32(...) __builtin_sve_svld1sh_gather_u32base_offset_u32(__VA_ARGS__)
+#define svld1sh_gather_u64base_offset_u64(...) __builtin_sve_svld1sh_gather_u64base_offset_u64(__VA_ARGS__)
+#define svld1sh_gather_u32base_offset_s32(...) __builtin_sve_svld1sh_gather_u32base_offset_s32(__VA_ARGS__)
+#define svld1sh_gather_u64base_offset_s64(...) __builtin_sve_svld1sh_gather_u64base_offset_s64(__VA_ARGS__)
+#define svld1sh_gather_u32base_u32(...) __builtin_sve_svld1sh_gather_u32base_u32(__VA_ARGS__)
+#define svld1sh_gather_u64base_u64(...) __builtin_sve_svld1sh_gather_u64base_u64(__VA_ARGS__)
+#define svld1sh_gather_u32base_s32(...) __builtin_sve_svld1sh_gather_u32base_s32(__VA_ARGS__)
+#define svld1sh_gather_u64base_s64(...) __builtin_sve_svld1sh_gather_u64base_s64(__VA_ARGS__)
+#define svld1sh_gather_s32index_u32(...) __builtin_sve_svld1sh_gather_s32index_u32(__VA_ARGS__)
+#define svld1sh_gather_s32index_s32(...) __builtin_sve_svld1sh_gather_s32index_s32(__VA_ARGS__)
+#define svld1sh_gather_u32index_u32(...) __builtin_sve_svld1sh_gather_u32index_u32(__VA_ARGS__)
+#define svld1sh_gather_u32index_s32(...) __builtin_sve_svld1sh_gather_u32index_s32(__VA_ARGS__)
+#define svld1sh_gather_s64index_u64(...) __builtin_sve_svld1sh_gather_s64index_u64(__VA_ARGS__)
+#define svld1sh_gather_s64index_s64(...) __builtin_sve_svld1sh_gather_s64index_s64(__VA_ARGS__)
+#define svld1sh_gather_u64index_u64(...) __builtin_sve_svld1sh_gather_u64index_u64(__VA_ARGS__)
+#define svld1sh_gather_u64index_s64(...) __builtin_sve_svld1sh_gather_u64index_s64(__VA_ARGS__)
+#define svld1sh_gather_s32offset_u32(...) __builtin_sve_svld1sh_gather_s32offset_u32(__VA_ARGS__)
+#define svld1sh_gather_s32offset_s32(...) __builtin_sve_svld1sh_gather_s32offset_s32(__VA_ARGS__)
+#define svld1sh_gather_u32offset_u32(...) __builtin_sve_svld1sh_gather_u32offset_u32(__VA_ARGS__)
+#define svld1sh_gather_u32offset_s32(...) __builtin_sve_svld1sh_gather_u32offset_s32(__VA_ARGS__)
+#define svld1sh_gather_s64offset_u64(...) __builtin_sve_svld1sh_gather_s64offset_u64(__VA_ARGS__)
+#define svld1sh_gather_s64offset_s64(...) __builtin_sve_svld1sh_gather_s64offset_s64(__VA_ARGS__)
+#define svld1sh_gather_u64offset_u64(...) __builtin_sve_svld1sh_gather_u64offset_u64(__VA_ARGS__)
+#define svld1sh_gather_u64offset_s64(...) __builtin_sve_svld1sh_gather_u64offset_s64(__VA_ARGS__)
+#define svld1sh_vnum_u32(...) __builtin_sve_svld1sh_vnum_u32(__VA_ARGS__)
+#define svld1sh_vnum_u64(...) __builtin_sve_svld1sh_vnum_u64(__VA_ARGS__)
+#define svld1sh_vnum_s32(...) __builtin_sve_svld1sh_vnum_s32(__VA_ARGS__)
+#define svld1sh_vnum_s64(...) __builtin_sve_svld1sh_vnum_s64(__VA_ARGS__)
+#define svld1sh_u32(...) __builtin_sve_svld1sh_u32(__VA_ARGS__)
+#define svld1sh_u64(...) __builtin_sve_svld1sh_u64(__VA_ARGS__)
+#define svld1sh_s32(...) __builtin_sve_svld1sh_s32(__VA_ARGS__)
+#define svld1sh_s64(...) __builtin_sve_svld1sh_s64(__VA_ARGS__)
+#define svld1sw_gather_u64base_index_u64(...) __builtin_sve_svld1sw_gather_u64base_index_u64(__VA_ARGS__)
+#define svld1sw_gather_u64base_index_s64(...) __builtin_sve_svld1sw_gather_u64base_index_s64(__VA_ARGS__)
+#define svld1sw_gather_u64base_offset_u64(...) __builtin_sve_svld1sw_gather_u64base_offset_u64(__VA_ARGS__)
+#define svld1sw_gather_u64base_offset_s64(...) __builtin_sve_svld1sw_gather_u64base_offset_s64(__VA_ARGS__)
+#define svld1sw_gather_u64base_u64(...) __builtin_sve_svld1sw_gather_u64base_u64(__VA_ARGS__)
+#define svld1sw_gather_u64base_s64(...) __builtin_sve_svld1sw_gather_u64base_s64(__VA_ARGS__)
+#define svld1sw_gather_s64index_u64(...) __builtin_sve_svld1sw_gather_s64index_u64(__VA_ARGS__)
+#define svld1sw_gather_s64index_s64(...) __builtin_sve_svld1sw_gather_s64index_s64(__VA_ARGS__)
+#define svld1sw_gather_u64index_u64(...) __builtin_sve_svld1sw_gather_u64index_u64(__VA_ARGS__)
+#define svld1sw_gather_u64index_s64(...) __builtin_sve_svld1sw_gather_u64index_s64(__VA_ARGS__)
+#define svld1sw_gather_s64offset_u64(...) __builtin_sve_svld1sw_gather_s64offset_u64(__VA_ARGS__)
+#define svld1sw_gather_s64offset_s64(...) __builtin_sve_svld1sw_gather_s64offset_s64(__VA_ARGS__)
+#define svld1sw_gather_u64offset_u64(...) __builtin_sve_svld1sw_gather_u64offset_u64(__VA_ARGS__)
+#define svld1sw_gather_u64offset_s64(...) __builtin_sve_svld1sw_gather_u64offset_s64(__VA_ARGS__)
+#define svld1sw_vnum_u64(...) __builtin_sve_svld1sw_vnum_u64(__VA_ARGS__)
+#define svld1sw_vnum_s64(...) __builtin_sve_svld1sw_vnum_s64(__VA_ARGS__)
+#define svld1sw_u64(...) __builtin_sve_svld1sw_u64(__VA_ARGS__)
+#define svld1sw_s64(...) __builtin_sve_svld1sw_s64(__VA_ARGS__)
+#define svld1ub_gather_u32base_offset_u32(...) __builtin_sve_svld1ub_gather_u32base_offset_u32(__VA_ARGS__)
+#define svld1ub_gather_u64base_offset_u64(...) __builtin_sve_svld1ub_gather_u64base_offset_u64(__VA_ARGS__)
+#define svld1ub_gather_u32base_offset_s32(...) __builtin_sve_svld1ub_gather_u32base_offset_s32(__VA_ARGS__)
+#define svld1ub_gather_u64base_offset_s64(...) __builtin_sve_svld1ub_gather_u64base_offset_s64(__VA_ARGS__)
+#define svld1ub_gather_u32base_u32(...) __builtin_sve_svld1ub_gather_u32base_u32(__VA_ARGS__)
+#define svld1ub_gather_u64base_u64(...) __builtin_sve_svld1ub_gather_u64base_u64(__VA_ARGS__)
+#define svld1ub_gather_u32base_s32(...) __builtin_sve_svld1ub_gather_u32base_s32(__VA_ARGS__)
+#define svld1ub_gather_u64base_s64(...) __builtin_sve_svld1ub_gather_u64base_s64(__VA_ARGS__)
+#define svld1ub_gather_s32offset_u32(...) __builtin_sve_svld1ub_gather_s32offset_u32(__VA_ARGS__)
+#define svld1ub_gather_s32offset_s32(...) __builtin_sve_svld1ub_gather_s32offset_s32(__VA_ARGS__)
+#define svld1ub_gather_u32offset_u32(...) __builtin_sve_svld1ub_gather_u32offset_u32(__VA_ARGS__)
+#define svld1ub_gather_u32offset_s32(...) __builtin_sve_svld1ub_gather_u32offset_s32(__VA_ARGS__)
+#define svld1ub_gather_s64offset_u64(...) __builtin_sve_svld1ub_gather_s64offset_u64(__VA_ARGS__)
+#define svld1ub_gather_s64offset_s64(...) __builtin_sve_svld1ub_gather_s64offset_s64(__VA_ARGS__)
+#define svld1ub_gather_u64offset_u64(...) __builtin_sve_svld1ub_gather_u64offset_u64(__VA_ARGS__)
+#define svld1ub_gather_u64offset_s64(...) __builtin_sve_svld1ub_gather_u64offset_s64(__VA_ARGS__)
+#define svld1ub_vnum_u32(...) __builtin_sve_svld1ub_vnum_u32(__VA_ARGS__)
+#define svld1ub_vnum_u64(...) __builtin_sve_svld1ub_vnum_u64(__VA_ARGS__)
+#define svld1ub_vnum_u16(...) __builtin_sve_svld1ub_vnum_u16(__VA_ARGS__)
+#define svld1ub_vnum_s32(...) __builtin_sve_svld1ub_vnum_s32(__VA_ARGS__)
+#define svld1ub_vnum_s64(...) __builtin_sve_svld1ub_vnum_s64(__VA_ARGS__)
+#define svld1ub_vnum_s16(...) __builtin_sve_svld1ub_vnum_s16(__VA_ARGS__)
+#define svld1ub_u32(...) __builtin_sve_svld1ub_u32(__VA_ARGS__)
+#define svld1ub_u64(...) __builtin_sve_svld1ub_u64(__VA_ARGS__)
+#define svld1ub_u16(...) __builtin_sve_svld1ub_u16(__VA_ARGS__)
+#define svld1ub_s32(...) __builtin_sve_svld1ub_s32(__VA_ARGS__)
+#define svld1ub_s64(...) __builtin_sve_svld1ub_s64(__VA_ARGS__)
+#define svld1ub_s16(...) __builtin_sve_svld1ub_s16(__VA_ARGS__)
+#define svld1uh_gather_u32base_index_u32(...) __builtin_sve_svld1uh_gather_u32base_index_u32(__VA_ARGS__)
+#define svld1uh_gather_u64base_index_u64(...) __builtin_sve_svld1uh_gather_u64base_index_u64(__VA_ARGS__)
+#define svld1uh_gather_u32base_index_s32(...) __builtin_sve_svld1uh_gather_u32base_index_s32(__VA_ARGS__)
+#define svld1uh_gather_u64base_index_s64(...) __builtin_sve_svld1uh_gather_u64base_index_s64(__VA_ARGS__)
+#define svld1uh_gather_u32base_offset_u32(...) __builtin_sve_svld1uh_gather_u32base_offset_u32(__VA_ARGS__)
+#define svld1uh_gather_u64base_offset_u64(...) __builtin_sve_svld1uh_gather_u64base_offset_u64(__VA_ARGS__)
+#define svld1uh_gather_u32base_offset_s32(...) __builtin_sve_svld1uh_gather_u32base_offset_s32(__VA_ARGS__)
+#define svld1uh_gather_u64base_offset_s64(...) __builtin_sve_svld1uh_gather_u64base_offset_s64(__VA_ARGS__)
+#define svld1uh_gather_u32base_u32(...) __builtin_sve_svld1uh_gather_u32base_u32(__VA_ARGS__)
+#define svld1uh_gather_u64base_u64(...) __builtin_sve_svld1uh_gather_u64base_u64(__VA_ARGS__)
+#define svld1uh_gather_u32base_s32(...) __builtin_sve_svld1uh_gather_u32base_s32(__VA_ARGS__)
+#define svld1uh_gather_u64base_s64(...) __builtin_sve_svld1uh_gather_u64base_s64(__VA_ARGS__)
+#define svld1uh_gather_s32index_u32(...) __builtin_sve_svld1uh_gather_s32index_u32(__VA_ARGS__)
+#define svld1uh_gather_s32index_s32(...) __builtin_sve_svld1uh_gather_s32index_s32(__VA_ARGS__)
+#define svld1uh_gather_u32index_u32(...) __builtin_sve_svld1uh_gather_u32index_u32(__VA_ARGS__)
+#define svld1uh_gather_u32index_s32(...) __builtin_sve_svld1uh_gather_u32index_s32(__VA_ARGS__)
+#define svld1uh_gather_s64index_u64(...) __builtin_sve_svld1uh_gather_s64index_u64(__VA_ARGS__)
+#define svld1uh_gather_s64index_s64(...) __builtin_sve_svld1uh_gather_s64index_s64(__VA_ARGS__)
+#define svld1uh_gather_u64index_u64(...) __builtin_sve_svld1uh_gather_u64index_u64(__VA_ARGS__)
+#define svld1uh_gather_u64index_s64(...) __builtin_sve_svld1uh_gather_u64index_s64(__VA_ARGS__)
+#define svld1uh_gather_s32offset_u32(...) __builtin_sve_svld1uh_gather_s32offset_u32(__VA_ARGS__)
+#define svld1uh_gather_s32offset_s32(...) __builtin_sve_svld1uh_gather_s32offset_s32(__VA_ARGS__)
+#define svld1uh_gather_u32offset_u32(...) __builtin_sve_svld1uh_gather_u32offset_u32(__VA_ARGS__)
+#define svld1uh_gather_u32offset_s32(...) __builtin_sve_svld1uh_gather_u32offset_s32(__VA_ARGS__)
+#define svld1uh_gather_s64offset_u64(...) __builtin_sve_svld1uh_gather_s64offset_u64(__VA_ARGS__)
+#define svld1uh_gather_s64offset_s64(...) __builtin_sve_svld1uh_gather_s64offset_s64(__VA_ARGS__)
+#define svld1uh_gather_u64offset_u64(...) __builtin_sve_svld1uh_gather_u64offset_u64(__VA_ARGS__)
+#define svld1uh_gather_u64offset_s64(...) __builtin_sve_svld1uh_gather_u64offset_s64(__VA_ARGS__)
+#define svld1uh_vnum_u32(...) __builtin_sve_svld1uh_vnum_u32(__VA_ARGS__)
+#define svld1uh_vnum_u64(...) __builtin_sve_svld1uh_vnum_u64(__VA_ARGS__)
+#define svld1uh_vnum_s32(...) __builtin_sve_svld1uh_vnum_s32(__VA_ARGS__)
+#define svld1uh_vnum_s64(...) __builtin_sve_svld1uh_vnum_s64(__VA_ARGS__)
+#define svld1uh_u32(...) __builtin_sve_svld1uh_u32(__VA_ARGS__)
+#define svld1uh_u64(...) __builtin_sve_svld1uh_u64(__VA_ARGS__)
+#define svld1uh_s32(...) __builtin_sve_svld1uh_s32(__VA_ARGS__)
+#define svld1uh_s64(...) __builtin_sve_svld1uh_s64(__VA_ARGS__)
+#define svld1uw_gather_u64base_index_u64(...) __builtin_sve_svld1uw_gather_u64base_index_u64(__VA_ARGS__)
+#define svld1uw_gather_u64base_index_s64(...) __builtin_sve_svld1uw_gather_u64base_index_s64(__VA_ARGS__)
+#define svld1uw_gather_u64base_offset_u64(...) __builtin_sve_svld1uw_gather_u64base_offset_u64(__VA_ARGS__)
+#define svld1uw_gather_u64base_offset_s64(...) __builtin_sve_svld1uw_gather_u64base_offset_s64(__VA_ARGS__)
+#define svld1uw_gather_u64base_u64(...) __builtin_sve_svld1uw_gather_u64base_u64(__VA_ARGS__)
+#define svld1uw_gather_u64base_s64(...) __builtin_sve_svld1uw_gather_u64base_s64(__VA_ARGS__)
+#define svld1uw_gather_s64index_u64(...) __builtin_sve_svld1uw_gather_s64index_u64(__VA_ARGS__)
+#define svld1uw_gather_s64index_s64(...) __builtin_sve_svld1uw_gather_s64index_s64(__VA_ARGS__)
+#define svld1uw_gather_u64index_u64(...) __builtin_sve_svld1uw_gather_u64index_u64(__VA_ARGS__)
+#define svld1uw_gather_u64index_s64(...) __builtin_sve_svld1uw_gather_u64index_s64(__VA_ARGS__)
+#define svld1uw_gather_s64offset_u64(...) __builtin_sve_svld1uw_gather_s64offset_u64(__VA_ARGS__)
+#define svld1uw_gather_s64offset_s64(...) __builtin_sve_svld1uw_gather_s64offset_s64(__VA_ARGS__)
+#define svld1uw_gather_u64offset_u64(...) __builtin_sve_svld1uw_gather_u64offset_u64(__VA_ARGS__)
+#define svld1uw_gather_u64offset_s64(...) __builtin_sve_svld1uw_gather_u64offset_s64(__VA_ARGS__)
+#define svld1uw_vnum_u64(...) __builtin_sve_svld1uw_vnum_u64(__VA_ARGS__)
+#define svld1uw_vnum_s64(...) __builtin_sve_svld1uw_vnum_s64(__VA_ARGS__)
+#define svld1uw_u64(...) __builtin_sve_svld1uw_u64(__VA_ARGS__)
+#define svld1uw_s64(...) __builtin_sve_svld1uw_s64(__VA_ARGS__)
+#define svld2_u8(...) __builtin_sve_svld2_u8(__VA_ARGS__)
+#define svld2_u32(...) __builtin_sve_svld2_u32(__VA_ARGS__)
+#define svld2_u64(...) __builtin_sve_svld2_u64(__VA_ARGS__)
+#define svld2_u16(...) __builtin_sve_svld2_u16(__VA_ARGS__)
+#define svld2_s8(...) __builtin_sve_svld2_s8(__VA_ARGS__)
+#define svld2_f64(...) __builtin_sve_svld2_f64(__VA_ARGS__)
+#define svld2_f32(...) __builtin_sve_svld2_f32(__VA_ARGS__)
+#define svld2_f16(...) __builtin_sve_svld2_f16(__VA_ARGS__)
+#define svld2_s32(...) __builtin_sve_svld2_s32(__VA_ARGS__)
+#define svld2_s64(...) __builtin_sve_svld2_s64(__VA_ARGS__)
+#define svld2_s16(...) __builtin_sve_svld2_s16(__VA_ARGS__)
+#define svld2_vnum_u8(...) __builtin_sve_svld2_vnum_u8(__VA_ARGS__)
+#define svld2_vnum_u32(...) __builtin_sve_svld2_vnum_u32(__VA_ARGS__)
+#define svld2_vnum_u64(...) __builtin_sve_svld2_vnum_u64(__VA_ARGS__)
+#define svld2_vnum_u16(...) __builtin_sve_svld2_vnum_u16(__VA_ARGS__)
+#define svld2_vnum_s8(...) __builtin_sve_svld2_vnum_s8(__VA_ARGS__)
+#define svld2_vnum_f64(...) __builtin_sve_svld2_vnum_f64(__VA_ARGS__)
+#define svld2_vnum_f32(...) __builtin_sve_svld2_vnum_f32(__VA_ARGS__)
+#define svld2_vnum_f16(...) __builtin_sve_svld2_vnum_f16(__VA_ARGS__)
+#define svld2_vnum_s32(...) __builtin_sve_svld2_vnum_s32(__VA_ARGS__)
+#define svld2_vnum_s64(...) __builtin_sve_svld2_vnum_s64(__VA_ARGS__)
+#define svld2_vnum_s16(...) __builtin_sve_svld2_vnum_s16(__VA_ARGS__)
+#define svld3_u8(...) __builtin_sve_svld3_u8(__VA_ARGS__)
+#define svld3_u32(...) __builtin_sve_svld3_u32(__VA_ARGS__)
+#define svld3_u64(...) __builtin_sve_svld3_u64(__VA_ARGS__)
+#define svld3_u16(...) __builtin_sve_svld3_u16(__VA_ARGS__)
+#define svld3_s8(...) __builtin_sve_svld3_s8(__VA_ARGS__)
+#define svld3_f64(...) __builtin_sve_svld3_f64(__VA_ARGS__)
+#define svld3_f32(...) __builtin_sve_svld3_f32(__VA_ARGS__)
+#define svld3_f16(...) __builtin_sve_svld3_f16(__VA_ARGS__)
+#define svld3_s32(...) __builtin_sve_svld3_s32(__VA_ARGS__)
+#define svld3_s64(...) __builtin_sve_svld3_s64(__VA_ARGS__)
+#define svld3_s16(...) __builtin_sve_svld3_s16(__VA_ARGS__)
+#define svld3_vnum_u8(...) __builtin_sve_svld3_vnum_u8(__VA_ARGS__)
+#define svld3_vnum_u32(...) __builtin_sve_svld3_vnum_u32(__VA_ARGS__)
+#define svld3_vnum_u64(...) __builtin_sve_svld3_vnum_u64(__VA_ARGS__)
+#define svld3_vnum_u16(...) __builtin_sve_svld3_vnum_u16(__VA_ARGS__)
+#define svld3_vnum_s8(...) __builtin_sve_svld3_vnum_s8(__VA_ARGS__)
+#define svld3_vnum_f64(...) __builtin_sve_svld3_vnum_f64(__VA_ARGS__)
+#define svld3_vnum_f32(...) __builtin_sve_svld3_vnum_f32(__VA_ARGS__)
+#define svld3_vnum_f16(...) __builtin_sve_svld3_vnum_f16(__VA_ARGS__)
+#define svld3_vnum_s32(...) __builtin_sve_svld3_vnum_s32(__VA_ARGS__)
+#define svld3_vnum_s64(...) __builtin_sve_svld3_vnum_s64(__VA_ARGS__)
+#define svld3_vnum_s16(...) __builtin_sve_svld3_vnum_s16(__VA_ARGS__)
+#define svld4_u8(...) __builtin_sve_svld4_u8(__VA_ARGS__)
+#define svld4_u32(...) __builtin_sve_svld4_u32(__VA_ARGS__)
+#define svld4_u64(...) __builtin_sve_svld4_u64(__VA_ARGS__)
+#define svld4_u16(...) __builtin_sve_svld4_u16(__VA_ARGS__)
+#define svld4_s8(...) __builtin_sve_svld4_s8(__VA_ARGS__)
+#define svld4_f64(...) __builtin_sve_svld4_f64(__VA_ARGS__)
+#define svld4_f32(...) __builtin_sve_svld4_f32(__VA_ARGS__)
+#define svld4_f16(...) __builtin_sve_svld4_f16(__VA_ARGS__)
+#define svld4_s32(...) __builtin_sve_svld4_s32(__VA_ARGS__)
+#define svld4_s64(...) __builtin_sve_svld4_s64(__VA_ARGS__)
+#define svld4_s16(...) __builtin_sve_svld4_s16(__VA_ARGS__)
+#define svld4_vnum_u8(...) __builtin_sve_svld4_vnum_u8(__VA_ARGS__)
+#define svld4_vnum_u32(...) __builtin_sve_svld4_vnum_u32(__VA_ARGS__)
+#define svld4_vnum_u64(...) __builtin_sve_svld4_vnum_u64(__VA_ARGS__)
+#define svld4_vnum_u16(...) __builtin_sve_svld4_vnum_u16(__VA_ARGS__)
+#define svld4_vnum_s8(...) __builtin_sve_svld4_vnum_s8(__VA_ARGS__)
+#define svld4_vnum_f64(...) __builtin_sve_svld4_vnum_f64(__VA_ARGS__)
+#define svld4_vnum_f32(...) __builtin_sve_svld4_vnum_f32(__VA_ARGS__)
+#define svld4_vnum_f16(...) __builtin_sve_svld4_vnum_f16(__VA_ARGS__)
+#define svld4_vnum_s32(...) __builtin_sve_svld4_vnum_s32(__VA_ARGS__)
+#define svld4_vnum_s64(...) __builtin_sve_svld4_vnum_s64(__VA_ARGS__)
+#define svld4_vnum_s16(...) __builtin_sve_svld4_vnum_s16(__VA_ARGS__)
+#define svldff1_u8(...) __builtin_sve_svldff1_u8(__VA_ARGS__)
+#define svldff1_u32(...) __builtin_sve_svldff1_u32(__VA_ARGS__)
+#define svldff1_u64(...) __builtin_sve_svldff1_u64(__VA_ARGS__)
+#define svldff1_u16(...) __builtin_sve_svldff1_u16(__VA_ARGS__)
+#define svldff1_s8(...) __builtin_sve_svldff1_s8(__VA_ARGS__)
+#define svldff1_f64(...) __builtin_sve_svldff1_f64(__VA_ARGS__)
+#define svldff1_f32(...) __builtin_sve_svldff1_f32(__VA_ARGS__)
+#define svldff1_f16(...) __builtin_sve_svldff1_f16(__VA_ARGS__)
+#define svldff1_s32(...) __builtin_sve_svldff1_s32(__VA_ARGS__)
+#define svldff1_s64(...) __builtin_sve_svldff1_s64(__VA_ARGS__)
+#define svldff1_s16(...) __builtin_sve_svldff1_s16(__VA_ARGS__)
+#define svldff1_gather_u32base_index_u32(...) __builtin_sve_svldff1_gather_u32base_index_u32(__VA_ARGS__)
+#define svldff1_gather_u64base_index_u64(...) __builtin_sve_svldff1_gather_u64base_index_u64(__VA_ARGS__)
+#define svldff1_gather_u64base_index_f64(...) __builtin_sve_svldff1_gather_u64base_index_f64(__VA_ARGS__)
+#define svldff1_gather_u32base_index_f32(...) __builtin_sve_svldff1_gather_u32base_index_f32(__VA_ARGS__)
+#define svldff1_gather_u32base_index_s32(...) __builtin_sve_svldff1_gather_u32base_index_s32(__VA_ARGS__)
+#define svldff1_gather_u64base_index_s64(...) __builtin_sve_svldff1_gather_u64base_index_s64(__VA_ARGS__)
+#define svldff1_gather_u32base_offset_u32(...) __builtin_sve_svldff1_gather_u32base_offset_u32(__VA_ARGS__)
+#define svldff1_gather_u64base_offset_u64(...) __builtin_sve_svldff1_gather_u64base_offset_u64(__VA_ARGS__)
+#define svldff1_gather_u64base_offset_f64(...) __builtin_sve_svldff1_gather_u64base_offset_f64(__VA_ARGS__)
+#define svldff1_gather_u32base_offset_f32(...) __builtin_sve_svldff1_gather_u32base_offset_f32(__VA_ARGS__)
+#define svldff1_gather_u32base_offset_s32(...) __builtin_sve_svldff1_gather_u32base_offset_s32(__VA_ARGS__)
+#define svldff1_gather_u64base_offset_s64(...) __builtin_sve_svldff1_gather_u64base_offset_s64(__VA_ARGS__)
+#define svldff1_gather_u32base_u32(...) __builtin_sve_svldff1_gather_u32base_u32(__VA_ARGS__)
+#define svldff1_gather_u64base_u64(...) __builtin_sve_svldff1_gather_u64base_u64(__VA_ARGS__)
+#define svldff1_gather_u64base_f64(...) __builtin_sve_svldff1_gather_u64base_f64(__VA_ARGS__)
+#define svldff1_gather_u32base_f32(...) __builtin_sve_svldff1_gather_u32base_f32(__VA_ARGS__)
+#define svldff1_gather_u32base_s32(...) __builtin_sve_svldff1_gather_u32base_s32(__VA_ARGS__)
+#define svldff1_gather_u64base_s64(...) __builtin_sve_svldff1_gather_u64base_s64(__VA_ARGS__)
+#define svldff1_gather_s32index_u32(...) __builtin_sve_svldff1_gather_s32index_u32(__VA_ARGS__)
+#define svldff1_gather_s32index_f32(...) __builtin_sve_svldff1_gather_s32index_f32(__VA_ARGS__)
+#define svldff1_gather_s32index_s32(...) __builtin_sve_svldff1_gather_s32index_s32(__VA_ARGS__)
+#define svldff1_gather_u32index_u32(...) __builtin_sve_svldff1_gather_u32index_u32(__VA_ARGS__)
+#define svldff1_gather_u32index_f32(...) __builtin_sve_svldff1_gather_u32index_f32(__VA_ARGS__)
+#define svldff1_gather_u32index_s32(...) __builtin_sve_svldff1_gather_u32index_s32(__VA_ARGS__)
+#define svldff1_gather_s64index_u64(...) __builtin_sve_svldff1_gather_s64index_u64(__VA_ARGS__)
+#define svldff1_gather_s64index_f64(...) __builtin_sve_svldff1_gather_s64index_f64(__VA_ARGS__)
+#define svldff1_gather_s64index_s64(...) __builtin_sve_svldff1_gather_s64index_s64(__VA_ARGS__)
+#define svldff1_gather_u64index_u64(...) __builtin_sve_svldff1_gather_u64index_u64(__VA_ARGS__)
+#define svldff1_gather_u64index_f64(...) __builtin_sve_svldff1_gather_u64index_f64(__VA_ARGS__)
+#define svldff1_gather_u64index_s64(...) __builtin_sve_svldff1_gather_u64index_s64(__VA_ARGS__)
+#define svldff1_gather_s32offset_u32(...) __builtin_sve_svldff1_gather_s32offset_u32(__VA_ARGS__)
+#define svldff1_gather_s32offset_f32(...) __builtin_sve_svldff1_gather_s32offset_f32(__VA_ARGS__)
+#define svldff1_gather_s32offset_s32(...) __builtin_sve_svldff1_gather_s32offset_s32(__VA_ARGS__)
+#define svldff1_gather_u32offset_u32(...) __builtin_sve_svldff1_gather_u32offset_u32(__VA_ARGS__)
+#define svldff1_gather_u32offset_f32(...) __builtin_sve_svldff1_gather_u32offset_f32(__VA_ARGS__)
+#define svldff1_gather_u32offset_s32(...) __builtin_sve_svldff1_gather_u32offset_s32(__VA_ARGS__)
+#define svldff1_gather_s64offset_u64(...) __builtin_sve_svldff1_gather_s64offset_u64(__VA_ARGS__)
+#define svldff1_gather_s64offset_f64(...) __builtin_sve_svldff1_gather_s64offset_f64(__VA_ARGS__)
+#define svldff1_gather_s64offset_s64(...) __builtin_sve_svldff1_gather_s64offset_s64(__VA_ARGS__)
+#define svldff1_gather_u64offset_u64(...) __builtin_sve_svldff1_gather_u64offset_u64(__VA_ARGS__)
+#define svldff1_gather_u64offset_f64(...) __builtin_sve_svldff1_gather_u64offset_f64(__VA_ARGS__)
+#define svldff1_gather_u64offset_s64(...) __builtin_sve_svldff1_gather_u64offset_s64(__VA_ARGS__)
+#define svldff1_vnum_u8(...) __builtin_sve_svldff1_vnum_u8(__VA_ARGS__)
+#define svldff1_vnum_u32(...) __builtin_sve_svldff1_vnum_u32(__VA_ARGS__)
+#define svldff1_vnum_u64(...) __builtin_sve_svldff1_vnum_u64(__VA_ARGS__)
+#define svldff1_vnum_u16(...) __builtin_sve_svldff1_vnum_u16(__VA_ARGS__)
+#define svldff1_vnum_s8(...) __builtin_sve_svldff1_vnum_s8(__VA_ARGS__)
+#define svldff1_vnum_f64(...) __builtin_sve_svldff1_vnum_f64(__VA_ARGS__)
+#define svldff1_vnum_f32(...) __builtin_sve_svldff1_vnum_f32(__VA_ARGS__)
+#define svldff1_vnum_f16(...) __builtin_sve_svldff1_vnum_f16(__VA_ARGS__)
+#define svldff1_vnum_s32(...) __builtin_sve_svldff1_vnum_s32(__VA_ARGS__)
+#define svldff1_vnum_s64(...) __builtin_sve_svldff1_vnum_s64(__VA_ARGS__)
+#define svldff1_vnum_s16(...) __builtin_sve_svldff1_vnum_s16(__VA_ARGS__)
+#define svldff1sb_gather_u32base_offset_u32(...) __builtin_sve_svldff1sb_gather_u32base_offset_u32(__VA_ARGS__)
+#define svldff1sb_gather_u64base_offset_u64(...) __builtin_sve_svldff1sb_gather_u64base_offset_u64(__VA_ARGS__)
+#define svldff1sb_gather_u32base_offset_s32(...) __builtin_sve_svldff1sb_gather_u32base_offset_s32(__VA_ARGS__)
+#define svldff1sb_gather_u64base_offset_s64(...) __builtin_sve_svldff1sb_gather_u64base_offset_s64(__VA_ARGS__)
+#define svldff1sb_gather_u32base_u32(...) __builtin_sve_svldff1sb_gather_u32base_u32(__VA_ARGS__)
+#define svldff1sb_gather_u64base_u64(...) __builtin_sve_svldff1sb_gather_u64base_u64(__VA_ARGS__)
+#define svldff1sb_gather_u32base_s32(...) __builtin_sve_svldff1sb_gather_u32base_s32(__VA_ARGS__)
+#define svldff1sb_gather_u64base_s64(...) __builtin_sve_svldff1sb_gather_u64base_s64(__VA_ARGS__)
+#define svldff1sb_gather_s32offset_u32(...) __builtin_sve_svldff1sb_gather_s32offset_u32(__VA_ARGS__)
+#define svldff1sb_gather_s32offset_s32(...) __builtin_sve_svldff1sb_gather_s32offset_s32(__VA_ARGS__)
+#define svldff1sb_gather_u32offset_u32(...) __builtin_sve_svldff1sb_gather_u32offset_u32(__VA_ARGS__)
+#define svldff1sb_gather_u32offset_s32(...) __builtin_sve_svldff1sb_gather_u32offset_s32(__VA_ARGS__)
+#define svldff1sb_gather_s64offset_u64(...) __builtin_sve_svldff1sb_gather_s64offset_u64(__VA_ARGS__)
+#define svldff1sb_gather_s64offset_s64(...) __builtin_sve_svldff1sb_gather_s64offset_s64(__VA_ARGS__)
+#define svldff1sb_gather_u64offset_u64(...) __builtin_sve_svldff1sb_gather_u64offset_u64(__VA_ARGS__)
+#define svldff1sb_gather_u64offset_s64(...) __builtin_sve_svldff1sb_gather_u64offset_s64(__VA_ARGS__)
+#define svldff1sb_vnum_u32(...) __builtin_sve_svldff1sb_vnum_u32(__VA_ARGS__)
+#define svldff1sb_vnum_u64(...) __builtin_sve_svldff1sb_vnum_u64(__VA_ARGS__)
+#define svldff1sb_vnum_u16(...) __builtin_sve_svldff1sb_vnum_u16(__VA_ARGS__)
+#define svldff1sb_vnum_s32(...) __builtin_sve_svldff1sb_vnum_s32(__VA_ARGS__)
+#define svldff1sb_vnum_s64(...) __builtin_sve_svldff1sb_vnum_s64(__VA_ARGS__)
+#define svldff1sb_vnum_s16(...) __builtin_sve_svldff1sb_vnum_s16(__VA_ARGS__)
+#define svldff1sb_u32(...) __builtin_sve_svldff1sb_u32(__VA_ARGS__)
+#define svldff1sb_u64(...) __builtin_sve_svldff1sb_u64(__VA_ARGS__)
+#define svldff1sb_u16(...) __builtin_sve_svldff1sb_u16(__VA_ARGS__)
+#define svldff1sb_s32(...) __builtin_sve_svldff1sb_s32(__VA_ARGS__)
+#define svldff1sb_s64(...) __builtin_sve_svldff1sb_s64(__VA_ARGS__)
+#define svldff1sb_s16(...) __builtin_sve_svldff1sb_s16(__VA_ARGS__)
+#define svldff1sh_gather_u32base_index_u32(...) __builtin_sve_svldff1sh_gather_u32base_index_u32(__VA_ARGS__)
+#define svldff1sh_gather_u64base_index_u64(...) __builtin_sve_svldff1sh_gather_u64base_index_u64(__VA_ARGS__)
+#define svldff1sh_gather_u32base_index_s32(...) __builtin_sve_svldff1sh_gather_u32base_index_s32(__VA_ARGS__)
+#define svldff1sh_gather_u64base_index_s64(...) __builtin_sve_svldff1sh_gather_u64base_index_s64(__VA_ARGS__)
+#define svldff1sh_gather_u32base_offset_u32(...) __builtin_sve_svldff1sh_gather_u32base_offset_u32(__VA_ARGS__)
+#define svldff1sh_gather_u64base_offset_u64(...) __builtin_sve_svldff1sh_gather_u64base_offset_u64(__VA_ARGS__)
+#define svldff1sh_gather_u32base_offset_s32(...) __builtin_sve_svldff1sh_gather_u32base_offset_s32(__VA_ARGS__)
+#define svldff1sh_gather_u64base_offset_s64(...) __builtin_sve_svldff1sh_gather_u64base_offset_s64(__VA_ARGS__)
+#define svldff1sh_gather_u32base_u32(...) __builtin_sve_svldff1sh_gather_u32base_u32(__VA_ARGS__)
+#define svldff1sh_gather_u64base_u64(...) __builtin_sve_svldff1sh_gather_u64base_u64(__VA_ARGS__)
+#define svldff1sh_gather_u32base_s32(...) __builtin_sve_svldff1sh_gather_u32base_s32(__VA_ARGS__)
+#define svldff1sh_gather_u64base_s64(...) __builtin_sve_svldff1sh_gather_u64base_s64(__VA_ARGS__)
+#define svldff1sh_gather_s32index_u32(...) __builtin_sve_svldff1sh_gather_s32index_u32(__VA_ARGS__)
+#define svldff1sh_gather_s32index_s32(...) __builtin_sve_svldff1sh_gather_s32index_s32(__VA_ARGS__)
+#define svldff1sh_gather_u32index_u32(...) __builtin_sve_svldff1sh_gather_u32index_u32(__VA_ARGS__)
+#define svldff1sh_gather_u32index_s32(...) __builtin_sve_svldff1sh_gather_u32index_s32(__VA_ARGS__)
+#define svldff1sh_gather_s64index_u64(...) __builtin_sve_svldff1sh_gather_s64index_u64(__VA_ARGS__)
+#define svldff1sh_gather_s64index_s64(...) __builtin_sve_svldff1sh_gather_s64index_s64(__VA_ARGS__)
+#define svldff1sh_gather_u64index_u64(...) __builtin_sve_svldff1sh_gather_u64index_u64(__VA_ARGS__)
+#define svldff1sh_gather_u64index_s64(...) __builtin_sve_svldff1sh_gather_u64index_s64(__VA_ARGS__)
+#define svldff1sh_gather_s32offset_u32(...) __builtin_sve_svldff1sh_gather_s32offset_u32(__VA_ARGS__)
+#define svldff1sh_gather_s32offset_s32(...) __builtin_sve_svldff1sh_gather_s32offset_s32(__VA_ARGS__)
+#define svldff1sh_gather_u32offset_u32(...) __builtin_sve_svldff1sh_gather_u32offset_u32(__VA_ARGS__)
+#define svldff1sh_gather_u32offset_s32(...) __builtin_sve_svldff1sh_gather_u32offset_s32(__VA_ARGS__)
+#define svldff1sh_gather_s64offset_u64(...) __builtin_sve_svldff1sh_gather_s64offset_u64(__VA_ARGS__)
+#define svldff1sh_gather_s64offset_s64(...) __builtin_sve_svldff1sh_gather_s64offset_s64(__VA_ARGS__)
+#define svldff1sh_gather_u64offset_u64(...) __builtin_sve_svldff1sh_gather_u64offset_u64(__VA_ARGS__)
+#define svldff1sh_gather_u64offset_s64(...) __builtin_sve_svldff1sh_gather_u64offset_s64(__VA_ARGS__)
+#define svldff1sh_vnum_u32(...) __builtin_sve_svldff1sh_vnum_u32(__VA_ARGS__)
+#define svldff1sh_vnum_u64(...) __builtin_sve_svldff1sh_vnum_u64(__VA_ARGS__)
+#define svldff1sh_vnum_s32(...) __builtin_sve_svldff1sh_vnum_s32(__VA_ARGS__)
+#define svldff1sh_vnum_s64(...) __builtin_sve_svldff1sh_vnum_s64(__VA_ARGS__)
+#define svldff1sh_u32(...) __builtin_sve_svldff1sh_u32(__VA_ARGS__)
+#define svldff1sh_u64(...) __builtin_sve_svldff1sh_u64(__VA_ARGS__)
+#define svldff1sh_s32(...) __builtin_sve_svldff1sh_s32(__VA_ARGS__)
+#define svldff1sh_s64(...) __builtin_sve_svldff1sh_s64(__VA_ARGS__)
+#define svldff1sw_gather_u64base_index_u64(...) __builtin_sve_svldff1sw_gather_u64base_index_u64(__VA_ARGS__)
+#define svldff1sw_gather_u64base_index_s64(...) __builtin_sve_svldff1sw_gather_u64base_index_s64(__VA_ARGS__)
+#define svldff1sw_gather_u64base_offset_u64(...) __builtin_sve_svldff1sw_gather_u64base_offset_u64(__VA_ARGS__)
+#define svldff1sw_gather_u64base_offset_s64(...) __builtin_sve_svldff1sw_gather_u64base_offset_s64(__VA_ARGS__)
+#define svldff1sw_gather_u64base_u64(...) __builtin_sve_svldff1sw_gather_u64base_u64(__VA_ARGS__)
+#define svldff1sw_gather_u64base_s64(...) __builtin_sve_svldff1sw_gather_u64base_s64(__VA_ARGS__)
+#define svldff1sw_gather_s64index_u64(...) __builtin_sve_svldff1sw_gather_s64index_u64(__VA_ARGS__)
+#define svldff1sw_gather_s64index_s64(...) __builtin_sve_svldff1sw_gather_s64index_s64(__VA_ARGS__)
+#define svldff1sw_gather_u64index_u64(...) __builtin_sve_svldff1sw_gather_u64index_u64(__VA_ARGS__)
+#define svldff1sw_gather_u64index_s64(...) __builtin_sve_svldff1sw_gather_u64index_s64(__VA_ARGS__)
+#define svldff1sw_gather_s64offset_u64(...) __builtin_sve_svldff1sw_gather_s64offset_u64(__VA_ARGS__)
+#define svldff1sw_gather_s64offset_s64(...) __builtin_sve_svldff1sw_gather_s64offset_s64(__VA_ARGS__)
+#define svldff1sw_gather_u64offset_u64(...) __builtin_sve_svldff1sw_gather_u64offset_u64(__VA_ARGS__)
+#define svldff1sw_gather_u64offset_s64(...) __builtin_sve_svldff1sw_gather_u64offset_s64(__VA_ARGS__)
+#define svldff1sw_vnum_u64(...) __builtin_sve_svldff1sw_vnum_u64(__VA_ARGS__)
+#define svldff1sw_vnum_s64(...) __builtin_sve_svldff1sw_vnum_s64(__VA_ARGS__)
+#define svldff1sw_u64(...) __builtin_sve_svldff1sw_u64(__VA_ARGS__)
+#define svldff1sw_s64(...) __builtin_sve_svldff1sw_s64(__VA_ARGS__)
+#define svldff1ub_gather_u32base_offset_u32(...) __builtin_sve_svldff1ub_gather_u32base_offset_u32(__VA_ARGS__)
+#define svldff1ub_gather_u64base_offset_u64(...) __builtin_sve_svldff1ub_gather_u64base_offset_u64(__VA_ARGS__)
+#define svldff1ub_gather_u32base_offset_s32(...) __builtin_sve_svldff1ub_gather_u32base_offset_s32(__VA_ARGS__)
+#define svldff1ub_gather_u64base_offset_s64(...) __builtin_sve_svldff1ub_gather_u64base_offset_s64(__VA_ARGS__)
+#define svldff1ub_gather_u32base_u32(...) __builtin_sve_svldff1ub_gather_u32base_u32(__VA_ARGS__)
+#define svldff1ub_gather_u64base_u64(...) __builtin_sve_svldff1ub_gather_u64base_u64(__VA_ARGS__)
+#define svldff1ub_gather_u32base_s32(...) __builtin_sve_svldff1ub_gather_u32base_s32(__VA_ARGS__)
+#define svldff1ub_gather_u64base_s64(...) __builtin_sve_svldff1ub_gather_u64base_s64(__VA_ARGS__)
+#define svldff1ub_gather_s32offset_u32(...) __builtin_sve_svldff1ub_gather_s32offset_u32(__VA_ARGS__)
+#define svldff1ub_gather_s32offset_s32(...) __builtin_sve_svldff1ub_gather_s32offset_s32(__VA_ARGS__)
+#define svldff1ub_gather_u32offset_u32(...) __builtin_sve_svldff1ub_gather_u32offset_u32(__VA_ARGS__)
+#define svldff1ub_gather_u32offset_s32(...) __builtin_sve_svldff1ub_gather_u32offset_s32(__VA_ARGS__)
+#define svldff1ub_gather_s64offset_u64(...) __builtin_sve_svldff1ub_gather_s64offset_u64(__VA_ARGS__)
+#define svldff1ub_gather_s64offset_s64(...) __builtin_sve_svldff1ub_gather_s64offset_s64(__VA_ARGS__)
+#define svldff1ub_gather_u64offset_u64(...) __builtin_sve_svldff1ub_gather_u64offset_u64(__VA_ARGS__)
+#define svldff1ub_gather_u64offset_s64(...) __builtin_sve_svldff1ub_gather_u64offset_s64(__VA_ARGS__)
+#define svldff1ub_vnum_u32(...) __builtin_sve_svldff1ub_vnum_u32(__VA_ARGS__)
+#define svldff1ub_vnum_u64(...) __builtin_sve_svldff1ub_vnum_u64(__VA_ARGS__)
+#define svldff1ub_vnum_u16(...) __builtin_sve_svldff1ub_vnum_u16(__VA_ARGS__)
+#define svldff1ub_vnum_s32(...) __builtin_sve_svldff1ub_vnum_s32(__VA_ARGS__)
+#define svldff1ub_vnum_s64(...) __builtin_sve_svldff1ub_vnum_s64(__VA_ARGS__)
+#define svldff1ub_vnum_s16(...) __builtin_sve_svldff1ub_vnum_s16(__VA_ARGS__)
+#define svldff1ub_u32(...) __builtin_sve_svldff1ub_u32(__VA_ARGS__)
+#define svldff1ub_u64(...) __builtin_sve_svldff1ub_u64(__VA_ARGS__)
+#define svldff1ub_u16(...) __builtin_sve_svldff1ub_u16(__VA_ARGS__)
+#define svldff1ub_s32(...) __builtin_sve_svldff1ub_s32(__VA_ARGS__)
+#define svldff1ub_s64(...) __builtin_sve_svldff1ub_s64(__VA_ARGS__)
+#define svldff1ub_s16(...) __builtin_sve_svldff1ub_s16(__VA_ARGS__)
+#define svldff1uh_gather_u32base_index_u32(...) __builtin_sve_svldff1uh_gather_u32base_index_u32(__VA_ARGS__)
+#define svldff1uh_gather_u64base_index_u64(...) __builtin_sve_svldff1uh_gather_u64base_index_u64(__VA_ARGS__)
+#define svldff1uh_gather_u32base_index_s32(...) __builtin_sve_svldff1uh_gather_u32base_index_s32(__VA_ARGS__)
+#define svldff1uh_gather_u64base_index_s64(...) __builtin_sve_svldff1uh_gather_u64base_index_s64(__VA_ARGS__)
+#define svldff1uh_gather_u32base_offset_u32(...) __builtin_sve_svldff1uh_gather_u32base_offset_u32(__VA_ARGS__)
+#define svldff1uh_gather_u64base_offset_u64(...) __builtin_sve_svldff1uh_gather_u64base_offset_u64(__VA_ARGS__)
+#define svldff1uh_gather_u32base_offset_s32(...) __builtin_sve_svldff1uh_gather_u32base_offset_s32(__VA_ARGS__)
+#define svldff1uh_gather_u64base_offset_s64(...) __builtin_sve_svldff1uh_gather_u64base_offset_s64(__VA_ARGS__)
+#define svldff1uh_gather_u32base_u32(...) __builtin_sve_svldff1uh_gather_u32base_u32(__VA_ARGS__)
+#define svldff1uh_gather_u64base_u64(...) __builtin_sve_svldff1uh_gather_u64base_u64(__VA_ARGS__)
+#define svldff1uh_gather_u32base_s32(...) __builtin_sve_svldff1uh_gather_u32base_s32(__VA_ARGS__)
+#define svldff1uh_gather_u64base_s64(...) __builtin_sve_svldff1uh_gather_u64base_s64(__VA_ARGS__)
+#define svldff1uh_gather_s32index_u32(...) __builtin_sve_svldff1uh_gather_s32index_u32(__VA_ARGS__)
+#define svldff1uh_gather_s32index_s32(...) __builtin_sve_svldff1uh_gather_s32index_s32(__VA_ARGS__)
+#define svldff1uh_gather_u32index_u32(...) __builtin_sve_svldff1uh_gather_u32index_u32(__VA_ARGS__)
+#define svldff1uh_gather_u32index_s32(...) __builtin_sve_svldff1uh_gather_u32index_s32(__VA_ARGS__)
+#define svldff1uh_gather_s64index_u64(...) __builtin_sve_svldff1uh_gather_s64index_u64(__VA_ARGS__)
+#define svldff1uh_gather_s64index_s64(...) __builtin_sve_svldff1uh_gather_s64index_s64(__VA_ARGS__)
+#define svldff1uh_gather_u64index_u64(...) __builtin_sve_svldff1uh_gather_u64index_u64(__VA_ARGS__)
+#define svldff1uh_gather_u64index_s64(...) __builtin_sve_svldff1uh_gather_u64index_s64(__VA_ARGS__)
+#define svldff1uh_gather_s32offset_u32(...) __builtin_sve_svldff1uh_gather_s32offset_u32(__VA_ARGS__)
+#define svldff1uh_gather_s32offset_s32(...) __builtin_sve_svldff1uh_gather_s32offset_s32(__VA_ARGS__)
+#define svldff1uh_gather_u32offset_u32(...) __builtin_sve_svldff1uh_gather_u32offset_u32(__VA_ARGS__)
+#define svldff1uh_gather_u32offset_s32(...) __builtin_sve_svldff1uh_gather_u32offset_s32(__VA_ARGS__)
+#define svldff1uh_gather_s64offset_u64(...) __builtin_sve_svldff1uh_gather_s64offset_u64(__VA_ARGS__)
+#define svldff1uh_gather_s64offset_s64(...) __builtin_sve_svldff1uh_gather_s64offset_s64(__VA_ARGS__)
+#define svldff1uh_gather_u64offset_u64(...) __builtin_sve_svldff1uh_gather_u64offset_u64(__VA_ARGS__)
+#define svldff1uh_gather_u64offset_s64(...) __builtin_sve_svldff1uh_gather_u64offset_s64(__VA_ARGS__)
+#define svldff1uh_vnum_u32(...) __builtin_sve_svldff1uh_vnum_u32(__VA_ARGS__)
+#define svldff1uh_vnum_u64(...) __builtin_sve_svldff1uh_vnum_u64(__VA_ARGS__)
+#define svldff1uh_vnum_s32(...) __builtin_sve_svldff1uh_vnum_s32(__VA_ARGS__)
+#define svldff1uh_vnum_s64(...) __builtin_sve_svldff1uh_vnum_s64(__VA_ARGS__)
+#define svldff1uh_u32(...) __builtin_sve_svldff1uh_u32(__VA_ARGS__)
+#define svldff1uh_u64(...) __builtin_sve_svldff1uh_u64(__VA_ARGS__)
+#define svldff1uh_s32(...) __builtin_sve_svldff1uh_s32(__VA_ARGS__)
+#define svldff1uh_s64(...) __builtin_sve_svldff1uh_s64(__VA_ARGS__)
+#define svldff1uw_gather_u64base_index_u64(...) __builtin_sve_svldff1uw_gather_u64base_index_u64(__VA_ARGS__)
+#define svldff1uw_gather_u64base_index_s64(...) __builtin_sve_svldff1uw_gather_u64base_index_s64(__VA_ARGS__)
+#define svldff1uw_gather_u64base_offset_u64(...) __builtin_sve_svldff1uw_gather_u64base_offset_u64(__VA_ARGS__)
+#define svldff1uw_gather_u64base_offset_s64(...) __builtin_sve_svldff1uw_gather_u64base_offset_s64(__VA_ARGS__)
+#define svldff1uw_gather_u64base_u64(...) __builtin_sve_svldff1uw_gather_u64base_u64(__VA_ARGS__)
+#define svldff1uw_gather_u64base_s64(...) __builtin_sve_svldff1uw_gather_u64base_s64(__VA_ARGS__)
+#define svldff1uw_gather_s64index_u64(...) __builtin_sve_svldff1uw_gather_s64index_u64(__VA_ARGS__)
+#define svldff1uw_gather_s64index_s64(...) __builtin_sve_svldff1uw_gather_s64index_s64(__VA_ARGS__)
+#define svldff1uw_gather_u64index_u64(...) __builtin_sve_svldff1uw_gather_u64index_u64(__VA_ARGS__)
+#define svldff1uw_gather_u64index_s64(...) __builtin_sve_svldff1uw_gather_u64index_s64(__VA_ARGS__)
+#define svldff1uw_gather_s64offset_u64(...) __builtin_sve_svldff1uw_gather_s64offset_u64(__VA_ARGS__)
+#define svldff1uw_gather_s64offset_s64(...) __builtin_sve_svldff1uw_gather_s64offset_s64(__VA_ARGS__)
+#define svldff1uw_gather_u64offset_u64(...) __builtin_sve_svldff1uw_gather_u64offset_u64(__VA_ARGS__)
+#define svldff1uw_gather_u64offset_s64(...) __builtin_sve_svldff1uw_gather_u64offset_s64(__VA_ARGS__)
+#define svldff1uw_vnum_u64(...) __builtin_sve_svldff1uw_vnum_u64(__VA_ARGS__)
+#define svldff1uw_vnum_s64(...) __builtin_sve_svldff1uw_vnum_s64(__VA_ARGS__)
+#define svldff1uw_u64(...) __builtin_sve_svldff1uw_u64(__VA_ARGS__)
+#define svldff1uw_s64(...) __builtin_sve_svldff1uw_s64(__VA_ARGS__)
+#define svldnf1_u8(...) __builtin_sve_svldnf1_u8(__VA_ARGS__)
+#define svldnf1_u32(...) __builtin_sve_svldnf1_u32(__VA_ARGS__)
+#define svldnf1_u64(...) __builtin_sve_svldnf1_u64(__VA_ARGS__)
+#define svldnf1_u16(...) __builtin_sve_svldnf1_u16(__VA_ARGS__)
+#define svldnf1_s8(...) __builtin_sve_svldnf1_s8(__VA_ARGS__)
+#define svldnf1_f64(...) __builtin_sve_svldnf1_f64(__VA_ARGS__)
+#define svldnf1_f32(...) __builtin_sve_svldnf1_f32(__VA_ARGS__)
+#define svldnf1_f16(...) __builtin_sve_svldnf1_f16(__VA_ARGS__)
+#define svldnf1_s32(...) __builtin_sve_svldnf1_s32(__VA_ARGS__)
+#define svldnf1_s64(...) __builtin_sve_svldnf1_s64(__VA_ARGS__)
+#define svldnf1_s16(...) __builtin_sve_svldnf1_s16(__VA_ARGS__)
+#define svldnf1_vnum_u8(...) __builtin_sve_svldnf1_vnum_u8(__VA_ARGS__)
+#define svldnf1_vnum_u32(...) __builtin_sve_svldnf1_vnum_u32(__VA_ARGS__)
+#define svldnf1_vnum_u64(...) __builtin_sve_svldnf1_vnum_u64(__VA_ARGS__)
+#define svldnf1_vnum_u16(...) __builtin_sve_svldnf1_vnum_u16(__VA_ARGS__)
+#define svldnf1_vnum_s8(...) __builtin_sve_svldnf1_vnum_s8(__VA_ARGS__)
+#define svldnf1_vnum_f64(...) __builtin_sve_svldnf1_vnum_f64(__VA_ARGS__)
+#define svldnf1_vnum_f32(...) __builtin_sve_svldnf1_vnum_f32(__VA_ARGS__)
+#define svldnf1_vnum_f16(...) __builtin_sve_svldnf1_vnum_f16(__VA_ARGS__)
+#define svldnf1_vnum_s32(...) __builtin_sve_svldnf1_vnum_s32(__VA_ARGS__)
+#define svldnf1_vnum_s64(...) __builtin_sve_svldnf1_vnum_s64(__VA_ARGS__)
+#define svldnf1_vnum_s16(...) __builtin_sve_svldnf1_vnum_s16(__VA_ARGS__)
+#define svldnf1sb_vnum_u32(...) __builtin_sve_svldnf1sb_vnum_u32(__VA_ARGS__)
+#define svldnf1sb_vnum_u64(...) __builtin_sve_svldnf1sb_vnum_u64(__VA_ARGS__)
+#define svldnf1sb_vnum_u16(...) __builtin_sve_svldnf1sb_vnum_u16(__VA_ARGS__)
+#define svldnf1sb_vnum_s32(...) __builtin_sve_svldnf1sb_vnum_s32(__VA_ARGS__)
+#define svldnf1sb_vnum_s64(...) __builtin_sve_svldnf1sb_vnum_s64(__VA_ARGS__)
+#define svldnf1sb_vnum_s16(...) __builtin_sve_svldnf1sb_vnum_s16(__VA_ARGS__)
+#define svldnf1sb_u32(...) __builtin_sve_svldnf1sb_u32(__VA_ARGS__)
+#define svldnf1sb_u64(...) __builtin_sve_svldnf1sb_u64(__VA_ARGS__)
+#define svldnf1sb_u16(...) __builtin_sve_svldnf1sb_u16(__VA_ARGS__)
+#define svldnf1sb_s32(...) __builtin_sve_svldnf1sb_s32(__VA_ARGS__)
+#define svldnf1sb_s64(...) __builtin_sve_svldnf1sb_s64(__VA_ARGS__)
+#define svldnf1sb_s16(...) __builtin_sve_svldnf1sb_s16(__VA_ARGS__)
+#define svldnf1sh_vnum_u32(...) __builtin_sve_svldnf1sh_vnum_u32(__VA_ARGS__)
+#define svldnf1sh_vnum_u64(...) __builtin_sve_svldnf1sh_vnum_u64(__VA_ARGS__)
+#define svldnf1sh_vnum_s32(...) __builtin_sve_svldnf1sh_vnum_s32(__VA_ARGS__)
+#define svldnf1sh_vnum_s64(...) __builtin_sve_svldnf1sh_vnum_s64(__VA_ARGS__)
+#define svldnf1sh_u32(...) __builtin_sve_svldnf1sh_u32(__VA_ARGS__)
+#define svldnf1sh_u64(...) __builtin_sve_svldnf1sh_u64(__VA_ARGS__)
+#define svldnf1sh_s32(...) __builtin_sve_svldnf1sh_s32(__VA_ARGS__)
+#define svldnf1sh_s64(...) __builtin_sve_svldnf1sh_s64(__VA_ARGS__)
+#define svldnf1sw_vnum_u64(...) __builtin_sve_svldnf1sw_vnum_u64(__VA_ARGS__)
+#define svldnf1sw_vnum_s64(...) __builtin_sve_svldnf1sw_vnum_s64(__VA_ARGS__)
+#define svldnf1sw_u64(...) __builtin_sve_svldnf1sw_u64(__VA_ARGS__)
+#define svldnf1sw_s64(...) __builtin_sve_svldnf1sw_s64(__VA_ARGS__)
+#define svldnf1ub_vnum_u32(...) __builtin_sve_svldnf1ub_vnum_u32(__VA_ARGS__)
+#define svldnf1ub_vnum_u64(...) __builtin_sve_svldnf1ub_vnum_u64(__VA_ARGS__)
+#define svldnf1ub_vnum_u16(...) __builtin_sve_svldnf1ub_vnum_u16(__VA_ARGS__)
+#define svldnf1ub_vnum_s32(...) __builtin_sve_svldnf1ub_vnum_s32(__VA_ARGS__)
+#define svldnf1ub_vnum_s64(...) __builtin_sve_svldnf1ub_vnum_s64(__VA_ARGS__)
+#define svldnf1ub_vnum_s16(...) __builtin_sve_svldnf1ub_vnum_s16(__VA_ARGS__)
+#define svldnf1ub_u32(...) __builtin_sve_svldnf1ub_u32(__VA_ARGS__)
+#define svldnf1ub_u64(...) __builtin_sve_svldnf1ub_u64(__VA_ARGS__)
+#define svldnf1ub_u16(...) __builtin_sve_svldnf1ub_u16(__VA_ARGS__)
+#define svldnf1ub_s32(...) __builtin_sve_svldnf1ub_s32(__VA_ARGS__)
+#define svldnf1ub_s64(...) __builtin_sve_svldnf1ub_s64(__VA_ARGS__)
+#define svldnf1ub_s16(...) __builtin_sve_svldnf1ub_s16(__VA_ARGS__)
+#define svldnf1uh_vnum_u32(...) __builtin_sve_svldnf1uh_vnum_u32(__VA_ARGS__)
+#define svldnf1uh_vnum_u64(...) __builtin_sve_svldnf1uh_vnum_u64(__VA_ARGS__)
+#define svldnf1uh_vnum_s32(...) __builtin_sve_svldnf1uh_vnum_s32(__VA_ARGS__)
+#define svldnf1uh_vnum_s64(...) __builtin_sve_svldnf1uh_vnum_s64(__VA_ARGS__)
+#define svldnf1uh_u32(...) __builtin_sve_svldnf1uh_u32(__VA_ARGS__)
+#define svldnf1uh_u64(...) __builtin_sve_svldnf1uh_u64(__VA_ARGS__)
+#define svldnf1uh_s32(...) __builtin_sve_svldnf1uh_s32(__VA_ARGS__)
+#define svldnf1uh_s64(...) __builtin_sve_svldnf1uh_s64(__VA_ARGS__)
+#define svldnf1uw_vnum_u64(...) __builtin_sve_svldnf1uw_vnum_u64(__VA_ARGS__)
+#define svldnf1uw_vnum_s64(...) __builtin_sve_svldnf1uw_vnum_s64(__VA_ARGS__)
+#define svldnf1uw_u64(...) __builtin_sve_svldnf1uw_u64(__VA_ARGS__)
+#define svldnf1uw_s64(...) __builtin_sve_svldnf1uw_s64(__VA_ARGS__)
+#define svldnt1_u8(...) __builtin_sve_svldnt1_u8(__VA_ARGS__)
+#define svldnt1_u32(...) __builtin_sve_svldnt1_u32(__VA_ARGS__)
+#define svldnt1_u64(...) __builtin_sve_svldnt1_u64(__VA_ARGS__)
+#define svldnt1_u16(...) __builtin_sve_svldnt1_u16(__VA_ARGS__)
+#define svldnt1_s8(...) __builtin_sve_svldnt1_s8(__VA_ARGS__)
+#define svldnt1_f64(...) __builtin_sve_svldnt1_f64(__VA_ARGS__)
+#define svldnt1_f32(...) __builtin_sve_svldnt1_f32(__VA_ARGS__)
+#define svldnt1_f16(...) __builtin_sve_svldnt1_f16(__VA_ARGS__)
+#define svldnt1_s32(...) __builtin_sve_svldnt1_s32(__VA_ARGS__)
+#define svldnt1_s64(...) __builtin_sve_svldnt1_s64(__VA_ARGS__)
+#define svldnt1_s16(...) __builtin_sve_svldnt1_s16(__VA_ARGS__)
+#define svldnt1_vnum_u8(...) __builtin_sve_svldnt1_vnum_u8(__VA_ARGS__)
+#define svldnt1_vnum_u32(...) __builtin_sve_svldnt1_vnum_u32(__VA_ARGS__)
+#define svldnt1_vnum_u64(...) __builtin_sve_svldnt1_vnum_u64(__VA_ARGS__)
+#define svldnt1_vnum_u16(...) __builtin_sve_svldnt1_vnum_u16(__VA_ARGS__)
+#define svldnt1_vnum_s8(...) __builtin_sve_svldnt1_vnum_s8(__VA_ARGS__)
+#define svldnt1_vnum_f64(...) __builtin_sve_svldnt1_vnum_f64(__VA_ARGS__)
+#define svldnt1_vnum_f32(...) __builtin_sve_svldnt1_vnum_f32(__VA_ARGS__)
+#define svldnt1_vnum_f16(...) __builtin_sve_svldnt1_vnum_f16(__VA_ARGS__)
+#define svldnt1_vnum_s32(...) __builtin_sve_svldnt1_vnum_s32(__VA_ARGS__)
+#define svldnt1_vnum_s64(...) __builtin_sve_svldnt1_vnum_s64(__VA_ARGS__)
+#define svldnt1_vnum_s16(...) __builtin_sve_svldnt1_vnum_s16(__VA_ARGS__)
+#define svlen_u8(...) __builtin_sve_svlen_u8(__VA_ARGS__)
+#define svlen_u32(...) __builtin_sve_svlen_u32(__VA_ARGS__)
+#define svlen_u64(...) __builtin_sve_svlen_u64(__VA_ARGS__)
+#define svlen_u16(...) __builtin_sve_svlen_u16(__VA_ARGS__)
+#define svlen_s8(...) __builtin_sve_svlen_s8(__VA_ARGS__)
+#define svlen_f64(...) __builtin_sve_svlen_f64(__VA_ARGS__)
+#define svlen_f32(...) __builtin_sve_svlen_f32(__VA_ARGS__)
+#define svlen_f16(...) __builtin_sve_svlen_f16(__VA_ARGS__)
+#define svlen_s32(...) __builtin_sve_svlen_s32(__VA_ARGS__)
+#define svlen_s64(...) __builtin_sve_svlen_s64(__VA_ARGS__)
+#define svlen_s16(...) __builtin_sve_svlen_s16(__VA_ARGS__)
+#define svlsl_n_u8_m(...) __builtin_sve_svlsl_n_u8_m(__VA_ARGS__)
+#define svlsl_n_u32_m(...) __builtin_sve_svlsl_n_u32_m(__VA_ARGS__)
+#define svlsl_n_u64_m(...) __builtin_sve_svlsl_n_u64_m(__VA_ARGS__)
+#define svlsl_n_u16_m(...) __builtin_sve_svlsl_n_u16_m(__VA_ARGS__)
+#define svlsl_n_s8_m(...) __builtin_sve_svlsl_n_s8_m(__VA_ARGS__)
+#define svlsl_n_s32_m(...) __builtin_sve_svlsl_n_s32_m(__VA_ARGS__)
+#define svlsl_n_s64_m(...) __builtin_sve_svlsl_n_s64_m(__VA_ARGS__)
+#define svlsl_n_s16_m(...) __builtin_sve_svlsl_n_s16_m(__VA_ARGS__)
+#define svlsl_n_u8_x(...) __builtin_sve_svlsl_n_u8_x(__VA_ARGS__)
+#define svlsl_n_u32_x(...) __builtin_sve_svlsl_n_u32_x(__VA_ARGS__)
+#define svlsl_n_u64_x(...) __builtin_sve_svlsl_n_u64_x(__VA_ARGS__)
+#define svlsl_n_u16_x(...) __builtin_sve_svlsl_n_u16_x(__VA_ARGS__)
+#define svlsl_n_s8_x(...) __builtin_sve_svlsl_n_s8_x(__VA_ARGS__)
+#define svlsl_n_s32_x(...) __builtin_sve_svlsl_n_s32_x(__VA_ARGS__)
+#define svlsl_n_s64_x(...) __builtin_sve_svlsl_n_s64_x(__VA_ARGS__)
+#define svlsl_n_s16_x(...) __builtin_sve_svlsl_n_s16_x(__VA_ARGS__)
+#define svlsl_n_u8_z(...) __builtin_sve_svlsl_n_u8_z(__VA_ARGS__)
+#define svlsl_n_u32_z(...) __builtin_sve_svlsl_n_u32_z(__VA_ARGS__)
+#define svlsl_n_u64_z(...) __builtin_sve_svlsl_n_u64_z(__VA_ARGS__)
+#define svlsl_n_u16_z(...) __builtin_sve_svlsl_n_u16_z(__VA_ARGS__)
+#define svlsl_n_s8_z(...) __builtin_sve_svlsl_n_s8_z(__VA_ARGS__)
+#define svlsl_n_s32_z(...) __builtin_sve_svlsl_n_s32_z(__VA_ARGS__)
+#define svlsl_n_s64_z(...) __builtin_sve_svlsl_n_s64_z(__VA_ARGS__)
+#define svlsl_n_s16_z(...) __builtin_sve_svlsl_n_s16_z(__VA_ARGS__)
+#define svlsl_u8_m(...) __builtin_sve_svlsl_u8_m(__VA_ARGS__)
+#define svlsl_u32_m(...) __builtin_sve_svlsl_u32_m(__VA_ARGS__)
+#define svlsl_u64_m(...) __builtin_sve_svlsl_u64_m(__VA_ARGS__)
+#define svlsl_u16_m(...) __builtin_sve_svlsl_u16_m(__VA_ARGS__)
+#define svlsl_s8_m(...) __builtin_sve_svlsl_s8_m(__VA_ARGS__)
+#define svlsl_s32_m(...) __builtin_sve_svlsl_s32_m(__VA_ARGS__)
+#define svlsl_s64_m(...) __builtin_sve_svlsl_s64_m(__VA_ARGS__)
+#define svlsl_s16_m(...) __builtin_sve_svlsl_s16_m(__VA_ARGS__)
+#define svlsl_u8_x(...) __builtin_sve_svlsl_u8_x(__VA_ARGS__)
+#define svlsl_u32_x(...) __builtin_sve_svlsl_u32_x(__VA_ARGS__)
+#define svlsl_u64_x(...) __builtin_sve_svlsl_u64_x(__VA_ARGS__)
+#define svlsl_u16_x(...) __builtin_sve_svlsl_u16_x(__VA_ARGS__)
+#define svlsl_s8_x(...) __builtin_sve_svlsl_s8_x(__VA_ARGS__)
+#define svlsl_s32_x(...) __builtin_sve_svlsl_s32_x(__VA_ARGS__)
+#define svlsl_s64_x(...) __builtin_sve_svlsl_s64_x(__VA_ARGS__)
+#define svlsl_s16_x(...) __builtin_sve_svlsl_s16_x(__VA_ARGS__)
+#define svlsl_u8_z(...) __builtin_sve_svlsl_u8_z(__VA_ARGS__)
+#define svlsl_u32_z(...) __builtin_sve_svlsl_u32_z(__VA_ARGS__)
+#define svlsl_u64_z(...) __builtin_sve_svlsl_u64_z(__VA_ARGS__)
+#define svlsl_u16_z(...) __builtin_sve_svlsl_u16_z(__VA_ARGS__)
+#define svlsl_s8_z(...) __builtin_sve_svlsl_s8_z(__VA_ARGS__)
+#define svlsl_s32_z(...) __builtin_sve_svlsl_s32_z(__VA_ARGS__)
+#define svlsl_s64_z(...) __builtin_sve_svlsl_s64_z(__VA_ARGS__)
+#define svlsl_s16_z(...) __builtin_sve_svlsl_s16_z(__VA_ARGS__)
+#define svlsl_wide_n_u8_m(...) __builtin_sve_svlsl_wide_n_u8_m(__VA_ARGS__)
+#define svlsl_wide_n_u32_m(...) __builtin_sve_svlsl_wide_n_u32_m(__VA_ARGS__)
+#define svlsl_wide_n_u16_m(...) __builtin_sve_svlsl_wide_n_u16_m(__VA_ARGS__)
+#define svlsl_wide_n_s8_m(...) __builtin_sve_svlsl_wide_n_s8_m(__VA_ARGS__)
+#define svlsl_wide_n_s32_m(...) __builtin_sve_svlsl_wide_n_s32_m(__VA_ARGS__)
+#define svlsl_wide_n_s16_m(...) __builtin_sve_svlsl_wide_n_s16_m(__VA_ARGS__)
+#define svlsl_wide_n_u8_x(...) __builtin_sve_svlsl_wide_n_u8_x(__VA_ARGS__)
+#define svlsl_wide_n_u32_x(...) __builtin_sve_svlsl_wide_n_u32_x(__VA_ARGS__)
+#define svlsl_wide_n_u16_x(...) __builtin_sve_svlsl_wide_n_u16_x(__VA_ARGS__)
+#define svlsl_wide_n_s8_x(...) __builtin_sve_svlsl_wide_n_s8_x(__VA_ARGS__)
+#define svlsl_wide_n_s32_x(...) __builtin_sve_svlsl_wide_n_s32_x(__VA_ARGS__)
+#define svlsl_wide_n_s16_x(...) __builtin_sve_svlsl_wide_n_s16_x(__VA_ARGS__)
+#define svlsl_wide_n_u8_z(...) __builtin_sve_svlsl_wide_n_u8_z(__VA_ARGS__)
+#define svlsl_wide_n_u32_z(...) __builtin_sve_svlsl_wide_n_u32_z(__VA_ARGS__)
+#define svlsl_wide_n_u16_z(...) __builtin_sve_svlsl_wide_n_u16_z(__VA_ARGS__)
+#define svlsl_wide_n_s8_z(...) __builtin_sve_svlsl_wide_n_s8_z(__VA_ARGS__)
+#define svlsl_wide_n_s32_z(...) __builtin_sve_svlsl_wide_n_s32_z(__VA_ARGS__)
+#define svlsl_wide_n_s16_z(...) __builtin_sve_svlsl_wide_n_s16_z(__VA_ARGS__)
+#define svlsl_wide_u8_m(...) __builtin_sve_svlsl_wide_u8_m(__VA_ARGS__)
+#define svlsl_wide_u32_m(...) __builtin_sve_svlsl_wide_u32_m(__VA_ARGS__)
+#define svlsl_wide_u16_m(...) __builtin_sve_svlsl_wide_u16_m(__VA_ARGS__)
+#define svlsl_wide_s8_m(...) __builtin_sve_svlsl_wide_s8_m(__VA_ARGS__)
+#define svlsl_wide_s32_m(...) __builtin_sve_svlsl_wide_s32_m(__VA_ARGS__)
+#define svlsl_wide_s16_m(...) __builtin_sve_svlsl_wide_s16_m(__VA_ARGS__)
+#define svlsl_wide_u8_x(...) __builtin_sve_svlsl_wide_u8_x(__VA_ARGS__)
+#define svlsl_wide_u32_x(...) __builtin_sve_svlsl_wide_u32_x(__VA_ARGS__)
+#define svlsl_wide_u16_x(...) __builtin_sve_svlsl_wide_u16_x(__VA_ARGS__)
+#define svlsl_wide_s8_x(...) __builtin_sve_svlsl_wide_s8_x(__VA_ARGS__)
+#define svlsl_wide_s32_x(...) __builtin_sve_svlsl_wide_s32_x(__VA_ARGS__)
+#define svlsl_wide_s16_x(...) __builtin_sve_svlsl_wide_s16_x(__VA_ARGS__)
+#define svlsl_wide_u8_z(...) __builtin_sve_svlsl_wide_u8_z(__VA_ARGS__)
+#define svlsl_wide_u32_z(...) __builtin_sve_svlsl_wide_u32_z(__VA_ARGS__)
+#define svlsl_wide_u16_z(...) __builtin_sve_svlsl_wide_u16_z(__VA_ARGS__)
+#define svlsl_wide_s8_z(...) __builtin_sve_svlsl_wide_s8_z(__VA_ARGS__)
+#define svlsl_wide_s32_z(...) __builtin_sve_svlsl_wide_s32_z(__VA_ARGS__)
+#define svlsl_wide_s16_z(...) __builtin_sve_svlsl_wide_s16_z(__VA_ARGS__)
+#define svlsr_n_u8_m(...) __builtin_sve_svlsr_n_u8_m(__VA_ARGS__)
+#define svlsr_n_u32_m(...) __builtin_sve_svlsr_n_u32_m(__VA_ARGS__)
+#define svlsr_n_u64_m(...) __builtin_sve_svlsr_n_u64_m(__VA_ARGS__)
+#define svlsr_n_u16_m(...) __builtin_sve_svlsr_n_u16_m(__VA_ARGS__)
+#define svlsr_n_u8_x(...) __builtin_sve_svlsr_n_u8_x(__VA_ARGS__)
+#define svlsr_n_u32_x(...) __builtin_sve_svlsr_n_u32_x(__VA_ARGS__)
+#define svlsr_n_u64_x(...) __builtin_sve_svlsr_n_u64_x(__VA_ARGS__)
+#define svlsr_n_u16_x(...) __builtin_sve_svlsr_n_u16_x(__VA_ARGS__)
+#define svlsr_n_u8_z(...) __builtin_sve_svlsr_n_u8_z(__VA_ARGS__)
+#define svlsr_n_u32_z(...) __builtin_sve_svlsr_n_u32_z(__VA_ARGS__)
+#define svlsr_n_u64_z(...) __builtin_sve_svlsr_n_u64_z(__VA_ARGS__)
+#define svlsr_n_u16_z(...) __builtin_sve_svlsr_n_u16_z(__VA_ARGS__)
+#define svlsr_u8_m(...) __builtin_sve_svlsr_u8_m(__VA_ARGS__)
+#define svlsr_u32_m(...) __builtin_sve_svlsr_u32_m(__VA_ARGS__)
+#define svlsr_u64_m(...) __builtin_sve_svlsr_u64_m(__VA_ARGS__)
+#define svlsr_u16_m(...) __builtin_sve_svlsr_u16_m(__VA_ARGS__)
+#define svlsr_u8_x(...) __builtin_sve_svlsr_u8_x(__VA_ARGS__)
+#define svlsr_u32_x(...) __builtin_sve_svlsr_u32_x(__VA_ARGS__)
+#define svlsr_u64_x(...) __builtin_sve_svlsr_u64_x(__VA_ARGS__)
+#define svlsr_u16_x(...) __builtin_sve_svlsr_u16_x(__VA_ARGS__)
+#define svlsr_u8_z(...) __builtin_sve_svlsr_u8_z(__VA_ARGS__)
+#define svlsr_u32_z(...) __builtin_sve_svlsr_u32_z(__VA_ARGS__)
+#define svlsr_u64_z(...) __builtin_sve_svlsr_u64_z(__VA_ARGS__)
+#define svlsr_u16_z(...) __builtin_sve_svlsr_u16_z(__VA_ARGS__)
+#define svlsr_wide_n_u8_m(...) __builtin_sve_svlsr_wide_n_u8_m(__VA_ARGS__)
+#define svlsr_wide_n_u32_m(...) __builtin_sve_svlsr_wide_n_u32_m(__VA_ARGS__)
+#define svlsr_wide_n_u16_m(...) __builtin_sve_svlsr_wide_n_u16_m(__VA_ARGS__)
+#define svlsr_wide_n_u8_x(...) __builtin_sve_svlsr_wide_n_u8_x(__VA_ARGS__)
+#define svlsr_wide_n_u32_x(...) __builtin_sve_svlsr_wide_n_u32_x(__VA_ARGS__)
+#define svlsr_wide_n_u16_x(...) __builtin_sve_svlsr_wide_n_u16_x(__VA_ARGS__)
+#define svlsr_wide_n_u8_z(...) __builtin_sve_svlsr_wide_n_u8_z(__VA_ARGS__)
+#define svlsr_wide_n_u32_z(...) __builtin_sve_svlsr_wide_n_u32_z(__VA_ARGS__)
+#define svlsr_wide_n_u16_z(...) __builtin_sve_svlsr_wide_n_u16_z(__VA_ARGS__)
+#define svlsr_wide_u8_m(...) __builtin_sve_svlsr_wide_u8_m(__VA_ARGS__)
+#define svlsr_wide_u32_m(...) __builtin_sve_svlsr_wide_u32_m(__VA_ARGS__)
+#define svlsr_wide_u16_m(...) __builtin_sve_svlsr_wide_u16_m(__VA_ARGS__)
+#define svlsr_wide_u8_x(...) __builtin_sve_svlsr_wide_u8_x(__VA_ARGS__)
+#define svlsr_wide_u32_x(...) __builtin_sve_svlsr_wide_u32_x(__VA_ARGS__)
+#define svlsr_wide_u16_x(...) __builtin_sve_svlsr_wide_u16_x(__VA_ARGS__)
+#define svlsr_wide_u8_z(...) __builtin_sve_svlsr_wide_u8_z(__VA_ARGS__)
+#define svlsr_wide_u32_z(...) __builtin_sve_svlsr_wide_u32_z(__VA_ARGS__)
+#define svlsr_wide_u16_z(...) __builtin_sve_svlsr_wide_u16_z(__VA_ARGS__)
+#define svmad_n_f64_m(...) __builtin_sve_svmad_n_f64_m(__VA_ARGS__)
+#define svmad_n_f32_m(...) __builtin_sve_svmad_n_f32_m(__VA_ARGS__)
+#define svmad_n_f16_m(...) __builtin_sve_svmad_n_f16_m(__VA_ARGS__)
+#define svmad_n_f64_x(...) __builtin_sve_svmad_n_f64_x(__VA_ARGS__)
+#define svmad_n_f32_x(...) __builtin_sve_svmad_n_f32_x(__VA_ARGS__)
+#define svmad_n_f16_x(...) __builtin_sve_svmad_n_f16_x(__VA_ARGS__)
+#define svmad_n_f64_z(...) __builtin_sve_svmad_n_f64_z(__VA_ARGS__)
+#define svmad_n_f32_z(...) __builtin_sve_svmad_n_f32_z(__VA_ARGS__)
+#define svmad_n_f16_z(...) __builtin_sve_svmad_n_f16_z(__VA_ARGS__)
+#define svmad_n_u8_m(...) __builtin_sve_svmad_n_u8_m(__VA_ARGS__)
+#define svmad_n_u32_m(...) __builtin_sve_svmad_n_u32_m(__VA_ARGS__)
+#define svmad_n_u64_m(...) __builtin_sve_svmad_n_u64_m(__VA_ARGS__)
+#define svmad_n_u16_m(...) __builtin_sve_svmad_n_u16_m(__VA_ARGS__)
+#define svmad_n_s8_m(...) __builtin_sve_svmad_n_s8_m(__VA_ARGS__)
+#define svmad_n_s32_m(...) __builtin_sve_svmad_n_s32_m(__VA_ARGS__)
+#define svmad_n_s64_m(...) __builtin_sve_svmad_n_s64_m(__VA_ARGS__)
+#define svmad_n_s16_m(...) __builtin_sve_svmad_n_s16_m(__VA_ARGS__)
+#define svmad_n_u8_x(...) __builtin_sve_svmad_n_u8_x(__VA_ARGS__)
+#define svmad_n_u32_x(...) __builtin_sve_svmad_n_u32_x(__VA_ARGS__)
+#define svmad_n_u64_x(...) __builtin_sve_svmad_n_u64_x(__VA_ARGS__)
+#define svmad_n_u16_x(...) __builtin_sve_svmad_n_u16_x(__VA_ARGS__)
+#define svmad_n_s8_x(...) __builtin_sve_svmad_n_s8_x(__VA_ARGS__)
+#define svmad_n_s32_x(...) __builtin_sve_svmad_n_s32_x(__VA_ARGS__)
+#define svmad_n_s64_x(...) __builtin_sve_svmad_n_s64_x(__VA_ARGS__)
+#define svmad_n_s16_x(...) __builtin_sve_svmad_n_s16_x(__VA_ARGS__)
+#define svmad_n_u8_z(...) __builtin_sve_svmad_n_u8_z(__VA_ARGS__)
+#define svmad_n_u32_z(...) __builtin_sve_svmad_n_u32_z(__VA_ARGS__)
+#define svmad_n_u64_z(...) __builtin_sve_svmad_n_u64_z(__VA_ARGS__)
+#define svmad_n_u16_z(...) __builtin_sve_svmad_n_u16_z(__VA_ARGS__)
+#define svmad_n_s8_z(...) __builtin_sve_svmad_n_s8_z(__VA_ARGS__)
+#define svmad_n_s32_z(...) __builtin_sve_svmad_n_s32_z(__VA_ARGS__)
+#define svmad_n_s64_z(...) __builtin_sve_svmad_n_s64_z(__VA_ARGS__)
+#define svmad_n_s16_z(...) __builtin_sve_svmad_n_s16_z(__VA_ARGS__)
+#define svmad_f64_m(...) __builtin_sve_svmad_f64_m(__VA_ARGS__)
+#define svmad_f32_m(...) __builtin_sve_svmad_f32_m(__VA_ARGS__)
+#define svmad_f16_m(...) __builtin_sve_svmad_f16_m(__VA_ARGS__)
+#define svmad_f64_x(...) __builtin_sve_svmad_f64_x(__VA_ARGS__)
+#define svmad_f32_x(...) __builtin_sve_svmad_f32_x(__VA_ARGS__)
+#define svmad_f16_x(...) __builtin_sve_svmad_f16_x(__VA_ARGS__)
+#define svmad_f64_z(...) __builtin_sve_svmad_f64_z(__VA_ARGS__)
+#define svmad_f32_z(...) __builtin_sve_svmad_f32_z(__VA_ARGS__)
+#define svmad_f16_z(...) __builtin_sve_svmad_f16_z(__VA_ARGS__)
+#define svmad_u8_m(...) __builtin_sve_svmad_u8_m(__VA_ARGS__)
+#define svmad_u32_m(...) __builtin_sve_svmad_u32_m(__VA_ARGS__)
+#define svmad_u64_m(...) __builtin_sve_svmad_u64_m(__VA_ARGS__)
+#define svmad_u16_m(...) __builtin_sve_svmad_u16_m(__VA_ARGS__)
+#define svmad_s8_m(...) __builtin_sve_svmad_s8_m(__VA_ARGS__)
+#define svmad_s32_m(...) __builtin_sve_svmad_s32_m(__VA_ARGS__)
+#define svmad_s64_m(...) __builtin_sve_svmad_s64_m(__VA_ARGS__)
+#define svmad_s16_m(...) __builtin_sve_svmad_s16_m(__VA_ARGS__)
+#define svmad_u8_x(...) __builtin_sve_svmad_u8_x(__VA_ARGS__)
+#define svmad_u32_x(...) __builtin_sve_svmad_u32_x(__VA_ARGS__)
+#define svmad_u64_x(...) __builtin_sve_svmad_u64_x(__VA_ARGS__)
+#define svmad_u16_x(...) __builtin_sve_svmad_u16_x(__VA_ARGS__)
+#define svmad_s8_x(...) __builtin_sve_svmad_s8_x(__VA_ARGS__)
+#define svmad_s32_x(...) __builtin_sve_svmad_s32_x(__VA_ARGS__)
+#define svmad_s64_x(...) __builtin_sve_svmad_s64_x(__VA_ARGS__)
+#define svmad_s16_x(...) __builtin_sve_svmad_s16_x(__VA_ARGS__)
+#define svmad_u8_z(...) __builtin_sve_svmad_u8_z(__VA_ARGS__)
+#define svmad_u32_z(...) __builtin_sve_svmad_u32_z(__VA_ARGS__)
+#define svmad_u64_z(...) __builtin_sve_svmad_u64_z(__VA_ARGS__)
+#define svmad_u16_z(...) __builtin_sve_svmad_u16_z(__VA_ARGS__)
+#define svmad_s8_z(...) __builtin_sve_svmad_s8_z(__VA_ARGS__)
+#define svmad_s32_z(...) __builtin_sve_svmad_s32_z(__VA_ARGS__)
+#define svmad_s64_z(...) __builtin_sve_svmad_s64_z(__VA_ARGS__)
+#define svmad_s16_z(...) __builtin_sve_svmad_s16_z(__VA_ARGS__)
+#define svmax_n_f64_m(...) __builtin_sve_svmax_n_f64_m(__VA_ARGS__)
+#define svmax_n_f32_m(...) __builtin_sve_svmax_n_f32_m(__VA_ARGS__)
+#define svmax_n_f16_m(...) __builtin_sve_svmax_n_f16_m(__VA_ARGS__)
+#define svmax_n_f64_x(...) __builtin_sve_svmax_n_f64_x(__VA_ARGS__)
+#define svmax_n_f32_x(...) __builtin_sve_svmax_n_f32_x(__VA_ARGS__)
+#define svmax_n_f16_x(...) __builtin_sve_svmax_n_f16_x(__VA_ARGS__)
+#define svmax_n_f64_z(...) __builtin_sve_svmax_n_f64_z(__VA_ARGS__)
+#define svmax_n_f32_z(...) __builtin_sve_svmax_n_f32_z(__VA_ARGS__)
+#define svmax_n_f16_z(...) __builtin_sve_svmax_n_f16_z(__VA_ARGS__)
+#define svmax_n_s8_m(...) __builtin_sve_svmax_n_s8_m(__VA_ARGS__)
+#define svmax_n_s32_m(...) __builtin_sve_svmax_n_s32_m(__VA_ARGS__)
+#define svmax_n_s64_m(...) __builtin_sve_svmax_n_s64_m(__VA_ARGS__)
+#define svmax_n_s16_m(...) __builtin_sve_svmax_n_s16_m(__VA_ARGS__)
+#define svmax_n_s8_x(...) __builtin_sve_svmax_n_s8_x(__VA_ARGS__)
+#define svmax_n_s32_x(...) __builtin_sve_svmax_n_s32_x(__VA_ARGS__)
+#define svmax_n_s64_x(...) __builtin_sve_svmax_n_s64_x(__VA_ARGS__)
+#define svmax_n_s16_x(...) __builtin_sve_svmax_n_s16_x(__VA_ARGS__)
+#define svmax_n_s8_z(...) __builtin_sve_svmax_n_s8_z(__VA_ARGS__)
+#define svmax_n_s32_z(...) __builtin_sve_svmax_n_s32_z(__VA_ARGS__)
+#define svmax_n_s64_z(...) __builtin_sve_svmax_n_s64_z(__VA_ARGS__)
+#define svmax_n_s16_z(...) __builtin_sve_svmax_n_s16_z(__VA_ARGS__)
+#define svmax_n_u8_m(...) __builtin_sve_svmax_n_u8_m(__VA_ARGS__)
+#define svmax_n_u32_m(...) __builtin_sve_svmax_n_u32_m(__VA_ARGS__)
+#define svmax_n_u64_m(...) __builtin_sve_svmax_n_u64_m(__VA_ARGS__)
+#define svmax_n_u16_m(...) __builtin_sve_svmax_n_u16_m(__VA_ARGS__)
+#define svmax_n_u8_x(...) __builtin_sve_svmax_n_u8_x(__VA_ARGS__)
+#define svmax_n_u32_x(...) __builtin_sve_svmax_n_u32_x(__VA_ARGS__)
+#define svmax_n_u64_x(...) __builtin_sve_svmax_n_u64_x(__VA_ARGS__)
+#define svmax_n_u16_x(...) __builtin_sve_svmax_n_u16_x(__VA_ARGS__)
+#define svmax_n_u8_z(...) __builtin_sve_svmax_n_u8_z(__VA_ARGS__)
+#define svmax_n_u32_z(...) __builtin_sve_svmax_n_u32_z(__VA_ARGS__)
+#define svmax_n_u64_z(...) __builtin_sve_svmax_n_u64_z(__VA_ARGS__)
+#define svmax_n_u16_z(...) __builtin_sve_svmax_n_u16_z(__VA_ARGS__)
+#define svmax_f64_m(...) __builtin_sve_svmax_f64_m(__VA_ARGS__)
+#define svmax_f32_m(...) __builtin_sve_svmax_f32_m(__VA_ARGS__)
+#define svmax_f16_m(...) __builtin_sve_svmax_f16_m(__VA_ARGS__)
+#define svmax_f64_x(...) __builtin_sve_svmax_f64_x(__VA_ARGS__)
+#define svmax_f32_x(...) __builtin_sve_svmax_f32_x(__VA_ARGS__)
+#define svmax_f16_x(...) __builtin_sve_svmax_f16_x(__VA_ARGS__)
+#define svmax_f64_z(...) __builtin_sve_svmax_f64_z(__VA_ARGS__)
+#define svmax_f32_z(...) __builtin_sve_svmax_f32_z(__VA_ARGS__)
+#define svmax_f16_z(...) __builtin_sve_svmax_f16_z(__VA_ARGS__)
+#define svmax_s8_m(...) __builtin_sve_svmax_s8_m(__VA_ARGS__)
+#define svmax_s32_m(...) __builtin_sve_svmax_s32_m(__VA_ARGS__)
+#define svmax_s64_m(...) __builtin_sve_svmax_s64_m(__VA_ARGS__)
+#define svmax_s16_m(...) __builtin_sve_svmax_s16_m(__VA_ARGS__)
+#define svmax_s8_x(...) __builtin_sve_svmax_s8_x(__VA_ARGS__)
+#define svmax_s32_x(...) __builtin_sve_svmax_s32_x(__VA_ARGS__)
+#define svmax_s64_x(...) __builtin_sve_svmax_s64_x(__VA_ARGS__)
+#define svmax_s16_x(...) __builtin_sve_svmax_s16_x(__VA_ARGS__)
+#define svmax_s8_z(...) __builtin_sve_svmax_s8_z(__VA_ARGS__)
+#define svmax_s32_z(...) __builtin_sve_svmax_s32_z(__VA_ARGS__)
+#define svmax_s64_z(...) __builtin_sve_svmax_s64_z(__VA_ARGS__)
+#define svmax_s16_z(...) __builtin_sve_svmax_s16_z(__VA_ARGS__)
+#define svmax_u8_m(...) __builtin_sve_svmax_u8_m(__VA_ARGS__)
+#define svmax_u32_m(...) __builtin_sve_svmax_u32_m(__VA_ARGS__)
+#define svmax_u64_m(...) __builtin_sve_svmax_u64_m(__VA_ARGS__)
+#define svmax_u16_m(...) __builtin_sve_svmax_u16_m(__VA_ARGS__)
+#define svmax_u8_x(...) __builtin_sve_svmax_u8_x(__VA_ARGS__)
+#define svmax_u32_x(...) __builtin_sve_svmax_u32_x(__VA_ARGS__)
+#define svmax_u64_x(...) __builtin_sve_svmax_u64_x(__VA_ARGS__)
+#define svmax_u16_x(...) __builtin_sve_svmax_u16_x(__VA_ARGS__)
+#define svmax_u8_z(...) __builtin_sve_svmax_u8_z(__VA_ARGS__)
+#define svmax_u32_z(...) __builtin_sve_svmax_u32_z(__VA_ARGS__)
+#define svmax_u64_z(...) __builtin_sve_svmax_u64_z(__VA_ARGS__)
+#define svmax_u16_z(...) __builtin_sve_svmax_u16_z(__VA_ARGS__)
+#define svmaxnm_n_f64_m(...) __builtin_sve_svmaxnm_n_f64_m(__VA_ARGS__)
+#define svmaxnm_n_f32_m(...) __builtin_sve_svmaxnm_n_f32_m(__VA_ARGS__)
+#define svmaxnm_n_f16_m(...) __builtin_sve_svmaxnm_n_f16_m(__VA_ARGS__)
+#define svmaxnm_n_f64_x(...) __builtin_sve_svmaxnm_n_f64_x(__VA_ARGS__)
+#define svmaxnm_n_f32_x(...) __builtin_sve_svmaxnm_n_f32_x(__VA_ARGS__)
+#define svmaxnm_n_f16_x(...) __builtin_sve_svmaxnm_n_f16_x(__VA_ARGS__)
+#define svmaxnm_n_f64_z(...) __builtin_sve_svmaxnm_n_f64_z(__VA_ARGS__)
+#define svmaxnm_n_f32_z(...) __builtin_sve_svmaxnm_n_f32_z(__VA_ARGS__)
+#define svmaxnm_n_f16_z(...) __builtin_sve_svmaxnm_n_f16_z(__VA_ARGS__)
+#define svmaxnm_f64_m(...) __builtin_sve_svmaxnm_f64_m(__VA_ARGS__)
+#define svmaxnm_f32_m(...) __builtin_sve_svmaxnm_f32_m(__VA_ARGS__)
+#define svmaxnm_f16_m(...) __builtin_sve_svmaxnm_f16_m(__VA_ARGS__)
+#define svmaxnm_f64_x(...) __builtin_sve_svmaxnm_f64_x(__VA_ARGS__)
+#define svmaxnm_f32_x(...) __builtin_sve_svmaxnm_f32_x(__VA_ARGS__)
+#define svmaxnm_f16_x(...) __builtin_sve_svmaxnm_f16_x(__VA_ARGS__)
+#define svmaxnm_f64_z(...) __builtin_sve_svmaxnm_f64_z(__VA_ARGS__)
+#define svmaxnm_f32_z(...) __builtin_sve_svmaxnm_f32_z(__VA_ARGS__)
+#define svmaxnm_f16_z(...) __builtin_sve_svmaxnm_f16_z(__VA_ARGS__)
+#define svmaxnmv_f64(...) __builtin_sve_svmaxnmv_f64(__VA_ARGS__)
+#define svmaxnmv_f32(...) __builtin_sve_svmaxnmv_f32(__VA_ARGS__)
+#define svmaxnmv_f16(...) __builtin_sve_svmaxnmv_f16(__VA_ARGS__)
+#define svmaxv_f64(...) __builtin_sve_svmaxv_f64(__VA_ARGS__)
+#define svmaxv_f32(...) __builtin_sve_svmaxv_f32(__VA_ARGS__)
+#define svmaxv_f16(...) __builtin_sve_svmaxv_f16(__VA_ARGS__)
+#define svmaxv_s8(...) __builtin_sve_svmaxv_s8(__VA_ARGS__)
+#define svmaxv_s32(...) __builtin_sve_svmaxv_s32(__VA_ARGS__)
+#define svmaxv_s64(...) __builtin_sve_svmaxv_s64(__VA_ARGS__)
+#define svmaxv_s16(...) __builtin_sve_svmaxv_s16(__VA_ARGS__)
+#define svmaxv_u8(...) __builtin_sve_svmaxv_u8(__VA_ARGS__)
+#define svmaxv_u32(...) __builtin_sve_svmaxv_u32(__VA_ARGS__)
+#define svmaxv_u64(...) __builtin_sve_svmaxv_u64(__VA_ARGS__)
+#define svmaxv_u16(...) __builtin_sve_svmaxv_u16(__VA_ARGS__)
+#define svmin_n_f64_m(...) __builtin_sve_svmin_n_f64_m(__VA_ARGS__)
+#define svmin_n_f32_m(...) __builtin_sve_svmin_n_f32_m(__VA_ARGS__)
+#define svmin_n_f16_m(...) __builtin_sve_svmin_n_f16_m(__VA_ARGS__)
+#define svmin_n_f64_x(...) __builtin_sve_svmin_n_f64_x(__VA_ARGS__)
+#define svmin_n_f32_x(...) __builtin_sve_svmin_n_f32_x(__VA_ARGS__)
+#define svmin_n_f16_x(...) __builtin_sve_svmin_n_f16_x(__VA_ARGS__)
+#define svmin_n_f64_z(...) __builtin_sve_svmin_n_f64_z(__VA_ARGS__)
+#define svmin_n_f32_z(...) __builtin_sve_svmin_n_f32_z(__VA_ARGS__)
+#define svmin_n_f16_z(...) __builtin_sve_svmin_n_f16_z(__VA_ARGS__)
+#define svmin_n_s8_m(...) __builtin_sve_svmin_n_s8_m(__VA_ARGS__)
+#define svmin_n_s32_m(...) __builtin_sve_svmin_n_s32_m(__VA_ARGS__)
+#define svmin_n_s64_m(...) __builtin_sve_svmin_n_s64_m(__VA_ARGS__)
+#define svmin_n_s16_m(...) __builtin_sve_svmin_n_s16_m(__VA_ARGS__)
+#define svmin_n_s8_x(...) __builtin_sve_svmin_n_s8_x(__VA_ARGS__)
+#define svmin_n_s32_x(...) __builtin_sve_svmin_n_s32_x(__VA_ARGS__)
+#define svmin_n_s64_x(...) __builtin_sve_svmin_n_s64_x(__VA_ARGS__)
+#define svmin_n_s16_x(...) __builtin_sve_svmin_n_s16_x(__VA_ARGS__)
+#define svmin_n_s8_z(...) __builtin_sve_svmin_n_s8_z(__VA_ARGS__)
+#define svmin_n_s32_z(...) __builtin_sve_svmin_n_s32_z(__VA_ARGS__)
+#define svmin_n_s64_z(...) __builtin_sve_svmin_n_s64_z(__VA_ARGS__)
+#define svmin_n_s16_z(...) __builtin_sve_svmin_n_s16_z(__VA_ARGS__)
+#define svmin_n_u8_m(...) __builtin_sve_svmin_n_u8_m(__VA_ARGS__)
+#define svmin_n_u32_m(...) __builtin_sve_svmin_n_u32_m(__VA_ARGS__)
+#define svmin_n_u64_m(...) __builtin_sve_svmin_n_u64_m(__VA_ARGS__)
+#define svmin_n_u16_m(...) __builtin_sve_svmin_n_u16_m(__VA_ARGS__)
+#define svmin_n_u8_x(...) __builtin_sve_svmin_n_u8_x(__VA_ARGS__)
+#define svmin_n_u32_x(...) __builtin_sve_svmin_n_u32_x(__VA_ARGS__)
+#define svmin_n_u64_x(...) __builtin_sve_svmin_n_u64_x(__VA_ARGS__)
+#define svmin_n_u16_x(...) __builtin_sve_svmin_n_u16_x(__VA_ARGS__)
+#define svmin_n_u8_z(...) __builtin_sve_svmin_n_u8_z(__VA_ARGS__)
+#define svmin_n_u32_z(...) __builtin_sve_svmin_n_u32_z(__VA_ARGS__)
+#define svmin_n_u64_z(...) __builtin_sve_svmin_n_u64_z(__VA_ARGS__)
+#define svmin_n_u16_z(...) __builtin_sve_svmin_n_u16_z(__VA_ARGS__)
+#define svmin_f64_m(...) __builtin_sve_svmin_f64_m(__VA_ARGS__)
+#define svmin_f32_m(...) __builtin_sve_svmin_f32_m(__VA_ARGS__)
+#define svmin_f16_m(...) __builtin_sve_svmin_f16_m(__VA_ARGS__)
+#define svmin_f64_x(...) __builtin_sve_svmin_f64_x(__VA_ARGS__)
+#define svmin_f32_x(...) __builtin_sve_svmin_f32_x(__VA_ARGS__)
+#define svmin_f16_x(...) __builtin_sve_svmin_f16_x(__VA_ARGS__)
+#define svmin_f64_z(...) __builtin_sve_svmin_f64_z(__VA_ARGS__)
+#define svmin_f32_z(...) __builtin_sve_svmin_f32_z(__VA_ARGS__)
+#define svmin_f16_z(...) __builtin_sve_svmin_f16_z(__VA_ARGS__)
+#define svmin_s8_m(...) __builtin_sve_svmin_s8_m(__VA_ARGS__)
+#define svmin_s32_m(...) __builtin_sve_svmin_s32_m(__VA_ARGS__)
+#define svmin_s64_m(...) __builtin_sve_svmin_s64_m(__VA_ARGS__)
+#define svmin_s16_m(...) __builtin_sve_svmin_s16_m(__VA_ARGS__)
+#define svmin_s8_x(...) __builtin_sve_svmin_s8_x(__VA_ARGS__)
+#define svmin_s32_x(...) __builtin_sve_svmin_s32_x(__VA_ARGS__)
+#define svmin_s64_x(...) __builtin_sve_svmin_s64_x(__VA_ARGS__)
+#define svmin_s16_x(...) __builtin_sve_svmin_s16_x(__VA_ARGS__)
+#define svmin_s8_z(...) __builtin_sve_svmin_s8_z(__VA_ARGS__)
+#define svmin_s32_z(...) __builtin_sve_svmin_s32_z(__VA_ARGS__)
+#define svmin_s64_z(...) __builtin_sve_svmin_s64_z(__VA_ARGS__)
+#define svmin_s16_z(...) __builtin_sve_svmin_s16_z(__VA_ARGS__)
+#define svmin_u8_m(...) __builtin_sve_svmin_u8_m(__VA_ARGS__)
+#define svmin_u32_m(...) __builtin_sve_svmin_u32_m(__VA_ARGS__)
+#define svmin_u64_m(...) __builtin_sve_svmin_u64_m(__VA_ARGS__)
+#define svmin_u16_m(...) __builtin_sve_svmin_u16_m(__VA_ARGS__)
+#define svmin_u8_x(...) __builtin_sve_svmin_u8_x(__VA_ARGS__)
+#define svmin_u32_x(...) __builtin_sve_svmin_u32_x(__VA_ARGS__)
+#define svmin_u64_x(...) __builtin_sve_svmin_u64_x(__VA_ARGS__)
+#define svmin_u16_x(...) __builtin_sve_svmin_u16_x(__VA_ARGS__)
+#define svmin_u8_z(...) __builtin_sve_svmin_u8_z(__VA_ARGS__)
+#define svmin_u32_z(...) __builtin_sve_svmin_u32_z(__VA_ARGS__)
+#define svmin_u64_z(...) __builtin_sve_svmin_u64_z(__VA_ARGS__)
+#define svmin_u16_z(...) __builtin_sve_svmin_u16_z(__VA_ARGS__)
+#define svminnm_n_f64_m(...) __builtin_sve_svminnm_n_f64_m(__VA_ARGS__)
+#define svminnm_n_f32_m(...) __builtin_sve_svminnm_n_f32_m(__VA_ARGS__)
+#define svminnm_n_f16_m(...) __builtin_sve_svminnm_n_f16_m(__VA_ARGS__)
+#define svminnm_n_f64_x(...) __builtin_sve_svminnm_n_f64_x(__VA_ARGS__)
+#define svminnm_n_f32_x(...) __builtin_sve_svminnm_n_f32_x(__VA_ARGS__)
+#define svminnm_n_f16_x(...) __builtin_sve_svminnm_n_f16_x(__VA_ARGS__)
+#define svminnm_n_f64_z(...) __builtin_sve_svminnm_n_f64_z(__VA_ARGS__)
+#define svminnm_n_f32_z(...) __builtin_sve_svminnm_n_f32_z(__VA_ARGS__)
+#define svminnm_n_f16_z(...) __builtin_sve_svminnm_n_f16_z(__VA_ARGS__)
+#define svminnm_f64_m(...) __builtin_sve_svminnm_f64_m(__VA_ARGS__)
+#define svminnm_f32_m(...) __builtin_sve_svminnm_f32_m(__VA_ARGS__)
+#define svminnm_f16_m(...) __builtin_sve_svminnm_f16_m(__VA_ARGS__)
+#define svminnm_f64_x(...) __builtin_sve_svminnm_f64_x(__VA_ARGS__)
+#define svminnm_f32_x(...) __builtin_sve_svminnm_f32_x(__VA_ARGS__)
+#define svminnm_f16_x(...) __builtin_sve_svminnm_f16_x(__VA_ARGS__)
+#define svminnm_f64_z(...) __builtin_sve_svminnm_f64_z(__VA_ARGS__)
+#define svminnm_f32_z(...) __builtin_sve_svminnm_f32_z(__VA_ARGS__)
+#define svminnm_f16_z(...) __builtin_sve_svminnm_f16_z(__VA_ARGS__)
+#define svminnmv_f64(...) __builtin_sve_svminnmv_f64(__VA_ARGS__)
+#define svminnmv_f32(...) __builtin_sve_svminnmv_f32(__VA_ARGS__)
+#define svminnmv_f16(...) __builtin_sve_svminnmv_f16(__VA_ARGS__)
+#define svminv_f64(...) __builtin_sve_svminv_f64(__VA_ARGS__)
+#define svminv_f32(...) __builtin_sve_svminv_f32(__VA_ARGS__)
+#define svminv_f16(...) __builtin_sve_svminv_f16(__VA_ARGS__)
+#define svminv_s8(...) __builtin_sve_svminv_s8(__VA_ARGS__)
+#define svminv_s32(...) __builtin_sve_svminv_s32(__VA_ARGS__)
+#define svminv_s64(...) __builtin_sve_svminv_s64(__VA_ARGS__)
+#define svminv_s16(...) __builtin_sve_svminv_s16(__VA_ARGS__)
+#define svminv_u8(...) __builtin_sve_svminv_u8(__VA_ARGS__)
+#define svminv_u32(...) __builtin_sve_svminv_u32(__VA_ARGS__)
+#define svminv_u64(...) __builtin_sve_svminv_u64(__VA_ARGS__)
+#define svminv_u16(...) __builtin_sve_svminv_u16(__VA_ARGS__)
+#define svmla_n_f64_m(...) __builtin_sve_svmla_n_f64_m(__VA_ARGS__)
+#define svmla_n_f32_m(...) __builtin_sve_svmla_n_f32_m(__VA_ARGS__)
+#define svmla_n_f16_m(...) __builtin_sve_svmla_n_f16_m(__VA_ARGS__)
+#define svmla_n_f64_x(...) __builtin_sve_svmla_n_f64_x(__VA_ARGS__)
+#define svmla_n_f32_x(...) __builtin_sve_svmla_n_f32_x(__VA_ARGS__)
+#define svmla_n_f16_x(...) __builtin_sve_svmla_n_f16_x(__VA_ARGS__)
+#define svmla_n_f64_z(...) __builtin_sve_svmla_n_f64_z(__VA_ARGS__)
+#define svmla_n_f32_z(...) __builtin_sve_svmla_n_f32_z(__VA_ARGS__)
+#define svmla_n_f16_z(...) __builtin_sve_svmla_n_f16_z(__VA_ARGS__)
+#define svmla_n_u8_m(...) __builtin_sve_svmla_n_u8_m(__VA_ARGS__)
+#define svmla_n_u32_m(...) __builtin_sve_svmla_n_u32_m(__VA_ARGS__)
+#define svmla_n_u64_m(...) __builtin_sve_svmla_n_u64_m(__VA_ARGS__)
+#define svmla_n_u16_m(...) __builtin_sve_svmla_n_u16_m(__VA_ARGS__)
+#define svmla_n_s8_m(...) __builtin_sve_svmla_n_s8_m(__VA_ARGS__)
+#define svmla_n_s32_m(...) __builtin_sve_svmla_n_s32_m(__VA_ARGS__)
+#define svmla_n_s64_m(...) __builtin_sve_svmla_n_s64_m(__VA_ARGS__)
+#define svmla_n_s16_m(...) __builtin_sve_svmla_n_s16_m(__VA_ARGS__)
+#define svmla_n_u8_x(...) __builtin_sve_svmla_n_u8_x(__VA_ARGS__)
+#define svmla_n_u32_x(...) __builtin_sve_svmla_n_u32_x(__VA_ARGS__)
+#define svmla_n_u64_x(...) __builtin_sve_svmla_n_u64_x(__VA_ARGS__)
+#define svmla_n_u16_x(...) __builtin_sve_svmla_n_u16_x(__VA_ARGS__)
+#define svmla_n_s8_x(...) __builtin_sve_svmla_n_s8_x(__VA_ARGS__)
+#define svmla_n_s32_x(...) __builtin_sve_svmla_n_s32_x(__VA_ARGS__)
+#define svmla_n_s64_x(...) __builtin_sve_svmla_n_s64_x(__VA_ARGS__)
+#define svmla_n_s16_x(...) __builtin_sve_svmla_n_s16_x(__VA_ARGS__)
+#define svmla_n_u8_z(...) __builtin_sve_svmla_n_u8_z(__VA_ARGS__)
+#define svmla_n_u32_z(...) __builtin_sve_svmla_n_u32_z(__VA_ARGS__)
+#define svmla_n_u64_z(...) __builtin_sve_svmla_n_u64_z(__VA_ARGS__)
+#define svmla_n_u16_z(...) __builtin_sve_svmla_n_u16_z(__VA_ARGS__)
+#define svmla_n_s8_z(...) __builtin_sve_svmla_n_s8_z(__VA_ARGS__)
+#define svmla_n_s32_z(...) __builtin_sve_svmla_n_s32_z(__VA_ARGS__)
+#define svmla_n_s64_z(...) __builtin_sve_svmla_n_s64_z(__VA_ARGS__)
+#define svmla_n_s16_z(...) __builtin_sve_svmla_n_s16_z(__VA_ARGS__)
+#define svmla_f64_m(...) __builtin_sve_svmla_f64_m(__VA_ARGS__)
+#define svmla_f32_m(...) __builtin_sve_svmla_f32_m(__VA_ARGS__)
+#define svmla_f16_m(...) __builtin_sve_svmla_f16_m(__VA_ARGS__)
+#define svmla_f64_x(...) __builtin_sve_svmla_f64_x(__VA_ARGS__)
+#define svmla_f32_x(...) __builtin_sve_svmla_f32_x(__VA_ARGS__)
+#define svmla_f16_x(...) __builtin_sve_svmla_f16_x(__VA_ARGS__)
+#define svmla_f64_z(...) __builtin_sve_svmla_f64_z(__VA_ARGS__)
+#define svmla_f32_z(...) __builtin_sve_svmla_f32_z(__VA_ARGS__)
+#define svmla_f16_z(...) __builtin_sve_svmla_f16_z(__VA_ARGS__)
+#define svmla_u8_m(...) __builtin_sve_svmla_u8_m(__VA_ARGS__)
+#define svmla_u32_m(...) __builtin_sve_svmla_u32_m(__VA_ARGS__)
+#define svmla_u64_m(...) __builtin_sve_svmla_u64_m(__VA_ARGS__)
+#define svmla_u16_m(...) __builtin_sve_svmla_u16_m(__VA_ARGS__)
+#define svmla_s8_m(...) __builtin_sve_svmla_s8_m(__VA_ARGS__)
+#define svmla_s32_m(...) __builtin_sve_svmla_s32_m(__VA_ARGS__)
+#define svmla_s64_m(...) __builtin_sve_svmla_s64_m(__VA_ARGS__)
+#define svmla_s16_m(...) __builtin_sve_svmla_s16_m(__VA_ARGS__)
+#define svmla_u8_x(...) __builtin_sve_svmla_u8_x(__VA_ARGS__)
+#define svmla_u32_x(...) __builtin_sve_svmla_u32_x(__VA_ARGS__)
+#define svmla_u64_x(...) __builtin_sve_svmla_u64_x(__VA_ARGS__)
+#define svmla_u16_x(...) __builtin_sve_svmla_u16_x(__VA_ARGS__)
+#define svmla_s8_x(...) __builtin_sve_svmla_s8_x(__VA_ARGS__)
+#define svmla_s32_x(...) __builtin_sve_svmla_s32_x(__VA_ARGS__)
+#define svmla_s64_x(...) __builtin_sve_svmla_s64_x(__VA_ARGS__)
+#define svmla_s16_x(...) __builtin_sve_svmla_s16_x(__VA_ARGS__)
+#define svmla_u8_z(...) __builtin_sve_svmla_u8_z(__VA_ARGS__)
+#define svmla_u32_z(...) __builtin_sve_svmla_u32_z(__VA_ARGS__)
+#define svmla_u64_z(...) __builtin_sve_svmla_u64_z(__VA_ARGS__)
+#define svmla_u16_z(...) __builtin_sve_svmla_u16_z(__VA_ARGS__)
+#define svmla_s8_z(...) __builtin_sve_svmla_s8_z(__VA_ARGS__)
+#define svmla_s32_z(...) __builtin_sve_svmla_s32_z(__VA_ARGS__)
+#define svmla_s64_z(...) __builtin_sve_svmla_s64_z(__VA_ARGS__)
+#define svmla_s16_z(...) __builtin_sve_svmla_s16_z(__VA_ARGS__)
+#define svmla_lane_f64(...) __builtin_sve_svmla_lane_f64(__VA_ARGS__)
+#define svmla_lane_f32(...) __builtin_sve_svmla_lane_f32(__VA_ARGS__)
+#define svmla_lane_f16(...) __builtin_sve_svmla_lane_f16(__VA_ARGS__)
+#define svmls_n_f64_m(...) __builtin_sve_svmls_n_f64_m(__VA_ARGS__)
+#define svmls_n_f32_m(...) __builtin_sve_svmls_n_f32_m(__VA_ARGS__)
+#define svmls_n_f16_m(...) __builtin_sve_svmls_n_f16_m(__VA_ARGS__)
+#define svmls_n_f64_x(...) __builtin_sve_svmls_n_f64_x(__VA_ARGS__)
+#define svmls_n_f32_x(...) __builtin_sve_svmls_n_f32_x(__VA_ARGS__)
+#define svmls_n_f16_x(...) __builtin_sve_svmls_n_f16_x(__VA_ARGS__)
+#define svmls_n_f64_z(...) __builtin_sve_svmls_n_f64_z(__VA_ARGS__)
+#define svmls_n_f32_z(...) __builtin_sve_svmls_n_f32_z(__VA_ARGS__)
+#define svmls_n_f16_z(...) __builtin_sve_svmls_n_f16_z(__VA_ARGS__)
+#define svmls_n_u8_m(...) __builtin_sve_svmls_n_u8_m(__VA_ARGS__)
+#define svmls_n_u32_m(...) __builtin_sve_svmls_n_u32_m(__VA_ARGS__)
+#define svmls_n_u64_m(...) __builtin_sve_svmls_n_u64_m(__VA_ARGS__)
+#define svmls_n_u16_m(...) __builtin_sve_svmls_n_u16_m(__VA_ARGS__)
+#define svmls_n_s8_m(...) __builtin_sve_svmls_n_s8_m(__VA_ARGS__)
+#define svmls_n_s32_m(...) __builtin_sve_svmls_n_s32_m(__VA_ARGS__)
+#define svmls_n_s64_m(...) __builtin_sve_svmls_n_s64_m(__VA_ARGS__)
+#define svmls_n_s16_m(...) __builtin_sve_svmls_n_s16_m(__VA_ARGS__)
+#define svmls_n_u8_x(...) __builtin_sve_svmls_n_u8_x(__VA_ARGS__)
+#define svmls_n_u32_x(...) __builtin_sve_svmls_n_u32_x(__VA_ARGS__)
+#define svmls_n_u64_x(...) __builtin_sve_svmls_n_u64_x(__VA_ARGS__)
+#define svmls_n_u16_x(...) __builtin_sve_svmls_n_u16_x(__VA_ARGS__)
+#define svmls_n_s8_x(...) __builtin_sve_svmls_n_s8_x(__VA_ARGS__)
+#define svmls_n_s32_x(...) __builtin_sve_svmls_n_s32_x(__VA_ARGS__)
+#define svmls_n_s64_x(...) __builtin_sve_svmls_n_s64_x(__VA_ARGS__)
+#define svmls_n_s16_x(...) __builtin_sve_svmls_n_s16_x(__VA_ARGS__)
+#define svmls_n_u8_z(...) __builtin_sve_svmls_n_u8_z(__VA_ARGS__)
+#define svmls_n_u32_z(...) __builtin_sve_svmls_n_u32_z(__VA_ARGS__)
+#define svmls_n_u64_z(...) __builtin_sve_svmls_n_u64_z(__VA_ARGS__)
+#define svmls_n_u16_z(...) __builtin_sve_svmls_n_u16_z(__VA_ARGS__)
+#define svmls_n_s8_z(...) __builtin_sve_svmls_n_s8_z(__VA_ARGS__)
+#define svmls_n_s32_z(...) __builtin_sve_svmls_n_s32_z(__VA_ARGS__)
+#define svmls_n_s64_z(...) __builtin_sve_svmls_n_s64_z(__VA_ARGS__)
+#define svmls_n_s16_z(...) __builtin_sve_svmls_n_s16_z(__VA_ARGS__)
+#define svmls_f64_m(...) __builtin_sve_svmls_f64_m(__VA_ARGS__)
+#define svmls_f32_m(...) __builtin_sve_svmls_f32_m(__VA_ARGS__)
+#define svmls_f16_m(...) __builtin_sve_svmls_f16_m(__VA_ARGS__)
+#define svmls_f64_x(...) __builtin_sve_svmls_f64_x(__VA_ARGS__)
+#define svmls_f32_x(...) __builtin_sve_svmls_f32_x(__VA_ARGS__)
+#define svmls_f16_x(...) __builtin_sve_svmls_f16_x(__VA_ARGS__)
+#define svmls_f64_z(...) __builtin_sve_svmls_f64_z(__VA_ARGS__)
+#define svmls_f32_z(...) __builtin_sve_svmls_f32_z(__VA_ARGS__)
+#define svmls_f16_z(...) __builtin_sve_svmls_f16_z(__VA_ARGS__)
+#define svmls_u8_m(...) __builtin_sve_svmls_u8_m(__VA_ARGS__)
+#define svmls_u32_m(...) __builtin_sve_svmls_u32_m(__VA_ARGS__)
+#define svmls_u64_m(...) __builtin_sve_svmls_u64_m(__VA_ARGS__)
+#define svmls_u16_m(...) __builtin_sve_svmls_u16_m(__VA_ARGS__)
+#define svmls_s8_m(...) __builtin_sve_svmls_s8_m(__VA_ARGS__)
+#define svmls_s32_m(...) __builtin_sve_svmls_s32_m(__VA_ARGS__)
+#define svmls_s64_m(...) __builtin_sve_svmls_s64_m(__VA_ARGS__)
+#define svmls_s16_m(...) __builtin_sve_svmls_s16_m(__VA_ARGS__)
+#define svmls_u8_x(...) __builtin_sve_svmls_u8_x(__VA_ARGS__)
+#define svmls_u32_x(...) __builtin_sve_svmls_u32_x(__VA_ARGS__)
+#define svmls_u64_x(...) __builtin_sve_svmls_u64_x(__VA_ARGS__)
+#define svmls_u16_x(...) __builtin_sve_svmls_u16_x(__VA_ARGS__)
+#define svmls_s8_x(...) __builtin_sve_svmls_s8_x(__VA_ARGS__)
+#define svmls_s32_x(...) __builtin_sve_svmls_s32_x(__VA_ARGS__)
+#define svmls_s64_x(...) __builtin_sve_svmls_s64_x(__VA_ARGS__)
+#define svmls_s16_x(...) __builtin_sve_svmls_s16_x(__VA_ARGS__)
+#define svmls_u8_z(...) __builtin_sve_svmls_u8_z(__VA_ARGS__)
+#define svmls_u32_z(...) __builtin_sve_svmls_u32_z(__VA_ARGS__)
+#define svmls_u64_z(...) __builtin_sve_svmls_u64_z(__VA_ARGS__)
+#define svmls_u16_z(...) __builtin_sve_svmls_u16_z(__VA_ARGS__)
+#define svmls_s8_z(...) __builtin_sve_svmls_s8_z(__VA_ARGS__)
+#define svmls_s32_z(...) __builtin_sve_svmls_s32_z(__VA_ARGS__)
+#define svmls_s64_z(...) __builtin_sve_svmls_s64_z(__VA_ARGS__)
+#define svmls_s16_z(...) __builtin_sve_svmls_s16_z(__VA_ARGS__)
+#define svmls_lane_f64(...) __builtin_sve_svmls_lane_f64(__VA_ARGS__)
+#define svmls_lane_f32(...) __builtin_sve_svmls_lane_f32(__VA_ARGS__)
+#define svmls_lane_f16(...) __builtin_sve_svmls_lane_f16(__VA_ARGS__)
+#define svmov_b_z(...) __builtin_sve_svmov_b_z(__VA_ARGS__)
+#define svmsb_n_f64_m(...) __builtin_sve_svmsb_n_f64_m(__VA_ARGS__)
+#define svmsb_n_f32_m(...) __builtin_sve_svmsb_n_f32_m(__VA_ARGS__)
+#define svmsb_n_f16_m(...) __builtin_sve_svmsb_n_f16_m(__VA_ARGS__)
+#define svmsb_n_f64_x(...) __builtin_sve_svmsb_n_f64_x(__VA_ARGS__)
+#define svmsb_n_f32_x(...) __builtin_sve_svmsb_n_f32_x(__VA_ARGS__)
+#define svmsb_n_f16_x(...) __builtin_sve_svmsb_n_f16_x(__VA_ARGS__)
+#define svmsb_n_f64_z(...) __builtin_sve_svmsb_n_f64_z(__VA_ARGS__)
+#define svmsb_n_f32_z(...) __builtin_sve_svmsb_n_f32_z(__VA_ARGS__)
+#define svmsb_n_f16_z(...) __builtin_sve_svmsb_n_f16_z(__VA_ARGS__)
+#define svmsb_n_u8_m(...) __builtin_sve_svmsb_n_u8_m(__VA_ARGS__)
+#define svmsb_n_u32_m(...) __builtin_sve_svmsb_n_u32_m(__VA_ARGS__)
+#define svmsb_n_u64_m(...) __builtin_sve_svmsb_n_u64_m(__VA_ARGS__)
+#define svmsb_n_u16_m(...) __builtin_sve_svmsb_n_u16_m(__VA_ARGS__)
+#define svmsb_n_s8_m(...) __builtin_sve_svmsb_n_s8_m(__VA_ARGS__)
+#define svmsb_n_s32_m(...) __builtin_sve_svmsb_n_s32_m(__VA_ARGS__)
+#define svmsb_n_s64_m(...) __builtin_sve_svmsb_n_s64_m(__VA_ARGS__)
+#define svmsb_n_s16_m(...) __builtin_sve_svmsb_n_s16_m(__VA_ARGS__)
+#define svmsb_n_u8_x(...) __builtin_sve_svmsb_n_u8_x(__VA_ARGS__)
+#define svmsb_n_u32_x(...) __builtin_sve_svmsb_n_u32_x(__VA_ARGS__)
+#define svmsb_n_u64_x(...) __builtin_sve_svmsb_n_u64_x(__VA_ARGS__)
+#define svmsb_n_u16_x(...) __builtin_sve_svmsb_n_u16_x(__VA_ARGS__)
+#define svmsb_n_s8_x(...) __builtin_sve_svmsb_n_s8_x(__VA_ARGS__)
+#define svmsb_n_s32_x(...) __builtin_sve_svmsb_n_s32_x(__VA_ARGS__)
+#define svmsb_n_s64_x(...) __builtin_sve_svmsb_n_s64_x(__VA_ARGS__)
+#define svmsb_n_s16_x(...) __builtin_sve_svmsb_n_s16_x(__VA_ARGS__)
+#define svmsb_n_u8_z(...) __builtin_sve_svmsb_n_u8_z(__VA_ARGS__)
+#define svmsb_n_u32_z(...) __builtin_sve_svmsb_n_u32_z(__VA_ARGS__)
+#define svmsb_n_u64_z(...) __builtin_sve_svmsb_n_u64_z(__VA_ARGS__)
+#define svmsb_n_u16_z(...) __builtin_sve_svmsb_n_u16_z(__VA_ARGS__)
+#define svmsb_n_s8_z(...) __builtin_sve_svmsb_n_s8_z(__VA_ARGS__)
+#define svmsb_n_s32_z(...) __builtin_sve_svmsb_n_s32_z(__VA_ARGS__)
+#define svmsb_n_s64_z(...) __builtin_sve_svmsb_n_s64_z(__VA_ARGS__)
+#define svmsb_n_s16_z(...) __builtin_sve_svmsb_n_s16_z(__VA_ARGS__)
+#define svmsb_f64_m(...) __builtin_sve_svmsb_f64_m(__VA_ARGS__)
+#define svmsb_f32_m(...) __builtin_sve_svmsb_f32_m(__VA_ARGS__)
+#define svmsb_f16_m(...) __builtin_sve_svmsb_f16_m(__VA_ARGS__)
+#define svmsb_f64_x(...) __builtin_sve_svmsb_f64_x(__VA_ARGS__)
+#define svmsb_f32_x(...) __builtin_sve_svmsb_f32_x(__VA_ARGS__)
+#define svmsb_f16_x(...) __builtin_sve_svmsb_f16_x(__VA_ARGS__)
+#define svmsb_f64_z(...) __builtin_sve_svmsb_f64_z(__VA_ARGS__)
+#define svmsb_f32_z(...) __builtin_sve_svmsb_f32_z(__VA_ARGS__)
+#define svmsb_f16_z(...) __builtin_sve_svmsb_f16_z(__VA_ARGS__)
+#define svmsb_u8_m(...) __builtin_sve_svmsb_u8_m(__VA_ARGS__)
+#define svmsb_u32_m(...) __builtin_sve_svmsb_u32_m(__VA_ARGS__)
+#define svmsb_u64_m(...) __builtin_sve_svmsb_u64_m(__VA_ARGS__)
+#define svmsb_u16_m(...) __builtin_sve_svmsb_u16_m(__VA_ARGS__)
+#define svmsb_s8_m(...) __builtin_sve_svmsb_s8_m(__VA_ARGS__)
+#define svmsb_s32_m(...) __builtin_sve_svmsb_s32_m(__VA_ARGS__)
+#define svmsb_s64_m(...) __builtin_sve_svmsb_s64_m(__VA_ARGS__)
+#define svmsb_s16_m(...) __builtin_sve_svmsb_s16_m(__VA_ARGS__)
+#define svmsb_u8_x(...) __builtin_sve_svmsb_u8_x(__VA_ARGS__)
+#define svmsb_u32_x(...) __builtin_sve_svmsb_u32_x(__VA_ARGS__)
+#define svmsb_u64_x(...) __builtin_sve_svmsb_u64_x(__VA_ARGS__)
+#define svmsb_u16_x(...) __builtin_sve_svmsb_u16_x(__VA_ARGS__)
+#define svmsb_s8_x(...) __builtin_sve_svmsb_s8_x(__VA_ARGS__)
+#define svmsb_s32_x(...) __builtin_sve_svmsb_s32_x(__VA_ARGS__)
+#define svmsb_s64_x(...) __builtin_sve_svmsb_s64_x(__VA_ARGS__)
+#define svmsb_s16_x(...) __builtin_sve_svmsb_s16_x(__VA_ARGS__)
+#define svmsb_u8_z(...) __builtin_sve_svmsb_u8_z(__VA_ARGS__)
+#define svmsb_u32_z(...) __builtin_sve_svmsb_u32_z(__VA_ARGS__)
+#define svmsb_u64_z(...) __builtin_sve_svmsb_u64_z(__VA_ARGS__)
+#define svmsb_u16_z(...) __builtin_sve_svmsb_u16_z(__VA_ARGS__)
+#define svmsb_s8_z(...) __builtin_sve_svmsb_s8_z(__VA_ARGS__)
+#define svmsb_s32_z(...) __builtin_sve_svmsb_s32_z(__VA_ARGS__)
+#define svmsb_s64_z(...) __builtin_sve_svmsb_s64_z(__VA_ARGS__)
+#define svmsb_s16_z(...) __builtin_sve_svmsb_s16_z(__VA_ARGS__)
+#define svmul_n_f64_m(...) __builtin_sve_svmul_n_f64_m(__VA_ARGS__)
+#define svmul_n_f32_m(...) __builtin_sve_svmul_n_f32_m(__VA_ARGS__)
+#define svmul_n_f16_m(...) __builtin_sve_svmul_n_f16_m(__VA_ARGS__)
+#define svmul_n_f64_x(...) __builtin_sve_svmul_n_f64_x(__VA_ARGS__)
+#define svmul_n_f32_x(...) __builtin_sve_svmul_n_f32_x(__VA_ARGS__)
+#define svmul_n_f16_x(...) __builtin_sve_svmul_n_f16_x(__VA_ARGS__)
+#define svmul_n_f64_z(...) __builtin_sve_svmul_n_f64_z(__VA_ARGS__)
+#define svmul_n_f32_z(...) __builtin_sve_svmul_n_f32_z(__VA_ARGS__)
+#define svmul_n_f16_z(...) __builtin_sve_svmul_n_f16_z(__VA_ARGS__)
+#define svmul_n_u8_m(...) __builtin_sve_svmul_n_u8_m(__VA_ARGS__)
+#define svmul_n_u32_m(...) __builtin_sve_svmul_n_u32_m(__VA_ARGS__)
+#define svmul_n_u64_m(...) __builtin_sve_svmul_n_u64_m(__VA_ARGS__)
+#define svmul_n_u16_m(...) __builtin_sve_svmul_n_u16_m(__VA_ARGS__)
+#define svmul_n_s8_m(...) __builtin_sve_svmul_n_s8_m(__VA_ARGS__)
+#define svmul_n_s32_m(...) __builtin_sve_svmul_n_s32_m(__VA_ARGS__)
+#define svmul_n_s64_m(...) __builtin_sve_svmul_n_s64_m(__VA_ARGS__)
+#define svmul_n_s16_m(...) __builtin_sve_svmul_n_s16_m(__VA_ARGS__)
+#define svmul_n_u8_x(...) __builtin_sve_svmul_n_u8_x(__VA_ARGS__)
+#define svmul_n_u32_x(...) __builtin_sve_svmul_n_u32_x(__VA_ARGS__)
+#define svmul_n_u64_x(...) __builtin_sve_svmul_n_u64_x(__VA_ARGS__)
+#define svmul_n_u16_x(...) __builtin_sve_svmul_n_u16_x(__VA_ARGS__)
+#define svmul_n_s8_x(...) __builtin_sve_svmul_n_s8_x(__VA_ARGS__)
+#define svmul_n_s32_x(...) __builtin_sve_svmul_n_s32_x(__VA_ARGS__)
+#define svmul_n_s64_x(...) __builtin_sve_svmul_n_s64_x(__VA_ARGS__)
+#define svmul_n_s16_x(...) __builtin_sve_svmul_n_s16_x(__VA_ARGS__)
+#define svmul_n_u8_z(...) __builtin_sve_svmul_n_u8_z(__VA_ARGS__)
+#define svmul_n_u32_z(...) __builtin_sve_svmul_n_u32_z(__VA_ARGS__)
+#define svmul_n_u64_z(...) __builtin_sve_svmul_n_u64_z(__VA_ARGS__)
+#define svmul_n_u16_z(...) __builtin_sve_svmul_n_u16_z(__VA_ARGS__)
+#define svmul_n_s8_z(...) __builtin_sve_svmul_n_s8_z(__VA_ARGS__)
+#define svmul_n_s32_z(...) __builtin_sve_svmul_n_s32_z(__VA_ARGS__)
+#define svmul_n_s64_z(...) __builtin_sve_svmul_n_s64_z(__VA_ARGS__)
+#define svmul_n_s16_z(...) __builtin_sve_svmul_n_s16_z(__VA_ARGS__)
+#define svmul_f64_m(...) __builtin_sve_svmul_f64_m(__VA_ARGS__)
+#define svmul_f32_m(...) __builtin_sve_svmul_f32_m(__VA_ARGS__)
+#define svmul_f16_m(...) __builtin_sve_svmul_f16_m(__VA_ARGS__)
+#define svmul_f64_x(...) __builtin_sve_svmul_f64_x(__VA_ARGS__)
+#define svmul_f32_x(...) __builtin_sve_svmul_f32_x(__VA_ARGS__)
+#define svmul_f16_x(...) __builtin_sve_svmul_f16_x(__VA_ARGS__)
+#define svmul_f64_z(...) __builtin_sve_svmul_f64_z(__VA_ARGS__)
+#define svmul_f32_z(...) __builtin_sve_svmul_f32_z(__VA_ARGS__)
+#define svmul_f16_z(...) __builtin_sve_svmul_f16_z(__VA_ARGS__)
+#define svmul_u8_m(...) __builtin_sve_svmul_u8_m(__VA_ARGS__)
+#define svmul_u32_m(...) __builtin_sve_svmul_u32_m(__VA_ARGS__)
+#define svmul_u64_m(...) __builtin_sve_svmul_u64_m(__VA_ARGS__)
+#define svmul_u16_m(...) __builtin_sve_svmul_u16_m(__VA_ARGS__)
+#define svmul_s8_m(...) __builtin_sve_svmul_s8_m(__VA_ARGS__)
+#define svmul_s32_m(...) __builtin_sve_svmul_s32_m(__VA_ARGS__)
+#define svmul_s64_m(...) __builtin_sve_svmul_s64_m(__VA_ARGS__)
+#define svmul_s16_m(...) __builtin_sve_svmul_s16_m(__VA_ARGS__)
+#define svmul_u8_x(...) __builtin_sve_svmul_u8_x(__VA_ARGS__)
+#define svmul_u32_x(...) __builtin_sve_svmul_u32_x(__VA_ARGS__)
+#define svmul_u64_x(...) __builtin_sve_svmul_u64_x(__VA_ARGS__)
+#define svmul_u16_x(...) __builtin_sve_svmul_u16_x(__VA_ARGS__)
+#define svmul_s8_x(...) __builtin_sve_svmul_s8_x(__VA_ARGS__)
+#define svmul_s32_x(...) __builtin_sve_svmul_s32_x(__VA_ARGS__)
+#define svmul_s64_x(...) __builtin_sve_svmul_s64_x(__VA_ARGS__)
+#define svmul_s16_x(...) __builtin_sve_svmul_s16_x(__VA_ARGS__)
+#define svmul_u8_z(...) __builtin_sve_svmul_u8_z(__VA_ARGS__)
+#define svmul_u32_z(...) __builtin_sve_svmul_u32_z(__VA_ARGS__)
+#define svmul_u64_z(...) __builtin_sve_svmul_u64_z(__VA_ARGS__)
+#define svmul_u16_z(...) __builtin_sve_svmul_u16_z(__VA_ARGS__)
+#define svmul_s8_z(...) __builtin_sve_svmul_s8_z(__VA_ARGS__)
+#define svmul_s32_z(...) __builtin_sve_svmul_s32_z(__VA_ARGS__)
+#define svmul_s64_z(...) __builtin_sve_svmul_s64_z(__VA_ARGS__)
+#define svmul_s16_z(...) __builtin_sve_svmul_s16_z(__VA_ARGS__)
+#define svmul_lane_f64(...) __builtin_sve_svmul_lane_f64(__VA_ARGS__)
+#define svmul_lane_f32(...) __builtin_sve_svmul_lane_f32(__VA_ARGS__)
+#define svmul_lane_f16(...) __builtin_sve_svmul_lane_f16(__VA_ARGS__)
+#define svmulh_n_s8_m(...) __builtin_sve_svmulh_n_s8_m(__VA_ARGS__)
+#define svmulh_n_s32_m(...) __builtin_sve_svmulh_n_s32_m(__VA_ARGS__)
+#define svmulh_n_s64_m(...) __builtin_sve_svmulh_n_s64_m(__VA_ARGS__)
+#define svmulh_n_s16_m(...) __builtin_sve_svmulh_n_s16_m(__VA_ARGS__)
+#define svmulh_n_s8_x(...) __builtin_sve_svmulh_n_s8_x(__VA_ARGS__)
+#define svmulh_n_s32_x(...) __builtin_sve_svmulh_n_s32_x(__VA_ARGS__)
+#define svmulh_n_s64_x(...) __builtin_sve_svmulh_n_s64_x(__VA_ARGS__)
+#define svmulh_n_s16_x(...) __builtin_sve_svmulh_n_s16_x(__VA_ARGS__)
+#define svmulh_n_s8_z(...) __builtin_sve_svmulh_n_s8_z(__VA_ARGS__)
+#define svmulh_n_s32_z(...) __builtin_sve_svmulh_n_s32_z(__VA_ARGS__)
+#define svmulh_n_s64_z(...) __builtin_sve_svmulh_n_s64_z(__VA_ARGS__)
+#define svmulh_n_s16_z(...) __builtin_sve_svmulh_n_s16_z(__VA_ARGS__)
+#define svmulh_n_u8_m(...) __builtin_sve_svmulh_n_u8_m(__VA_ARGS__)
+#define svmulh_n_u32_m(...) __builtin_sve_svmulh_n_u32_m(__VA_ARGS__)
+#define svmulh_n_u64_m(...) __builtin_sve_svmulh_n_u64_m(__VA_ARGS__)
+#define svmulh_n_u16_m(...) __builtin_sve_svmulh_n_u16_m(__VA_ARGS__)
+#define svmulh_n_u8_x(...) __builtin_sve_svmulh_n_u8_x(__VA_ARGS__)
+#define svmulh_n_u32_x(...) __builtin_sve_svmulh_n_u32_x(__VA_ARGS__)
+#define svmulh_n_u64_x(...) __builtin_sve_svmulh_n_u64_x(__VA_ARGS__)
+#define svmulh_n_u16_x(...) __builtin_sve_svmulh_n_u16_x(__VA_ARGS__)
+#define svmulh_n_u8_z(...) __builtin_sve_svmulh_n_u8_z(__VA_ARGS__)
+#define svmulh_n_u32_z(...) __builtin_sve_svmulh_n_u32_z(__VA_ARGS__)
+#define svmulh_n_u64_z(...) __builtin_sve_svmulh_n_u64_z(__VA_ARGS__)
+#define svmulh_n_u16_z(...) __builtin_sve_svmulh_n_u16_z(__VA_ARGS__)
+#define svmulh_s8_m(...) __builtin_sve_svmulh_s8_m(__VA_ARGS__)
+#define svmulh_s32_m(...) __builtin_sve_svmulh_s32_m(__VA_ARGS__)
+#define svmulh_s64_m(...) __builtin_sve_svmulh_s64_m(__VA_ARGS__)
+#define svmulh_s16_m(...) __builtin_sve_svmulh_s16_m(__VA_ARGS__)
+#define svmulh_s8_x(...) __builtin_sve_svmulh_s8_x(__VA_ARGS__)
+#define svmulh_s32_x(...) __builtin_sve_svmulh_s32_x(__VA_ARGS__)
+#define svmulh_s64_x(...) __builtin_sve_svmulh_s64_x(__VA_ARGS__)
+#define svmulh_s16_x(...) __builtin_sve_svmulh_s16_x(__VA_ARGS__)
+#define svmulh_s8_z(...) __builtin_sve_svmulh_s8_z(__VA_ARGS__)
+#define svmulh_s32_z(...) __builtin_sve_svmulh_s32_z(__VA_ARGS__)
+#define svmulh_s64_z(...) __builtin_sve_svmulh_s64_z(__VA_ARGS__)
+#define svmulh_s16_z(...) __builtin_sve_svmulh_s16_z(__VA_ARGS__)
+#define svmulh_u8_m(...) __builtin_sve_svmulh_u8_m(__VA_ARGS__)
+#define svmulh_u32_m(...) __builtin_sve_svmulh_u32_m(__VA_ARGS__)
+#define svmulh_u64_m(...) __builtin_sve_svmulh_u64_m(__VA_ARGS__)
+#define svmulh_u16_m(...) __builtin_sve_svmulh_u16_m(__VA_ARGS__)
+#define svmulh_u8_x(...) __builtin_sve_svmulh_u8_x(__VA_ARGS__)
+#define svmulh_u32_x(...) __builtin_sve_svmulh_u32_x(__VA_ARGS__)
+#define svmulh_u64_x(...) __builtin_sve_svmulh_u64_x(__VA_ARGS__)
+#define svmulh_u16_x(...) __builtin_sve_svmulh_u16_x(__VA_ARGS__)
+#define svmulh_u8_z(...) __builtin_sve_svmulh_u8_z(__VA_ARGS__)
+#define svmulh_u32_z(...) __builtin_sve_svmulh_u32_z(__VA_ARGS__)
+#define svmulh_u64_z(...) __builtin_sve_svmulh_u64_z(__VA_ARGS__)
+#define svmulh_u16_z(...) __builtin_sve_svmulh_u16_z(__VA_ARGS__)
+#define svmulx_n_f64_m(...) __builtin_sve_svmulx_n_f64_m(__VA_ARGS__)
+#define svmulx_n_f32_m(...) __builtin_sve_svmulx_n_f32_m(__VA_ARGS__)
+#define svmulx_n_f16_m(...) __builtin_sve_svmulx_n_f16_m(__VA_ARGS__)
+#define svmulx_n_f64_x(...) __builtin_sve_svmulx_n_f64_x(__VA_ARGS__)
+#define svmulx_n_f32_x(...) __builtin_sve_svmulx_n_f32_x(__VA_ARGS__)
+#define svmulx_n_f16_x(...) __builtin_sve_svmulx_n_f16_x(__VA_ARGS__)
+#define svmulx_n_f64_z(...) __builtin_sve_svmulx_n_f64_z(__VA_ARGS__)
+#define svmulx_n_f32_z(...) __builtin_sve_svmulx_n_f32_z(__VA_ARGS__)
+#define svmulx_n_f16_z(...) __builtin_sve_svmulx_n_f16_z(__VA_ARGS__)
+#define svmulx_f64_m(...) __builtin_sve_svmulx_f64_m(__VA_ARGS__)
+#define svmulx_f32_m(...) __builtin_sve_svmulx_f32_m(__VA_ARGS__)
+#define svmulx_f16_m(...) __builtin_sve_svmulx_f16_m(__VA_ARGS__)
+#define svmulx_f64_x(...) __builtin_sve_svmulx_f64_x(__VA_ARGS__)
+#define svmulx_f32_x(...) __builtin_sve_svmulx_f32_x(__VA_ARGS__)
+#define svmulx_f16_x(...) __builtin_sve_svmulx_f16_x(__VA_ARGS__)
+#define svmulx_f64_z(...) __builtin_sve_svmulx_f64_z(__VA_ARGS__)
+#define svmulx_f32_z(...) __builtin_sve_svmulx_f32_z(__VA_ARGS__)
+#define svmulx_f16_z(...) __builtin_sve_svmulx_f16_z(__VA_ARGS__)
+#define svnand_b_z(...) __builtin_sve_svnand_b_z(__VA_ARGS__)
+#define svneg_f64_m(...) __builtin_sve_svneg_f64_m(__VA_ARGS__)
+#define svneg_f32_m(...) __builtin_sve_svneg_f32_m(__VA_ARGS__)
+#define svneg_f16_m(...) __builtin_sve_svneg_f16_m(__VA_ARGS__)
+#define svneg_f64_x(...) __builtin_sve_svneg_f64_x(__VA_ARGS__)
+#define svneg_f32_x(...) __builtin_sve_svneg_f32_x(__VA_ARGS__)
+#define svneg_f16_x(...) __builtin_sve_svneg_f16_x(__VA_ARGS__)
+#define svneg_f64_z(...) __builtin_sve_svneg_f64_z(__VA_ARGS__)
+#define svneg_f32_z(...) __builtin_sve_svneg_f32_z(__VA_ARGS__)
+#define svneg_f16_z(...) __builtin_sve_svneg_f16_z(__VA_ARGS__)
+#define svneg_s8_m(...) __builtin_sve_svneg_s8_m(__VA_ARGS__)
+#define svneg_s32_m(...) __builtin_sve_svneg_s32_m(__VA_ARGS__)
+#define svneg_s64_m(...) __builtin_sve_svneg_s64_m(__VA_ARGS__)
+#define svneg_s16_m(...) __builtin_sve_svneg_s16_m(__VA_ARGS__)
+#define svneg_s8_x(...) __builtin_sve_svneg_s8_x(__VA_ARGS__)
+#define svneg_s32_x(...) __builtin_sve_svneg_s32_x(__VA_ARGS__)
+#define svneg_s64_x(...) __builtin_sve_svneg_s64_x(__VA_ARGS__)
+#define svneg_s16_x(...) __builtin_sve_svneg_s16_x(__VA_ARGS__)
+#define svneg_s8_z(...) __builtin_sve_svneg_s8_z(__VA_ARGS__)
+#define svneg_s32_z(...) __builtin_sve_svneg_s32_z(__VA_ARGS__)
+#define svneg_s64_z(...) __builtin_sve_svneg_s64_z(__VA_ARGS__)
+#define svneg_s16_z(...) __builtin_sve_svneg_s16_z(__VA_ARGS__)
+#define svnmad_n_f64_m(...) __builtin_sve_svnmad_n_f64_m(__VA_ARGS__)
+#define svnmad_n_f32_m(...) __builtin_sve_svnmad_n_f32_m(__VA_ARGS__)
+#define svnmad_n_f16_m(...) __builtin_sve_svnmad_n_f16_m(__VA_ARGS__)
+#define svnmad_n_f64_x(...) __builtin_sve_svnmad_n_f64_x(__VA_ARGS__)
+#define svnmad_n_f32_x(...) __builtin_sve_svnmad_n_f32_x(__VA_ARGS__)
+#define svnmad_n_f16_x(...) __builtin_sve_svnmad_n_f16_x(__VA_ARGS__)
+#define svnmad_n_f64_z(...) __builtin_sve_svnmad_n_f64_z(__VA_ARGS__)
+#define svnmad_n_f32_z(...) __builtin_sve_svnmad_n_f32_z(__VA_ARGS__)
+#define svnmad_n_f16_z(...) __builtin_sve_svnmad_n_f16_z(__VA_ARGS__)
+#define svnmad_f64_m(...) __builtin_sve_svnmad_f64_m(__VA_ARGS__)
+#define svnmad_f32_m(...) __builtin_sve_svnmad_f32_m(__VA_ARGS__)
+#define svnmad_f16_m(...) __builtin_sve_svnmad_f16_m(__VA_ARGS__)
+#define svnmad_f64_x(...) __builtin_sve_svnmad_f64_x(__VA_ARGS__)
+#define svnmad_f32_x(...) __builtin_sve_svnmad_f32_x(__VA_ARGS__)
+#define svnmad_f16_x(...) __builtin_sve_svnmad_f16_x(__VA_ARGS__)
+#define svnmad_f64_z(...) __builtin_sve_svnmad_f64_z(__VA_ARGS__)
+#define svnmad_f32_z(...) __builtin_sve_svnmad_f32_z(__VA_ARGS__)
+#define svnmad_f16_z(...) __builtin_sve_svnmad_f16_z(__VA_ARGS__)
+#define svnmla_n_f64_m(...) __builtin_sve_svnmla_n_f64_m(__VA_ARGS__)
+#define svnmla_n_f32_m(...) __builtin_sve_svnmla_n_f32_m(__VA_ARGS__)
+#define svnmla_n_f16_m(...) __builtin_sve_svnmla_n_f16_m(__VA_ARGS__)
+#define svnmla_n_f64_x(...) __builtin_sve_svnmla_n_f64_x(__VA_ARGS__)
+#define svnmla_n_f32_x(...) __builtin_sve_svnmla_n_f32_x(__VA_ARGS__)
+#define svnmla_n_f16_x(...) __builtin_sve_svnmla_n_f16_x(__VA_ARGS__)
+#define svnmla_n_f64_z(...) __builtin_sve_svnmla_n_f64_z(__VA_ARGS__)
+#define svnmla_n_f32_z(...) __builtin_sve_svnmla_n_f32_z(__VA_ARGS__)
+#define svnmla_n_f16_z(...) __builtin_sve_svnmla_n_f16_z(__VA_ARGS__)
+#define svnmla_f64_m(...) __builtin_sve_svnmla_f64_m(__VA_ARGS__)
+#define svnmla_f32_m(...) __builtin_sve_svnmla_f32_m(__VA_ARGS__)
+#define svnmla_f16_m(...) __builtin_sve_svnmla_f16_m(__VA_ARGS__)
+#define svnmla_f64_x(...) __builtin_sve_svnmla_f64_x(__VA_ARGS__)
+#define svnmla_f32_x(...) __builtin_sve_svnmla_f32_x(__VA_ARGS__)
+#define svnmla_f16_x(...) __builtin_sve_svnmla_f16_x(__VA_ARGS__)
+#define svnmla_f64_z(...) __builtin_sve_svnmla_f64_z(__VA_ARGS__)
+#define svnmla_f32_z(...) __builtin_sve_svnmla_f32_z(__VA_ARGS__)
+#define svnmla_f16_z(...) __builtin_sve_svnmla_f16_z(__VA_ARGS__)
+#define svnmls_n_f64_m(...) __builtin_sve_svnmls_n_f64_m(__VA_ARGS__)
+#define svnmls_n_f32_m(...) __builtin_sve_svnmls_n_f32_m(__VA_ARGS__)
+#define svnmls_n_f16_m(...) __builtin_sve_svnmls_n_f16_m(__VA_ARGS__)
+#define svnmls_n_f64_x(...) __builtin_sve_svnmls_n_f64_x(__VA_ARGS__)
+#define svnmls_n_f32_x(...) __builtin_sve_svnmls_n_f32_x(__VA_ARGS__)
+#define svnmls_n_f16_x(...) __builtin_sve_svnmls_n_f16_x(__VA_ARGS__)
+#define svnmls_n_f64_z(...) __builtin_sve_svnmls_n_f64_z(__VA_ARGS__)
+#define svnmls_n_f32_z(...) __builtin_sve_svnmls_n_f32_z(__VA_ARGS__)
+#define svnmls_n_f16_z(...) __builtin_sve_svnmls_n_f16_z(__VA_ARGS__)
+#define svnmls_f64_m(...) __builtin_sve_svnmls_f64_m(__VA_ARGS__)
+#define svnmls_f32_m(...) __builtin_sve_svnmls_f32_m(__VA_ARGS__)
+#define svnmls_f16_m(...) __builtin_sve_svnmls_f16_m(__VA_ARGS__)
+#define svnmls_f64_x(...) __builtin_sve_svnmls_f64_x(__VA_ARGS__)
+#define svnmls_f32_x(...) __builtin_sve_svnmls_f32_x(__VA_ARGS__)
+#define svnmls_f16_x(...) __builtin_sve_svnmls_f16_x(__VA_ARGS__)
+#define svnmls_f64_z(...) __builtin_sve_svnmls_f64_z(__VA_ARGS__)
+#define svnmls_f32_z(...) __builtin_sve_svnmls_f32_z(__VA_ARGS__)
+#define svnmls_f16_z(...) __builtin_sve_svnmls_f16_z(__VA_ARGS__)
+#define svnmsb_n_f64_m(...) __builtin_sve_svnmsb_n_f64_m(__VA_ARGS__)
+#define svnmsb_n_f32_m(...) __builtin_sve_svnmsb_n_f32_m(__VA_ARGS__)
+#define svnmsb_n_f16_m(...) __builtin_sve_svnmsb_n_f16_m(__VA_ARGS__)
+#define svnmsb_n_f64_x(...) __builtin_sve_svnmsb_n_f64_x(__VA_ARGS__)
+#define svnmsb_n_f32_x(...) __builtin_sve_svnmsb_n_f32_x(__VA_ARGS__)
+#define svnmsb_n_f16_x(...) __builtin_sve_svnmsb_n_f16_x(__VA_ARGS__)
+#define svnmsb_n_f64_z(...) __builtin_sve_svnmsb_n_f64_z(__VA_ARGS__)
+#define svnmsb_n_f32_z(...) __builtin_sve_svnmsb_n_f32_z(__VA_ARGS__)
+#define svnmsb_n_f16_z(...) __builtin_sve_svnmsb_n_f16_z(__VA_ARGS__)
+#define svnmsb_f64_m(...) __builtin_sve_svnmsb_f64_m(__VA_ARGS__)
+#define svnmsb_f32_m(...) __builtin_sve_svnmsb_f32_m(__VA_ARGS__)
+#define svnmsb_f16_m(...) __builtin_sve_svnmsb_f16_m(__VA_ARGS__)
+#define svnmsb_f64_x(...) __builtin_sve_svnmsb_f64_x(__VA_ARGS__)
+#define svnmsb_f32_x(...) __builtin_sve_svnmsb_f32_x(__VA_ARGS__)
+#define svnmsb_f16_x(...) __builtin_sve_svnmsb_f16_x(__VA_ARGS__)
+#define svnmsb_f64_z(...) __builtin_sve_svnmsb_f64_z(__VA_ARGS__)
+#define svnmsb_f32_z(...) __builtin_sve_svnmsb_f32_z(__VA_ARGS__)
+#define svnmsb_f16_z(...) __builtin_sve_svnmsb_f16_z(__VA_ARGS__)
+#define svnor_b_z(...) __builtin_sve_svnor_b_z(__VA_ARGS__)
+#define svnot_b_z(...) __builtin_sve_svnot_b_z(__VA_ARGS__)
+#define svnot_u8_m(...) __builtin_sve_svnot_u8_m(__VA_ARGS__)
+#define svnot_u32_m(...) __builtin_sve_svnot_u32_m(__VA_ARGS__)
+#define svnot_u64_m(...) __builtin_sve_svnot_u64_m(__VA_ARGS__)
+#define svnot_u16_m(...) __builtin_sve_svnot_u16_m(__VA_ARGS__)
+#define svnot_s8_m(...) __builtin_sve_svnot_s8_m(__VA_ARGS__)
+#define svnot_s32_m(...) __builtin_sve_svnot_s32_m(__VA_ARGS__)
+#define svnot_s64_m(...) __builtin_sve_svnot_s64_m(__VA_ARGS__)
+#define svnot_s16_m(...) __builtin_sve_svnot_s16_m(__VA_ARGS__)
+#define svnot_u8_x(...) __builtin_sve_svnot_u8_x(__VA_ARGS__)
+#define svnot_u32_x(...) __builtin_sve_svnot_u32_x(__VA_ARGS__)
+#define svnot_u64_x(...) __builtin_sve_svnot_u64_x(__VA_ARGS__)
+#define svnot_u16_x(...) __builtin_sve_svnot_u16_x(__VA_ARGS__)
+#define svnot_s8_x(...) __builtin_sve_svnot_s8_x(__VA_ARGS__)
+#define svnot_s32_x(...) __builtin_sve_svnot_s32_x(__VA_ARGS__)
+#define svnot_s64_x(...) __builtin_sve_svnot_s64_x(__VA_ARGS__)
+#define svnot_s16_x(...) __builtin_sve_svnot_s16_x(__VA_ARGS__)
+#define svnot_u8_z(...) __builtin_sve_svnot_u8_z(__VA_ARGS__)
+#define svnot_u32_z(...) __builtin_sve_svnot_u32_z(__VA_ARGS__)
+#define svnot_u64_z(...) __builtin_sve_svnot_u64_z(__VA_ARGS__)
+#define svnot_u16_z(...) __builtin_sve_svnot_u16_z(__VA_ARGS__)
+#define svnot_s8_z(...) __builtin_sve_svnot_s8_z(__VA_ARGS__)
+#define svnot_s32_z(...) __builtin_sve_svnot_s32_z(__VA_ARGS__)
+#define svnot_s64_z(...) __builtin_sve_svnot_s64_z(__VA_ARGS__)
+#define svnot_s16_z(...) __builtin_sve_svnot_s16_z(__VA_ARGS__)
+#define svorn_b_z(...) __builtin_sve_svorn_b_z(__VA_ARGS__)
+#define svorr_b_z(...) __builtin_sve_svorr_b_z(__VA_ARGS__)
+#define svorr_n_u8_m(...) __builtin_sve_svorr_n_u8_m(__VA_ARGS__)
+#define svorr_n_u32_m(...) __builtin_sve_svorr_n_u32_m(__VA_ARGS__)
+#define svorr_n_u64_m(...) __builtin_sve_svorr_n_u64_m(__VA_ARGS__)
+#define svorr_n_u16_m(...) __builtin_sve_svorr_n_u16_m(__VA_ARGS__)
+#define svorr_n_s8_m(...) __builtin_sve_svorr_n_s8_m(__VA_ARGS__)
+#define svorr_n_s32_m(...) __builtin_sve_svorr_n_s32_m(__VA_ARGS__)
+#define svorr_n_s64_m(...) __builtin_sve_svorr_n_s64_m(__VA_ARGS__)
+#define svorr_n_s16_m(...) __builtin_sve_svorr_n_s16_m(__VA_ARGS__)
+#define svorr_n_u8_x(...) __builtin_sve_svorr_n_u8_x(__VA_ARGS__)
+#define svorr_n_u32_x(...) __builtin_sve_svorr_n_u32_x(__VA_ARGS__)
+#define svorr_n_u64_x(...) __builtin_sve_svorr_n_u64_x(__VA_ARGS__)
+#define svorr_n_u16_x(...) __builtin_sve_svorr_n_u16_x(__VA_ARGS__)
+#define svorr_n_s8_x(...) __builtin_sve_svorr_n_s8_x(__VA_ARGS__)
+#define svorr_n_s32_x(...) __builtin_sve_svorr_n_s32_x(__VA_ARGS__)
+#define svorr_n_s64_x(...) __builtin_sve_svorr_n_s64_x(__VA_ARGS__)
+#define svorr_n_s16_x(...) __builtin_sve_svorr_n_s16_x(__VA_ARGS__)
+#define svorr_n_u8_z(...) __builtin_sve_svorr_n_u8_z(__VA_ARGS__)
+#define svorr_n_u32_z(...) __builtin_sve_svorr_n_u32_z(__VA_ARGS__)
+#define svorr_n_u64_z(...) __builtin_sve_svorr_n_u64_z(__VA_ARGS__)
+#define svorr_n_u16_z(...) __builtin_sve_svorr_n_u16_z(__VA_ARGS__)
+#define svorr_n_s8_z(...) __builtin_sve_svorr_n_s8_z(__VA_ARGS__)
+#define svorr_n_s32_z(...) __builtin_sve_svorr_n_s32_z(__VA_ARGS__)
+#define svorr_n_s64_z(...) __builtin_sve_svorr_n_s64_z(__VA_ARGS__)
+#define svorr_n_s16_z(...) __builtin_sve_svorr_n_s16_z(__VA_ARGS__)
+#define svorr_u8_m(...) __builtin_sve_svorr_u8_m(__VA_ARGS__)
+#define svorr_u32_m(...) __builtin_sve_svorr_u32_m(__VA_ARGS__)
+#define svorr_u64_m(...) __builtin_sve_svorr_u64_m(__VA_ARGS__)
+#define svorr_u16_m(...) __builtin_sve_svorr_u16_m(__VA_ARGS__)
+#define svorr_s8_m(...) __builtin_sve_svorr_s8_m(__VA_ARGS__)
+#define svorr_s32_m(...) __builtin_sve_svorr_s32_m(__VA_ARGS__)
+#define svorr_s64_m(...) __builtin_sve_svorr_s64_m(__VA_ARGS__)
+#define svorr_s16_m(...) __builtin_sve_svorr_s16_m(__VA_ARGS__)
+#define svorr_u8_x(...) __builtin_sve_svorr_u8_x(__VA_ARGS__)
+#define svorr_u32_x(...) __builtin_sve_svorr_u32_x(__VA_ARGS__)
+#define svorr_u64_x(...) __builtin_sve_svorr_u64_x(__VA_ARGS__)
+#define svorr_u16_x(...) __builtin_sve_svorr_u16_x(__VA_ARGS__)
+#define svorr_s8_x(...) __builtin_sve_svorr_s8_x(__VA_ARGS__)
+#define svorr_s32_x(...) __builtin_sve_svorr_s32_x(__VA_ARGS__)
+#define svorr_s64_x(...) __builtin_sve_svorr_s64_x(__VA_ARGS__)
+#define svorr_s16_x(...) __builtin_sve_svorr_s16_x(__VA_ARGS__)
+#define svorr_u8_z(...) __builtin_sve_svorr_u8_z(__VA_ARGS__)
+#define svorr_u32_z(...) __builtin_sve_svorr_u32_z(__VA_ARGS__)
+#define svorr_u64_z(...) __builtin_sve_svorr_u64_z(__VA_ARGS__)
+#define svorr_u16_z(...) __builtin_sve_svorr_u16_z(__VA_ARGS__)
+#define svorr_s8_z(...) __builtin_sve_svorr_s8_z(__VA_ARGS__)
+#define svorr_s32_z(...) __builtin_sve_svorr_s32_z(__VA_ARGS__)
+#define svorr_s64_z(...) __builtin_sve_svorr_s64_z(__VA_ARGS__)
+#define svorr_s16_z(...) __builtin_sve_svorr_s16_z(__VA_ARGS__)
+#define svorv_u8(...) __builtin_sve_svorv_u8(__VA_ARGS__)
+#define svorv_u32(...) __builtin_sve_svorv_u32(__VA_ARGS__)
+#define svorv_u64(...) __builtin_sve_svorv_u64(__VA_ARGS__)
+#define svorv_u16(...) __builtin_sve_svorv_u16(__VA_ARGS__)
+#define svorv_s8(...) __builtin_sve_svorv_s8(__VA_ARGS__)
+#define svorv_s32(...) __builtin_sve_svorv_s32(__VA_ARGS__)
+#define svorv_s64(...) __builtin_sve_svorv_s64(__VA_ARGS__)
+#define svorv_s16(...) __builtin_sve_svorv_s16(__VA_ARGS__)
+#define svpfalse_b(...) __builtin_sve_svpfalse_b(__VA_ARGS__)
+#define svpfirst_b(...) __builtin_sve_svpfirst_b(__VA_ARGS__)
+#define svpnext_b8(...) __builtin_sve_svpnext_b8(__VA_ARGS__)
+#define svpnext_b32(...) __builtin_sve_svpnext_b32(__VA_ARGS__)
+#define svpnext_b64(...) __builtin_sve_svpnext_b64(__VA_ARGS__)
+#define svpnext_b16(...) __builtin_sve_svpnext_b16(__VA_ARGS__)
+#define svprfb(...) __builtin_sve_svprfb(__VA_ARGS__)
+#define svprfb_gather_u32base(...) __builtin_sve_svprfb_gather_u32base(__VA_ARGS__)
+#define svprfb_gather_u64base(...) __builtin_sve_svprfb_gather_u64base(__VA_ARGS__)
+#define svprfb_gather_u32base_offset(...) __builtin_sve_svprfb_gather_u32base_offset(__VA_ARGS__)
+#define svprfb_gather_u64base_offset(...) __builtin_sve_svprfb_gather_u64base_offset(__VA_ARGS__)
+#define svprfb_gather_s32offset(...) __builtin_sve_svprfb_gather_s32offset(__VA_ARGS__)
+#define svprfb_gather_u32offset(...) __builtin_sve_svprfb_gather_u32offset(__VA_ARGS__)
+#define svprfb_gather_s64offset(...) __builtin_sve_svprfb_gather_s64offset(__VA_ARGS__)
+#define svprfb_gather_u64offset(...) __builtin_sve_svprfb_gather_u64offset(__VA_ARGS__)
+#define svprfb_vnum(...) __builtin_sve_svprfb_vnum(__VA_ARGS__)
+#define svprfd(...) __builtin_sve_svprfd(__VA_ARGS__)
+#define svprfd_gather_u32base(...) __builtin_sve_svprfd_gather_u32base(__VA_ARGS__)
+#define svprfd_gather_u64base(...) __builtin_sve_svprfd_gather_u64base(__VA_ARGS__)
+#define svprfd_gather_u32base_index(...) __builtin_sve_svprfd_gather_u32base_index(__VA_ARGS__)
+#define svprfd_gather_u64base_index(...) __builtin_sve_svprfd_gather_u64base_index(__VA_ARGS__)
+#define svprfd_gather_s32index(...) __builtin_sve_svprfd_gather_s32index(__VA_ARGS__)
+#define svprfd_gather_u32index(...) __builtin_sve_svprfd_gather_u32index(__VA_ARGS__)
+#define svprfd_gather_s64index(...) __builtin_sve_svprfd_gather_s64index(__VA_ARGS__)
+#define svprfd_gather_u64index(...) __builtin_sve_svprfd_gather_u64index(__VA_ARGS__)
+#define svprfd_vnum(...) __builtin_sve_svprfd_vnum(__VA_ARGS__)
+#define svprfh(...) __builtin_sve_svprfh(__VA_ARGS__)
+#define svprfh_gather_u32base(...) __builtin_sve_svprfh_gather_u32base(__VA_ARGS__)
+#define svprfh_gather_u64base(...) __builtin_sve_svprfh_gather_u64base(__VA_ARGS__)
+#define svprfh_gather_u32base_index(...) __builtin_sve_svprfh_gather_u32base_index(__VA_ARGS__)
+#define svprfh_gather_u64base_index(...) __builtin_sve_svprfh_gather_u64base_index(__VA_ARGS__)
+#define svprfh_gather_s32index(...) __builtin_sve_svprfh_gather_s32index(__VA_ARGS__)
+#define svprfh_gather_u32index(...) __builtin_sve_svprfh_gather_u32index(__VA_ARGS__)
+#define svprfh_gather_s64index(...) __builtin_sve_svprfh_gather_s64index(__VA_ARGS__)
+#define svprfh_gather_u64index(...) __builtin_sve_svprfh_gather_u64index(__VA_ARGS__)
+#define svprfh_vnum(...) __builtin_sve_svprfh_vnum(__VA_ARGS__)
+#define svprfw(...) __builtin_sve_svprfw(__VA_ARGS__)
+#define svprfw_gather_u32base(...) __builtin_sve_svprfw_gather_u32base(__VA_ARGS__)
+#define svprfw_gather_u64base(...) __builtin_sve_svprfw_gather_u64base(__VA_ARGS__)
+#define svprfw_gather_u32base_index(...) __builtin_sve_svprfw_gather_u32base_index(__VA_ARGS__)
+#define svprfw_gather_u64base_index(...) __builtin_sve_svprfw_gather_u64base_index(__VA_ARGS__)
+#define svprfw_gather_s32index(...) __builtin_sve_svprfw_gather_s32index(__VA_ARGS__)
+#define svprfw_gather_u32index(...) __builtin_sve_svprfw_gather_u32index(__VA_ARGS__)
+#define svprfw_gather_s64index(...) __builtin_sve_svprfw_gather_s64index(__VA_ARGS__)
+#define svprfw_gather_u64index(...) __builtin_sve_svprfw_gather_u64index(__VA_ARGS__)
+#define svprfw_vnum(...) __builtin_sve_svprfw_vnum(__VA_ARGS__)
+#define svptest_any(...) __builtin_sve_svptest_any(__VA_ARGS__)
+#define svptest_first(...) __builtin_sve_svptest_first(__VA_ARGS__)
+#define svptest_last(...) __builtin_sve_svptest_last(__VA_ARGS__)
+#define svptrue_pat_b8(...) __builtin_sve_svptrue_pat_b8(__VA_ARGS__)
+#define svptrue_pat_b32(...) __builtin_sve_svptrue_pat_b32(__VA_ARGS__)
+#define svptrue_pat_b64(...) __builtin_sve_svptrue_pat_b64(__VA_ARGS__)
+#define svptrue_pat_b16(...) __builtin_sve_svptrue_pat_b16(__VA_ARGS__)
+#define svptrue_b8(...) __builtin_sve_svptrue_b8(__VA_ARGS__)
+#define svptrue_b32(...) __builtin_sve_svptrue_b32(__VA_ARGS__)
+#define svptrue_b64(...) __builtin_sve_svptrue_b64(__VA_ARGS__)
+#define svptrue_b16(...) __builtin_sve_svptrue_b16(__VA_ARGS__)
+#define svqadd_n_s8(...) __builtin_sve_svqadd_n_s8(__VA_ARGS__)
+#define svqadd_n_s32(...) __builtin_sve_svqadd_n_s32(__VA_ARGS__)
+#define svqadd_n_s64(...) __builtin_sve_svqadd_n_s64(__VA_ARGS__)
+#define svqadd_n_s16(...) __builtin_sve_svqadd_n_s16(__VA_ARGS__)
+#define svqadd_n_u8(...) __builtin_sve_svqadd_n_u8(__VA_ARGS__)
+#define svqadd_n_u32(...) __builtin_sve_svqadd_n_u32(__VA_ARGS__)
+#define svqadd_n_u64(...) __builtin_sve_svqadd_n_u64(__VA_ARGS__)
+#define svqadd_n_u16(...) __builtin_sve_svqadd_n_u16(__VA_ARGS__)
+#define svqadd_s8(...) __builtin_sve_svqadd_s8(__VA_ARGS__)
+#define svqadd_s32(...) __builtin_sve_svqadd_s32(__VA_ARGS__)
+#define svqadd_s64(...) __builtin_sve_svqadd_s64(__VA_ARGS__)
+#define svqadd_s16(...) __builtin_sve_svqadd_s16(__VA_ARGS__)
+#define svqadd_u8(...) __builtin_sve_svqadd_u8(__VA_ARGS__)
+#define svqadd_u32(...) __builtin_sve_svqadd_u32(__VA_ARGS__)
+#define svqadd_u64(...) __builtin_sve_svqadd_u64(__VA_ARGS__)
+#define svqadd_u16(...) __builtin_sve_svqadd_u16(__VA_ARGS__)
+#define svqdecb_n_s32(...) __builtin_sve_svqdecb_n_s32(__VA_ARGS__)
+#define svqdecb_n_s64(...) __builtin_sve_svqdecb_n_s64(__VA_ARGS__)
+#define svqdecb_n_u32(...) __builtin_sve_svqdecb_n_u32(__VA_ARGS__)
+#define svqdecb_n_u64(...) __builtin_sve_svqdecb_n_u64(__VA_ARGS__)
+#define svqdecb_pat_n_s32(...) __builtin_sve_svqdecb_pat_n_s32(__VA_ARGS__)
+#define svqdecb_pat_n_s64(...) __builtin_sve_svqdecb_pat_n_s64(__VA_ARGS__)
+#define svqdecb_pat_n_u32(...) __builtin_sve_svqdecb_pat_n_u32(__VA_ARGS__)
+#define svqdecb_pat_n_u64(...) __builtin_sve_svqdecb_pat_n_u64(__VA_ARGS__)
+#define svqdecd_n_s32(...) __builtin_sve_svqdecd_n_s32(__VA_ARGS__)
+#define svqdecd_n_s64(...) __builtin_sve_svqdecd_n_s64(__VA_ARGS__)
+#define svqdecd_n_u32(...) __builtin_sve_svqdecd_n_u32(__VA_ARGS__)
+#define svqdecd_n_u64(...) __builtin_sve_svqdecd_n_u64(__VA_ARGS__)
+#define svqdecd_s64(...) __builtin_sve_svqdecd_s64(__VA_ARGS__)
+#define svqdecd_u64(...) __builtin_sve_svqdecd_u64(__VA_ARGS__)
+#define svqdecd_pat_n_s32(...) __builtin_sve_svqdecd_pat_n_s32(__VA_ARGS__)
+#define svqdecd_pat_n_s64(...) __builtin_sve_svqdecd_pat_n_s64(__VA_ARGS__)
+#define svqdecd_pat_n_u32(...) __builtin_sve_svqdecd_pat_n_u32(__VA_ARGS__)
+#define svqdecd_pat_n_u64(...) __builtin_sve_svqdecd_pat_n_u64(__VA_ARGS__)
+#define svqdecd_pat_s64(...) __builtin_sve_svqdecd_pat_s64(__VA_ARGS__)
+#define svqdecd_pat_u64(...) __builtin_sve_svqdecd_pat_u64(__VA_ARGS__)
+#define svqdech_n_s32(...) __builtin_sve_svqdech_n_s32(__VA_ARGS__)
+#define svqdech_n_s64(...) __builtin_sve_svqdech_n_s64(__VA_ARGS__)
+#define svqdech_n_u32(...) __builtin_sve_svqdech_n_u32(__VA_ARGS__)
+#define svqdech_n_u64(...) __builtin_sve_svqdech_n_u64(__VA_ARGS__)
+#define svqdech_s16(...) __builtin_sve_svqdech_s16(__VA_ARGS__)
+#define svqdech_u16(...) __builtin_sve_svqdech_u16(__VA_ARGS__)
+#define svqdech_pat_n_s32(...) __builtin_sve_svqdech_pat_n_s32(__VA_ARGS__)
+#define svqdech_pat_n_s64(...) __builtin_sve_svqdech_pat_n_s64(__VA_ARGS__)
+#define svqdech_pat_n_u32(...) __builtin_sve_svqdech_pat_n_u32(__VA_ARGS__)
+#define svqdech_pat_n_u64(...) __builtin_sve_svqdech_pat_n_u64(__VA_ARGS__)
+#define svqdech_pat_s16(...) __builtin_sve_svqdech_pat_s16(__VA_ARGS__)
+#define svqdech_pat_u16(...) __builtin_sve_svqdech_pat_u16(__VA_ARGS__)
+#define svqdecp_n_s32_b8(...) __builtin_sve_svqdecp_n_s32_b8(__VA_ARGS__)
+#define svqdecp_n_s32_b32(...) __builtin_sve_svqdecp_n_s32_b32(__VA_ARGS__)
+#define svqdecp_n_s32_b64(...) __builtin_sve_svqdecp_n_s32_b64(__VA_ARGS__)
+#define svqdecp_n_s32_b16(...) __builtin_sve_svqdecp_n_s32_b16(__VA_ARGS__)
+#define svqdecp_n_s64_b8(...) __builtin_sve_svqdecp_n_s64_b8(__VA_ARGS__)
+#define svqdecp_n_s64_b32(...) __builtin_sve_svqdecp_n_s64_b32(__VA_ARGS__)
+#define svqdecp_n_s64_b64(...) __builtin_sve_svqdecp_n_s64_b64(__VA_ARGS__)
+#define svqdecp_n_s64_b16(...) __builtin_sve_svqdecp_n_s64_b16(__VA_ARGS__)
+#define svqdecp_n_u32_b8(...) __builtin_sve_svqdecp_n_u32_b8(__VA_ARGS__)
+#define svqdecp_n_u32_b32(...) __builtin_sve_svqdecp_n_u32_b32(__VA_ARGS__)
+#define svqdecp_n_u32_b64(...) __builtin_sve_svqdecp_n_u32_b64(__VA_ARGS__)
+#define svqdecp_n_u32_b16(...) __builtin_sve_svqdecp_n_u32_b16(__VA_ARGS__)
+#define svqdecp_n_u64_b8(...) __builtin_sve_svqdecp_n_u64_b8(__VA_ARGS__)
+#define svqdecp_n_u64_b32(...) __builtin_sve_svqdecp_n_u64_b32(__VA_ARGS__)
+#define svqdecp_n_u64_b64(...) __builtin_sve_svqdecp_n_u64_b64(__VA_ARGS__)
+#define svqdecp_n_u64_b16(...) __builtin_sve_svqdecp_n_u64_b16(__VA_ARGS__)
+#define svqdecp_s32(...) __builtin_sve_svqdecp_s32(__VA_ARGS__)
+#define svqdecp_s64(...) __builtin_sve_svqdecp_s64(__VA_ARGS__)
+#define svqdecp_s16(...) __builtin_sve_svqdecp_s16(__VA_ARGS__)
+#define svqdecp_u32(...) __builtin_sve_svqdecp_u32(__VA_ARGS__)
+#define svqdecp_u64(...) __builtin_sve_svqdecp_u64(__VA_ARGS__)
+#define svqdecp_u16(...) __builtin_sve_svqdecp_u16(__VA_ARGS__)
+#define svqdecw_n_s32(...) __builtin_sve_svqdecw_n_s32(__VA_ARGS__)
+#define svqdecw_n_s64(...) __builtin_sve_svqdecw_n_s64(__VA_ARGS__)
+#define svqdecw_n_u32(...) __builtin_sve_svqdecw_n_u32(__VA_ARGS__)
+#define svqdecw_n_u64(...) __builtin_sve_svqdecw_n_u64(__VA_ARGS__)
+#define svqdecw_s32(...) __builtin_sve_svqdecw_s32(__VA_ARGS__)
+#define svqdecw_u32(...) __builtin_sve_svqdecw_u32(__VA_ARGS__)
+#define svqdecw_pat_n_s32(...) __builtin_sve_svqdecw_pat_n_s32(__VA_ARGS__)
+#define svqdecw_pat_n_s64(...) __builtin_sve_svqdecw_pat_n_s64(__VA_ARGS__)
+#define svqdecw_pat_n_u32(...) __builtin_sve_svqdecw_pat_n_u32(__VA_ARGS__)
+#define svqdecw_pat_n_u64(...) __builtin_sve_svqdecw_pat_n_u64(__VA_ARGS__)
+#define svqdecw_pat_s32(...) __builtin_sve_svqdecw_pat_s32(__VA_ARGS__)
+#define svqdecw_pat_u32(...) __builtin_sve_svqdecw_pat_u32(__VA_ARGS__)
+#define svqincb_n_s32(...) __builtin_sve_svqincb_n_s32(__VA_ARGS__)
+#define svqincb_n_s64(...) __builtin_sve_svqincb_n_s64(__VA_ARGS__)
+#define svqincb_n_u32(...) __builtin_sve_svqincb_n_u32(__VA_ARGS__)
+#define svqincb_n_u64(...) __builtin_sve_svqincb_n_u64(__VA_ARGS__)
+#define svqincb_pat_n_s32(...) __builtin_sve_svqincb_pat_n_s32(__VA_ARGS__)
+#define svqincb_pat_n_s64(...) __builtin_sve_svqincb_pat_n_s64(__VA_ARGS__)
+#define svqincb_pat_n_u32(...) __builtin_sve_svqincb_pat_n_u32(__VA_ARGS__)
+#define svqincb_pat_n_u64(...) __builtin_sve_svqincb_pat_n_u64(__VA_ARGS__)
+#define svqincd_n_s32(...) __builtin_sve_svqincd_n_s32(__VA_ARGS__)
+#define svqincd_n_s64(...) __builtin_sve_svqincd_n_s64(__VA_ARGS__)
+#define svqincd_n_u32(...) __builtin_sve_svqincd_n_u32(__VA_ARGS__)
+#define svqincd_n_u64(...) __builtin_sve_svqincd_n_u64(__VA_ARGS__)
+#define svqincd_s64(...) __builtin_sve_svqincd_s64(__VA_ARGS__)
+#define svqincd_u64(...) __builtin_sve_svqincd_u64(__VA_ARGS__)
+#define svqincd_pat_n_s32(...) __builtin_sve_svqincd_pat_n_s32(__VA_ARGS__)
+#define svqincd_pat_n_s64(...) __builtin_sve_svqincd_pat_n_s64(__VA_ARGS__)
+#define svqincd_pat_n_u32(...) __builtin_sve_svqincd_pat_n_u32(__VA_ARGS__)
+#define svqincd_pat_n_u64(...) __builtin_sve_svqincd_pat_n_u64(__VA_ARGS__)
+#define svqincd_pat_s64(...) __builtin_sve_svqincd_pat_s64(__VA_ARGS__)
+#define svqincd_pat_u64(...) __builtin_sve_svqincd_pat_u64(__VA_ARGS__)
+#define svqinch_n_s32(...) __builtin_sve_svqinch_n_s32(__VA_ARGS__)
+#define svqinch_n_s64(...) __builtin_sve_svqinch_n_s64(__VA_ARGS__)
+#define svqinch_n_u32(...) __builtin_sve_svqinch_n_u32(__VA_ARGS__)
+#define svqinch_n_u64(...) __builtin_sve_svqinch_n_u64(__VA_ARGS__)
+#define svqinch_s16(...) __builtin_sve_svqinch_s16(__VA_ARGS__)
+#define svqinch_u16(...) __builtin_sve_svqinch_u16(__VA_ARGS__)
+#define svqinch_pat_n_s32(...) __builtin_sve_svqinch_pat_n_s32(__VA_ARGS__)
+#define svqinch_pat_n_s64(...) __builtin_sve_svqinch_pat_n_s64(__VA_ARGS__)
+#define svqinch_pat_n_u32(...) __builtin_sve_svqinch_pat_n_u32(__VA_ARGS__)
+#define svqinch_pat_n_u64(...) __builtin_sve_svqinch_pat_n_u64(__VA_ARGS__)
+#define svqinch_pat_s16(...) __builtin_sve_svqinch_pat_s16(__VA_ARGS__)
+#define svqinch_pat_u16(...) __builtin_sve_svqinch_pat_u16(__VA_ARGS__)
+#define svqincp_n_s32_b8(...) __builtin_sve_svqincp_n_s32_b8(__VA_ARGS__)
+#define svqincp_n_s32_b32(...) __builtin_sve_svqincp_n_s32_b32(__VA_ARGS__)
+#define svqincp_n_s32_b64(...) __builtin_sve_svqincp_n_s32_b64(__VA_ARGS__)
+#define svqincp_n_s32_b16(...) __builtin_sve_svqincp_n_s32_b16(__VA_ARGS__)
+#define svqincp_n_s64_b8(...) __builtin_sve_svqincp_n_s64_b8(__VA_ARGS__)
+#define svqincp_n_s64_b32(...) __builtin_sve_svqincp_n_s64_b32(__VA_ARGS__)
+#define svqincp_n_s64_b64(...) __builtin_sve_svqincp_n_s64_b64(__VA_ARGS__)
+#define svqincp_n_s64_b16(...) __builtin_sve_svqincp_n_s64_b16(__VA_ARGS__)
+#define svqincp_n_u32_b8(...) __builtin_sve_svqincp_n_u32_b8(__VA_ARGS__)
+#define svqincp_n_u32_b32(...) __builtin_sve_svqincp_n_u32_b32(__VA_ARGS__)
+#define svqincp_n_u32_b64(...) __builtin_sve_svqincp_n_u32_b64(__VA_ARGS__)
+#define svqincp_n_u32_b16(...) __builtin_sve_svqincp_n_u32_b16(__VA_ARGS__)
+#define svqincp_n_u64_b8(...) __builtin_sve_svqincp_n_u64_b8(__VA_ARGS__)
+#define svqincp_n_u64_b32(...) __builtin_sve_svqincp_n_u64_b32(__VA_ARGS__)
+#define svqincp_n_u64_b64(...) __builtin_sve_svqincp_n_u64_b64(__VA_ARGS__)
+#define svqincp_n_u64_b16(...) __builtin_sve_svqincp_n_u64_b16(__VA_ARGS__)
+#define svqincp_s32(...) __builtin_sve_svqincp_s32(__VA_ARGS__)
+#define svqincp_s64(...) __builtin_sve_svqincp_s64(__VA_ARGS__)
+#define svqincp_s16(...) __builtin_sve_svqincp_s16(__VA_ARGS__)
+#define svqincp_u32(...) __builtin_sve_svqincp_u32(__VA_ARGS__)
+#define svqincp_u64(...) __builtin_sve_svqincp_u64(__VA_ARGS__)
+#define svqincp_u16(...) __builtin_sve_svqincp_u16(__VA_ARGS__)
+#define svqincw_n_s32(...) __builtin_sve_svqincw_n_s32(__VA_ARGS__)
+#define svqincw_n_s64(...) __builtin_sve_svqincw_n_s64(__VA_ARGS__)
+#define svqincw_n_u32(...) __builtin_sve_svqincw_n_u32(__VA_ARGS__)
+#define svqincw_n_u64(...) __builtin_sve_svqincw_n_u64(__VA_ARGS__)
+#define svqincw_s32(...) __builtin_sve_svqincw_s32(__VA_ARGS__)
+#define svqincw_u32(...) __builtin_sve_svqincw_u32(__VA_ARGS__)
+#define svqincw_pat_n_s32(...) __builtin_sve_svqincw_pat_n_s32(__VA_ARGS__)
+#define svqincw_pat_n_s64(...) __builtin_sve_svqincw_pat_n_s64(__VA_ARGS__)
+#define svqincw_pat_n_u32(...) __builtin_sve_svqincw_pat_n_u32(__VA_ARGS__)
+#define svqincw_pat_n_u64(...) __builtin_sve_svqincw_pat_n_u64(__VA_ARGS__)
+#define svqincw_pat_s32(...) __builtin_sve_svqincw_pat_s32(__VA_ARGS__)
+#define svqincw_pat_u32(...) __builtin_sve_svqincw_pat_u32(__VA_ARGS__)
+#define svqsub_n_s8(...) __builtin_sve_svqsub_n_s8(__VA_ARGS__)
+#define svqsub_n_s32(...) __builtin_sve_svqsub_n_s32(__VA_ARGS__)
+#define svqsub_n_s64(...) __builtin_sve_svqsub_n_s64(__VA_ARGS__)
+#define svqsub_n_s16(...) __builtin_sve_svqsub_n_s16(__VA_ARGS__)
+#define svqsub_n_u8(...) __builtin_sve_svqsub_n_u8(__VA_ARGS__)
+#define svqsub_n_u32(...) __builtin_sve_svqsub_n_u32(__VA_ARGS__)
+#define svqsub_n_u64(...) __builtin_sve_svqsub_n_u64(__VA_ARGS__)
+#define svqsub_n_u16(...) __builtin_sve_svqsub_n_u16(__VA_ARGS__)
+#define svqsub_s8(...) __builtin_sve_svqsub_s8(__VA_ARGS__)
+#define svqsub_s32(...) __builtin_sve_svqsub_s32(__VA_ARGS__)
+#define svqsub_s64(...) __builtin_sve_svqsub_s64(__VA_ARGS__)
+#define svqsub_s16(...) __builtin_sve_svqsub_s16(__VA_ARGS__)
+#define svqsub_u8(...) __builtin_sve_svqsub_u8(__VA_ARGS__)
+#define svqsub_u32(...) __builtin_sve_svqsub_u32(__VA_ARGS__)
+#define svqsub_u64(...) __builtin_sve_svqsub_u64(__VA_ARGS__)
+#define svqsub_u16(...) __builtin_sve_svqsub_u16(__VA_ARGS__)
+#define svrbit_u8_m(...) __builtin_sve_svrbit_u8_m(__VA_ARGS__)
+#define svrbit_u32_m(...) __builtin_sve_svrbit_u32_m(__VA_ARGS__)
+#define svrbit_u64_m(...) __builtin_sve_svrbit_u64_m(__VA_ARGS__)
+#define svrbit_u16_m(...) __builtin_sve_svrbit_u16_m(__VA_ARGS__)
+#define svrbit_s8_m(...) __builtin_sve_svrbit_s8_m(__VA_ARGS__)
+#define svrbit_s32_m(...) __builtin_sve_svrbit_s32_m(__VA_ARGS__)
+#define svrbit_s64_m(...) __builtin_sve_svrbit_s64_m(__VA_ARGS__)
+#define svrbit_s16_m(...) __builtin_sve_svrbit_s16_m(__VA_ARGS__)
+#define svrbit_u8_x(...) __builtin_sve_svrbit_u8_x(__VA_ARGS__)
+#define svrbit_u32_x(...) __builtin_sve_svrbit_u32_x(__VA_ARGS__)
+#define svrbit_u64_x(...) __builtin_sve_svrbit_u64_x(__VA_ARGS__)
+#define svrbit_u16_x(...) __builtin_sve_svrbit_u16_x(__VA_ARGS__)
+#define svrbit_s8_x(...) __builtin_sve_svrbit_s8_x(__VA_ARGS__)
+#define svrbit_s32_x(...) __builtin_sve_svrbit_s32_x(__VA_ARGS__)
+#define svrbit_s64_x(...) __builtin_sve_svrbit_s64_x(__VA_ARGS__)
+#define svrbit_s16_x(...) __builtin_sve_svrbit_s16_x(__VA_ARGS__)
+#define svrbit_u8_z(...) __builtin_sve_svrbit_u8_z(__VA_ARGS__)
+#define svrbit_u32_z(...) __builtin_sve_svrbit_u32_z(__VA_ARGS__)
+#define svrbit_u64_z(...) __builtin_sve_svrbit_u64_z(__VA_ARGS__)
+#define svrbit_u16_z(...) __builtin_sve_svrbit_u16_z(__VA_ARGS__)
+#define svrbit_s8_z(...) __builtin_sve_svrbit_s8_z(__VA_ARGS__)
+#define svrbit_s32_z(...) __builtin_sve_svrbit_s32_z(__VA_ARGS__)
+#define svrbit_s64_z(...) __builtin_sve_svrbit_s64_z(__VA_ARGS__)
+#define svrbit_s16_z(...) __builtin_sve_svrbit_s16_z(__VA_ARGS__)
+#define svrdffr(...) __builtin_sve_svrdffr(__VA_ARGS__)
+#define svrdffr_z(...) __builtin_sve_svrdffr_z(__VA_ARGS__)
+#define svrecpe_f64(...) __builtin_sve_svrecpe_f64(__VA_ARGS__)
+#define svrecpe_f32(...) __builtin_sve_svrecpe_f32(__VA_ARGS__)
+#define svrecpe_f16(...) __builtin_sve_svrecpe_f16(__VA_ARGS__)
+#define svrecps_f64(...) __builtin_sve_svrecps_f64(__VA_ARGS__)
+#define svrecps_f32(...) __builtin_sve_svrecps_f32(__VA_ARGS__)
+#define svrecps_f16(...) __builtin_sve_svrecps_f16(__VA_ARGS__)
+#define svrecpx_f64_m(...) __builtin_sve_svrecpx_f64_m(__VA_ARGS__)
+#define svrecpx_f32_m(...) __builtin_sve_svrecpx_f32_m(__VA_ARGS__)
+#define svrecpx_f16_m(...) __builtin_sve_svrecpx_f16_m(__VA_ARGS__)
+#define svrecpx_f64_x(...) __builtin_sve_svrecpx_f64_x(__VA_ARGS__)
+#define svrecpx_f32_x(...) __builtin_sve_svrecpx_f32_x(__VA_ARGS__)
+#define svrecpx_f16_x(...) __builtin_sve_svrecpx_f16_x(__VA_ARGS__)
+#define svrecpx_f64_z(...) __builtin_sve_svrecpx_f64_z(__VA_ARGS__)
+#define svrecpx_f32_z(...) __builtin_sve_svrecpx_f32_z(__VA_ARGS__)
+#define svrecpx_f16_z(...) __builtin_sve_svrecpx_f16_z(__VA_ARGS__)
+#define svrev_u8(...) __builtin_sve_svrev_u8(__VA_ARGS__)
+#define svrev_u32(...) __builtin_sve_svrev_u32(__VA_ARGS__)
+#define svrev_u64(...) __builtin_sve_svrev_u64(__VA_ARGS__)
+#define svrev_u16(...) __builtin_sve_svrev_u16(__VA_ARGS__)
+#define svrev_s8(...) __builtin_sve_svrev_s8(__VA_ARGS__)
+#define svrev_f64(...) __builtin_sve_svrev_f64(__VA_ARGS__)
+#define svrev_f32(...) __builtin_sve_svrev_f32(__VA_ARGS__)
+#define svrev_f16(...) __builtin_sve_svrev_f16(__VA_ARGS__)
+#define svrev_s32(...) __builtin_sve_svrev_s32(__VA_ARGS__)
+#define svrev_s64(...) __builtin_sve_svrev_s64(__VA_ARGS__)
+#define svrev_s16(...) __builtin_sve_svrev_s16(__VA_ARGS__)
+#define svrev_b8(...) __builtin_sve_svrev_b8(__VA_ARGS__)
+#define svrev_b32(...) __builtin_sve_svrev_b32(__VA_ARGS__)
+#define svrev_b64(...) __builtin_sve_svrev_b64(__VA_ARGS__)
+#define svrev_b16(...) __builtin_sve_svrev_b16(__VA_ARGS__)
+#define svrevb_u32_m(...) __builtin_sve_svrevb_u32_m(__VA_ARGS__)
+#define svrevb_u64_m(...) __builtin_sve_svrevb_u64_m(__VA_ARGS__)
+#define svrevb_u16_m(...) __builtin_sve_svrevb_u16_m(__VA_ARGS__)
+#define svrevb_s32_m(...) __builtin_sve_svrevb_s32_m(__VA_ARGS__)
+#define svrevb_s64_m(...) __builtin_sve_svrevb_s64_m(__VA_ARGS__)
+#define svrevb_s16_m(...) __builtin_sve_svrevb_s16_m(__VA_ARGS__)
+#define svrevb_u32_x(...) __builtin_sve_svrevb_u32_x(__VA_ARGS__)
+#define svrevb_u64_x(...) __builtin_sve_svrevb_u64_x(__VA_ARGS__)
+#define svrevb_u16_x(...) __builtin_sve_svrevb_u16_x(__VA_ARGS__)
+#define svrevb_s32_x(...) __builtin_sve_svrevb_s32_x(__VA_ARGS__)
+#define svrevb_s64_x(...) __builtin_sve_svrevb_s64_x(__VA_ARGS__)
+#define svrevb_s16_x(...) __builtin_sve_svrevb_s16_x(__VA_ARGS__)
+#define svrevb_u32_z(...) __builtin_sve_svrevb_u32_z(__VA_ARGS__)
+#define svrevb_u64_z(...) __builtin_sve_svrevb_u64_z(__VA_ARGS__)
+#define svrevb_u16_z(...) __builtin_sve_svrevb_u16_z(__VA_ARGS__)
+#define svrevb_s32_z(...) __builtin_sve_svrevb_s32_z(__VA_ARGS__)
+#define svrevb_s64_z(...) __builtin_sve_svrevb_s64_z(__VA_ARGS__)
+#define svrevb_s16_z(...) __builtin_sve_svrevb_s16_z(__VA_ARGS__)
+#define svrevh_u32_m(...) __builtin_sve_svrevh_u32_m(__VA_ARGS__)
+#define svrevh_u64_m(...) __builtin_sve_svrevh_u64_m(__VA_ARGS__)
+#define svrevh_s32_m(...) __builtin_sve_svrevh_s32_m(__VA_ARGS__)
+#define svrevh_s64_m(...) __builtin_sve_svrevh_s64_m(__VA_ARGS__)
+#define svrevh_u32_x(...) __builtin_sve_svrevh_u32_x(__VA_ARGS__)
+#define svrevh_u64_x(...) __builtin_sve_svrevh_u64_x(__VA_ARGS__)
+#define svrevh_s32_x(...) __builtin_sve_svrevh_s32_x(__VA_ARGS__)
+#define svrevh_s64_x(...) __builtin_sve_svrevh_s64_x(__VA_ARGS__)
+#define svrevh_u32_z(...) __builtin_sve_svrevh_u32_z(__VA_ARGS__)
+#define svrevh_u64_z(...) __builtin_sve_svrevh_u64_z(__VA_ARGS__)
+#define svrevh_s32_z(...) __builtin_sve_svrevh_s32_z(__VA_ARGS__)
+#define svrevh_s64_z(...) __builtin_sve_svrevh_s64_z(__VA_ARGS__)
+#define svrevw_u64_m(...) __builtin_sve_svrevw_u64_m(__VA_ARGS__)
+#define svrevw_s64_m(...) __builtin_sve_svrevw_s64_m(__VA_ARGS__)
+#define svrevw_u64_x(...) __builtin_sve_svrevw_u64_x(__VA_ARGS__)
+#define svrevw_s64_x(...) __builtin_sve_svrevw_s64_x(__VA_ARGS__)
+#define svrevw_u64_z(...) __builtin_sve_svrevw_u64_z(__VA_ARGS__)
+#define svrevw_s64_z(...) __builtin_sve_svrevw_s64_z(__VA_ARGS__)
+#define svrinta_f64_m(...) __builtin_sve_svrinta_f64_m(__VA_ARGS__)
+#define svrinta_f32_m(...) __builtin_sve_svrinta_f32_m(__VA_ARGS__)
+#define svrinta_f16_m(...) __builtin_sve_svrinta_f16_m(__VA_ARGS__)
+#define svrinta_f64_x(...) __builtin_sve_svrinta_f64_x(__VA_ARGS__)
+#define svrinta_f32_x(...) __builtin_sve_svrinta_f32_x(__VA_ARGS__)
+#define svrinta_f16_x(...) __builtin_sve_svrinta_f16_x(__VA_ARGS__)
+#define svrinta_f64_z(...) __builtin_sve_svrinta_f64_z(__VA_ARGS__)
+#define svrinta_f32_z(...) __builtin_sve_svrinta_f32_z(__VA_ARGS__)
+#define svrinta_f16_z(...) __builtin_sve_svrinta_f16_z(__VA_ARGS__)
+#define svrinti_f64_m(...) __builtin_sve_svrinti_f64_m(__VA_ARGS__)
+#define svrinti_f32_m(...) __builtin_sve_svrinti_f32_m(__VA_ARGS__)
+#define svrinti_f16_m(...) __builtin_sve_svrinti_f16_m(__VA_ARGS__)
+#define svrinti_f64_x(...) __builtin_sve_svrinti_f64_x(__VA_ARGS__)
+#define svrinti_f32_x(...) __builtin_sve_svrinti_f32_x(__VA_ARGS__)
+#define svrinti_f16_x(...) __builtin_sve_svrinti_f16_x(__VA_ARGS__)
+#define svrinti_f64_z(...) __builtin_sve_svrinti_f64_z(__VA_ARGS__)
+#define svrinti_f32_z(...) __builtin_sve_svrinti_f32_z(__VA_ARGS__)
+#define svrinti_f16_z(...) __builtin_sve_svrinti_f16_z(__VA_ARGS__)
+#define svrintm_f64_m(...) __builtin_sve_svrintm_f64_m(__VA_ARGS__)
+#define svrintm_f32_m(...) __builtin_sve_svrintm_f32_m(__VA_ARGS__)
+#define svrintm_f16_m(...) __builtin_sve_svrintm_f16_m(__VA_ARGS__)
+#define svrintm_f64_x(...) __builtin_sve_svrintm_f64_x(__VA_ARGS__)
+#define svrintm_f32_x(...) __builtin_sve_svrintm_f32_x(__VA_ARGS__)
+#define svrintm_f16_x(...) __builtin_sve_svrintm_f16_x(__VA_ARGS__)
+#define svrintm_f64_z(...) __builtin_sve_svrintm_f64_z(__VA_ARGS__)
+#define svrintm_f32_z(...) __builtin_sve_svrintm_f32_z(__VA_ARGS__)
+#define svrintm_f16_z(...) __builtin_sve_svrintm_f16_z(__VA_ARGS__)
+#define svrintn_f64_m(...) __builtin_sve_svrintn_f64_m(__VA_ARGS__)
+#define svrintn_f32_m(...) __builtin_sve_svrintn_f32_m(__VA_ARGS__)
+#define svrintn_f16_m(...) __builtin_sve_svrintn_f16_m(__VA_ARGS__)
+#define svrintn_f64_x(...) __builtin_sve_svrintn_f64_x(__VA_ARGS__)
+#define svrintn_f32_x(...) __builtin_sve_svrintn_f32_x(__VA_ARGS__)
+#define svrintn_f16_x(...) __builtin_sve_svrintn_f16_x(__VA_ARGS__)
+#define svrintn_f64_z(...) __builtin_sve_svrintn_f64_z(__VA_ARGS__)
+#define svrintn_f32_z(...) __builtin_sve_svrintn_f32_z(__VA_ARGS__)
+#define svrintn_f16_z(...) __builtin_sve_svrintn_f16_z(__VA_ARGS__)
+#define svrintp_f64_m(...) __builtin_sve_svrintp_f64_m(__VA_ARGS__)
+#define svrintp_f32_m(...) __builtin_sve_svrintp_f32_m(__VA_ARGS__)
+#define svrintp_f16_m(...) __builtin_sve_svrintp_f16_m(__VA_ARGS__)
+#define svrintp_f64_x(...) __builtin_sve_svrintp_f64_x(__VA_ARGS__)
+#define svrintp_f32_x(...) __builtin_sve_svrintp_f32_x(__VA_ARGS__)
+#define svrintp_f16_x(...) __builtin_sve_svrintp_f16_x(__VA_ARGS__)
+#define svrintp_f64_z(...) __builtin_sve_svrintp_f64_z(__VA_ARGS__)
+#define svrintp_f32_z(...) __builtin_sve_svrintp_f32_z(__VA_ARGS__)
+#define svrintp_f16_z(...) __builtin_sve_svrintp_f16_z(__VA_ARGS__)
+#define svrintx_f64_m(...) __builtin_sve_svrintx_f64_m(__VA_ARGS__)
+#define svrintx_f32_m(...) __builtin_sve_svrintx_f32_m(__VA_ARGS__)
+#define svrintx_f16_m(...) __builtin_sve_svrintx_f16_m(__VA_ARGS__)
+#define svrintx_f64_x(...) __builtin_sve_svrintx_f64_x(__VA_ARGS__)
+#define svrintx_f32_x(...) __builtin_sve_svrintx_f32_x(__VA_ARGS__)
+#define svrintx_f16_x(...) __builtin_sve_svrintx_f16_x(__VA_ARGS__)
+#define svrintx_f64_z(...) __builtin_sve_svrintx_f64_z(__VA_ARGS__)
+#define svrintx_f32_z(...) __builtin_sve_svrintx_f32_z(__VA_ARGS__)
+#define svrintx_f16_z(...) __builtin_sve_svrintx_f16_z(__VA_ARGS__)
+#define svrintz_f64_m(...) __builtin_sve_svrintz_f64_m(__VA_ARGS__)
+#define svrintz_f32_m(...) __builtin_sve_svrintz_f32_m(__VA_ARGS__)
+#define svrintz_f16_m(...) __builtin_sve_svrintz_f16_m(__VA_ARGS__)
+#define svrintz_f64_x(...) __builtin_sve_svrintz_f64_x(__VA_ARGS__)
+#define svrintz_f32_x(...) __builtin_sve_svrintz_f32_x(__VA_ARGS__)
+#define svrintz_f16_x(...) __builtin_sve_svrintz_f16_x(__VA_ARGS__)
+#define svrintz_f64_z(...) __builtin_sve_svrintz_f64_z(__VA_ARGS__)
+#define svrintz_f32_z(...) __builtin_sve_svrintz_f32_z(__VA_ARGS__)
+#define svrintz_f16_z(...) __builtin_sve_svrintz_f16_z(__VA_ARGS__)
+#define svrsqrte_f64(...) __builtin_sve_svrsqrte_f64(__VA_ARGS__)
+#define svrsqrte_f32(...) __builtin_sve_svrsqrte_f32(__VA_ARGS__)
+#define svrsqrte_f16(...) __builtin_sve_svrsqrte_f16(__VA_ARGS__)
+#define svrsqrts_f64(...) __builtin_sve_svrsqrts_f64(__VA_ARGS__)
+#define svrsqrts_f32(...) __builtin_sve_svrsqrts_f32(__VA_ARGS__)
+#define svrsqrts_f16(...) __builtin_sve_svrsqrts_f16(__VA_ARGS__)
+#define svscale_n_f64_m(...) __builtin_sve_svscale_n_f64_m(__VA_ARGS__)
+#define svscale_n_f32_m(...) __builtin_sve_svscale_n_f32_m(__VA_ARGS__)
+#define svscale_n_f16_m(...) __builtin_sve_svscale_n_f16_m(__VA_ARGS__)
+#define svscale_n_f64_x(...) __builtin_sve_svscale_n_f64_x(__VA_ARGS__)
+#define svscale_n_f32_x(...) __builtin_sve_svscale_n_f32_x(__VA_ARGS__)
+#define svscale_n_f16_x(...) __builtin_sve_svscale_n_f16_x(__VA_ARGS__)
+#define svscale_n_f64_z(...) __builtin_sve_svscale_n_f64_z(__VA_ARGS__)
+#define svscale_n_f32_z(...) __builtin_sve_svscale_n_f32_z(__VA_ARGS__)
+#define svscale_n_f16_z(...) __builtin_sve_svscale_n_f16_z(__VA_ARGS__)
+#define svscale_f64_m(...) __builtin_sve_svscale_f64_m(__VA_ARGS__)
+#define svscale_f32_m(...) __builtin_sve_svscale_f32_m(__VA_ARGS__)
+#define svscale_f16_m(...) __builtin_sve_svscale_f16_m(__VA_ARGS__)
+#define svscale_f64_x(...) __builtin_sve_svscale_f64_x(__VA_ARGS__)
+#define svscale_f32_x(...) __builtin_sve_svscale_f32_x(__VA_ARGS__)
+#define svscale_f16_x(...) __builtin_sve_svscale_f16_x(__VA_ARGS__)
+#define svscale_f64_z(...) __builtin_sve_svscale_f64_z(__VA_ARGS__)
+#define svscale_f32_z(...) __builtin_sve_svscale_f32_z(__VA_ARGS__)
+#define svscale_f16_z(...) __builtin_sve_svscale_f16_z(__VA_ARGS__)
+#define svsel_b(...) __builtin_sve_svsel_b(__VA_ARGS__)
+#define svsel_u8(...) __builtin_sve_svsel_u8(__VA_ARGS__)
+#define svsel_u32(...) __builtin_sve_svsel_u32(__VA_ARGS__)
+#define svsel_u64(...) __builtin_sve_svsel_u64(__VA_ARGS__)
+#define svsel_u16(...) __builtin_sve_svsel_u16(__VA_ARGS__)
+#define svsel_s8(...) __builtin_sve_svsel_s8(__VA_ARGS__)
+#define svsel_f64(...) __builtin_sve_svsel_f64(__VA_ARGS__)
+#define svsel_f32(...) __builtin_sve_svsel_f32(__VA_ARGS__)
+#define svsel_f16(...) __builtin_sve_svsel_f16(__VA_ARGS__)
+#define svsel_s32(...) __builtin_sve_svsel_s32(__VA_ARGS__)
+#define svsel_s64(...) __builtin_sve_svsel_s64(__VA_ARGS__)
+#define svsel_s16(...) __builtin_sve_svsel_s16(__VA_ARGS__)
+#define svset2_u8(...) __builtin_sve_svset2_u8(__VA_ARGS__)
+#define svset2_u32(...) __builtin_sve_svset2_u32(__VA_ARGS__)
+#define svset2_u64(...) __builtin_sve_svset2_u64(__VA_ARGS__)
+#define svset2_u16(...) __builtin_sve_svset2_u16(__VA_ARGS__)
+#define svset2_s8(...) __builtin_sve_svset2_s8(__VA_ARGS__)
+#define svset2_f64(...) __builtin_sve_svset2_f64(__VA_ARGS__)
+#define svset2_f32(...) __builtin_sve_svset2_f32(__VA_ARGS__)
+#define svset2_f16(...) __builtin_sve_svset2_f16(__VA_ARGS__)
+#define svset2_s32(...) __builtin_sve_svset2_s32(__VA_ARGS__)
+#define svset2_s64(...) __builtin_sve_svset2_s64(__VA_ARGS__)
+#define svset2_s16(...) __builtin_sve_svset2_s16(__VA_ARGS__)
+#define svset3_u8(...) __builtin_sve_svset3_u8(__VA_ARGS__)
+#define svset3_u32(...) __builtin_sve_svset3_u32(__VA_ARGS__)
+#define svset3_u64(...) __builtin_sve_svset3_u64(__VA_ARGS__)
+#define svset3_u16(...) __builtin_sve_svset3_u16(__VA_ARGS__)
+#define svset3_s8(...) __builtin_sve_svset3_s8(__VA_ARGS__)
+#define svset3_f64(...) __builtin_sve_svset3_f64(__VA_ARGS__)
+#define svset3_f32(...) __builtin_sve_svset3_f32(__VA_ARGS__)
+#define svset3_f16(...) __builtin_sve_svset3_f16(__VA_ARGS__)
+#define svset3_s32(...) __builtin_sve_svset3_s32(__VA_ARGS__)
+#define svset3_s64(...) __builtin_sve_svset3_s64(__VA_ARGS__)
+#define svset3_s16(...) __builtin_sve_svset3_s16(__VA_ARGS__)
+#define svset4_u8(...) __builtin_sve_svset4_u8(__VA_ARGS__)
+#define svset4_u32(...) __builtin_sve_svset4_u32(__VA_ARGS__)
+#define svset4_u64(...) __builtin_sve_svset4_u64(__VA_ARGS__)
+#define svset4_u16(...) __builtin_sve_svset4_u16(__VA_ARGS__)
+#define svset4_s8(...) __builtin_sve_svset4_s8(__VA_ARGS__)
+#define svset4_f64(...) __builtin_sve_svset4_f64(__VA_ARGS__)
+#define svset4_f32(...) __builtin_sve_svset4_f32(__VA_ARGS__)
+#define svset4_f16(...) __builtin_sve_svset4_f16(__VA_ARGS__)
+#define svset4_s32(...) __builtin_sve_svset4_s32(__VA_ARGS__)
+#define svset4_s64(...) __builtin_sve_svset4_s64(__VA_ARGS__)
+#define svset4_s16(...) __builtin_sve_svset4_s16(__VA_ARGS__)
+#define svsetffr(...) __builtin_sve_svsetffr(__VA_ARGS__)
+#define svsplice_u8(...) __builtin_sve_svsplice_u8(__VA_ARGS__)
+#define svsplice_u32(...) __builtin_sve_svsplice_u32(__VA_ARGS__)
+#define svsplice_u64(...) __builtin_sve_svsplice_u64(__VA_ARGS__)
+#define svsplice_u16(...) __builtin_sve_svsplice_u16(__VA_ARGS__)
+#define svsplice_s8(...) __builtin_sve_svsplice_s8(__VA_ARGS__)
+#define svsplice_f64(...) __builtin_sve_svsplice_f64(__VA_ARGS__)
+#define svsplice_f32(...) __builtin_sve_svsplice_f32(__VA_ARGS__)
+#define svsplice_f16(...) __builtin_sve_svsplice_f16(__VA_ARGS__)
+#define svsplice_s32(...) __builtin_sve_svsplice_s32(__VA_ARGS__)
+#define svsplice_s64(...) __builtin_sve_svsplice_s64(__VA_ARGS__)
+#define svsplice_s16(...) __builtin_sve_svsplice_s16(__VA_ARGS__)
+#define svsqrt_f64_m(...) __builtin_sve_svsqrt_f64_m(__VA_ARGS__)
+#define svsqrt_f32_m(...) __builtin_sve_svsqrt_f32_m(__VA_ARGS__)
+#define svsqrt_f16_m(...) __builtin_sve_svsqrt_f16_m(__VA_ARGS__)
+#define svsqrt_f64_x(...) __builtin_sve_svsqrt_f64_x(__VA_ARGS__)
+#define svsqrt_f32_x(...) __builtin_sve_svsqrt_f32_x(__VA_ARGS__)
+#define svsqrt_f16_x(...) __builtin_sve_svsqrt_f16_x(__VA_ARGS__)
+#define svsqrt_f64_z(...) __builtin_sve_svsqrt_f64_z(__VA_ARGS__)
+#define svsqrt_f32_z(...) __builtin_sve_svsqrt_f32_z(__VA_ARGS__)
+#define svsqrt_f16_z(...) __builtin_sve_svsqrt_f16_z(__VA_ARGS__)
+#define svst1_u8(...) __builtin_sve_svst1_u8(__VA_ARGS__)
+#define svst1_u32(...) __builtin_sve_svst1_u32(__VA_ARGS__)
+#define svst1_u64(...) __builtin_sve_svst1_u64(__VA_ARGS__)
+#define svst1_u16(...) __builtin_sve_svst1_u16(__VA_ARGS__)
+#define svst1_s8(...) __builtin_sve_svst1_s8(__VA_ARGS__)
+#define svst1_f64(...) __builtin_sve_svst1_f64(__VA_ARGS__)
+#define svst1_f32(...) __builtin_sve_svst1_f32(__VA_ARGS__)
+#define svst1_f16(...) __builtin_sve_svst1_f16(__VA_ARGS__)
+#define svst1_s32(...) __builtin_sve_svst1_s32(__VA_ARGS__)
+#define svst1_s64(...) __builtin_sve_svst1_s64(__VA_ARGS__)
+#define svst1_s16(...) __builtin_sve_svst1_s16(__VA_ARGS__)
+#define svst1_scatter_u32base_index_u32(...) __builtin_sve_svst1_scatter_u32base_index_u32(__VA_ARGS__)
+#define svst1_scatter_u64base_index_u64(...) __builtin_sve_svst1_scatter_u64base_index_u64(__VA_ARGS__)
+#define svst1_scatter_u64base_index_f64(...) __builtin_sve_svst1_scatter_u64base_index_f64(__VA_ARGS__)
+#define svst1_scatter_u32base_index_f32(...) __builtin_sve_svst1_scatter_u32base_index_f32(__VA_ARGS__)
+#define svst1_scatter_u32base_index_s32(...) __builtin_sve_svst1_scatter_u32base_index_s32(__VA_ARGS__)
+#define svst1_scatter_u64base_index_s64(...) __builtin_sve_svst1_scatter_u64base_index_s64(__VA_ARGS__)
+#define svst1_scatter_u32base_offset_u32(...) __builtin_sve_svst1_scatter_u32base_offset_u32(__VA_ARGS__)
+#define svst1_scatter_u64base_offset_u64(...) __builtin_sve_svst1_scatter_u64base_offset_u64(__VA_ARGS__)
+#define svst1_scatter_u64base_offset_f64(...) __builtin_sve_svst1_scatter_u64base_offset_f64(__VA_ARGS__)
+#define svst1_scatter_u32base_offset_f32(...) __builtin_sve_svst1_scatter_u32base_offset_f32(__VA_ARGS__)
+#define svst1_scatter_u32base_offset_s32(...) __builtin_sve_svst1_scatter_u32base_offset_s32(__VA_ARGS__)
+#define svst1_scatter_u64base_offset_s64(...) __builtin_sve_svst1_scatter_u64base_offset_s64(__VA_ARGS__)
+#define svst1_scatter_u32base_u32(...) __builtin_sve_svst1_scatter_u32base_u32(__VA_ARGS__)
+#define svst1_scatter_u64base_u64(...) __builtin_sve_svst1_scatter_u64base_u64(__VA_ARGS__)
+#define svst1_scatter_u64base_f64(...) __builtin_sve_svst1_scatter_u64base_f64(__VA_ARGS__)
+#define svst1_scatter_u32base_f32(...) __builtin_sve_svst1_scatter_u32base_f32(__VA_ARGS__)
+#define svst1_scatter_u32base_s32(...) __builtin_sve_svst1_scatter_u32base_s32(__VA_ARGS__)
+#define svst1_scatter_u64base_s64(...) __builtin_sve_svst1_scatter_u64base_s64(__VA_ARGS__)
+#define svst1_scatter_s32index_u32(...) __builtin_sve_svst1_scatter_s32index_u32(__VA_ARGS__)
+#define svst1_scatter_s32index_f32(...) __builtin_sve_svst1_scatter_s32index_f32(__VA_ARGS__)
+#define svst1_scatter_s32index_s32(...) __builtin_sve_svst1_scatter_s32index_s32(__VA_ARGS__)
+#define svst1_scatter_u32index_u32(...) __builtin_sve_svst1_scatter_u32index_u32(__VA_ARGS__)
+#define svst1_scatter_u32index_f32(...) __builtin_sve_svst1_scatter_u32index_f32(__VA_ARGS__)
+#define svst1_scatter_u32index_s32(...) __builtin_sve_svst1_scatter_u32index_s32(__VA_ARGS__)
+#define svst1_scatter_s64index_u64(...) __builtin_sve_svst1_scatter_s64index_u64(__VA_ARGS__)
+#define svst1_scatter_s64index_f64(...) __builtin_sve_svst1_scatter_s64index_f64(__VA_ARGS__)
+#define svst1_scatter_s64index_s64(...) __builtin_sve_svst1_scatter_s64index_s64(__VA_ARGS__)
+#define svst1_scatter_u64index_u64(...) __builtin_sve_svst1_scatter_u64index_u64(__VA_ARGS__)
+#define svst1_scatter_u64index_f64(...) __builtin_sve_svst1_scatter_u64index_f64(__VA_ARGS__)
+#define svst1_scatter_u64index_s64(...) __builtin_sve_svst1_scatter_u64index_s64(__VA_ARGS__)
+#define svst1_scatter_s32offset_u32(...) __builtin_sve_svst1_scatter_s32offset_u32(__VA_ARGS__)
+#define svst1_scatter_s32offset_f32(...) __builtin_sve_svst1_scatter_s32offset_f32(__VA_ARGS__)
+#define svst1_scatter_s32offset_s32(...) __builtin_sve_svst1_scatter_s32offset_s32(__VA_ARGS__)
+#define svst1_scatter_u32offset_u32(...) __builtin_sve_svst1_scatter_u32offset_u32(__VA_ARGS__)
+#define svst1_scatter_u32offset_f32(...) __builtin_sve_svst1_scatter_u32offset_f32(__VA_ARGS__)
+#define svst1_scatter_u32offset_s32(...) __builtin_sve_svst1_scatter_u32offset_s32(__VA_ARGS__)
+#define svst1_scatter_s64offset_u64(...) __builtin_sve_svst1_scatter_s64offset_u64(__VA_ARGS__)
+#define svst1_scatter_s64offset_f64(...) __builtin_sve_svst1_scatter_s64offset_f64(__VA_ARGS__)
+#define svst1_scatter_s64offset_s64(...) __builtin_sve_svst1_scatter_s64offset_s64(__VA_ARGS__)
+#define svst1_scatter_u64offset_u64(...) __builtin_sve_svst1_scatter_u64offset_u64(__VA_ARGS__)
+#define svst1_scatter_u64offset_f64(...) __builtin_sve_svst1_scatter_u64offset_f64(__VA_ARGS__)
+#define svst1_scatter_u64offset_s64(...) __builtin_sve_svst1_scatter_u64offset_s64(__VA_ARGS__)
+#define svst1_vnum_u8(...) __builtin_sve_svst1_vnum_u8(__VA_ARGS__)
+#define svst1_vnum_u32(...) __builtin_sve_svst1_vnum_u32(__VA_ARGS__)
+#define svst1_vnum_u64(...) __builtin_sve_svst1_vnum_u64(__VA_ARGS__)
+#define svst1_vnum_u16(...) __builtin_sve_svst1_vnum_u16(__VA_ARGS__)
+#define svst1_vnum_s8(...) __builtin_sve_svst1_vnum_s8(__VA_ARGS__)
+#define svst1_vnum_f64(...) __builtin_sve_svst1_vnum_f64(__VA_ARGS__)
+#define svst1_vnum_f32(...) __builtin_sve_svst1_vnum_f32(__VA_ARGS__)
+#define svst1_vnum_f16(...) __builtin_sve_svst1_vnum_f16(__VA_ARGS__)
+#define svst1_vnum_s32(...) __builtin_sve_svst1_vnum_s32(__VA_ARGS__)
+#define svst1_vnum_s64(...) __builtin_sve_svst1_vnum_s64(__VA_ARGS__)
+#define svst1_vnum_s16(...) __builtin_sve_svst1_vnum_s16(__VA_ARGS__)
+#define svst1b_s32(...) __builtin_sve_svst1b_s32(__VA_ARGS__)
+#define svst1b_s64(...) __builtin_sve_svst1b_s64(__VA_ARGS__)
+#define svst1b_s16(...) __builtin_sve_svst1b_s16(__VA_ARGS__)
+#define svst1b_u32(...) __builtin_sve_svst1b_u32(__VA_ARGS__)
+#define svst1b_u64(...) __builtin_sve_svst1b_u64(__VA_ARGS__)
+#define svst1b_u16(...) __builtin_sve_svst1b_u16(__VA_ARGS__)
+#define svst1b_scatter_u32base_offset_u32(...) __builtin_sve_svst1b_scatter_u32base_offset_u32(__VA_ARGS__)
+#define svst1b_scatter_u64base_offset_u64(...) __builtin_sve_svst1b_scatter_u64base_offset_u64(__VA_ARGS__)
+#define svst1b_scatter_u32base_offset_s32(...) __builtin_sve_svst1b_scatter_u32base_offset_s32(__VA_ARGS__)
+#define svst1b_scatter_u64base_offset_s64(...) __builtin_sve_svst1b_scatter_u64base_offset_s64(__VA_ARGS__)
+#define svst1b_scatter_u32base_u32(...) __builtin_sve_svst1b_scatter_u32base_u32(__VA_ARGS__)
+#define svst1b_scatter_u64base_u64(...) __builtin_sve_svst1b_scatter_u64base_u64(__VA_ARGS__)
+#define svst1b_scatter_u32base_s32(...) __builtin_sve_svst1b_scatter_u32base_s32(__VA_ARGS__)
+#define svst1b_scatter_u64base_s64(...) __builtin_sve_svst1b_scatter_u64base_s64(__VA_ARGS__)
+#define svst1b_scatter_s32offset_s32(...) __builtin_sve_svst1b_scatter_s32offset_s32(__VA_ARGS__)
+#define svst1b_scatter_s32offset_u32(...) __builtin_sve_svst1b_scatter_s32offset_u32(__VA_ARGS__)
+#define svst1b_scatter_u32offset_s32(...) __builtin_sve_svst1b_scatter_u32offset_s32(__VA_ARGS__)
+#define svst1b_scatter_u32offset_u32(...) __builtin_sve_svst1b_scatter_u32offset_u32(__VA_ARGS__)
+#define svst1b_scatter_s64offset_s64(...) __builtin_sve_svst1b_scatter_s64offset_s64(__VA_ARGS__)
+#define svst1b_scatter_s64offset_u64(...) __builtin_sve_svst1b_scatter_s64offset_u64(__VA_ARGS__)
+#define svst1b_scatter_u64offset_s64(...) __builtin_sve_svst1b_scatter_u64offset_s64(__VA_ARGS__)
+#define svst1b_scatter_u64offset_u64(...) __builtin_sve_svst1b_scatter_u64offset_u64(__VA_ARGS__)
+#define svst1b_vnum_s32(...) __builtin_sve_svst1b_vnum_s32(__VA_ARGS__)
+#define svst1b_vnum_s64(...) __builtin_sve_svst1b_vnum_s64(__VA_ARGS__)
+#define svst1b_vnum_s16(...) __builtin_sve_svst1b_vnum_s16(__VA_ARGS__)
+#define svst1b_vnum_u32(...) __builtin_sve_svst1b_vnum_u32(__VA_ARGS__)
+#define svst1b_vnum_u64(...) __builtin_sve_svst1b_vnum_u64(__VA_ARGS__)
+#define svst1b_vnum_u16(...) __builtin_sve_svst1b_vnum_u16(__VA_ARGS__)
+#define svst1h_s32(...) __builtin_sve_svst1h_s32(__VA_ARGS__)
+#define svst1h_s64(...) __builtin_sve_svst1h_s64(__VA_ARGS__)
+#define svst1h_u32(...) __builtin_sve_svst1h_u32(__VA_ARGS__)
+#define svst1h_u64(...) __builtin_sve_svst1h_u64(__VA_ARGS__)
+#define svst1h_scatter_u32base_index_u32(...) __builtin_sve_svst1h_scatter_u32base_index_u32(__VA_ARGS__)
+#define svst1h_scatter_u64base_index_u64(...) __builtin_sve_svst1h_scatter_u64base_index_u64(__VA_ARGS__)
+#define svst1h_scatter_u32base_index_s32(...) __builtin_sve_svst1h_scatter_u32base_index_s32(__VA_ARGS__)
+#define svst1h_scatter_u64base_index_s64(...) __builtin_sve_svst1h_scatter_u64base_index_s64(__VA_ARGS__)
+#define svst1h_scatter_u32base_offset_u32(...) __builtin_sve_svst1h_scatter_u32base_offset_u32(__VA_ARGS__)
+#define svst1h_scatter_u64base_offset_u64(...) __builtin_sve_svst1h_scatter_u64base_offset_u64(__VA_ARGS__)
+#define svst1h_scatter_u32base_offset_s32(...) __builtin_sve_svst1h_scatter_u32base_offset_s32(__VA_ARGS__)
+#define svst1h_scatter_u64base_offset_s64(...) __builtin_sve_svst1h_scatter_u64base_offset_s64(__VA_ARGS__)
+#define svst1h_scatter_u32base_u32(...) __builtin_sve_svst1h_scatter_u32base_u32(__VA_ARGS__)
+#define svst1h_scatter_u64base_u64(...) __builtin_sve_svst1h_scatter_u64base_u64(__VA_ARGS__)
+#define svst1h_scatter_u32base_s32(...) __builtin_sve_svst1h_scatter_u32base_s32(__VA_ARGS__)
+#define svst1h_scatter_u64base_s64(...) __builtin_sve_svst1h_scatter_u64base_s64(__VA_ARGS__)
+#define svst1h_scatter_s32index_s32(...) __builtin_sve_svst1h_scatter_s32index_s32(__VA_ARGS__)
+#define svst1h_scatter_s32index_u32(...) __builtin_sve_svst1h_scatter_s32index_u32(__VA_ARGS__)
+#define svst1h_scatter_u32index_s32(...) __builtin_sve_svst1h_scatter_u32index_s32(__VA_ARGS__)
+#define svst1h_scatter_u32index_u32(...) __builtin_sve_svst1h_scatter_u32index_u32(__VA_ARGS__)
+#define svst1h_scatter_s64index_s64(...) __builtin_sve_svst1h_scatter_s64index_s64(__VA_ARGS__)
+#define svst1h_scatter_s64index_u64(...) __builtin_sve_svst1h_scatter_s64index_u64(__VA_ARGS__)
+#define svst1h_scatter_u64index_s64(...) __builtin_sve_svst1h_scatter_u64index_s64(__VA_ARGS__)
+#define svst1h_scatter_u64index_u64(...) __builtin_sve_svst1h_scatter_u64index_u64(__VA_ARGS__)
+#define svst1h_scatter_s32offset_s32(...) __builtin_sve_svst1h_scatter_s32offset_s32(__VA_ARGS__)
+#define svst1h_scatter_s32offset_u32(...) __builtin_sve_svst1h_scatter_s32offset_u32(__VA_ARGS__)
+#define svst1h_scatter_u32offset_s32(...) __builtin_sve_svst1h_scatter_u32offset_s32(__VA_ARGS__)
+#define svst1h_scatter_u32offset_u32(...) __builtin_sve_svst1h_scatter_u32offset_u32(__VA_ARGS__)
+#define svst1h_scatter_s64offset_s64(...) __builtin_sve_svst1h_scatter_s64offset_s64(__VA_ARGS__)
+#define svst1h_scatter_s64offset_u64(...) __builtin_sve_svst1h_scatter_s64offset_u64(__VA_ARGS__)
+#define svst1h_scatter_u64offset_s64(...) __builtin_sve_svst1h_scatter_u64offset_s64(__VA_ARGS__)
+#define svst1h_scatter_u64offset_u64(...) __builtin_sve_svst1h_scatter_u64offset_u64(__VA_ARGS__)
+#define svst1h_vnum_s32(...) __builtin_sve_svst1h_vnum_s32(__VA_ARGS__)
+#define svst1h_vnum_s64(...) __builtin_sve_svst1h_vnum_s64(__VA_ARGS__)
+#define svst1h_vnum_u32(...) __builtin_sve_svst1h_vnum_u32(__VA_ARGS__)
+#define svst1h_vnum_u64(...) __builtin_sve_svst1h_vnum_u64(__VA_ARGS__)
+#define svst1w_s64(...) __builtin_sve_svst1w_s64(__VA_ARGS__)
+#define svst1w_u64(...) __builtin_sve_svst1w_u64(__VA_ARGS__)
+#define svst1w_scatter_u64base_index_u64(...) __builtin_sve_svst1w_scatter_u64base_index_u64(__VA_ARGS__)
+#define svst1w_scatter_u64base_index_s64(...) __builtin_sve_svst1w_scatter_u64base_index_s64(__VA_ARGS__)
+#define svst1w_scatter_u64base_offset_u64(...) __builtin_sve_svst1w_scatter_u64base_offset_u64(__VA_ARGS__)
+#define svst1w_scatter_u64base_offset_s64(...) __builtin_sve_svst1w_scatter_u64base_offset_s64(__VA_ARGS__)
+#define svst1w_scatter_u64base_u64(...) __builtin_sve_svst1w_scatter_u64base_u64(__VA_ARGS__)
+#define svst1w_scatter_u64base_s64(...) __builtin_sve_svst1w_scatter_u64base_s64(__VA_ARGS__)
+#define svst1w_scatter_s64index_s64(...) __builtin_sve_svst1w_scatter_s64index_s64(__VA_ARGS__)
+#define svst1w_scatter_s64index_u64(...) __builtin_sve_svst1w_scatter_s64index_u64(__VA_ARGS__)
+#define svst1w_scatter_u64index_s64(...) __builtin_sve_svst1w_scatter_u64index_s64(__VA_ARGS__)
+#define svst1w_scatter_u64index_u64(...) __builtin_sve_svst1w_scatter_u64index_u64(__VA_ARGS__)
+#define svst1w_scatter_s64offset_s64(...) __builtin_sve_svst1w_scatter_s64offset_s64(__VA_ARGS__)
+#define svst1w_scatter_s64offset_u64(...) __builtin_sve_svst1w_scatter_s64offset_u64(__VA_ARGS__)
+#define svst1w_scatter_u64offset_s64(...) __builtin_sve_svst1w_scatter_u64offset_s64(__VA_ARGS__)
+#define svst1w_scatter_u64offset_u64(...) __builtin_sve_svst1w_scatter_u64offset_u64(__VA_ARGS__)
+#define svst1w_vnum_s64(...) __builtin_sve_svst1w_vnum_s64(__VA_ARGS__)
+#define svst1w_vnum_u64(...) __builtin_sve_svst1w_vnum_u64(__VA_ARGS__)
+#define svst2_u8(...) __builtin_sve_svst2_u8(__VA_ARGS__)
+#define svst2_u32(...) __builtin_sve_svst2_u32(__VA_ARGS__)
+#define svst2_u64(...) __builtin_sve_svst2_u64(__VA_ARGS__)
+#define svst2_u16(...) __builtin_sve_svst2_u16(__VA_ARGS__)
+#define svst2_s8(...) __builtin_sve_svst2_s8(__VA_ARGS__)
+#define svst2_f64(...) __builtin_sve_svst2_f64(__VA_ARGS__)
+#define svst2_f32(...) __builtin_sve_svst2_f32(__VA_ARGS__)
+#define svst2_f16(...) __builtin_sve_svst2_f16(__VA_ARGS__)
+#define svst2_s32(...) __builtin_sve_svst2_s32(__VA_ARGS__)
+#define svst2_s64(...) __builtin_sve_svst2_s64(__VA_ARGS__)
+#define svst2_s16(...) __builtin_sve_svst2_s16(__VA_ARGS__)
+#define svst2_vnum_u8(...) __builtin_sve_svst2_vnum_u8(__VA_ARGS__)
+#define svst2_vnum_u32(...) __builtin_sve_svst2_vnum_u32(__VA_ARGS__)
+#define svst2_vnum_u64(...) __builtin_sve_svst2_vnum_u64(__VA_ARGS__)
+#define svst2_vnum_u16(...) __builtin_sve_svst2_vnum_u16(__VA_ARGS__)
+#define svst2_vnum_s8(...) __builtin_sve_svst2_vnum_s8(__VA_ARGS__)
+#define svst2_vnum_f64(...) __builtin_sve_svst2_vnum_f64(__VA_ARGS__)
+#define svst2_vnum_f32(...) __builtin_sve_svst2_vnum_f32(__VA_ARGS__)
+#define svst2_vnum_f16(...) __builtin_sve_svst2_vnum_f16(__VA_ARGS__)
+#define svst2_vnum_s32(...) __builtin_sve_svst2_vnum_s32(__VA_ARGS__)
+#define svst2_vnum_s64(...) __builtin_sve_svst2_vnum_s64(__VA_ARGS__)
+#define svst2_vnum_s16(...) __builtin_sve_svst2_vnum_s16(__VA_ARGS__)
+#define svst3_u8(...) __builtin_sve_svst3_u8(__VA_ARGS__)
+#define svst3_u32(...) __builtin_sve_svst3_u32(__VA_ARGS__)
+#define svst3_u64(...) __builtin_sve_svst3_u64(__VA_ARGS__)
+#define svst3_u16(...) __builtin_sve_svst3_u16(__VA_ARGS__)
+#define svst3_s8(...) __builtin_sve_svst3_s8(__VA_ARGS__)
+#define svst3_f64(...) __builtin_sve_svst3_f64(__VA_ARGS__)
+#define svst3_f32(...) __builtin_sve_svst3_f32(__VA_ARGS__)
+#define svst3_f16(...) __builtin_sve_svst3_f16(__VA_ARGS__)
+#define svst3_s32(...) __builtin_sve_svst3_s32(__VA_ARGS__)
+#define svst3_s64(...) __builtin_sve_svst3_s64(__VA_ARGS__)
+#define svst3_s16(...) __builtin_sve_svst3_s16(__VA_ARGS__)
+#define svst3_vnum_u8(...) __builtin_sve_svst3_vnum_u8(__VA_ARGS__)
+#define svst3_vnum_u32(...) __builtin_sve_svst3_vnum_u32(__VA_ARGS__)
+#define svst3_vnum_u64(...) __builtin_sve_svst3_vnum_u64(__VA_ARGS__)
+#define svst3_vnum_u16(...) __builtin_sve_svst3_vnum_u16(__VA_ARGS__)
+#define svst3_vnum_s8(...) __builtin_sve_svst3_vnum_s8(__VA_ARGS__)
+#define svst3_vnum_f64(...) __builtin_sve_svst3_vnum_f64(__VA_ARGS__)
+#define svst3_vnum_f32(...) __builtin_sve_svst3_vnum_f32(__VA_ARGS__)
+#define svst3_vnum_f16(...) __builtin_sve_svst3_vnum_f16(__VA_ARGS__)
+#define svst3_vnum_s32(...) __builtin_sve_svst3_vnum_s32(__VA_ARGS__)
+#define svst3_vnum_s64(...) __builtin_sve_svst3_vnum_s64(__VA_ARGS__)
+#define svst3_vnum_s16(...) __builtin_sve_svst3_vnum_s16(__VA_ARGS__)
+#define svst4_u8(...) __builtin_sve_svst4_u8(__VA_ARGS__)
+#define svst4_u32(...) __builtin_sve_svst4_u32(__VA_ARGS__)
+#define svst4_u64(...) __builtin_sve_svst4_u64(__VA_ARGS__)
+#define svst4_u16(...) __builtin_sve_svst4_u16(__VA_ARGS__)
+#define svst4_s8(...) __builtin_sve_svst4_s8(__VA_ARGS__)
+#define svst4_f64(...) __builtin_sve_svst4_f64(__VA_ARGS__)
+#define svst4_f32(...) __builtin_sve_svst4_f32(__VA_ARGS__)
+#define svst4_f16(...) __builtin_sve_svst4_f16(__VA_ARGS__)
+#define svst4_s32(...) __builtin_sve_svst4_s32(__VA_ARGS__)
+#define svst4_s64(...) __builtin_sve_svst4_s64(__VA_ARGS__)
+#define svst4_s16(...) __builtin_sve_svst4_s16(__VA_ARGS__)
+#define svst4_vnum_u8(...) __builtin_sve_svst4_vnum_u8(__VA_ARGS__)
+#define svst4_vnum_u32(...) __builtin_sve_svst4_vnum_u32(__VA_ARGS__)
+#define svst4_vnum_u64(...) __builtin_sve_svst4_vnum_u64(__VA_ARGS__)
+#define svst4_vnum_u16(...) __builtin_sve_svst4_vnum_u16(__VA_ARGS__)
+#define svst4_vnum_s8(...) __builtin_sve_svst4_vnum_s8(__VA_ARGS__)
+#define svst4_vnum_f64(...) __builtin_sve_svst4_vnum_f64(__VA_ARGS__)
+#define svst4_vnum_f32(...) __builtin_sve_svst4_vnum_f32(__VA_ARGS__)
+#define svst4_vnum_f16(...) __builtin_sve_svst4_vnum_f16(__VA_ARGS__)
+#define svst4_vnum_s32(...) __builtin_sve_svst4_vnum_s32(__VA_ARGS__)
+#define svst4_vnum_s64(...) __builtin_sve_svst4_vnum_s64(__VA_ARGS__)
+#define svst4_vnum_s16(...) __builtin_sve_svst4_vnum_s16(__VA_ARGS__)
+#define svstnt1_u8(...) __builtin_sve_svstnt1_u8(__VA_ARGS__)
+#define svstnt1_u32(...) __builtin_sve_svstnt1_u32(__VA_ARGS__)
+#define svstnt1_u64(...) __builtin_sve_svstnt1_u64(__VA_ARGS__)
+#define svstnt1_u16(...) __builtin_sve_svstnt1_u16(__VA_ARGS__)
+#define svstnt1_s8(...) __builtin_sve_svstnt1_s8(__VA_ARGS__)
+#define svstnt1_f64(...) __builtin_sve_svstnt1_f64(__VA_ARGS__)
+#define svstnt1_f32(...) __builtin_sve_svstnt1_f32(__VA_ARGS__)
+#define svstnt1_f16(...) __builtin_sve_svstnt1_f16(__VA_ARGS__)
+#define svstnt1_s32(...) __builtin_sve_svstnt1_s32(__VA_ARGS__)
+#define svstnt1_s64(...) __builtin_sve_svstnt1_s64(__VA_ARGS__)
+#define svstnt1_s16(...) __builtin_sve_svstnt1_s16(__VA_ARGS__)
+#define svstnt1_vnum_u8(...) __builtin_sve_svstnt1_vnum_u8(__VA_ARGS__)
+#define svstnt1_vnum_u32(...) __builtin_sve_svstnt1_vnum_u32(__VA_ARGS__)
+#define svstnt1_vnum_u64(...) __builtin_sve_svstnt1_vnum_u64(__VA_ARGS__)
+#define svstnt1_vnum_u16(...) __builtin_sve_svstnt1_vnum_u16(__VA_ARGS__)
+#define svstnt1_vnum_s8(...) __builtin_sve_svstnt1_vnum_s8(__VA_ARGS__)
+#define svstnt1_vnum_f64(...) __builtin_sve_svstnt1_vnum_f64(__VA_ARGS__)
+#define svstnt1_vnum_f32(...) __builtin_sve_svstnt1_vnum_f32(__VA_ARGS__)
+#define svstnt1_vnum_f16(...) __builtin_sve_svstnt1_vnum_f16(__VA_ARGS__)
+#define svstnt1_vnum_s32(...) __builtin_sve_svstnt1_vnum_s32(__VA_ARGS__)
+#define svstnt1_vnum_s64(...) __builtin_sve_svstnt1_vnum_s64(__VA_ARGS__)
+#define svstnt1_vnum_s16(...) __builtin_sve_svstnt1_vnum_s16(__VA_ARGS__)
+#define svsub_n_f64_m(...) __builtin_sve_svsub_n_f64_m(__VA_ARGS__)
+#define svsub_n_f32_m(...) __builtin_sve_svsub_n_f32_m(__VA_ARGS__)
+#define svsub_n_f16_m(...) __builtin_sve_svsub_n_f16_m(__VA_ARGS__)
+#define svsub_n_f64_x(...) __builtin_sve_svsub_n_f64_x(__VA_ARGS__)
+#define svsub_n_f32_x(...) __builtin_sve_svsub_n_f32_x(__VA_ARGS__)
+#define svsub_n_f16_x(...) __builtin_sve_svsub_n_f16_x(__VA_ARGS__)
+#define svsub_n_f64_z(...) __builtin_sve_svsub_n_f64_z(__VA_ARGS__)
+#define svsub_n_f32_z(...) __builtin_sve_svsub_n_f32_z(__VA_ARGS__)
+#define svsub_n_f16_z(...) __builtin_sve_svsub_n_f16_z(__VA_ARGS__)
+#define svsub_n_u8_m(...) __builtin_sve_svsub_n_u8_m(__VA_ARGS__)
+#define svsub_n_u32_m(...) __builtin_sve_svsub_n_u32_m(__VA_ARGS__)
+#define svsub_n_u64_m(...) __builtin_sve_svsub_n_u64_m(__VA_ARGS__)
+#define svsub_n_u16_m(...) __builtin_sve_svsub_n_u16_m(__VA_ARGS__)
+#define svsub_n_s8_m(...) __builtin_sve_svsub_n_s8_m(__VA_ARGS__)
+#define svsub_n_s32_m(...) __builtin_sve_svsub_n_s32_m(__VA_ARGS__)
+#define svsub_n_s64_m(...) __builtin_sve_svsub_n_s64_m(__VA_ARGS__)
+#define svsub_n_s16_m(...) __builtin_sve_svsub_n_s16_m(__VA_ARGS__)
+#define svsub_n_u8_x(...) __builtin_sve_svsub_n_u8_x(__VA_ARGS__)
+#define svsub_n_u32_x(...) __builtin_sve_svsub_n_u32_x(__VA_ARGS__)
+#define svsub_n_u64_x(...) __builtin_sve_svsub_n_u64_x(__VA_ARGS__)
+#define svsub_n_u16_x(...) __builtin_sve_svsub_n_u16_x(__VA_ARGS__)
+#define svsub_n_s8_x(...) __builtin_sve_svsub_n_s8_x(__VA_ARGS__)
+#define svsub_n_s32_x(...) __builtin_sve_svsub_n_s32_x(__VA_ARGS__)
+#define svsub_n_s64_x(...) __builtin_sve_svsub_n_s64_x(__VA_ARGS__)
+#define svsub_n_s16_x(...) __builtin_sve_svsub_n_s16_x(__VA_ARGS__)
+#define svsub_n_u8_z(...) __builtin_sve_svsub_n_u8_z(__VA_ARGS__)
+#define svsub_n_u32_z(...) __builtin_sve_svsub_n_u32_z(__VA_ARGS__)
+#define svsub_n_u64_z(...) __builtin_sve_svsub_n_u64_z(__VA_ARGS__)
+#define svsub_n_u16_z(...) __builtin_sve_svsub_n_u16_z(__VA_ARGS__)
+#define svsub_n_s8_z(...) __builtin_sve_svsub_n_s8_z(__VA_ARGS__)
+#define svsub_n_s32_z(...) __builtin_sve_svsub_n_s32_z(__VA_ARGS__)
+#define svsub_n_s64_z(...) __builtin_sve_svsub_n_s64_z(__VA_ARGS__)
+#define svsub_n_s16_z(...) __builtin_sve_svsub_n_s16_z(__VA_ARGS__)
+#define svsub_f64_m(...) __builtin_sve_svsub_f64_m(__VA_ARGS__)
+#define svsub_f32_m(...) __builtin_sve_svsub_f32_m(__VA_ARGS__)
+#define svsub_f16_m(...) __builtin_sve_svsub_f16_m(__VA_ARGS__)
+#define svsub_f64_x(...) __builtin_sve_svsub_f64_x(__VA_ARGS__)
+#define svsub_f32_x(...) __builtin_sve_svsub_f32_x(__VA_ARGS__)
+#define svsub_f16_x(...) __builtin_sve_svsub_f16_x(__VA_ARGS__)
+#define svsub_f64_z(...) __builtin_sve_svsub_f64_z(__VA_ARGS__)
+#define svsub_f32_z(...) __builtin_sve_svsub_f32_z(__VA_ARGS__)
+#define svsub_f16_z(...) __builtin_sve_svsub_f16_z(__VA_ARGS__)
+#define svsub_u8_m(...) __builtin_sve_svsub_u8_m(__VA_ARGS__)
+#define svsub_u32_m(...) __builtin_sve_svsub_u32_m(__VA_ARGS__)
+#define svsub_u64_m(...) __builtin_sve_svsub_u64_m(__VA_ARGS__)
+#define svsub_u16_m(...) __builtin_sve_svsub_u16_m(__VA_ARGS__)
+#define svsub_s8_m(...) __builtin_sve_svsub_s8_m(__VA_ARGS__)
+#define svsub_s32_m(...) __builtin_sve_svsub_s32_m(__VA_ARGS__)
+#define svsub_s64_m(...) __builtin_sve_svsub_s64_m(__VA_ARGS__)
+#define svsub_s16_m(...) __builtin_sve_svsub_s16_m(__VA_ARGS__)
+#define svsub_u8_x(...) __builtin_sve_svsub_u8_x(__VA_ARGS__)
+#define svsub_u32_x(...) __builtin_sve_svsub_u32_x(__VA_ARGS__)
+#define svsub_u64_x(...) __builtin_sve_svsub_u64_x(__VA_ARGS__)
+#define svsub_u16_x(...) __builtin_sve_svsub_u16_x(__VA_ARGS__)
+#define svsub_s8_x(...) __builtin_sve_svsub_s8_x(__VA_ARGS__)
+#define svsub_s32_x(...) __builtin_sve_svsub_s32_x(__VA_ARGS__)
+#define svsub_s64_x(...) __builtin_sve_svsub_s64_x(__VA_ARGS__)
+#define svsub_s16_x(...) __builtin_sve_svsub_s16_x(__VA_ARGS__)
+#define svsub_u8_z(...) __builtin_sve_svsub_u8_z(__VA_ARGS__)
+#define svsub_u32_z(...) __builtin_sve_svsub_u32_z(__VA_ARGS__)
+#define svsub_u64_z(...) __builtin_sve_svsub_u64_z(__VA_ARGS__)
+#define svsub_u16_z(...) __builtin_sve_svsub_u16_z(__VA_ARGS__)
+#define svsub_s8_z(...) __builtin_sve_svsub_s8_z(__VA_ARGS__)
+#define svsub_s32_z(...) __builtin_sve_svsub_s32_z(__VA_ARGS__)
+#define svsub_s64_z(...) __builtin_sve_svsub_s64_z(__VA_ARGS__)
+#define svsub_s16_z(...) __builtin_sve_svsub_s16_z(__VA_ARGS__)
+#define svsubr_n_f64_m(...) __builtin_sve_svsubr_n_f64_m(__VA_ARGS__)
+#define svsubr_n_f32_m(...) __builtin_sve_svsubr_n_f32_m(__VA_ARGS__)
+#define svsubr_n_f16_m(...) __builtin_sve_svsubr_n_f16_m(__VA_ARGS__)
+#define svsubr_n_f64_x(...) __builtin_sve_svsubr_n_f64_x(__VA_ARGS__)
+#define svsubr_n_f32_x(...) __builtin_sve_svsubr_n_f32_x(__VA_ARGS__)
+#define svsubr_n_f16_x(...) __builtin_sve_svsubr_n_f16_x(__VA_ARGS__)
+#define svsubr_n_f64_z(...) __builtin_sve_svsubr_n_f64_z(__VA_ARGS__)
+#define svsubr_n_f32_z(...) __builtin_sve_svsubr_n_f32_z(__VA_ARGS__)
+#define svsubr_n_f16_z(...) __builtin_sve_svsubr_n_f16_z(__VA_ARGS__)
+#define svsubr_n_u8_m(...) __builtin_sve_svsubr_n_u8_m(__VA_ARGS__)
+#define svsubr_n_u32_m(...) __builtin_sve_svsubr_n_u32_m(__VA_ARGS__)
+#define svsubr_n_u64_m(...) __builtin_sve_svsubr_n_u64_m(__VA_ARGS__)
+#define svsubr_n_u16_m(...) __builtin_sve_svsubr_n_u16_m(__VA_ARGS__)
+#define svsubr_n_s8_m(...) __builtin_sve_svsubr_n_s8_m(__VA_ARGS__)
+#define svsubr_n_s32_m(...) __builtin_sve_svsubr_n_s32_m(__VA_ARGS__)
+#define svsubr_n_s64_m(...) __builtin_sve_svsubr_n_s64_m(__VA_ARGS__)
+#define svsubr_n_s16_m(...) __builtin_sve_svsubr_n_s16_m(__VA_ARGS__)
+#define svsubr_n_u8_x(...) __builtin_sve_svsubr_n_u8_x(__VA_ARGS__)
+#define svsubr_n_u32_x(...) __builtin_sve_svsubr_n_u32_x(__VA_ARGS__)
+#define svsubr_n_u64_x(...) __builtin_sve_svsubr_n_u64_x(__VA_ARGS__)
+#define svsubr_n_u16_x(...) __builtin_sve_svsubr_n_u16_x(__VA_ARGS__)
+#define svsubr_n_s8_x(...) __builtin_sve_svsubr_n_s8_x(__VA_ARGS__)
+#define svsubr_n_s32_x(...) __builtin_sve_svsubr_n_s32_x(__VA_ARGS__)
+#define svsubr_n_s64_x(...) __builtin_sve_svsubr_n_s64_x(__VA_ARGS__)
+#define svsubr_n_s16_x(...) __builtin_sve_svsubr_n_s16_x(__VA_ARGS__)
+#define svsubr_n_u8_z(...) __builtin_sve_svsubr_n_u8_z(__VA_ARGS__)
+#define svsubr_n_u32_z(...) __builtin_sve_svsubr_n_u32_z(__VA_ARGS__)
+#define svsubr_n_u64_z(...) __builtin_sve_svsubr_n_u64_z(__VA_ARGS__)
+#define svsubr_n_u16_z(...) __builtin_sve_svsubr_n_u16_z(__VA_ARGS__)
+#define svsubr_n_s8_z(...) __builtin_sve_svsubr_n_s8_z(__VA_ARGS__)
+#define svsubr_n_s32_z(...) __builtin_sve_svsubr_n_s32_z(__VA_ARGS__)
+#define svsubr_n_s64_z(...) __builtin_sve_svsubr_n_s64_z(__VA_ARGS__)
+#define svsubr_n_s16_z(...) __builtin_sve_svsubr_n_s16_z(__VA_ARGS__)
+#define svsubr_f64_m(...) __builtin_sve_svsubr_f64_m(__VA_ARGS__)
+#define svsubr_f32_m(...) __builtin_sve_svsubr_f32_m(__VA_ARGS__)
+#define svsubr_f16_m(...) __builtin_sve_svsubr_f16_m(__VA_ARGS__)
+#define svsubr_f64_x(...) __builtin_sve_svsubr_f64_x(__VA_ARGS__)
+#define svsubr_f32_x(...) __builtin_sve_svsubr_f32_x(__VA_ARGS__)
+#define svsubr_f16_x(...) __builtin_sve_svsubr_f16_x(__VA_ARGS__)
+#define svsubr_f64_z(...) __builtin_sve_svsubr_f64_z(__VA_ARGS__)
+#define svsubr_f32_z(...) __builtin_sve_svsubr_f32_z(__VA_ARGS__)
+#define svsubr_f16_z(...) __builtin_sve_svsubr_f16_z(__VA_ARGS__)
+#define svsubr_u8_m(...) __builtin_sve_svsubr_u8_m(__VA_ARGS__)
+#define svsubr_u32_m(...) __builtin_sve_svsubr_u32_m(__VA_ARGS__)
+#define svsubr_u64_m(...) __builtin_sve_svsubr_u64_m(__VA_ARGS__)
+#define svsubr_u16_m(...) __builtin_sve_svsubr_u16_m(__VA_ARGS__)
+#define svsubr_s8_m(...) __builtin_sve_svsubr_s8_m(__VA_ARGS__)
+#define svsubr_s32_m(...) __builtin_sve_svsubr_s32_m(__VA_ARGS__)
+#define svsubr_s64_m(...) __builtin_sve_svsubr_s64_m(__VA_ARGS__)
+#define svsubr_s16_m(...) __builtin_sve_svsubr_s16_m(__VA_ARGS__)
+#define svsubr_u8_x(...) __builtin_sve_svsubr_u8_x(__VA_ARGS__)
+#define svsubr_u32_x(...) __builtin_sve_svsubr_u32_x(__VA_ARGS__)
+#define svsubr_u64_x(...) __builtin_sve_svsubr_u64_x(__VA_ARGS__)
+#define svsubr_u16_x(...) __builtin_sve_svsubr_u16_x(__VA_ARGS__)
+#define svsubr_s8_x(...) __builtin_sve_svsubr_s8_x(__VA_ARGS__)
+#define svsubr_s32_x(...) __builtin_sve_svsubr_s32_x(__VA_ARGS__)
+#define svsubr_s64_x(...) __builtin_sve_svsubr_s64_x(__VA_ARGS__)
+#define svsubr_s16_x(...) __builtin_sve_svsubr_s16_x(__VA_ARGS__)
+#define svsubr_u8_z(...) __builtin_sve_svsubr_u8_z(__VA_ARGS__)
+#define svsubr_u32_z(...) __builtin_sve_svsubr_u32_z(__VA_ARGS__)
+#define svsubr_u64_z(...) __builtin_sve_svsubr_u64_z(__VA_ARGS__)
+#define svsubr_u16_z(...) __builtin_sve_svsubr_u16_z(__VA_ARGS__)
+#define svsubr_s8_z(...) __builtin_sve_svsubr_s8_z(__VA_ARGS__)
+#define svsubr_s32_z(...) __builtin_sve_svsubr_s32_z(__VA_ARGS__)
+#define svsubr_s64_z(...) __builtin_sve_svsubr_s64_z(__VA_ARGS__)
+#define svsubr_s16_z(...) __builtin_sve_svsubr_s16_z(__VA_ARGS__)
+#define svtbl_u8(...) __builtin_sve_svtbl_u8(__VA_ARGS__)
+#define svtbl_u32(...) __builtin_sve_svtbl_u32(__VA_ARGS__)
+#define svtbl_u64(...) __builtin_sve_svtbl_u64(__VA_ARGS__)
+#define svtbl_u16(...) __builtin_sve_svtbl_u16(__VA_ARGS__)
+#define svtbl_s8(...) __builtin_sve_svtbl_s8(__VA_ARGS__)
+#define svtbl_f64(...) __builtin_sve_svtbl_f64(__VA_ARGS__)
+#define svtbl_f32(...) __builtin_sve_svtbl_f32(__VA_ARGS__)
+#define svtbl_f16(...) __builtin_sve_svtbl_f16(__VA_ARGS__)
+#define svtbl_s32(...) __builtin_sve_svtbl_s32(__VA_ARGS__)
+#define svtbl_s64(...) __builtin_sve_svtbl_s64(__VA_ARGS__)
+#define svtbl_s16(...) __builtin_sve_svtbl_s16(__VA_ARGS__)
+#define svtmad_f64(...) __builtin_sve_svtmad_f64(__VA_ARGS__)
+#define svtmad_f32(...) __builtin_sve_svtmad_f32(__VA_ARGS__)
+#define svtmad_f16(...) __builtin_sve_svtmad_f16(__VA_ARGS__)
+#define svtrn1_u8(...) __builtin_sve_svtrn1_u8(__VA_ARGS__)
+#define svtrn1_u32(...) __builtin_sve_svtrn1_u32(__VA_ARGS__)
+#define svtrn1_u64(...) __builtin_sve_svtrn1_u64(__VA_ARGS__)
+#define svtrn1_u16(...) __builtin_sve_svtrn1_u16(__VA_ARGS__)
+#define svtrn1_s8(...) __builtin_sve_svtrn1_s8(__VA_ARGS__)
+#define svtrn1_f64(...) __builtin_sve_svtrn1_f64(__VA_ARGS__)
+#define svtrn1_f32(...) __builtin_sve_svtrn1_f32(__VA_ARGS__)
+#define svtrn1_f16(...) __builtin_sve_svtrn1_f16(__VA_ARGS__)
+#define svtrn1_s32(...) __builtin_sve_svtrn1_s32(__VA_ARGS__)
+#define svtrn1_s64(...) __builtin_sve_svtrn1_s64(__VA_ARGS__)
+#define svtrn1_s16(...) __builtin_sve_svtrn1_s16(__VA_ARGS__)
+#define svtrn1_b8(...) __builtin_sve_svtrn1_b8(__VA_ARGS__)
+#define svtrn1_b32(...) __builtin_sve_svtrn1_b32(__VA_ARGS__)
+#define svtrn1_b64(...) __builtin_sve_svtrn1_b64(__VA_ARGS__)
+#define svtrn1_b16(...) __builtin_sve_svtrn1_b16(__VA_ARGS__)
+#define svtrn2_u8(...) __builtin_sve_svtrn2_u8(__VA_ARGS__)
+#define svtrn2_u32(...) __builtin_sve_svtrn2_u32(__VA_ARGS__)
+#define svtrn2_u64(...) __builtin_sve_svtrn2_u64(__VA_ARGS__)
+#define svtrn2_u16(...) __builtin_sve_svtrn2_u16(__VA_ARGS__)
+#define svtrn2_s8(...) __builtin_sve_svtrn2_s8(__VA_ARGS__)
+#define svtrn2_f64(...) __builtin_sve_svtrn2_f64(__VA_ARGS__)
+#define svtrn2_f32(...) __builtin_sve_svtrn2_f32(__VA_ARGS__)
+#define svtrn2_f16(...) __builtin_sve_svtrn2_f16(__VA_ARGS__)
+#define svtrn2_s32(...) __builtin_sve_svtrn2_s32(__VA_ARGS__)
+#define svtrn2_s64(...) __builtin_sve_svtrn2_s64(__VA_ARGS__)
+#define svtrn2_s16(...) __builtin_sve_svtrn2_s16(__VA_ARGS__)
+#define svtrn2_b8(...) __builtin_sve_svtrn2_b8(__VA_ARGS__)
+#define svtrn2_b32(...) __builtin_sve_svtrn2_b32(__VA_ARGS__)
+#define svtrn2_b64(...) __builtin_sve_svtrn2_b64(__VA_ARGS__)
+#define svtrn2_b16(...) __builtin_sve_svtrn2_b16(__VA_ARGS__)
+#define svtsmul_f64(...) __builtin_sve_svtsmul_f64(__VA_ARGS__)
+#define svtsmul_f32(...) __builtin_sve_svtsmul_f32(__VA_ARGS__)
+#define svtsmul_f16(...) __builtin_sve_svtsmul_f16(__VA_ARGS__)
+#define svtssel_f64(...) __builtin_sve_svtssel_f64(__VA_ARGS__)
+#define svtssel_f32(...) __builtin_sve_svtssel_f32(__VA_ARGS__)
+#define svtssel_f16(...) __builtin_sve_svtssel_f16(__VA_ARGS__)
+#define svundef2_u8(...) __builtin_sve_svundef2_u8(__VA_ARGS__)
+#define svundef2_u32(...) __builtin_sve_svundef2_u32(__VA_ARGS__)
+#define svundef2_u64(...) __builtin_sve_svundef2_u64(__VA_ARGS__)
+#define svundef2_u16(...) __builtin_sve_svundef2_u16(__VA_ARGS__)
+#define svundef2_s8(...) __builtin_sve_svundef2_s8(__VA_ARGS__)
+#define svundef2_f64(...) __builtin_sve_svundef2_f64(__VA_ARGS__)
+#define svundef2_f32(...) __builtin_sve_svundef2_f32(__VA_ARGS__)
+#define svundef2_f16(...) __builtin_sve_svundef2_f16(__VA_ARGS__)
+#define svundef2_s32(...) __builtin_sve_svundef2_s32(__VA_ARGS__)
+#define svundef2_s64(...) __builtin_sve_svundef2_s64(__VA_ARGS__)
+#define svundef2_s16(...) __builtin_sve_svundef2_s16(__VA_ARGS__)
+#define svundef3_u8(...) __builtin_sve_svundef3_u8(__VA_ARGS__)
+#define svundef3_u32(...) __builtin_sve_svundef3_u32(__VA_ARGS__)
+#define svundef3_u64(...) __builtin_sve_svundef3_u64(__VA_ARGS__)
+#define svundef3_u16(...) __builtin_sve_svundef3_u16(__VA_ARGS__)
+#define svundef3_s8(...) __builtin_sve_svundef3_s8(__VA_ARGS__)
+#define svundef3_f64(...) __builtin_sve_svundef3_f64(__VA_ARGS__)
+#define svundef3_f32(...) __builtin_sve_svundef3_f32(__VA_ARGS__)
+#define svundef3_f16(...) __builtin_sve_svundef3_f16(__VA_ARGS__)
+#define svundef3_s32(...) __builtin_sve_svundef3_s32(__VA_ARGS__)
+#define svundef3_s64(...) __builtin_sve_svundef3_s64(__VA_ARGS__)
+#define svundef3_s16(...) __builtin_sve_svundef3_s16(__VA_ARGS__)
+#define svundef4_u8(...) __builtin_sve_svundef4_u8(__VA_ARGS__)
+#define svundef4_u32(...) __builtin_sve_svundef4_u32(__VA_ARGS__)
+#define svundef4_u64(...) __builtin_sve_svundef4_u64(__VA_ARGS__)
+#define svundef4_u16(...) __builtin_sve_svundef4_u16(__VA_ARGS__)
+#define svundef4_s8(...) __builtin_sve_svundef4_s8(__VA_ARGS__)
+#define svundef4_f64(...) __builtin_sve_svundef4_f64(__VA_ARGS__)
+#define svundef4_f32(...) __builtin_sve_svundef4_f32(__VA_ARGS__)
+#define svundef4_f16(...) __builtin_sve_svundef4_f16(__VA_ARGS__)
+#define svundef4_s32(...) __builtin_sve_svundef4_s32(__VA_ARGS__)
+#define svundef4_s64(...) __builtin_sve_svundef4_s64(__VA_ARGS__)
+#define svundef4_s16(...) __builtin_sve_svundef4_s16(__VA_ARGS__)
+#define svundef_u8(...) __builtin_sve_svundef_u8(__VA_ARGS__)
+#define svundef_u32(...) __builtin_sve_svundef_u32(__VA_ARGS__)
+#define svundef_u64(...) __builtin_sve_svundef_u64(__VA_ARGS__)
+#define svundef_u16(...) __builtin_sve_svundef_u16(__VA_ARGS__)
+#define svundef_s8(...) __builtin_sve_svundef_s8(__VA_ARGS__)
+#define svundef_f64(...) __builtin_sve_svundef_f64(__VA_ARGS__)
+#define svundef_f32(...) __builtin_sve_svundef_f32(__VA_ARGS__)
+#define svundef_f16(...) __builtin_sve_svundef_f16(__VA_ARGS__)
+#define svundef_s32(...) __builtin_sve_svundef_s32(__VA_ARGS__)
+#define svundef_s64(...) __builtin_sve_svundef_s64(__VA_ARGS__)
+#define svundef_s16(...) __builtin_sve_svundef_s16(__VA_ARGS__)
+#define svunpkhi_b(...) __builtin_sve_svunpkhi_b(__VA_ARGS__)
+#define svunpkhi_s32(...) __builtin_sve_svunpkhi_s32(__VA_ARGS__)
+#define svunpkhi_s64(...) __builtin_sve_svunpkhi_s64(__VA_ARGS__)
+#define svunpkhi_s16(...) __builtin_sve_svunpkhi_s16(__VA_ARGS__)
+#define svunpkhi_u32(...) __builtin_sve_svunpkhi_u32(__VA_ARGS__)
+#define svunpkhi_u64(...) __builtin_sve_svunpkhi_u64(__VA_ARGS__)
+#define svunpkhi_u16(...) __builtin_sve_svunpkhi_u16(__VA_ARGS__)
+#define svunpklo_b(...) __builtin_sve_svunpklo_b(__VA_ARGS__)
+#define svunpklo_s32(...) __builtin_sve_svunpklo_s32(__VA_ARGS__)
+#define svunpklo_s64(...) __builtin_sve_svunpklo_s64(__VA_ARGS__)
+#define svunpklo_s16(...) __builtin_sve_svunpklo_s16(__VA_ARGS__)
+#define svunpklo_u32(...) __builtin_sve_svunpklo_u32(__VA_ARGS__)
+#define svunpklo_u64(...) __builtin_sve_svunpklo_u64(__VA_ARGS__)
+#define svunpklo_u16(...) __builtin_sve_svunpklo_u16(__VA_ARGS__)
+#define svuzp1_u8(...) __builtin_sve_svuzp1_u8(__VA_ARGS__)
+#define svuzp1_u32(...) __builtin_sve_svuzp1_u32(__VA_ARGS__)
+#define svuzp1_u64(...) __builtin_sve_svuzp1_u64(__VA_ARGS__)
+#define svuzp1_u16(...) __builtin_sve_svuzp1_u16(__VA_ARGS__)
+#define svuzp1_s8(...) __builtin_sve_svuzp1_s8(__VA_ARGS__)
+#define svuzp1_f64(...) __builtin_sve_svuzp1_f64(__VA_ARGS__)
+#define svuzp1_f32(...) __builtin_sve_svuzp1_f32(__VA_ARGS__)
+#define svuzp1_f16(...) __builtin_sve_svuzp1_f16(__VA_ARGS__)
+#define svuzp1_s32(...) __builtin_sve_svuzp1_s32(__VA_ARGS__)
+#define svuzp1_s64(...) __builtin_sve_svuzp1_s64(__VA_ARGS__)
+#define svuzp1_s16(...) __builtin_sve_svuzp1_s16(__VA_ARGS__)
+#define svuzp1_b8(...) __builtin_sve_svuzp1_b8(__VA_ARGS__)
+#define svuzp1_b32(...) __builtin_sve_svuzp1_b32(__VA_ARGS__)
+#define svuzp1_b64(...) __builtin_sve_svuzp1_b64(__VA_ARGS__)
+#define svuzp1_b16(...) __builtin_sve_svuzp1_b16(__VA_ARGS__)
+#define svuzp2_u8(...) __builtin_sve_svuzp2_u8(__VA_ARGS__)
+#define svuzp2_u32(...) __builtin_sve_svuzp2_u32(__VA_ARGS__)
+#define svuzp2_u64(...) __builtin_sve_svuzp2_u64(__VA_ARGS__)
+#define svuzp2_u16(...) __builtin_sve_svuzp2_u16(__VA_ARGS__)
+#define svuzp2_s8(...) __builtin_sve_svuzp2_s8(__VA_ARGS__)
+#define svuzp2_f64(...) __builtin_sve_svuzp2_f64(__VA_ARGS__)
+#define svuzp2_f32(...) __builtin_sve_svuzp2_f32(__VA_ARGS__)
+#define svuzp2_f16(...) __builtin_sve_svuzp2_f16(__VA_ARGS__)
+#define svuzp2_s32(...) __builtin_sve_svuzp2_s32(__VA_ARGS__)
+#define svuzp2_s64(...) __builtin_sve_svuzp2_s64(__VA_ARGS__)
+#define svuzp2_s16(...) __builtin_sve_svuzp2_s16(__VA_ARGS__)
+#define svuzp2_b8(...) __builtin_sve_svuzp2_b8(__VA_ARGS__)
+#define svuzp2_b32(...) __builtin_sve_svuzp2_b32(__VA_ARGS__)
+#define svuzp2_b64(...) __builtin_sve_svuzp2_b64(__VA_ARGS__)
+#define svuzp2_b16(...) __builtin_sve_svuzp2_b16(__VA_ARGS__)
+#define svwhilele_b8_s32(...) __builtin_sve_svwhilele_b8_s32(__VA_ARGS__)
+#define svwhilele_b32_s32(...) __builtin_sve_svwhilele_b32_s32(__VA_ARGS__)
+#define svwhilele_b64_s32(...) __builtin_sve_svwhilele_b64_s32(__VA_ARGS__)
+#define svwhilele_b16_s32(...) __builtin_sve_svwhilele_b16_s32(__VA_ARGS__)
+#define svwhilele_b8_s64(...) __builtin_sve_svwhilele_b8_s64(__VA_ARGS__)
+#define svwhilele_b32_s64(...) __builtin_sve_svwhilele_b32_s64(__VA_ARGS__)
+#define svwhilele_b64_s64(...) __builtin_sve_svwhilele_b64_s64(__VA_ARGS__)
+#define svwhilele_b16_s64(...) __builtin_sve_svwhilele_b16_s64(__VA_ARGS__)
+#define svwhilele_b8_u32(...) __builtin_sve_svwhilele_b8_u32(__VA_ARGS__)
+#define svwhilele_b32_u32(...) __builtin_sve_svwhilele_b32_u32(__VA_ARGS__)
+#define svwhilele_b64_u32(...) __builtin_sve_svwhilele_b64_u32(__VA_ARGS__)
+#define svwhilele_b16_u32(...) __builtin_sve_svwhilele_b16_u32(__VA_ARGS__)
+#define svwhilele_b8_u64(...) __builtin_sve_svwhilele_b8_u64(__VA_ARGS__)
+#define svwhilele_b32_u64(...) __builtin_sve_svwhilele_b32_u64(__VA_ARGS__)
+#define svwhilele_b64_u64(...) __builtin_sve_svwhilele_b64_u64(__VA_ARGS__)
+#define svwhilele_b16_u64(...) __builtin_sve_svwhilele_b16_u64(__VA_ARGS__)
+#define svwhilelt_b8_u32(...) __builtin_sve_svwhilelt_b8_u32(__VA_ARGS__)
+#define svwhilelt_b32_u32(...) __builtin_sve_svwhilelt_b32_u32(__VA_ARGS__)
+#define svwhilelt_b64_u32(...) __builtin_sve_svwhilelt_b64_u32(__VA_ARGS__)
+#define svwhilelt_b16_u32(...) __builtin_sve_svwhilelt_b16_u32(__VA_ARGS__)
+#define svwhilelt_b8_u64(...) __builtin_sve_svwhilelt_b8_u64(__VA_ARGS__)
+#define svwhilelt_b32_u64(...) __builtin_sve_svwhilelt_b32_u64(__VA_ARGS__)
+#define svwhilelt_b64_u64(...) __builtin_sve_svwhilelt_b64_u64(__VA_ARGS__)
+#define svwhilelt_b16_u64(...) __builtin_sve_svwhilelt_b16_u64(__VA_ARGS__)
+#define svwhilelt_b8_s32(...) __builtin_sve_svwhilelt_b8_s32(__VA_ARGS__)
+#define svwhilelt_b32_s32(...) __builtin_sve_svwhilelt_b32_s32(__VA_ARGS__)
+#define svwhilelt_b64_s32(...) __builtin_sve_svwhilelt_b64_s32(__VA_ARGS__)
+#define svwhilelt_b16_s32(...) __builtin_sve_svwhilelt_b16_s32(__VA_ARGS__)
+#define svwhilelt_b8_s64(...) __builtin_sve_svwhilelt_b8_s64(__VA_ARGS__)
+#define svwhilelt_b32_s64(...) __builtin_sve_svwhilelt_b32_s64(__VA_ARGS__)
+#define svwhilelt_b64_s64(...) __builtin_sve_svwhilelt_b64_s64(__VA_ARGS__)
+#define svwhilelt_b16_s64(...) __builtin_sve_svwhilelt_b16_s64(__VA_ARGS__)
+#define svwrffr(...) __builtin_sve_svwrffr(__VA_ARGS__)
+#define svzip1_u8(...) __builtin_sve_svzip1_u8(__VA_ARGS__)
+#define svzip1_u32(...) __builtin_sve_svzip1_u32(__VA_ARGS__)
+#define svzip1_u64(...) __builtin_sve_svzip1_u64(__VA_ARGS__)
+#define svzip1_u16(...) __builtin_sve_svzip1_u16(__VA_ARGS__)
+#define svzip1_s8(...) __builtin_sve_svzip1_s8(__VA_ARGS__)
+#define svzip1_f64(...) __builtin_sve_svzip1_f64(__VA_ARGS__)
+#define svzip1_f32(...) __builtin_sve_svzip1_f32(__VA_ARGS__)
+#define svzip1_f16(...) __builtin_sve_svzip1_f16(__VA_ARGS__)
+#define svzip1_s32(...) __builtin_sve_svzip1_s32(__VA_ARGS__)
+#define svzip1_s64(...) __builtin_sve_svzip1_s64(__VA_ARGS__)
+#define svzip1_s16(...) __builtin_sve_svzip1_s16(__VA_ARGS__)
+#define svzip1_b8(...) __builtin_sve_svzip1_b8(__VA_ARGS__)
+#define svzip1_b32(...) __builtin_sve_svzip1_b32(__VA_ARGS__)
+#define svzip1_b64(...) __builtin_sve_svzip1_b64(__VA_ARGS__)
+#define svzip1_b16(...) __builtin_sve_svzip1_b16(__VA_ARGS__)
+#define svzip2_u8(...) __builtin_sve_svzip2_u8(__VA_ARGS__)
+#define svzip2_u32(...) __builtin_sve_svzip2_u32(__VA_ARGS__)
+#define svzip2_u64(...) __builtin_sve_svzip2_u64(__VA_ARGS__)
+#define svzip2_u16(...) __builtin_sve_svzip2_u16(__VA_ARGS__)
+#define svzip2_s8(...) __builtin_sve_svzip2_s8(__VA_ARGS__)
+#define svzip2_f64(...) __builtin_sve_svzip2_f64(__VA_ARGS__)
+#define svzip2_f32(...) __builtin_sve_svzip2_f32(__VA_ARGS__)
+#define svzip2_f16(...) __builtin_sve_svzip2_f16(__VA_ARGS__)
+#define svzip2_s32(...) __builtin_sve_svzip2_s32(__VA_ARGS__)
+#define svzip2_s64(...) __builtin_sve_svzip2_s64(__VA_ARGS__)
+#define svzip2_s16(...) __builtin_sve_svzip2_s16(__VA_ARGS__)
+#define svzip2_b8(...) __builtin_sve_svzip2_b8(__VA_ARGS__)
+#define svzip2_b32(...) __builtin_sve_svzip2_b32(__VA_ARGS__)
+#define svzip2_b64(...) __builtin_sve_svzip2_b64(__VA_ARGS__)
+#define svzip2_b16(...) __builtin_sve_svzip2_b16(__VA_ARGS__)
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_f64_m)))
+svfloat64_t svabd_m(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_f32_m)))
+svfloat32_t svabd_m(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_f16_m)))
+svfloat16_t svabd_m(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_f64_x)))
+svfloat64_t svabd_x(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_f32_x)))
+svfloat32_t svabd_x(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_f16_x)))
+svfloat16_t svabd_x(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_f64_z)))
+svfloat64_t svabd_z(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_f32_z)))
+svfloat32_t svabd_z(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_f16_z)))
+svfloat16_t svabd_z(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_s8_m)))
+svint8_t svabd_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_s32_m)))
+svint32_t svabd_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_s64_m)))
+svint64_t svabd_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_s16_m)))
+svint16_t svabd_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_s8_x)))
+svint8_t svabd_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_s32_x)))
+svint32_t svabd_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_s64_x)))
+svint64_t svabd_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_s16_x)))
+svint16_t svabd_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_s8_z)))
+svint8_t svabd_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_s32_z)))
+svint32_t svabd_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_s64_z)))
+svint64_t svabd_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_s16_z)))
+svint16_t svabd_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_u8_m)))
+svuint8_t svabd_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_u32_m)))
+svuint32_t svabd_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_u64_m)))
+svuint64_t svabd_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_u16_m)))
+svuint16_t svabd_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_u8_x)))
+svuint8_t svabd_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_u32_x)))
+svuint32_t svabd_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_u64_x)))
+svuint64_t svabd_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_u16_x)))
+svuint16_t svabd_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_u8_z)))
+svuint8_t svabd_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_u32_z)))
+svuint32_t svabd_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_u64_z)))
+svuint64_t svabd_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_n_u16_z)))
+svuint16_t svabd_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_f64_m)))
+svfloat64_t svabd_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_f32_m)))
+svfloat32_t svabd_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_f16_m)))
+svfloat16_t svabd_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_f64_x)))
+svfloat64_t svabd_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_f32_x)))
+svfloat32_t svabd_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_f16_x)))
+svfloat16_t svabd_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_f64_z)))
+svfloat64_t svabd_z(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_f32_z)))
+svfloat32_t svabd_z(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_f16_z)))
+svfloat16_t svabd_z(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_s8_m)))
+svint8_t svabd_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_s32_m)))
+svint32_t svabd_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_s64_m)))
+svint64_t svabd_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_s16_m)))
+svint16_t svabd_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_s8_x)))
+svint8_t svabd_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_s32_x)))
+svint32_t svabd_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_s64_x)))
+svint64_t svabd_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_s16_x)))
+svint16_t svabd_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_s8_z)))
+svint8_t svabd_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_s32_z)))
+svint32_t svabd_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_s64_z)))
+svint64_t svabd_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_s16_z)))
+svint16_t svabd_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_u8_m)))
+svuint8_t svabd_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_u32_m)))
+svuint32_t svabd_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_u64_m)))
+svuint64_t svabd_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_u16_m)))
+svuint16_t svabd_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_u8_x)))
+svuint8_t svabd_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_u32_x)))
+svuint32_t svabd_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_u64_x)))
+svuint64_t svabd_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_u16_x)))
+svuint16_t svabd_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_u8_z)))
+svuint8_t svabd_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_u32_z)))
+svuint32_t svabd_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_u64_z)))
+svuint64_t svabd_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabd_u16_z)))
+svuint16_t svabd_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_f64_m)))
+svfloat64_t svabs_m(svfloat64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_f32_m)))
+svfloat32_t svabs_m(svfloat32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_f16_m)))
+svfloat16_t svabs_m(svfloat16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_f64_x)))
+svfloat64_t svabs_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_f32_x)))
+svfloat32_t svabs_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_f16_x)))
+svfloat16_t svabs_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_f64_z)))
+svfloat64_t svabs_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_f32_z)))
+svfloat32_t svabs_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_f16_z)))
+svfloat16_t svabs_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_s8_m)))
+svint8_t svabs_m(svint8_t, svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_s32_m)))
+svint32_t svabs_m(svint32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_s64_m)))
+svint64_t svabs_m(svint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_s16_m)))
+svint16_t svabs_m(svint16_t, svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_s8_x)))
+svint8_t svabs_x(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_s32_x)))
+svint32_t svabs_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_s64_x)))
+svint64_t svabs_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_s16_x)))
+svint16_t svabs_x(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_s8_z)))
+svint8_t svabs_z(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_s32_z)))
+svint32_t svabs_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_s64_z)))
+svint64_t svabs_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabs_s16_z)))
+svint16_t svabs_z(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacge_n_f64)))
+svbool_t svacge(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacge_n_f32)))
+svbool_t svacge(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacge_n_f16)))
+svbool_t svacge(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacge_f64)))
+svbool_t svacge(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacge_f32)))
+svbool_t svacge(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacge_f16)))
+svbool_t svacge(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacgt_n_f64)))
+svbool_t svacgt(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacgt_n_f32)))
+svbool_t svacgt(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacgt_n_f16)))
+svbool_t svacgt(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacgt_f64)))
+svbool_t svacgt(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacgt_f32)))
+svbool_t svacgt(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacgt_f16)))
+svbool_t svacgt(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacle_n_f64)))
+svbool_t svacle(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacle_n_f32)))
+svbool_t svacle(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacle_n_f16)))
+svbool_t svacle(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacle_f64)))
+svbool_t svacle(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacle_f32)))
+svbool_t svacle(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svacle_f16)))
+svbool_t svacle(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaclt_n_f64)))
+svbool_t svaclt(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaclt_n_f32)))
+svbool_t svaclt(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaclt_n_f16)))
+svbool_t svaclt(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaclt_f64)))
+svbool_t svaclt(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaclt_f32)))
+svbool_t svaclt(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaclt_f16)))
+svbool_t svaclt(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_f64_m)))
+svfloat64_t svadd_m(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_f32_m)))
+svfloat32_t svadd_m(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_f16_m)))
+svfloat16_t svadd_m(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_f64_x)))
+svfloat64_t svadd_x(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_f32_x)))
+svfloat32_t svadd_x(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_f16_x)))
+svfloat16_t svadd_x(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_f64_z)))
+svfloat64_t svadd_z(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_f32_z)))
+svfloat32_t svadd_z(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_f16_z)))
+svfloat16_t svadd_z(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_u8_m)))
+svuint8_t svadd_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_u32_m)))
+svuint32_t svadd_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_u64_m)))
+svuint64_t svadd_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_u16_m)))
+svuint16_t svadd_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_s8_m)))
+svint8_t svadd_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_s32_m)))
+svint32_t svadd_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_s64_m)))
+svint64_t svadd_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_s16_m)))
+svint16_t svadd_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_u8_x)))
+svuint8_t svadd_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_u32_x)))
+svuint32_t svadd_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_u64_x)))
+svuint64_t svadd_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_u16_x)))
+svuint16_t svadd_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_s8_x)))
+svint8_t svadd_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_s32_x)))
+svint32_t svadd_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_s64_x)))
+svint64_t svadd_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_s16_x)))
+svint16_t svadd_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_u8_z)))
+svuint8_t svadd_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_u32_z)))
+svuint32_t svadd_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_u64_z)))
+svuint64_t svadd_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_u16_z)))
+svuint16_t svadd_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_s8_z)))
+svint8_t svadd_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_s32_z)))
+svint32_t svadd_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_s64_z)))
+svint64_t svadd_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_n_s16_z)))
+svint16_t svadd_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_f64_m)))
+svfloat64_t svadd_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_f32_m)))
+svfloat32_t svadd_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_f16_m)))
+svfloat16_t svadd_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_f64_x)))
+svfloat64_t svadd_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_f32_x)))
+svfloat32_t svadd_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_f16_x)))
+svfloat16_t svadd_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_f64_z)))
+svfloat64_t svadd_z(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_f32_z)))
+svfloat32_t svadd_z(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_f16_z)))
+svfloat16_t svadd_z(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_u8_m)))
+svuint8_t svadd_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_u32_m)))
+svuint32_t svadd_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_u64_m)))
+svuint64_t svadd_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_u16_m)))
+svuint16_t svadd_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_s8_m)))
+svint8_t svadd_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_s32_m)))
+svint32_t svadd_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_s64_m)))
+svint64_t svadd_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_s16_m)))
+svint16_t svadd_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_u8_x)))
+svuint8_t svadd_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_u32_x)))
+svuint32_t svadd_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_u64_x)))
+svuint64_t svadd_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_u16_x)))
+svuint16_t svadd_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_s8_x)))
+svint8_t svadd_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_s32_x)))
+svint32_t svadd_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_s64_x)))
+svint64_t svadd_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_s16_x)))
+svint16_t svadd_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_u8_z)))
+svuint8_t svadd_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_u32_z)))
+svuint32_t svadd_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_u64_z)))
+svuint64_t svadd_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_u16_z)))
+svuint16_t svadd_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_s8_z)))
+svint8_t svadd_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_s32_z)))
+svint32_t svadd_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_s64_z)))
+svint64_t svadd_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_s16_z)))
+svint16_t svadd_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadda_f64)))
+float64_t svadda(svbool_t, float64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadda_f32)))
+float32_t svadda(svbool_t, float32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadda_f16)))
+float16_t svadda(svbool_t, float16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddv_s8)))
+int64_t svaddv(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddv_s32)))
+int64_t svaddv(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddv_s64)))
+int64_t svaddv(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddv_s16)))
+int64_t svaddv(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddv_u8)))
+uint64_t svaddv(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddv_u32)))
+uint64_t svaddv(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddv_u64)))
+uint64_t svaddv(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddv_u16)))
+uint64_t svaddv(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddv_f64)))
+float64_t svaddv(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddv_f32)))
+float32_t svaddv(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddv_f16)))
+float16_t svaddv(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrb_u32base_u32offset)))
+svuint32_t svadrb_offset(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrb_u64base_u64offset)))
+svuint64_t svadrb_offset(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrb_u32base_s32offset)))
+svuint32_t svadrb_offset(svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrb_u64base_s64offset)))
+svuint64_t svadrb_offset(svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrd_u32base_u32index)))
+svuint32_t svadrd_index(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrd_u64base_u64index)))
+svuint64_t svadrd_index(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrd_u32base_s32index)))
+svuint32_t svadrd_index(svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrd_u64base_s64index)))
+svuint64_t svadrd_index(svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrh_u32base_u32index)))
+svuint32_t svadrh_index(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrh_u64base_u64index)))
+svuint64_t svadrh_index(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrh_u32base_s32index)))
+svuint32_t svadrh_index(svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrh_u64base_s64index)))
+svuint64_t svadrh_index(svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrw_u32base_u32index)))
+svuint32_t svadrw_index(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrw_u64base_u64index)))
+svuint64_t svadrw_index(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrw_u32base_s32index)))
+svuint32_t svadrw_index(svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadrw_u64base_s64index)))
+svuint64_t svadrw_index(svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_b_z)))
+svbool_t svand_z(svbool_t, svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_u8_m)))
+svuint8_t svand_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_u32_m)))
+svuint32_t svand_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_u64_m)))
+svuint64_t svand_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_u16_m)))
+svuint16_t svand_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_s8_m)))
+svint8_t svand_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_s32_m)))
+svint32_t svand_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_s64_m)))
+svint64_t svand_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_s16_m)))
+svint16_t svand_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_u8_x)))
+svuint8_t svand_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_u32_x)))
+svuint32_t svand_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_u64_x)))
+svuint64_t svand_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_u16_x)))
+svuint16_t svand_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_s8_x)))
+svint8_t svand_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_s32_x)))
+svint32_t svand_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_s64_x)))
+svint64_t svand_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_s16_x)))
+svint16_t svand_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_u8_z)))
+svuint8_t svand_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_u32_z)))
+svuint32_t svand_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_u64_z)))
+svuint64_t svand_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_u16_z)))
+svuint16_t svand_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_s8_z)))
+svint8_t svand_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_s32_z)))
+svint32_t svand_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_s64_z)))
+svint64_t svand_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_n_s16_z)))
+svint16_t svand_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_u8_m)))
+svuint8_t svand_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_u32_m)))
+svuint32_t svand_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_u64_m)))
+svuint64_t svand_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_u16_m)))
+svuint16_t svand_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_s8_m)))
+svint8_t svand_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_s32_m)))
+svint32_t svand_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_s64_m)))
+svint64_t svand_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_s16_m)))
+svint16_t svand_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_u8_x)))
+svuint8_t svand_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_u32_x)))
+svuint32_t svand_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_u64_x)))
+svuint64_t svand_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_u16_x)))
+svuint16_t svand_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_s8_x)))
+svint8_t svand_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_s32_x)))
+svint32_t svand_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_s64_x)))
+svint64_t svand_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_s16_x)))
+svint16_t svand_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_u8_z)))
+svuint8_t svand_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_u32_z)))
+svuint32_t svand_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_u64_z)))
+svuint64_t svand_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_u16_z)))
+svuint16_t svand_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_s8_z)))
+svint8_t svand_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_s32_z)))
+svint32_t svand_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_s64_z)))
+svint64_t svand_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svand_s16_z)))
+svint16_t svand_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandv_u8)))
+uint8_t svandv(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandv_u32)))
+uint32_t svandv(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandv_u64)))
+uint64_t svandv(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandv_u16)))
+uint16_t svandv(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandv_s8)))
+int8_t svandv(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandv_s32)))
+int32_t svandv(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandv_s64)))
+int64_t svandv(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandv_s16)))
+int16_t svandv(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_n_s8_m)))
+svint8_t svasr_m(svbool_t, svint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_n_s32_m)))
+svint32_t svasr_m(svbool_t, svint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_n_s64_m)))
+svint64_t svasr_m(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_n_s16_m)))
+svint16_t svasr_m(svbool_t, svint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_n_s8_x)))
+svint8_t svasr_x(svbool_t, svint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_n_s32_x)))
+svint32_t svasr_x(svbool_t, svint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_n_s64_x)))
+svint64_t svasr_x(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_n_s16_x)))
+svint16_t svasr_x(svbool_t, svint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_n_s8_z)))
+svint8_t svasr_z(svbool_t, svint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_n_s32_z)))
+svint32_t svasr_z(svbool_t, svint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_n_s64_z)))
+svint64_t svasr_z(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_n_s16_z)))
+svint16_t svasr_z(svbool_t, svint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_s8_m)))
+svint8_t svasr_m(svbool_t, svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_s32_m)))
+svint32_t svasr_m(svbool_t, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_s64_m)))
+svint64_t svasr_m(svbool_t, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_s16_m)))
+svint16_t svasr_m(svbool_t, svint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_s8_x)))
+svint8_t svasr_x(svbool_t, svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_s32_x)))
+svint32_t svasr_x(svbool_t, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_s64_x)))
+svint64_t svasr_x(svbool_t, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_s16_x)))
+svint16_t svasr_x(svbool_t, svint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_s8_z)))
+svint8_t svasr_z(svbool_t, svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_s32_z)))
+svint32_t svasr_z(svbool_t, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_s64_z)))
+svint64_t svasr_z(svbool_t, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_s16_z)))
+svint16_t svasr_z(svbool_t, svint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_n_s8_m)))
+svint8_t svasr_wide_m(svbool_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_n_s32_m)))
+svint32_t svasr_wide_m(svbool_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_n_s16_m)))
+svint16_t svasr_wide_m(svbool_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_n_s8_x)))
+svint8_t svasr_wide_x(svbool_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_n_s32_x)))
+svint32_t svasr_wide_x(svbool_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_n_s16_x)))
+svint16_t svasr_wide_x(svbool_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_n_s8_z)))
+svint8_t svasr_wide_z(svbool_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_n_s32_z)))
+svint32_t svasr_wide_z(svbool_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_n_s16_z)))
+svint16_t svasr_wide_z(svbool_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_s8_m)))
+svint8_t svasr_wide_m(svbool_t, svint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_s32_m)))
+svint32_t svasr_wide_m(svbool_t, svint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_s16_m)))
+svint16_t svasr_wide_m(svbool_t, svint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_s8_x)))
+svint8_t svasr_wide_x(svbool_t, svint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_s32_x)))
+svint32_t svasr_wide_x(svbool_t, svint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_s16_x)))
+svint16_t svasr_wide_x(svbool_t, svint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_s8_z)))
+svint8_t svasr_wide_z(svbool_t, svint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_s32_z)))
+svint32_t svasr_wide_z(svbool_t, svint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasr_wide_s16_z)))
+svint16_t svasr_wide_z(svbool_t, svint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasrd_n_s8_m)))
+svint8_t svasrd_m(svbool_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasrd_n_s32_m)))
+svint32_t svasrd_m(svbool_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasrd_n_s64_m)))
+svint64_t svasrd_m(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasrd_n_s16_m)))
+svint16_t svasrd_m(svbool_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasrd_n_s8_x)))
+svint8_t svasrd_x(svbool_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasrd_n_s32_x)))
+svint32_t svasrd_x(svbool_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasrd_n_s64_x)))
+svint64_t svasrd_x(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasrd_n_s16_x)))
+svint16_t svasrd_x(svbool_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasrd_n_s8_z)))
+svint8_t svasrd_z(svbool_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasrd_n_s32_z)))
+svint32_t svasrd_z(svbool_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasrd_n_s64_z)))
+svint64_t svasrd_z(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svasrd_n_s16_z)))
+svint16_t svasrd_z(svbool_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_b_z)))
+svbool_t svbic_z(svbool_t, svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_u8_m)))
+svuint8_t svbic_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_u32_m)))
+svuint32_t svbic_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_u64_m)))
+svuint64_t svbic_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_u16_m)))
+svuint16_t svbic_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_s8_m)))
+svint8_t svbic_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_s32_m)))
+svint32_t svbic_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_s64_m)))
+svint64_t svbic_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_s16_m)))
+svint16_t svbic_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_u8_x)))
+svuint8_t svbic_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_u32_x)))
+svuint32_t svbic_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_u64_x)))
+svuint64_t svbic_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_u16_x)))
+svuint16_t svbic_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_s8_x)))
+svint8_t svbic_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_s32_x)))
+svint32_t svbic_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_s64_x)))
+svint64_t svbic_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_s16_x)))
+svint16_t svbic_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_u8_z)))
+svuint8_t svbic_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_u32_z)))
+svuint32_t svbic_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_u64_z)))
+svuint64_t svbic_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_u16_z)))
+svuint16_t svbic_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_s8_z)))
+svint8_t svbic_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_s32_z)))
+svint32_t svbic_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_s64_z)))
+svint64_t svbic_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_n_s16_z)))
+svint16_t svbic_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_u8_m)))
+svuint8_t svbic_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_u32_m)))
+svuint32_t svbic_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_u64_m)))
+svuint64_t svbic_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_u16_m)))
+svuint16_t svbic_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_s8_m)))
+svint8_t svbic_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_s32_m)))
+svint32_t svbic_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_s64_m)))
+svint64_t svbic_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_s16_m)))
+svint16_t svbic_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_u8_x)))
+svuint8_t svbic_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_u32_x)))
+svuint32_t svbic_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_u64_x)))
+svuint64_t svbic_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_u16_x)))
+svuint16_t svbic_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_s8_x)))
+svint8_t svbic_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_s32_x)))
+svint32_t svbic_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_s64_x)))
+svint64_t svbic_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_s16_x)))
+svint16_t svbic_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_u8_z)))
+svuint8_t svbic_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_u32_z)))
+svuint32_t svbic_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_u64_z)))
+svuint64_t svbic_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_u16_z)))
+svuint16_t svbic_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_s8_z)))
+svint8_t svbic_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_s32_z)))
+svint32_t svbic_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_s64_z)))
+svint64_t svbic_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbic_s16_z)))
+svint16_t svbic_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbrka_b_m)))
+svbool_t svbrka_m(svbool_t, svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbrka_b_z)))
+svbool_t svbrka_z(svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbrkb_b_m)))
+svbool_t svbrkb_m(svbool_t, svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbrkb_b_z)))
+svbool_t svbrkb_z(svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbrkn_b_z)))
+svbool_t svbrkn_z(svbool_t, svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbrkpa_b_z)))
+svbool_t svbrkpa_z(svbool_t, svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbrkpb_b_z)))
+svbool_t svbrkpb_z(svbool_t, svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_f64_m)))
+svfloat64_t svcadd_m(svbool_t, svfloat64_t, svfloat64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_f32_m)))
+svfloat32_t svcadd_m(svbool_t, svfloat32_t, svfloat32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_f16_m)))
+svfloat16_t svcadd_m(svbool_t, svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_f64_x)))
+svfloat64_t svcadd_x(svbool_t, svfloat64_t, svfloat64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_f32_x)))
+svfloat32_t svcadd_x(svbool_t, svfloat32_t, svfloat32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_f16_x)))
+svfloat16_t svcadd_x(svbool_t, svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_f64_z)))
+svfloat64_t svcadd_z(svbool_t, svfloat64_t, svfloat64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_f32_z)))
+svfloat32_t svcadd_z(svbool_t, svfloat32_t, svfloat32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_f16_z)))
+svfloat16_t svcadd_z(svbool_t, svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_u8)))
+uint8_t svclasta(svbool_t, uint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_u32)))
+uint32_t svclasta(svbool_t, uint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_u64)))
+uint64_t svclasta(svbool_t, uint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_u16)))
+uint16_t svclasta(svbool_t, uint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_s8)))
+int8_t svclasta(svbool_t, int8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_f64)))
+float64_t svclasta(svbool_t, float64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_f32)))
+float32_t svclasta(svbool_t, float32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_f16)))
+float16_t svclasta(svbool_t, float16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_s32)))
+int32_t svclasta(svbool_t, int32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_s64)))
+int64_t svclasta(svbool_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_s16)))
+int16_t svclasta(svbool_t, int16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_u8)))
+svuint8_t svclasta(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_u32)))
+svuint32_t svclasta(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_u64)))
+svuint64_t svclasta(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_u16)))
+svuint16_t svclasta(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_s8)))
+svint8_t svclasta(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_f64)))
+svfloat64_t svclasta(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_f32)))
+svfloat32_t svclasta(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_f16)))
+svfloat16_t svclasta(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_s32)))
+svint32_t svclasta(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_s64)))
+svint64_t svclasta(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_s16)))
+svint16_t svclasta(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_u8)))
+uint8_t svclastb(svbool_t, uint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_u32)))
+uint32_t svclastb(svbool_t, uint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_u64)))
+uint64_t svclastb(svbool_t, uint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_u16)))
+uint16_t svclastb(svbool_t, uint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_s8)))
+int8_t svclastb(svbool_t, int8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_f64)))
+float64_t svclastb(svbool_t, float64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_f32)))
+float32_t svclastb(svbool_t, float32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_f16)))
+float16_t svclastb(svbool_t, float16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_s32)))
+int32_t svclastb(svbool_t, int32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_s64)))
+int64_t svclastb(svbool_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_s16)))
+int16_t svclastb(svbool_t, int16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_u8)))
+svuint8_t svclastb(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_u32)))
+svuint32_t svclastb(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_u64)))
+svuint64_t svclastb(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_u16)))
+svuint16_t svclastb(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_s8)))
+svint8_t svclastb(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_f64)))
+svfloat64_t svclastb(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_f32)))
+svfloat32_t svclastb(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_f16)))
+svfloat16_t svclastb(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_s32)))
+svint32_t svclastb(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_s64)))
+svint64_t svclastb(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_s16)))
+svint16_t svclastb(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcls_s8_m)))
+svuint8_t svcls_m(svuint8_t, svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcls_s32_m)))
+svuint32_t svcls_m(svuint32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcls_s64_m)))
+svuint64_t svcls_m(svuint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcls_s16_m)))
+svuint16_t svcls_m(svuint16_t, svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcls_s8_x)))
+svuint8_t svcls_x(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcls_s32_x)))
+svuint32_t svcls_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcls_s64_x)))
+svuint64_t svcls_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcls_s16_x)))
+svuint16_t svcls_x(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcls_s8_z)))
+svuint8_t svcls_z(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcls_s32_z)))
+svuint32_t svcls_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcls_s64_z)))
+svuint64_t svcls_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcls_s16_z)))
+svuint16_t svcls_z(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_u8_m)))
+svuint8_t svclz_m(svuint8_t, svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_u32_m)))
+svuint32_t svclz_m(svuint32_t, svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_u64_m)))
+svuint64_t svclz_m(svuint64_t, svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_u16_m)))
+svuint16_t svclz_m(svuint16_t, svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_s8_m)))
+svuint8_t svclz_m(svuint8_t, svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_s32_m)))
+svuint32_t svclz_m(svuint32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_s64_m)))
+svuint64_t svclz_m(svuint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_s16_m)))
+svuint16_t svclz_m(svuint16_t, svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_u8_x)))
+svuint8_t svclz_x(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_u32_x)))
+svuint32_t svclz_x(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_u64_x)))
+svuint64_t svclz_x(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_u16_x)))
+svuint16_t svclz_x(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_s8_x)))
+svuint8_t svclz_x(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_s32_x)))
+svuint32_t svclz_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_s64_x)))
+svuint64_t svclz_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_s16_x)))
+svuint16_t svclz_x(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_u8_z)))
+svuint8_t svclz_z(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_u32_z)))
+svuint32_t svclz_z(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_u64_z)))
+svuint64_t svclz_z(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_u16_z)))
+svuint16_t svclz_z(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_s8_z)))
+svuint8_t svclz_z(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_s32_z)))
+svuint32_t svclz_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_s64_z)))
+svuint64_t svclz_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclz_s16_z)))
+svuint16_t svclz_z(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_f64_m)))
+svfloat64_t svcmla_m(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_f32_m)))
+svfloat32_t svcmla_m(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_f16_m)))
+svfloat16_t svcmla_m(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_f64_x)))
+svfloat64_t svcmla_x(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_f32_x)))
+svfloat32_t svcmla_x(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_f16_x)))
+svfloat16_t svcmla_x(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_f64_z)))
+svfloat64_t svcmla_z(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_f32_z)))
+svfloat32_t svcmla_z(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_f16_z)))
+svfloat16_t svcmla_z(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_f32)))
+svfloat32_t svcmla_lane(svfloat32_t, svfloat32_t, svfloat32_t, uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_f16)))
+svfloat16_t svcmla_lane(svfloat16_t, svfloat16_t, svfloat16_t, uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_n_f64)))
+svbool_t svcmpeq(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_n_f32)))
+svbool_t svcmpeq(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_n_f16)))
+svbool_t svcmpeq(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_n_u8)))
+svbool_t svcmpeq(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_n_u32)))
+svbool_t svcmpeq(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_n_u64)))
+svbool_t svcmpeq(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_n_u16)))
+svbool_t svcmpeq(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_n_s8)))
+svbool_t svcmpeq(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_n_s32)))
+svbool_t svcmpeq(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_n_s64)))
+svbool_t svcmpeq(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_n_s16)))
+svbool_t svcmpeq(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_u8)))
+svbool_t svcmpeq(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_u32)))
+svbool_t svcmpeq(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_u64)))
+svbool_t svcmpeq(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_u16)))
+svbool_t svcmpeq(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_s8)))
+svbool_t svcmpeq(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_s32)))
+svbool_t svcmpeq(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_s64)))
+svbool_t svcmpeq(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_s16)))
+svbool_t svcmpeq(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_f64)))
+svbool_t svcmpeq(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_f32)))
+svbool_t svcmpeq(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_f16)))
+svbool_t svcmpeq(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_wide_n_s8)))
+svbool_t svcmpeq_wide(svbool_t, svint8_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_wide_n_s32)))
+svbool_t svcmpeq_wide(svbool_t, svint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_wide_n_s16)))
+svbool_t svcmpeq_wide(svbool_t, svint16_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_wide_s8)))
+svbool_t svcmpeq_wide(svbool_t, svint8_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_wide_s32)))
+svbool_t svcmpeq_wide(svbool_t, svint32_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpeq_wide_s16)))
+svbool_t svcmpeq_wide(svbool_t, svint16_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_n_f64)))
+svbool_t svcmpge(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_n_f32)))
+svbool_t svcmpge(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_n_f16)))
+svbool_t svcmpge(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_n_s8)))
+svbool_t svcmpge(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_n_s32)))
+svbool_t svcmpge(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_n_s64)))
+svbool_t svcmpge(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_n_s16)))
+svbool_t svcmpge(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_n_u8)))
+svbool_t svcmpge(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_n_u32)))
+svbool_t svcmpge(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_n_u64)))
+svbool_t svcmpge(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_n_u16)))
+svbool_t svcmpge(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_s8)))
+svbool_t svcmpge(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_s32)))
+svbool_t svcmpge(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_s64)))
+svbool_t svcmpge(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_s16)))
+svbool_t svcmpge(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_f64)))
+svbool_t svcmpge(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_f32)))
+svbool_t svcmpge(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_f16)))
+svbool_t svcmpge(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_u8)))
+svbool_t svcmpge(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_u32)))
+svbool_t svcmpge(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_u64)))
+svbool_t svcmpge(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_u16)))
+svbool_t svcmpge(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_wide_n_s8)))
+svbool_t svcmpge_wide(svbool_t, svint8_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_wide_n_s32)))
+svbool_t svcmpge_wide(svbool_t, svint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_wide_n_s16)))
+svbool_t svcmpge_wide(svbool_t, svint16_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_wide_n_u8)))
+svbool_t svcmpge_wide(svbool_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_wide_n_u32)))
+svbool_t svcmpge_wide(svbool_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_wide_n_u16)))
+svbool_t svcmpge_wide(svbool_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_wide_s8)))
+svbool_t svcmpge_wide(svbool_t, svint8_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_wide_s32)))
+svbool_t svcmpge_wide(svbool_t, svint32_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_wide_s16)))
+svbool_t svcmpge_wide(svbool_t, svint16_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_wide_u8)))
+svbool_t svcmpge_wide(svbool_t, svuint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_wide_u32)))
+svbool_t svcmpge_wide(svbool_t, svuint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpge_wide_u16)))
+svbool_t svcmpge_wide(svbool_t, svuint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_n_f64)))
+svbool_t svcmpgt(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_n_f32)))
+svbool_t svcmpgt(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_n_f16)))
+svbool_t svcmpgt(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_n_s8)))
+svbool_t svcmpgt(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_n_s32)))
+svbool_t svcmpgt(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_n_s64)))
+svbool_t svcmpgt(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_n_s16)))
+svbool_t svcmpgt(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_n_u8)))
+svbool_t svcmpgt(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_n_u32)))
+svbool_t svcmpgt(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_n_u64)))
+svbool_t svcmpgt(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_n_u16)))
+svbool_t svcmpgt(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_s8)))
+svbool_t svcmpgt(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_s32)))
+svbool_t svcmpgt(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_s64)))
+svbool_t svcmpgt(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_s16)))
+svbool_t svcmpgt(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_f64)))
+svbool_t svcmpgt(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_f32)))
+svbool_t svcmpgt(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_f16)))
+svbool_t svcmpgt(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_u8)))
+svbool_t svcmpgt(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_u32)))
+svbool_t svcmpgt(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_u64)))
+svbool_t svcmpgt(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_u16)))
+svbool_t svcmpgt(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_wide_n_s8)))
+svbool_t svcmpgt_wide(svbool_t, svint8_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_wide_n_s32)))
+svbool_t svcmpgt_wide(svbool_t, svint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_wide_n_s16)))
+svbool_t svcmpgt_wide(svbool_t, svint16_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_wide_n_u8)))
+svbool_t svcmpgt_wide(svbool_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_wide_n_u32)))
+svbool_t svcmpgt_wide(svbool_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_wide_n_u16)))
+svbool_t svcmpgt_wide(svbool_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_wide_s8)))
+svbool_t svcmpgt_wide(svbool_t, svint8_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_wide_s32)))
+svbool_t svcmpgt_wide(svbool_t, svint32_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_wide_s16)))
+svbool_t svcmpgt_wide(svbool_t, svint16_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_wide_u8)))
+svbool_t svcmpgt_wide(svbool_t, svuint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_wide_u32)))
+svbool_t svcmpgt_wide(svbool_t, svuint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpgt_wide_u16)))
+svbool_t svcmpgt_wide(svbool_t, svuint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_n_f64)))
+svbool_t svcmple(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_n_f32)))
+svbool_t svcmple(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_n_f16)))
+svbool_t svcmple(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_n_s8)))
+svbool_t svcmple(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_n_s32)))
+svbool_t svcmple(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_n_s64)))
+svbool_t svcmple(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_n_s16)))
+svbool_t svcmple(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_n_u8)))
+svbool_t svcmple(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_n_u32)))
+svbool_t svcmple(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_n_u64)))
+svbool_t svcmple(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_n_u16)))
+svbool_t svcmple(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_s8)))
+svbool_t svcmple(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_s32)))
+svbool_t svcmple(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_s64)))
+svbool_t svcmple(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_s16)))
+svbool_t svcmple(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_f64)))
+svbool_t svcmple(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_f32)))
+svbool_t svcmple(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_f16)))
+svbool_t svcmple(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_u8)))
+svbool_t svcmple(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_u32)))
+svbool_t svcmple(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_u64)))
+svbool_t svcmple(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_u16)))
+svbool_t svcmple(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_wide_n_s8)))
+svbool_t svcmple_wide(svbool_t, svint8_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_wide_n_s32)))
+svbool_t svcmple_wide(svbool_t, svint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_wide_n_s16)))
+svbool_t svcmple_wide(svbool_t, svint16_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_wide_n_u8)))
+svbool_t svcmple_wide(svbool_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_wide_n_u32)))
+svbool_t svcmple_wide(svbool_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_wide_n_u16)))
+svbool_t svcmple_wide(svbool_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_wide_s8)))
+svbool_t svcmple_wide(svbool_t, svint8_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_wide_s32)))
+svbool_t svcmple_wide(svbool_t, svint32_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_wide_s16)))
+svbool_t svcmple_wide(svbool_t, svint16_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_wide_u8)))
+svbool_t svcmple_wide(svbool_t, svuint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_wide_u32)))
+svbool_t svcmple_wide(svbool_t, svuint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmple_wide_u16)))
+svbool_t svcmple_wide(svbool_t, svuint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_n_u8)))
+svbool_t svcmplt(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_n_u32)))
+svbool_t svcmplt(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_n_u64)))
+svbool_t svcmplt(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_n_u16)))
+svbool_t svcmplt(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_n_f64)))
+svbool_t svcmplt(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_n_f32)))
+svbool_t svcmplt(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_n_f16)))
+svbool_t svcmplt(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_n_s8)))
+svbool_t svcmplt(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_n_s32)))
+svbool_t svcmplt(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_n_s64)))
+svbool_t svcmplt(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_n_s16)))
+svbool_t svcmplt(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_u8)))
+svbool_t svcmplt(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_u32)))
+svbool_t svcmplt(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_u64)))
+svbool_t svcmplt(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_u16)))
+svbool_t svcmplt(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_s8)))
+svbool_t svcmplt(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_s32)))
+svbool_t svcmplt(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_s64)))
+svbool_t svcmplt(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_s16)))
+svbool_t svcmplt(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_f64)))
+svbool_t svcmplt(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_f32)))
+svbool_t svcmplt(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_f16)))
+svbool_t svcmplt(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_wide_n_u8)))
+svbool_t svcmplt_wide(svbool_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_wide_n_u32)))
+svbool_t svcmplt_wide(svbool_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_wide_n_u16)))
+svbool_t svcmplt_wide(svbool_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_wide_n_s8)))
+svbool_t svcmplt_wide(svbool_t, svint8_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_wide_n_s32)))
+svbool_t svcmplt_wide(svbool_t, svint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_wide_n_s16)))
+svbool_t svcmplt_wide(svbool_t, svint16_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_wide_u8)))
+svbool_t svcmplt_wide(svbool_t, svuint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_wide_u32)))
+svbool_t svcmplt_wide(svbool_t, svuint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_wide_u16)))
+svbool_t svcmplt_wide(svbool_t, svuint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_wide_s8)))
+svbool_t svcmplt_wide(svbool_t, svint8_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_wide_s32)))
+svbool_t svcmplt_wide(svbool_t, svint32_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmplt_wide_s16)))
+svbool_t svcmplt_wide(svbool_t, svint16_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_n_f64)))
+svbool_t svcmpne(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_n_f32)))
+svbool_t svcmpne(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_n_f16)))
+svbool_t svcmpne(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_n_u8)))
+svbool_t svcmpne(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_n_u32)))
+svbool_t svcmpne(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_n_u64)))
+svbool_t svcmpne(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_n_u16)))
+svbool_t svcmpne(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_n_s8)))
+svbool_t svcmpne(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_n_s32)))
+svbool_t svcmpne(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_n_s64)))
+svbool_t svcmpne(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_n_s16)))
+svbool_t svcmpne(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_u8)))
+svbool_t svcmpne(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_u32)))
+svbool_t svcmpne(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_u64)))
+svbool_t svcmpne(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_u16)))
+svbool_t svcmpne(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_s8)))
+svbool_t svcmpne(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_s32)))
+svbool_t svcmpne(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_s64)))
+svbool_t svcmpne(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_s16)))
+svbool_t svcmpne(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_f64)))
+svbool_t svcmpne(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_f32)))
+svbool_t svcmpne(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_f16)))
+svbool_t svcmpne(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_wide_n_s8)))
+svbool_t svcmpne_wide(svbool_t, svint8_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_wide_n_s32)))
+svbool_t svcmpne_wide(svbool_t, svint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_wide_n_s16)))
+svbool_t svcmpne_wide(svbool_t, svint16_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_wide_s8)))
+svbool_t svcmpne_wide(svbool_t, svint8_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_wide_s32)))
+svbool_t svcmpne_wide(svbool_t, svint32_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpne_wide_s16)))
+svbool_t svcmpne_wide(svbool_t, svint16_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpuo_n_f64)))
+svbool_t svcmpuo(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpuo_n_f32)))
+svbool_t svcmpuo(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpuo_n_f16)))
+svbool_t svcmpuo(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpuo_f64)))
+svbool_t svcmpuo(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpuo_f32)))
+svbool_t svcmpuo(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmpuo_f16)))
+svbool_t svcmpuo(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_u8_m)))
+svuint8_t svcnot_m(svuint8_t, svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_u32_m)))
+svuint32_t svcnot_m(svuint32_t, svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_u64_m)))
+svuint64_t svcnot_m(svuint64_t, svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_u16_m)))
+svuint16_t svcnot_m(svuint16_t, svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_s8_m)))
+svint8_t svcnot_m(svint8_t, svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_s32_m)))
+svint32_t svcnot_m(svint32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_s64_m)))
+svint64_t svcnot_m(svint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_s16_m)))
+svint16_t svcnot_m(svint16_t, svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_u8_x)))
+svuint8_t svcnot_x(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_u32_x)))
+svuint32_t svcnot_x(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_u64_x)))
+svuint64_t svcnot_x(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_u16_x)))
+svuint16_t svcnot_x(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_s8_x)))
+svint8_t svcnot_x(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_s32_x)))
+svint32_t svcnot_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_s64_x)))
+svint64_t svcnot_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_s16_x)))
+svint16_t svcnot_x(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_u8_z)))
+svuint8_t svcnot_z(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_u32_z)))
+svuint32_t svcnot_z(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_u64_z)))
+svuint64_t svcnot_z(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_u16_z)))
+svuint16_t svcnot_z(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_s8_z)))
+svint8_t svcnot_z(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_s32_z)))
+svint32_t svcnot_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_s64_z)))
+svint64_t svcnot_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnot_s16_z)))
+svint16_t svcnot_z(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u8_m)))
+svuint8_t svcnt_m(svuint8_t, svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u32_m)))
+svuint32_t svcnt_m(svuint32_t, svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u64_m)))
+svuint64_t svcnt_m(svuint64_t, svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u16_m)))
+svuint16_t svcnt_m(svuint16_t, svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s8_m)))
+svuint8_t svcnt_m(svuint8_t, svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_f64_m)))
+svuint64_t svcnt_m(svuint64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_f32_m)))
+svuint32_t svcnt_m(svuint32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_f16_m)))
+svuint16_t svcnt_m(svuint16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s32_m)))
+svuint32_t svcnt_m(svuint32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s64_m)))
+svuint64_t svcnt_m(svuint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s16_m)))
+svuint16_t svcnt_m(svuint16_t, svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u8_x)))
+svuint8_t svcnt_x(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u32_x)))
+svuint32_t svcnt_x(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u64_x)))
+svuint64_t svcnt_x(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u16_x)))
+svuint16_t svcnt_x(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s8_x)))
+svuint8_t svcnt_x(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_f64_x)))
+svuint64_t svcnt_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_f32_x)))
+svuint32_t svcnt_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_f16_x)))
+svuint16_t svcnt_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s32_x)))
+svuint32_t svcnt_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s64_x)))
+svuint64_t svcnt_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s16_x)))
+svuint16_t svcnt_x(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u8_z)))
+svuint8_t svcnt_z(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u32_z)))
+svuint32_t svcnt_z(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u64_z)))
+svuint64_t svcnt_z(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u16_z)))
+svuint16_t svcnt_z(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s8_z)))
+svuint8_t svcnt_z(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_f64_z)))
+svuint64_t svcnt_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_f32_z)))
+svuint32_t svcnt_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_f16_z)))
+svuint16_t svcnt_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s32_z)))
+svuint32_t svcnt_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s64_z)))
+svuint64_t svcnt_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s16_z)))
+svuint16_t svcnt_z(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcompact_u32)))
+svuint32_t svcompact(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcompact_u64)))
+svuint64_t svcompact(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcompact_f64)))
+svfloat64_t svcompact(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcompact_f32)))
+svfloat32_t svcompact(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcompact_s32)))
+svint32_t svcompact(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcompact_s64)))
+svint64_t svcompact(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_u8)))
+svuint8x2_t svcreate2(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_u32)))
+svuint32x2_t svcreate2(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_u64)))
+svuint64x2_t svcreate2(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_u16)))
+svuint16x2_t svcreate2(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_s8)))
+svint8x2_t svcreate2(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_f64)))
+svfloat64x2_t svcreate2(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_f32)))
+svfloat32x2_t svcreate2(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_f16)))
+svfloat16x2_t svcreate2(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_s32)))
+svint32x2_t svcreate2(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_s64)))
+svint64x2_t svcreate2(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_s16)))
+svint16x2_t svcreate2(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_u8)))
+svuint8x3_t svcreate3(svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_u32)))
+svuint32x3_t svcreate3(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_u64)))
+svuint64x3_t svcreate3(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_u16)))
+svuint16x3_t svcreate3(svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_s8)))
+svint8x3_t svcreate3(svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_f64)))
+svfloat64x3_t svcreate3(svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_f32)))
+svfloat32x3_t svcreate3(svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_f16)))
+svfloat16x3_t svcreate3(svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_s32)))
+svint32x3_t svcreate3(svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_s64)))
+svint64x3_t svcreate3(svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_s16)))
+svint16x3_t svcreate3(svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_u8)))
+svuint8x4_t svcreate4(svuint8_t, svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_u32)))
+svuint32x4_t svcreate4(svuint32_t, svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_u64)))
+svuint64x4_t svcreate4(svuint64_t, svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_u16)))
+svuint16x4_t svcreate4(svuint16_t, svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_s8)))
+svint8x4_t svcreate4(svint8_t, svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_f64)))
+svfloat64x4_t svcreate4(svfloat64_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_f32)))
+svfloat32x4_t svcreate4(svfloat32_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_f16)))
+svfloat16x4_t svcreate4(svfloat16_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_s32)))
+svint32x4_t svcreate4(svint32_t, svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_s64)))
+svint64x4_t svcreate4(svint64_t, svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_s16)))
+svint16x4_t svcreate4(svint16_t, svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_f32_m)))
+svfloat16_t svcvt_f16_m(svfloat16_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_f32_x)))
+svfloat16_t svcvt_f16_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_f32_z)))
+svfloat16_t svcvt_f16_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_f64_m)))
+svfloat16_t svcvt_f16_m(svfloat16_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_f64_x)))
+svfloat16_t svcvt_f16_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_f64_z)))
+svfloat16_t svcvt_f16_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_s16_m)))
+svfloat16_t svcvt_f16_m(svfloat16_t, svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_s16_x)))
+svfloat16_t svcvt_f16_x(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_s16_z)))
+svfloat16_t svcvt_f16_z(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_s32_m)))
+svfloat16_t svcvt_f16_m(svfloat16_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_s32_x)))
+svfloat16_t svcvt_f16_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_s32_z)))
+svfloat16_t svcvt_f16_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_s64_m)))
+svfloat16_t svcvt_f16_m(svfloat16_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_s64_x)))
+svfloat16_t svcvt_f16_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_s64_z)))
+svfloat16_t svcvt_f16_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_u16_m)))
+svfloat16_t svcvt_f16_m(svfloat16_t, svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_u16_x)))
+svfloat16_t svcvt_f16_x(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_u16_z)))
+svfloat16_t svcvt_f16_z(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_u32_m)))
+svfloat16_t svcvt_f16_m(svfloat16_t, svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_u32_x)))
+svfloat16_t svcvt_f16_x(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_u32_z)))
+svfloat16_t svcvt_f16_z(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_u64_m)))
+svfloat16_t svcvt_f16_m(svfloat16_t, svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_u64_x)))
+svfloat16_t svcvt_f16_x(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_u64_z)))
+svfloat16_t svcvt_f16_z(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_f16_m)))
+svfloat32_t svcvt_f32_m(svfloat32_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_f16_x)))
+svfloat32_t svcvt_f32_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_f16_z)))
+svfloat32_t svcvt_f32_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_f64_m)))
+svfloat32_t svcvt_f32_m(svfloat32_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_f64_x)))
+svfloat32_t svcvt_f32_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_f64_z)))
+svfloat32_t svcvt_f32_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_s32_m)))
+svfloat32_t svcvt_f32_m(svfloat32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_s32_x)))
+svfloat32_t svcvt_f32_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_s32_z)))
+svfloat32_t svcvt_f32_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_s64_m)))
+svfloat32_t svcvt_f32_m(svfloat32_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_s64_x)))
+svfloat32_t svcvt_f32_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_s64_z)))
+svfloat32_t svcvt_f32_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_u32_m)))
+svfloat32_t svcvt_f32_m(svfloat32_t, svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_u32_x)))
+svfloat32_t svcvt_f32_x(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_u32_z)))
+svfloat32_t svcvt_f32_z(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_u64_m)))
+svfloat32_t svcvt_f32_m(svfloat32_t, svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_u64_x)))
+svfloat32_t svcvt_f32_x(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f32_u64_z)))
+svfloat32_t svcvt_f32_z(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_f16_m)))
+svfloat64_t svcvt_f64_m(svfloat64_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_f16_x)))
+svfloat64_t svcvt_f64_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_f16_z)))
+svfloat64_t svcvt_f64_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_f32_m)))
+svfloat64_t svcvt_f64_m(svfloat64_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_f32_x)))
+svfloat64_t svcvt_f64_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_f32_z)))
+svfloat64_t svcvt_f64_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_s32_m)))
+svfloat64_t svcvt_f64_m(svfloat64_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_s32_x)))
+svfloat64_t svcvt_f64_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_s32_z)))
+svfloat64_t svcvt_f64_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_s64_m)))
+svfloat64_t svcvt_f64_m(svfloat64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_s64_x)))
+svfloat64_t svcvt_f64_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_s64_z)))
+svfloat64_t svcvt_f64_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_u32_m)))
+svfloat64_t svcvt_f64_m(svfloat64_t, svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_u32_x)))
+svfloat64_t svcvt_f64_x(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_u32_z)))
+svfloat64_t svcvt_f64_z(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_u64_m)))
+svfloat64_t svcvt_f64_m(svfloat64_t, svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_u64_x)))
+svfloat64_t svcvt_f64_x(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f64_u64_z)))
+svfloat64_t svcvt_f64_z(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s16_f16_m)))
+svint16_t svcvt_s16_m(svint16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s16_f16_x)))
+svint16_t svcvt_s16_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s16_f16_z)))
+svint16_t svcvt_s16_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s32_f16_m)))
+svint32_t svcvt_s32_m(svint32_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s32_f16_x)))
+svint32_t svcvt_s32_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s32_f16_z)))
+svint32_t svcvt_s32_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s32_f32_m)))
+svint32_t svcvt_s32_m(svint32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s32_f32_x)))
+svint32_t svcvt_s32_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s32_f32_z)))
+svint32_t svcvt_s32_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s32_f64_m)))
+svint32_t svcvt_s32_m(svint32_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s32_f64_x)))
+svint32_t svcvt_s32_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s32_f64_z)))
+svint32_t svcvt_s32_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s64_f16_m)))
+svint64_t svcvt_s64_m(svint64_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s64_f16_x)))
+svint64_t svcvt_s64_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s64_f16_z)))
+svint64_t svcvt_s64_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s64_f32_m)))
+svint64_t svcvt_s64_m(svint64_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s64_f32_x)))
+svint64_t svcvt_s64_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s64_f32_z)))
+svint64_t svcvt_s64_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s64_f64_m)))
+svint64_t svcvt_s64_m(svint64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s64_f64_x)))
+svint64_t svcvt_s64_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_s64_f64_z)))
+svint64_t svcvt_s64_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u16_f16_m)))
+svuint16_t svcvt_u16_m(svuint16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u16_f16_x)))
+svuint16_t svcvt_u16_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u16_f16_z)))
+svuint16_t svcvt_u16_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u32_f16_m)))
+svuint32_t svcvt_u32_m(svuint32_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u32_f16_x)))
+svuint32_t svcvt_u32_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u32_f16_z)))
+svuint32_t svcvt_u32_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u32_f32_m)))
+svuint32_t svcvt_u32_m(svuint32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u32_f32_x)))
+svuint32_t svcvt_u32_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u32_f32_z)))
+svuint32_t svcvt_u32_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u32_f64_m)))
+svuint32_t svcvt_u32_m(svuint32_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u32_f64_x)))
+svuint32_t svcvt_u32_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u32_f64_z)))
+svuint32_t svcvt_u32_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u64_f16_m)))
+svuint64_t svcvt_u64_m(svuint64_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u64_f16_x)))
+svuint64_t svcvt_u64_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u64_f16_z)))
+svuint64_t svcvt_u64_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u64_f32_m)))
+svuint64_t svcvt_u64_m(svuint64_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u64_f32_x)))
+svuint64_t svcvt_u64_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u64_f32_z)))
+svuint64_t svcvt_u64_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u64_f64_m)))
+svuint64_t svcvt_u64_m(svuint64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u64_f64_x)))
+svuint64_t svcvt_u64_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_u64_f64_z)))
+svuint64_t svcvt_u64_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_f64_m)))
+svfloat64_t svdiv_m(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_f32_m)))
+svfloat32_t svdiv_m(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_f16_m)))
+svfloat16_t svdiv_m(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_f64_x)))
+svfloat64_t svdiv_x(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_f32_x)))
+svfloat32_t svdiv_x(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_f16_x)))
+svfloat16_t svdiv_x(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_f64_z)))
+svfloat64_t svdiv_z(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_f32_z)))
+svfloat32_t svdiv_z(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_f16_z)))
+svfloat16_t svdiv_z(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_s32_m)))
+svint32_t svdiv_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_s64_m)))
+svint64_t svdiv_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_s32_x)))
+svint32_t svdiv_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_s64_x)))
+svint64_t svdiv_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_s32_z)))
+svint32_t svdiv_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_s64_z)))
+svint64_t svdiv_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_u32_m)))
+svuint32_t svdiv_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_u64_m)))
+svuint64_t svdiv_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_u32_x)))
+svuint32_t svdiv_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_u64_x)))
+svuint64_t svdiv_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_u32_z)))
+svuint32_t svdiv_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_n_u64_z)))
+svuint64_t svdiv_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_f64_m)))
+svfloat64_t svdiv_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_f32_m)))
+svfloat32_t svdiv_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_f16_m)))
+svfloat16_t svdiv_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_f64_x)))
+svfloat64_t svdiv_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_f32_x)))
+svfloat32_t svdiv_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_f16_x)))
+svfloat16_t svdiv_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_f64_z)))
+svfloat64_t svdiv_z(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_f32_z)))
+svfloat32_t svdiv_z(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_f16_z)))
+svfloat16_t svdiv_z(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_s32_m)))
+svint32_t svdiv_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_s64_m)))
+svint64_t svdiv_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_s32_x)))
+svint32_t svdiv_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_s64_x)))
+svint64_t svdiv_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_s32_z)))
+svint32_t svdiv_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_s64_z)))
+svint64_t svdiv_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_u32_m)))
+svuint32_t svdiv_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_u64_m)))
+svuint64_t svdiv_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_u32_x)))
+svuint32_t svdiv_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_u64_x)))
+svuint64_t svdiv_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_u32_z)))
+svuint32_t svdiv_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdiv_u64_z)))
+svuint64_t svdiv_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_f64_m)))
+svfloat64_t svdivr_m(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_f32_m)))
+svfloat32_t svdivr_m(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_f16_m)))
+svfloat16_t svdivr_m(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_f64_x)))
+svfloat64_t svdivr_x(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_f32_x)))
+svfloat32_t svdivr_x(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_f16_x)))
+svfloat16_t svdivr_x(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_f64_z)))
+svfloat64_t svdivr_z(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_f32_z)))
+svfloat32_t svdivr_z(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_f16_z)))
+svfloat16_t svdivr_z(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_s32_m)))
+svint32_t svdivr_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_s64_m)))
+svint64_t svdivr_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_s32_x)))
+svint32_t svdivr_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_s64_x)))
+svint64_t svdivr_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_s32_z)))
+svint32_t svdivr_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_s64_z)))
+svint64_t svdivr_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_u32_m)))
+svuint32_t svdivr_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_u64_m)))
+svuint64_t svdivr_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_u32_x)))
+svuint32_t svdivr_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_u64_x)))
+svuint64_t svdivr_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_u32_z)))
+svuint32_t svdivr_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_n_u64_z)))
+svuint64_t svdivr_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_f64_m)))
+svfloat64_t svdivr_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_f32_m)))
+svfloat32_t svdivr_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_f16_m)))
+svfloat16_t svdivr_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_f64_x)))
+svfloat64_t svdivr_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_f32_x)))
+svfloat32_t svdivr_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_f16_x)))
+svfloat16_t svdivr_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_f64_z)))
+svfloat64_t svdivr_z(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_f32_z)))
+svfloat32_t svdivr_z(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_f16_z)))
+svfloat16_t svdivr_z(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_s32_m)))
+svint32_t svdivr_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_s64_m)))
+svint64_t svdivr_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_s32_x)))
+svint32_t svdivr_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_s64_x)))
+svint64_t svdivr_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_s32_z)))
+svint32_t svdivr_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_s64_z)))
+svint64_t svdivr_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_u32_m)))
+svuint32_t svdivr_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_u64_m)))
+svuint64_t svdivr_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_u32_x)))
+svuint32_t svdivr_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_u64_x)))
+svuint64_t svdivr_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_u32_z)))
+svuint32_t svdivr_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdivr_u64_z)))
+svuint64_t svdivr_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_n_s32)))
+svint32_t svdot(svint32_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_n_s64)))
+svint64_t svdot(svint64_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_n_u32)))
+svuint32_t svdot(svuint32_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_n_u64)))
+svuint64_t svdot(svuint64_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_s32)))
+svint32_t svdot(svint32_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_s64)))
+svint64_t svdot(svint64_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_u32)))
+svuint32_t svdot(svuint32_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_u64)))
+svuint64_t svdot(svuint64_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_s32)))
+svint32_t svdot_lane(svint32_t, svint8_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_s64)))
+svint64_t svdot_lane(svint64_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_u32)))
+svuint32_t svdot_lane(svuint32_t, svuint8_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_u64)))
+svuint64_t svdot_lane(svuint64_t, svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u8)))
+svuint8_t svdup_u8(uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u32)))
+svuint32_t svdup_u32(uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u64)))
+svuint64_t svdup_u64(uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u16)))
+svuint16_t svdup_u16(uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s8)))
+svint8_t svdup_s8(int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f64)))
+svfloat64_t svdup_f64(float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f32)))
+svfloat32_t svdup_f32(float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f16)))
+svfloat16_t svdup_f16(float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s32)))
+svint32_t svdup_s32(int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s64)))
+svint64_t svdup_s64(int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s16)))
+svint16_t svdup_s16(int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u8_m)))
+svuint8_t svdup_u8_m(svuint8_t, svbool_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u32_m)))
+svuint32_t svdup_u32_m(svuint32_t, svbool_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u64_m)))
+svuint64_t svdup_u64_m(svuint64_t, svbool_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u16_m)))
+svuint16_t svdup_u16_m(svuint16_t, svbool_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s8_m)))
+svint8_t svdup_s8_m(svint8_t, svbool_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f64_m)))
+svfloat64_t svdup_f64_m(svfloat64_t, svbool_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f32_m)))
+svfloat32_t svdup_f32_m(svfloat32_t, svbool_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f16_m)))
+svfloat16_t svdup_f16_m(svfloat16_t, svbool_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s32_m)))
+svint32_t svdup_s32_m(svint32_t, svbool_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s64_m)))
+svint64_t svdup_s64_m(svint64_t, svbool_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s16_m)))
+svint16_t svdup_s16_m(svint16_t, svbool_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_b8)))
+svbool_t svdup_b8(bool);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_b32)))
+svbool_t svdup_b32(bool);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_b64)))
+svbool_t svdup_b64(bool);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_b16)))
+svbool_t svdup_b16(bool);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u8_x)))
+svuint8_t svdup_u8_x(svbool_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u32_x)))
+svuint32_t svdup_u32_x(svbool_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u64_x)))
+svuint64_t svdup_u64_x(svbool_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u16_x)))
+svuint16_t svdup_u16_x(svbool_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s8_x)))
+svint8_t svdup_s8_x(svbool_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f64_x)))
+svfloat64_t svdup_f64_x(svbool_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f32_x)))
+svfloat32_t svdup_f32_x(svbool_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f16_x)))
+svfloat16_t svdup_f16_x(svbool_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s32_x)))
+svint32_t svdup_s32_x(svbool_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s64_x)))
+svint64_t svdup_s64_x(svbool_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s16_x)))
+svint16_t svdup_s16_x(svbool_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u8_z)))
+svuint8_t svdup_u8_z(svbool_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u32_z)))
+svuint32_t svdup_u32_z(svbool_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u64_z)))
+svuint64_t svdup_u64_z(svbool_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u16_z)))
+svuint16_t svdup_u16_z(svbool_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s8_z)))
+svint8_t svdup_s8_z(svbool_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f64_z)))
+svfloat64_t svdup_f64_z(svbool_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f32_z)))
+svfloat32_t svdup_f32_z(svbool_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f16_z)))
+svfloat16_t svdup_f16_z(svbool_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s32_z)))
+svint32_t svdup_s32_z(svbool_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s64_z)))
+svint64_t svdup_s64_z(svbool_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s16_z)))
+svint16_t svdup_s16_z(svbool_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_u8)))
+svuint8_t svdup_lane(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_u32)))
+svuint32_t svdup_lane(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_u64)))
+svuint64_t svdup_lane(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_u16)))
+svuint16_t svdup_lane(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_s8)))
+svint8_t svdup_lane(svint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_f64)))
+svfloat64_t svdup_lane(svfloat64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_f32)))
+svfloat32_t svdup_lane(svfloat32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_f16)))
+svfloat16_t svdup_lane(svfloat16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_s32)))
+svint32_t svdup_lane(svint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_s64)))
+svint64_t svdup_lane(svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_s16)))
+svint16_t svdup_lane(svint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_u16)))
+svuint16_t svdupq_u16(uint16_t, uint16_t, uint16_t, uint16_t, uint16_t, uint16_t, uint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_f16)))
+svfloat16_t svdupq_f16(float16_t, float16_t, float16_t, float16_t, float16_t, float16_t, float16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_s16)))
+svint16_t svdupq_s16(int16_t, int16_t, int16_t, int16_t, int16_t, int16_t, int16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_u32)))
+svuint32_t svdupq_u32(uint32_t, uint32_t, uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_f32)))
+svfloat32_t svdupq_f32(float32_t, float32_t, float32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_s32)))
+svint32_t svdupq_s32(int32_t, int32_t, int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_u64)))
+svuint64_t svdupq_u64(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_f64)))
+svfloat64_t svdupq_f64(float64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_s64)))
+svint64_t svdupq_s64(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_u8)))
+svuint8_t svdupq_u8(uint8_t, uint8_t, uint8_t, uint8_t, uint8_t, uint8_t, uint8_t, uint8_t, uint8_t, uint8_t, uint8_t, uint8_t, uint8_t, uint8_t, uint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_s8)))
+svint8_t svdupq_s8(int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_b16)))
+svbool_t svdupq_b16(bool, bool, bool, bool, bool, bool, bool, bool);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_b32)))
+svbool_t svdupq_b32(bool, bool, bool, bool);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_b64)))
+svbool_t svdupq_b64(bool, bool);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_b8)))
+svbool_t svdupq_b8(bool, bool, bool, bool, bool, bool, bool, bool, bool, bool, bool, bool, bool, bool, bool, bool);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_u8)))
+svuint8_t svdupq_lane(svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_u32)))
+svuint32_t svdupq_lane(svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_u64)))
+svuint64_t svdupq_lane(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_u16)))
+svuint16_t svdupq_lane(svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_s8)))
+svint8_t svdupq_lane(svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_f64)))
+svfloat64_t svdupq_lane(svfloat64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_f32)))
+svfloat32_t svdupq_lane(svfloat32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_f16)))
+svfloat16_t svdupq_lane(svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_s32)))
+svint32_t svdupq_lane(svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_s64)))
+svint64_t svdupq_lane(svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_s16)))
+svint16_t svdupq_lane(svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_b_z)))
+svbool_t sveor_z(svbool_t, svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_u8_m)))
+svuint8_t sveor_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_u32_m)))
+svuint32_t sveor_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_u64_m)))
+svuint64_t sveor_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_u16_m)))
+svuint16_t sveor_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_s8_m)))
+svint8_t sveor_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_s32_m)))
+svint32_t sveor_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_s64_m)))
+svint64_t sveor_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_s16_m)))
+svint16_t sveor_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_u8_x)))
+svuint8_t sveor_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_u32_x)))
+svuint32_t sveor_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_u64_x)))
+svuint64_t sveor_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_u16_x)))
+svuint16_t sveor_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_s8_x)))
+svint8_t sveor_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_s32_x)))
+svint32_t sveor_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_s64_x)))
+svint64_t sveor_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_s16_x)))
+svint16_t sveor_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_u8_z)))
+svuint8_t sveor_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_u32_z)))
+svuint32_t sveor_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_u64_z)))
+svuint64_t sveor_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_u16_z)))
+svuint16_t sveor_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_s8_z)))
+svint8_t sveor_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_s32_z)))
+svint32_t sveor_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_s64_z)))
+svint64_t sveor_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_n_s16_z)))
+svint16_t sveor_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_u8_m)))
+svuint8_t sveor_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_u32_m)))
+svuint32_t sveor_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_u64_m)))
+svuint64_t sveor_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_u16_m)))
+svuint16_t sveor_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_s8_m)))
+svint8_t sveor_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_s32_m)))
+svint32_t sveor_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_s64_m)))
+svint64_t sveor_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_s16_m)))
+svint16_t sveor_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_u8_x)))
+svuint8_t sveor_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_u32_x)))
+svuint32_t sveor_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_u64_x)))
+svuint64_t sveor_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_u16_x)))
+svuint16_t sveor_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_s8_x)))
+svint8_t sveor_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_s32_x)))
+svint32_t sveor_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_s64_x)))
+svint64_t sveor_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_s16_x)))
+svint16_t sveor_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_u8_z)))
+svuint8_t sveor_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_u32_z)))
+svuint32_t sveor_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_u64_z)))
+svuint64_t sveor_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_u16_z)))
+svuint16_t sveor_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_s8_z)))
+svint8_t sveor_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_s32_z)))
+svint32_t sveor_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_s64_z)))
+svint64_t sveor_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor_s16_z)))
+svint16_t sveor_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorv_u8)))
+uint8_t sveorv(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorv_u32)))
+uint32_t sveorv(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorv_u64)))
+uint64_t sveorv(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorv_u16)))
+uint16_t sveorv(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorv_s8)))
+int8_t sveorv(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorv_s32)))
+int32_t sveorv(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorv_s64)))
+int64_t sveorv(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorv_s16)))
+int16_t sveorv(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexpa_f64)))
+svfloat64_t svexpa(svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexpa_f32)))
+svfloat32_t svexpa(svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexpa_f16)))
+svfloat16_t svexpa(svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_u8)))
+svuint8_t svext(svuint8_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_u32)))
+svuint32_t svext(svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_u64)))
+svuint64_t svext(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_u16)))
+svuint16_t svext(svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_s8)))
+svint8_t svext(svint8_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_f64)))
+svfloat64_t svext(svfloat64_t, svfloat64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_f32)))
+svfloat32_t svext(svfloat32_t, svfloat32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_f16)))
+svfloat16_t svext(svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_s32)))
+svint32_t svext(svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_s64)))
+svint64_t svext(svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_s16)))
+svint16_t svext(svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_s32_m)))
+svint32_t svextb_m(svint32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_s64_m)))
+svint64_t svextb_m(svint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_s16_m)))
+svint16_t svextb_m(svint16_t, svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_s32_x)))
+svint32_t svextb_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_s64_x)))
+svint64_t svextb_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_s16_x)))
+svint16_t svextb_x(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_s32_z)))
+svint32_t svextb_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_s64_z)))
+svint64_t svextb_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_s16_z)))
+svint16_t svextb_z(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_u32_m)))
+svuint32_t svextb_m(svuint32_t, svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_u64_m)))
+svuint64_t svextb_m(svuint64_t, svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_u16_m)))
+svuint16_t svextb_m(svuint16_t, svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_u32_x)))
+svuint32_t svextb_x(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_u64_x)))
+svuint64_t svextb_x(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_u16_x)))
+svuint16_t svextb_x(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_u32_z)))
+svuint32_t svextb_z(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_u64_z)))
+svuint64_t svextb_z(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextb_u16_z)))
+svuint16_t svextb_z(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexth_s32_m)))
+svint32_t svexth_m(svint32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexth_s64_m)))
+svint64_t svexth_m(svint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexth_s32_x)))
+svint32_t svexth_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexth_s64_x)))
+svint64_t svexth_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexth_s32_z)))
+svint32_t svexth_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexth_s64_z)))
+svint64_t svexth_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexth_u32_m)))
+svuint32_t svexth_m(svuint32_t, svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexth_u64_m)))
+svuint64_t svexth_m(svuint64_t, svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexth_u32_x)))
+svuint32_t svexth_x(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexth_u64_x)))
+svuint64_t svexth_x(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexth_u32_z)))
+svuint32_t svexth_z(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexth_u64_z)))
+svuint64_t svexth_z(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextw_s64_m)))
+svint64_t svextw_m(svint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextw_s64_x)))
+svint64_t svextw_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextw_s64_z)))
+svint64_t svextw_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextw_u64_m)))
+svuint64_t svextw_m(svuint64_t, svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextw_u64_x)))
+svuint64_t svextw_x(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextw_u64_z)))
+svuint64_t svextw_z(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_u8)))
+svuint8_t svget2(svuint8x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_u32)))
+svuint32_t svget2(svuint32x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_u64)))
+svuint64_t svget2(svuint64x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_u16)))
+svuint16_t svget2(svuint16x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_s8)))
+svint8_t svget2(svint8x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_f64)))
+svfloat64_t svget2(svfloat64x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_f32)))
+svfloat32_t svget2(svfloat32x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_f16)))
+svfloat16_t svget2(svfloat16x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_s32)))
+svint32_t svget2(svint32x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_s64)))
+svint64_t svget2(svint64x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_s16)))
+svint16_t svget2(svint16x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_u8)))
+svuint8_t svget3(svuint8x3_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_u32)))
+svuint32_t svget3(svuint32x3_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_u64)))
+svuint64_t svget3(svuint64x3_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_u16)))
+svuint16_t svget3(svuint16x3_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_s8)))
+svint8_t svget3(svint8x3_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_f64)))
+svfloat64_t svget3(svfloat64x3_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_f32)))
+svfloat32_t svget3(svfloat32x3_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_f16)))
+svfloat16_t svget3(svfloat16x3_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_s32)))
+svint32_t svget3(svint32x3_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_s64)))
+svint64_t svget3(svint64x3_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_s16)))
+svint16_t svget3(svint16x3_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_u8)))
+svuint8_t svget4(svuint8x4_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_u32)))
+svuint32_t svget4(svuint32x4_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_u64)))
+svuint64_t svget4(svuint64x4_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_u16)))
+svuint16_t svget4(svuint16x4_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_s8)))
+svint8_t svget4(svint8x4_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_f64)))
+svfloat64_t svget4(svfloat64x4_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_f32)))
+svfloat32_t svget4(svfloat32x4_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_f16)))
+svfloat16_t svget4(svfloat16x4_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_s32)))
+svint32_t svget4(svint32x4_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_s64)))
+svint64_t svget4(svint64x4_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_s16)))
+svint16_t svget4(svint16x4_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_u8)))
+svuint8_t svinsr(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_u32)))
+svuint32_t svinsr(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_u64)))
+svuint64_t svinsr(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_u16)))
+svuint16_t svinsr(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_s8)))
+svint8_t svinsr(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_f64)))
+svfloat64_t svinsr(svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_f32)))
+svfloat32_t svinsr(svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_f16)))
+svfloat16_t svinsr(svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_s32)))
+svint32_t svinsr(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_s64)))
+svint64_t svinsr(svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_s16)))
+svint16_t svinsr(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_u8)))
+uint8_t svlasta(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_u32)))
+uint32_t svlasta(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_u64)))
+uint64_t svlasta(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_u16)))
+uint16_t svlasta(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_s8)))
+int8_t svlasta(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_f64)))
+float64_t svlasta(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_f32)))
+float32_t svlasta(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_f16)))
+float16_t svlasta(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_s32)))
+int32_t svlasta(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_s64)))
+int64_t svlasta(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_s16)))
+int16_t svlasta(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_u8)))
+uint8_t svlastb(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_u32)))
+uint32_t svlastb(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_u64)))
+uint64_t svlastb(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_u16)))
+uint16_t svlastb(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_s8)))
+int8_t svlastb(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_f64)))
+float64_t svlastb(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_f32)))
+float32_t svlastb(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_f16)))
+float16_t svlastb(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_s32)))
+int32_t svlastb(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_s64)))
+int64_t svlastb(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_s16)))
+int16_t svlastb(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u8)))
+svuint8_t svld1(svbool_t, uint8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u32)))
+svuint32_t svld1(svbool_t, uint32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u64)))
+svuint64_t svld1(svbool_t, uint64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u16)))
+svuint16_t svld1(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s8)))
+svint8_t svld1(svbool_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_f64)))
+svfloat64_t svld1(svbool_t, float64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_f32)))
+svfloat32_t svld1(svbool_t, float32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_f16)))
+svfloat16_t svld1(svbool_t, float16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s32)))
+svint32_t svld1(svbool_t, int32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s64)))
+svint64_t svld1(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s16)))
+svint16_t svld1(svbool_t, int16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32base_index_u32)))
+svuint32_t svld1_gather_index_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64base_index_u64)))
+svuint64_t svld1_gather_index_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64base_index_f64)))
+svfloat64_t svld1_gather_index_f64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32base_index_f32)))
+svfloat32_t svld1_gather_index_f32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32base_index_s32)))
+svint32_t svld1_gather_index_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64base_index_s64)))
+svint64_t svld1_gather_index_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32base_offset_u32)))
+svuint32_t svld1_gather_offset_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64base_offset_u64)))
+svuint64_t svld1_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64base_offset_f64)))
+svfloat64_t svld1_gather_offset_f64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32base_offset_f32)))
+svfloat32_t svld1_gather_offset_f32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32base_offset_s32)))
+svint32_t svld1_gather_offset_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64base_offset_s64)))
+svint64_t svld1_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32base_u32)))
+svuint32_t svld1_gather_u32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64base_u64)))
+svuint64_t svld1_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64base_f64)))
+svfloat64_t svld1_gather_f64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32base_f32)))
+svfloat32_t svld1_gather_f32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32base_s32)))
+svint32_t svld1_gather_s32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64base_s64)))
+svint64_t svld1_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_s32index_u32)))
+svuint32_t svld1_gather_index(svbool_t, uint32_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_s32index_f32)))
+svfloat32_t svld1_gather_index(svbool_t, float32_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_s32index_s32)))
+svint32_t svld1_gather_index(svbool_t, int32_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32index_u32)))
+svuint32_t svld1_gather_index(svbool_t, uint32_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32index_f32)))
+svfloat32_t svld1_gather_index(svbool_t, float32_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32index_s32)))
+svint32_t svld1_gather_index(svbool_t, int32_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_s64index_u64)))
+svuint64_t svld1_gather_index(svbool_t, uint64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_s64index_f64)))
+svfloat64_t svld1_gather_index(svbool_t, float64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_s64index_s64)))
+svint64_t svld1_gather_index(svbool_t, int64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64index_u64)))
+svuint64_t svld1_gather_index(svbool_t, uint64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64index_f64)))
+svfloat64_t svld1_gather_index(svbool_t, float64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64index_s64)))
+svint64_t svld1_gather_index(svbool_t, int64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_s32offset_u32)))
+svuint32_t svld1_gather_offset(svbool_t, uint32_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_s32offset_f32)))
+svfloat32_t svld1_gather_offset(svbool_t, float32_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_s32offset_s32)))
+svint32_t svld1_gather_offset(svbool_t, int32_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32offset_u32)))
+svuint32_t svld1_gather_offset(svbool_t, uint32_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32offset_f32)))
+svfloat32_t svld1_gather_offset(svbool_t, float32_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32offset_s32)))
+svint32_t svld1_gather_offset(svbool_t, int32_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_s64offset_u64)))
+svuint64_t svld1_gather_offset(svbool_t, uint64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_s64offset_f64)))
+svfloat64_t svld1_gather_offset(svbool_t, float64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_s64offset_s64)))
+svint64_t svld1_gather_offset(svbool_t, int64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64offset_u64)))
+svuint64_t svld1_gather_offset(svbool_t, uint64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64offset_f64)))
+svfloat64_t svld1_gather_offset(svbool_t, float64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64offset_s64)))
+svint64_t svld1_gather_offset(svbool_t, int64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u8)))
+svuint8_t svld1_vnum(svbool_t, uint8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u32)))
+svuint32_t svld1_vnum(svbool_t, uint32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u64)))
+svuint64_t svld1_vnum(svbool_t, uint64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u16)))
+svuint16_t svld1_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s8)))
+svint8_t svld1_vnum(svbool_t, int8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_f64)))
+svfloat64_t svld1_vnum(svbool_t, float64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_f32)))
+svfloat32_t svld1_vnum(svbool_t, float32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_f16)))
+svfloat16_t svld1_vnum(svbool_t, float16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s32)))
+svint32_t svld1_vnum(svbool_t, int32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s64)))
+svint64_t svld1_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s16)))
+svint16_t svld1_vnum(svbool_t, int16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_u8)))
+svuint8_t svld1rq(svbool_t, uint8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_u32)))
+svuint32_t svld1rq(svbool_t, uint32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_u64)))
+svuint64_t svld1rq(svbool_t, uint64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_u16)))
+svuint16_t svld1rq(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_s8)))
+svint8_t svld1rq(svbool_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_f64)))
+svfloat64_t svld1rq(svbool_t, float64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_f32)))
+svfloat32_t svld1rq(svbool_t, float32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_f16)))
+svfloat16_t svld1rq(svbool_t, float16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_s32)))
+svint32_t svld1rq(svbool_t, int32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_s64)))
+svint64_t svld1rq(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_s16)))
+svint16_t svld1rq(svbool_t, int16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_u32base_offset_u32)))
+svuint32_t svld1sb_gather_offset_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_u64base_offset_u64)))
+svuint64_t svld1sb_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_u32base_offset_s32)))
+svint32_t svld1sb_gather_offset_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_u64base_offset_s64)))
+svint64_t svld1sb_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_u32base_u32)))
+svuint32_t svld1sb_gather_u32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_u64base_u64)))
+svuint64_t svld1sb_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_u32base_s32)))
+svint32_t svld1sb_gather_s32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_u64base_s64)))
+svint64_t svld1sb_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_s32offset_u32)))
+svuint32_t svld1sb_gather_offset_u32(svbool_t, int8_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_s32offset_s32)))
+svint32_t svld1sb_gather_offset_s32(svbool_t, int8_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_u32offset_u32)))
+svuint32_t svld1sb_gather_offset_u32(svbool_t, int8_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_u32offset_s32)))
+svint32_t svld1sb_gather_offset_s32(svbool_t, int8_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_s64offset_u64)))
+svuint64_t svld1sb_gather_offset_u64(svbool_t, int8_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_s64offset_s64)))
+svint64_t svld1sb_gather_offset_s64(svbool_t, int8_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_u64offset_u64)))
+svuint64_t svld1sb_gather_offset_u64(svbool_t, int8_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_gather_u64offset_s64)))
+svint64_t svld1sb_gather_offset_s64(svbool_t, int8_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u32base_index_u32)))
+svuint32_t svld1sh_gather_index_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u64base_index_u64)))
+svuint64_t svld1sh_gather_index_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u32base_index_s32)))
+svint32_t svld1sh_gather_index_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u64base_index_s64)))
+svint64_t svld1sh_gather_index_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u32base_offset_u32)))
+svuint32_t svld1sh_gather_offset_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u64base_offset_u64)))
+svuint64_t svld1sh_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u32base_offset_s32)))
+svint32_t svld1sh_gather_offset_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u64base_offset_s64)))
+svint64_t svld1sh_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u32base_u32)))
+svuint32_t svld1sh_gather_u32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u64base_u64)))
+svuint64_t svld1sh_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u32base_s32)))
+svint32_t svld1sh_gather_s32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u64base_s64)))
+svint64_t svld1sh_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_s32index_u32)))
+svuint32_t svld1sh_gather_index_u32(svbool_t, int16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_s32index_s32)))
+svint32_t svld1sh_gather_index_s32(svbool_t, int16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u32index_u32)))
+svuint32_t svld1sh_gather_index_u32(svbool_t, int16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u32index_s32)))
+svint32_t svld1sh_gather_index_s32(svbool_t, int16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_s64index_u64)))
+svuint64_t svld1sh_gather_index_u64(svbool_t, int16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_s64index_s64)))
+svint64_t svld1sh_gather_index_s64(svbool_t, int16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u64index_u64)))
+svuint64_t svld1sh_gather_index_u64(svbool_t, int16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u64index_s64)))
+svint64_t svld1sh_gather_index_s64(svbool_t, int16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_s32offset_u32)))
+svuint32_t svld1sh_gather_offset_u32(svbool_t, int16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_s32offset_s32)))
+svint32_t svld1sh_gather_offset_s32(svbool_t, int16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u32offset_u32)))
+svuint32_t svld1sh_gather_offset_u32(svbool_t, int16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u32offset_s32)))
+svint32_t svld1sh_gather_offset_s32(svbool_t, int16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_s64offset_u64)))
+svuint64_t svld1sh_gather_offset_u64(svbool_t, int16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_s64offset_s64)))
+svint64_t svld1sh_gather_offset_s64(svbool_t, int16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u64offset_u64)))
+svuint64_t svld1sh_gather_offset_u64(svbool_t, int16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sh_gather_u64offset_s64)))
+svint64_t svld1sh_gather_offset_s64(svbool_t, int16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sw_gather_u64base_index_u64)))
+svuint64_t svld1sw_gather_index_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sw_gather_u64base_index_s64)))
+svint64_t svld1sw_gather_index_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sw_gather_u64base_offset_u64)))
+svuint64_t svld1sw_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sw_gather_u64base_offset_s64)))
+svint64_t svld1sw_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sw_gather_u64base_u64)))
+svuint64_t svld1sw_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sw_gather_u64base_s64)))
+svint64_t svld1sw_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sw_gather_s64index_u64)))
+svuint64_t svld1sw_gather_index_u64(svbool_t, int32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sw_gather_s64index_s64)))
+svint64_t svld1sw_gather_index_s64(svbool_t, int32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sw_gather_u64index_u64)))
+svuint64_t svld1sw_gather_index_u64(svbool_t, int32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sw_gather_u64index_s64)))
+svint64_t svld1sw_gather_index_s64(svbool_t, int32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sw_gather_s64offset_u64)))
+svuint64_t svld1sw_gather_offset_u64(svbool_t, int32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sw_gather_s64offset_s64)))
+svint64_t svld1sw_gather_offset_s64(svbool_t, int32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sw_gather_u64offset_u64)))
+svuint64_t svld1sw_gather_offset_u64(svbool_t, int32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sw_gather_u64offset_s64)))
+svint64_t svld1sw_gather_offset_s64(svbool_t, int32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_u32base_offset_u32)))
+svuint32_t svld1ub_gather_offset_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_u64base_offset_u64)))
+svuint64_t svld1ub_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_u32base_offset_s32)))
+svint32_t svld1ub_gather_offset_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_u64base_offset_s64)))
+svint64_t svld1ub_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_u32base_u32)))
+svuint32_t svld1ub_gather_u32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_u64base_u64)))
+svuint64_t svld1ub_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_u32base_s32)))
+svint32_t svld1ub_gather_s32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_u64base_s64)))
+svint64_t svld1ub_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_s32offset_u32)))
+svuint32_t svld1ub_gather_offset_u32(svbool_t, uint8_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_s32offset_s32)))
+svint32_t svld1ub_gather_offset_s32(svbool_t, uint8_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_u32offset_u32)))
+svuint32_t svld1ub_gather_offset_u32(svbool_t, uint8_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_u32offset_s32)))
+svint32_t svld1ub_gather_offset_s32(svbool_t, uint8_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_s64offset_u64)))
+svuint64_t svld1ub_gather_offset_u64(svbool_t, uint8_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_s64offset_s64)))
+svint64_t svld1ub_gather_offset_s64(svbool_t, uint8_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_u64offset_u64)))
+svuint64_t svld1ub_gather_offset_u64(svbool_t, uint8_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ub_gather_u64offset_s64)))
+svint64_t svld1ub_gather_offset_s64(svbool_t, uint8_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u32base_index_u32)))
+svuint32_t svld1uh_gather_index_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u64base_index_u64)))
+svuint64_t svld1uh_gather_index_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u32base_index_s32)))
+svint32_t svld1uh_gather_index_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u64base_index_s64)))
+svint64_t svld1uh_gather_index_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u32base_offset_u32)))
+svuint32_t svld1uh_gather_offset_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u64base_offset_u64)))
+svuint64_t svld1uh_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u32base_offset_s32)))
+svint32_t svld1uh_gather_offset_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u64base_offset_s64)))
+svint64_t svld1uh_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u32base_u32)))
+svuint32_t svld1uh_gather_u32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u64base_u64)))
+svuint64_t svld1uh_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u32base_s32)))
+svint32_t svld1uh_gather_s32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u64base_s64)))
+svint64_t svld1uh_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_s32index_u32)))
+svuint32_t svld1uh_gather_index_u32(svbool_t, uint16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_s32index_s32)))
+svint32_t svld1uh_gather_index_s32(svbool_t, uint16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u32index_u32)))
+svuint32_t svld1uh_gather_index_u32(svbool_t, uint16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u32index_s32)))
+svint32_t svld1uh_gather_index_s32(svbool_t, uint16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_s64index_u64)))
+svuint64_t svld1uh_gather_index_u64(svbool_t, uint16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_s64index_s64)))
+svint64_t svld1uh_gather_index_s64(svbool_t, uint16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u64index_u64)))
+svuint64_t svld1uh_gather_index_u64(svbool_t, uint16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u64index_s64)))
+svint64_t svld1uh_gather_index_s64(svbool_t, uint16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_s32offset_u32)))
+svuint32_t svld1uh_gather_offset_u32(svbool_t, uint16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_s32offset_s32)))
+svint32_t svld1uh_gather_offset_s32(svbool_t, uint16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u32offset_u32)))
+svuint32_t svld1uh_gather_offset_u32(svbool_t, uint16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u32offset_s32)))
+svint32_t svld1uh_gather_offset_s32(svbool_t, uint16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_s64offset_u64)))
+svuint64_t svld1uh_gather_offset_u64(svbool_t, uint16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_s64offset_s64)))
+svint64_t svld1uh_gather_offset_s64(svbool_t, uint16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u64offset_u64)))
+svuint64_t svld1uh_gather_offset_u64(svbool_t, uint16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uh_gather_u64offset_s64)))
+svint64_t svld1uh_gather_offset_s64(svbool_t, uint16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uw_gather_u64base_index_u64)))
+svuint64_t svld1uw_gather_index_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uw_gather_u64base_index_s64)))
+svint64_t svld1uw_gather_index_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uw_gather_u64base_offset_u64)))
+svuint64_t svld1uw_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uw_gather_u64base_offset_s64)))
+svint64_t svld1uw_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uw_gather_u64base_u64)))
+svuint64_t svld1uw_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uw_gather_u64base_s64)))
+svint64_t svld1uw_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uw_gather_s64index_u64)))
+svuint64_t svld1uw_gather_index_u64(svbool_t, uint32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uw_gather_s64index_s64)))
+svint64_t svld1uw_gather_index_s64(svbool_t, uint32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uw_gather_u64index_u64)))
+svuint64_t svld1uw_gather_index_u64(svbool_t, uint32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uw_gather_u64index_s64)))
+svint64_t svld1uw_gather_index_s64(svbool_t, uint32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uw_gather_s64offset_u64)))
+svuint64_t svld1uw_gather_offset_u64(svbool_t, uint32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uw_gather_s64offset_s64)))
+svint64_t svld1uw_gather_offset_s64(svbool_t, uint32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uw_gather_u64offset_u64)))
+svuint64_t svld1uw_gather_offset_u64(svbool_t, uint32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uw_gather_u64offset_s64)))
+svint64_t svld1uw_gather_offset_s64(svbool_t, uint32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_u8)))
+svuint8x2_t svld2(svbool_t, uint8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_u32)))
+svuint32x2_t svld2(svbool_t, uint32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_u64)))
+svuint64x2_t svld2(svbool_t, uint64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_u16)))
+svuint16x2_t svld2(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_s8)))
+svint8x2_t svld2(svbool_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_f64)))
+svfloat64x2_t svld2(svbool_t, float64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_f32)))
+svfloat32x2_t svld2(svbool_t, float32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_f16)))
+svfloat16x2_t svld2(svbool_t, float16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_s32)))
+svint32x2_t svld2(svbool_t, int32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_s64)))
+svint64x2_t svld2(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_s16)))
+svint16x2_t svld2(svbool_t, int16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_u8)))
+svuint8x2_t svld2_vnum(svbool_t, uint8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_u32)))
+svuint32x2_t svld2_vnum(svbool_t, uint32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_u64)))
+svuint64x2_t svld2_vnum(svbool_t, uint64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_u16)))
+svuint16x2_t svld2_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_s8)))
+svint8x2_t svld2_vnum(svbool_t, int8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_f64)))
+svfloat64x2_t svld2_vnum(svbool_t, float64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_f32)))
+svfloat32x2_t svld2_vnum(svbool_t, float32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_f16)))
+svfloat16x2_t svld2_vnum(svbool_t, float16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_s32)))
+svint32x2_t svld2_vnum(svbool_t, int32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_s64)))
+svint64x2_t svld2_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_s16)))
+svint16x2_t svld2_vnum(svbool_t, int16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_u8)))
+svuint8x3_t svld3(svbool_t, uint8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_u32)))
+svuint32x3_t svld3(svbool_t, uint32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_u64)))
+svuint64x3_t svld3(svbool_t, uint64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_u16)))
+svuint16x3_t svld3(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_s8)))
+svint8x3_t svld3(svbool_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_f64)))
+svfloat64x3_t svld3(svbool_t, float64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_f32)))
+svfloat32x3_t svld3(svbool_t, float32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_f16)))
+svfloat16x3_t svld3(svbool_t, float16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_s32)))
+svint32x3_t svld3(svbool_t, int32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_s64)))
+svint64x3_t svld3(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_s16)))
+svint16x3_t svld3(svbool_t, int16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_u8)))
+svuint8x3_t svld3_vnum(svbool_t, uint8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_u32)))
+svuint32x3_t svld3_vnum(svbool_t, uint32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_u64)))
+svuint64x3_t svld3_vnum(svbool_t, uint64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_u16)))
+svuint16x3_t svld3_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_s8)))
+svint8x3_t svld3_vnum(svbool_t, int8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_f64)))
+svfloat64x3_t svld3_vnum(svbool_t, float64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_f32)))
+svfloat32x3_t svld3_vnum(svbool_t, float32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_f16)))
+svfloat16x3_t svld3_vnum(svbool_t, float16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_s32)))
+svint32x3_t svld3_vnum(svbool_t, int32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_s64)))
+svint64x3_t svld3_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_s16)))
+svint16x3_t svld3_vnum(svbool_t, int16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_u8)))
+svuint8x4_t svld4(svbool_t, uint8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_u32)))
+svuint32x4_t svld4(svbool_t, uint32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_u64)))
+svuint64x4_t svld4(svbool_t, uint64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_u16)))
+svuint16x4_t svld4(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_s8)))
+svint8x4_t svld4(svbool_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_f64)))
+svfloat64x4_t svld4(svbool_t, float64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_f32)))
+svfloat32x4_t svld4(svbool_t, float32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_f16)))
+svfloat16x4_t svld4(svbool_t, float16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_s32)))
+svint32x4_t svld4(svbool_t, int32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_s64)))
+svint64x4_t svld4(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_s16)))
+svint16x4_t svld4(svbool_t, int16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_u8)))
+svuint8x4_t svld4_vnum(svbool_t, uint8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_u32)))
+svuint32x4_t svld4_vnum(svbool_t, uint32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_u64)))
+svuint64x4_t svld4_vnum(svbool_t, uint64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_u16)))
+svuint16x4_t svld4_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_s8)))
+svint8x4_t svld4_vnum(svbool_t, int8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_f64)))
+svfloat64x4_t svld4_vnum(svbool_t, float64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_f32)))
+svfloat32x4_t svld4_vnum(svbool_t, float32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_f16)))
+svfloat16x4_t svld4_vnum(svbool_t, float16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_s32)))
+svint32x4_t svld4_vnum(svbool_t, int32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_s64)))
+svint64x4_t svld4_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_s16)))
+svint16x4_t svld4_vnum(svbool_t, int16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_u8)))
+svuint8_t svldff1(svbool_t, uint8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_u32)))
+svuint32_t svldff1(svbool_t, uint32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_u64)))
+svuint64_t svldff1(svbool_t, uint64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_u16)))
+svuint16_t svldff1(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_s8)))
+svint8_t svldff1(svbool_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_f64)))
+svfloat64_t svldff1(svbool_t, float64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_f32)))
+svfloat32_t svldff1(svbool_t, float32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_f16)))
+svfloat16_t svldff1(svbool_t, float16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_s32)))
+svint32_t svldff1(svbool_t, int32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_s64)))
+svint64_t svldff1(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_s16)))
+svint16_t svldff1(svbool_t, int16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32base_index_u32)))
+svuint32_t svldff1_gather_index_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u64base_index_u64)))
+svuint64_t svldff1_gather_index_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u64base_index_f64)))
+svfloat64_t svldff1_gather_index_f64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32base_index_f32)))
+svfloat32_t svldff1_gather_index_f32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32base_index_s32)))
+svint32_t svldff1_gather_index_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u64base_index_s64)))
+svint64_t svldff1_gather_index_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32base_offset_u32)))
+svuint32_t svldff1_gather_offset_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u64base_offset_u64)))
+svuint64_t svldff1_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u64base_offset_f64)))
+svfloat64_t svldff1_gather_offset_f64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32base_offset_f32)))
+svfloat32_t svldff1_gather_offset_f32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32base_offset_s32)))
+svint32_t svldff1_gather_offset_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u64base_offset_s64)))
+svint64_t svldff1_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32base_u32)))
+svuint32_t svldff1_gather_u32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u64base_u64)))
+svuint64_t svldff1_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u64base_f64)))
+svfloat64_t svldff1_gather_f64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32base_f32)))
+svfloat32_t svldff1_gather_f32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32base_s32)))
+svint32_t svldff1_gather_s32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u64base_s64)))
+svint64_t svldff1_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_s32index_u32)))
+svuint32_t svldff1_gather_index(svbool_t, uint32_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_s32index_f32)))
+svfloat32_t svldff1_gather_index(svbool_t, float32_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_s32index_s32)))
+svint32_t svldff1_gather_index(svbool_t, int32_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32index_u32)))
+svuint32_t svldff1_gather_index(svbool_t, uint32_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32index_f32)))
+svfloat32_t svldff1_gather_index(svbool_t, float32_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32index_s32)))
+svint32_t svldff1_gather_index(svbool_t, int32_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_s64index_u64)))
+svuint64_t svldff1_gather_index(svbool_t, uint64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_s64index_f64)))
+svfloat64_t svldff1_gather_index(svbool_t, float64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_s64index_s64)))
+svint64_t svldff1_gather_index(svbool_t, int64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u64index_u64)))
+svuint64_t svldff1_gather_index(svbool_t, uint64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u64index_f64)))
+svfloat64_t svldff1_gather_index(svbool_t, float64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u64index_s64)))
+svint64_t svldff1_gather_index(svbool_t, int64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_s32offset_u32)))
+svuint32_t svldff1_gather_offset(svbool_t, uint32_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_s32offset_f32)))
+svfloat32_t svldff1_gather_offset(svbool_t, float32_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_s32offset_s32)))
+svint32_t svldff1_gather_offset(svbool_t, int32_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32offset_u32)))
+svuint32_t svldff1_gather_offset(svbool_t, uint32_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32offset_f32)))
+svfloat32_t svldff1_gather_offset(svbool_t, float32_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32offset_s32)))
+svint32_t svldff1_gather_offset(svbool_t, int32_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_s64offset_u64)))
+svuint64_t svldff1_gather_offset(svbool_t, uint64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_s64offset_f64)))
+svfloat64_t svldff1_gather_offset(svbool_t, float64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_s64offset_s64)))
+svint64_t svldff1_gather_offset(svbool_t, int64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u64offset_u64)))
+svuint64_t svldff1_gather_offset(svbool_t, uint64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u64offset_f64)))
+svfloat64_t svldff1_gather_offset(svbool_t, float64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u64offset_s64)))
+svint64_t svldff1_gather_offset(svbool_t, int64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_u8)))
+svuint8_t svldff1_vnum(svbool_t, uint8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_u32)))
+svuint32_t svldff1_vnum(svbool_t, uint32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_u64)))
+svuint64_t svldff1_vnum(svbool_t, uint64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_u16)))
+svuint16_t svldff1_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_s8)))
+svint8_t svldff1_vnum(svbool_t, int8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_f64)))
+svfloat64_t svldff1_vnum(svbool_t, float64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_f32)))
+svfloat32_t svldff1_vnum(svbool_t, float32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_f16)))
+svfloat16_t svldff1_vnum(svbool_t, float16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_s32)))
+svint32_t svldff1_vnum(svbool_t, int32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_s64)))
+svint64_t svldff1_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_s16)))
+svint16_t svldff1_vnum(svbool_t, int16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_u32base_offset_u32)))
+svuint32_t svldff1sb_gather_offset_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_u64base_offset_u64)))
+svuint64_t svldff1sb_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_u32base_offset_s32)))
+svint32_t svldff1sb_gather_offset_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_u64base_offset_s64)))
+svint64_t svldff1sb_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_u32base_u32)))
+svuint32_t svldff1sb_gather_u32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_u64base_u64)))
+svuint64_t svldff1sb_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_u32base_s32)))
+svint32_t svldff1sb_gather_s32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_u64base_s64)))
+svint64_t svldff1sb_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_s32offset_u32)))
+svuint32_t svldff1sb_gather_offset_u32(svbool_t, int8_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_s32offset_s32)))
+svint32_t svldff1sb_gather_offset_s32(svbool_t, int8_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_u32offset_u32)))
+svuint32_t svldff1sb_gather_offset_u32(svbool_t, int8_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_u32offset_s32)))
+svint32_t svldff1sb_gather_offset_s32(svbool_t, int8_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_s64offset_u64)))
+svuint64_t svldff1sb_gather_offset_u64(svbool_t, int8_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_s64offset_s64)))
+svint64_t svldff1sb_gather_offset_s64(svbool_t, int8_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_u64offset_u64)))
+svuint64_t svldff1sb_gather_offset_u64(svbool_t, int8_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_u64offset_s64)))
+svint64_t svldff1sb_gather_offset_s64(svbool_t, int8_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u32base_index_u32)))
+svuint32_t svldff1sh_gather_index_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u64base_index_u64)))
+svuint64_t svldff1sh_gather_index_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u32base_index_s32)))
+svint32_t svldff1sh_gather_index_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u64base_index_s64)))
+svint64_t svldff1sh_gather_index_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u32base_offset_u32)))
+svuint32_t svldff1sh_gather_offset_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u64base_offset_u64)))
+svuint64_t svldff1sh_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u32base_offset_s32)))
+svint32_t svldff1sh_gather_offset_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u64base_offset_s64)))
+svint64_t svldff1sh_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u32base_u32)))
+svuint32_t svldff1sh_gather_u32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u64base_u64)))
+svuint64_t svldff1sh_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u32base_s32)))
+svint32_t svldff1sh_gather_s32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u64base_s64)))
+svint64_t svldff1sh_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_s32index_u32)))
+svuint32_t svldff1sh_gather_index_u32(svbool_t, int16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_s32index_s32)))
+svint32_t svldff1sh_gather_index_s32(svbool_t, int16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u32index_u32)))
+svuint32_t svldff1sh_gather_index_u32(svbool_t, int16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u32index_s32)))
+svint32_t svldff1sh_gather_index_s32(svbool_t, int16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_s64index_u64)))
+svuint64_t svldff1sh_gather_index_u64(svbool_t, int16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_s64index_s64)))
+svint64_t svldff1sh_gather_index_s64(svbool_t, int16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u64index_u64)))
+svuint64_t svldff1sh_gather_index_u64(svbool_t, int16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u64index_s64)))
+svint64_t svldff1sh_gather_index_s64(svbool_t, int16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_s32offset_u32)))
+svuint32_t svldff1sh_gather_offset_u32(svbool_t, int16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_s32offset_s32)))
+svint32_t svldff1sh_gather_offset_s32(svbool_t, int16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u32offset_u32)))
+svuint32_t svldff1sh_gather_offset_u32(svbool_t, int16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u32offset_s32)))
+svint32_t svldff1sh_gather_offset_s32(svbool_t, int16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_s64offset_u64)))
+svuint64_t svldff1sh_gather_offset_u64(svbool_t, int16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_s64offset_s64)))
+svint64_t svldff1sh_gather_offset_s64(svbool_t, int16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u64offset_u64)))
+svuint64_t svldff1sh_gather_offset_u64(svbool_t, int16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sh_gather_u64offset_s64)))
+svint64_t svldff1sh_gather_offset_s64(svbool_t, int16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sw_gather_u64base_index_u64)))
+svuint64_t svldff1sw_gather_index_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sw_gather_u64base_index_s64)))
+svint64_t svldff1sw_gather_index_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sw_gather_u64base_offset_u64)))
+svuint64_t svldff1sw_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sw_gather_u64base_offset_s64)))
+svint64_t svldff1sw_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sw_gather_u64base_u64)))
+svuint64_t svldff1sw_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sw_gather_u64base_s64)))
+svint64_t svldff1sw_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sw_gather_s64index_u64)))
+svuint64_t svldff1sw_gather_index_u64(svbool_t, int32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sw_gather_s64index_s64)))
+svint64_t svldff1sw_gather_index_s64(svbool_t, int32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sw_gather_u64index_u64)))
+svuint64_t svldff1sw_gather_index_u64(svbool_t, int32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sw_gather_u64index_s64)))
+svint64_t svldff1sw_gather_index_s64(svbool_t, int32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sw_gather_s64offset_u64)))
+svuint64_t svldff1sw_gather_offset_u64(svbool_t, int32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sw_gather_s64offset_s64)))
+svint64_t svldff1sw_gather_offset_s64(svbool_t, int32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sw_gather_u64offset_u64)))
+svuint64_t svldff1sw_gather_offset_u64(svbool_t, int32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sw_gather_u64offset_s64)))
+svint64_t svldff1sw_gather_offset_s64(svbool_t, int32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_u32base_offset_u32)))
+svuint32_t svldff1ub_gather_offset_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_u64base_offset_u64)))
+svuint64_t svldff1ub_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_u32base_offset_s32)))
+svint32_t svldff1ub_gather_offset_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_u64base_offset_s64)))
+svint64_t svldff1ub_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_u32base_u32)))
+svuint32_t svldff1ub_gather_u32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_u64base_u64)))
+svuint64_t svldff1ub_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_u32base_s32)))
+svint32_t svldff1ub_gather_s32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_u64base_s64)))
+svint64_t svldff1ub_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_s32offset_u32)))
+svuint32_t svldff1ub_gather_offset_u32(svbool_t, uint8_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_s32offset_s32)))
+svint32_t svldff1ub_gather_offset_s32(svbool_t, uint8_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_u32offset_u32)))
+svuint32_t svldff1ub_gather_offset_u32(svbool_t, uint8_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_u32offset_s32)))
+svint32_t svldff1ub_gather_offset_s32(svbool_t, uint8_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_s64offset_u64)))
+svuint64_t svldff1ub_gather_offset_u64(svbool_t, uint8_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_s64offset_s64)))
+svint64_t svldff1ub_gather_offset_s64(svbool_t, uint8_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_u64offset_u64)))
+svuint64_t svldff1ub_gather_offset_u64(svbool_t, uint8_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1ub_gather_u64offset_s64)))
+svint64_t svldff1ub_gather_offset_s64(svbool_t, uint8_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u32base_index_u32)))
+svuint32_t svldff1uh_gather_index_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u64base_index_u64)))
+svuint64_t svldff1uh_gather_index_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u32base_index_s32)))
+svint32_t svldff1uh_gather_index_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u64base_index_s64)))
+svint64_t svldff1uh_gather_index_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u32base_offset_u32)))
+svuint32_t svldff1uh_gather_offset_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u64base_offset_u64)))
+svuint64_t svldff1uh_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u32base_offset_s32)))
+svint32_t svldff1uh_gather_offset_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u64base_offset_s64)))
+svint64_t svldff1uh_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u32base_u32)))
+svuint32_t svldff1uh_gather_u32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u64base_u64)))
+svuint64_t svldff1uh_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u32base_s32)))
+svint32_t svldff1uh_gather_s32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u64base_s64)))
+svint64_t svldff1uh_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_s32index_u32)))
+svuint32_t svldff1uh_gather_index_u32(svbool_t, uint16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_s32index_s32)))
+svint32_t svldff1uh_gather_index_s32(svbool_t, uint16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u32index_u32)))
+svuint32_t svldff1uh_gather_index_u32(svbool_t, uint16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u32index_s32)))
+svint32_t svldff1uh_gather_index_s32(svbool_t, uint16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_s64index_u64)))
+svuint64_t svldff1uh_gather_index_u64(svbool_t, uint16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_s64index_s64)))
+svint64_t svldff1uh_gather_index_s64(svbool_t, uint16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u64index_u64)))
+svuint64_t svldff1uh_gather_index_u64(svbool_t, uint16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u64index_s64)))
+svint64_t svldff1uh_gather_index_s64(svbool_t, uint16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_s32offset_u32)))
+svuint32_t svldff1uh_gather_offset_u32(svbool_t, uint16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_s32offset_s32)))
+svint32_t svldff1uh_gather_offset_s32(svbool_t, uint16_t const *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u32offset_u32)))
+svuint32_t svldff1uh_gather_offset_u32(svbool_t, uint16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u32offset_s32)))
+svint32_t svldff1uh_gather_offset_s32(svbool_t, uint16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_s64offset_u64)))
+svuint64_t svldff1uh_gather_offset_u64(svbool_t, uint16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_s64offset_s64)))
+svint64_t svldff1uh_gather_offset_s64(svbool_t, uint16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u64offset_u64)))
+svuint64_t svldff1uh_gather_offset_u64(svbool_t, uint16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uh_gather_u64offset_s64)))
+svint64_t svldff1uh_gather_offset_s64(svbool_t, uint16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uw_gather_u64base_index_u64)))
+svuint64_t svldff1uw_gather_index_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uw_gather_u64base_index_s64)))
+svint64_t svldff1uw_gather_index_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uw_gather_u64base_offset_u64)))
+svuint64_t svldff1uw_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uw_gather_u64base_offset_s64)))
+svint64_t svldff1uw_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uw_gather_u64base_u64)))
+svuint64_t svldff1uw_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uw_gather_u64base_s64)))
+svint64_t svldff1uw_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uw_gather_s64index_u64)))
+svuint64_t svldff1uw_gather_index_u64(svbool_t, uint32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uw_gather_s64index_s64)))
+svint64_t svldff1uw_gather_index_s64(svbool_t, uint32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uw_gather_u64index_u64)))
+svuint64_t svldff1uw_gather_index_u64(svbool_t, uint32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uw_gather_u64index_s64)))
+svint64_t svldff1uw_gather_index_s64(svbool_t, uint32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uw_gather_s64offset_u64)))
+svuint64_t svldff1uw_gather_offset_u64(svbool_t, uint32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uw_gather_s64offset_s64)))
+svint64_t svldff1uw_gather_offset_s64(svbool_t, uint32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uw_gather_u64offset_u64)))
+svuint64_t svldff1uw_gather_offset_u64(svbool_t, uint32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1uw_gather_u64offset_s64)))
+svint64_t svldff1uw_gather_offset_s64(svbool_t, uint32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_u8)))
+svuint8_t svldnf1(svbool_t, uint8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_u32)))
+svuint32_t svldnf1(svbool_t, uint32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_u64)))
+svuint64_t svldnf1(svbool_t, uint64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_u16)))
+svuint16_t svldnf1(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_s8)))
+svint8_t svldnf1(svbool_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_f64)))
+svfloat64_t svldnf1(svbool_t, float64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_f32)))
+svfloat32_t svldnf1(svbool_t, float32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_f16)))
+svfloat16_t svldnf1(svbool_t, float16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_s32)))
+svint32_t svldnf1(svbool_t, int32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_s64)))
+svint64_t svldnf1(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_s16)))
+svint16_t svldnf1(svbool_t, int16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_u8)))
+svuint8_t svldnf1_vnum(svbool_t, uint8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_u32)))
+svuint32_t svldnf1_vnum(svbool_t, uint32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_u64)))
+svuint64_t svldnf1_vnum(svbool_t, uint64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_u16)))
+svuint16_t svldnf1_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_s8)))
+svint8_t svldnf1_vnum(svbool_t, int8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_f64)))
+svfloat64_t svldnf1_vnum(svbool_t, float64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_f32)))
+svfloat32_t svldnf1_vnum(svbool_t, float32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_f16)))
+svfloat16_t svldnf1_vnum(svbool_t, float16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_s32)))
+svint32_t svldnf1_vnum(svbool_t, int32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_s64)))
+svint64_t svldnf1_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_s16)))
+svint16_t svldnf1_vnum(svbool_t, int16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u8)))
+svuint8_t svldnt1(svbool_t, uint8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u32)))
+svuint32_t svldnt1(svbool_t, uint32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u64)))
+svuint64_t svldnt1(svbool_t, uint64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u16)))
+svuint16_t svldnt1(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s8)))
+svint8_t svldnt1(svbool_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_f64)))
+svfloat64_t svldnt1(svbool_t, float64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_f32)))
+svfloat32_t svldnt1(svbool_t, float32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_f16)))
+svfloat16_t svldnt1(svbool_t, float16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s32)))
+svint32_t svldnt1(svbool_t, int32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s64)))
+svint64_t svldnt1(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s16)))
+svint16_t svldnt1(svbool_t, int16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u8)))
+svuint8_t svldnt1_vnum(svbool_t, uint8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u32)))
+svuint32_t svldnt1_vnum(svbool_t, uint32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u64)))
+svuint64_t svldnt1_vnum(svbool_t, uint64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u16)))
+svuint16_t svldnt1_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s8)))
+svint8_t svldnt1_vnum(svbool_t, int8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_f64)))
+svfloat64_t svldnt1_vnum(svbool_t, float64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_f32)))
+svfloat32_t svldnt1_vnum(svbool_t, float32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_f16)))
+svfloat16_t svldnt1_vnum(svbool_t, float16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s32)))
+svint32_t svldnt1_vnum(svbool_t, int32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s64)))
+svint64_t svldnt1_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s16)))
+svint16_t svldnt1_vnum(svbool_t, int16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_u8)))
+uint64_t svlen(svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_u32)))
+uint64_t svlen(svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_u64)))
+uint64_t svlen(svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_u16)))
+uint64_t svlen(svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_s8)))
+uint64_t svlen(svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_f64)))
+uint64_t svlen(svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_f32)))
+uint64_t svlen(svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_f16)))
+uint64_t svlen(svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_s32)))
+uint64_t svlen(svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_s64)))
+uint64_t svlen(svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_s16)))
+uint64_t svlen(svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_u8_m)))
+svuint8_t svlsl_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_u32_m)))
+svuint32_t svlsl_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_u64_m)))
+svuint64_t svlsl_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_u16_m)))
+svuint16_t svlsl_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_s8_m)))
+svint8_t svlsl_m(svbool_t, svint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_s32_m)))
+svint32_t svlsl_m(svbool_t, svint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_s64_m)))
+svint64_t svlsl_m(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_s16_m)))
+svint16_t svlsl_m(svbool_t, svint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_u8_x)))
+svuint8_t svlsl_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_u32_x)))
+svuint32_t svlsl_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_u64_x)))
+svuint64_t svlsl_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_u16_x)))
+svuint16_t svlsl_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_s8_x)))
+svint8_t svlsl_x(svbool_t, svint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_s32_x)))
+svint32_t svlsl_x(svbool_t, svint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_s64_x)))
+svint64_t svlsl_x(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_s16_x)))
+svint16_t svlsl_x(svbool_t, svint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_u8_z)))
+svuint8_t svlsl_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_u32_z)))
+svuint32_t svlsl_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_u64_z)))
+svuint64_t svlsl_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_u16_z)))
+svuint16_t svlsl_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_s8_z)))
+svint8_t svlsl_z(svbool_t, svint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_s32_z)))
+svint32_t svlsl_z(svbool_t, svint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_s64_z)))
+svint64_t svlsl_z(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_n_s16_z)))
+svint16_t svlsl_z(svbool_t, svint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_u8_m)))
+svuint8_t svlsl_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_u32_m)))
+svuint32_t svlsl_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_u64_m)))
+svuint64_t svlsl_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_u16_m)))
+svuint16_t svlsl_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_s8_m)))
+svint8_t svlsl_m(svbool_t, svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_s32_m)))
+svint32_t svlsl_m(svbool_t, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_s64_m)))
+svint64_t svlsl_m(svbool_t, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_s16_m)))
+svint16_t svlsl_m(svbool_t, svint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_u8_x)))
+svuint8_t svlsl_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_u32_x)))
+svuint32_t svlsl_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_u64_x)))
+svuint64_t svlsl_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_u16_x)))
+svuint16_t svlsl_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_s8_x)))
+svint8_t svlsl_x(svbool_t, svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_s32_x)))
+svint32_t svlsl_x(svbool_t, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_s64_x)))
+svint64_t svlsl_x(svbool_t, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_s16_x)))
+svint16_t svlsl_x(svbool_t, svint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_u8_z)))
+svuint8_t svlsl_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_u32_z)))
+svuint32_t svlsl_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_u64_z)))
+svuint64_t svlsl_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_u16_z)))
+svuint16_t svlsl_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_s8_z)))
+svint8_t svlsl_z(svbool_t, svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_s32_z)))
+svint32_t svlsl_z(svbool_t, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_s64_z)))
+svint64_t svlsl_z(svbool_t, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_s16_z)))
+svint16_t svlsl_z(svbool_t, svint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_u8_m)))
+svuint8_t svlsl_wide_m(svbool_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_u32_m)))
+svuint32_t svlsl_wide_m(svbool_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_u16_m)))
+svuint16_t svlsl_wide_m(svbool_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_s8_m)))
+svint8_t svlsl_wide_m(svbool_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_s32_m)))
+svint32_t svlsl_wide_m(svbool_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_s16_m)))
+svint16_t svlsl_wide_m(svbool_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_u8_x)))
+svuint8_t svlsl_wide_x(svbool_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_u32_x)))
+svuint32_t svlsl_wide_x(svbool_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_u16_x)))
+svuint16_t svlsl_wide_x(svbool_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_s8_x)))
+svint8_t svlsl_wide_x(svbool_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_s32_x)))
+svint32_t svlsl_wide_x(svbool_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_s16_x)))
+svint16_t svlsl_wide_x(svbool_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_u8_z)))
+svuint8_t svlsl_wide_z(svbool_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_u32_z)))
+svuint32_t svlsl_wide_z(svbool_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_u16_z)))
+svuint16_t svlsl_wide_z(svbool_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_s8_z)))
+svint8_t svlsl_wide_z(svbool_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_s32_z)))
+svint32_t svlsl_wide_z(svbool_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_n_s16_z)))
+svint16_t svlsl_wide_z(svbool_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_u8_m)))
+svuint8_t svlsl_wide_m(svbool_t, svuint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_u32_m)))
+svuint32_t svlsl_wide_m(svbool_t, svuint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_u16_m)))
+svuint16_t svlsl_wide_m(svbool_t, svuint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_s8_m)))
+svint8_t svlsl_wide_m(svbool_t, svint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_s32_m)))
+svint32_t svlsl_wide_m(svbool_t, svint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_s16_m)))
+svint16_t svlsl_wide_m(svbool_t, svint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_u8_x)))
+svuint8_t svlsl_wide_x(svbool_t, svuint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_u32_x)))
+svuint32_t svlsl_wide_x(svbool_t, svuint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_u16_x)))
+svuint16_t svlsl_wide_x(svbool_t, svuint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_s8_x)))
+svint8_t svlsl_wide_x(svbool_t, svint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_s32_x)))
+svint32_t svlsl_wide_x(svbool_t, svint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_s16_x)))
+svint16_t svlsl_wide_x(svbool_t, svint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_u8_z)))
+svuint8_t svlsl_wide_z(svbool_t, svuint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_u32_z)))
+svuint32_t svlsl_wide_z(svbool_t, svuint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_u16_z)))
+svuint16_t svlsl_wide_z(svbool_t, svuint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_s8_z)))
+svint8_t svlsl_wide_z(svbool_t, svint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_s32_z)))
+svint32_t svlsl_wide_z(svbool_t, svint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsl_wide_s16_z)))
+svint16_t svlsl_wide_z(svbool_t, svint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_n_u8_m)))
+svuint8_t svlsr_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_n_u32_m)))
+svuint32_t svlsr_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_n_u64_m)))
+svuint64_t svlsr_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_n_u16_m)))
+svuint16_t svlsr_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_n_u8_x)))
+svuint8_t svlsr_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_n_u32_x)))
+svuint32_t svlsr_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_n_u64_x)))
+svuint64_t svlsr_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_n_u16_x)))
+svuint16_t svlsr_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_n_u8_z)))
+svuint8_t svlsr_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_n_u32_z)))
+svuint32_t svlsr_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_n_u64_z)))
+svuint64_t svlsr_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_n_u16_z)))
+svuint16_t svlsr_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_u8_m)))
+svuint8_t svlsr_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_u32_m)))
+svuint32_t svlsr_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_u64_m)))
+svuint64_t svlsr_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_u16_m)))
+svuint16_t svlsr_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_u8_x)))
+svuint8_t svlsr_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_u32_x)))
+svuint32_t svlsr_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_u64_x)))
+svuint64_t svlsr_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_u16_x)))
+svuint16_t svlsr_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_u8_z)))
+svuint8_t svlsr_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_u32_z)))
+svuint32_t svlsr_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_u64_z)))
+svuint64_t svlsr_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_u16_z)))
+svuint16_t svlsr_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_n_u8_m)))
+svuint8_t svlsr_wide_m(svbool_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_n_u32_m)))
+svuint32_t svlsr_wide_m(svbool_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_n_u16_m)))
+svuint16_t svlsr_wide_m(svbool_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_n_u8_x)))
+svuint8_t svlsr_wide_x(svbool_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_n_u32_x)))
+svuint32_t svlsr_wide_x(svbool_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_n_u16_x)))
+svuint16_t svlsr_wide_x(svbool_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_n_u8_z)))
+svuint8_t svlsr_wide_z(svbool_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_n_u32_z)))
+svuint32_t svlsr_wide_z(svbool_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_n_u16_z)))
+svuint16_t svlsr_wide_z(svbool_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_u8_m)))
+svuint8_t svlsr_wide_m(svbool_t, svuint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_u32_m)))
+svuint32_t svlsr_wide_m(svbool_t, svuint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_u16_m)))
+svuint16_t svlsr_wide_m(svbool_t, svuint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_u8_x)))
+svuint8_t svlsr_wide_x(svbool_t, svuint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_u32_x)))
+svuint32_t svlsr_wide_x(svbool_t, svuint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_u16_x)))
+svuint16_t svlsr_wide_x(svbool_t, svuint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_u8_z)))
+svuint8_t svlsr_wide_z(svbool_t, svuint8_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_u32_z)))
+svuint32_t svlsr_wide_z(svbool_t, svuint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlsr_wide_u16_z)))
+svuint16_t svlsr_wide_z(svbool_t, svuint16_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_f64_m)))
+svfloat64_t svmad_m(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_f32_m)))
+svfloat32_t svmad_m(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_f16_m)))
+svfloat16_t svmad_m(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_f64_x)))
+svfloat64_t svmad_x(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_f32_x)))
+svfloat32_t svmad_x(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_f16_x)))
+svfloat16_t svmad_x(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_f64_z)))
+svfloat64_t svmad_z(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_f32_z)))
+svfloat32_t svmad_z(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_f16_z)))
+svfloat16_t svmad_z(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_u8_m)))
+svuint8_t svmad_m(svbool_t, svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_u32_m)))
+svuint32_t svmad_m(svbool_t, svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_u64_m)))
+svuint64_t svmad_m(svbool_t, svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_u16_m)))
+svuint16_t svmad_m(svbool_t, svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_s8_m)))
+svint8_t svmad_m(svbool_t, svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_s32_m)))
+svint32_t svmad_m(svbool_t, svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_s64_m)))
+svint64_t svmad_m(svbool_t, svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_s16_m)))
+svint16_t svmad_m(svbool_t, svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_u8_x)))
+svuint8_t svmad_x(svbool_t, svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_u32_x)))
+svuint32_t svmad_x(svbool_t, svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_u64_x)))
+svuint64_t svmad_x(svbool_t, svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_u16_x)))
+svuint16_t svmad_x(svbool_t, svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_s8_x)))
+svint8_t svmad_x(svbool_t, svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_s32_x)))
+svint32_t svmad_x(svbool_t, svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_s64_x)))
+svint64_t svmad_x(svbool_t, svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_s16_x)))
+svint16_t svmad_x(svbool_t, svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_u8_z)))
+svuint8_t svmad_z(svbool_t, svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_u32_z)))
+svuint32_t svmad_z(svbool_t, svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_u64_z)))
+svuint64_t svmad_z(svbool_t, svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_u16_z)))
+svuint16_t svmad_z(svbool_t, svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_s8_z)))
+svint8_t svmad_z(svbool_t, svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_s32_z)))
+svint32_t svmad_z(svbool_t, svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_s64_z)))
+svint64_t svmad_z(svbool_t, svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_n_s16_z)))
+svint16_t svmad_z(svbool_t, svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_f64_m)))
+svfloat64_t svmad_m(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_f32_m)))
+svfloat32_t svmad_m(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_f16_m)))
+svfloat16_t svmad_m(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_f64_x)))
+svfloat64_t svmad_x(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_f32_x)))
+svfloat32_t svmad_x(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_f16_x)))
+svfloat16_t svmad_x(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_f64_z)))
+svfloat64_t svmad_z(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_f32_z)))
+svfloat32_t svmad_z(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_f16_z)))
+svfloat16_t svmad_z(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_u8_m)))
+svuint8_t svmad_m(svbool_t, svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_u32_m)))
+svuint32_t svmad_m(svbool_t, svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_u64_m)))
+svuint64_t svmad_m(svbool_t, svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_u16_m)))
+svuint16_t svmad_m(svbool_t, svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_s8_m)))
+svint8_t svmad_m(svbool_t, svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_s32_m)))
+svint32_t svmad_m(svbool_t, svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_s64_m)))
+svint64_t svmad_m(svbool_t, svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_s16_m)))
+svint16_t svmad_m(svbool_t, svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_u8_x)))
+svuint8_t svmad_x(svbool_t, svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_u32_x)))
+svuint32_t svmad_x(svbool_t, svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_u64_x)))
+svuint64_t svmad_x(svbool_t, svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_u16_x)))
+svuint16_t svmad_x(svbool_t, svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_s8_x)))
+svint8_t svmad_x(svbool_t, svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_s32_x)))
+svint32_t svmad_x(svbool_t, svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_s64_x)))
+svint64_t svmad_x(svbool_t, svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_s16_x)))
+svint16_t svmad_x(svbool_t, svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_u8_z)))
+svuint8_t svmad_z(svbool_t, svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_u32_z)))
+svuint32_t svmad_z(svbool_t, svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_u64_z)))
+svuint64_t svmad_z(svbool_t, svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_u16_z)))
+svuint16_t svmad_z(svbool_t, svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_s8_z)))
+svint8_t svmad_z(svbool_t, svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_s32_z)))
+svint32_t svmad_z(svbool_t, svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_s64_z)))
+svint64_t svmad_z(svbool_t, svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmad_s16_z)))
+svint16_t svmad_z(svbool_t, svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_f64_m)))
+svfloat64_t svmax_m(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_f32_m)))
+svfloat32_t svmax_m(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_f16_m)))
+svfloat16_t svmax_m(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_f64_x)))
+svfloat64_t svmax_x(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_f32_x)))
+svfloat32_t svmax_x(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_f16_x)))
+svfloat16_t svmax_x(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_f64_z)))
+svfloat64_t svmax_z(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_f32_z)))
+svfloat32_t svmax_z(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_f16_z)))
+svfloat16_t svmax_z(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_s8_m)))
+svint8_t svmax_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_s32_m)))
+svint32_t svmax_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_s64_m)))
+svint64_t svmax_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_s16_m)))
+svint16_t svmax_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_s8_x)))
+svint8_t svmax_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_s32_x)))
+svint32_t svmax_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_s64_x)))
+svint64_t svmax_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_s16_x)))
+svint16_t svmax_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_s8_z)))
+svint8_t svmax_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_s32_z)))
+svint32_t svmax_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_s64_z)))
+svint64_t svmax_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_s16_z)))
+svint16_t svmax_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_u8_m)))
+svuint8_t svmax_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_u32_m)))
+svuint32_t svmax_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_u64_m)))
+svuint64_t svmax_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_u16_m)))
+svuint16_t svmax_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_u8_x)))
+svuint8_t svmax_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_u32_x)))
+svuint32_t svmax_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_u64_x)))
+svuint64_t svmax_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_u16_x)))
+svuint16_t svmax_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_u8_z)))
+svuint8_t svmax_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_u32_z)))
+svuint32_t svmax_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_u64_z)))
+svuint64_t svmax_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_n_u16_z)))
+svuint16_t svmax_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_f64_m)))
+svfloat64_t svmax_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_f32_m)))
+svfloat32_t svmax_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_f16_m)))
+svfloat16_t svmax_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_f64_x)))
+svfloat64_t svmax_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_f32_x)))
+svfloat32_t svmax_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_f16_x)))
+svfloat16_t svmax_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_f64_z)))
+svfloat64_t svmax_z(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_f32_z)))
+svfloat32_t svmax_z(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_f16_z)))
+svfloat16_t svmax_z(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_s8_m)))
+svint8_t svmax_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_s32_m)))
+svint32_t svmax_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_s64_m)))
+svint64_t svmax_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_s16_m)))
+svint16_t svmax_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_s8_x)))
+svint8_t svmax_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_s32_x)))
+svint32_t svmax_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_s64_x)))
+svint64_t svmax_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_s16_x)))
+svint16_t svmax_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_s8_z)))
+svint8_t svmax_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_s32_z)))
+svint32_t svmax_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_s64_z)))
+svint64_t svmax_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_s16_z)))
+svint16_t svmax_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_u8_m)))
+svuint8_t svmax_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_u32_m)))
+svuint32_t svmax_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_u64_m)))
+svuint64_t svmax_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_u16_m)))
+svuint16_t svmax_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_u8_x)))
+svuint8_t svmax_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_u32_x)))
+svuint32_t svmax_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_u64_x)))
+svuint64_t svmax_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_u16_x)))
+svuint16_t svmax_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_u8_z)))
+svuint8_t svmax_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_u32_z)))
+svuint32_t svmax_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_u64_z)))
+svuint64_t svmax_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmax_u16_z)))
+svuint16_t svmax_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_n_f64_m)))
+svfloat64_t svmaxnm_m(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_n_f32_m)))
+svfloat32_t svmaxnm_m(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_n_f16_m)))
+svfloat16_t svmaxnm_m(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_n_f64_x)))
+svfloat64_t svmaxnm_x(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_n_f32_x)))
+svfloat32_t svmaxnm_x(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_n_f16_x)))
+svfloat16_t svmaxnm_x(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_n_f64_z)))
+svfloat64_t svmaxnm_z(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_n_f32_z)))
+svfloat32_t svmaxnm_z(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_n_f16_z)))
+svfloat16_t svmaxnm_z(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_f64_m)))
+svfloat64_t svmaxnm_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_f32_m)))
+svfloat32_t svmaxnm_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_f16_m)))
+svfloat16_t svmaxnm_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_f64_x)))
+svfloat64_t svmaxnm_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_f32_x)))
+svfloat32_t svmaxnm_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_f16_x)))
+svfloat16_t svmaxnm_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_f64_z)))
+svfloat64_t svmaxnm_z(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_f32_z)))
+svfloat32_t svmaxnm_z(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnm_f16_z)))
+svfloat16_t svmaxnm_z(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmv_f64)))
+float64_t svmaxnmv(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmv_f32)))
+float32_t svmaxnmv(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmv_f16)))
+float16_t svmaxnmv(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxv_f64)))
+float64_t svmaxv(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxv_f32)))
+float32_t svmaxv(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxv_f16)))
+float16_t svmaxv(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxv_s8)))
+int8_t svmaxv(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxv_s32)))
+int32_t svmaxv(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxv_s64)))
+int64_t svmaxv(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxv_s16)))
+int16_t svmaxv(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxv_u8)))
+uint8_t svmaxv(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxv_u32)))
+uint32_t svmaxv(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxv_u64)))
+uint64_t svmaxv(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxv_u16)))
+uint16_t svmaxv(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_f64_m)))
+svfloat64_t svmin_m(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_f32_m)))
+svfloat32_t svmin_m(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_f16_m)))
+svfloat16_t svmin_m(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_f64_x)))
+svfloat64_t svmin_x(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_f32_x)))
+svfloat32_t svmin_x(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_f16_x)))
+svfloat16_t svmin_x(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_f64_z)))
+svfloat64_t svmin_z(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_f32_z)))
+svfloat32_t svmin_z(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_f16_z)))
+svfloat16_t svmin_z(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_s8_m)))
+svint8_t svmin_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_s32_m)))
+svint32_t svmin_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_s64_m)))
+svint64_t svmin_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_s16_m)))
+svint16_t svmin_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_s8_x)))
+svint8_t svmin_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_s32_x)))
+svint32_t svmin_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_s64_x)))
+svint64_t svmin_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_s16_x)))
+svint16_t svmin_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_s8_z)))
+svint8_t svmin_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_s32_z)))
+svint32_t svmin_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_s64_z)))
+svint64_t svmin_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_s16_z)))
+svint16_t svmin_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_u8_m)))
+svuint8_t svmin_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_u32_m)))
+svuint32_t svmin_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_u64_m)))
+svuint64_t svmin_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_u16_m)))
+svuint16_t svmin_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_u8_x)))
+svuint8_t svmin_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_u32_x)))
+svuint32_t svmin_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_u64_x)))
+svuint64_t svmin_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_u16_x)))
+svuint16_t svmin_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_u8_z)))
+svuint8_t svmin_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_u32_z)))
+svuint32_t svmin_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_u64_z)))
+svuint64_t svmin_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_n_u16_z)))
+svuint16_t svmin_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_f64_m)))
+svfloat64_t svmin_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_f32_m)))
+svfloat32_t svmin_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_f16_m)))
+svfloat16_t svmin_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_f64_x)))
+svfloat64_t svmin_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_f32_x)))
+svfloat32_t svmin_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_f16_x)))
+svfloat16_t svmin_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_f64_z)))
+svfloat64_t svmin_z(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_f32_z)))
+svfloat32_t svmin_z(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_f16_z)))
+svfloat16_t svmin_z(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_s8_m)))
+svint8_t svmin_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_s32_m)))
+svint32_t svmin_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_s64_m)))
+svint64_t svmin_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_s16_m)))
+svint16_t svmin_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_s8_x)))
+svint8_t svmin_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_s32_x)))
+svint32_t svmin_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_s64_x)))
+svint64_t svmin_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_s16_x)))
+svint16_t svmin_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_s8_z)))
+svint8_t svmin_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_s32_z)))
+svint32_t svmin_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_s64_z)))
+svint64_t svmin_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_s16_z)))
+svint16_t svmin_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_u8_m)))
+svuint8_t svmin_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_u32_m)))
+svuint32_t svmin_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_u64_m)))
+svuint64_t svmin_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_u16_m)))
+svuint16_t svmin_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_u8_x)))
+svuint8_t svmin_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_u32_x)))
+svuint32_t svmin_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_u64_x)))
+svuint64_t svmin_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_u16_x)))
+svuint16_t svmin_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_u8_z)))
+svuint8_t svmin_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_u32_z)))
+svuint32_t svmin_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_u64_z)))
+svuint64_t svmin_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmin_u16_z)))
+svuint16_t svmin_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_n_f64_m)))
+svfloat64_t svminnm_m(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_n_f32_m)))
+svfloat32_t svminnm_m(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_n_f16_m)))
+svfloat16_t svminnm_m(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_n_f64_x)))
+svfloat64_t svminnm_x(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_n_f32_x)))
+svfloat32_t svminnm_x(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_n_f16_x)))
+svfloat16_t svminnm_x(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_n_f64_z)))
+svfloat64_t svminnm_z(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_n_f32_z)))
+svfloat32_t svminnm_z(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_n_f16_z)))
+svfloat16_t svminnm_z(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_f64_m)))
+svfloat64_t svminnm_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_f32_m)))
+svfloat32_t svminnm_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_f16_m)))
+svfloat16_t svminnm_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_f64_x)))
+svfloat64_t svminnm_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_f32_x)))
+svfloat32_t svminnm_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_f16_x)))
+svfloat16_t svminnm_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_f64_z)))
+svfloat64_t svminnm_z(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_f32_z)))
+svfloat32_t svminnm_z(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnm_f16_z)))
+svfloat16_t svminnm_z(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmv_f64)))
+float64_t svminnmv(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmv_f32)))
+float32_t svminnmv(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmv_f16)))
+float16_t svminnmv(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminv_f64)))
+float64_t svminv(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminv_f32)))
+float32_t svminv(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminv_f16)))
+float16_t svminv(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminv_s8)))
+int8_t svminv(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminv_s32)))
+int32_t svminv(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminv_s64)))
+int64_t svminv(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminv_s16)))
+int16_t svminv(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminv_u8)))
+uint8_t svminv(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminv_u32)))
+uint32_t svminv(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminv_u64)))
+uint64_t svminv(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminv_u16)))
+uint16_t svminv(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_f64_m)))
+svfloat64_t svmla_m(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_f32_m)))
+svfloat32_t svmla_m(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_f16_m)))
+svfloat16_t svmla_m(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_f64_x)))
+svfloat64_t svmla_x(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_f32_x)))
+svfloat32_t svmla_x(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_f16_x)))
+svfloat16_t svmla_x(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_f64_z)))
+svfloat64_t svmla_z(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_f32_z)))
+svfloat32_t svmla_z(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_f16_z)))
+svfloat16_t svmla_z(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_u8_m)))
+svuint8_t svmla_m(svbool_t, svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_u32_m)))
+svuint32_t svmla_m(svbool_t, svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_u64_m)))
+svuint64_t svmla_m(svbool_t, svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_u16_m)))
+svuint16_t svmla_m(svbool_t, svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_s8_m)))
+svint8_t svmla_m(svbool_t, svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_s32_m)))
+svint32_t svmla_m(svbool_t, svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_s64_m)))
+svint64_t svmla_m(svbool_t, svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_s16_m)))
+svint16_t svmla_m(svbool_t, svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_u8_x)))
+svuint8_t svmla_x(svbool_t, svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_u32_x)))
+svuint32_t svmla_x(svbool_t, svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_u64_x)))
+svuint64_t svmla_x(svbool_t, svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_u16_x)))
+svuint16_t svmla_x(svbool_t, svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_s8_x)))
+svint8_t svmla_x(svbool_t, svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_s32_x)))
+svint32_t svmla_x(svbool_t, svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_s64_x)))
+svint64_t svmla_x(svbool_t, svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_s16_x)))
+svint16_t svmla_x(svbool_t, svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_u8_z)))
+svuint8_t svmla_z(svbool_t, svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_u32_z)))
+svuint32_t svmla_z(svbool_t, svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_u64_z)))
+svuint64_t svmla_z(svbool_t, svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_u16_z)))
+svuint16_t svmla_z(svbool_t, svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_s8_z)))
+svint8_t svmla_z(svbool_t, svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_s32_z)))
+svint32_t svmla_z(svbool_t, svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_s64_z)))
+svint64_t svmla_z(svbool_t, svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_n_s16_z)))
+svint16_t svmla_z(svbool_t, svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_f64_m)))
+svfloat64_t svmla_m(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_f32_m)))
+svfloat32_t svmla_m(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_f16_m)))
+svfloat16_t svmla_m(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_f64_x)))
+svfloat64_t svmla_x(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_f32_x)))
+svfloat32_t svmla_x(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_f16_x)))
+svfloat16_t svmla_x(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_f64_z)))
+svfloat64_t svmla_z(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_f32_z)))
+svfloat32_t svmla_z(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_f16_z)))
+svfloat16_t svmla_z(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_u8_m)))
+svuint8_t svmla_m(svbool_t, svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_u32_m)))
+svuint32_t svmla_m(svbool_t, svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_u64_m)))
+svuint64_t svmla_m(svbool_t, svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_u16_m)))
+svuint16_t svmla_m(svbool_t, svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_s8_m)))
+svint8_t svmla_m(svbool_t, svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_s32_m)))
+svint32_t svmla_m(svbool_t, svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_s64_m)))
+svint64_t svmla_m(svbool_t, svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_s16_m)))
+svint16_t svmla_m(svbool_t, svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_u8_x)))
+svuint8_t svmla_x(svbool_t, svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_u32_x)))
+svuint32_t svmla_x(svbool_t, svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_u64_x)))
+svuint64_t svmla_x(svbool_t, svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_u16_x)))
+svuint16_t svmla_x(svbool_t, svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_s8_x)))
+svint8_t svmla_x(svbool_t, svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_s32_x)))
+svint32_t svmla_x(svbool_t, svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_s64_x)))
+svint64_t svmla_x(svbool_t, svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_s16_x)))
+svint16_t svmla_x(svbool_t, svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_u8_z)))
+svuint8_t svmla_z(svbool_t, svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_u32_z)))
+svuint32_t svmla_z(svbool_t, svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_u64_z)))
+svuint64_t svmla_z(svbool_t, svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_u16_z)))
+svuint16_t svmla_z(svbool_t, svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_s8_z)))
+svint8_t svmla_z(svbool_t, svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_s32_z)))
+svint32_t svmla_z(svbool_t, svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_s64_z)))
+svint64_t svmla_z(svbool_t, svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_s16_z)))
+svint16_t svmla_z(svbool_t, svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_f64)))
+svfloat64_t svmla_lane(svfloat64_t, svfloat64_t, svfloat64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_f32)))
+svfloat32_t svmla_lane(svfloat32_t, svfloat32_t, svfloat32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_f16)))
+svfloat16_t svmla_lane(svfloat16_t, svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_f64_m)))
+svfloat64_t svmls_m(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_f32_m)))
+svfloat32_t svmls_m(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_f16_m)))
+svfloat16_t svmls_m(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_f64_x)))
+svfloat64_t svmls_x(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_f32_x)))
+svfloat32_t svmls_x(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_f16_x)))
+svfloat16_t svmls_x(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_f64_z)))
+svfloat64_t svmls_z(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_f32_z)))
+svfloat32_t svmls_z(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_f16_z)))
+svfloat16_t svmls_z(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_u8_m)))
+svuint8_t svmls_m(svbool_t, svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_u32_m)))
+svuint32_t svmls_m(svbool_t, svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_u64_m)))
+svuint64_t svmls_m(svbool_t, svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_u16_m)))
+svuint16_t svmls_m(svbool_t, svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_s8_m)))
+svint8_t svmls_m(svbool_t, svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_s32_m)))
+svint32_t svmls_m(svbool_t, svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_s64_m)))
+svint64_t svmls_m(svbool_t, svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_s16_m)))
+svint16_t svmls_m(svbool_t, svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_u8_x)))
+svuint8_t svmls_x(svbool_t, svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_u32_x)))
+svuint32_t svmls_x(svbool_t, svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_u64_x)))
+svuint64_t svmls_x(svbool_t, svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_u16_x)))
+svuint16_t svmls_x(svbool_t, svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_s8_x)))
+svint8_t svmls_x(svbool_t, svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_s32_x)))
+svint32_t svmls_x(svbool_t, svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_s64_x)))
+svint64_t svmls_x(svbool_t, svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_s16_x)))
+svint16_t svmls_x(svbool_t, svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_u8_z)))
+svuint8_t svmls_z(svbool_t, svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_u32_z)))
+svuint32_t svmls_z(svbool_t, svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_u64_z)))
+svuint64_t svmls_z(svbool_t, svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_u16_z)))
+svuint16_t svmls_z(svbool_t, svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_s8_z)))
+svint8_t svmls_z(svbool_t, svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_s32_z)))
+svint32_t svmls_z(svbool_t, svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_s64_z)))
+svint64_t svmls_z(svbool_t, svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_n_s16_z)))
+svint16_t svmls_z(svbool_t, svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_f64_m)))
+svfloat64_t svmls_m(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_f32_m)))
+svfloat32_t svmls_m(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_f16_m)))
+svfloat16_t svmls_m(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_f64_x)))
+svfloat64_t svmls_x(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_f32_x)))
+svfloat32_t svmls_x(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_f16_x)))
+svfloat16_t svmls_x(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_f64_z)))
+svfloat64_t svmls_z(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_f32_z)))
+svfloat32_t svmls_z(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_f16_z)))
+svfloat16_t svmls_z(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_u8_m)))
+svuint8_t svmls_m(svbool_t, svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_u32_m)))
+svuint32_t svmls_m(svbool_t, svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_u64_m)))
+svuint64_t svmls_m(svbool_t, svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_u16_m)))
+svuint16_t svmls_m(svbool_t, svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_s8_m)))
+svint8_t svmls_m(svbool_t, svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_s32_m)))
+svint32_t svmls_m(svbool_t, svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_s64_m)))
+svint64_t svmls_m(svbool_t, svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_s16_m)))
+svint16_t svmls_m(svbool_t, svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_u8_x)))
+svuint8_t svmls_x(svbool_t, svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_u32_x)))
+svuint32_t svmls_x(svbool_t, svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_u64_x)))
+svuint64_t svmls_x(svbool_t, svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_u16_x)))
+svuint16_t svmls_x(svbool_t, svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_s8_x)))
+svint8_t svmls_x(svbool_t, svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_s32_x)))
+svint32_t svmls_x(svbool_t, svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_s64_x)))
+svint64_t svmls_x(svbool_t, svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_s16_x)))
+svint16_t svmls_x(svbool_t, svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_u8_z)))
+svuint8_t svmls_z(svbool_t, svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_u32_z)))
+svuint32_t svmls_z(svbool_t, svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_u64_z)))
+svuint64_t svmls_z(svbool_t, svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_u16_z)))
+svuint16_t svmls_z(svbool_t, svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_s8_z)))
+svint8_t svmls_z(svbool_t, svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_s32_z)))
+svint32_t svmls_z(svbool_t, svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_s64_z)))
+svint64_t svmls_z(svbool_t, svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_s16_z)))
+svint16_t svmls_z(svbool_t, svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_f64)))
+svfloat64_t svmls_lane(svfloat64_t, svfloat64_t, svfloat64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_f32)))
+svfloat32_t svmls_lane(svfloat32_t, svfloat32_t, svfloat32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_f16)))
+svfloat16_t svmls_lane(svfloat16_t, svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmov_b_z)))
+svbool_t svmov_z(svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_f64_m)))
+svfloat64_t svmsb_m(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_f32_m)))
+svfloat32_t svmsb_m(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_f16_m)))
+svfloat16_t svmsb_m(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_f64_x)))
+svfloat64_t svmsb_x(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_f32_x)))
+svfloat32_t svmsb_x(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_f16_x)))
+svfloat16_t svmsb_x(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_f64_z)))
+svfloat64_t svmsb_z(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_f32_z)))
+svfloat32_t svmsb_z(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_f16_z)))
+svfloat16_t svmsb_z(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_u8_m)))
+svuint8_t svmsb_m(svbool_t, svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_u32_m)))
+svuint32_t svmsb_m(svbool_t, svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_u64_m)))
+svuint64_t svmsb_m(svbool_t, svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_u16_m)))
+svuint16_t svmsb_m(svbool_t, svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_s8_m)))
+svint8_t svmsb_m(svbool_t, svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_s32_m)))
+svint32_t svmsb_m(svbool_t, svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_s64_m)))
+svint64_t svmsb_m(svbool_t, svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_s16_m)))
+svint16_t svmsb_m(svbool_t, svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_u8_x)))
+svuint8_t svmsb_x(svbool_t, svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_u32_x)))
+svuint32_t svmsb_x(svbool_t, svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_u64_x)))
+svuint64_t svmsb_x(svbool_t, svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_u16_x)))
+svuint16_t svmsb_x(svbool_t, svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_s8_x)))
+svint8_t svmsb_x(svbool_t, svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_s32_x)))
+svint32_t svmsb_x(svbool_t, svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_s64_x)))
+svint64_t svmsb_x(svbool_t, svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_s16_x)))
+svint16_t svmsb_x(svbool_t, svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_u8_z)))
+svuint8_t svmsb_z(svbool_t, svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_u32_z)))
+svuint32_t svmsb_z(svbool_t, svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_u64_z)))
+svuint64_t svmsb_z(svbool_t, svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_u16_z)))
+svuint16_t svmsb_z(svbool_t, svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_s8_z)))
+svint8_t svmsb_z(svbool_t, svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_s32_z)))
+svint32_t svmsb_z(svbool_t, svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_s64_z)))
+svint64_t svmsb_z(svbool_t, svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_n_s16_z)))
+svint16_t svmsb_z(svbool_t, svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_f64_m)))
+svfloat64_t svmsb_m(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_f32_m)))
+svfloat32_t svmsb_m(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_f16_m)))
+svfloat16_t svmsb_m(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_f64_x)))
+svfloat64_t svmsb_x(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_f32_x)))
+svfloat32_t svmsb_x(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_f16_x)))
+svfloat16_t svmsb_x(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_f64_z)))
+svfloat64_t svmsb_z(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_f32_z)))
+svfloat32_t svmsb_z(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_f16_z)))
+svfloat16_t svmsb_z(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_u8_m)))
+svuint8_t svmsb_m(svbool_t, svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_u32_m)))
+svuint32_t svmsb_m(svbool_t, svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_u64_m)))
+svuint64_t svmsb_m(svbool_t, svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_u16_m)))
+svuint16_t svmsb_m(svbool_t, svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_s8_m)))
+svint8_t svmsb_m(svbool_t, svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_s32_m)))
+svint32_t svmsb_m(svbool_t, svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_s64_m)))
+svint64_t svmsb_m(svbool_t, svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_s16_m)))
+svint16_t svmsb_m(svbool_t, svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_u8_x)))
+svuint8_t svmsb_x(svbool_t, svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_u32_x)))
+svuint32_t svmsb_x(svbool_t, svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_u64_x)))
+svuint64_t svmsb_x(svbool_t, svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_u16_x)))
+svuint16_t svmsb_x(svbool_t, svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_s8_x)))
+svint8_t svmsb_x(svbool_t, svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_s32_x)))
+svint32_t svmsb_x(svbool_t, svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_s64_x)))
+svint64_t svmsb_x(svbool_t, svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_s16_x)))
+svint16_t svmsb_x(svbool_t, svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_u8_z)))
+svuint8_t svmsb_z(svbool_t, svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_u32_z)))
+svuint32_t svmsb_z(svbool_t, svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_u64_z)))
+svuint64_t svmsb_z(svbool_t, svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_u16_z)))
+svuint16_t svmsb_z(svbool_t, svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_s8_z)))
+svint8_t svmsb_z(svbool_t, svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_s32_z)))
+svint32_t svmsb_z(svbool_t, svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_s64_z)))
+svint64_t svmsb_z(svbool_t, svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmsb_s16_z)))
+svint16_t svmsb_z(svbool_t, svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_f64_m)))
+svfloat64_t svmul_m(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_f32_m)))
+svfloat32_t svmul_m(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_f16_m)))
+svfloat16_t svmul_m(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_f64_x)))
+svfloat64_t svmul_x(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_f32_x)))
+svfloat32_t svmul_x(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_f16_x)))
+svfloat16_t svmul_x(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_f64_z)))
+svfloat64_t svmul_z(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_f32_z)))
+svfloat32_t svmul_z(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_f16_z)))
+svfloat16_t svmul_z(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_u8_m)))
+svuint8_t svmul_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_u32_m)))
+svuint32_t svmul_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_u64_m)))
+svuint64_t svmul_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_u16_m)))
+svuint16_t svmul_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_s8_m)))
+svint8_t svmul_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_s32_m)))
+svint32_t svmul_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_s64_m)))
+svint64_t svmul_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_s16_m)))
+svint16_t svmul_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_u8_x)))
+svuint8_t svmul_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_u32_x)))
+svuint32_t svmul_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_u64_x)))
+svuint64_t svmul_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_u16_x)))
+svuint16_t svmul_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_s8_x)))
+svint8_t svmul_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_s32_x)))
+svint32_t svmul_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_s64_x)))
+svint64_t svmul_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_s16_x)))
+svint16_t svmul_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_u8_z)))
+svuint8_t svmul_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_u32_z)))
+svuint32_t svmul_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_u64_z)))
+svuint64_t svmul_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_u16_z)))
+svuint16_t svmul_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_s8_z)))
+svint8_t svmul_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_s32_z)))
+svint32_t svmul_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_s64_z)))
+svint64_t svmul_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_n_s16_z)))
+svint16_t svmul_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_f64_m)))
+svfloat64_t svmul_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_f32_m)))
+svfloat32_t svmul_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_f16_m)))
+svfloat16_t svmul_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_f64_x)))
+svfloat64_t svmul_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_f32_x)))
+svfloat32_t svmul_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_f16_x)))
+svfloat16_t svmul_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_f64_z)))
+svfloat64_t svmul_z(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_f32_z)))
+svfloat32_t svmul_z(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_f16_z)))
+svfloat16_t svmul_z(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_u8_m)))
+svuint8_t svmul_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_u32_m)))
+svuint32_t svmul_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_u64_m)))
+svuint64_t svmul_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_u16_m)))
+svuint16_t svmul_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_s8_m)))
+svint8_t svmul_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_s32_m)))
+svint32_t svmul_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_s64_m)))
+svint64_t svmul_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_s16_m)))
+svint16_t svmul_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_u8_x)))
+svuint8_t svmul_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_u32_x)))
+svuint32_t svmul_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_u64_x)))
+svuint64_t svmul_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_u16_x)))
+svuint16_t svmul_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_s8_x)))
+svint8_t svmul_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_s32_x)))
+svint32_t svmul_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_s64_x)))
+svint64_t svmul_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_s16_x)))
+svint16_t svmul_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_u8_z)))
+svuint8_t svmul_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_u32_z)))
+svuint32_t svmul_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_u64_z)))
+svuint64_t svmul_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_u16_z)))
+svuint16_t svmul_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_s8_z)))
+svint8_t svmul_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_s32_z)))
+svint32_t svmul_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_s64_z)))
+svint64_t svmul_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_s16_z)))
+svint16_t svmul_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_f64)))
+svfloat64_t svmul_lane(svfloat64_t, svfloat64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_f32)))
+svfloat32_t svmul_lane(svfloat32_t, svfloat32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_f16)))
+svfloat16_t svmul_lane(svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_s8_m)))
+svint8_t svmulh_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_s32_m)))
+svint32_t svmulh_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_s64_m)))
+svint64_t svmulh_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_s16_m)))
+svint16_t svmulh_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_s8_x)))
+svint8_t svmulh_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_s32_x)))
+svint32_t svmulh_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_s64_x)))
+svint64_t svmulh_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_s16_x)))
+svint16_t svmulh_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_s8_z)))
+svint8_t svmulh_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_s32_z)))
+svint32_t svmulh_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_s64_z)))
+svint64_t svmulh_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_s16_z)))
+svint16_t svmulh_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_u8_m)))
+svuint8_t svmulh_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_u32_m)))
+svuint32_t svmulh_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_u64_m)))
+svuint64_t svmulh_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_u16_m)))
+svuint16_t svmulh_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_u8_x)))
+svuint8_t svmulh_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_u32_x)))
+svuint32_t svmulh_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_u64_x)))
+svuint64_t svmulh_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_u16_x)))
+svuint16_t svmulh_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_u8_z)))
+svuint8_t svmulh_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_u32_z)))
+svuint32_t svmulh_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_u64_z)))
+svuint64_t svmulh_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_n_u16_z)))
+svuint16_t svmulh_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_s8_m)))
+svint8_t svmulh_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_s32_m)))
+svint32_t svmulh_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_s64_m)))
+svint64_t svmulh_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_s16_m)))
+svint16_t svmulh_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_s8_x)))
+svint8_t svmulh_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_s32_x)))
+svint32_t svmulh_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_s64_x)))
+svint64_t svmulh_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_s16_x)))
+svint16_t svmulh_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_s8_z)))
+svint8_t svmulh_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_s32_z)))
+svint32_t svmulh_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_s64_z)))
+svint64_t svmulh_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_s16_z)))
+svint16_t svmulh_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_u8_m)))
+svuint8_t svmulh_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_u32_m)))
+svuint32_t svmulh_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_u64_m)))
+svuint64_t svmulh_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_u16_m)))
+svuint16_t svmulh_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_u8_x)))
+svuint8_t svmulh_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_u32_x)))
+svuint32_t svmulh_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_u64_x)))
+svuint64_t svmulh_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_u16_x)))
+svuint16_t svmulh_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_u8_z)))
+svuint8_t svmulh_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_u32_z)))
+svuint32_t svmulh_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_u64_z)))
+svuint64_t svmulh_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulh_u16_z)))
+svuint16_t svmulh_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_n_f64_m)))
+svfloat64_t svmulx_m(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_n_f32_m)))
+svfloat32_t svmulx_m(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_n_f16_m)))
+svfloat16_t svmulx_m(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_n_f64_x)))
+svfloat64_t svmulx_x(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_n_f32_x)))
+svfloat32_t svmulx_x(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_n_f16_x)))
+svfloat16_t svmulx_x(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_n_f64_z)))
+svfloat64_t svmulx_z(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_n_f32_z)))
+svfloat32_t svmulx_z(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_n_f16_z)))
+svfloat16_t svmulx_z(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_f64_m)))
+svfloat64_t svmulx_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_f32_m)))
+svfloat32_t svmulx_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_f16_m)))
+svfloat16_t svmulx_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_f64_x)))
+svfloat64_t svmulx_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_f32_x)))
+svfloat32_t svmulx_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_f16_x)))
+svfloat16_t svmulx_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_f64_z)))
+svfloat64_t svmulx_z(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_f32_z)))
+svfloat32_t svmulx_z(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmulx_f16_z)))
+svfloat16_t svmulx_z(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnand_b_z)))
+svbool_t svnand_z(svbool_t, svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_f64_m)))
+svfloat64_t svneg_m(svfloat64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_f32_m)))
+svfloat32_t svneg_m(svfloat32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_f16_m)))
+svfloat16_t svneg_m(svfloat16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_f64_x)))
+svfloat64_t svneg_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_f32_x)))
+svfloat32_t svneg_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_f16_x)))
+svfloat16_t svneg_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_f64_z)))
+svfloat64_t svneg_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_f32_z)))
+svfloat32_t svneg_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_f16_z)))
+svfloat16_t svneg_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_s8_m)))
+svint8_t svneg_m(svint8_t, svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_s32_m)))
+svint32_t svneg_m(svint32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_s64_m)))
+svint64_t svneg_m(svint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_s16_m)))
+svint16_t svneg_m(svint16_t, svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_s8_x)))
+svint8_t svneg_x(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_s32_x)))
+svint32_t svneg_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_s64_x)))
+svint64_t svneg_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_s16_x)))
+svint16_t svneg_x(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_s8_z)))
+svint8_t svneg_z(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_s32_z)))
+svint32_t svneg_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_s64_z)))
+svint64_t svneg_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svneg_s16_z)))
+svint16_t svneg_z(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_n_f64_m)))
+svfloat64_t svnmad_m(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_n_f32_m)))
+svfloat32_t svnmad_m(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_n_f16_m)))
+svfloat16_t svnmad_m(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_n_f64_x)))
+svfloat64_t svnmad_x(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_n_f32_x)))
+svfloat32_t svnmad_x(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_n_f16_x)))
+svfloat16_t svnmad_x(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_n_f64_z)))
+svfloat64_t svnmad_z(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_n_f32_z)))
+svfloat32_t svnmad_z(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_n_f16_z)))
+svfloat16_t svnmad_z(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_f64_m)))
+svfloat64_t svnmad_m(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_f32_m)))
+svfloat32_t svnmad_m(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_f16_m)))
+svfloat16_t svnmad_m(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_f64_x)))
+svfloat64_t svnmad_x(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_f32_x)))
+svfloat32_t svnmad_x(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_f16_x)))
+svfloat16_t svnmad_x(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_f64_z)))
+svfloat64_t svnmad_z(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_f32_z)))
+svfloat32_t svnmad_z(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmad_f16_z)))
+svfloat16_t svnmad_z(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_n_f64_m)))
+svfloat64_t svnmla_m(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_n_f32_m)))
+svfloat32_t svnmla_m(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_n_f16_m)))
+svfloat16_t svnmla_m(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_n_f64_x)))
+svfloat64_t svnmla_x(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_n_f32_x)))
+svfloat32_t svnmla_x(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_n_f16_x)))
+svfloat16_t svnmla_x(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_n_f64_z)))
+svfloat64_t svnmla_z(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_n_f32_z)))
+svfloat32_t svnmla_z(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_n_f16_z)))
+svfloat16_t svnmla_z(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_f64_m)))
+svfloat64_t svnmla_m(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_f32_m)))
+svfloat32_t svnmla_m(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_f16_m)))
+svfloat16_t svnmla_m(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_f64_x)))
+svfloat64_t svnmla_x(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_f32_x)))
+svfloat32_t svnmla_x(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_f16_x)))
+svfloat16_t svnmla_x(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_f64_z)))
+svfloat64_t svnmla_z(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_f32_z)))
+svfloat32_t svnmla_z(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmla_f16_z)))
+svfloat16_t svnmla_z(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_n_f64_m)))
+svfloat64_t svnmls_m(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_n_f32_m)))
+svfloat32_t svnmls_m(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_n_f16_m)))
+svfloat16_t svnmls_m(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_n_f64_x)))
+svfloat64_t svnmls_x(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_n_f32_x)))
+svfloat32_t svnmls_x(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_n_f16_x)))
+svfloat16_t svnmls_x(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_n_f64_z)))
+svfloat64_t svnmls_z(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_n_f32_z)))
+svfloat32_t svnmls_z(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_n_f16_z)))
+svfloat16_t svnmls_z(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_f64_m)))
+svfloat64_t svnmls_m(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_f32_m)))
+svfloat32_t svnmls_m(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_f16_m)))
+svfloat16_t svnmls_m(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_f64_x)))
+svfloat64_t svnmls_x(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_f32_x)))
+svfloat32_t svnmls_x(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_f16_x)))
+svfloat16_t svnmls_x(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_f64_z)))
+svfloat64_t svnmls_z(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_f32_z)))
+svfloat32_t svnmls_z(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmls_f16_z)))
+svfloat16_t svnmls_z(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_n_f64_m)))
+svfloat64_t svnmsb_m(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_n_f32_m)))
+svfloat32_t svnmsb_m(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_n_f16_m)))
+svfloat16_t svnmsb_m(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_n_f64_x)))
+svfloat64_t svnmsb_x(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_n_f32_x)))
+svfloat32_t svnmsb_x(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_n_f16_x)))
+svfloat16_t svnmsb_x(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_n_f64_z)))
+svfloat64_t svnmsb_z(svbool_t, svfloat64_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_n_f32_z)))
+svfloat32_t svnmsb_z(svbool_t, svfloat32_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_n_f16_z)))
+svfloat16_t svnmsb_z(svbool_t, svfloat16_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_f64_m)))
+svfloat64_t svnmsb_m(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_f32_m)))
+svfloat32_t svnmsb_m(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_f16_m)))
+svfloat16_t svnmsb_m(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_f64_x)))
+svfloat64_t svnmsb_x(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_f32_x)))
+svfloat32_t svnmsb_x(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_f16_x)))
+svfloat16_t svnmsb_x(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_f64_z)))
+svfloat64_t svnmsb_z(svbool_t, svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_f32_z)))
+svfloat32_t svnmsb_z(svbool_t, svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmsb_f16_z)))
+svfloat16_t svnmsb_z(svbool_t, svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnor_b_z)))
+svbool_t svnor_z(svbool_t, svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_b_z)))
+svbool_t svnot_z(svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_u8_m)))
+svuint8_t svnot_m(svuint8_t, svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_u32_m)))
+svuint32_t svnot_m(svuint32_t, svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_u64_m)))
+svuint64_t svnot_m(svuint64_t, svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_u16_m)))
+svuint16_t svnot_m(svuint16_t, svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_s8_m)))
+svint8_t svnot_m(svint8_t, svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_s32_m)))
+svint32_t svnot_m(svint32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_s64_m)))
+svint64_t svnot_m(svint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_s16_m)))
+svint16_t svnot_m(svint16_t, svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_u8_x)))
+svuint8_t svnot_x(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_u32_x)))
+svuint32_t svnot_x(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_u64_x)))
+svuint64_t svnot_x(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_u16_x)))
+svuint16_t svnot_x(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_s8_x)))
+svint8_t svnot_x(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_s32_x)))
+svint32_t svnot_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_s64_x)))
+svint64_t svnot_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_s16_x)))
+svint16_t svnot_x(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_u8_z)))
+svuint8_t svnot_z(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_u32_z)))
+svuint32_t svnot_z(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_u64_z)))
+svuint64_t svnot_z(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_u16_z)))
+svuint16_t svnot_z(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_s8_z)))
+svint8_t svnot_z(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_s32_z)))
+svint32_t svnot_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_s64_z)))
+svint64_t svnot_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnot_s16_z)))
+svint16_t svnot_z(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorn_b_z)))
+svbool_t svorn_z(svbool_t, svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_b_z)))
+svbool_t svorr_z(svbool_t, svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_u8_m)))
+svuint8_t svorr_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_u32_m)))
+svuint32_t svorr_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_u64_m)))
+svuint64_t svorr_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_u16_m)))
+svuint16_t svorr_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_s8_m)))
+svint8_t svorr_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_s32_m)))
+svint32_t svorr_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_s64_m)))
+svint64_t svorr_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_s16_m)))
+svint16_t svorr_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_u8_x)))
+svuint8_t svorr_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_u32_x)))
+svuint32_t svorr_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_u64_x)))
+svuint64_t svorr_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_u16_x)))
+svuint16_t svorr_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_s8_x)))
+svint8_t svorr_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_s32_x)))
+svint32_t svorr_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_s64_x)))
+svint64_t svorr_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_s16_x)))
+svint16_t svorr_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_u8_z)))
+svuint8_t svorr_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_u32_z)))
+svuint32_t svorr_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_u64_z)))
+svuint64_t svorr_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_u16_z)))
+svuint16_t svorr_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_s8_z)))
+svint8_t svorr_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_s32_z)))
+svint32_t svorr_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_s64_z)))
+svint64_t svorr_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_n_s16_z)))
+svint16_t svorr_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_u8_m)))
+svuint8_t svorr_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_u32_m)))
+svuint32_t svorr_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_u64_m)))
+svuint64_t svorr_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_u16_m)))
+svuint16_t svorr_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_s8_m)))
+svint8_t svorr_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_s32_m)))
+svint32_t svorr_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_s64_m)))
+svint64_t svorr_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_s16_m)))
+svint16_t svorr_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_u8_x)))
+svuint8_t svorr_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_u32_x)))
+svuint32_t svorr_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_u64_x)))
+svuint64_t svorr_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_u16_x)))
+svuint16_t svorr_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_s8_x)))
+svint8_t svorr_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_s32_x)))
+svint32_t svorr_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_s64_x)))
+svint64_t svorr_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_s16_x)))
+svint16_t svorr_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_u8_z)))
+svuint8_t svorr_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_u32_z)))
+svuint32_t svorr_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_u64_z)))
+svuint64_t svorr_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_u16_z)))
+svuint16_t svorr_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_s8_z)))
+svint8_t svorr_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_s32_z)))
+svint32_t svorr_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_s64_z)))
+svint64_t svorr_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorr_s16_z)))
+svint16_t svorr_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorv_u8)))
+uint8_t svorv(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorv_u32)))
+uint32_t svorv(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorv_u64)))
+uint64_t svorv(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorv_u16)))
+uint16_t svorv(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorv_s8)))
+int8_t svorv(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorv_s32)))
+int32_t svorv(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorv_s64)))
+int64_t svorv(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorv_s16)))
+int16_t svorv(svbool_t, svint16_t);
+#define svpfalse(...) __builtin_sve_svpfalse_b(__VA_ARGS__)
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpfirst_b)))
+svbool_t svpfirst(svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfb_gather_u32base)))
+void svprfb_gather(svbool_t, svuint32_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfb_gather_u64base)))
+void svprfb_gather(svbool_t, svuint64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfb_gather_u32base_offset)))
+void svprfb_gather_offset(svbool_t, svuint32_t, int64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfb_gather_u64base_offset)))
+void svprfb_gather_offset(svbool_t, svuint64_t, int64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfb_gather_s32offset)))
+void svprfb_gather_offset(svbool_t, void const *, svint32_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfb_gather_u32offset)))
+void svprfb_gather_offset(svbool_t, void const *, svuint32_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfb_gather_s64offset)))
+void svprfb_gather_offset(svbool_t, void const *, svint64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfb_gather_u64offset)))
+void svprfb_gather_offset(svbool_t, void const *, svuint64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfd_gather_u32base)))
+void svprfd_gather(svbool_t, svuint32_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfd_gather_u64base)))
+void svprfd_gather(svbool_t, svuint64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfd_gather_u32base_index)))
+void svprfd_gather_index(svbool_t, svuint32_t, int64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfd_gather_u64base_index)))
+void svprfd_gather_index(svbool_t, svuint64_t, int64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfd_gather_s32index)))
+void svprfd_gather_index(svbool_t, void const *, svint32_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfd_gather_u32index)))
+void svprfd_gather_index(svbool_t, void const *, svuint32_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfd_gather_s64index)))
+void svprfd_gather_index(svbool_t, void const *, svint64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfd_gather_u64index)))
+void svprfd_gather_index(svbool_t, void const *, svuint64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfh_gather_u32base)))
+void svprfh_gather(svbool_t, svuint32_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfh_gather_u64base)))
+void svprfh_gather(svbool_t, svuint64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfh_gather_u32base_index)))
+void svprfh_gather_index(svbool_t, svuint32_t, int64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfh_gather_u64base_index)))
+void svprfh_gather_index(svbool_t, svuint64_t, int64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfh_gather_s32index)))
+void svprfh_gather_index(svbool_t, void const *, svint32_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfh_gather_u32index)))
+void svprfh_gather_index(svbool_t, void const *, svuint32_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfh_gather_s64index)))
+void svprfh_gather_index(svbool_t, void const *, svint64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfh_gather_u64index)))
+void svprfh_gather_index(svbool_t, void const *, svuint64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfw_gather_u32base)))
+void svprfw_gather(svbool_t, svuint32_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfw_gather_u64base)))
+void svprfw_gather(svbool_t, svuint64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfw_gather_u32base_index)))
+void svprfw_gather_index(svbool_t, svuint32_t, int64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfw_gather_u64base_index)))
+void svprfw_gather_index(svbool_t, svuint64_t, int64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfw_gather_s32index)))
+void svprfw_gather_index(svbool_t, void const *, svint32_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfw_gather_u32index)))
+void svprfw_gather_index(svbool_t, void const *, svuint32_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfw_gather_s64index)))
+void svprfw_gather_index(svbool_t, void const *, svint64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfw_gather_u64index)))
+void svprfw_gather_index(svbool_t, void const *, svuint64_t, sv_prfop);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8)))
+svint8_t svqadd(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32)))
+svint32_t svqadd(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64)))
+svint64_t svqadd(svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16)))
+svint16_t svqadd(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8)))
+svuint8_t svqadd(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32)))
+svuint32_t svqadd(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64)))
+svuint64_t svqadd(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16)))
+svuint16_t svqadd(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8)))
+svint8_t svqadd(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32)))
+svint32_t svqadd(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64)))
+svint64_t svqadd(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16)))
+svint16_t svqadd(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8)))
+svuint8_t svqadd(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32)))
+svuint32_t svqadd(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64)))
+svuint64_t svqadd(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16)))
+svuint16_t svqadd(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecb_n_s32)))
+int32_t svqdecb(int32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecb_n_s64)))
+int64_t svqdecb(int64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecb_n_u32)))
+uint32_t svqdecb(uint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecb_n_u64)))
+uint64_t svqdecb(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecb_pat_n_s32)))
+int32_t svqdecb_pat(int32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecb_pat_n_s64)))
+int64_t svqdecb_pat(int64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecb_pat_n_u32)))
+uint32_t svqdecb_pat(uint32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecb_pat_n_u64)))
+uint64_t svqdecb_pat(uint64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecd_n_s32)))
+int32_t svqdecd(int32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecd_n_s64)))
+int64_t svqdecd(int64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecd_n_u32)))
+uint32_t svqdecd(uint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecd_n_u64)))
+uint64_t svqdecd(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecd_s64)))
+svint64_t svqdecd(svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecd_u64)))
+svuint64_t svqdecd(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecd_pat_n_s32)))
+int32_t svqdecd_pat(int32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecd_pat_n_s64)))
+int64_t svqdecd_pat(int64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecd_pat_n_u32)))
+uint32_t svqdecd_pat(uint32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecd_pat_n_u64)))
+uint64_t svqdecd_pat(uint64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecd_pat_s64)))
+svint64_t svqdecd_pat(svint64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecd_pat_u64)))
+svuint64_t svqdecd_pat(svuint64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdech_n_s32)))
+int32_t svqdech(int32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdech_n_s64)))
+int64_t svqdech(int64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdech_n_u32)))
+uint32_t svqdech(uint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdech_n_u64)))
+uint64_t svqdech(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdech_s16)))
+svint16_t svqdech(svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdech_u16)))
+svuint16_t svqdech(svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdech_pat_n_s32)))
+int32_t svqdech_pat(int32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdech_pat_n_s64)))
+int64_t svqdech_pat(int64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdech_pat_n_u32)))
+uint32_t svqdech_pat(uint32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdech_pat_n_u64)))
+uint64_t svqdech_pat(uint64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdech_pat_s16)))
+svint16_t svqdech_pat(svint16_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdech_pat_u16)))
+svuint16_t svqdech_pat(svuint16_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_s32_b8)))
+int32_t svqdecp_b8(int32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_s32_b32)))
+int32_t svqdecp_b32(int32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_s32_b64)))
+int32_t svqdecp_b64(int32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_s32_b16)))
+int32_t svqdecp_b16(int32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_s64_b8)))
+int64_t svqdecp_b8(int64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_s64_b32)))
+int64_t svqdecp_b32(int64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_s64_b64)))
+int64_t svqdecp_b64(int64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_s64_b16)))
+int64_t svqdecp_b16(int64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_u32_b8)))
+uint32_t svqdecp_b8(uint32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_u32_b32)))
+uint32_t svqdecp_b32(uint32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_u32_b64)))
+uint32_t svqdecp_b64(uint32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_u32_b16)))
+uint32_t svqdecp_b16(uint32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_u64_b8)))
+uint64_t svqdecp_b8(uint64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_u64_b32)))
+uint64_t svqdecp_b32(uint64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_u64_b64)))
+uint64_t svqdecp_b64(uint64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_n_u64_b16)))
+uint64_t svqdecp_b16(uint64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_s32)))
+svint32_t svqdecp(svint32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_s64)))
+svint64_t svqdecp(svint64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_s16)))
+svint16_t svqdecp(svint16_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_u32)))
+svuint32_t svqdecp(svuint32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_u64)))
+svuint64_t svqdecp(svuint64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecp_u16)))
+svuint16_t svqdecp(svuint16_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecw_n_s32)))
+int32_t svqdecw(int32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecw_n_s64)))
+int64_t svqdecw(int64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecw_n_u32)))
+uint32_t svqdecw(uint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecw_n_u64)))
+uint64_t svqdecw(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecw_s32)))
+svint32_t svqdecw(svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecw_u32)))
+svuint32_t svqdecw(svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecw_pat_n_s32)))
+int32_t svqdecw_pat(int32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecw_pat_n_s64)))
+int64_t svqdecw_pat(int64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecw_pat_n_u32)))
+uint32_t svqdecw_pat(uint32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecw_pat_n_u64)))
+uint64_t svqdecw_pat(uint64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecw_pat_s32)))
+svint32_t svqdecw_pat(svint32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdecw_pat_u32)))
+svuint32_t svqdecw_pat(svuint32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincb_n_s32)))
+int32_t svqincb(int32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincb_n_s64)))
+int64_t svqincb(int64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincb_n_u32)))
+uint32_t svqincb(uint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincb_n_u64)))
+uint64_t svqincb(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincb_pat_n_s32)))
+int32_t svqincb_pat(int32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincb_pat_n_s64)))
+int64_t svqincb_pat(int64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincb_pat_n_u32)))
+uint32_t svqincb_pat(uint32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincb_pat_n_u64)))
+uint64_t svqincb_pat(uint64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincd_n_s32)))
+int32_t svqincd(int32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincd_n_s64)))
+int64_t svqincd(int64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincd_n_u32)))
+uint32_t svqincd(uint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincd_n_u64)))
+uint64_t svqincd(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincd_s64)))
+svint64_t svqincd(svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincd_u64)))
+svuint64_t svqincd(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincd_pat_n_s32)))
+int32_t svqincd_pat(int32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincd_pat_n_s64)))
+int64_t svqincd_pat(int64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincd_pat_n_u32)))
+uint32_t svqincd_pat(uint32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincd_pat_n_u64)))
+uint64_t svqincd_pat(uint64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincd_pat_s64)))
+svint64_t svqincd_pat(svint64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincd_pat_u64)))
+svuint64_t svqincd_pat(svuint64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqinch_n_s32)))
+int32_t svqinch(int32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqinch_n_s64)))
+int64_t svqinch(int64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqinch_n_u32)))
+uint32_t svqinch(uint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqinch_n_u64)))
+uint64_t svqinch(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqinch_s16)))
+svint16_t svqinch(svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqinch_u16)))
+svuint16_t svqinch(svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqinch_pat_n_s32)))
+int32_t svqinch_pat(int32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqinch_pat_n_s64)))
+int64_t svqinch_pat(int64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqinch_pat_n_u32)))
+uint32_t svqinch_pat(uint32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqinch_pat_n_u64)))
+uint64_t svqinch_pat(uint64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqinch_pat_s16)))
+svint16_t svqinch_pat(svint16_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqinch_pat_u16)))
+svuint16_t svqinch_pat(svuint16_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_s32_b8)))
+int32_t svqincp_b8(int32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_s32_b32)))
+int32_t svqincp_b32(int32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_s32_b64)))
+int32_t svqincp_b64(int32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_s32_b16)))
+int32_t svqincp_b16(int32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_s64_b8)))
+int64_t svqincp_b8(int64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_s64_b32)))
+int64_t svqincp_b32(int64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_s64_b64)))
+int64_t svqincp_b64(int64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_s64_b16)))
+int64_t svqincp_b16(int64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_u32_b8)))
+uint32_t svqincp_b8(uint32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_u32_b32)))
+uint32_t svqincp_b32(uint32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_u32_b64)))
+uint32_t svqincp_b64(uint32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_u32_b16)))
+uint32_t svqincp_b16(uint32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_u64_b8)))
+uint64_t svqincp_b8(uint64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_u64_b32)))
+uint64_t svqincp_b32(uint64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_u64_b64)))
+uint64_t svqincp_b64(uint64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_n_u64_b16)))
+uint64_t svqincp_b16(uint64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_s32)))
+svint32_t svqincp(svint32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_s64)))
+svint64_t svqincp(svint64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_s16)))
+svint16_t svqincp(svint16_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_u32)))
+svuint32_t svqincp(svuint32_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_u64)))
+svuint64_t svqincp(svuint64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincp_u16)))
+svuint16_t svqincp(svuint16_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincw_n_s32)))
+int32_t svqincw(int32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincw_n_s64)))
+int64_t svqincw(int64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincw_n_u32)))
+uint32_t svqincw(uint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincw_n_u64)))
+uint64_t svqincw(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincw_s32)))
+svint32_t svqincw(svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincw_u32)))
+svuint32_t svqincw(svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincw_pat_n_s32)))
+int32_t svqincw_pat(int32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincw_pat_n_s64)))
+int64_t svqincw_pat(int64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincw_pat_n_u32)))
+uint32_t svqincw_pat(uint32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincw_pat_n_u64)))
+uint64_t svqincw_pat(uint64_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincw_pat_s32)))
+svint32_t svqincw_pat(svint32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqincw_pat_u32)))
+svuint32_t svqincw_pat(svuint32_t, sv_pattern, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8)))
+svint8_t svqsub(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32)))
+svint32_t svqsub(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64)))
+svint64_t svqsub(svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16)))
+svint16_t svqsub(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8)))
+svuint8_t svqsub(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32)))
+svuint32_t svqsub(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64)))
+svuint64_t svqsub(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16)))
+svuint16_t svqsub(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8)))
+svint8_t svqsub(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32)))
+svint32_t svqsub(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64)))
+svint64_t svqsub(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16)))
+svint16_t svqsub(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8)))
+svuint8_t svqsub(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32)))
+svuint32_t svqsub(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64)))
+svuint64_t svqsub(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16)))
+svuint16_t svqsub(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_u8_m)))
+svuint8_t svrbit_m(svuint8_t, svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_u32_m)))
+svuint32_t svrbit_m(svuint32_t, svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_u64_m)))
+svuint64_t svrbit_m(svuint64_t, svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_u16_m)))
+svuint16_t svrbit_m(svuint16_t, svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_s8_m)))
+svint8_t svrbit_m(svint8_t, svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_s32_m)))
+svint32_t svrbit_m(svint32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_s64_m)))
+svint64_t svrbit_m(svint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_s16_m)))
+svint16_t svrbit_m(svint16_t, svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_u8_x)))
+svuint8_t svrbit_x(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_u32_x)))
+svuint32_t svrbit_x(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_u64_x)))
+svuint64_t svrbit_x(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_u16_x)))
+svuint16_t svrbit_x(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_s8_x)))
+svint8_t svrbit_x(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_s32_x)))
+svint32_t svrbit_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_s64_x)))
+svint64_t svrbit_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_s16_x)))
+svint16_t svrbit_x(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_u8_z)))
+svuint8_t svrbit_z(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_u32_z)))
+svuint32_t svrbit_z(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_u64_z)))
+svuint64_t svrbit_z(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_u16_z)))
+svuint16_t svrbit_z(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_s8_z)))
+svint8_t svrbit_z(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_s32_z)))
+svint32_t svrbit_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_s64_z)))
+svint64_t svrbit_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrbit_s16_z)))
+svint16_t svrbit_z(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_f64)))
+svfloat64_t svrecpe(svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_f32)))
+svfloat32_t svrecpe(svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_f16)))
+svfloat16_t svrecpe(svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecps_f64)))
+svfloat64_t svrecps(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecps_f32)))
+svfloat32_t svrecps(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecps_f16)))
+svfloat16_t svrecps(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpx_f64_m)))
+svfloat64_t svrecpx_m(svfloat64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpx_f32_m)))
+svfloat32_t svrecpx_m(svfloat32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpx_f16_m)))
+svfloat16_t svrecpx_m(svfloat16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpx_f64_x)))
+svfloat64_t svrecpx_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpx_f32_x)))
+svfloat32_t svrecpx_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpx_f16_x)))
+svfloat16_t svrecpx_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpx_f64_z)))
+svfloat64_t svrecpx_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpx_f32_z)))
+svfloat32_t svrecpx_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpx_f16_z)))
+svfloat16_t svrecpx_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_u8)))
+svuint8_t svrev(svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_u32)))
+svuint32_t svrev(svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_u64)))
+svuint64_t svrev(svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_u16)))
+svuint16_t svrev(svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_s8)))
+svint8_t svrev(svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_f64)))
+svfloat64_t svrev(svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_f32)))
+svfloat32_t svrev(svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_f16)))
+svfloat16_t svrev(svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_s32)))
+svint32_t svrev(svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_s64)))
+svint64_t svrev(svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_s16)))
+svint16_t svrev(svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_u32_m)))
+svuint32_t svrevb_m(svuint32_t, svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_u64_m)))
+svuint64_t svrevb_m(svuint64_t, svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_u16_m)))
+svuint16_t svrevb_m(svuint16_t, svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_s32_m)))
+svint32_t svrevb_m(svint32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_s64_m)))
+svint64_t svrevb_m(svint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_s16_m)))
+svint16_t svrevb_m(svint16_t, svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_u32_x)))
+svuint32_t svrevb_x(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_u64_x)))
+svuint64_t svrevb_x(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_u16_x)))
+svuint16_t svrevb_x(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_s32_x)))
+svint32_t svrevb_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_s64_x)))
+svint64_t svrevb_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_s16_x)))
+svint16_t svrevb_x(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_u32_z)))
+svuint32_t svrevb_z(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_u64_z)))
+svuint64_t svrevb_z(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_u16_z)))
+svuint16_t svrevb_z(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_s32_z)))
+svint32_t svrevb_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_s64_z)))
+svint64_t svrevb_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevb_s16_z)))
+svint16_t svrevb_z(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevh_u32_m)))
+svuint32_t svrevh_m(svuint32_t, svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevh_u64_m)))
+svuint64_t svrevh_m(svuint64_t, svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevh_s32_m)))
+svint32_t svrevh_m(svint32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevh_s64_m)))
+svint64_t svrevh_m(svint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevh_u32_x)))
+svuint32_t svrevh_x(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevh_u64_x)))
+svuint64_t svrevh_x(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevh_s32_x)))
+svint32_t svrevh_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevh_s64_x)))
+svint64_t svrevh_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevh_u32_z)))
+svuint32_t svrevh_z(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevh_u64_z)))
+svuint64_t svrevh_z(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevh_s32_z)))
+svint32_t svrevh_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevh_s64_z)))
+svint64_t svrevh_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevw_u64_m)))
+svuint64_t svrevw_m(svuint64_t, svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevw_s64_m)))
+svint64_t svrevw_m(svint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevw_u64_x)))
+svuint64_t svrevw_x(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevw_s64_x)))
+svint64_t svrevw_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevw_u64_z)))
+svuint64_t svrevw_z(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevw_s64_z)))
+svint64_t svrevw_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinta_f64_m)))
+svfloat64_t svrinta_m(svfloat64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinta_f32_m)))
+svfloat32_t svrinta_m(svfloat32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinta_f16_m)))
+svfloat16_t svrinta_m(svfloat16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinta_f64_x)))
+svfloat64_t svrinta_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinta_f32_x)))
+svfloat32_t svrinta_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinta_f16_x)))
+svfloat16_t svrinta_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinta_f64_z)))
+svfloat64_t svrinta_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinta_f32_z)))
+svfloat32_t svrinta_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinta_f16_z)))
+svfloat16_t svrinta_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinti_f64_m)))
+svfloat64_t svrinti_m(svfloat64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinti_f32_m)))
+svfloat32_t svrinti_m(svfloat32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinti_f16_m)))
+svfloat16_t svrinti_m(svfloat16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinti_f64_x)))
+svfloat64_t svrinti_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinti_f32_x)))
+svfloat32_t svrinti_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinti_f16_x)))
+svfloat16_t svrinti_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinti_f64_z)))
+svfloat64_t svrinti_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinti_f32_z)))
+svfloat32_t svrinti_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrinti_f16_z)))
+svfloat16_t svrinti_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintm_f64_m)))
+svfloat64_t svrintm_m(svfloat64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintm_f32_m)))
+svfloat32_t svrintm_m(svfloat32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintm_f16_m)))
+svfloat16_t svrintm_m(svfloat16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintm_f64_x)))
+svfloat64_t svrintm_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintm_f32_x)))
+svfloat32_t svrintm_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintm_f16_x)))
+svfloat16_t svrintm_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintm_f64_z)))
+svfloat64_t svrintm_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintm_f32_z)))
+svfloat32_t svrintm_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintm_f16_z)))
+svfloat16_t svrintm_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintn_f64_m)))
+svfloat64_t svrintn_m(svfloat64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintn_f32_m)))
+svfloat32_t svrintn_m(svfloat32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintn_f16_m)))
+svfloat16_t svrintn_m(svfloat16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintn_f64_x)))
+svfloat64_t svrintn_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintn_f32_x)))
+svfloat32_t svrintn_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintn_f16_x)))
+svfloat16_t svrintn_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintn_f64_z)))
+svfloat64_t svrintn_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintn_f32_z)))
+svfloat32_t svrintn_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintn_f16_z)))
+svfloat16_t svrintn_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintp_f64_m)))
+svfloat64_t svrintp_m(svfloat64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintp_f32_m)))
+svfloat32_t svrintp_m(svfloat32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintp_f16_m)))
+svfloat16_t svrintp_m(svfloat16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintp_f64_x)))
+svfloat64_t svrintp_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintp_f32_x)))
+svfloat32_t svrintp_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintp_f16_x)))
+svfloat16_t svrintp_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintp_f64_z)))
+svfloat64_t svrintp_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintp_f32_z)))
+svfloat32_t svrintp_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintp_f16_z)))
+svfloat16_t svrintp_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintx_f64_m)))
+svfloat64_t svrintx_m(svfloat64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintx_f32_m)))
+svfloat32_t svrintx_m(svfloat32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintx_f16_m)))
+svfloat16_t svrintx_m(svfloat16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintx_f64_x)))
+svfloat64_t svrintx_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintx_f32_x)))
+svfloat32_t svrintx_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintx_f16_x)))
+svfloat16_t svrintx_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintx_f64_z)))
+svfloat64_t svrintx_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintx_f32_z)))
+svfloat32_t svrintx_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintx_f16_z)))
+svfloat16_t svrintx_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintz_f64_m)))
+svfloat64_t svrintz_m(svfloat64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintz_f32_m)))
+svfloat32_t svrintz_m(svfloat32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintz_f16_m)))
+svfloat16_t svrintz_m(svfloat16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintz_f64_x)))
+svfloat64_t svrintz_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintz_f32_x)))
+svfloat32_t svrintz_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintz_f16_x)))
+svfloat16_t svrintz_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintz_f64_z)))
+svfloat64_t svrintz_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintz_f32_z)))
+svfloat32_t svrintz_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrintz_f16_z)))
+svfloat16_t svrintz_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_f64)))
+svfloat64_t svrsqrte(svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_f32)))
+svfloat32_t svrsqrte(svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_f16)))
+svfloat16_t svrsqrte(svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrts_f64)))
+svfloat64_t svrsqrts(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrts_f32)))
+svfloat32_t svrsqrts(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrts_f16)))
+svfloat16_t svrsqrts(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_n_f64_m)))
+svfloat64_t svscale_m(svbool_t, svfloat64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_n_f32_m)))
+svfloat32_t svscale_m(svbool_t, svfloat32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_n_f16_m)))
+svfloat16_t svscale_m(svbool_t, svfloat16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_n_f64_x)))
+svfloat64_t svscale_x(svbool_t, svfloat64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_n_f32_x)))
+svfloat32_t svscale_x(svbool_t, svfloat32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_n_f16_x)))
+svfloat16_t svscale_x(svbool_t, svfloat16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_n_f64_z)))
+svfloat64_t svscale_z(svbool_t, svfloat64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_n_f32_z)))
+svfloat32_t svscale_z(svbool_t, svfloat32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_n_f16_z)))
+svfloat16_t svscale_z(svbool_t, svfloat16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_f64_m)))
+svfloat64_t svscale_m(svbool_t, svfloat64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_f32_m)))
+svfloat32_t svscale_m(svbool_t, svfloat32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_f16_m)))
+svfloat16_t svscale_m(svbool_t, svfloat16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_f64_x)))
+svfloat64_t svscale_x(svbool_t, svfloat64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_f32_x)))
+svfloat32_t svscale_x(svbool_t, svfloat32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_f16_x)))
+svfloat16_t svscale_x(svbool_t, svfloat16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_f64_z)))
+svfloat64_t svscale_z(svbool_t, svfloat64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_f32_z)))
+svfloat32_t svscale_z(svbool_t, svfloat32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svscale_f16_z)))
+svfloat16_t svscale_z(svbool_t, svfloat16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_b)))
+svbool_t svsel(svbool_t, svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_u8)))
+svuint8_t svsel(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_u32)))
+svuint32_t svsel(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_u64)))
+svuint64_t svsel(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_u16)))
+svuint16_t svsel(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s8)))
+svint8_t svsel(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_f64)))
+svfloat64_t svsel(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_f32)))
+svfloat32_t svsel(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_f16)))
+svfloat16_t svsel(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s32)))
+svint32_t svsel(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s64)))
+svint64_t svsel(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s16)))
+svint16_t svsel(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_u8)))
+svuint8x2_t svset2(svuint8x2_t, uint64_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_u32)))
+svuint32x2_t svset2(svuint32x2_t, uint64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_u64)))
+svuint64x2_t svset2(svuint64x2_t, uint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_u16)))
+svuint16x2_t svset2(svuint16x2_t, uint64_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_s8)))
+svint8x2_t svset2(svint8x2_t, uint64_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_f64)))
+svfloat64x2_t svset2(svfloat64x2_t, uint64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_f32)))
+svfloat32x2_t svset2(svfloat32x2_t, uint64_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_f16)))
+svfloat16x2_t svset2(svfloat16x2_t, uint64_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_s32)))
+svint32x2_t svset2(svint32x2_t, uint64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_s64)))
+svint64x2_t svset2(svint64x2_t, uint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_s16)))
+svint16x2_t svset2(svint16x2_t, uint64_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_u8)))
+svuint8x3_t svset3(svuint8x3_t, uint64_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_u32)))
+svuint32x3_t svset3(svuint32x3_t, uint64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_u64)))
+svuint64x3_t svset3(svuint64x3_t, uint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_u16)))
+svuint16x3_t svset3(svuint16x3_t, uint64_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_s8)))
+svint8x3_t svset3(svint8x3_t, uint64_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_f64)))
+svfloat64x3_t svset3(svfloat64x3_t, uint64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_f32)))
+svfloat32x3_t svset3(svfloat32x3_t, uint64_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_f16)))
+svfloat16x3_t svset3(svfloat16x3_t, uint64_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_s32)))
+svint32x3_t svset3(svint32x3_t, uint64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_s64)))
+svint64x3_t svset3(svint64x3_t, uint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_s16)))
+svint16x3_t svset3(svint16x3_t, uint64_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_u8)))
+svuint8x4_t svset4(svuint8x4_t, uint64_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_u32)))
+svuint32x4_t svset4(svuint32x4_t, uint64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_u64)))
+svuint64x4_t svset4(svuint64x4_t, uint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_u16)))
+svuint16x4_t svset4(svuint16x4_t, uint64_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_s8)))
+svint8x4_t svset4(svint8x4_t, uint64_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_f64)))
+svfloat64x4_t svset4(svfloat64x4_t, uint64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_f32)))
+svfloat32x4_t svset4(svfloat32x4_t, uint64_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_f16)))
+svfloat16x4_t svset4(svfloat16x4_t, uint64_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_s32)))
+svint32x4_t svset4(svint32x4_t, uint64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_s64)))
+svint64x4_t svset4(svint64x4_t, uint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_s16)))
+svint16x4_t svset4(svint16x4_t, uint64_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_u8)))
+svuint8_t svsplice(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_u32)))
+svuint32_t svsplice(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_u64)))
+svuint64_t svsplice(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_u16)))
+svuint16_t svsplice(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_s8)))
+svint8_t svsplice(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_f64)))
+svfloat64_t svsplice(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_f32)))
+svfloat32_t svsplice(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_f16)))
+svfloat16_t svsplice(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_s32)))
+svint32_t svsplice(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_s64)))
+svint64_t svsplice(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_s16)))
+svint16_t svsplice(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqrt_f64_m)))
+svfloat64_t svsqrt_m(svfloat64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqrt_f32_m)))
+svfloat32_t svsqrt_m(svfloat32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqrt_f16_m)))
+svfloat16_t svsqrt_m(svfloat16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqrt_f64_x)))
+svfloat64_t svsqrt_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqrt_f32_x)))
+svfloat32_t svsqrt_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqrt_f16_x)))
+svfloat16_t svsqrt_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqrt_f64_z)))
+svfloat64_t svsqrt_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqrt_f32_z)))
+svfloat32_t svsqrt_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqrt_f16_z)))
+svfloat16_t svsqrt_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u8)))
+void svst1(svbool_t, uint8_t *, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u32)))
+void svst1(svbool_t, uint32_t *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u64)))
+void svst1(svbool_t, uint64_t *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u16)))
+void svst1(svbool_t, uint16_t *, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s8)))
+void svst1(svbool_t, int8_t *, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_f64)))
+void svst1(svbool_t, float64_t *, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_f32)))
+void svst1(svbool_t, float32_t *, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_f16)))
+void svst1(svbool_t, float16_t *, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s32)))
+void svst1(svbool_t, int32_t *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s64)))
+void svst1(svbool_t, int64_t *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s16)))
+void svst1(svbool_t, int16_t *, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u32base_index_u32)))
+void svst1_scatter_index(svbool_t, svuint32_t, int64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u64base_index_u64)))
+void svst1_scatter_index(svbool_t, svuint64_t, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u64base_index_f64)))
+void svst1_scatter_index(svbool_t, svuint64_t, int64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u32base_index_f32)))
+void svst1_scatter_index(svbool_t, svuint32_t, int64_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u32base_index_s32)))
+void svst1_scatter_index(svbool_t, svuint32_t, int64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u64base_index_s64)))
+void svst1_scatter_index(svbool_t, svuint64_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u32base_offset_u32)))
+void svst1_scatter_offset(svbool_t, svuint32_t, int64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u64base_offset_u64)))
+void svst1_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u64base_offset_f64)))
+void svst1_scatter_offset(svbool_t, svuint64_t, int64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u32base_offset_f32)))
+void svst1_scatter_offset(svbool_t, svuint32_t, int64_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u32base_offset_s32)))
+void svst1_scatter_offset(svbool_t, svuint32_t, int64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u64base_offset_s64)))
+void svst1_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u32base_u32)))
+void svst1_scatter(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u64base_u64)))
+void svst1_scatter(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u64base_f64)))
+void svst1_scatter(svbool_t, svuint64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u32base_f32)))
+void svst1_scatter(svbool_t, svuint32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u32base_s32)))
+void svst1_scatter(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u64base_s64)))
+void svst1_scatter(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_s32index_u32)))
+void svst1_scatter_index(svbool_t, uint32_t *, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_s32index_f32)))
+void svst1_scatter_index(svbool_t, float32_t *, svint32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_s32index_s32)))
+void svst1_scatter_index(svbool_t, int32_t *, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u32index_u32)))
+void svst1_scatter_index(svbool_t, uint32_t *, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u32index_f32)))
+void svst1_scatter_index(svbool_t, float32_t *, svuint32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u32index_s32)))
+void svst1_scatter_index(svbool_t, int32_t *, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_s64index_u64)))
+void svst1_scatter_index(svbool_t, uint64_t *, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_s64index_f64)))
+void svst1_scatter_index(svbool_t, float64_t *, svint64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_s64index_s64)))
+void svst1_scatter_index(svbool_t, int64_t *, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u64index_u64)))
+void svst1_scatter_index(svbool_t, uint64_t *, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u64index_f64)))
+void svst1_scatter_index(svbool_t, float64_t *, svuint64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u64index_s64)))
+void svst1_scatter_index(svbool_t, int64_t *, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_s32offset_u32)))
+void svst1_scatter_offset(svbool_t, uint32_t *, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_s32offset_f32)))
+void svst1_scatter_offset(svbool_t, float32_t *, svint32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_s32offset_s32)))
+void svst1_scatter_offset(svbool_t, int32_t *, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u32offset_u32)))
+void svst1_scatter_offset(svbool_t, uint32_t *, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u32offset_f32)))
+void svst1_scatter_offset(svbool_t, float32_t *, svuint32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u32offset_s32)))
+void svst1_scatter_offset(svbool_t, int32_t *, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_s64offset_u64)))
+void svst1_scatter_offset(svbool_t, uint64_t *, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_s64offset_f64)))
+void svst1_scatter_offset(svbool_t, float64_t *, svint64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_s64offset_s64)))
+void svst1_scatter_offset(svbool_t, int64_t *, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u64offset_u64)))
+void svst1_scatter_offset(svbool_t, uint64_t *, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u64offset_f64)))
+void svst1_scatter_offset(svbool_t, float64_t *, svuint64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_scatter_u64offset_s64)))
+void svst1_scatter_offset(svbool_t, int64_t *, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u8)))
+void svst1_vnum(svbool_t, uint8_t *, int64_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u32)))
+void svst1_vnum(svbool_t, uint32_t *, int64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u64)))
+void svst1_vnum(svbool_t, uint64_t *, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u16)))
+void svst1_vnum(svbool_t, uint16_t *, int64_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s8)))
+void svst1_vnum(svbool_t, int8_t *, int64_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_f64)))
+void svst1_vnum(svbool_t, float64_t *, int64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_f32)))
+void svst1_vnum(svbool_t, float32_t *, int64_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_f16)))
+void svst1_vnum(svbool_t, float16_t *, int64_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s32)))
+void svst1_vnum(svbool_t, int32_t *, int64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s64)))
+void svst1_vnum(svbool_t, int64_t *, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s16)))
+void svst1_vnum(svbool_t, int16_t *, int64_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_s32)))
+void svst1b(svbool_t, int8_t *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_s64)))
+void svst1b(svbool_t, int8_t *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_s16)))
+void svst1b(svbool_t, int8_t *, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_u32)))
+void svst1b(svbool_t, uint8_t *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_u64)))
+void svst1b(svbool_t, uint8_t *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_u16)))
+void svst1b(svbool_t, uint8_t *, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_u32base_offset_u32)))
+void svst1b_scatter_offset(svbool_t, svuint32_t, int64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_u64base_offset_u64)))
+void svst1b_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_u32base_offset_s32)))
+void svst1b_scatter_offset(svbool_t, svuint32_t, int64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_u64base_offset_s64)))
+void svst1b_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_u32base_u32)))
+void svst1b_scatter(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_u64base_u64)))
+void svst1b_scatter(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_u32base_s32)))
+void svst1b_scatter(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_u64base_s64)))
+void svst1b_scatter(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_s32offset_s32)))
+void svst1b_scatter_offset(svbool_t, int8_t *, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_s32offset_u32)))
+void svst1b_scatter_offset(svbool_t, uint8_t *, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_u32offset_s32)))
+void svst1b_scatter_offset(svbool_t, int8_t *, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_u32offset_u32)))
+void svst1b_scatter_offset(svbool_t, uint8_t *, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_s64offset_s64)))
+void svst1b_scatter_offset(svbool_t, int8_t *, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_s64offset_u64)))
+void svst1b_scatter_offset(svbool_t, uint8_t *, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_u64offset_s64)))
+void svst1b_scatter_offset(svbool_t, int8_t *, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_scatter_u64offset_u64)))
+void svst1b_scatter_offset(svbool_t, uint8_t *, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_vnum_s32)))
+void svst1b_vnum(svbool_t, int8_t *, int64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_vnum_s64)))
+void svst1b_vnum(svbool_t, int8_t *, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_vnum_s16)))
+void svst1b_vnum(svbool_t, int8_t *, int64_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_vnum_u32)))
+void svst1b_vnum(svbool_t, uint8_t *, int64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_vnum_u64)))
+void svst1b_vnum(svbool_t, uint8_t *, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_vnum_u16)))
+void svst1b_vnum(svbool_t, uint8_t *, int64_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_s32)))
+void svst1h(svbool_t, int16_t *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_s64)))
+void svst1h(svbool_t, int16_t *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_u32)))
+void svst1h(svbool_t, uint16_t *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_u64)))
+void svst1h(svbool_t, uint16_t *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u32base_index_u32)))
+void svst1h_scatter_index(svbool_t, svuint32_t, int64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u64base_index_u64)))
+void svst1h_scatter_index(svbool_t, svuint64_t, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u32base_index_s32)))
+void svst1h_scatter_index(svbool_t, svuint32_t, int64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u64base_index_s64)))
+void svst1h_scatter_index(svbool_t, svuint64_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u32base_offset_u32)))
+void svst1h_scatter_offset(svbool_t, svuint32_t, int64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u64base_offset_u64)))
+void svst1h_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u32base_offset_s32)))
+void svst1h_scatter_offset(svbool_t, svuint32_t, int64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u64base_offset_s64)))
+void svst1h_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u32base_u32)))
+void svst1h_scatter(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u64base_u64)))
+void svst1h_scatter(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u32base_s32)))
+void svst1h_scatter(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u64base_s64)))
+void svst1h_scatter(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_s32index_s32)))
+void svst1h_scatter_index(svbool_t, int16_t *, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_s32index_u32)))
+void svst1h_scatter_index(svbool_t, uint16_t *, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u32index_s32)))
+void svst1h_scatter_index(svbool_t, int16_t *, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u32index_u32)))
+void svst1h_scatter_index(svbool_t, uint16_t *, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_s64index_s64)))
+void svst1h_scatter_index(svbool_t, int16_t *, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_s64index_u64)))
+void svst1h_scatter_index(svbool_t, uint16_t *, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u64index_s64)))
+void svst1h_scatter_index(svbool_t, int16_t *, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u64index_u64)))
+void svst1h_scatter_index(svbool_t, uint16_t *, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_s32offset_s32)))
+void svst1h_scatter_offset(svbool_t, int16_t *, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_s32offset_u32)))
+void svst1h_scatter_offset(svbool_t, uint16_t *, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u32offset_s32)))
+void svst1h_scatter_offset(svbool_t, int16_t *, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u32offset_u32)))
+void svst1h_scatter_offset(svbool_t, uint16_t *, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_s64offset_s64)))
+void svst1h_scatter_offset(svbool_t, int16_t *, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_s64offset_u64)))
+void svst1h_scatter_offset(svbool_t, uint16_t *, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u64offset_s64)))
+void svst1h_scatter_offset(svbool_t, int16_t *, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_scatter_u64offset_u64)))
+void svst1h_scatter_offset(svbool_t, uint16_t *, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_vnum_s32)))
+void svst1h_vnum(svbool_t, int16_t *, int64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_vnum_s64)))
+void svst1h_vnum(svbool_t, int16_t *, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_vnum_u32)))
+void svst1h_vnum(svbool_t, uint16_t *, int64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1h_vnum_u64)))
+void svst1h_vnum(svbool_t, uint16_t *, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_s64)))
+void svst1w(svbool_t, int32_t *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_u64)))
+void svst1w(svbool_t, uint32_t *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_scatter_u64base_index_u64)))
+void svst1w_scatter_index(svbool_t, svuint64_t, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_scatter_u64base_index_s64)))
+void svst1w_scatter_index(svbool_t, svuint64_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_scatter_u64base_offset_u64)))
+void svst1w_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_scatter_u64base_offset_s64)))
+void svst1w_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_scatter_u64base_u64)))
+void svst1w_scatter(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_scatter_u64base_s64)))
+void svst1w_scatter(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_scatter_s64index_s64)))
+void svst1w_scatter_index(svbool_t, int32_t *, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_scatter_s64index_u64)))
+void svst1w_scatter_index(svbool_t, uint32_t *, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_scatter_u64index_s64)))
+void svst1w_scatter_index(svbool_t, int32_t *, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_scatter_u64index_u64)))
+void svst1w_scatter_index(svbool_t, uint32_t *, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_scatter_s64offset_s64)))
+void svst1w_scatter_offset(svbool_t, int32_t *, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_scatter_s64offset_u64)))
+void svst1w_scatter_offset(svbool_t, uint32_t *, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_scatter_u64offset_s64)))
+void svst1w_scatter_offset(svbool_t, int32_t *, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_scatter_u64offset_u64)))
+void svst1w_scatter_offset(svbool_t, uint32_t *, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_vnum_s64)))
+void svst1w_vnum(svbool_t, int32_t *, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1w_vnum_u64)))
+void svst1w_vnum(svbool_t, uint32_t *, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_u8)))
+void svst2(svbool_t, uint8_t *, svuint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_u32)))
+void svst2(svbool_t, uint32_t *, svuint32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_u64)))
+void svst2(svbool_t, uint64_t *, svuint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_u16)))
+void svst2(svbool_t, uint16_t *, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_s8)))
+void svst2(svbool_t, int8_t *, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_f64)))
+void svst2(svbool_t, float64_t *, svfloat64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_f32)))
+void svst2(svbool_t, float32_t *, svfloat32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_f16)))
+void svst2(svbool_t, float16_t *, svfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_s32)))
+void svst2(svbool_t, int32_t *, svint32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_s64)))
+void svst2(svbool_t, int64_t *, svint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_s16)))
+void svst2(svbool_t, int16_t *, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_u8)))
+void svst2_vnum(svbool_t, uint8_t *, int64_t, svuint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_u32)))
+void svst2_vnum(svbool_t, uint32_t *, int64_t, svuint32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_u64)))
+void svst2_vnum(svbool_t, uint64_t *, int64_t, svuint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_u16)))
+void svst2_vnum(svbool_t, uint16_t *, int64_t, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_s8)))
+void svst2_vnum(svbool_t, int8_t *, int64_t, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_f64)))
+void svst2_vnum(svbool_t, float64_t *, int64_t, svfloat64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_f32)))
+void svst2_vnum(svbool_t, float32_t *, int64_t, svfloat32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_f16)))
+void svst2_vnum(svbool_t, float16_t *, int64_t, svfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_s32)))
+void svst2_vnum(svbool_t, int32_t *, int64_t, svint32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_s64)))
+void svst2_vnum(svbool_t, int64_t *, int64_t, svint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_s16)))
+void svst2_vnum(svbool_t, int16_t *, int64_t, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_u8)))
+void svst3(svbool_t, uint8_t *, svuint8x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_u32)))
+void svst3(svbool_t, uint32_t *, svuint32x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_u64)))
+void svst3(svbool_t, uint64_t *, svuint64x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_u16)))
+void svst3(svbool_t, uint16_t *, svuint16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_s8)))
+void svst3(svbool_t, int8_t *, svint8x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_f64)))
+void svst3(svbool_t, float64_t *, svfloat64x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_f32)))
+void svst3(svbool_t, float32_t *, svfloat32x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_f16)))
+void svst3(svbool_t, float16_t *, svfloat16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_s32)))
+void svst3(svbool_t, int32_t *, svint32x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_s64)))
+void svst3(svbool_t, int64_t *, svint64x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_s16)))
+void svst3(svbool_t, int16_t *, svint16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_u8)))
+void svst3_vnum(svbool_t, uint8_t *, int64_t, svuint8x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_u32)))
+void svst3_vnum(svbool_t, uint32_t *, int64_t, svuint32x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_u64)))
+void svst3_vnum(svbool_t, uint64_t *, int64_t, svuint64x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_u16)))
+void svst3_vnum(svbool_t, uint16_t *, int64_t, svuint16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_s8)))
+void svst3_vnum(svbool_t, int8_t *, int64_t, svint8x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_f64)))
+void svst3_vnum(svbool_t, float64_t *, int64_t, svfloat64x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_f32)))
+void svst3_vnum(svbool_t, float32_t *, int64_t, svfloat32x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_f16)))
+void svst3_vnum(svbool_t, float16_t *, int64_t, svfloat16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_s32)))
+void svst3_vnum(svbool_t, int32_t *, int64_t, svint32x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_s64)))
+void svst3_vnum(svbool_t, int64_t *, int64_t, svint64x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_s16)))
+void svst3_vnum(svbool_t, int16_t *, int64_t, svint16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_u8)))
+void svst4(svbool_t, uint8_t *, svuint8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_u32)))
+void svst4(svbool_t, uint32_t *, svuint32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_u64)))
+void svst4(svbool_t, uint64_t *, svuint64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_u16)))
+void svst4(svbool_t, uint16_t *, svuint16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_s8)))
+void svst4(svbool_t, int8_t *, svint8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_f64)))
+void svst4(svbool_t, float64_t *, svfloat64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_f32)))
+void svst4(svbool_t, float32_t *, svfloat32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_f16)))
+void svst4(svbool_t, float16_t *, svfloat16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_s32)))
+void svst4(svbool_t, int32_t *, svint32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_s64)))
+void svst4(svbool_t, int64_t *, svint64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_s16)))
+void svst4(svbool_t, int16_t *, svint16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_u8)))
+void svst4_vnum(svbool_t, uint8_t *, int64_t, svuint8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_u32)))
+void svst4_vnum(svbool_t, uint32_t *, int64_t, svuint32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_u64)))
+void svst4_vnum(svbool_t, uint64_t *, int64_t, svuint64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_u16)))
+void svst4_vnum(svbool_t, uint16_t *, int64_t, svuint16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_s8)))
+void svst4_vnum(svbool_t, int8_t *, int64_t, svint8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_f64)))
+void svst4_vnum(svbool_t, float64_t *, int64_t, svfloat64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_f32)))
+void svst4_vnum(svbool_t, float32_t *, int64_t, svfloat32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_f16)))
+void svst4_vnum(svbool_t, float16_t *, int64_t, svfloat16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_s32)))
+void svst4_vnum(svbool_t, int32_t *, int64_t, svint32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_s64)))
+void svst4_vnum(svbool_t, int64_t *, int64_t, svint64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_s16)))
+void svst4_vnum(svbool_t, int16_t *, int64_t, svint16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u8)))
+void svstnt1(svbool_t, uint8_t *, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u32)))
+void svstnt1(svbool_t, uint32_t *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u64)))
+void svstnt1(svbool_t, uint64_t *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u16)))
+void svstnt1(svbool_t, uint16_t *, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s8)))
+void svstnt1(svbool_t, int8_t *, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_f64)))
+void svstnt1(svbool_t, float64_t *, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_f32)))
+void svstnt1(svbool_t, float32_t *, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_f16)))
+void svstnt1(svbool_t, float16_t *, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s32)))
+void svstnt1(svbool_t, int32_t *, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s64)))
+void svstnt1(svbool_t, int64_t *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s16)))
+void svstnt1(svbool_t, int16_t *, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u8)))
+void svstnt1_vnum(svbool_t, uint8_t *, int64_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u32)))
+void svstnt1_vnum(svbool_t, uint32_t *, int64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u64)))
+void svstnt1_vnum(svbool_t, uint64_t *, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u16)))
+void svstnt1_vnum(svbool_t, uint16_t *, int64_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s8)))
+void svstnt1_vnum(svbool_t, int8_t *, int64_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_f64)))
+void svstnt1_vnum(svbool_t, float64_t *, int64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_f32)))
+void svstnt1_vnum(svbool_t, float32_t *, int64_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_f16)))
+void svstnt1_vnum(svbool_t, float16_t *, int64_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s32)))
+void svstnt1_vnum(svbool_t, int32_t *, int64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s64)))
+void svstnt1_vnum(svbool_t, int64_t *, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s16)))
+void svstnt1_vnum(svbool_t, int16_t *, int64_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_f64_m)))
+svfloat64_t svsub_m(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_f32_m)))
+svfloat32_t svsub_m(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_f16_m)))
+svfloat16_t svsub_m(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_f64_x)))
+svfloat64_t svsub_x(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_f32_x)))
+svfloat32_t svsub_x(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_f16_x)))
+svfloat16_t svsub_x(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_f64_z)))
+svfloat64_t svsub_z(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_f32_z)))
+svfloat32_t svsub_z(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_f16_z)))
+svfloat16_t svsub_z(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_u8_m)))
+svuint8_t svsub_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_u32_m)))
+svuint32_t svsub_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_u64_m)))
+svuint64_t svsub_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_u16_m)))
+svuint16_t svsub_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_s8_m)))
+svint8_t svsub_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_s32_m)))
+svint32_t svsub_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_s64_m)))
+svint64_t svsub_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_s16_m)))
+svint16_t svsub_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_u8_x)))
+svuint8_t svsub_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_u32_x)))
+svuint32_t svsub_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_u64_x)))
+svuint64_t svsub_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_u16_x)))
+svuint16_t svsub_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_s8_x)))
+svint8_t svsub_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_s32_x)))
+svint32_t svsub_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_s64_x)))
+svint64_t svsub_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_s16_x)))
+svint16_t svsub_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_u8_z)))
+svuint8_t svsub_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_u32_z)))
+svuint32_t svsub_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_u64_z)))
+svuint64_t svsub_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_u16_z)))
+svuint16_t svsub_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_s8_z)))
+svint8_t svsub_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_s32_z)))
+svint32_t svsub_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_s64_z)))
+svint64_t svsub_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_s16_z)))
+svint16_t svsub_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_f64_m)))
+svfloat64_t svsub_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_f32_m)))
+svfloat32_t svsub_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_f16_m)))
+svfloat16_t svsub_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_f64_x)))
+svfloat64_t svsub_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_f32_x)))
+svfloat32_t svsub_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_f16_x)))
+svfloat16_t svsub_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_f64_z)))
+svfloat64_t svsub_z(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_f32_z)))
+svfloat32_t svsub_z(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_f16_z)))
+svfloat16_t svsub_z(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_u8_m)))
+svuint8_t svsub_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_u32_m)))
+svuint32_t svsub_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_u64_m)))
+svuint64_t svsub_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_u16_m)))
+svuint16_t svsub_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_s8_m)))
+svint8_t svsub_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_s32_m)))
+svint32_t svsub_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_s64_m)))
+svint64_t svsub_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_s16_m)))
+svint16_t svsub_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_u8_x)))
+svuint8_t svsub_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_u32_x)))
+svuint32_t svsub_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_u64_x)))
+svuint64_t svsub_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_u16_x)))
+svuint16_t svsub_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_s8_x)))
+svint8_t svsub_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_s32_x)))
+svint32_t svsub_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_s64_x)))
+svint64_t svsub_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_s16_x)))
+svint16_t svsub_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_u8_z)))
+svuint8_t svsub_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_u32_z)))
+svuint32_t svsub_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_u64_z)))
+svuint64_t svsub_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_u16_z)))
+svuint16_t svsub_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_s8_z)))
+svint8_t svsub_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_s32_z)))
+svint32_t svsub_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_s64_z)))
+svint64_t svsub_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_s16_z)))
+svint16_t svsub_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_f64_m)))
+svfloat64_t svsubr_m(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_f32_m)))
+svfloat32_t svsubr_m(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_f16_m)))
+svfloat16_t svsubr_m(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_f64_x)))
+svfloat64_t svsubr_x(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_f32_x)))
+svfloat32_t svsubr_x(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_f16_x)))
+svfloat16_t svsubr_x(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_f64_z)))
+svfloat64_t svsubr_z(svbool_t, svfloat64_t, float64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_f32_z)))
+svfloat32_t svsubr_z(svbool_t, svfloat32_t, float32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_f16_z)))
+svfloat16_t svsubr_z(svbool_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_u8_m)))
+svuint8_t svsubr_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_u32_m)))
+svuint32_t svsubr_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_u64_m)))
+svuint64_t svsubr_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_u16_m)))
+svuint16_t svsubr_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_s8_m)))
+svint8_t svsubr_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_s32_m)))
+svint32_t svsubr_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_s64_m)))
+svint64_t svsubr_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_s16_m)))
+svint16_t svsubr_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_u8_x)))
+svuint8_t svsubr_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_u32_x)))
+svuint32_t svsubr_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_u64_x)))
+svuint64_t svsubr_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_u16_x)))
+svuint16_t svsubr_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_s8_x)))
+svint8_t svsubr_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_s32_x)))
+svint32_t svsubr_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_s64_x)))
+svint64_t svsubr_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_s16_x)))
+svint16_t svsubr_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_u8_z)))
+svuint8_t svsubr_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_u32_z)))
+svuint32_t svsubr_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_u64_z)))
+svuint64_t svsubr_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_u16_z)))
+svuint16_t svsubr_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_s8_z)))
+svint8_t svsubr_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_s32_z)))
+svint32_t svsubr_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_s64_z)))
+svint64_t svsubr_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_n_s16_z)))
+svint16_t svsubr_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_f64_m)))
+svfloat64_t svsubr_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_f32_m)))
+svfloat32_t svsubr_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_f16_m)))
+svfloat16_t svsubr_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_f64_x)))
+svfloat64_t svsubr_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_f32_x)))
+svfloat32_t svsubr_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_f16_x)))
+svfloat16_t svsubr_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_f64_z)))
+svfloat64_t svsubr_z(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_f32_z)))
+svfloat32_t svsubr_z(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_f16_z)))
+svfloat16_t svsubr_z(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_u8_m)))
+svuint8_t svsubr_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_u32_m)))
+svuint32_t svsubr_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_u64_m)))
+svuint64_t svsubr_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_u16_m)))
+svuint16_t svsubr_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_s8_m)))
+svint8_t svsubr_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_s32_m)))
+svint32_t svsubr_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_s64_m)))
+svint64_t svsubr_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_s16_m)))
+svint16_t svsubr_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_u8_x)))
+svuint8_t svsubr_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_u32_x)))
+svuint32_t svsubr_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_u64_x)))
+svuint64_t svsubr_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_u16_x)))
+svuint16_t svsubr_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_s8_x)))
+svint8_t svsubr_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_s32_x)))
+svint32_t svsubr_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_s64_x)))
+svint64_t svsubr_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_s16_x)))
+svint16_t svsubr_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_u8_z)))
+svuint8_t svsubr_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_u32_z)))
+svuint32_t svsubr_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_u64_z)))
+svuint64_t svsubr_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_u16_z)))
+svuint16_t svsubr_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_s8_z)))
+svint8_t svsubr_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_s32_z)))
+svint32_t svsubr_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_s64_z)))
+svint64_t svsubr_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubr_s16_z)))
+svint16_t svsubr_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_u8)))
+svuint8_t svtbl(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_u32)))
+svuint32_t svtbl(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_u64)))
+svuint64_t svtbl(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_u16)))
+svuint16_t svtbl(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_s8)))
+svint8_t svtbl(svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_f64)))
+svfloat64_t svtbl(svfloat64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_f32)))
+svfloat32_t svtbl(svfloat32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_f16)))
+svfloat16_t svtbl(svfloat16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_s32)))
+svint32_t svtbl(svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_s64)))
+svint64_t svtbl(svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_s16)))
+svint16_t svtbl(svint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtmad_f64)))
+svfloat64_t svtmad(svfloat64_t, svfloat64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtmad_f32)))
+svfloat32_t svtmad(svfloat32_t, svfloat32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtmad_f16)))
+svfloat16_t svtmad(svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_u8)))
+svuint8_t svtrn1(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_u32)))
+svuint32_t svtrn1(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_u64)))
+svuint64_t svtrn1(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_u16)))
+svuint16_t svtrn1(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_s8)))
+svint8_t svtrn1(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_f64)))
+svfloat64_t svtrn1(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_f32)))
+svfloat32_t svtrn1(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_f16)))
+svfloat16_t svtrn1(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_s32)))
+svint32_t svtrn1(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_s64)))
+svint64_t svtrn1(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_s16)))
+svint16_t svtrn1(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_u8)))
+svuint8_t svtrn2(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_u32)))
+svuint32_t svtrn2(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_u64)))
+svuint64_t svtrn2(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_u16)))
+svuint16_t svtrn2(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_s8)))
+svint8_t svtrn2(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_f64)))
+svfloat64_t svtrn2(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_f32)))
+svfloat32_t svtrn2(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_f16)))
+svfloat16_t svtrn2(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_s32)))
+svint32_t svtrn2(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_s64)))
+svint64_t svtrn2(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_s16)))
+svint16_t svtrn2(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtsmul_f64)))
+svfloat64_t svtsmul(svfloat64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtsmul_f32)))
+svfloat32_t svtsmul(svfloat32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtsmul_f16)))
+svfloat16_t svtsmul(svfloat16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtssel_f64)))
+svfloat64_t svtssel(svfloat64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtssel_f32)))
+svfloat32_t svtssel(svfloat32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtssel_f16)))
+svfloat16_t svtssel(svfloat16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpkhi_b)))
+svbool_t svunpkhi(svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpkhi_s32)))
+svint32_t svunpkhi(svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpkhi_s64)))
+svint64_t svunpkhi(svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpkhi_s16)))
+svint16_t svunpkhi(svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpkhi_u32)))
+svuint32_t svunpkhi(svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpkhi_u64)))
+svuint64_t svunpkhi(svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpkhi_u16)))
+svuint16_t svunpkhi(svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpklo_b)))
+svbool_t svunpklo(svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpklo_s32)))
+svint32_t svunpklo(svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpklo_s64)))
+svint64_t svunpklo(svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpklo_s16)))
+svint16_t svunpklo(svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpklo_u32)))
+svuint32_t svunpklo(svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpklo_u64)))
+svuint64_t svunpklo(svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpklo_u16)))
+svuint16_t svunpklo(svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_u8)))
+svuint8_t svuzp1(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_u32)))
+svuint32_t svuzp1(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_u64)))
+svuint64_t svuzp1(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_u16)))
+svuint16_t svuzp1(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_s8)))
+svint8_t svuzp1(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_f64)))
+svfloat64_t svuzp1(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_f32)))
+svfloat32_t svuzp1(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_f16)))
+svfloat16_t svuzp1(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_s32)))
+svint32_t svuzp1(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_s64)))
+svint64_t svuzp1(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_s16)))
+svint16_t svuzp1(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_u8)))
+svuint8_t svuzp2(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_u32)))
+svuint32_t svuzp2(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_u64)))
+svuint64_t svuzp2(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_u16)))
+svuint16_t svuzp2(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_s8)))
+svint8_t svuzp2(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_f64)))
+svfloat64_t svuzp2(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_f32)))
+svfloat32_t svuzp2(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_f16)))
+svfloat16_t svuzp2(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_s32)))
+svint32_t svuzp2(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_s64)))
+svint64_t svuzp2(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_s16)))
+svint16_t svuzp2(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b8_s32)))
+svbool_t svwhilele_b8(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b32_s32)))
+svbool_t svwhilele_b32(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b64_s32)))
+svbool_t svwhilele_b64(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b16_s32)))
+svbool_t svwhilele_b16(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b8_s64)))
+svbool_t svwhilele_b8(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b32_s64)))
+svbool_t svwhilele_b32(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b64_s64)))
+svbool_t svwhilele_b64(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b16_s64)))
+svbool_t svwhilele_b16(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b8_u32)))
+svbool_t svwhilele_b8(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b32_u32)))
+svbool_t svwhilele_b32(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b64_u32)))
+svbool_t svwhilele_b64(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b16_u32)))
+svbool_t svwhilele_b16(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b8_u64)))
+svbool_t svwhilele_b8(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b32_u64)))
+svbool_t svwhilele_b32(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b64_u64)))
+svbool_t svwhilele_b64(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b16_u64)))
+svbool_t svwhilele_b16(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b8_u32)))
+svbool_t svwhilelt_b8(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b32_u32)))
+svbool_t svwhilelt_b32(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b64_u32)))
+svbool_t svwhilelt_b64(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b16_u32)))
+svbool_t svwhilelt_b16(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b8_u64)))
+svbool_t svwhilelt_b8(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b32_u64)))
+svbool_t svwhilelt_b32(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b64_u64)))
+svbool_t svwhilelt_b64(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b16_u64)))
+svbool_t svwhilelt_b16(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b8_s32)))
+svbool_t svwhilelt_b8(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b32_s32)))
+svbool_t svwhilelt_b32(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b64_s32)))
+svbool_t svwhilelt_b64(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b16_s32)))
+svbool_t svwhilelt_b16(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b8_s64)))
+svbool_t svwhilelt_b8(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b32_s64)))
+svbool_t svwhilelt_b32(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b64_s64)))
+svbool_t svwhilelt_b64(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b16_s64)))
+svbool_t svwhilelt_b16(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_u8)))
+svuint8_t svzip1(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_u32)))
+svuint32_t svzip1(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_u64)))
+svuint64_t svzip1(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_u16)))
+svuint16_t svzip1(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_s8)))
+svint8_t svzip1(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_f64)))
+svfloat64_t svzip1(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_f32)))
+svfloat32_t svzip1(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_f16)))
+svfloat16_t svzip1(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_s32)))
+svint32_t svzip1(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_s64)))
+svint64_t svzip1(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_s16)))
+svint16_t svzip1(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_u8)))
+svuint8_t svzip2(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_u32)))
+svuint32_t svzip2(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_u64)))
+svuint64_t svzip2(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_u16)))
+svuint16_t svzip2(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_s8)))
+svint8_t svzip2(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_f64)))
+svfloat64_t svzip2(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_f32)))
+svfloat32_t svzip2(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_f16)))
+svfloat16_t svzip2(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_s32)))
+svint32_t svzip2(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_s64)))
+svint64_t svzip2(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_s16)))
+svint16_t svzip2(svint16_t, svint16_t);
+
+#if __ARM_FEATURE_SVE2_BITPERM
+#define svbdep_n_u8(...) __builtin_sve_svbdep_n_u8(__VA_ARGS__)
+#define svbdep_n_u32(...) __builtin_sve_svbdep_n_u32(__VA_ARGS__)
+#define svbdep_n_u64(...) __builtin_sve_svbdep_n_u64(__VA_ARGS__)
+#define svbdep_n_u16(...) __builtin_sve_svbdep_n_u16(__VA_ARGS__)
+#define svbdep_u8(...) __builtin_sve_svbdep_u8(__VA_ARGS__)
+#define svbdep_u32(...) __builtin_sve_svbdep_u32(__VA_ARGS__)
+#define svbdep_u64(...) __builtin_sve_svbdep_u64(__VA_ARGS__)
+#define svbdep_u16(...) __builtin_sve_svbdep_u16(__VA_ARGS__)
+#define svbext_n_u8(...) __builtin_sve_svbext_n_u8(__VA_ARGS__)
+#define svbext_n_u32(...) __builtin_sve_svbext_n_u32(__VA_ARGS__)
+#define svbext_n_u64(...) __builtin_sve_svbext_n_u64(__VA_ARGS__)
+#define svbext_n_u16(...) __builtin_sve_svbext_n_u16(__VA_ARGS__)
+#define svbext_u8(...) __builtin_sve_svbext_u8(__VA_ARGS__)
+#define svbext_u32(...) __builtin_sve_svbext_u32(__VA_ARGS__)
+#define svbext_u64(...) __builtin_sve_svbext_u64(__VA_ARGS__)
+#define svbext_u16(...) __builtin_sve_svbext_u16(__VA_ARGS__)
+#define svbgrp_n_u8(...) __builtin_sve_svbgrp_n_u8(__VA_ARGS__)
+#define svbgrp_n_u32(...) __builtin_sve_svbgrp_n_u32(__VA_ARGS__)
+#define svbgrp_n_u64(...) __builtin_sve_svbgrp_n_u64(__VA_ARGS__)
+#define svbgrp_n_u16(...) __builtin_sve_svbgrp_n_u16(__VA_ARGS__)
+#define svbgrp_u8(...) __builtin_sve_svbgrp_u8(__VA_ARGS__)
+#define svbgrp_u32(...) __builtin_sve_svbgrp_u32(__VA_ARGS__)
+#define svbgrp_u64(...) __builtin_sve_svbgrp_u64(__VA_ARGS__)
+#define svbgrp_u16(...) __builtin_sve_svbgrp_u16(__VA_ARGS__)
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u8)))
+svuint8_t svbdep(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u32)))
+svuint32_t svbdep(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u64)))
+svuint64_t svbdep(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u16)))
+svuint16_t svbdep(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u8)))
+svuint8_t svbdep(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u32)))
+svuint32_t svbdep(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u64)))
+svuint64_t svbdep(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u16)))
+svuint16_t svbdep(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u8)))
+svuint8_t svbext(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u32)))
+svuint32_t svbext(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u64)))
+svuint64_t svbext(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u16)))
+svuint16_t svbext(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u8)))
+svuint8_t svbext(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u32)))
+svuint32_t svbext(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u64)))
+svuint64_t svbext(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u16)))
+svuint16_t svbext(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u8)))
+svuint8_t svbgrp(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u32)))
+svuint32_t svbgrp(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u64)))
+svuint64_t svbgrp(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u16)))
+svuint16_t svbgrp(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u8)))
+svuint8_t svbgrp(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u32)))
+svuint32_t svbgrp(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u64)))
+svuint64_t svbgrp(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u16)))
+svuint16_t svbgrp(svuint16_t, svuint16_t);
+#endif //__ARM_FEATURE_SVE2_BITPERM
+
+#if defined(__ARM_FEATURE_SVE2)
+#define svaba_n_s8(...) __builtin_sve_svaba_n_s8(__VA_ARGS__)
+#define svaba_n_s32(...) __builtin_sve_svaba_n_s32(__VA_ARGS__)
+#define svaba_n_s64(...) __builtin_sve_svaba_n_s64(__VA_ARGS__)
+#define svaba_n_s16(...) __builtin_sve_svaba_n_s16(__VA_ARGS__)
+#define svaba_n_u8(...) __builtin_sve_svaba_n_u8(__VA_ARGS__)
+#define svaba_n_u32(...) __builtin_sve_svaba_n_u32(__VA_ARGS__)
+#define svaba_n_u64(...) __builtin_sve_svaba_n_u64(__VA_ARGS__)
+#define svaba_n_u16(...) __builtin_sve_svaba_n_u16(__VA_ARGS__)
+#define svaba_s8(...) __builtin_sve_svaba_s8(__VA_ARGS__)
+#define svaba_s32(...) __builtin_sve_svaba_s32(__VA_ARGS__)
+#define svaba_s64(...) __builtin_sve_svaba_s64(__VA_ARGS__)
+#define svaba_s16(...) __builtin_sve_svaba_s16(__VA_ARGS__)
+#define svaba_u8(...) __builtin_sve_svaba_u8(__VA_ARGS__)
+#define svaba_u32(...) __builtin_sve_svaba_u32(__VA_ARGS__)
+#define svaba_u64(...) __builtin_sve_svaba_u64(__VA_ARGS__)
+#define svaba_u16(...) __builtin_sve_svaba_u16(__VA_ARGS__)
+#define svabalb_n_s32(...) __builtin_sve_svabalb_n_s32(__VA_ARGS__)
+#define svabalb_n_s64(...) __builtin_sve_svabalb_n_s64(__VA_ARGS__)
+#define svabalb_n_s16(...) __builtin_sve_svabalb_n_s16(__VA_ARGS__)
+#define svabalb_n_u32(...) __builtin_sve_svabalb_n_u32(__VA_ARGS__)
+#define svabalb_n_u64(...) __builtin_sve_svabalb_n_u64(__VA_ARGS__)
+#define svabalb_n_u16(...) __builtin_sve_svabalb_n_u16(__VA_ARGS__)
+#define svabalb_s32(...) __builtin_sve_svabalb_s32(__VA_ARGS__)
+#define svabalb_s64(...) __builtin_sve_svabalb_s64(__VA_ARGS__)
+#define svabalb_s16(...) __builtin_sve_svabalb_s16(__VA_ARGS__)
+#define svabalb_u32(...) __builtin_sve_svabalb_u32(__VA_ARGS__)
+#define svabalb_u64(...) __builtin_sve_svabalb_u64(__VA_ARGS__)
+#define svabalb_u16(...) __builtin_sve_svabalb_u16(__VA_ARGS__)
+#define svabalt_n_s32(...) __builtin_sve_svabalt_n_s32(__VA_ARGS__)
+#define svabalt_n_s64(...) __builtin_sve_svabalt_n_s64(__VA_ARGS__)
+#define svabalt_n_s16(...) __builtin_sve_svabalt_n_s16(__VA_ARGS__)
+#define svabalt_n_u32(...) __builtin_sve_svabalt_n_u32(__VA_ARGS__)
+#define svabalt_n_u64(...) __builtin_sve_svabalt_n_u64(__VA_ARGS__)
+#define svabalt_n_u16(...) __builtin_sve_svabalt_n_u16(__VA_ARGS__)
+#define svabalt_s32(...) __builtin_sve_svabalt_s32(__VA_ARGS__)
+#define svabalt_s64(...) __builtin_sve_svabalt_s64(__VA_ARGS__)
+#define svabalt_s16(...) __builtin_sve_svabalt_s16(__VA_ARGS__)
+#define svabalt_u32(...) __builtin_sve_svabalt_u32(__VA_ARGS__)
+#define svabalt_u64(...) __builtin_sve_svabalt_u64(__VA_ARGS__)
+#define svabalt_u16(...) __builtin_sve_svabalt_u16(__VA_ARGS__)
+#define svabdlb_n_s32(...) __builtin_sve_svabdlb_n_s32(__VA_ARGS__)
+#define svabdlb_n_s64(...) __builtin_sve_svabdlb_n_s64(__VA_ARGS__)
+#define svabdlb_n_s16(...) __builtin_sve_svabdlb_n_s16(__VA_ARGS__)
+#define svabdlb_n_u32(...) __builtin_sve_svabdlb_n_u32(__VA_ARGS__)
+#define svabdlb_n_u64(...) __builtin_sve_svabdlb_n_u64(__VA_ARGS__)
+#define svabdlb_n_u16(...) __builtin_sve_svabdlb_n_u16(__VA_ARGS__)
+#define svabdlb_s32(...) __builtin_sve_svabdlb_s32(__VA_ARGS__)
+#define svabdlb_s64(...) __builtin_sve_svabdlb_s64(__VA_ARGS__)
+#define svabdlb_s16(...) __builtin_sve_svabdlb_s16(__VA_ARGS__)
+#define svabdlb_u32(...) __builtin_sve_svabdlb_u32(__VA_ARGS__)
+#define svabdlb_u64(...) __builtin_sve_svabdlb_u64(__VA_ARGS__)
+#define svabdlb_u16(...) __builtin_sve_svabdlb_u16(__VA_ARGS__)
+#define svabdlt_n_s32(...) __builtin_sve_svabdlt_n_s32(__VA_ARGS__)
+#define svabdlt_n_s64(...) __builtin_sve_svabdlt_n_s64(__VA_ARGS__)
+#define svabdlt_n_s16(...) __builtin_sve_svabdlt_n_s16(__VA_ARGS__)
+#define svabdlt_n_u32(...) __builtin_sve_svabdlt_n_u32(__VA_ARGS__)
+#define svabdlt_n_u64(...) __builtin_sve_svabdlt_n_u64(__VA_ARGS__)
+#define svabdlt_n_u16(...) __builtin_sve_svabdlt_n_u16(__VA_ARGS__)
+#define svabdlt_s32(...) __builtin_sve_svabdlt_s32(__VA_ARGS__)
+#define svabdlt_s64(...) __builtin_sve_svabdlt_s64(__VA_ARGS__)
+#define svabdlt_s16(...) __builtin_sve_svabdlt_s16(__VA_ARGS__)
+#define svabdlt_u32(...) __builtin_sve_svabdlt_u32(__VA_ARGS__)
+#define svabdlt_u64(...) __builtin_sve_svabdlt_u64(__VA_ARGS__)
+#define svabdlt_u16(...) __builtin_sve_svabdlt_u16(__VA_ARGS__)
+#define svadalp_s32_m(...) __builtin_sve_svadalp_s32_m(__VA_ARGS__)
+#define svadalp_s64_m(...) __builtin_sve_svadalp_s64_m(__VA_ARGS__)
+#define svadalp_s16_m(...) __builtin_sve_svadalp_s16_m(__VA_ARGS__)
+#define svadalp_s32_x(...) __builtin_sve_svadalp_s32_x(__VA_ARGS__)
+#define svadalp_s64_x(...) __builtin_sve_svadalp_s64_x(__VA_ARGS__)
+#define svadalp_s16_x(...) __builtin_sve_svadalp_s16_x(__VA_ARGS__)
+#define svadalp_s32_z(...) __builtin_sve_svadalp_s32_z(__VA_ARGS__)
+#define svadalp_s64_z(...) __builtin_sve_svadalp_s64_z(__VA_ARGS__)
+#define svadalp_s16_z(...) __builtin_sve_svadalp_s16_z(__VA_ARGS__)
+#define svadalp_u32_m(...) __builtin_sve_svadalp_u32_m(__VA_ARGS__)
+#define svadalp_u64_m(...) __builtin_sve_svadalp_u64_m(__VA_ARGS__)
+#define svadalp_u16_m(...) __builtin_sve_svadalp_u16_m(__VA_ARGS__)
+#define svadalp_u32_x(...) __builtin_sve_svadalp_u32_x(__VA_ARGS__)
+#define svadalp_u64_x(...) __builtin_sve_svadalp_u64_x(__VA_ARGS__)
+#define svadalp_u16_x(...) __builtin_sve_svadalp_u16_x(__VA_ARGS__)
+#define svadalp_u32_z(...) __builtin_sve_svadalp_u32_z(__VA_ARGS__)
+#define svadalp_u64_z(...) __builtin_sve_svadalp_u64_z(__VA_ARGS__)
+#define svadalp_u16_z(...) __builtin_sve_svadalp_u16_z(__VA_ARGS__)
+#define svadclb_n_u32(...) __builtin_sve_svadclb_n_u32(__VA_ARGS__)
+#define svadclb_n_u64(...) __builtin_sve_svadclb_n_u64(__VA_ARGS__)
+#define svadclb_u32(...) __builtin_sve_svadclb_u32(__VA_ARGS__)
+#define svadclb_u64(...) __builtin_sve_svadclb_u64(__VA_ARGS__)
+#define svadclt_n_u32(...) __builtin_sve_svadclt_n_u32(__VA_ARGS__)
+#define svadclt_n_u64(...) __builtin_sve_svadclt_n_u64(__VA_ARGS__)
+#define svadclt_u32(...) __builtin_sve_svadclt_u32(__VA_ARGS__)
+#define svadclt_u64(...) __builtin_sve_svadclt_u64(__VA_ARGS__)
+#define svaddhnb_n_u32(...) __builtin_sve_svaddhnb_n_u32(__VA_ARGS__)
+#define svaddhnb_n_u64(...) __builtin_sve_svaddhnb_n_u64(__VA_ARGS__)
+#define svaddhnb_n_u16(...) __builtin_sve_svaddhnb_n_u16(__VA_ARGS__)
+#define svaddhnb_n_s32(...) __builtin_sve_svaddhnb_n_s32(__VA_ARGS__)
+#define svaddhnb_n_s64(...) __builtin_sve_svaddhnb_n_s64(__VA_ARGS__)
+#define svaddhnb_n_s16(...) __builtin_sve_svaddhnb_n_s16(__VA_ARGS__)
+#define svaddhnb_u32(...) __builtin_sve_svaddhnb_u32(__VA_ARGS__)
+#define svaddhnb_u64(...) __builtin_sve_svaddhnb_u64(__VA_ARGS__)
+#define svaddhnb_u16(...) __builtin_sve_svaddhnb_u16(__VA_ARGS__)
+#define svaddhnb_s32(...) __builtin_sve_svaddhnb_s32(__VA_ARGS__)
+#define svaddhnb_s64(...) __builtin_sve_svaddhnb_s64(__VA_ARGS__)
+#define svaddhnb_s16(...) __builtin_sve_svaddhnb_s16(__VA_ARGS__)
+#define svaddhnt_n_u32(...) __builtin_sve_svaddhnt_n_u32(__VA_ARGS__)
+#define svaddhnt_n_u64(...) __builtin_sve_svaddhnt_n_u64(__VA_ARGS__)
+#define svaddhnt_n_u16(...) __builtin_sve_svaddhnt_n_u16(__VA_ARGS__)
+#define svaddhnt_n_s32(...) __builtin_sve_svaddhnt_n_s32(__VA_ARGS__)
+#define svaddhnt_n_s64(...) __builtin_sve_svaddhnt_n_s64(__VA_ARGS__)
+#define svaddhnt_n_s16(...) __builtin_sve_svaddhnt_n_s16(__VA_ARGS__)
+#define svaddhnt_u32(...) __builtin_sve_svaddhnt_u32(__VA_ARGS__)
+#define svaddhnt_u64(...) __builtin_sve_svaddhnt_u64(__VA_ARGS__)
+#define svaddhnt_u16(...) __builtin_sve_svaddhnt_u16(__VA_ARGS__)
+#define svaddhnt_s32(...) __builtin_sve_svaddhnt_s32(__VA_ARGS__)
+#define svaddhnt_s64(...) __builtin_sve_svaddhnt_s64(__VA_ARGS__)
+#define svaddhnt_s16(...) __builtin_sve_svaddhnt_s16(__VA_ARGS__)
+#define svaddlb_n_s32(...) __builtin_sve_svaddlb_n_s32(__VA_ARGS__)
+#define svaddlb_n_s64(...) __builtin_sve_svaddlb_n_s64(__VA_ARGS__)
+#define svaddlb_n_s16(...) __builtin_sve_svaddlb_n_s16(__VA_ARGS__)
+#define svaddlb_n_u32(...) __builtin_sve_svaddlb_n_u32(__VA_ARGS__)
+#define svaddlb_n_u64(...) __builtin_sve_svaddlb_n_u64(__VA_ARGS__)
+#define svaddlb_n_u16(...) __builtin_sve_svaddlb_n_u16(__VA_ARGS__)
+#define svaddlb_s32(...) __builtin_sve_svaddlb_s32(__VA_ARGS__)
+#define svaddlb_s64(...) __builtin_sve_svaddlb_s64(__VA_ARGS__)
+#define svaddlb_s16(...) __builtin_sve_svaddlb_s16(__VA_ARGS__)
+#define svaddlb_u32(...) __builtin_sve_svaddlb_u32(__VA_ARGS__)
+#define svaddlb_u64(...) __builtin_sve_svaddlb_u64(__VA_ARGS__)
+#define svaddlb_u16(...) __builtin_sve_svaddlb_u16(__VA_ARGS__)
+#define svaddlbt_n_s32(...) __builtin_sve_svaddlbt_n_s32(__VA_ARGS__)
+#define svaddlbt_n_s64(...) __builtin_sve_svaddlbt_n_s64(__VA_ARGS__)
+#define svaddlbt_n_s16(...) __builtin_sve_svaddlbt_n_s16(__VA_ARGS__)
+#define svaddlbt_s32(...) __builtin_sve_svaddlbt_s32(__VA_ARGS__)
+#define svaddlbt_s64(...) __builtin_sve_svaddlbt_s64(__VA_ARGS__)
+#define svaddlbt_s16(...) __builtin_sve_svaddlbt_s16(__VA_ARGS__)
+#define svaddlt_n_s32(...) __builtin_sve_svaddlt_n_s32(__VA_ARGS__)
+#define svaddlt_n_s64(...) __builtin_sve_svaddlt_n_s64(__VA_ARGS__)
+#define svaddlt_n_s16(...) __builtin_sve_svaddlt_n_s16(__VA_ARGS__)
+#define svaddlt_n_u32(...) __builtin_sve_svaddlt_n_u32(__VA_ARGS__)
+#define svaddlt_n_u64(...) __builtin_sve_svaddlt_n_u64(__VA_ARGS__)
+#define svaddlt_n_u16(...) __builtin_sve_svaddlt_n_u16(__VA_ARGS__)
+#define svaddlt_s32(...) __builtin_sve_svaddlt_s32(__VA_ARGS__)
+#define svaddlt_s64(...) __builtin_sve_svaddlt_s64(__VA_ARGS__)
+#define svaddlt_s16(...) __builtin_sve_svaddlt_s16(__VA_ARGS__)
+#define svaddlt_u32(...) __builtin_sve_svaddlt_u32(__VA_ARGS__)
+#define svaddlt_u64(...) __builtin_sve_svaddlt_u64(__VA_ARGS__)
+#define svaddlt_u16(...) __builtin_sve_svaddlt_u16(__VA_ARGS__)
+#define svaddp_f64_m(...) __builtin_sve_svaddp_f64_m(__VA_ARGS__)
+#define svaddp_f32_m(...) __builtin_sve_svaddp_f32_m(__VA_ARGS__)
+#define svaddp_f16_m(...) __builtin_sve_svaddp_f16_m(__VA_ARGS__)
+#define svaddp_f64_x(...) __builtin_sve_svaddp_f64_x(__VA_ARGS__)
+#define svaddp_f32_x(...) __builtin_sve_svaddp_f32_x(__VA_ARGS__)
+#define svaddp_f16_x(...) __builtin_sve_svaddp_f16_x(__VA_ARGS__)
+#define svaddp_u8_m(...) __builtin_sve_svaddp_u8_m(__VA_ARGS__)
+#define svaddp_u32_m(...) __builtin_sve_svaddp_u32_m(__VA_ARGS__)
+#define svaddp_u64_m(...) __builtin_sve_svaddp_u64_m(__VA_ARGS__)
+#define svaddp_u16_m(...) __builtin_sve_svaddp_u16_m(__VA_ARGS__)
+#define svaddp_s8_m(...) __builtin_sve_svaddp_s8_m(__VA_ARGS__)
+#define svaddp_s32_m(...) __builtin_sve_svaddp_s32_m(__VA_ARGS__)
+#define svaddp_s64_m(...) __builtin_sve_svaddp_s64_m(__VA_ARGS__)
+#define svaddp_s16_m(...) __builtin_sve_svaddp_s16_m(__VA_ARGS__)
+#define svaddp_u8_x(...) __builtin_sve_svaddp_u8_x(__VA_ARGS__)
+#define svaddp_u32_x(...) __builtin_sve_svaddp_u32_x(__VA_ARGS__)
+#define svaddp_u64_x(...) __builtin_sve_svaddp_u64_x(__VA_ARGS__)
+#define svaddp_u16_x(...) __builtin_sve_svaddp_u16_x(__VA_ARGS__)
+#define svaddp_s8_x(...) __builtin_sve_svaddp_s8_x(__VA_ARGS__)
+#define svaddp_s32_x(...) __builtin_sve_svaddp_s32_x(__VA_ARGS__)
+#define svaddp_s64_x(...) __builtin_sve_svaddp_s64_x(__VA_ARGS__)
+#define svaddp_s16_x(...) __builtin_sve_svaddp_s16_x(__VA_ARGS__)
+#define svaddwb_n_s32(...) __builtin_sve_svaddwb_n_s32(__VA_ARGS__)
+#define svaddwb_n_s64(...) __builtin_sve_svaddwb_n_s64(__VA_ARGS__)
+#define svaddwb_n_s16(...) __builtin_sve_svaddwb_n_s16(__VA_ARGS__)
+#define svaddwb_n_u32(...) __builtin_sve_svaddwb_n_u32(__VA_ARGS__)
+#define svaddwb_n_u64(...) __builtin_sve_svaddwb_n_u64(__VA_ARGS__)
+#define svaddwb_n_u16(...) __builtin_sve_svaddwb_n_u16(__VA_ARGS__)
+#define svaddwb_s32(...) __builtin_sve_svaddwb_s32(__VA_ARGS__)
+#define svaddwb_s64(...) __builtin_sve_svaddwb_s64(__VA_ARGS__)
+#define svaddwb_s16(...) __builtin_sve_svaddwb_s16(__VA_ARGS__)
+#define svaddwb_u32(...) __builtin_sve_svaddwb_u32(__VA_ARGS__)
+#define svaddwb_u64(...) __builtin_sve_svaddwb_u64(__VA_ARGS__)
+#define svaddwb_u16(...) __builtin_sve_svaddwb_u16(__VA_ARGS__)
+#define svaddwt_n_s32(...) __builtin_sve_svaddwt_n_s32(__VA_ARGS__)
+#define svaddwt_n_s64(...) __builtin_sve_svaddwt_n_s64(__VA_ARGS__)
+#define svaddwt_n_s16(...) __builtin_sve_svaddwt_n_s16(__VA_ARGS__)
+#define svaddwt_n_u32(...) __builtin_sve_svaddwt_n_u32(__VA_ARGS__)
+#define svaddwt_n_u64(...) __builtin_sve_svaddwt_n_u64(__VA_ARGS__)
+#define svaddwt_n_u16(...) __builtin_sve_svaddwt_n_u16(__VA_ARGS__)
+#define svaddwt_s32(...) __builtin_sve_svaddwt_s32(__VA_ARGS__)
+#define svaddwt_s64(...) __builtin_sve_svaddwt_s64(__VA_ARGS__)
+#define svaddwt_s16(...) __builtin_sve_svaddwt_s16(__VA_ARGS__)
+#define svaddwt_u32(...) __builtin_sve_svaddwt_u32(__VA_ARGS__)
+#define svaddwt_u64(...) __builtin_sve_svaddwt_u64(__VA_ARGS__)
+#define svaddwt_u16(...) __builtin_sve_svaddwt_u16(__VA_ARGS__)
+#define svbcax_n_u8(...) __builtin_sve_svbcax_n_u8(__VA_ARGS__)
+#define svbcax_n_u32(...) __builtin_sve_svbcax_n_u32(__VA_ARGS__)
+#define svbcax_n_u64(...) __builtin_sve_svbcax_n_u64(__VA_ARGS__)
+#define svbcax_n_u16(...) __builtin_sve_svbcax_n_u16(__VA_ARGS__)
+#define svbcax_n_s8(...) __builtin_sve_svbcax_n_s8(__VA_ARGS__)
+#define svbcax_n_s32(...) __builtin_sve_svbcax_n_s32(__VA_ARGS__)
+#define svbcax_n_s64(...) __builtin_sve_svbcax_n_s64(__VA_ARGS__)
+#define svbcax_n_s16(...) __builtin_sve_svbcax_n_s16(__VA_ARGS__)
+#define svbcax_u8(...) __builtin_sve_svbcax_u8(__VA_ARGS__)
+#define svbcax_u32(...) __builtin_sve_svbcax_u32(__VA_ARGS__)
+#define svbcax_u64(...) __builtin_sve_svbcax_u64(__VA_ARGS__)
+#define svbcax_u16(...) __builtin_sve_svbcax_u16(__VA_ARGS__)
+#define svbcax_s8(...) __builtin_sve_svbcax_s8(__VA_ARGS__)
+#define svbcax_s32(...) __builtin_sve_svbcax_s32(__VA_ARGS__)
+#define svbcax_s64(...) __builtin_sve_svbcax_s64(__VA_ARGS__)
+#define svbcax_s16(...) __builtin_sve_svbcax_s16(__VA_ARGS__)
+#define svbsl1n_n_u8(...) __builtin_sve_svbsl1n_n_u8(__VA_ARGS__)
+#define svbsl1n_n_u32(...) __builtin_sve_svbsl1n_n_u32(__VA_ARGS__)
+#define svbsl1n_n_u64(...) __builtin_sve_svbsl1n_n_u64(__VA_ARGS__)
+#define svbsl1n_n_u16(...) __builtin_sve_svbsl1n_n_u16(__VA_ARGS__)
+#define svbsl1n_n_s8(...) __builtin_sve_svbsl1n_n_s8(__VA_ARGS__)
+#define svbsl1n_n_s32(...) __builtin_sve_svbsl1n_n_s32(__VA_ARGS__)
+#define svbsl1n_n_s64(...) __builtin_sve_svbsl1n_n_s64(__VA_ARGS__)
+#define svbsl1n_n_s16(...) __builtin_sve_svbsl1n_n_s16(__VA_ARGS__)
+#define svbsl1n_u8(...) __builtin_sve_svbsl1n_u8(__VA_ARGS__)
+#define svbsl1n_u32(...) __builtin_sve_svbsl1n_u32(__VA_ARGS__)
+#define svbsl1n_u64(...) __builtin_sve_svbsl1n_u64(__VA_ARGS__)
+#define svbsl1n_u16(...) __builtin_sve_svbsl1n_u16(__VA_ARGS__)
+#define svbsl1n_s8(...) __builtin_sve_svbsl1n_s8(__VA_ARGS__)
+#define svbsl1n_s32(...) __builtin_sve_svbsl1n_s32(__VA_ARGS__)
+#define svbsl1n_s64(...) __builtin_sve_svbsl1n_s64(__VA_ARGS__)
+#define svbsl1n_s16(...) __builtin_sve_svbsl1n_s16(__VA_ARGS__)
+#define svbsl2n_n_u8(...) __builtin_sve_svbsl2n_n_u8(__VA_ARGS__)
+#define svbsl2n_n_u32(...) __builtin_sve_svbsl2n_n_u32(__VA_ARGS__)
+#define svbsl2n_n_u64(...) __builtin_sve_svbsl2n_n_u64(__VA_ARGS__)
+#define svbsl2n_n_u16(...) __builtin_sve_svbsl2n_n_u16(__VA_ARGS__)
+#define svbsl2n_n_s8(...) __builtin_sve_svbsl2n_n_s8(__VA_ARGS__)
+#define svbsl2n_n_s32(...) __builtin_sve_svbsl2n_n_s32(__VA_ARGS__)
+#define svbsl2n_n_s64(...) __builtin_sve_svbsl2n_n_s64(__VA_ARGS__)
+#define svbsl2n_n_s16(...) __builtin_sve_svbsl2n_n_s16(__VA_ARGS__)
+#define svbsl2n_u8(...) __builtin_sve_svbsl2n_u8(__VA_ARGS__)
+#define svbsl2n_u32(...) __builtin_sve_svbsl2n_u32(__VA_ARGS__)
+#define svbsl2n_u64(...) __builtin_sve_svbsl2n_u64(__VA_ARGS__)
+#define svbsl2n_u16(...) __builtin_sve_svbsl2n_u16(__VA_ARGS__)
+#define svbsl2n_s8(...) __builtin_sve_svbsl2n_s8(__VA_ARGS__)
+#define svbsl2n_s32(...) __builtin_sve_svbsl2n_s32(__VA_ARGS__)
+#define svbsl2n_s64(...) __builtin_sve_svbsl2n_s64(__VA_ARGS__)
+#define svbsl2n_s16(...) __builtin_sve_svbsl2n_s16(__VA_ARGS__)
+#define svbsl_n_u8(...) __builtin_sve_svbsl_n_u8(__VA_ARGS__)
+#define svbsl_n_u32(...) __builtin_sve_svbsl_n_u32(__VA_ARGS__)
+#define svbsl_n_u64(...) __builtin_sve_svbsl_n_u64(__VA_ARGS__)
+#define svbsl_n_u16(...) __builtin_sve_svbsl_n_u16(__VA_ARGS__)
+#define svbsl_n_s8(...) __builtin_sve_svbsl_n_s8(__VA_ARGS__)
+#define svbsl_n_s32(...) __builtin_sve_svbsl_n_s32(__VA_ARGS__)
+#define svbsl_n_s64(...) __builtin_sve_svbsl_n_s64(__VA_ARGS__)
+#define svbsl_n_s16(...) __builtin_sve_svbsl_n_s16(__VA_ARGS__)
+#define svbsl_u8(...) __builtin_sve_svbsl_u8(__VA_ARGS__)
+#define svbsl_u32(...) __builtin_sve_svbsl_u32(__VA_ARGS__)
+#define svbsl_u64(...) __builtin_sve_svbsl_u64(__VA_ARGS__)
+#define svbsl_u16(...) __builtin_sve_svbsl_u16(__VA_ARGS__)
+#define svbsl_s8(...) __builtin_sve_svbsl_s8(__VA_ARGS__)
+#define svbsl_s32(...) __builtin_sve_svbsl_s32(__VA_ARGS__)
+#define svbsl_s64(...) __builtin_sve_svbsl_s64(__VA_ARGS__)
+#define svbsl_s16(...) __builtin_sve_svbsl_s16(__VA_ARGS__)
+#define svcadd_u8(...) __builtin_sve_svcadd_u8(__VA_ARGS__)
+#define svcadd_u32(...) __builtin_sve_svcadd_u32(__VA_ARGS__)
+#define svcadd_u64(...) __builtin_sve_svcadd_u64(__VA_ARGS__)
+#define svcadd_u16(...) __builtin_sve_svcadd_u16(__VA_ARGS__)
+#define svcadd_s8(...) __builtin_sve_svcadd_s8(__VA_ARGS__)
+#define svcadd_s32(...) __builtin_sve_svcadd_s32(__VA_ARGS__)
+#define svcadd_s64(...) __builtin_sve_svcadd_s64(__VA_ARGS__)
+#define svcadd_s16(...) __builtin_sve_svcadd_s16(__VA_ARGS__)
+#define svcdot_s32(...) __builtin_sve_svcdot_s32(__VA_ARGS__)
+#define svcdot_s64(...) __builtin_sve_svcdot_s64(__VA_ARGS__)
+#define svcdot_lane_s32(...) __builtin_sve_svcdot_lane_s32(__VA_ARGS__)
+#define svcdot_lane_s64(...) __builtin_sve_svcdot_lane_s64(__VA_ARGS__)
+#define svcmla_u8(...) __builtin_sve_svcmla_u8(__VA_ARGS__)
+#define svcmla_u32(...) __builtin_sve_svcmla_u32(__VA_ARGS__)
+#define svcmla_u64(...) __builtin_sve_svcmla_u64(__VA_ARGS__)
+#define svcmla_u16(...) __builtin_sve_svcmla_u16(__VA_ARGS__)
+#define svcmla_s8(...) __builtin_sve_svcmla_s8(__VA_ARGS__)
+#define svcmla_s32(...) __builtin_sve_svcmla_s32(__VA_ARGS__)
+#define svcmla_s64(...) __builtin_sve_svcmla_s64(__VA_ARGS__)
+#define svcmla_s16(...) __builtin_sve_svcmla_s16(__VA_ARGS__)
+#define svcmla_lane_u32(...) __builtin_sve_svcmla_lane_u32(__VA_ARGS__)
+#define svcmla_lane_u16(...) __builtin_sve_svcmla_lane_u16(__VA_ARGS__)
+#define svcmla_lane_s32(...) __builtin_sve_svcmla_lane_s32(__VA_ARGS__)
+#define svcmla_lane_s16(...) __builtin_sve_svcmla_lane_s16(__VA_ARGS__)
+#define svcvtlt_f32_f16_m(...) __builtin_sve_svcvtlt_f32_f16_m(__VA_ARGS__)
+#define svcvtlt_f32_f16_x(...) __builtin_sve_svcvtlt_f32_f16_x(__VA_ARGS__)
+#define svcvtlt_f64_f32_m(...) __builtin_sve_svcvtlt_f64_f32_m(__VA_ARGS__)
+#define svcvtlt_f64_f32_x(...) __builtin_sve_svcvtlt_f64_f32_x(__VA_ARGS__)
+#define svcvtnt_f16_f32_m(...) __builtin_sve_svcvtnt_f16_f32_m(__VA_ARGS__)
+#define svcvtnt_f32_f64_m(...) __builtin_sve_svcvtnt_f32_f64_m(__VA_ARGS__)
+#define svcvtx_f32_f64_m(...) __builtin_sve_svcvtx_f32_f64_m(__VA_ARGS__)
+#define svcvtx_f32_f64_x(...) __builtin_sve_svcvtx_f32_f64_x(__VA_ARGS__)
+#define svcvtx_f32_f64_z(...) __builtin_sve_svcvtx_f32_f64_z(__VA_ARGS__)
+#define svcvtxnt_f32_f64_m(...) __builtin_sve_svcvtxnt_f32_f64_m(__VA_ARGS__)
+#define sveor3_n_u8(...) __builtin_sve_sveor3_n_u8(__VA_ARGS__)
+#define sveor3_n_u32(...) __builtin_sve_sveor3_n_u32(__VA_ARGS__)
+#define sveor3_n_u64(...) __builtin_sve_sveor3_n_u64(__VA_ARGS__)
+#define sveor3_n_u16(...) __builtin_sve_sveor3_n_u16(__VA_ARGS__)
+#define sveor3_n_s8(...) __builtin_sve_sveor3_n_s8(__VA_ARGS__)
+#define sveor3_n_s32(...) __builtin_sve_sveor3_n_s32(__VA_ARGS__)
+#define sveor3_n_s64(...) __builtin_sve_sveor3_n_s64(__VA_ARGS__)
+#define sveor3_n_s16(...) __builtin_sve_sveor3_n_s16(__VA_ARGS__)
+#define sveor3_u8(...) __builtin_sve_sveor3_u8(__VA_ARGS__)
+#define sveor3_u32(...) __builtin_sve_sveor3_u32(__VA_ARGS__)
+#define sveor3_u64(...) __builtin_sve_sveor3_u64(__VA_ARGS__)
+#define sveor3_u16(...) __builtin_sve_sveor3_u16(__VA_ARGS__)
+#define sveor3_s8(...) __builtin_sve_sveor3_s8(__VA_ARGS__)
+#define sveor3_s32(...) __builtin_sve_sveor3_s32(__VA_ARGS__)
+#define sveor3_s64(...) __builtin_sve_sveor3_s64(__VA_ARGS__)
+#define sveor3_s16(...) __builtin_sve_sveor3_s16(__VA_ARGS__)
+#define sveorbt_n_u8(...) __builtin_sve_sveorbt_n_u8(__VA_ARGS__)
+#define sveorbt_n_u32(...) __builtin_sve_sveorbt_n_u32(__VA_ARGS__)
+#define sveorbt_n_u64(...) __builtin_sve_sveorbt_n_u64(__VA_ARGS__)
+#define sveorbt_n_u16(...) __builtin_sve_sveorbt_n_u16(__VA_ARGS__)
+#define sveorbt_n_s8(...) __builtin_sve_sveorbt_n_s8(__VA_ARGS__)
+#define sveorbt_n_s32(...) __builtin_sve_sveorbt_n_s32(__VA_ARGS__)
+#define sveorbt_n_s64(...) __builtin_sve_sveorbt_n_s64(__VA_ARGS__)
+#define sveorbt_n_s16(...) __builtin_sve_sveorbt_n_s16(__VA_ARGS__)
+#define sveorbt_u8(...) __builtin_sve_sveorbt_u8(__VA_ARGS__)
+#define sveorbt_u32(...) __builtin_sve_sveorbt_u32(__VA_ARGS__)
+#define sveorbt_u64(...) __builtin_sve_sveorbt_u64(__VA_ARGS__)
+#define sveorbt_u16(...) __builtin_sve_sveorbt_u16(__VA_ARGS__)
+#define sveorbt_s8(...) __builtin_sve_sveorbt_s8(__VA_ARGS__)
+#define sveorbt_s32(...) __builtin_sve_sveorbt_s32(__VA_ARGS__)
+#define sveorbt_s64(...) __builtin_sve_sveorbt_s64(__VA_ARGS__)
+#define sveorbt_s16(...) __builtin_sve_sveorbt_s16(__VA_ARGS__)
+#define sveortb_n_u8(...) __builtin_sve_sveortb_n_u8(__VA_ARGS__)
+#define sveortb_n_u32(...) __builtin_sve_sveortb_n_u32(__VA_ARGS__)
+#define sveortb_n_u64(...) __builtin_sve_sveortb_n_u64(__VA_ARGS__)
+#define sveortb_n_u16(...) __builtin_sve_sveortb_n_u16(__VA_ARGS__)
+#define sveortb_n_s8(...) __builtin_sve_sveortb_n_s8(__VA_ARGS__)
+#define sveortb_n_s32(...) __builtin_sve_sveortb_n_s32(__VA_ARGS__)
+#define sveortb_n_s64(...) __builtin_sve_sveortb_n_s64(__VA_ARGS__)
+#define sveortb_n_s16(...) __builtin_sve_sveortb_n_s16(__VA_ARGS__)
+#define sveortb_u8(...) __builtin_sve_sveortb_u8(__VA_ARGS__)
+#define sveortb_u32(...) __builtin_sve_sveortb_u32(__VA_ARGS__)
+#define sveortb_u64(...) __builtin_sve_sveortb_u64(__VA_ARGS__)
+#define sveortb_u16(...) __builtin_sve_sveortb_u16(__VA_ARGS__)
+#define sveortb_s8(...) __builtin_sve_sveortb_s8(__VA_ARGS__)
+#define sveortb_s32(...) __builtin_sve_sveortb_s32(__VA_ARGS__)
+#define sveortb_s64(...) __builtin_sve_sveortb_s64(__VA_ARGS__)
+#define sveortb_s16(...) __builtin_sve_sveortb_s16(__VA_ARGS__)
+#define svhadd_n_s8_m(...) __builtin_sve_svhadd_n_s8_m(__VA_ARGS__)
+#define svhadd_n_s32_m(...) __builtin_sve_svhadd_n_s32_m(__VA_ARGS__)
+#define svhadd_n_s64_m(...) __builtin_sve_svhadd_n_s64_m(__VA_ARGS__)
+#define svhadd_n_s16_m(...) __builtin_sve_svhadd_n_s16_m(__VA_ARGS__)
+#define svhadd_n_s8_x(...) __builtin_sve_svhadd_n_s8_x(__VA_ARGS__)
+#define svhadd_n_s32_x(...) __builtin_sve_svhadd_n_s32_x(__VA_ARGS__)
+#define svhadd_n_s64_x(...) __builtin_sve_svhadd_n_s64_x(__VA_ARGS__)
+#define svhadd_n_s16_x(...) __builtin_sve_svhadd_n_s16_x(__VA_ARGS__)
+#define svhadd_n_s8_z(...) __builtin_sve_svhadd_n_s8_z(__VA_ARGS__)
+#define svhadd_n_s32_z(...) __builtin_sve_svhadd_n_s32_z(__VA_ARGS__)
+#define svhadd_n_s64_z(...) __builtin_sve_svhadd_n_s64_z(__VA_ARGS__)
+#define svhadd_n_s16_z(...) __builtin_sve_svhadd_n_s16_z(__VA_ARGS__)
+#define svhadd_n_u8_m(...) __builtin_sve_svhadd_n_u8_m(__VA_ARGS__)
+#define svhadd_n_u32_m(...) __builtin_sve_svhadd_n_u32_m(__VA_ARGS__)
+#define svhadd_n_u64_m(...) __builtin_sve_svhadd_n_u64_m(__VA_ARGS__)
+#define svhadd_n_u16_m(...) __builtin_sve_svhadd_n_u16_m(__VA_ARGS__)
+#define svhadd_n_u8_x(...) __builtin_sve_svhadd_n_u8_x(__VA_ARGS__)
+#define svhadd_n_u32_x(...) __builtin_sve_svhadd_n_u32_x(__VA_ARGS__)
+#define svhadd_n_u64_x(...) __builtin_sve_svhadd_n_u64_x(__VA_ARGS__)
+#define svhadd_n_u16_x(...) __builtin_sve_svhadd_n_u16_x(__VA_ARGS__)
+#define svhadd_n_u8_z(...) __builtin_sve_svhadd_n_u8_z(__VA_ARGS__)
+#define svhadd_n_u32_z(...) __builtin_sve_svhadd_n_u32_z(__VA_ARGS__)
+#define svhadd_n_u64_z(...) __builtin_sve_svhadd_n_u64_z(__VA_ARGS__)
+#define svhadd_n_u16_z(...) __builtin_sve_svhadd_n_u16_z(__VA_ARGS__)
+#define svhadd_s8_m(...) __builtin_sve_svhadd_s8_m(__VA_ARGS__)
+#define svhadd_s32_m(...) __builtin_sve_svhadd_s32_m(__VA_ARGS__)
+#define svhadd_s64_m(...) __builtin_sve_svhadd_s64_m(__VA_ARGS__)
+#define svhadd_s16_m(...) __builtin_sve_svhadd_s16_m(__VA_ARGS__)
+#define svhadd_s8_x(...) __builtin_sve_svhadd_s8_x(__VA_ARGS__)
+#define svhadd_s32_x(...) __builtin_sve_svhadd_s32_x(__VA_ARGS__)
+#define svhadd_s64_x(...) __builtin_sve_svhadd_s64_x(__VA_ARGS__)
+#define svhadd_s16_x(...) __builtin_sve_svhadd_s16_x(__VA_ARGS__)
+#define svhadd_s8_z(...) __builtin_sve_svhadd_s8_z(__VA_ARGS__)
+#define svhadd_s32_z(...) __builtin_sve_svhadd_s32_z(__VA_ARGS__)
+#define svhadd_s64_z(...) __builtin_sve_svhadd_s64_z(__VA_ARGS__)
+#define svhadd_s16_z(...) __builtin_sve_svhadd_s16_z(__VA_ARGS__)
+#define svhadd_u8_m(...) __builtin_sve_svhadd_u8_m(__VA_ARGS__)
+#define svhadd_u32_m(...) __builtin_sve_svhadd_u32_m(__VA_ARGS__)
+#define svhadd_u64_m(...) __builtin_sve_svhadd_u64_m(__VA_ARGS__)
+#define svhadd_u16_m(...) __builtin_sve_svhadd_u16_m(__VA_ARGS__)
+#define svhadd_u8_x(...) __builtin_sve_svhadd_u8_x(__VA_ARGS__)
+#define svhadd_u32_x(...) __builtin_sve_svhadd_u32_x(__VA_ARGS__)
+#define svhadd_u64_x(...) __builtin_sve_svhadd_u64_x(__VA_ARGS__)
+#define svhadd_u16_x(...) __builtin_sve_svhadd_u16_x(__VA_ARGS__)
+#define svhadd_u8_z(...) __builtin_sve_svhadd_u8_z(__VA_ARGS__)
+#define svhadd_u32_z(...) __builtin_sve_svhadd_u32_z(__VA_ARGS__)
+#define svhadd_u64_z(...) __builtin_sve_svhadd_u64_z(__VA_ARGS__)
+#define svhadd_u16_z(...) __builtin_sve_svhadd_u16_z(__VA_ARGS__)
+#define svhistcnt_u32_z(...) __builtin_sve_svhistcnt_u32_z(__VA_ARGS__)
+#define svhistcnt_u64_z(...) __builtin_sve_svhistcnt_u64_z(__VA_ARGS__)
+#define svhistcnt_s32_z(...) __builtin_sve_svhistcnt_s32_z(__VA_ARGS__)
+#define svhistcnt_s64_z(...) __builtin_sve_svhistcnt_s64_z(__VA_ARGS__)
+#define svhistseg_u8(...) __builtin_sve_svhistseg_u8(__VA_ARGS__)
+#define svhistseg_s8(...) __builtin_sve_svhistseg_s8(__VA_ARGS__)
+#define svhsub_n_s8_m(...) __builtin_sve_svhsub_n_s8_m(__VA_ARGS__)
+#define svhsub_n_s32_m(...) __builtin_sve_svhsub_n_s32_m(__VA_ARGS__)
+#define svhsub_n_s64_m(...) __builtin_sve_svhsub_n_s64_m(__VA_ARGS__)
+#define svhsub_n_s16_m(...) __builtin_sve_svhsub_n_s16_m(__VA_ARGS__)
+#define svhsub_n_s8_x(...) __builtin_sve_svhsub_n_s8_x(__VA_ARGS__)
+#define svhsub_n_s32_x(...) __builtin_sve_svhsub_n_s32_x(__VA_ARGS__)
+#define svhsub_n_s64_x(...) __builtin_sve_svhsub_n_s64_x(__VA_ARGS__)
+#define svhsub_n_s16_x(...) __builtin_sve_svhsub_n_s16_x(__VA_ARGS__)
+#define svhsub_n_s8_z(...) __builtin_sve_svhsub_n_s8_z(__VA_ARGS__)
+#define svhsub_n_s32_z(...) __builtin_sve_svhsub_n_s32_z(__VA_ARGS__)
+#define svhsub_n_s64_z(...) __builtin_sve_svhsub_n_s64_z(__VA_ARGS__)
+#define svhsub_n_s16_z(...) __builtin_sve_svhsub_n_s16_z(__VA_ARGS__)
+#define svhsub_n_u8_m(...) __builtin_sve_svhsub_n_u8_m(__VA_ARGS__)
+#define svhsub_n_u32_m(...) __builtin_sve_svhsub_n_u32_m(__VA_ARGS__)
+#define svhsub_n_u64_m(...) __builtin_sve_svhsub_n_u64_m(__VA_ARGS__)
+#define svhsub_n_u16_m(...) __builtin_sve_svhsub_n_u16_m(__VA_ARGS__)
+#define svhsub_n_u8_x(...) __builtin_sve_svhsub_n_u8_x(__VA_ARGS__)
+#define svhsub_n_u32_x(...) __builtin_sve_svhsub_n_u32_x(__VA_ARGS__)
+#define svhsub_n_u64_x(...) __builtin_sve_svhsub_n_u64_x(__VA_ARGS__)
+#define svhsub_n_u16_x(...) __builtin_sve_svhsub_n_u16_x(__VA_ARGS__)
+#define svhsub_n_u8_z(...) __builtin_sve_svhsub_n_u8_z(__VA_ARGS__)
+#define svhsub_n_u32_z(...) __builtin_sve_svhsub_n_u32_z(__VA_ARGS__)
+#define svhsub_n_u64_z(...) __builtin_sve_svhsub_n_u64_z(__VA_ARGS__)
+#define svhsub_n_u16_z(...) __builtin_sve_svhsub_n_u16_z(__VA_ARGS__)
+#define svhsub_s8_m(...) __builtin_sve_svhsub_s8_m(__VA_ARGS__)
+#define svhsub_s32_m(...) __builtin_sve_svhsub_s32_m(__VA_ARGS__)
+#define svhsub_s64_m(...) __builtin_sve_svhsub_s64_m(__VA_ARGS__)
+#define svhsub_s16_m(...) __builtin_sve_svhsub_s16_m(__VA_ARGS__)
+#define svhsub_s8_x(...) __builtin_sve_svhsub_s8_x(__VA_ARGS__)
+#define svhsub_s32_x(...) __builtin_sve_svhsub_s32_x(__VA_ARGS__)
+#define svhsub_s64_x(...) __builtin_sve_svhsub_s64_x(__VA_ARGS__)
+#define svhsub_s16_x(...) __builtin_sve_svhsub_s16_x(__VA_ARGS__)
+#define svhsub_s8_z(...) __builtin_sve_svhsub_s8_z(__VA_ARGS__)
+#define svhsub_s32_z(...) __builtin_sve_svhsub_s32_z(__VA_ARGS__)
+#define svhsub_s64_z(...) __builtin_sve_svhsub_s64_z(__VA_ARGS__)
+#define svhsub_s16_z(...) __builtin_sve_svhsub_s16_z(__VA_ARGS__)
+#define svhsub_u8_m(...) __builtin_sve_svhsub_u8_m(__VA_ARGS__)
+#define svhsub_u32_m(...) __builtin_sve_svhsub_u32_m(__VA_ARGS__)
+#define svhsub_u64_m(...) __builtin_sve_svhsub_u64_m(__VA_ARGS__)
+#define svhsub_u16_m(...) __builtin_sve_svhsub_u16_m(__VA_ARGS__)
+#define svhsub_u8_x(...) __builtin_sve_svhsub_u8_x(__VA_ARGS__)
+#define svhsub_u32_x(...) __builtin_sve_svhsub_u32_x(__VA_ARGS__)
+#define svhsub_u64_x(...) __builtin_sve_svhsub_u64_x(__VA_ARGS__)
+#define svhsub_u16_x(...) __builtin_sve_svhsub_u16_x(__VA_ARGS__)
+#define svhsub_u8_z(...) __builtin_sve_svhsub_u8_z(__VA_ARGS__)
+#define svhsub_u32_z(...) __builtin_sve_svhsub_u32_z(__VA_ARGS__)
+#define svhsub_u64_z(...) __builtin_sve_svhsub_u64_z(__VA_ARGS__)
+#define svhsub_u16_z(...) __builtin_sve_svhsub_u16_z(__VA_ARGS__)
+#define svhsubr_n_s8_m(...) __builtin_sve_svhsubr_n_s8_m(__VA_ARGS__)
+#define svhsubr_n_s32_m(...) __builtin_sve_svhsubr_n_s32_m(__VA_ARGS__)
+#define svhsubr_n_s64_m(...) __builtin_sve_svhsubr_n_s64_m(__VA_ARGS__)
+#define svhsubr_n_s16_m(...) __builtin_sve_svhsubr_n_s16_m(__VA_ARGS__)
+#define svhsubr_n_s8_x(...) __builtin_sve_svhsubr_n_s8_x(__VA_ARGS__)
+#define svhsubr_n_s32_x(...) __builtin_sve_svhsubr_n_s32_x(__VA_ARGS__)
+#define svhsubr_n_s64_x(...) __builtin_sve_svhsubr_n_s64_x(__VA_ARGS__)
+#define svhsubr_n_s16_x(...) __builtin_sve_svhsubr_n_s16_x(__VA_ARGS__)
+#define svhsubr_n_s8_z(...) __builtin_sve_svhsubr_n_s8_z(__VA_ARGS__)
+#define svhsubr_n_s32_z(...) __builtin_sve_svhsubr_n_s32_z(__VA_ARGS__)
+#define svhsubr_n_s64_z(...) __builtin_sve_svhsubr_n_s64_z(__VA_ARGS__)
+#define svhsubr_n_s16_z(...) __builtin_sve_svhsubr_n_s16_z(__VA_ARGS__)
+#define svhsubr_n_u8_m(...) __builtin_sve_svhsubr_n_u8_m(__VA_ARGS__)
+#define svhsubr_n_u32_m(...) __builtin_sve_svhsubr_n_u32_m(__VA_ARGS__)
+#define svhsubr_n_u64_m(...) __builtin_sve_svhsubr_n_u64_m(__VA_ARGS__)
+#define svhsubr_n_u16_m(...) __builtin_sve_svhsubr_n_u16_m(__VA_ARGS__)
+#define svhsubr_n_u8_x(...) __builtin_sve_svhsubr_n_u8_x(__VA_ARGS__)
+#define svhsubr_n_u32_x(...) __builtin_sve_svhsubr_n_u32_x(__VA_ARGS__)
+#define svhsubr_n_u64_x(...) __builtin_sve_svhsubr_n_u64_x(__VA_ARGS__)
+#define svhsubr_n_u16_x(...) __builtin_sve_svhsubr_n_u16_x(__VA_ARGS__)
+#define svhsubr_n_u8_z(...) __builtin_sve_svhsubr_n_u8_z(__VA_ARGS__)
+#define svhsubr_n_u32_z(...) __builtin_sve_svhsubr_n_u32_z(__VA_ARGS__)
+#define svhsubr_n_u64_z(...) __builtin_sve_svhsubr_n_u64_z(__VA_ARGS__)
+#define svhsubr_n_u16_z(...) __builtin_sve_svhsubr_n_u16_z(__VA_ARGS__)
+#define svhsubr_s8_m(...) __builtin_sve_svhsubr_s8_m(__VA_ARGS__)
+#define svhsubr_s32_m(...) __builtin_sve_svhsubr_s32_m(__VA_ARGS__)
+#define svhsubr_s64_m(...) __builtin_sve_svhsubr_s64_m(__VA_ARGS__)
+#define svhsubr_s16_m(...) __builtin_sve_svhsubr_s16_m(__VA_ARGS__)
+#define svhsubr_s8_x(...) __builtin_sve_svhsubr_s8_x(__VA_ARGS__)
+#define svhsubr_s32_x(...) __builtin_sve_svhsubr_s32_x(__VA_ARGS__)
+#define svhsubr_s64_x(...) __builtin_sve_svhsubr_s64_x(__VA_ARGS__)
+#define svhsubr_s16_x(...) __builtin_sve_svhsubr_s16_x(__VA_ARGS__)
+#define svhsubr_s8_z(...) __builtin_sve_svhsubr_s8_z(__VA_ARGS__)
+#define svhsubr_s32_z(...) __builtin_sve_svhsubr_s32_z(__VA_ARGS__)
+#define svhsubr_s64_z(...) __builtin_sve_svhsubr_s64_z(__VA_ARGS__)
+#define svhsubr_s16_z(...) __builtin_sve_svhsubr_s16_z(__VA_ARGS__)
+#define svhsubr_u8_m(...) __builtin_sve_svhsubr_u8_m(__VA_ARGS__)
+#define svhsubr_u32_m(...) __builtin_sve_svhsubr_u32_m(__VA_ARGS__)
+#define svhsubr_u64_m(...) __builtin_sve_svhsubr_u64_m(__VA_ARGS__)
+#define svhsubr_u16_m(...) __builtin_sve_svhsubr_u16_m(__VA_ARGS__)
+#define svhsubr_u8_x(...) __builtin_sve_svhsubr_u8_x(__VA_ARGS__)
+#define svhsubr_u32_x(...) __builtin_sve_svhsubr_u32_x(__VA_ARGS__)
+#define svhsubr_u64_x(...) __builtin_sve_svhsubr_u64_x(__VA_ARGS__)
+#define svhsubr_u16_x(...) __builtin_sve_svhsubr_u16_x(__VA_ARGS__)
+#define svhsubr_u8_z(...) __builtin_sve_svhsubr_u8_z(__VA_ARGS__)
+#define svhsubr_u32_z(...) __builtin_sve_svhsubr_u32_z(__VA_ARGS__)
+#define svhsubr_u64_z(...) __builtin_sve_svhsubr_u64_z(__VA_ARGS__)
+#define svhsubr_u16_z(...) __builtin_sve_svhsubr_u16_z(__VA_ARGS__)
+#define svldnt1_gather_u32base_index_u32(...) __builtin_sve_svldnt1_gather_u32base_index_u32(__VA_ARGS__)
+#define svldnt1_gather_u64base_index_u64(...) __builtin_sve_svldnt1_gather_u64base_index_u64(__VA_ARGS__)
+#define svldnt1_gather_u64base_index_f64(...) __builtin_sve_svldnt1_gather_u64base_index_f64(__VA_ARGS__)
+#define svldnt1_gather_u32base_index_f32(...) __builtin_sve_svldnt1_gather_u32base_index_f32(__VA_ARGS__)
+#define svldnt1_gather_u32base_index_s32(...) __builtin_sve_svldnt1_gather_u32base_index_s32(__VA_ARGS__)
+#define svldnt1_gather_u64base_index_s64(...) __builtin_sve_svldnt1_gather_u64base_index_s64(__VA_ARGS__)
+#define svldnt1_gather_u32base_offset_u32(...) __builtin_sve_svldnt1_gather_u32base_offset_u32(__VA_ARGS__)
+#define svldnt1_gather_u64base_offset_u64(...) __builtin_sve_svldnt1_gather_u64base_offset_u64(__VA_ARGS__)
+#define svldnt1_gather_u64base_offset_f64(...) __builtin_sve_svldnt1_gather_u64base_offset_f64(__VA_ARGS__)
+#define svldnt1_gather_u32base_offset_f32(...) __builtin_sve_svldnt1_gather_u32base_offset_f32(__VA_ARGS__)
+#define svldnt1_gather_u32base_offset_s32(...) __builtin_sve_svldnt1_gather_u32base_offset_s32(__VA_ARGS__)
+#define svldnt1_gather_u64base_offset_s64(...) __builtin_sve_svldnt1_gather_u64base_offset_s64(__VA_ARGS__)
+#define svldnt1_gather_u32base_u32(...) __builtin_sve_svldnt1_gather_u32base_u32(__VA_ARGS__)
+#define svldnt1_gather_u64base_u64(...) __builtin_sve_svldnt1_gather_u64base_u64(__VA_ARGS__)
+#define svldnt1_gather_u64base_f64(...) __builtin_sve_svldnt1_gather_u64base_f64(__VA_ARGS__)
+#define svldnt1_gather_u32base_f32(...) __builtin_sve_svldnt1_gather_u32base_f32(__VA_ARGS__)
+#define svldnt1_gather_u32base_s32(...) __builtin_sve_svldnt1_gather_u32base_s32(__VA_ARGS__)
+#define svldnt1_gather_u64base_s64(...) __builtin_sve_svldnt1_gather_u64base_s64(__VA_ARGS__)
+#define svldnt1_gather_s64index_u64(...) __builtin_sve_svldnt1_gather_s64index_u64(__VA_ARGS__)
+#define svldnt1_gather_s64index_f64(...) __builtin_sve_svldnt1_gather_s64index_f64(__VA_ARGS__)
+#define svldnt1_gather_s64index_s64(...) __builtin_sve_svldnt1_gather_s64index_s64(__VA_ARGS__)
+#define svldnt1_gather_u64index_u64(...) __builtin_sve_svldnt1_gather_u64index_u64(__VA_ARGS__)
+#define svldnt1_gather_u64index_f64(...) __builtin_sve_svldnt1_gather_u64index_f64(__VA_ARGS__)
+#define svldnt1_gather_u64index_s64(...) __builtin_sve_svldnt1_gather_u64index_s64(__VA_ARGS__)
+#define svldnt1_gather_u32offset_u32(...) __builtin_sve_svldnt1_gather_u32offset_u32(__VA_ARGS__)
+#define svldnt1_gather_u32offset_f32(...) __builtin_sve_svldnt1_gather_u32offset_f32(__VA_ARGS__)
+#define svldnt1_gather_u32offset_s32(...) __builtin_sve_svldnt1_gather_u32offset_s32(__VA_ARGS__)
+#define svldnt1_gather_s64offset_u64(...) __builtin_sve_svldnt1_gather_s64offset_u64(__VA_ARGS__)
+#define svldnt1_gather_s64offset_f64(...) __builtin_sve_svldnt1_gather_s64offset_f64(__VA_ARGS__)
+#define svldnt1_gather_s64offset_s64(...) __builtin_sve_svldnt1_gather_s64offset_s64(__VA_ARGS__)
+#define svldnt1_gather_u64offset_u64(...) __builtin_sve_svldnt1_gather_u64offset_u64(__VA_ARGS__)
+#define svldnt1_gather_u64offset_f64(...) __builtin_sve_svldnt1_gather_u64offset_f64(__VA_ARGS__)
+#define svldnt1_gather_u64offset_s64(...) __builtin_sve_svldnt1_gather_u64offset_s64(__VA_ARGS__)
+#define svldnt1sb_gather_u32base_offset_u32(...) __builtin_sve_svldnt1sb_gather_u32base_offset_u32(__VA_ARGS__)
+#define svldnt1sb_gather_u64base_offset_u64(...) __builtin_sve_svldnt1sb_gather_u64base_offset_u64(__VA_ARGS__)
+#define svldnt1sb_gather_u32base_offset_s32(...) __builtin_sve_svldnt1sb_gather_u32base_offset_s32(__VA_ARGS__)
+#define svldnt1sb_gather_u64base_offset_s64(...) __builtin_sve_svldnt1sb_gather_u64base_offset_s64(__VA_ARGS__)
+#define svldnt1sb_gather_u32base_u32(...) __builtin_sve_svldnt1sb_gather_u32base_u32(__VA_ARGS__)
+#define svldnt1sb_gather_u64base_u64(...) __builtin_sve_svldnt1sb_gather_u64base_u64(__VA_ARGS__)
+#define svldnt1sb_gather_u32base_s32(...) __builtin_sve_svldnt1sb_gather_u32base_s32(__VA_ARGS__)
+#define svldnt1sb_gather_u64base_s64(...) __builtin_sve_svldnt1sb_gather_u64base_s64(__VA_ARGS__)
+#define svldnt1sb_gather_u32offset_u32(...) __builtin_sve_svldnt1sb_gather_u32offset_u32(__VA_ARGS__)
+#define svldnt1sb_gather_u32offset_s32(...) __builtin_sve_svldnt1sb_gather_u32offset_s32(__VA_ARGS__)
+#define svldnt1sb_gather_s64offset_u64(...) __builtin_sve_svldnt1sb_gather_s64offset_u64(__VA_ARGS__)
+#define svldnt1sb_gather_s64offset_s64(...) __builtin_sve_svldnt1sb_gather_s64offset_s64(__VA_ARGS__)
+#define svldnt1sb_gather_u64offset_u64(...) __builtin_sve_svldnt1sb_gather_u64offset_u64(__VA_ARGS__)
+#define svldnt1sb_gather_u64offset_s64(...) __builtin_sve_svldnt1sb_gather_u64offset_s64(__VA_ARGS__)
+#define svldnt1sh_gather_u32base_index_u32(...) __builtin_sve_svldnt1sh_gather_u32base_index_u32(__VA_ARGS__)
+#define svldnt1sh_gather_u64base_index_u64(...) __builtin_sve_svldnt1sh_gather_u64base_index_u64(__VA_ARGS__)
+#define svldnt1sh_gather_u32base_index_s32(...) __builtin_sve_svldnt1sh_gather_u32base_index_s32(__VA_ARGS__)
+#define svldnt1sh_gather_u64base_index_s64(...) __builtin_sve_svldnt1sh_gather_u64base_index_s64(__VA_ARGS__)
+#define svldnt1sh_gather_u32base_offset_u32(...) __builtin_sve_svldnt1sh_gather_u32base_offset_u32(__VA_ARGS__)
+#define svldnt1sh_gather_u64base_offset_u64(...) __builtin_sve_svldnt1sh_gather_u64base_offset_u64(__VA_ARGS__)
+#define svldnt1sh_gather_u32base_offset_s32(...) __builtin_sve_svldnt1sh_gather_u32base_offset_s32(__VA_ARGS__)
+#define svldnt1sh_gather_u64base_offset_s64(...) __builtin_sve_svldnt1sh_gather_u64base_offset_s64(__VA_ARGS__)
+#define svldnt1sh_gather_u32base_u32(...) __builtin_sve_svldnt1sh_gather_u32base_u32(__VA_ARGS__)
+#define svldnt1sh_gather_u64base_u64(...) __builtin_sve_svldnt1sh_gather_u64base_u64(__VA_ARGS__)
+#define svldnt1sh_gather_u32base_s32(...) __builtin_sve_svldnt1sh_gather_u32base_s32(__VA_ARGS__)
+#define svldnt1sh_gather_u64base_s64(...) __builtin_sve_svldnt1sh_gather_u64base_s64(__VA_ARGS__)
+#define svldnt1sh_gather_s64index_u64(...) __builtin_sve_svldnt1sh_gather_s64index_u64(__VA_ARGS__)
+#define svldnt1sh_gather_s64index_s64(...) __builtin_sve_svldnt1sh_gather_s64index_s64(__VA_ARGS__)
+#define svldnt1sh_gather_u64index_u64(...) __builtin_sve_svldnt1sh_gather_u64index_u64(__VA_ARGS__)
+#define svldnt1sh_gather_u64index_s64(...) __builtin_sve_svldnt1sh_gather_u64index_s64(__VA_ARGS__)
+#define svldnt1sh_gather_u32offset_u32(...) __builtin_sve_svldnt1sh_gather_u32offset_u32(__VA_ARGS__)
+#define svldnt1sh_gather_u32offset_s32(...) __builtin_sve_svldnt1sh_gather_u32offset_s32(__VA_ARGS__)
+#define svldnt1sh_gather_s64offset_u64(...) __builtin_sve_svldnt1sh_gather_s64offset_u64(__VA_ARGS__)
+#define svldnt1sh_gather_s64offset_s64(...) __builtin_sve_svldnt1sh_gather_s64offset_s64(__VA_ARGS__)
+#define svldnt1sh_gather_u64offset_u64(...) __builtin_sve_svldnt1sh_gather_u64offset_u64(__VA_ARGS__)
+#define svldnt1sh_gather_u64offset_s64(...) __builtin_sve_svldnt1sh_gather_u64offset_s64(__VA_ARGS__)
+#define svldnt1sw_gather_u64base_index_u64(...) __builtin_sve_svldnt1sw_gather_u64base_index_u64(__VA_ARGS__)
+#define svldnt1sw_gather_u64base_index_s64(...) __builtin_sve_svldnt1sw_gather_u64base_index_s64(__VA_ARGS__)
+#define svldnt1sw_gather_u64base_offset_u64(...) __builtin_sve_svldnt1sw_gather_u64base_offset_u64(__VA_ARGS__)
+#define svldnt1sw_gather_u64base_offset_s64(...) __builtin_sve_svldnt1sw_gather_u64base_offset_s64(__VA_ARGS__)
+#define svldnt1sw_gather_u64base_u64(...) __builtin_sve_svldnt1sw_gather_u64base_u64(__VA_ARGS__)
+#define svldnt1sw_gather_u64base_s64(...) __builtin_sve_svldnt1sw_gather_u64base_s64(__VA_ARGS__)
+#define svldnt1sw_gather_s64index_u64(...) __builtin_sve_svldnt1sw_gather_s64index_u64(__VA_ARGS__)
+#define svldnt1sw_gather_s64index_s64(...) __builtin_sve_svldnt1sw_gather_s64index_s64(__VA_ARGS__)
+#define svldnt1sw_gather_u64index_u64(...) __builtin_sve_svldnt1sw_gather_u64index_u64(__VA_ARGS__)
+#define svldnt1sw_gather_u64index_s64(...) __builtin_sve_svldnt1sw_gather_u64index_s64(__VA_ARGS__)
+#define svldnt1sw_gather_s64offset_u64(...) __builtin_sve_svldnt1sw_gather_s64offset_u64(__VA_ARGS__)
+#define svldnt1sw_gather_s64offset_s64(...) __builtin_sve_svldnt1sw_gather_s64offset_s64(__VA_ARGS__)
+#define svldnt1sw_gather_u64offset_u64(...) __builtin_sve_svldnt1sw_gather_u64offset_u64(__VA_ARGS__)
+#define svldnt1sw_gather_u64offset_s64(...) __builtin_sve_svldnt1sw_gather_u64offset_s64(__VA_ARGS__)
+#define svldnt1ub_gather_u32base_offset_u32(...) __builtin_sve_svldnt1ub_gather_u32base_offset_u32(__VA_ARGS__)
+#define svldnt1ub_gather_u64base_offset_u64(...) __builtin_sve_svldnt1ub_gather_u64base_offset_u64(__VA_ARGS__)
+#define svldnt1ub_gather_u32base_offset_s32(...) __builtin_sve_svldnt1ub_gather_u32base_offset_s32(__VA_ARGS__)
+#define svldnt1ub_gather_u64base_offset_s64(...) __builtin_sve_svldnt1ub_gather_u64base_offset_s64(__VA_ARGS__)
+#define svldnt1ub_gather_u32base_u32(...) __builtin_sve_svldnt1ub_gather_u32base_u32(__VA_ARGS__)
+#define svldnt1ub_gather_u64base_u64(...) __builtin_sve_svldnt1ub_gather_u64base_u64(__VA_ARGS__)
+#define svldnt1ub_gather_u32base_s32(...) __builtin_sve_svldnt1ub_gather_u32base_s32(__VA_ARGS__)
+#define svldnt1ub_gather_u64base_s64(...) __builtin_sve_svldnt1ub_gather_u64base_s64(__VA_ARGS__)
+#define svldnt1ub_gather_u32offset_u32(...) __builtin_sve_svldnt1ub_gather_u32offset_u32(__VA_ARGS__)
+#define svldnt1ub_gather_u32offset_s32(...) __builtin_sve_svldnt1ub_gather_u32offset_s32(__VA_ARGS__)
+#define svldnt1ub_gather_s64offset_u64(...) __builtin_sve_svldnt1ub_gather_s64offset_u64(__VA_ARGS__)
+#define svldnt1ub_gather_s64offset_s64(...) __builtin_sve_svldnt1ub_gather_s64offset_s64(__VA_ARGS__)
+#define svldnt1ub_gather_u64offset_u64(...) __builtin_sve_svldnt1ub_gather_u64offset_u64(__VA_ARGS__)
+#define svldnt1ub_gather_u64offset_s64(...) __builtin_sve_svldnt1ub_gather_u64offset_s64(__VA_ARGS__)
+#define svldnt1uh_gather_u32base_index_u32(...) __builtin_sve_svldnt1uh_gather_u32base_index_u32(__VA_ARGS__)
+#define svldnt1uh_gather_u64base_index_u64(...) __builtin_sve_svldnt1uh_gather_u64base_index_u64(__VA_ARGS__)
+#define svldnt1uh_gather_u32base_index_s32(...) __builtin_sve_svldnt1uh_gather_u32base_index_s32(__VA_ARGS__)
+#define svldnt1uh_gather_u64base_index_s64(...) __builtin_sve_svldnt1uh_gather_u64base_index_s64(__VA_ARGS__)
+#define svldnt1uh_gather_u32base_offset_u32(...) __builtin_sve_svldnt1uh_gather_u32base_offset_u32(__VA_ARGS__)
+#define svldnt1uh_gather_u64base_offset_u64(...) __builtin_sve_svldnt1uh_gather_u64base_offset_u64(__VA_ARGS__)
+#define svldnt1uh_gather_u32base_offset_s32(...) __builtin_sve_svldnt1uh_gather_u32base_offset_s32(__VA_ARGS__)
+#define svldnt1uh_gather_u64base_offset_s64(...) __builtin_sve_svldnt1uh_gather_u64base_offset_s64(__VA_ARGS__)
+#define svldnt1uh_gather_u32base_u32(...) __builtin_sve_svldnt1uh_gather_u32base_u32(__VA_ARGS__)
+#define svldnt1uh_gather_u64base_u64(...) __builtin_sve_svldnt1uh_gather_u64base_u64(__VA_ARGS__)
+#define svldnt1uh_gather_u32base_s32(...) __builtin_sve_svldnt1uh_gather_u32base_s32(__VA_ARGS__)
+#define svldnt1uh_gather_u64base_s64(...) __builtin_sve_svldnt1uh_gather_u64base_s64(__VA_ARGS__)
+#define svldnt1uh_gather_s64index_u64(...) __builtin_sve_svldnt1uh_gather_s64index_u64(__VA_ARGS__)
+#define svldnt1uh_gather_s64index_s64(...) __builtin_sve_svldnt1uh_gather_s64index_s64(__VA_ARGS__)
+#define svldnt1uh_gather_u64index_u64(...) __builtin_sve_svldnt1uh_gather_u64index_u64(__VA_ARGS__)
+#define svldnt1uh_gather_u64index_s64(...) __builtin_sve_svldnt1uh_gather_u64index_s64(__VA_ARGS__)
+#define svldnt1uh_gather_u32offset_u32(...) __builtin_sve_svldnt1uh_gather_u32offset_u32(__VA_ARGS__)
+#define svldnt1uh_gather_u32offset_s32(...) __builtin_sve_svldnt1uh_gather_u32offset_s32(__VA_ARGS__)
+#define svldnt1uh_gather_s64offset_u64(...) __builtin_sve_svldnt1uh_gather_s64offset_u64(__VA_ARGS__)
+#define svldnt1uh_gather_s64offset_s64(...) __builtin_sve_svldnt1uh_gather_s64offset_s64(__VA_ARGS__)
+#define svldnt1uh_gather_u64offset_u64(...) __builtin_sve_svldnt1uh_gather_u64offset_u64(__VA_ARGS__)
+#define svldnt1uh_gather_u64offset_s64(...) __builtin_sve_svldnt1uh_gather_u64offset_s64(__VA_ARGS__)
+#define svldnt1uw_gather_u64base_index_u64(...) __builtin_sve_svldnt1uw_gather_u64base_index_u64(__VA_ARGS__)
+#define svldnt1uw_gather_u64base_index_s64(...) __builtin_sve_svldnt1uw_gather_u64base_index_s64(__VA_ARGS__)
+#define svldnt1uw_gather_u64base_offset_u64(...) __builtin_sve_svldnt1uw_gather_u64base_offset_u64(__VA_ARGS__)
+#define svldnt1uw_gather_u64base_offset_s64(...) __builtin_sve_svldnt1uw_gather_u64base_offset_s64(__VA_ARGS__)
+#define svldnt1uw_gather_u64base_u64(...) __builtin_sve_svldnt1uw_gather_u64base_u64(__VA_ARGS__)
+#define svldnt1uw_gather_u64base_s64(...) __builtin_sve_svldnt1uw_gather_u64base_s64(__VA_ARGS__)
+#define svldnt1uw_gather_s64index_u64(...) __builtin_sve_svldnt1uw_gather_s64index_u64(__VA_ARGS__)
+#define svldnt1uw_gather_s64index_s64(...) __builtin_sve_svldnt1uw_gather_s64index_s64(__VA_ARGS__)
+#define svldnt1uw_gather_u64index_u64(...) __builtin_sve_svldnt1uw_gather_u64index_u64(__VA_ARGS__)
+#define svldnt1uw_gather_u64index_s64(...) __builtin_sve_svldnt1uw_gather_u64index_s64(__VA_ARGS__)
+#define svldnt1uw_gather_s64offset_u64(...) __builtin_sve_svldnt1uw_gather_s64offset_u64(__VA_ARGS__)
+#define svldnt1uw_gather_s64offset_s64(...) __builtin_sve_svldnt1uw_gather_s64offset_s64(__VA_ARGS__)
+#define svldnt1uw_gather_u64offset_u64(...) __builtin_sve_svldnt1uw_gather_u64offset_u64(__VA_ARGS__)
+#define svldnt1uw_gather_u64offset_s64(...) __builtin_sve_svldnt1uw_gather_u64offset_s64(__VA_ARGS__)
+#define svlogb_f64_m(...) __builtin_sve_svlogb_f64_m(__VA_ARGS__)
+#define svlogb_f32_m(...) __builtin_sve_svlogb_f32_m(__VA_ARGS__)
+#define svlogb_f16_m(...) __builtin_sve_svlogb_f16_m(__VA_ARGS__)
+#define svlogb_f64_x(...) __builtin_sve_svlogb_f64_x(__VA_ARGS__)
+#define svlogb_f32_x(...) __builtin_sve_svlogb_f32_x(__VA_ARGS__)
+#define svlogb_f16_x(...) __builtin_sve_svlogb_f16_x(__VA_ARGS__)
+#define svlogb_f64_z(...) __builtin_sve_svlogb_f64_z(__VA_ARGS__)
+#define svlogb_f32_z(...) __builtin_sve_svlogb_f32_z(__VA_ARGS__)
+#define svlogb_f16_z(...) __builtin_sve_svlogb_f16_z(__VA_ARGS__)
+#define svmatch_u8(...) __builtin_sve_svmatch_u8(__VA_ARGS__)
+#define svmatch_u16(...) __builtin_sve_svmatch_u16(__VA_ARGS__)
+#define svmatch_s8(...) __builtin_sve_svmatch_s8(__VA_ARGS__)
+#define svmatch_s16(...) __builtin_sve_svmatch_s16(__VA_ARGS__)
+#define svmaxnmp_f64_m(...) __builtin_sve_svmaxnmp_f64_m(__VA_ARGS__)
+#define svmaxnmp_f32_m(...) __builtin_sve_svmaxnmp_f32_m(__VA_ARGS__)
+#define svmaxnmp_f16_m(...) __builtin_sve_svmaxnmp_f16_m(__VA_ARGS__)
+#define svmaxnmp_f64_x(...) __builtin_sve_svmaxnmp_f64_x(__VA_ARGS__)
+#define svmaxnmp_f32_x(...) __builtin_sve_svmaxnmp_f32_x(__VA_ARGS__)
+#define svmaxnmp_f16_x(...) __builtin_sve_svmaxnmp_f16_x(__VA_ARGS__)
+#define svmaxp_f64_m(...) __builtin_sve_svmaxp_f64_m(__VA_ARGS__)
+#define svmaxp_f32_m(...) __builtin_sve_svmaxp_f32_m(__VA_ARGS__)
+#define svmaxp_f16_m(...) __builtin_sve_svmaxp_f16_m(__VA_ARGS__)
+#define svmaxp_f64_x(...) __builtin_sve_svmaxp_f64_x(__VA_ARGS__)
+#define svmaxp_f32_x(...) __builtin_sve_svmaxp_f32_x(__VA_ARGS__)
+#define svmaxp_f16_x(...) __builtin_sve_svmaxp_f16_x(__VA_ARGS__)
+#define svmaxp_s8_m(...) __builtin_sve_svmaxp_s8_m(__VA_ARGS__)
+#define svmaxp_s32_m(...) __builtin_sve_svmaxp_s32_m(__VA_ARGS__)
+#define svmaxp_s64_m(...) __builtin_sve_svmaxp_s64_m(__VA_ARGS__)
+#define svmaxp_s16_m(...) __builtin_sve_svmaxp_s16_m(__VA_ARGS__)
+#define svmaxp_s8_x(...) __builtin_sve_svmaxp_s8_x(__VA_ARGS__)
+#define svmaxp_s32_x(...) __builtin_sve_svmaxp_s32_x(__VA_ARGS__)
+#define svmaxp_s64_x(...) __builtin_sve_svmaxp_s64_x(__VA_ARGS__)
+#define svmaxp_s16_x(...) __builtin_sve_svmaxp_s16_x(__VA_ARGS__)
+#define svmaxp_u8_m(...) __builtin_sve_svmaxp_u8_m(__VA_ARGS__)
+#define svmaxp_u32_m(...) __builtin_sve_svmaxp_u32_m(__VA_ARGS__)
+#define svmaxp_u64_m(...) __builtin_sve_svmaxp_u64_m(__VA_ARGS__)
+#define svmaxp_u16_m(...) __builtin_sve_svmaxp_u16_m(__VA_ARGS__)
+#define svmaxp_u8_x(...) __builtin_sve_svmaxp_u8_x(__VA_ARGS__)
+#define svmaxp_u32_x(...) __builtin_sve_svmaxp_u32_x(__VA_ARGS__)
+#define svmaxp_u64_x(...) __builtin_sve_svmaxp_u64_x(__VA_ARGS__)
+#define svmaxp_u16_x(...) __builtin_sve_svmaxp_u16_x(__VA_ARGS__)
+#define svminnmp_f64_m(...) __builtin_sve_svminnmp_f64_m(__VA_ARGS__)
+#define svminnmp_f32_m(...) __builtin_sve_svminnmp_f32_m(__VA_ARGS__)
+#define svminnmp_f16_m(...) __builtin_sve_svminnmp_f16_m(__VA_ARGS__)
+#define svminnmp_f64_x(...) __builtin_sve_svminnmp_f64_x(__VA_ARGS__)
+#define svminnmp_f32_x(...) __builtin_sve_svminnmp_f32_x(__VA_ARGS__)
+#define svminnmp_f16_x(...) __builtin_sve_svminnmp_f16_x(__VA_ARGS__)
+#define svminp_f64_m(...) __builtin_sve_svminp_f64_m(__VA_ARGS__)
+#define svminp_f32_m(...) __builtin_sve_svminp_f32_m(__VA_ARGS__)
+#define svminp_f16_m(...) __builtin_sve_svminp_f16_m(__VA_ARGS__)
+#define svminp_f64_x(...) __builtin_sve_svminp_f64_x(__VA_ARGS__)
+#define svminp_f32_x(...) __builtin_sve_svminp_f32_x(__VA_ARGS__)
+#define svminp_f16_x(...) __builtin_sve_svminp_f16_x(__VA_ARGS__)
+#define svminp_s8_m(...) __builtin_sve_svminp_s8_m(__VA_ARGS__)
+#define svminp_s32_m(...) __builtin_sve_svminp_s32_m(__VA_ARGS__)
+#define svminp_s64_m(...) __builtin_sve_svminp_s64_m(__VA_ARGS__)
+#define svminp_s16_m(...) __builtin_sve_svminp_s16_m(__VA_ARGS__)
+#define svminp_s8_x(...) __builtin_sve_svminp_s8_x(__VA_ARGS__)
+#define svminp_s32_x(...) __builtin_sve_svminp_s32_x(__VA_ARGS__)
+#define svminp_s64_x(...) __builtin_sve_svminp_s64_x(__VA_ARGS__)
+#define svminp_s16_x(...) __builtin_sve_svminp_s16_x(__VA_ARGS__)
+#define svminp_u8_m(...) __builtin_sve_svminp_u8_m(__VA_ARGS__)
+#define svminp_u32_m(...) __builtin_sve_svminp_u32_m(__VA_ARGS__)
+#define svminp_u64_m(...) __builtin_sve_svminp_u64_m(__VA_ARGS__)
+#define svminp_u16_m(...) __builtin_sve_svminp_u16_m(__VA_ARGS__)
+#define svminp_u8_x(...) __builtin_sve_svminp_u8_x(__VA_ARGS__)
+#define svminp_u32_x(...) __builtin_sve_svminp_u32_x(__VA_ARGS__)
+#define svminp_u64_x(...) __builtin_sve_svminp_u64_x(__VA_ARGS__)
+#define svminp_u16_x(...) __builtin_sve_svminp_u16_x(__VA_ARGS__)
+#define svmla_lane_u32(...) __builtin_sve_svmla_lane_u32(__VA_ARGS__)
+#define svmla_lane_u64(...) __builtin_sve_svmla_lane_u64(__VA_ARGS__)
+#define svmla_lane_u16(...) __builtin_sve_svmla_lane_u16(__VA_ARGS__)
+#define svmla_lane_s32(...) __builtin_sve_svmla_lane_s32(__VA_ARGS__)
+#define svmla_lane_s64(...) __builtin_sve_svmla_lane_s64(__VA_ARGS__)
+#define svmla_lane_s16(...) __builtin_sve_svmla_lane_s16(__VA_ARGS__)
+#define svmlalb_n_f32(...) __builtin_sve_svmlalb_n_f32(__VA_ARGS__)
+#define svmlalb_n_s32(...) __builtin_sve_svmlalb_n_s32(__VA_ARGS__)
+#define svmlalb_n_s64(...) __builtin_sve_svmlalb_n_s64(__VA_ARGS__)
+#define svmlalb_n_s16(...) __builtin_sve_svmlalb_n_s16(__VA_ARGS__)
+#define svmlalb_n_u32(...) __builtin_sve_svmlalb_n_u32(__VA_ARGS__)
+#define svmlalb_n_u64(...) __builtin_sve_svmlalb_n_u64(__VA_ARGS__)
+#define svmlalb_n_u16(...) __builtin_sve_svmlalb_n_u16(__VA_ARGS__)
+#define svmlalb_f32(...) __builtin_sve_svmlalb_f32(__VA_ARGS__)
+#define svmlalb_s32(...) __builtin_sve_svmlalb_s32(__VA_ARGS__)
+#define svmlalb_s64(...) __builtin_sve_svmlalb_s64(__VA_ARGS__)
+#define svmlalb_s16(...) __builtin_sve_svmlalb_s16(__VA_ARGS__)
+#define svmlalb_u32(...) __builtin_sve_svmlalb_u32(__VA_ARGS__)
+#define svmlalb_u64(...) __builtin_sve_svmlalb_u64(__VA_ARGS__)
+#define svmlalb_u16(...) __builtin_sve_svmlalb_u16(__VA_ARGS__)
+#define svmlalb_lane_f32(...) __builtin_sve_svmlalb_lane_f32(__VA_ARGS__)
+#define svmlalb_lane_s32(...) __builtin_sve_svmlalb_lane_s32(__VA_ARGS__)
+#define svmlalb_lane_s64(...) __builtin_sve_svmlalb_lane_s64(__VA_ARGS__)
+#define svmlalb_lane_u32(...) __builtin_sve_svmlalb_lane_u32(__VA_ARGS__)
+#define svmlalb_lane_u64(...) __builtin_sve_svmlalb_lane_u64(__VA_ARGS__)
+#define svmlalt_n_f32(...) __builtin_sve_svmlalt_n_f32(__VA_ARGS__)
+#define svmlalt_n_s32(...) __builtin_sve_svmlalt_n_s32(__VA_ARGS__)
+#define svmlalt_n_s64(...) __builtin_sve_svmlalt_n_s64(__VA_ARGS__)
+#define svmlalt_n_s16(...) __builtin_sve_svmlalt_n_s16(__VA_ARGS__)
+#define svmlalt_n_u32(...) __builtin_sve_svmlalt_n_u32(__VA_ARGS__)
+#define svmlalt_n_u64(...) __builtin_sve_svmlalt_n_u64(__VA_ARGS__)
+#define svmlalt_n_u16(...) __builtin_sve_svmlalt_n_u16(__VA_ARGS__)
+#define svmlalt_f32(...) __builtin_sve_svmlalt_f32(__VA_ARGS__)
+#define svmlalt_s32(...) __builtin_sve_svmlalt_s32(__VA_ARGS__)
+#define svmlalt_s64(...) __builtin_sve_svmlalt_s64(__VA_ARGS__)
+#define svmlalt_s16(...) __builtin_sve_svmlalt_s16(__VA_ARGS__)
+#define svmlalt_u32(...) __builtin_sve_svmlalt_u32(__VA_ARGS__)
+#define svmlalt_u64(...) __builtin_sve_svmlalt_u64(__VA_ARGS__)
+#define svmlalt_u16(...) __builtin_sve_svmlalt_u16(__VA_ARGS__)
+#define svmlalt_lane_f32(...) __builtin_sve_svmlalt_lane_f32(__VA_ARGS__)
+#define svmlalt_lane_s32(...) __builtin_sve_svmlalt_lane_s32(__VA_ARGS__)
+#define svmlalt_lane_s64(...) __builtin_sve_svmlalt_lane_s64(__VA_ARGS__)
+#define svmlalt_lane_u32(...) __builtin_sve_svmlalt_lane_u32(__VA_ARGS__)
+#define svmlalt_lane_u64(...) __builtin_sve_svmlalt_lane_u64(__VA_ARGS__)
+#define svmls_lane_u32(...) __builtin_sve_svmls_lane_u32(__VA_ARGS__)
+#define svmls_lane_u64(...) __builtin_sve_svmls_lane_u64(__VA_ARGS__)
+#define svmls_lane_u16(...) __builtin_sve_svmls_lane_u16(__VA_ARGS__)
+#define svmls_lane_s32(...) __builtin_sve_svmls_lane_s32(__VA_ARGS__)
+#define svmls_lane_s64(...) __builtin_sve_svmls_lane_s64(__VA_ARGS__)
+#define svmls_lane_s16(...) __builtin_sve_svmls_lane_s16(__VA_ARGS__)
+#define svmlslb_n_f32(...) __builtin_sve_svmlslb_n_f32(__VA_ARGS__)
+#define svmlslb_n_s32(...) __builtin_sve_svmlslb_n_s32(__VA_ARGS__)
+#define svmlslb_n_s64(...) __builtin_sve_svmlslb_n_s64(__VA_ARGS__)
+#define svmlslb_n_s16(...) __builtin_sve_svmlslb_n_s16(__VA_ARGS__)
+#define svmlslb_n_u32(...) __builtin_sve_svmlslb_n_u32(__VA_ARGS__)
+#define svmlslb_n_u64(...) __builtin_sve_svmlslb_n_u64(__VA_ARGS__)
+#define svmlslb_n_u16(...) __builtin_sve_svmlslb_n_u16(__VA_ARGS__)
+#define svmlslb_f32(...) __builtin_sve_svmlslb_f32(__VA_ARGS__)
+#define svmlslb_s32(...) __builtin_sve_svmlslb_s32(__VA_ARGS__)
+#define svmlslb_s64(...) __builtin_sve_svmlslb_s64(__VA_ARGS__)
+#define svmlslb_s16(...) __builtin_sve_svmlslb_s16(__VA_ARGS__)
+#define svmlslb_u32(...) __builtin_sve_svmlslb_u32(__VA_ARGS__)
+#define svmlslb_u64(...) __builtin_sve_svmlslb_u64(__VA_ARGS__)
+#define svmlslb_u16(...) __builtin_sve_svmlslb_u16(__VA_ARGS__)
+#define svmlslb_lane_f32(...) __builtin_sve_svmlslb_lane_f32(__VA_ARGS__)
+#define svmlslb_lane_s32(...) __builtin_sve_svmlslb_lane_s32(__VA_ARGS__)
+#define svmlslb_lane_s64(...) __builtin_sve_svmlslb_lane_s64(__VA_ARGS__)
+#define svmlslb_lane_u32(...) __builtin_sve_svmlslb_lane_u32(__VA_ARGS__)
+#define svmlslb_lane_u64(...) __builtin_sve_svmlslb_lane_u64(__VA_ARGS__)
+#define svmlslt_n_f32(...) __builtin_sve_svmlslt_n_f32(__VA_ARGS__)
+#define svmlslt_n_s32(...) __builtin_sve_svmlslt_n_s32(__VA_ARGS__)
+#define svmlslt_n_s64(...) __builtin_sve_svmlslt_n_s64(__VA_ARGS__)
+#define svmlslt_n_s16(...) __builtin_sve_svmlslt_n_s16(__VA_ARGS__)
+#define svmlslt_n_u32(...) __builtin_sve_svmlslt_n_u32(__VA_ARGS__)
+#define svmlslt_n_u64(...) __builtin_sve_svmlslt_n_u64(__VA_ARGS__)
+#define svmlslt_n_u16(...) __builtin_sve_svmlslt_n_u16(__VA_ARGS__)
+#define svmlslt_f32(...) __builtin_sve_svmlslt_f32(__VA_ARGS__)
+#define svmlslt_s32(...) __builtin_sve_svmlslt_s32(__VA_ARGS__)
+#define svmlslt_s64(...) __builtin_sve_svmlslt_s64(__VA_ARGS__)
+#define svmlslt_s16(...) __builtin_sve_svmlslt_s16(__VA_ARGS__)
+#define svmlslt_u32(...) __builtin_sve_svmlslt_u32(__VA_ARGS__)
+#define svmlslt_u64(...) __builtin_sve_svmlslt_u64(__VA_ARGS__)
+#define svmlslt_u16(...) __builtin_sve_svmlslt_u16(__VA_ARGS__)
+#define svmlslt_lane_f32(...) __builtin_sve_svmlslt_lane_f32(__VA_ARGS__)
+#define svmlslt_lane_s32(...) __builtin_sve_svmlslt_lane_s32(__VA_ARGS__)
+#define svmlslt_lane_s64(...) __builtin_sve_svmlslt_lane_s64(__VA_ARGS__)
+#define svmlslt_lane_u32(...) __builtin_sve_svmlslt_lane_u32(__VA_ARGS__)
+#define svmlslt_lane_u64(...) __builtin_sve_svmlslt_lane_u64(__VA_ARGS__)
+#define svmovlb_s32(...) __builtin_sve_svmovlb_s32(__VA_ARGS__)
+#define svmovlb_s64(...) __builtin_sve_svmovlb_s64(__VA_ARGS__)
+#define svmovlb_s16(...) __builtin_sve_svmovlb_s16(__VA_ARGS__)
+#define svmovlb_u32(...) __builtin_sve_svmovlb_u32(__VA_ARGS__)
+#define svmovlb_u64(...) __builtin_sve_svmovlb_u64(__VA_ARGS__)
+#define svmovlb_u16(...) __builtin_sve_svmovlb_u16(__VA_ARGS__)
+#define svmovlt_s32(...) __builtin_sve_svmovlt_s32(__VA_ARGS__)
+#define svmovlt_s64(...) __builtin_sve_svmovlt_s64(__VA_ARGS__)
+#define svmovlt_s16(...) __builtin_sve_svmovlt_s16(__VA_ARGS__)
+#define svmovlt_u32(...) __builtin_sve_svmovlt_u32(__VA_ARGS__)
+#define svmovlt_u64(...) __builtin_sve_svmovlt_u64(__VA_ARGS__)
+#define svmovlt_u16(...) __builtin_sve_svmovlt_u16(__VA_ARGS__)
+#define svmul_lane_u32(...) __builtin_sve_svmul_lane_u32(__VA_ARGS__)
+#define svmul_lane_u64(...) __builtin_sve_svmul_lane_u64(__VA_ARGS__)
+#define svmul_lane_u16(...) __builtin_sve_svmul_lane_u16(__VA_ARGS__)
+#define svmul_lane_s32(...) __builtin_sve_svmul_lane_s32(__VA_ARGS__)
+#define svmul_lane_s64(...) __builtin_sve_svmul_lane_s64(__VA_ARGS__)
+#define svmul_lane_s16(...) __builtin_sve_svmul_lane_s16(__VA_ARGS__)
+#define svmullb_n_s32(...) __builtin_sve_svmullb_n_s32(__VA_ARGS__)
+#define svmullb_n_s64(...) __builtin_sve_svmullb_n_s64(__VA_ARGS__)
+#define svmullb_n_s16(...) __builtin_sve_svmullb_n_s16(__VA_ARGS__)
+#define svmullb_n_u32(...) __builtin_sve_svmullb_n_u32(__VA_ARGS__)
+#define svmullb_n_u64(...) __builtin_sve_svmullb_n_u64(__VA_ARGS__)
+#define svmullb_n_u16(...) __builtin_sve_svmullb_n_u16(__VA_ARGS__)
+#define svmullb_s32(...) __builtin_sve_svmullb_s32(__VA_ARGS__)
+#define svmullb_s64(...) __builtin_sve_svmullb_s64(__VA_ARGS__)
+#define svmullb_s16(...) __builtin_sve_svmullb_s16(__VA_ARGS__)
+#define svmullb_u32(...) __builtin_sve_svmullb_u32(__VA_ARGS__)
+#define svmullb_u64(...) __builtin_sve_svmullb_u64(__VA_ARGS__)
+#define svmullb_u16(...) __builtin_sve_svmullb_u16(__VA_ARGS__)
+#define svmullb_lane_s32(...) __builtin_sve_svmullb_lane_s32(__VA_ARGS__)
+#define svmullb_lane_s64(...) __builtin_sve_svmullb_lane_s64(__VA_ARGS__)
+#define svmullb_lane_u32(...) __builtin_sve_svmullb_lane_u32(__VA_ARGS__)
+#define svmullb_lane_u64(...) __builtin_sve_svmullb_lane_u64(__VA_ARGS__)
+#define svmullt_n_s32(...) __builtin_sve_svmullt_n_s32(__VA_ARGS__)
+#define svmullt_n_s64(...) __builtin_sve_svmullt_n_s64(__VA_ARGS__)
+#define svmullt_n_s16(...) __builtin_sve_svmullt_n_s16(__VA_ARGS__)
+#define svmullt_n_u32(...) __builtin_sve_svmullt_n_u32(__VA_ARGS__)
+#define svmullt_n_u64(...) __builtin_sve_svmullt_n_u64(__VA_ARGS__)
+#define svmullt_n_u16(...) __builtin_sve_svmullt_n_u16(__VA_ARGS__)
+#define svmullt_s32(...) __builtin_sve_svmullt_s32(__VA_ARGS__)
+#define svmullt_s64(...) __builtin_sve_svmullt_s64(__VA_ARGS__)
+#define svmullt_s16(...) __builtin_sve_svmullt_s16(__VA_ARGS__)
+#define svmullt_u32(...) __builtin_sve_svmullt_u32(__VA_ARGS__)
+#define svmullt_u64(...) __builtin_sve_svmullt_u64(__VA_ARGS__)
+#define svmullt_u16(...) __builtin_sve_svmullt_u16(__VA_ARGS__)
+#define svmullt_lane_s32(...) __builtin_sve_svmullt_lane_s32(__VA_ARGS__)
+#define svmullt_lane_s64(...) __builtin_sve_svmullt_lane_s64(__VA_ARGS__)
+#define svmullt_lane_u32(...) __builtin_sve_svmullt_lane_u32(__VA_ARGS__)
+#define svmullt_lane_u64(...) __builtin_sve_svmullt_lane_u64(__VA_ARGS__)
+#define svnbsl_n_u8(...) __builtin_sve_svnbsl_n_u8(__VA_ARGS__)
+#define svnbsl_n_u32(...) __builtin_sve_svnbsl_n_u32(__VA_ARGS__)
+#define svnbsl_n_u64(...) __builtin_sve_svnbsl_n_u64(__VA_ARGS__)
+#define svnbsl_n_u16(...) __builtin_sve_svnbsl_n_u16(__VA_ARGS__)
+#define svnbsl_n_s8(...) __builtin_sve_svnbsl_n_s8(__VA_ARGS__)
+#define svnbsl_n_s32(...) __builtin_sve_svnbsl_n_s32(__VA_ARGS__)
+#define svnbsl_n_s64(...) __builtin_sve_svnbsl_n_s64(__VA_ARGS__)
+#define svnbsl_n_s16(...) __builtin_sve_svnbsl_n_s16(__VA_ARGS__)
+#define svnbsl_u8(...) __builtin_sve_svnbsl_u8(__VA_ARGS__)
+#define svnbsl_u32(...) __builtin_sve_svnbsl_u32(__VA_ARGS__)
+#define svnbsl_u64(...) __builtin_sve_svnbsl_u64(__VA_ARGS__)
+#define svnbsl_u16(...) __builtin_sve_svnbsl_u16(__VA_ARGS__)
+#define svnbsl_s8(...) __builtin_sve_svnbsl_s8(__VA_ARGS__)
+#define svnbsl_s32(...) __builtin_sve_svnbsl_s32(__VA_ARGS__)
+#define svnbsl_s64(...) __builtin_sve_svnbsl_s64(__VA_ARGS__)
+#define svnbsl_s16(...) __builtin_sve_svnbsl_s16(__VA_ARGS__)
+#define svnmatch_u8(...) __builtin_sve_svnmatch_u8(__VA_ARGS__)
+#define svnmatch_u16(...) __builtin_sve_svnmatch_u16(__VA_ARGS__)
+#define svnmatch_s8(...) __builtin_sve_svnmatch_s8(__VA_ARGS__)
+#define svnmatch_s16(...) __builtin_sve_svnmatch_s16(__VA_ARGS__)
+#define svpmul_n_u8(...) __builtin_sve_svpmul_n_u8(__VA_ARGS__)
+#define svpmul_u8(...) __builtin_sve_svpmul_u8(__VA_ARGS__)
+#define svpmullb_n_u64(...) __builtin_sve_svpmullb_n_u64(__VA_ARGS__)
+#define svpmullb_n_u16(...) __builtin_sve_svpmullb_n_u16(__VA_ARGS__)
+#define svpmullb_u64(...) __builtin_sve_svpmullb_u64(__VA_ARGS__)
+#define svpmullb_u16(...) __builtin_sve_svpmullb_u16(__VA_ARGS__)
+#define svpmullb_pair_n_u8(...) __builtin_sve_svpmullb_pair_n_u8(__VA_ARGS__)
+#define svpmullb_pair_n_u32(...) __builtin_sve_svpmullb_pair_n_u32(__VA_ARGS__)
+#define svpmullb_pair_u8(...) __builtin_sve_svpmullb_pair_u8(__VA_ARGS__)
+#define svpmullb_pair_u32(...) __builtin_sve_svpmullb_pair_u32(__VA_ARGS__)
+#define svpmullt_n_u64(...) __builtin_sve_svpmullt_n_u64(__VA_ARGS__)
+#define svpmullt_n_u16(...) __builtin_sve_svpmullt_n_u16(__VA_ARGS__)
+#define svpmullt_u64(...) __builtin_sve_svpmullt_u64(__VA_ARGS__)
+#define svpmullt_u16(...) __builtin_sve_svpmullt_u16(__VA_ARGS__)
+#define svpmullt_pair_n_u8(...) __builtin_sve_svpmullt_pair_n_u8(__VA_ARGS__)
+#define svpmullt_pair_n_u32(...) __builtin_sve_svpmullt_pair_n_u32(__VA_ARGS__)
+#define svpmullt_pair_u8(...) __builtin_sve_svpmullt_pair_u8(__VA_ARGS__)
+#define svpmullt_pair_u32(...) __builtin_sve_svpmullt_pair_u32(__VA_ARGS__)
+#define svqabs_s8_m(...) __builtin_sve_svqabs_s8_m(__VA_ARGS__)
+#define svqabs_s32_m(...) __builtin_sve_svqabs_s32_m(__VA_ARGS__)
+#define svqabs_s64_m(...) __builtin_sve_svqabs_s64_m(__VA_ARGS__)
+#define svqabs_s16_m(...) __builtin_sve_svqabs_s16_m(__VA_ARGS__)
+#define svqabs_s8_x(...) __builtin_sve_svqabs_s8_x(__VA_ARGS__)
+#define svqabs_s32_x(...) __builtin_sve_svqabs_s32_x(__VA_ARGS__)
+#define svqabs_s64_x(...) __builtin_sve_svqabs_s64_x(__VA_ARGS__)
+#define svqabs_s16_x(...) __builtin_sve_svqabs_s16_x(__VA_ARGS__)
+#define svqabs_s8_z(...) __builtin_sve_svqabs_s8_z(__VA_ARGS__)
+#define svqabs_s32_z(...) __builtin_sve_svqabs_s32_z(__VA_ARGS__)
+#define svqabs_s64_z(...) __builtin_sve_svqabs_s64_z(__VA_ARGS__)
+#define svqabs_s16_z(...) __builtin_sve_svqabs_s16_z(__VA_ARGS__)
+#define svqadd_n_s8_m(...) __builtin_sve_svqadd_n_s8_m(__VA_ARGS__)
+#define svqadd_n_s32_m(...) __builtin_sve_svqadd_n_s32_m(__VA_ARGS__)
+#define svqadd_n_s64_m(...) __builtin_sve_svqadd_n_s64_m(__VA_ARGS__)
+#define svqadd_n_s16_m(...) __builtin_sve_svqadd_n_s16_m(__VA_ARGS__)
+#define svqadd_n_s8_x(...) __builtin_sve_svqadd_n_s8_x(__VA_ARGS__)
+#define svqadd_n_s32_x(...) __builtin_sve_svqadd_n_s32_x(__VA_ARGS__)
+#define svqadd_n_s64_x(...) __builtin_sve_svqadd_n_s64_x(__VA_ARGS__)
+#define svqadd_n_s16_x(...) __builtin_sve_svqadd_n_s16_x(__VA_ARGS__)
+#define svqadd_n_s8_z(...) __builtin_sve_svqadd_n_s8_z(__VA_ARGS__)
+#define svqadd_n_s32_z(...) __builtin_sve_svqadd_n_s32_z(__VA_ARGS__)
+#define svqadd_n_s64_z(...) __builtin_sve_svqadd_n_s64_z(__VA_ARGS__)
+#define svqadd_n_s16_z(...) __builtin_sve_svqadd_n_s16_z(__VA_ARGS__)
+#define svqadd_n_u8_m(...) __builtin_sve_svqadd_n_u8_m(__VA_ARGS__)
+#define svqadd_n_u32_m(...) __builtin_sve_svqadd_n_u32_m(__VA_ARGS__)
+#define svqadd_n_u64_m(...) __builtin_sve_svqadd_n_u64_m(__VA_ARGS__)
+#define svqadd_n_u16_m(...) __builtin_sve_svqadd_n_u16_m(__VA_ARGS__)
+#define svqadd_n_u8_x(...) __builtin_sve_svqadd_n_u8_x(__VA_ARGS__)
+#define svqadd_n_u32_x(...) __builtin_sve_svqadd_n_u32_x(__VA_ARGS__)
+#define svqadd_n_u64_x(...) __builtin_sve_svqadd_n_u64_x(__VA_ARGS__)
+#define svqadd_n_u16_x(...) __builtin_sve_svqadd_n_u16_x(__VA_ARGS__)
+#define svqadd_n_u8_z(...) __builtin_sve_svqadd_n_u8_z(__VA_ARGS__)
+#define svqadd_n_u32_z(...) __builtin_sve_svqadd_n_u32_z(__VA_ARGS__)
+#define svqadd_n_u64_z(...) __builtin_sve_svqadd_n_u64_z(__VA_ARGS__)
+#define svqadd_n_u16_z(...) __builtin_sve_svqadd_n_u16_z(__VA_ARGS__)
+#define svqadd_s8_m(...) __builtin_sve_svqadd_s8_m(__VA_ARGS__)
+#define svqadd_s32_m(...) __builtin_sve_svqadd_s32_m(__VA_ARGS__)
+#define svqadd_s64_m(...) __builtin_sve_svqadd_s64_m(__VA_ARGS__)
+#define svqadd_s16_m(...) __builtin_sve_svqadd_s16_m(__VA_ARGS__)
+#define svqadd_s8_x(...) __builtin_sve_svqadd_s8_x(__VA_ARGS__)
+#define svqadd_s32_x(...) __builtin_sve_svqadd_s32_x(__VA_ARGS__)
+#define svqadd_s64_x(...) __builtin_sve_svqadd_s64_x(__VA_ARGS__)
+#define svqadd_s16_x(...) __builtin_sve_svqadd_s16_x(__VA_ARGS__)
+#define svqadd_s8_z(...) __builtin_sve_svqadd_s8_z(__VA_ARGS__)
+#define svqadd_s32_z(...) __builtin_sve_svqadd_s32_z(__VA_ARGS__)
+#define svqadd_s64_z(...) __builtin_sve_svqadd_s64_z(__VA_ARGS__)
+#define svqadd_s16_z(...) __builtin_sve_svqadd_s16_z(__VA_ARGS__)
+#define svqadd_u8_m(...) __builtin_sve_svqadd_u8_m(__VA_ARGS__)
+#define svqadd_u32_m(...) __builtin_sve_svqadd_u32_m(__VA_ARGS__)
+#define svqadd_u64_m(...) __builtin_sve_svqadd_u64_m(__VA_ARGS__)
+#define svqadd_u16_m(...) __builtin_sve_svqadd_u16_m(__VA_ARGS__)
+#define svqadd_u8_x(...) __builtin_sve_svqadd_u8_x(__VA_ARGS__)
+#define svqadd_u32_x(...) __builtin_sve_svqadd_u32_x(__VA_ARGS__)
+#define svqadd_u64_x(...) __builtin_sve_svqadd_u64_x(__VA_ARGS__)
+#define svqadd_u16_x(...) __builtin_sve_svqadd_u16_x(__VA_ARGS__)
+#define svqadd_u8_z(...) __builtin_sve_svqadd_u8_z(__VA_ARGS__)
+#define svqadd_u32_z(...) __builtin_sve_svqadd_u32_z(__VA_ARGS__)
+#define svqadd_u64_z(...) __builtin_sve_svqadd_u64_z(__VA_ARGS__)
+#define svqadd_u16_z(...) __builtin_sve_svqadd_u16_z(__VA_ARGS__)
+#define svqcadd_s8(...) __builtin_sve_svqcadd_s8(__VA_ARGS__)
+#define svqcadd_s32(...) __builtin_sve_svqcadd_s32(__VA_ARGS__)
+#define svqcadd_s64(...) __builtin_sve_svqcadd_s64(__VA_ARGS__)
+#define svqcadd_s16(...) __builtin_sve_svqcadd_s16(__VA_ARGS__)
+#define svqdmlalb_n_s32(...) __builtin_sve_svqdmlalb_n_s32(__VA_ARGS__)
+#define svqdmlalb_n_s64(...) __builtin_sve_svqdmlalb_n_s64(__VA_ARGS__)
+#define svqdmlalb_n_s16(...) __builtin_sve_svqdmlalb_n_s16(__VA_ARGS__)
+#define svqdmlalb_s32(...) __builtin_sve_svqdmlalb_s32(__VA_ARGS__)
+#define svqdmlalb_s64(...) __builtin_sve_svqdmlalb_s64(__VA_ARGS__)
+#define svqdmlalb_s16(...) __builtin_sve_svqdmlalb_s16(__VA_ARGS__)
+#define svqdmlalb_lane_s32(...) __builtin_sve_svqdmlalb_lane_s32(__VA_ARGS__)
+#define svqdmlalb_lane_s64(...) __builtin_sve_svqdmlalb_lane_s64(__VA_ARGS__)
+#define svqdmlalbt_n_s32(...) __builtin_sve_svqdmlalbt_n_s32(__VA_ARGS__)
+#define svqdmlalbt_n_s64(...) __builtin_sve_svqdmlalbt_n_s64(__VA_ARGS__)
+#define svqdmlalbt_n_s16(...) __builtin_sve_svqdmlalbt_n_s16(__VA_ARGS__)
+#define svqdmlalbt_s32(...) __builtin_sve_svqdmlalbt_s32(__VA_ARGS__)
+#define svqdmlalbt_s64(...) __builtin_sve_svqdmlalbt_s64(__VA_ARGS__)
+#define svqdmlalbt_s16(...) __builtin_sve_svqdmlalbt_s16(__VA_ARGS__)
+#define svqdmlalt_n_s32(...) __builtin_sve_svqdmlalt_n_s32(__VA_ARGS__)
+#define svqdmlalt_n_s64(...) __builtin_sve_svqdmlalt_n_s64(__VA_ARGS__)
+#define svqdmlalt_n_s16(...) __builtin_sve_svqdmlalt_n_s16(__VA_ARGS__)
+#define svqdmlalt_s32(...) __builtin_sve_svqdmlalt_s32(__VA_ARGS__)
+#define svqdmlalt_s64(...) __builtin_sve_svqdmlalt_s64(__VA_ARGS__)
+#define svqdmlalt_s16(...) __builtin_sve_svqdmlalt_s16(__VA_ARGS__)
+#define svqdmlalt_lane_s32(...) __builtin_sve_svqdmlalt_lane_s32(__VA_ARGS__)
+#define svqdmlalt_lane_s64(...) __builtin_sve_svqdmlalt_lane_s64(__VA_ARGS__)
+#define svqdmlslb_n_s32(...) __builtin_sve_svqdmlslb_n_s32(__VA_ARGS__)
+#define svqdmlslb_n_s64(...) __builtin_sve_svqdmlslb_n_s64(__VA_ARGS__)
+#define svqdmlslb_n_s16(...) __builtin_sve_svqdmlslb_n_s16(__VA_ARGS__)
+#define svqdmlslb_s32(...) __builtin_sve_svqdmlslb_s32(__VA_ARGS__)
+#define svqdmlslb_s64(...) __builtin_sve_svqdmlslb_s64(__VA_ARGS__)
+#define svqdmlslb_s16(...) __builtin_sve_svqdmlslb_s16(__VA_ARGS__)
+#define svqdmlslb_lane_s32(...) __builtin_sve_svqdmlslb_lane_s32(__VA_ARGS__)
+#define svqdmlslb_lane_s64(...) __builtin_sve_svqdmlslb_lane_s64(__VA_ARGS__)
+#define svqdmlslbt_n_s32(...) __builtin_sve_svqdmlslbt_n_s32(__VA_ARGS__)
+#define svqdmlslbt_n_s64(...) __builtin_sve_svqdmlslbt_n_s64(__VA_ARGS__)
+#define svqdmlslbt_n_s16(...) __builtin_sve_svqdmlslbt_n_s16(__VA_ARGS__)
+#define svqdmlslbt_s32(...) __builtin_sve_svqdmlslbt_s32(__VA_ARGS__)
+#define svqdmlslbt_s64(...) __builtin_sve_svqdmlslbt_s64(__VA_ARGS__)
+#define svqdmlslbt_s16(...) __builtin_sve_svqdmlslbt_s16(__VA_ARGS__)
+#define svqdmlslt_n_s32(...) __builtin_sve_svqdmlslt_n_s32(__VA_ARGS__)
+#define svqdmlslt_n_s64(...) __builtin_sve_svqdmlslt_n_s64(__VA_ARGS__)
+#define svqdmlslt_n_s16(...) __builtin_sve_svqdmlslt_n_s16(__VA_ARGS__)
+#define svqdmlslt_s32(...) __builtin_sve_svqdmlslt_s32(__VA_ARGS__)
+#define svqdmlslt_s64(...) __builtin_sve_svqdmlslt_s64(__VA_ARGS__)
+#define svqdmlslt_s16(...) __builtin_sve_svqdmlslt_s16(__VA_ARGS__)
+#define svqdmlslt_lane_s32(...) __builtin_sve_svqdmlslt_lane_s32(__VA_ARGS__)
+#define svqdmlslt_lane_s64(...) __builtin_sve_svqdmlslt_lane_s64(__VA_ARGS__)
+#define svqdmulh_n_s8(...) __builtin_sve_svqdmulh_n_s8(__VA_ARGS__)
+#define svqdmulh_n_s32(...) __builtin_sve_svqdmulh_n_s32(__VA_ARGS__)
+#define svqdmulh_n_s64(...) __builtin_sve_svqdmulh_n_s64(__VA_ARGS__)
+#define svqdmulh_n_s16(...) __builtin_sve_svqdmulh_n_s16(__VA_ARGS__)
+#define svqdmulh_s8(...) __builtin_sve_svqdmulh_s8(__VA_ARGS__)
+#define svqdmulh_s32(...) __builtin_sve_svqdmulh_s32(__VA_ARGS__)
+#define svqdmulh_s64(...) __builtin_sve_svqdmulh_s64(__VA_ARGS__)
+#define svqdmulh_s16(...) __builtin_sve_svqdmulh_s16(__VA_ARGS__)
+#define svqdmulh_lane_s32(...) __builtin_sve_svqdmulh_lane_s32(__VA_ARGS__)
+#define svqdmulh_lane_s64(...) __builtin_sve_svqdmulh_lane_s64(__VA_ARGS__)
+#define svqdmulh_lane_s16(...) __builtin_sve_svqdmulh_lane_s16(__VA_ARGS__)
+#define svqdmullb_n_s32(...) __builtin_sve_svqdmullb_n_s32(__VA_ARGS__)
+#define svqdmullb_n_s64(...) __builtin_sve_svqdmullb_n_s64(__VA_ARGS__)
+#define svqdmullb_n_s16(...) __builtin_sve_svqdmullb_n_s16(__VA_ARGS__)
+#define svqdmullb_s32(...) __builtin_sve_svqdmullb_s32(__VA_ARGS__)
+#define svqdmullb_s64(...) __builtin_sve_svqdmullb_s64(__VA_ARGS__)
+#define svqdmullb_s16(...) __builtin_sve_svqdmullb_s16(__VA_ARGS__)
+#define svqdmullb_lane_s32(...) __builtin_sve_svqdmullb_lane_s32(__VA_ARGS__)
+#define svqdmullb_lane_s64(...) __builtin_sve_svqdmullb_lane_s64(__VA_ARGS__)
+#define svqdmullt_n_s32(...) __builtin_sve_svqdmullt_n_s32(__VA_ARGS__)
+#define svqdmullt_n_s64(...) __builtin_sve_svqdmullt_n_s64(__VA_ARGS__)
+#define svqdmullt_n_s16(...) __builtin_sve_svqdmullt_n_s16(__VA_ARGS__)
+#define svqdmullt_s32(...) __builtin_sve_svqdmullt_s32(__VA_ARGS__)
+#define svqdmullt_s64(...) __builtin_sve_svqdmullt_s64(__VA_ARGS__)
+#define svqdmullt_s16(...) __builtin_sve_svqdmullt_s16(__VA_ARGS__)
+#define svqdmullt_lane_s32(...) __builtin_sve_svqdmullt_lane_s32(__VA_ARGS__)
+#define svqdmullt_lane_s64(...) __builtin_sve_svqdmullt_lane_s64(__VA_ARGS__)
+#define svqneg_s8_m(...) __builtin_sve_svqneg_s8_m(__VA_ARGS__)
+#define svqneg_s32_m(...) __builtin_sve_svqneg_s32_m(__VA_ARGS__)
+#define svqneg_s64_m(...) __builtin_sve_svqneg_s64_m(__VA_ARGS__)
+#define svqneg_s16_m(...) __builtin_sve_svqneg_s16_m(__VA_ARGS__)
+#define svqneg_s8_x(...) __builtin_sve_svqneg_s8_x(__VA_ARGS__)
+#define svqneg_s32_x(...) __builtin_sve_svqneg_s32_x(__VA_ARGS__)
+#define svqneg_s64_x(...) __builtin_sve_svqneg_s64_x(__VA_ARGS__)
+#define svqneg_s16_x(...) __builtin_sve_svqneg_s16_x(__VA_ARGS__)
+#define svqneg_s8_z(...) __builtin_sve_svqneg_s8_z(__VA_ARGS__)
+#define svqneg_s32_z(...) __builtin_sve_svqneg_s32_z(__VA_ARGS__)
+#define svqneg_s64_z(...) __builtin_sve_svqneg_s64_z(__VA_ARGS__)
+#define svqneg_s16_z(...) __builtin_sve_svqneg_s16_z(__VA_ARGS__)
+#define svqrdcmlah_s8(...) __builtin_sve_svqrdcmlah_s8(__VA_ARGS__)
+#define svqrdcmlah_s32(...) __builtin_sve_svqrdcmlah_s32(__VA_ARGS__)
+#define svqrdcmlah_s64(...) __builtin_sve_svqrdcmlah_s64(__VA_ARGS__)
+#define svqrdcmlah_s16(...) __builtin_sve_svqrdcmlah_s16(__VA_ARGS__)
+#define svqrdcmlah_lane_s32(...) __builtin_sve_svqrdcmlah_lane_s32(__VA_ARGS__)
+#define svqrdcmlah_lane_s16(...) __builtin_sve_svqrdcmlah_lane_s16(__VA_ARGS__)
+#define svqrdmlah_n_s8(...) __builtin_sve_svqrdmlah_n_s8(__VA_ARGS__)
+#define svqrdmlah_n_s32(...) __builtin_sve_svqrdmlah_n_s32(__VA_ARGS__)
+#define svqrdmlah_n_s64(...) __builtin_sve_svqrdmlah_n_s64(__VA_ARGS__)
+#define svqrdmlah_n_s16(...) __builtin_sve_svqrdmlah_n_s16(__VA_ARGS__)
+#define svqrdmlah_s8(...) __builtin_sve_svqrdmlah_s8(__VA_ARGS__)
+#define svqrdmlah_s32(...) __builtin_sve_svqrdmlah_s32(__VA_ARGS__)
+#define svqrdmlah_s64(...) __builtin_sve_svqrdmlah_s64(__VA_ARGS__)
+#define svqrdmlah_s16(...) __builtin_sve_svqrdmlah_s16(__VA_ARGS__)
+#define svqrdmlah_lane_s32(...) __builtin_sve_svqrdmlah_lane_s32(__VA_ARGS__)
+#define svqrdmlah_lane_s64(...) __builtin_sve_svqrdmlah_lane_s64(__VA_ARGS__)
+#define svqrdmlah_lane_s16(...) __builtin_sve_svqrdmlah_lane_s16(__VA_ARGS__)
+#define svqrdmlsh_n_s8(...) __builtin_sve_svqrdmlsh_n_s8(__VA_ARGS__)
+#define svqrdmlsh_n_s32(...) __builtin_sve_svqrdmlsh_n_s32(__VA_ARGS__)
+#define svqrdmlsh_n_s64(...) __builtin_sve_svqrdmlsh_n_s64(__VA_ARGS__)
+#define svqrdmlsh_n_s16(...) __builtin_sve_svqrdmlsh_n_s16(__VA_ARGS__)
+#define svqrdmlsh_s8(...) __builtin_sve_svqrdmlsh_s8(__VA_ARGS__)
+#define svqrdmlsh_s32(...) __builtin_sve_svqrdmlsh_s32(__VA_ARGS__)
+#define svqrdmlsh_s64(...) __builtin_sve_svqrdmlsh_s64(__VA_ARGS__)
+#define svqrdmlsh_s16(...) __builtin_sve_svqrdmlsh_s16(__VA_ARGS__)
+#define svqrdmlsh_lane_s32(...) __builtin_sve_svqrdmlsh_lane_s32(__VA_ARGS__)
+#define svqrdmlsh_lane_s64(...) __builtin_sve_svqrdmlsh_lane_s64(__VA_ARGS__)
+#define svqrdmlsh_lane_s16(...) __builtin_sve_svqrdmlsh_lane_s16(__VA_ARGS__)
+#define svqrdmulh_n_s8(...) __builtin_sve_svqrdmulh_n_s8(__VA_ARGS__)
+#define svqrdmulh_n_s32(...) __builtin_sve_svqrdmulh_n_s32(__VA_ARGS__)
+#define svqrdmulh_n_s64(...) __builtin_sve_svqrdmulh_n_s64(__VA_ARGS__)
+#define svqrdmulh_n_s16(...) __builtin_sve_svqrdmulh_n_s16(__VA_ARGS__)
+#define svqrdmulh_s8(...) __builtin_sve_svqrdmulh_s8(__VA_ARGS__)
+#define svqrdmulh_s32(...) __builtin_sve_svqrdmulh_s32(__VA_ARGS__)
+#define svqrdmulh_s64(...) __builtin_sve_svqrdmulh_s64(__VA_ARGS__)
+#define svqrdmulh_s16(...) __builtin_sve_svqrdmulh_s16(__VA_ARGS__)
+#define svqrdmulh_lane_s32(...) __builtin_sve_svqrdmulh_lane_s32(__VA_ARGS__)
+#define svqrdmulh_lane_s64(...) __builtin_sve_svqrdmulh_lane_s64(__VA_ARGS__)
+#define svqrdmulh_lane_s16(...) __builtin_sve_svqrdmulh_lane_s16(__VA_ARGS__)
+#define svqrshl_n_s8_m(...) __builtin_sve_svqrshl_n_s8_m(__VA_ARGS__)
+#define svqrshl_n_s32_m(...) __builtin_sve_svqrshl_n_s32_m(__VA_ARGS__)
+#define svqrshl_n_s64_m(...) __builtin_sve_svqrshl_n_s64_m(__VA_ARGS__)
+#define svqrshl_n_s16_m(...) __builtin_sve_svqrshl_n_s16_m(__VA_ARGS__)
+#define svqrshl_n_s8_x(...) __builtin_sve_svqrshl_n_s8_x(__VA_ARGS__)
+#define svqrshl_n_s32_x(...) __builtin_sve_svqrshl_n_s32_x(__VA_ARGS__)
+#define svqrshl_n_s64_x(...) __builtin_sve_svqrshl_n_s64_x(__VA_ARGS__)
+#define svqrshl_n_s16_x(...) __builtin_sve_svqrshl_n_s16_x(__VA_ARGS__)
+#define svqrshl_n_s8_z(...) __builtin_sve_svqrshl_n_s8_z(__VA_ARGS__)
+#define svqrshl_n_s32_z(...) __builtin_sve_svqrshl_n_s32_z(__VA_ARGS__)
+#define svqrshl_n_s64_z(...) __builtin_sve_svqrshl_n_s64_z(__VA_ARGS__)
+#define svqrshl_n_s16_z(...) __builtin_sve_svqrshl_n_s16_z(__VA_ARGS__)
+#define svqrshl_n_u8_m(...) __builtin_sve_svqrshl_n_u8_m(__VA_ARGS__)
+#define svqrshl_n_u32_m(...) __builtin_sve_svqrshl_n_u32_m(__VA_ARGS__)
+#define svqrshl_n_u64_m(...) __builtin_sve_svqrshl_n_u64_m(__VA_ARGS__)
+#define svqrshl_n_u16_m(...) __builtin_sve_svqrshl_n_u16_m(__VA_ARGS__)
+#define svqrshl_n_u8_x(...) __builtin_sve_svqrshl_n_u8_x(__VA_ARGS__)
+#define svqrshl_n_u32_x(...) __builtin_sve_svqrshl_n_u32_x(__VA_ARGS__)
+#define svqrshl_n_u64_x(...) __builtin_sve_svqrshl_n_u64_x(__VA_ARGS__)
+#define svqrshl_n_u16_x(...) __builtin_sve_svqrshl_n_u16_x(__VA_ARGS__)
+#define svqrshl_n_u8_z(...) __builtin_sve_svqrshl_n_u8_z(__VA_ARGS__)
+#define svqrshl_n_u32_z(...) __builtin_sve_svqrshl_n_u32_z(__VA_ARGS__)
+#define svqrshl_n_u64_z(...) __builtin_sve_svqrshl_n_u64_z(__VA_ARGS__)
+#define svqrshl_n_u16_z(...) __builtin_sve_svqrshl_n_u16_z(__VA_ARGS__)
+#define svqrshl_s8_m(...) __builtin_sve_svqrshl_s8_m(__VA_ARGS__)
+#define svqrshl_s32_m(...) __builtin_sve_svqrshl_s32_m(__VA_ARGS__)
+#define svqrshl_s64_m(...) __builtin_sve_svqrshl_s64_m(__VA_ARGS__)
+#define svqrshl_s16_m(...) __builtin_sve_svqrshl_s16_m(__VA_ARGS__)
+#define svqrshl_s8_x(...) __builtin_sve_svqrshl_s8_x(__VA_ARGS__)
+#define svqrshl_s32_x(...) __builtin_sve_svqrshl_s32_x(__VA_ARGS__)
+#define svqrshl_s64_x(...) __builtin_sve_svqrshl_s64_x(__VA_ARGS__)
+#define svqrshl_s16_x(...) __builtin_sve_svqrshl_s16_x(__VA_ARGS__)
+#define svqrshl_s8_z(...) __builtin_sve_svqrshl_s8_z(__VA_ARGS__)
+#define svqrshl_s32_z(...) __builtin_sve_svqrshl_s32_z(__VA_ARGS__)
+#define svqrshl_s64_z(...) __builtin_sve_svqrshl_s64_z(__VA_ARGS__)
+#define svqrshl_s16_z(...) __builtin_sve_svqrshl_s16_z(__VA_ARGS__)
+#define svqrshl_u8_m(...) __builtin_sve_svqrshl_u8_m(__VA_ARGS__)
+#define svqrshl_u32_m(...) __builtin_sve_svqrshl_u32_m(__VA_ARGS__)
+#define svqrshl_u64_m(...) __builtin_sve_svqrshl_u64_m(__VA_ARGS__)
+#define svqrshl_u16_m(...) __builtin_sve_svqrshl_u16_m(__VA_ARGS__)
+#define svqrshl_u8_x(...) __builtin_sve_svqrshl_u8_x(__VA_ARGS__)
+#define svqrshl_u32_x(...) __builtin_sve_svqrshl_u32_x(__VA_ARGS__)
+#define svqrshl_u64_x(...) __builtin_sve_svqrshl_u64_x(__VA_ARGS__)
+#define svqrshl_u16_x(...) __builtin_sve_svqrshl_u16_x(__VA_ARGS__)
+#define svqrshl_u8_z(...) __builtin_sve_svqrshl_u8_z(__VA_ARGS__)
+#define svqrshl_u32_z(...) __builtin_sve_svqrshl_u32_z(__VA_ARGS__)
+#define svqrshl_u64_z(...) __builtin_sve_svqrshl_u64_z(__VA_ARGS__)
+#define svqrshl_u16_z(...) __builtin_sve_svqrshl_u16_z(__VA_ARGS__)
+#define svqrshrnb_n_s32(...) __builtin_sve_svqrshrnb_n_s32(__VA_ARGS__)
+#define svqrshrnb_n_s64(...) __builtin_sve_svqrshrnb_n_s64(__VA_ARGS__)
+#define svqrshrnb_n_s16(...) __builtin_sve_svqrshrnb_n_s16(__VA_ARGS__)
+#define svqrshrnb_n_u32(...) __builtin_sve_svqrshrnb_n_u32(__VA_ARGS__)
+#define svqrshrnb_n_u64(...) __builtin_sve_svqrshrnb_n_u64(__VA_ARGS__)
+#define svqrshrnb_n_u16(...) __builtin_sve_svqrshrnb_n_u16(__VA_ARGS__)
+#define svqrshrnt_n_s32(...) __builtin_sve_svqrshrnt_n_s32(__VA_ARGS__)
+#define svqrshrnt_n_s64(...) __builtin_sve_svqrshrnt_n_s64(__VA_ARGS__)
+#define svqrshrnt_n_s16(...) __builtin_sve_svqrshrnt_n_s16(__VA_ARGS__)
+#define svqrshrnt_n_u32(...) __builtin_sve_svqrshrnt_n_u32(__VA_ARGS__)
+#define svqrshrnt_n_u64(...) __builtin_sve_svqrshrnt_n_u64(__VA_ARGS__)
+#define svqrshrnt_n_u16(...) __builtin_sve_svqrshrnt_n_u16(__VA_ARGS__)
+#define svqrshrunb_n_s32(...) __builtin_sve_svqrshrunb_n_s32(__VA_ARGS__)
+#define svqrshrunb_n_s64(...) __builtin_sve_svqrshrunb_n_s64(__VA_ARGS__)
+#define svqrshrunb_n_s16(...) __builtin_sve_svqrshrunb_n_s16(__VA_ARGS__)
+#define svqrshrunt_n_s32(...) __builtin_sve_svqrshrunt_n_s32(__VA_ARGS__)
+#define svqrshrunt_n_s64(...) __builtin_sve_svqrshrunt_n_s64(__VA_ARGS__)
+#define svqrshrunt_n_s16(...) __builtin_sve_svqrshrunt_n_s16(__VA_ARGS__)
+#define svqshl_n_s8_m(...) __builtin_sve_svqshl_n_s8_m(__VA_ARGS__)
+#define svqshl_n_s32_m(...) __builtin_sve_svqshl_n_s32_m(__VA_ARGS__)
+#define svqshl_n_s64_m(...) __builtin_sve_svqshl_n_s64_m(__VA_ARGS__)
+#define svqshl_n_s16_m(...) __builtin_sve_svqshl_n_s16_m(__VA_ARGS__)
+#define svqshl_n_s8_x(...) __builtin_sve_svqshl_n_s8_x(__VA_ARGS__)
+#define svqshl_n_s32_x(...) __builtin_sve_svqshl_n_s32_x(__VA_ARGS__)
+#define svqshl_n_s64_x(...) __builtin_sve_svqshl_n_s64_x(__VA_ARGS__)
+#define svqshl_n_s16_x(...) __builtin_sve_svqshl_n_s16_x(__VA_ARGS__)
+#define svqshl_n_s8_z(...) __builtin_sve_svqshl_n_s8_z(__VA_ARGS__)
+#define svqshl_n_s32_z(...) __builtin_sve_svqshl_n_s32_z(__VA_ARGS__)
+#define svqshl_n_s64_z(...) __builtin_sve_svqshl_n_s64_z(__VA_ARGS__)
+#define svqshl_n_s16_z(...) __builtin_sve_svqshl_n_s16_z(__VA_ARGS__)
+#define svqshl_n_u8_m(...) __builtin_sve_svqshl_n_u8_m(__VA_ARGS__)
+#define svqshl_n_u32_m(...) __builtin_sve_svqshl_n_u32_m(__VA_ARGS__)
+#define svqshl_n_u64_m(...) __builtin_sve_svqshl_n_u64_m(__VA_ARGS__)
+#define svqshl_n_u16_m(...) __builtin_sve_svqshl_n_u16_m(__VA_ARGS__)
+#define svqshl_n_u8_x(...) __builtin_sve_svqshl_n_u8_x(__VA_ARGS__)
+#define svqshl_n_u32_x(...) __builtin_sve_svqshl_n_u32_x(__VA_ARGS__)
+#define svqshl_n_u64_x(...) __builtin_sve_svqshl_n_u64_x(__VA_ARGS__)
+#define svqshl_n_u16_x(...) __builtin_sve_svqshl_n_u16_x(__VA_ARGS__)
+#define svqshl_n_u8_z(...) __builtin_sve_svqshl_n_u8_z(__VA_ARGS__)
+#define svqshl_n_u32_z(...) __builtin_sve_svqshl_n_u32_z(__VA_ARGS__)
+#define svqshl_n_u64_z(...) __builtin_sve_svqshl_n_u64_z(__VA_ARGS__)
+#define svqshl_n_u16_z(...) __builtin_sve_svqshl_n_u16_z(__VA_ARGS__)
+#define svqshl_s8_m(...) __builtin_sve_svqshl_s8_m(__VA_ARGS__)
+#define svqshl_s32_m(...) __builtin_sve_svqshl_s32_m(__VA_ARGS__)
+#define svqshl_s64_m(...) __builtin_sve_svqshl_s64_m(__VA_ARGS__)
+#define svqshl_s16_m(...) __builtin_sve_svqshl_s16_m(__VA_ARGS__)
+#define svqshl_s8_x(...) __builtin_sve_svqshl_s8_x(__VA_ARGS__)
+#define svqshl_s32_x(...) __builtin_sve_svqshl_s32_x(__VA_ARGS__)
+#define svqshl_s64_x(...) __builtin_sve_svqshl_s64_x(__VA_ARGS__)
+#define svqshl_s16_x(...) __builtin_sve_svqshl_s16_x(__VA_ARGS__)
+#define svqshl_s8_z(...) __builtin_sve_svqshl_s8_z(__VA_ARGS__)
+#define svqshl_s32_z(...) __builtin_sve_svqshl_s32_z(__VA_ARGS__)
+#define svqshl_s64_z(...) __builtin_sve_svqshl_s64_z(__VA_ARGS__)
+#define svqshl_s16_z(...) __builtin_sve_svqshl_s16_z(__VA_ARGS__)
+#define svqshl_u8_m(...) __builtin_sve_svqshl_u8_m(__VA_ARGS__)
+#define svqshl_u32_m(...) __builtin_sve_svqshl_u32_m(__VA_ARGS__)
+#define svqshl_u64_m(...) __builtin_sve_svqshl_u64_m(__VA_ARGS__)
+#define svqshl_u16_m(...) __builtin_sve_svqshl_u16_m(__VA_ARGS__)
+#define svqshl_u8_x(...) __builtin_sve_svqshl_u8_x(__VA_ARGS__)
+#define svqshl_u32_x(...) __builtin_sve_svqshl_u32_x(__VA_ARGS__)
+#define svqshl_u64_x(...) __builtin_sve_svqshl_u64_x(__VA_ARGS__)
+#define svqshl_u16_x(...) __builtin_sve_svqshl_u16_x(__VA_ARGS__)
+#define svqshl_u8_z(...) __builtin_sve_svqshl_u8_z(__VA_ARGS__)
+#define svqshl_u32_z(...) __builtin_sve_svqshl_u32_z(__VA_ARGS__)
+#define svqshl_u64_z(...) __builtin_sve_svqshl_u64_z(__VA_ARGS__)
+#define svqshl_u16_z(...) __builtin_sve_svqshl_u16_z(__VA_ARGS__)
+#define svqshlu_n_s8_m(...) __builtin_sve_svqshlu_n_s8_m(__VA_ARGS__)
+#define svqshlu_n_s32_m(...) __builtin_sve_svqshlu_n_s32_m(__VA_ARGS__)
+#define svqshlu_n_s64_m(...) __builtin_sve_svqshlu_n_s64_m(__VA_ARGS__)
+#define svqshlu_n_s16_m(...) __builtin_sve_svqshlu_n_s16_m(__VA_ARGS__)
+#define svqshlu_n_s8_x(...) __builtin_sve_svqshlu_n_s8_x(__VA_ARGS__)
+#define svqshlu_n_s32_x(...) __builtin_sve_svqshlu_n_s32_x(__VA_ARGS__)
+#define svqshlu_n_s64_x(...) __builtin_sve_svqshlu_n_s64_x(__VA_ARGS__)
+#define svqshlu_n_s16_x(...) __builtin_sve_svqshlu_n_s16_x(__VA_ARGS__)
+#define svqshlu_n_s8_z(...) __builtin_sve_svqshlu_n_s8_z(__VA_ARGS__)
+#define svqshlu_n_s32_z(...) __builtin_sve_svqshlu_n_s32_z(__VA_ARGS__)
+#define svqshlu_n_s64_z(...) __builtin_sve_svqshlu_n_s64_z(__VA_ARGS__)
+#define svqshlu_n_s16_z(...) __builtin_sve_svqshlu_n_s16_z(__VA_ARGS__)
+#define svqshrnb_n_s32(...) __builtin_sve_svqshrnb_n_s32(__VA_ARGS__)
+#define svqshrnb_n_s64(...) __builtin_sve_svqshrnb_n_s64(__VA_ARGS__)
+#define svqshrnb_n_s16(...) __builtin_sve_svqshrnb_n_s16(__VA_ARGS__)
+#define svqshrnb_n_u32(...) __builtin_sve_svqshrnb_n_u32(__VA_ARGS__)
+#define svqshrnb_n_u64(...) __builtin_sve_svqshrnb_n_u64(__VA_ARGS__)
+#define svqshrnb_n_u16(...) __builtin_sve_svqshrnb_n_u16(__VA_ARGS__)
+#define svqshrnt_n_s32(...) __builtin_sve_svqshrnt_n_s32(__VA_ARGS__)
+#define svqshrnt_n_s64(...) __builtin_sve_svqshrnt_n_s64(__VA_ARGS__)
+#define svqshrnt_n_s16(...) __builtin_sve_svqshrnt_n_s16(__VA_ARGS__)
+#define svqshrnt_n_u32(...) __builtin_sve_svqshrnt_n_u32(__VA_ARGS__)
+#define svqshrnt_n_u64(...) __builtin_sve_svqshrnt_n_u64(__VA_ARGS__)
+#define svqshrnt_n_u16(...) __builtin_sve_svqshrnt_n_u16(__VA_ARGS__)
+#define svqshrunb_n_s32(...) __builtin_sve_svqshrunb_n_s32(__VA_ARGS__)
+#define svqshrunb_n_s64(...) __builtin_sve_svqshrunb_n_s64(__VA_ARGS__)
+#define svqshrunb_n_s16(...) __builtin_sve_svqshrunb_n_s16(__VA_ARGS__)
+#define svqshrunt_n_s32(...) __builtin_sve_svqshrunt_n_s32(__VA_ARGS__)
+#define svqshrunt_n_s64(...) __builtin_sve_svqshrunt_n_s64(__VA_ARGS__)
+#define svqshrunt_n_s16(...) __builtin_sve_svqshrunt_n_s16(__VA_ARGS__)
+#define svqsub_n_s8_m(...) __builtin_sve_svqsub_n_s8_m(__VA_ARGS__)
+#define svqsub_n_s32_m(...) __builtin_sve_svqsub_n_s32_m(__VA_ARGS__)
+#define svqsub_n_s64_m(...) __builtin_sve_svqsub_n_s64_m(__VA_ARGS__)
+#define svqsub_n_s16_m(...) __builtin_sve_svqsub_n_s16_m(__VA_ARGS__)
+#define svqsub_n_s8_x(...) __builtin_sve_svqsub_n_s8_x(__VA_ARGS__)
+#define svqsub_n_s32_x(...) __builtin_sve_svqsub_n_s32_x(__VA_ARGS__)
+#define svqsub_n_s64_x(...) __builtin_sve_svqsub_n_s64_x(__VA_ARGS__)
+#define svqsub_n_s16_x(...) __builtin_sve_svqsub_n_s16_x(__VA_ARGS__)
+#define svqsub_n_s8_z(...) __builtin_sve_svqsub_n_s8_z(__VA_ARGS__)
+#define svqsub_n_s32_z(...) __builtin_sve_svqsub_n_s32_z(__VA_ARGS__)
+#define svqsub_n_s64_z(...) __builtin_sve_svqsub_n_s64_z(__VA_ARGS__)
+#define svqsub_n_s16_z(...) __builtin_sve_svqsub_n_s16_z(__VA_ARGS__)
+#define svqsub_n_u8_m(...) __builtin_sve_svqsub_n_u8_m(__VA_ARGS__)
+#define svqsub_n_u32_m(...) __builtin_sve_svqsub_n_u32_m(__VA_ARGS__)
+#define svqsub_n_u64_m(...) __builtin_sve_svqsub_n_u64_m(__VA_ARGS__)
+#define svqsub_n_u16_m(...) __builtin_sve_svqsub_n_u16_m(__VA_ARGS__)
+#define svqsub_n_u8_x(...) __builtin_sve_svqsub_n_u8_x(__VA_ARGS__)
+#define svqsub_n_u32_x(...) __builtin_sve_svqsub_n_u32_x(__VA_ARGS__)
+#define svqsub_n_u64_x(...) __builtin_sve_svqsub_n_u64_x(__VA_ARGS__)
+#define svqsub_n_u16_x(...) __builtin_sve_svqsub_n_u16_x(__VA_ARGS__)
+#define svqsub_n_u8_z(...) __builtin_sve_svqsub_n_u8_z(__VA_ARGS__)
+#define svqsub_n_u32_z(...) __builtin_sve_svqsub_n_u32_z(__VA_ARGS__)
+#define svqsub_n_u64_z(...) __builtin_sve_svqsub_n_u64_z(__VA_ARGS__)
+#define svqsub_n_u16_z(...) __builtin_sve_svqsub_n_u16_z(__VA_ARGS__)
+#define svqsub_s8_m(...) __builtin_sve_svqsub_s8_m(__VA_ARGS__)
+#define svqsub_s32_m(...) __builtin_sve_svqsub_s32_m(__VA_ARGS__)
+#define svqsub_s64_m(...) __builtin_sve_svqsub_s64_m(__VA_ARGS__)
+#define svqsub_s16_m(...) __builtin_sve_svqsub_s16_m(__VA_ARGS__)
+#define svqsub_s8_x(...) __builtin_sve_svqsub_s8_x(__VA_ARGS__)
+#define svqsub_s32_x(...) __builtin_sve_svqsub_s32_x(__VA_ARGS__)
+#define svqsub_s64_x(...) __builtin_sve_svqsub_s64_x(__VA_ARGS__)
+#define svqsub_s16_x(...) __builtin_sve_svqsub_s16_x(__VA_ARGS__)
+#define svqsub_s8_z(...) __builtin_sve_svqsub_s8_z(__VA_ARGS__)
+#define svqsub_s32_z(...) __builtin_sve_svqsub_s32_z(__VA_ARGS__)
+#define svqsub_s64_z(...) __builtin_sve_svqsub_s64_z(__VA_ARGS__)
+#define svqsub_s16_z(...) __builtin_sve_svqsub_s16_z(__VA_ARGS__)
+#define svqsub_u8_m(...) __builtin_sve_svqsub_u8_m(__VA_ARGS__)
+#define svqsub_u32_m(...) __builtin_sve_svqsub_u32_m(__VA_ARGS__)
+#define svqsub_u64_m(...) __builtin_sve_svqsub_u64_m(__VA_ARGS__)
+#define svqsub_u16_m(...) __builtin_sve_svqsub_u16_m(__VA_ARGS__)
+#define svqsub_u8_x(...) __builtin_sve_svqsub_u8_x(__VA_ARGS__)
+#define svqsub_u32_x(...) __builtin_sve_svqsub_u32_x(__VA_ARGS__)
+#define svqsub_u64_x(...) __builtin_sve_svqsub_u64_x(__VA_ARGS__)
+#define svqsub_u16_x(...) __builtin_sve_svqsub_u16_x(__VA_ARGS__)
+#define svqsub_u8_z(...) __builtin_sve_svqsub_u8_z(__VA_ARGS__)
+#define svqsub_u32_z(...) __builtin_sve_svqsub_u32_z(__VA_ARGS__)
+#define svqsub_u64_z(...) __builtin_sve_svqsub_u64_z(__VA_ARGS__)
+#define svqsub_u16_z(...) __builtin_sve_svqsub_u16_z(__VA_ARGS__)
+#define svqsubr_n_s8_m(...) __builtin_sve_svqsubr_n_s8_m(__VA_ARGS__)
+#define svqsubr_n_s32_m(...) __builtin_sve_svqsubr_n_s32_m(__VA_ARGS__)
+#define svqsubr_n_s64_m(...) __builtin_sve_svqsubr_n_s64_m(__VA_ARGS__)
+#define svqsubr_n_s16_m(...) __builtin_sve_svqsubr_n_s16_m(__VA_ARGS__)
+#define svqsubr_n_s8_x(...) __builtin_sve_svqsubr_n_s8_x(__VA_ARGS__)
+#define svqsubr_n_s32_x(...) __builtin_sve_svqsubr_n_s32_x(__VA_ARGS__)
+#define svqsubr_n_s64_x(...) __builtin_sve_svqsubr_n_s64_x(__VA_ARGS__)
+#define svqsubr_n_s16_x(...) __builtin_sve_svqsubr_n_s16_x(__VA_ARGS__)
+#define svqsubr_n_s8_z(...) __builtin_sve_svqsubr_n_s8_z(__VA_ARGS__)
+#define svqsubr_n_s32_z(...) __builtin_sve_svqsubr_n_s32_z(__VA_ARGS__)
+#define svqsubr_n_s64_z(...) __builtin_sve_svqsubr_n_s64_z(__VA_ARGS__)
+#define svqsubr_n_s16_z(...) __builtin_sve_svqsubr_n_s16_z(__VA_ARGS__)
+#define svqsubr_n_u8_m(...) __builtin_sve_svqsubr_n_u8_m(__VA_ARGS__)
+#define svqsubr_n_u32_m(...) __builtin_sve_svqsubr_n_u32_m(__VA_ARGS__)
+#define svqsubr_n_u64_m(...) __builtin_sve_svqsubr_n_u64_m(__VA_ARGS__)
+#define svqsubr_n_u16_m(...) __builtin_sve_svqsubr_n_u16_m(__VA_ARGS__)
+#define svqsubr_n_u8_x(...) __builtin_sve_svqsubr_n_u8_x(__VA_ARGS__)
+#define svqsubr_n_u32_x(...) __builtin_sve_svqsubr_n_u32_x(__VA_ARGS__)
+#define svqsubr_n_u64_x(...) __builtin_sve_svqsubr_n_u64_x(__VA_ARGS__)
+#define svqsubr_n_u16_x(...) __builtin_sve_svqsubr_n_u16_x(__VA_ARGS__)
+#define svqsubr_n_u8_z(...) __builtin_sve_svqsubr_n_u8_z(__VA_ARGS__)
+#define svqsubr_n_u32_z(...) __builtin_sve_svqsubr_n_u32_z(__VA_ARGS__)
+#define svqsubr_n_u64_z(...) __builtin_sve_svqsubr_n_u64_z(__VA_ARGS__)
+#define svqsubr_n_u16_z(...) __builtin_sve_svqsubr_n_u16_z(__VA_ARGS__)
+#define svqsubr_s8_m(...) __builtin_sve_svqsubr_s8_m(__VA_ARGS__)
+#define svqsubr_s32_m(...) __builtin_sve_svqsubr_s32_m(__VA_ARGS__)
+#define svqsubr_s64_m(...) __builtin_sve_svqsubr_s64_m(__VA_ARGS__)
+#define svqsubr_s16_m(...) __builtin_sve_svqsubr_s16_m(__VA_ARGS__)
+#define svqsubr_s8_x(...) __builtin_sve_svqsubr_s8_x(__VA_ARGS__)
+#define svqsubr_s32_x(...) __builtin_sve_svqsubr_s32_x(__VA_ARGS__)
+#define svqsubr_s64_x(...) __builtin_sve_svqsubr_s64_x(__VA_ARGS__)
+#define svqsubr_s16_x(...) __builtin_sve_svqsubr_s16_x(__VA_ARGS__)
+#define svqsubr_s8_z(...) __builtin_sve_svqsubr_s8_z(__VA_ARGS__)
+#define svqsubr_s32_z(...) __builtin_sve_svqsubr_s32_z(__VA_ARGS__)
+#define svqsubr_s64_z(...) __builtin_sve_svqsubr_s64_z(__VA_ARGS__)
+#define svqsubr_s16_z(...) __builtin_sve_svqsubr_s16_z(__VA_ARGS__)
+#define svqsubr_u8_m(...) __builtin_sve_svqsubr_u8_m(__VA_ARGS__)
+#define svqsubr_u32_m(...) __builtin_sve_svqsubr_u32_m(__VA_ARGS__)
+#define svqsubr_u64_m(...) __builtin_sve_svqsubr_u64_m(__VA_ARGS__)
+#define svqsubr_u16_m(...) __builtin_sve_svqsubr_u16_m(__VA_ARGS__)
+#define svqsubr_u8_x(...) __builtin_sve_svqsubr_u8_x(__VA_ARGS__)
+#define svqsubr_u32_x(...) __builtin_sve_svqsubr_u32_x(__VA_ARGS__)
+#define svqsubr_u64_x(...) __builtin_sve_svqsubr_u64_x(__VA_ARGS__)
+#define svqsubr_u16_x(...) __builtin_sve_svqsubr_u16_x(__VA_ARGS__)
+#define svqsubr_u8_z(...) __builtin_sve_svqsubr_u8_z(__VA_ARGS__)
+#define svqsubr_u32_z(...) __builtin_sve_svqsubr_u32_z(__VA_ARGS__)
+#define svqsubr_u64_z(...) __builtin_sve_svqsubr_u64_z(__VA_ARGS__)
+#define svqsubr_u16_z(...) __builtin_sve_svqsubr_u16_z(__VA_ARGS__)
+#define svqxtnb_s32(...) __builtin_sve_svqxtnb_s32(__VA_ARGS__)
+#define svqxtnb_s64(...) __builtin_sve_svqxtnb_s64(__VA_ARGS__)
+#define svqxtnb_s16(...) __builtin_sve_svqxtnb_s16(__VA_ARGS__)
+#define svqxtnb_u32(...) __builtin_sve_svqxtnb_u32(__VA_ARGS__)
+#define svqxtnb_u64(...) __builtin_sve_svqxtnb_u64(__VA_ARGS__)
+#define svqxtnb_u16(...) __builtin_sve_svqxtnb_u16(__VA_ARGS__)
+#define svqxtnt_s32(...) __builtin_sve_svqxtnt_s32(__VA_ARGS__)
+#define svqxtnt_s64(...) __builtin_sve_svqxtnt_s64(__VA_ARGS__)
+#define svqxtnt_s16(...) __builtin_sve_svqxtnt_s16(__VA_ARGS__)
+#define svqxtnt_u32(...) __builtin_sve_svqxtnt_u32(__VA_ARGS__)
+#define svqxtnt_u64(...) __builtin_sve_svqxtnt_u64(__VA_ARGS__)
+#define svqxtnt_u16(...) __builtin_sve_svqxtnt_u16(__VA_ARGS__)
+#define svqxtunb_s32(...) __builtin_sve_svqxtunb_s32(__VA_ARGS__)
+#define svqxtunb_s64(...) __builtin_sve_svqxtunb_s64(__VA_ARGS__)
+#define svqxtunb_s16(...) __builtin_sve_svqxtunb_s16(__VA_ARGS__)
+#define svqxtunt_s32(...) __builtin_sve_svqxtunt_s32(__VA_ARGS__)
+#define svqxtunt_s64(...) __builtin_sve_svqxtunt_s64(__VA_ARGS__)
+#define svqxtunt_s16(...) __builtin_sve_svqxtunt_s16(__VA_ARGS__)
+#define svraddhnb_n_u32(...) __builtin_sve_svraddhnb_n_u32(__VA_ARGS__)
+#define svraddhnb_n_u64(...) __builtin_sve_svraddhnb_n_u64(__VA_ARGS__)
+#define svraddhnb_n_u16(...) __builtin_sve_svraddhnb_n_u16(__VA_ARGS__)
+#define svraddhnb_n_s32(...) __builtin_sve_svraddhnb_n_s32(__VA_ARGS__)
+#define svraddhnb_n_s64(...) __builtin_sve_svraddhnb_n_s64(__VA_ARGS__)
+#define svraddhnb_n_s16(...) __builtin_sve_svraddhnb_n_s16(__VA_ARGS__)
+#define svraddhnb_u32(...) __builtin_sve_svraddhnb_u32(__VA_ARGS__)
+#define svraddhnb_u64(...) __builtin_sve_svraddhnb_u64(__VA_ARGS__)
+#define svraddhnb_u16(...) __builtin_sve_svraddhnb_u16(__VA_ARGS__)
+#define svraddhnb_s32(...) __builtin_sve_svraddhnb_s32(__VA_ARGS__)
+#define svraddhnb_s64(...) __builtin_sve_svraddhnb_s64(__VA_ARGS__)
+#define svraddhnb_s16(...) __builtin_sve_svraddhnb_s16(__VA_ARGS__)
+#define svraddhnt_n_u32(...) __builtin_sve_svraddhnt_n_u32(__VA_ARGS__)
+#define svraddhnt_n_u64(...) __builtin_sve_svraddhnt_n_u64(__VA_ARGS__)
+#define svraddhnt_n_u16(...) __builtin_sve_svraddhnt_n_u16(__VA_ARGS__)
+#define svraddhnt_n_s32(...) __builtin_sve_svraddhnt_n_s32(__VA_ARGS__)
+#define svraddhnt_n_s64(...) __builtin_sve_svraddhnt_n_s64(__VA_ARGS__)
+#define svraddhnt_n_s16(...) __builtin_sve_svraddhnt_n_s16(__VA_ARGS__)
+#define svraddhnt_u32(...) __builtin_sve_svraddhnt_u32(__VA_ARGS__)
+#define svraddhnt_u64(...) __builtin_sve_svraddhnt_u64(__VA_ARGS__)
+#define svraddhnt_u16(...) __builtin_sve_svraddhnt_u16(__VA_ARGS__)
+#define svraddhnt_s32(...) __builtin_sve_svraddhnt_s32(__VA_ARGS__)
+#define svraddhnt_s64(...) __builtin_sve_svraddhnt_s64(__VA_ARGS__)
+#define svraddhnt_s16(...) __builtin_sve_svraddhnt_s16(__VA_ARGS__)
+#define svrecpe_u32_m(...) __builtin_sve_svrecpe_u32_m(__VA_ARGS__)
+#define svrecpe_u32_x(...) __builtin_sve_svrecpe_u32_x(__VA_ARGS__)
+#define svrecpe_u32_z(...) __builtin_sve_svrecpe_u32_z(__VA_ARGS__)
+#define svrhadd_n_s8_m(...) __builtin_sve_svrhadd_n_s8_m(__VA_ARGS__)
+#define svrhadd_n_s32_m(...) __builtin_sve_svrhadd_n_s32_m(__VA_ARGS__)
+#define svrhadd_n_s64_m(...) __builtin_sve_svrhadd_n_s64_m(__VA_ARGS__)
+#define svrhadd_n_s16_m(...) __builtin_sve_svrhadd_n_s16_m(__VA_ARGS__)
+#define svrhadd_n_s8_x(...) __builtin_sve_svrhadd_n_s8_x(__VA_ARGS__)
+#define svrhadd_n_s32_x(...) __builtin_sve_svrhadd_n_s32_x(__VA_ARGS__)
+#define svrhadd_n_s64_x(...) __builtin_sve_svrhadd_n_s64_x(__VA_ARGS__)
+#define svrhadd_n_s16_x(...) __builtin_sve_svrhadd_n_s16_x(__VA_ARGS__)
+#define svrhadd_n_s8_z(...) __builtin_sve_svrhadd_n_s8_z(__VA_ARGS__)
+#define svrhadd_n_s32_z(...) __builtin_sve_svrhadd_n_s32_z(__VA_ARGS__)
+#define svrhadd_n_s64_z(...) __builtin_sve_svrhadd_n_s64_z(__VA_ARGS__)
+#define svrhadd_n_s16_z(...) __builtin_sve_svrhadd_n_s16_z(__VA_ARGS__)
+#define svrhadd_n_u8_m(...) __builtin_sve_svrhadd_n_u8_m(__VA_ARGS__)
+#define svrhadd_n_u32_m(...) __builtin_sve_svrhadd_n_u32_m(__VA_ARGS__)
+#define svrhadd_n_u64_m(...) __builtin_sve_svrhadd_n_u64_m(__VA_ARGS__)
+#define svrhadd_n_u16_m(...) __builtin_sve_svrhadd_n_u16_m(__VA_ARGS__)
+#define svrhadd_n_u8_x(...) __builtin_sve_svrhadd_n_u8_x(__VA_ARGS__)
+#define svrhadd_n_u32_x(...) __builtin_sve_svrhadd_n_u32_x(__VA_ARGS__)
+#define svrhadd_n_u64_x(...) __builtin_sve_svrhadd_n_u64_x(__VA_ARGS__)
+#define svrhadd_n_u16_x(...) __builtin_sve_svrhadd_n_u16_x(__VA_ARGS__)
+#define svrhadd_n_u8_z(...) __builtin_sve_svrhadd_n_u8_z(__VA_ARGS__)
+#define svrhadd_n_u32_z(...) __builtin_sve_svrhadd_n_u32_z(__VA_ARGS__)
+#define svrhadd_n_u64_z(...) __builtin_sve_svrhadd_n_u64_z(__VA_ARGS__)
+#define svrhadd_n_u16_z(...) __builtin_sve_svrhadd_n_u16_z(__VA_ARGS__)
+#define svrhadd_s8_m(...) __builtin_sve_svrhadd_s8_m(__VA_ARGS__)
+#define svrhadd_s32_m(...) __builtin_sve_svrhadd_s32_m(__VA_ARGS__)
+#define svrhadd_s64_m(...) __builtin_sve_svrhadd_s64_m(__VA_ARGS__)
+#define svrhadd_s16_m(...) __builtin_sve_svrhadd_s16_m(__VA_ARGS__)
+#define svrhadd_s8_x(...) __builtin_sve_svrhadd_s8_x(__VA_ARGS__)
+#define svrhadd_s32_x(...) __builtin_sve_svrhadd_s32_x(__VA_ARGS__)
+#define svrhadd_s64_x(...) __builtin_sve_svrhadd_s64_x(__VA_ARGS__)
+#define svrhadd_s16_x(...) __builtin_sve_svrhadd_s16_x(__VA_ARGS__)
+#define svrhadd_s8_z(...) __builtin_sve_svrhadd_s8_z(__VA_ARGS__)
+#define svrhadd_s32_z(...) __builtin_sve_svrhadd_s32_z(__VA_ARGS__)
+#define svrhadd_s64_z(...) __builtin_sve_svrhadd_s64_z(__VA_ARGS__)
+#define svrhadd_s16_z(...) __builtin_sve_svrhadd_s16_z(__VA_ARGS__)
+#define svrhadd_u8_m(...) __builtin_sve_svrhadd_u8_m(__VA_ARGS__)
+#define svrhadd_u32_m(...) __builtin_sve_svrhadd_u32_m(__VA_ARGS__)
+#define svrhadd_u64_m(...) __builtin_sve_svrhadd_u64_m(__VA_ARGS__)
+#define svrhadd_u16_m(...) __builtin_sve_svrhadd_u16_m(__VA_ARGS__)
+#define svrhadd_u8_x(...) __builtin_sve_svrhadd_u8_x(__VA_ARGS__)
+#define svrhadd_u32_x(...) __builtin_sve_svrhadd_u32_x(__VA_ARGS__)
+#define svrhadd_u64_x(...) __builtin_sve_svrhadd_u64_x(__VA_ARGS__)
+#define svrhadd_u16_x(...) __builtin_sve_svrhadd_u16_x(__VA_ARGS__)
+#define svrhadd_u8_z(...) __builtin_sve_svrhadd_u8_z(__VA_ARGS__)
+#define svrhadd_u32_z(...) __builtin_sve_svrhadd_u32_z(__VA_ARGS__)
+#define svrhadd_u64_z(...) __builtin_sve_svrhadd_u64_z(__VA_ARGS__)
+#define svrhadd_u16_z(...) __builtin_sve_svrhadd_u16_z(__VA_ARGS__)
+#define svrshl_n_s8_m(...) __builtin_sve_svrshl_n_s8_m(__VA_ARGS__)
+#define svrshl_n_s32_m(...) __builtin_sve_svrshl_n_s32_m(__VA_ARGS__)
+#define svrshl_n_s64_m(...) __builtin_sve_svrshl_n_s64_m(__VA_ARGS__)
+#define svrshl_n_s16_m(...) __builtin_sve_svrshl_n_s16_m(__VA_ARGS__)
+#define svrshl_n_s8_x(...) __builtin_sve_svrshl_n_s8_x(__VA_ARGS__)
+#define svrshl_n_s32_x(...) __builtin_sve_svrshl_n_s32_x(__VA_ARGS__)
+#define svrshl_n_s64_x(...) __builtin_sve_svrshl_n_s64_x(__VA_ARGS__)
+#define svrshl_n_s16_x(...) __builtin_sve_svrshl_n_s16_x(__VA_ARGS__)
+#define svrshl_n_s8_z(...) __builtin_sve_svrshl_n_s8_z(__VA_ARGS__)
+#define svrshl_n_s32_z(...) __builtin_sve_svrshl_n_s32_z(__VA_ARGS__)
+#define svrshl_n_s64_z(...) __builtin_sve_svrshl_n_s64_z(__VA_ARGS__)
+#define svrshl_n_s16_z(...) __builtin_sve_svrshl_n_s16_z(__VA_ARGS__)
+#define svrshl_n_u8_m(...) __builtin_sve_svrshl_n_u8_m(__VA_ARGS__)
+#define svrshl_n_u32_m(...) __builtin_sve_svrshl_n_u32_m(__VA_ARGS__)
+#define svrshl_n_u64_m(...) __builtin_sve_svrshl_n_u64_m(__VA_ARGS__)
+#define svrshl_n_u16_m(...) __builtin_sve_svrshl_n_u16_m(__VA_ARGS__)
+#define svrshl_n_u8_x(...) __builtin_sve_svrshl_n_u8_x(__VA_ARGS__)
+#define svrshl_n_u32_x(...) __builtin_sve_svrshl_n_u32_x(__VA_ARGS__)
+#define svrshl_n_u64_x(...) __builtin_sve_svrshl_n_u64_x(__VA_ARGS__)
+#define svrshl_n_u16_x(...) __builtin_sve_svrshl_n_u16_x(__VA_ARGS__)
+#define svrshl_n_u8_z(...) __builtin_sve_svrshl_n_u8_z(__VA_ARGS__)
+#define svrshl_n_u32_z(...) __builtin_sve_svrshl_n_u32_z(__VA_ARGS__)
+#define svrshl_n_u64_z(...) __builtin_sve_svrshl_n_u64_z(__VA_ARGS__)
+#define svrshl_n_u16_z(...) __builtin_sve_svrshl_n_u16_z(__VA_ARGS__)
+#define svrshl_s8_m(...) __builtin_sve_svrshl_s8_m(__VA_ARGS__)
+#define svrshl_s32_m(...) __builtin_sve_svrshl_s32_m(__VA_ARGS__)
+#define svrshl_s64_m(...) __builtin_sve_svrshl_s64_m(__VA_ARGS__)
+#define svrshl_s16_m(...) __builtin_sve_svrshl_s16_m(__VA_ARGS__)
+#define svrshl_s8_x(...) __builtin_sve_svrshl_s8_x(__VA_ARGS__)
+#define svrshl_s32_x(...) __builtin_sve_svrshl_s32_x(__VA_ARGS__)
+#define svrshl_s64_x(...) __builtin_sve_svrshl_s64_x(__VA_ARGS__)
+#define svrshl_s16_x(...) __builtin_sve_svrshl_s16_x(__VA_ARGS__)
+#define svrshl_s8_z(...) __builtin_sve_svrshl_s8_z(__VA_ARGS__)
+#define svrshl_s32_z(...) __builtin_sve_svrshl_s32_z(__VA_ARGS__)
+#define svrshl_s64_z(...) __builtin_sve_svrshl_s64_z(__VA_ARGS__)
+#define svrshl_s16_z(...) __builtin_sve_svrshl_s16_z(__VA_ARGS__)
+#define svrshl_u8_m(...) __builtin_sve_svrshl_u8_m(__VA_ARGS__)
+#define svrshl_u32_m(...) __builtin_sve_svrshl_u32_m(__VA_ARGS__)
+#define svrshl_u64_m(...) __builtin_sve_svrshl_u64_m(__VA_ARGS__)
+#define svrshl_u16_m(...) __builtin_sve_svrshl_u16_m(__VA_ARGS__)
+#define svrshl_u8_x(...) __builtin_sve_svrshl_u8_x(__VA_ARGS__)
+#define svrshl_u32_x(...) __builtin_sve_svrshl_u32_x(__VA_ARGS__)
+#define svrshl_u64_x(...) __builtin_sve_svrshl_u64_x(__VA_ARGS__)
+#define svrshl_u16_x(...) __builtin_sve_svrshl_u16_x(__VA_ARGS__)
+#define svrshl_u8_z(...) __builtin_sve_svrshl_u8_z(__VA_ARGS__)
+#define svrshl_u32_z(...) __builtin_sve_svrshl_u32_z(__VA_ARGS__)
+#define svrshl_u64_z(...) __builtin_sve_svrshl_u64_z(__VA_ARGS__)
+#define svrshl_u16_z(...) __builtin_sve_svrshl_u16_z(__VA_ARGS__)
+#define svrshr_n_s8_m(...) __builtin_sve_svrshr_n_s8_m(__VA_ARGS__)
+#define svrshr_n_s32_m(...) __builtin_sve_svrshr_n_s32_m(__VA_ARGS__)
+#define svrshr_n_s64_m(...) __builtin_sve_svrshr_n_s64_m(__VA_ARGS__)
+#define svrshr_n_s16_m(...) __builtin_sve_svrshr_n_s16_m(__VA_ARGS__)
+#define svrshr_n_u8_m(...) __builtin_sve_svrshr_n_u8_m(__VA_ARGS__)
+#define svrshr_n_u32_m(...) __builtin_sve_svrshr_n_u32_m(__VA_ARGS__)
+#define svrshr_n_u64_m(...) __builtin_sve_svrshr_n_u64_m(__VA_ARGS__)
+#define svrshr_n_u16_m(...) __builtin_sve_svrshr_n_u16_m(__VA_ARGS__)
+#define svrshr_n_s8_x(...) __builtin_sve_svrshr_n_s8_x(__VA_ARGS__)
+#define svrshr_n_s32_x(...) __builtin_sve_svrshr_n_s32_x(__VA_ARGS__)
+#define svrshr_n_s64_x(...) __builtin_sve_svrshr_n_s64_x(__VA_ARGS__)
+#define svrshr_n_s16_x(...) __builtin_sve_svrshr_n_s16_x(__VA_ARGS__)
+#define svrshr_n_u8_x(...) __builtin_sve_svrshr_n_u8_x(__VA_ARGS__)
+#define svrshr_n_u32_x(...) __builtin_sve_svrshr_n_u32_x(__VA_ARGS__)
+#define svrshr_n_u64_x(...) __builtin_sve_svrshr_n_u64_x(__VA_ARGS__)
+#define svrshr_n_u16_x(...) __builtin_sve_svrshr_n_u16_x(__VA_ARGS__)
+#define svrshr_n_s8_z(...) __builtin_sve_svrshr_n_s8_z(__VA_ARGS__)
+#define svrshr_n_s32_z(...) __builtin_sve_svrshr_n_s32_z(__VA_ARGS__)
+#define svrshr_n_s64_z(...) __builtin_sve_svrshr_n_s64_z(__VA_ARGS__)
+#define svrshr_n_s16_z(...) __builtin_sve_svrshr_n_s16_z(__VA_ARGS__)
+#define svrshr_n_u8_z(...) __builtin_sve_svrshr_n_u8_z(__VA_ARGS__)
+#define svrshr_n_u32_z(...) __builtin_sve_svrshr_n_u32_z(__VA_ARGS__)
+#define svrshr_n_u64_z(...) __builtin_sve_svrshr_n_u64_z(__VA_ARGS__)
+#define svrshr_n_u16_z(...) __builtin_sve_svrshr_n_u16_z(__VA_ARGS__)
+#define svrshrnb_n_u32(...) __builtin_sve_svrshrnb_n_u32(__VA_ARGS__)
+#define svrshrnb_n_u64(...) __builtin_sve_svrshrnb_n_u64(__VA_ARGS__)
+#define svrshrnb_n_u16(...) __builtin_sve_svrshrnb_n_u16(__VA_ARGS__)
+#define svrshrnb_n_s32(...) __builtin_sve_svrshrnb_n_s32(__VA_ARGS__)
+#define svrshrnb_n_s64(...) __builtin_sve_svrshrnb_n_s64(__VA_ARGS__)
+#define svrshrnb_n_s16(...) __builtin_sve_svrshrnb_n_s16(__VA_ARGS__)
+#define svrshrnt_n_u32(...) __builtin_sve_svrshrnt_n_u32(__VA_ARGS__)
+#define svrshrnt_n_u64(...) __builtin_sve_svrshrnt_n_u64(__VA_ARGS__)
+#define svrshrnt_n_u16(...) __builtin_sve_svrshrnt_n_u16(__VA_ARGS__)
+#define svrshrnt_n_s32(...) __builtin_sve_svrshrnt_n_s32(__VA_ARGS__)
+#define svrshrnt_n_s64(...) __builtin_sve_svrshrnt_n_s64(__VA_ARGS__)
+#define svrshrnt_n_s16(...) __builtin_sve_svrshrnt_n_s16(__VA_ARGS__)
+#define svrsqrte_u32_m(...) __builtin_sve_svrsqrte_u32_m(__VA_ARGS__)
+#define svrsqrte_u32_x(...) __builtin_sve_svrsqrte_u32_x(__VA_ARGS__)
+#define svrsqrte_u32_z(...) __builtin_sve_svrsqrte_u32_z(__VA_ARGS__)
+#define svrsra_n_s8(...) __builtin_sve_svrsra_n_s8(__VA_ARGS__)
+#define svrsra_n_s32(...) __builtin_sve_svrsra_n_s32(__VA_ARGS__)
+#define svrsra_n_s64(...) __builtin_sve_svrsra_n_s64(__VA_ARGS__)
+#define svrsra_n_s16(...) __builtin_sve_svrsra_n_s16(__VA_ARGS__)
+#define svrsra_n_u8(...) __builtin_sve_svrsra_n_u8(__VA_ARGS__)
+#define svrsra_n_u32(...) __builtin_sve_svrsra_n_u32(__VA_ARGS__)
+#define svrsra_n_u64(...) __builtin_sve_svrsra_n_u64(__VA_ARGS__)
+#define svrsra_n_u16(...) __builtin_sve_svrsra_n_u16(__VA_ARGS__)
+#define svrsubhnb_n_u32(...) __builtin_sve_svrsubhnb_n_u32(__VA_ARGS__)
+#define svrsubhnb_n_u64(...) __builtin_sve_svrsubhnb_n_u64(__VA_ARGS__)
+#define svrsubhnb_n_u16(...) __builtin_sve_svrsubhnb_n_u16(__VA_ARGS__)
+#define svrsubhnb_n_s32(...) __builtin_sve_svrsubhnb_n_s32(__VA_ARGS__)
+#define svrsubhnb_n_s64(...) __builtin_sve_svrsubhnb_n_s64(__VA_ARGS__)
+#define svrsubhnb_n_s16(...) __builtin_sve_svrsubhnb_n_s16(__VA_ARGS__)
+#define svrsubhnb_u32(...) __builtin_sve_svrsubhnb_u32(__VA_ARGS__)
+#define svrsubhnb_u64(...) __builtin_sve_svrsubhnb_u64(__VA_ARGS__)
+#define svrsubhnb_u16(...) __builtin_sve_svrsubhnb_u16(__VA_ARGS__)
+#define svrsubhnb_s32(...) __builtin_sve_svrsubhnb_s32(__VA_ARGS__)
+#define svrsubhnb_s64(...) __builtin_sve_svrsubhnb_s64(__VA_ARGS__)
+#define svrsubhnb_s16(...) __builtin_sve_svrsubhnb_s16(__VA_ARGS__)
+#define svrsubhnt_n_u32(...) __builtin_sve_svrsubhnt_n_u32(__VA_ARGS__)
+#define svrsubhnt_n_u64(...) __builtin_sve_svrsubhnt_n_u64(__VA_ARGS__)
+#define svrsubhnt_n_u16(...) __builtin_sve_svrsubhnt_n_u16(__VA_ARGS__)
+#define svrsubhnt_n_s32(...) __builtin_sve_svrsubhnt_n_s32(__VA_ARGS__)
+#define svrsubhnt_n_s64(...) __builtin_sve_svrsubhnt_n_s64(__VA_ARGS__)
+#define svrsubhnt_n_s16(...) __builtin_sve_svrsubhnt_n_s16(__VA_ARGS__)
+#define svrsubhnt_u32(...) __builtin_sve_svrsubhnt_u32(__VA_ARGS__)
+#define svrsubhnt_u64(...) __builtin_sve_svrsubhnt_u64(__VA_ARGS__)
+#define svrsubhnt_u16(...) __builtin_sve_svrsubhnt_u16(__VA_ARGS__)
+#define svrsubhnt_s32(...) __builtin_sve_svrsubhnt_s32(__VA_ARGS__)
+#define svrsubhnt_s64(...) __builtin_sve_svrsubhnt_s64(__VA_ARGS__)
+#define svrsubhnt_s16(...) __builtin_sve_svrsubhnt_s16(__VA_ARGS__)
+#define svsbclb_n_u32(...) __builtin_sve_svsbclb_n_u32(__VA_ARGS__)
+#define svsbclb_n_u64(...) __builtin_sve_svsbclb_n_u64(__VA_ARGS__)
+#define svsbclb_u32(...) __builtin_sve_svsbclb_u32(__VA_ARGS__)
+#define svsbclb_u64(...) __builtin_sve_svsbclb_u64(__VA_ARGS__)
+#define svsbclt_n_u32(...) __builtin_sve_svsbclt_n_u32(__VA_ARGS__)
+#define svsbclt_n_u64(...) __builtin_sve_svsbclt_n_u64(__VA_ARGS__)
+#define svsbclt_u32(...) __builtin_sve_svsbclt_u32(__VA_ARGS__)
+#define svsbclt_u64(...) __builtin_sve_svsbclt_u64(__VA_ARGS__)
+#define svshllb_n_s32(...) __builtin_sve_svshllb_n_s32(__VA_ARGS__)
+#define svshllb_n_s64(...) __builtin_sve_svshllb_n_s64(__VA_ARGS__)
+#define svshllb_n_s16(...) __builtin_sve_svshllb_n_s16(__VA_ARGS__)
+#define svshllb_n_u32(...) __builtin_sve_svshllb_n_u32(__VA_ARGS__)
+#define svshllb_n_u64(...) __builtin_sve_svshllb_n_u64(__VA_ARGS__)
+#define svshllb_n_u16(...) __builtin_sve_svshllb_n_u16(__VA_ARGS__)
+#define svshllt_n_s32(...) __builtin_sve_svshllt_n_s32(__VA_ARGS__)
+#define svshllt_n_s64(...) __builtin_sve_svshllt_n_s64(__VA_ARGS__)
+#define svshllt_n_s16(...) __builtin_sve_svshllt_n_s16(__VA_ARGS__)
+#define svshllt_n_u32(...) __builtin_sve_svshllt_n_u32(__VA_ARGS__)
+#define svshllt_n_u64(...) __builtin_sve_svshllt_n_u64(__VA_ARGS__)
+#define svshllt_n_u16(...) __builtin_sve_svshllt_n_u16(__VA_ARGS__)
+#define svshrnb_n_u32(...) __builtin_sve_svshrnb_n_u32(__VA_ARGS__)
+#define svshrnb_n_u64(...) __builtin_sve_svshrnb_n_u64(__VA_ARGS__)
+#define svshrnb_n_u16(...) __builtin_sve_svshrnb_n_u16(__VA_ARGS__)
+#define svshrnb_n_s32(...) __builtin_sve_svshrnb_n_s32(__VA_ARGS__)
+#define svshrnb_n_s64(...) __builtin_sve_svshrnb_n_s64(__VA_ARGS__)
+#define svshrnb_n_s16(...) __builtin_sve_svshrnb_n_s16(__VA_ARGS__)
+#define svshrnt_n_u32(...) __builtin_sve_svshrnt_n_u32(__VA_ARGS__)
+#define svshrnt_n_u64(...) __builtin_sve_svshrnt_n_u64(__VA_ARGS__)
+#define svshrnt_n_u16(...) __builtin_sve_svshrnt_n_u16(__VA_ARGS__)
+#define svshrnt_n_s32(...) __builtin_sve_svshrnt_n_s32(__VA_ARGS__)
+#define svshrnt_n_s64(...) __builtin_sve_svshrnt_n_s64(__VA_ARGS__)
+#define svshrnt_n_s16(...) __builtin_sve_svshrnt_n_s16(__VA_ARGS__)
+#define svsli_n_u8(...) __builtin_sve_svsli_n_u8(__VA_ARGS__)
+#define svsli_n_u32(...) __builtin_sve_svsli_n_u32(__VA_ARGS__)
+#define svsli_n_u64(...) __builtin_sve_svsli_n_u64(__VA_ARGS__)
+#define svsli_n_u16(...) __builtin_sve_svsli_n_u16(__VA_ARGS__)
+#define svsli_n_s8(...) __builtin_sve_svsli_n_s8(__VA_ARGS__)
+#define svsli_n_s32(...) __builtin_sve_svsli_n_s32(__VA_ARGS__)
+#define svsli_n_s64(...) __builtin_sve_svsli_n_s64(__VA_ARGS__)
+#define svsli_n_s16(...) __builtin_sve_svsli_n_s16(__VA_ARGS__)
+#define svsqadd_n_u8_m(...) __builtin_sve_svsqadd_n_u8_m(__VA_ARGS__)
+#define svsqadd_n_u32_m(...) __builtin_sve_svsqadd_n_u32_m(__VA_ARGS__)
+#define svsqadd_n_u64_m(...) __builtin_sve_svsqadd_n_u64_m(__VA_ARGS__)
+#define svsqadd_n_u16_m(...) __builtin_sve_svsqadd_n_u16_m(__VA_ARGS__)
+#define svsqadd_n_u8_x(...) __builtin_sve_svsqadd_n_u8_x(__VA_ARGS__)
+#define svsqadd_n_u32_x(...) __builtin_sve_svsqadd_n_u32_x(__VA_ARGS__)
+#define svsqadd_n_u64_x(...) __builtin_sve_svsqadd_n_u64_x(__VA_ARGS__)
+#define svsqadd_n_u16_x(...) __builtin_sve_svsqadd_n_u16_x(__VA_ARGS__)
+#define svsqadd_n_u8_z(...) __builtin_sve_svsqadd_n_u8_z(__VA_ARGS__)
+#define svsqadd_n_u32_z(...) __builtin_sve_svsqadd_n_u32_z(__VA_ARGS__)
+#define svsqadd_n_u64_z(...) __builtin_sve_svsqadd_n_u64_z(__VA_ARGS__)
+#define svsqadd_n_u16_z(...) __builtin_sve_svsqadd_n_u16_z(__VA_ARGS__)
+#define svsqadd_u8_m(...) __builtin_sve_svsqadd_u8_m(__VA_ARGS__)
+#define svsqadd_u32_m(...) __builtin_sve_svsqadd_u32_m(__VA_ARGS__)
+#define svsqadd_u64_m(...) __builtin_sve_svsqadd_u64_m(__VA_ARGS__)
+#define svsqadd_u16_m(...) __builtin_sve_svsqadd_u16_m(__VA_ARGS__)
+#define svsqadd_u8_x(...) __builtin_sve_svsqadd_u8_x(__VA_ARGS__)
+#define svsqadd_u32_x(...) __builtin_sve_svsqadd_u32_x(__VA_ARGS__)
+#define svsqadd_u64_x(...) __builtin_sve_svsqadd_u64_x(__VA_ARGS__)
+#define svsqadd_u16_x(...) __builtin_sve_svsqadd_u16_x(__VA_ARGS__)
+#define svsqadd_u8_z(...) __builtin_sve_svsqadd_u8_z(__VA_ARGS__)
+#define svsqadd_u32_z(...) __builtin_sve_svsqadd_u32_z(__VA_ARGS__)
+#define svsqadd_u64_z(...) __builtin_sve_svsqadd_u64_z(__VA_ARGS__)
+#define svsqadd_u16_z(...) __builtin_sve_svsqadd_u16_z(__VA_ARGS__)
+#define svsra_n_s8(...) __builtin_sve_svsra_n_s8(__VA_ARGS__)
+#define svsra_n_s32(...) __builtin_sve_svsra_n_s32(__VA_ARGS__)
+#define svsra_n_s64(...) __builtin_sve_svsra_n_s64(__VA_ARGS__)
+#define svsra_n_s16(...) __builtin_sve_svsra_n_s16(__VA_ARGS__)
+#define svsra_n_u8(...) __builtin_sve_svsra_n_u8(__VA_ARGS__)
+#define svsra_n_u32(...) __builtin_sve_svsra_n_u32(__VA_ARGS__)
+#define svsra_n_u64(...) __builtin_sve_svsra_n_u64(__VA_ARGS__)
+#define svsra_n_u16(...) __builtin_sve_svsra_n_u16(__VA_ARGS__)
+#define svsri_n_u8(...) __builtin_sve_svsri_n_u8(__VA_ARGS__)
+#define svsri_n_u32(...) __builtin_sve_svsri_n_u32(__VA_ARGS__)
+#define svsri_n_u64(...) __builtin_sve_svsri_n_u64(__VA_ARGS__)
+#define svsri_n_u16(...) __builtin_sve_svsri_n_u16(__VA_ARGS__)
+#define svsri_n_s8(...) __builtin_sve_svsri_n_s8(__VA_ARGS__)
+#define svsri_n_s32(...) __builtin_sve_svsri_n_s32(__VA_ARGS__)
+#define svsri_n_s64(...) __builtin_sve_svsri_n_s64(__VA_ARGS__)
+#define svsri_n_s16(...) __builtin_sve_svsri_n_s16(__VA_ARGS__)
+#define svstnt1_scatter_u32base_index_u32(...) __builtin_sve_svstnt1_scatter_u32base_index_u32(__VA_ARGS__)
+#define svstnt1_scatter_u64base_index_u64(...) __builtin_sve_svstnt1_scatter_u64base_index_u64(__VA_ARGS__)
+#define svstnt1_scatter_u64base_index_f64(...) __builtin_sve_svstnt1_scatter_u64base_index_f64(__VA_ARGS__)
+#define svstnt1_scatter_u32base_index_f32(...) __builtin_sve_svstnt1_scatter_u32base_index_f32(__VA_ARGS__)
+#define svstnt1_scatter_u32base_index_s32(...) __builtin_sve_svstnt1_scatter_u32base_index_s32(__VA_ARGS__)
+#define svstnt1_scatter_u64base_index_s64(...) __builtin_sve_svstnt1_scatter_u64base_index_s64(__VA_ARGS__)
+#define svstnt1_scatter_u32base_offset_u32(...) __builtin_sve_svstnt1_scatter_u32base_offset_u32(__VA_ARGS__)
+#define svstnt1_scatter_u64base_offset_u64(...) __builtin_sve_svstnt1_scatter_u64base_offset_u64(__VA_ARGS__)
+#define svstnt1_scatter_u64base_offset_f64(...) __builtin_sve_svstnt1_scatter_u64base_offset_f64(__VA_ARGS__)
+#define svstnt1_scatter_u32base_offset_f32(...) __builtin_sve_svstnt1_scatter_u32base_offset_f32(__VA_ARGS__)
+#define svstnt1_scatter_u32base_offset_s32(...) __builtin_sve_svstnt1_scatter_u32base_offset_s32(__VA_ARGS__)
+#define svstnt1_scatter_u64base_offset_s64(...) __builtin_sve_svstnt1_scatter_u64base_offset_s64(__VA_ARGS__)
+#define svstnt1_scatter_u32base_u32(...) __builtin_sve_svstnt1_scatter_u32base_u32(__VA_ARGS__)
+#define svstnt1_scatter_u64base_u64(...) __builtin_sve_svstnt1_scatter_u64base_u64(__VA_ARGS__)
+#define svstnt1_scatter_u64base_f64(...) __builtin_sve_svstnt1_scatter_u64base_f64(__VA_ARGS__)
+#define svstnt1_scatter_u32base_f32(...) __builtin_sve_svstnt1_scatter_u32base_f32(__VA_ARGS__)
+#define svstnt1_scatter_u32base_s32(...) __builtin_sve_svstnt1_scatter_u32base_s32(__VA_ARGS__)
+#define svstnt1_scatter_u64base_s64(...) __builtin_sve_svstnt1_scatter_u64base_s64(__VA_ARGS__)
+#define svstnt1_scatter_s64index_u64(...) __builtin_sve_svstnt1_scatter_s64index_u64(__VA_ARGS__)
+#define svstnt1_scatter_s64index_f64(...) __builtin_sve_svstnt1_scatter_s64index_f64(__VA_ARGS__)
+#define svstnt1_scatter_s64index_s64(...) __builtin_sve_svstnt1_scatter_s64index_s64(__VA_ARGS__)
+#define svstnt1_scatter_u64index_u64(...) __builtin_sve_svstnt1_scatter_u64index_u64(__VA_ARGS__)
+#define svstnt1_scatter_u64index_f64(...) __builtin_sve_svstnt1_scatter_u64index_f64(__VA_ARGS__)
+#define svstnt1_scatter_u64index_s64(...) __builtin_sve_svstnt1_scatter_u64index_s64(__VA_ARGS__)
+#define svstnt1_scatter_u32offset_u32(...) __builtin_sve_svstnt1_scatter_u32offset_u32(__VA_ARGS__)
+#define svstnt1_scatter_u32offset_f32(...) __builtin_sve_svstnt1_scatter_u32offset_f32(__VA_ARGS__)
+#define svstnt1_scatter_u32offset_s32(...) __builtin_sve_svstnt1_scatter_u32offset_s32(__VA_ARGS__)
+#define svstnt1_scatter_s64offset_u64(...) __builtin_sve_svstnt1_scatter_s64offset_u64(__VA_ARGS__)
+#define svstnt1_scatter_s64offset_f64(...) __builtin_sve_svstnt1_scatter_s64offset_f64(__VA_ARGS__)
+#define svstnt1_scatter_s64offset_s64(...) __builtin_sve_svstnt1_scatter_s64offset_s64(__VA_ARGS__)
+#define svstnt1_scatter_u64offset_u64(...) __builtin_sve_svstnt1_scatter_u64offset_u64(__VA_ARGS__)
+#define svstnt1_scatter_u64offset_f64(...) __builtin_sve_svstnt1_scatter_u64offset_f64(__VA_ARGS__)
+#define svstnt1_scatter_u64offset_s64(...) __builtin_sve_svstnt1_scatter_u64offset_s64(__VA_ARGS__)
+#define svstnt1b_scatter_u32base_offset_u32(...) __builtin_sve_svstnt1b_scatter_u32base_offset_u32(__VA_ARGS__)
+#define svstnt1b_scatter_u64base_offset_u64(...) __builtin_sve_svstnt1b_scatter_u64base_offset_u64(__VA_ARGS__)
+#define svstnt1b_scatter_u32base_offset_s32(...) __builtin_sve_svstnt1b_scatter_u32base_offset_s32(__VA_ARGS__)
+#define svstnt1b_scatter_u64base_offset_s64(...) __builtin_sve_svstnt1b_scatter_u64base_offset_s64(__VA_ARGS__)
+#define svstnt1b_scatter_u32base_u32(...) __builtin_sve_svstnt1b_scatter_u32base_u32(__VA_ARGS__)
+#define svstnt1b_scatter_u64base_u64(...) __builtin_sve_svstnt1b_scatter_u64base_u64(__VA_ARGS__)
+#define svstnt1b_scatter_u32base_s32(...) __builtin_sve_svstnt1b_scatter_u32base_s32(__VA_ARGS__)
+#define svstnt1b_scatter_u64base_s64(...) __builtin_sve_svstnt1b_scatter_u64base_s64(__VA_ARGS__)
+#define svstnt1b_scatter_u32offset_s32(...) __builtin_sve_svstnt1b_scatter_u32offset_s32(__VA_ARGS__)
+#define svstnt1b_scatter_u32offset_u32(...) __builtin_sve_svstnt1b_scatter_u32offset_u32(__VA_ARGS__)
+#define svstnt1b_scatter_s64offset_s64(...) __builtin_sve_svstnt1b_scatter_s64offset_s64(__VA_ARGS__)
+#define svstnt1b_scatter_s64offset_u64(...) __builtin_sve_svstnt1b_scatter_s64offset_u64(__VA_ARGS__)
+#define svstnt1b_scatter_u64offset_s64(...) __builtin_sve_svstnt1b_scatter_u64offset_s64(__VA_ARGS__)
+#define svstnt1b_scatter_u64offset_u64(...) __builtin_sve_svstnt1b_scatter_u64offset_u64(__VA_ARGS__)
+#define svstnt1h_scatter_u32base_index_u32(...) __builtin_sve_svstnt1h_scatter_u32base_index_u32(__VA_ARGS__)
+#define svstnt1h_scatter_u64base_index_u64(...) __builtin_sve_svstnt1h_scatter_u64base_index_u64(__VA_ARGS__)
+#define svstnt1h_scatter_u32base_index_s32(...) __builtin_sve_svstnt1h_scatter_u32base_index_s32(__VA_ARGS__)
+#define svstnt1h_scatter_u64base_index_s64(...) __builtin_sve_svstnt1h_scatter_u64base_index_s64(__VA_ARGS__)
+#define svstnt1h_scatter_u32base_offset_u32(...) __builtin_sve_svstnt1h_scatter_u32base_offset_u32(__VA_ARGS__)
+#define svstnt1h_scatter_u64base_offset_u64(...) __builtin_sve_svstnt1h_scatter_u64base_offset_u64(__VA_ARGS__)
+#define svstnt1h_scatter_u32base_offset_s32(...) __builtin_sve_svstnt1h_scatter_u32base_offset_s32(__VA_ARGS__)
+#define svstnt1h_scatter_u64base_offset_s64(...) __builtin_sve_svstnt1h_scatter_u64base_offset_s64(__VA_ARGS__)
+#define svstnt1h_scatter_u32base_u32(...) __builtin_sve_svstnt1h_scatter_u32base_u32(__VA_ARGS__)
+#define svstnt1h_scatter_u64base_u64(...) __builtin_sve_svstnt1h_scatter_u64base_u64(__VA_ARGS__)
+#define svstnt1h_scatter_u32base_s32(...) __builtin_sve_svstnt1h_scatter_u32base_s32(__VA_ARGS__)
+#define svstnt1h_scatter_u64base_s64(...) __builtin_sve_svstnt1h_scatter_u64base_s64(__VA_ARGS__)
+#define svstnt1h_scatter_s64index_s64(...) __builtin_sve_svstnt1h_scatter_s64index_s64(__VA_ARGS__)
+#define svstnt1h_scatter_s64index_u64(...) __builtin_sve_svstnt1h_scatter_s64index_u64(__VA_ARGS__)
+#define svstnt1h_scatter_u64index_s64(...) __builtin_sve_svstnt1h_scatter_u64index_s64(__VA_ARGS__)
+#define svstnt1h_scatter_u64index_u64(...) __builtin_sve_svstnt1h_scatter_u64index_u64(__VA_ARGS__)
+#define svstnt1h_scatter_u32offset_s32(...) __builtin_sve_svstnt1h_scatter_u32offset_s32(__VA_ARGS__)
+#define svstnt1h_scatter_u32offset_u32(...) __builtin_sve_svstnt1h_scatter_u32offset_u32(__VA_ARGS__)
+#define svstnt1h_scatter_s64offset_s64(...) __builtin_sve_svstnt1h_scatter_s64offset_s64(__VA_ARGS__)
+#define svstnt1h_scatter_s64offset_u64(...) __builtin_sve_svstnt1h_scatter_s64offset_u64(__VA_ARGS__)
+#define svstnt1h_scatter_u64offset_s64(...) __builtin_sve_svstnt1h_scatter_u64offset_s64(__VA_ARGS__)
+#define svstnt1h_scatter_u64offset_u64(...) __builtin_sve_svstnt1h_scatter_u64offset_u64(__VA_ARGS__)
+#define svstnt1w_scatter_u64base_index_u64(...) __builtin_sve_svstnt1w_scatter_u64base_index_u64(__VA_ARGS__)
+#define svstnt1w_scatter_u64base_index_s64(...) __builtin_sve_svstnt1w_scatter_u64base_index_s64(__VA_ARGS__)
+#define svstnt1w_scatter_u64base_offset_u64(...) __builtin_sve_svstnt1w_scatter_u64base_offset_u64(__VA_ARGS__)
+#define svstnt1w_scatter_u64base_offset_s64(...) __builtin_sve_svstnt1w_scatter_u64base_offset_s64(__VA_ARGS__)
+#define svstnt1w_scatter_u64base_u64(...) __builtin_sve_svstnt1w_scatter_u64base_u64(__VA_ARGS__)
+#define svstnt1w_scatter_u64base_s64(...) __builtin_sve_svstnt1w_scatter_u64base_s64(__VA_ARGS__)
+#define svstnt1w_scatter_s64index_s64(...) __builtin_sve_svstnt1w_scatter_s64index_s64(__VA_ARGS__)
+#define svstnt1w_scatter_s64index_u64(...) __builtin_sve_svstnt1w_scatter_s64index_u64(__VA_ARGS__)
+#define svstnt1w_scatter_u64index_s64(...) __builtin_sve_svstnt1w_scatter_u64index_s64(__VA_ARGS__)
+#define svstnt1w_scatter_u64index_u64(...) __builtin_sve_svstnt1w_scatter_u64index_u64(__VA_ARGS__)
+#define svstnt1w_scatter_s64offset_s64(...) __builtin_sve_svstnt1w_scatter_s64offset_s64(__VA_ARGS__)
+#define svstnt1w_scatter_s64offset_u64(...) __builtin_sve_svstnt1w_scatter_s64offset_u64(__VA_ARGS__)
+#define svstnt1w_scatter_u64offset_s64(...) __builtin_sve_svstnt1w_scatter_u64offset_s64(__VA_ARGS__)
+#define svstnt1w_scatter_u64offset_u64(...) __builtin_sve_svstnt1w_scatter_u64offset_u64(__VA_ARGS__)
+#define svsubhnb_n_u32(...) __builtin_sve_svsubhnb_n_u32(__VA_ARGS__)
+#define svsubhnb_n_u64(...) __builtin_sve_svsubhnb_n_u64(__VA_ARGS__)
+#define svsubhnb_n_u16(...) __builtin_sve_svsubhnb_n_u16(__VA_ARGS__)
+#define svsubhnb_n_s32(...) __builtin_sve_svsubhnb_n_s32(__VA_ARGS__)
+#define svsubhnb_n_s64(...) __builtin_sve_svsubhnb_n_s64(__VA_ARGS__)
+#define svsubhnb_n_s16(...) __builtin_sve_svsubhnb_n_s16(__VA_ARGS__)
+#define svsubhnb_u32(...) __builtin_sve_svsubhnb_u32(__VA_ARGS__)
+#define svsubhnb_u64(...) __builtin_sve_svsubhnb_u64(__VA_ARGS__)
+#define svsubhnb_u16(...) __builtin_sve_svsubhnb_u16(__VA_ARGS__)
+#define svsubhnb_s32(...) __builtin_sve_svsubhnb_s32(__VA_ARGS__)
+#define svsubhnb_s64(...) __builtin_sve_svsubhnb_s64(__VA_ARGS__)
+#define svsubhnb_s16(...) __builtin_sve_svsubhnb_s16(__VA_ARGS__)
+#define svsubhnt_n_u32(...) __builtin_sve_svsubhnt_n_u32(__VA_ARGS__)
+#define svsubhnt_n_u64(...) __builtin_sve_svsubhnt_n_u64(__VA_ARGS__)
+#define svsubhnt_n_u16(...) __builtin_sve_svsubhnt_n_u16(__VA_ARGS__)
+#define svsubhnt_n_s32(...) __builtin_sve_svsubhnt_n_s32(__VA_ARGS__)
+#define svsubhnt_n_s64(...) __builtin_sve_svsubhnt_n_s64(__VA_ARGS__)
+#define svsubhnt_n_s16(...) __builtin_sve_svsubhnt_n_s16(__VA_ARGS__)
+#define svsubhnt_u32(...) __builtin_sve_svsubhnt_u32(__VA_ARGS__)
+#define svsubhnt_u64(...) __builtin_sve_svsubhnt_u64(__VA_ARGS__)
+#define svsubhnt_u16(...) __builtin_sve_svsubhnt_u16(__VA_ARGS__)
+#define svsubhnt_s32(...) __builtin_sve_svsubhnt_s32(__VA_ARGS__)
+#define svsubhnt_s64(...) __builtin_sve_svsubhnt_s64(__VA_ARGS__)
+#define svsubhnt_s16(...) __builtin_sve_svsubhnt_s16(__VA_ARGS__)
+#define svsublb_n_s32(...) __builtin_sve_svsublb_n_s32(__VA_ARGS__)
+#define svsublb_n_s64(...) __builtin_sve_svsublb_n_s64(__VA_ARGS__)
+#define svsublb_n_s16(...) __builtin_sve_svsublb_n_s16(__VA_ARGS__)
+#define svsublb_n_u32(...) __builtin_sve_svsublb_n_u32(__VA_ARGS__)
+#define svsublb_n_u64(...) __builtin_sve_svsublb_n_u64(__VA_ARGS__)
+#define svsublb_n_u16(...) __builtin_sve_svsublb_n_u16(__VA_ARGS__)
+#define svsublb_s32(...) __builtin_sve_svsublb_s32(__VA_ARGS__)
+#define svsublb_s64(...) __builtin_sve_svsublb_s64(__VA_ARGS__)
+#define svsublb_s16(...) __builtin_sve_svsublb_s16(__VA_ARGS__)
+#define svsublb_u32(...) __builtin_sve_svsublb_u32(__VA_ARGS__)
+#define svsublb_u64(...) __builtin_sve_svsublb_u64(__VA_ARGS__)
+#define svsublb_u16(...) __builtin_sve_svsublb_u16(__VA_ARGS__)
+#define svsublbt_n_s32(...) __builtin_sve_svsublbt_n_s32(__VA_ARGS__)
+#define svsublbt_n_s64(...) __builtin_sve_svsublbt_n_s64(__VA_ARGS__)
+#define svsublbt_n_s16(...) __builtin_sve_svsublbt_n_s16(__VA_ARGS__)
+#define svsublbt_s32(...) __builtin_sve_svsublbt_s32(__VA_ARGS__)
+#define svsublbt_s64(...) __builtin_sve_svsublbt_s64(__VA_ARGS__)
+#define svsublbt_s16(...) __builtin_sve_svsublbt_s16(__VA_ARGS__)
+#define svsublt_n_s32(...) __builtin_sve_svsublt_n_s32(__VA_ARGS__)
+#define svsublt_n_s64(...) __builtin_sve_svsublt_n_s64(__VA_ARGS__)
+#define svsublt_n_s16(...) __builtin_sve_svsublt_n_s16(__VA_ARGS__)
+#define svsublt_n_u32(...) __builtin_sve_svsublt_n_u32(__VA_ARGS__)
+#define svsublt_n_u64(...) __builtin_sve_svsublt_n_u64(__VA_ARGS__)
+#define svsublt_n_u16(...) __builtin_sve_svsublt_n_u16(__VA_ARGS__)
+#define svsublt_s32(...) __builtin_sve_svsublt_s32(__VA_ARGS__)
+#define svsublt_s64(...) __builtin_sve_svsublt_s64(__VA_ARGS__)
+#define svsublt_s16(...) __builtin_sve_svsublt_s16(__VA_ARGS__)
+#define svsublt_u32(...) __builtin_sve_svsublt_u32(__VA_ARGS__)
+#define svsublt_u64(...) __builtin_sve_svsublt_u64(__VA_ARGS__)
+#define svsublt_u16(...) __builtin_sve_svsublt_u16(__VA_ARGS__)
+#define svsubltb_n_s32(...) __builtin_sve_svsubltb_n_s32(__VA_ARGS__)
+#define svsubltb_n_s64(...) __builtin_sve_svsubltb_n_s64(__VA_ARGS__)
+#define svsubltb_n_s16(...) __builtin_sve_svsubltb_n_s16(__VA_ARGS__)
+#define svsubltb_s32(...) __builtin_sve_svsubltb_s32(__VA_ARGS__)
+#define svsubltb_s64(...) __builtin_sve_svsubltb_s64(__VA_ARGS__)
+#define svsubltb_s16(...) __builtin_sve_svsubltb_s16(__VA_ARGS__)
+#define svsubwb_n_s32(...) __builtin_sve_svsubwb_n_s32(__VA_ARGS__)
+#define svsubwb_n_s64(...) __builtin_sve_svsubwb_n_s64(__VA_ARGS__)
+#define svsubwb_n_s16(...) __builtin_sve_svsubwb_n_s16(__VA_ARGS__)
+#define svsubwb_n_u32(...) __builtin_sve_svsubwb_n_u32(__VA_ARGS__)
+#define svsubwb_n_u64(...) __builtin_sve_svsubwb_n_u64(__VA_ARGS__)
+#define svsubwb_n_u16(...) __builtin_sve_svsubwb_n_u16(__VA_ARGS__)
+#define svsubwb_s32(...) __builtin_sve_svsubwb_s32(__VA_ARGS__)
+#define svsubwb_s64(...) __builtin_sve_svsubwb_s64(__VA_ARGS__)
+#define svsubwb_s16(...) __builtin_sve_svsubwb_s16(__VA_ARGS__)
+#define svsubwb_u32(...) __builtin_sve_svsubwb_u32(__VA_ARGS__)
+#define svsubwb_u64(...) __builtin_sve_svsubwb_u64(__VA_ARGS__)
+#define svsubwb_u16(...) __builtin_sve_svsubwb_u16(__VA_ARGS__)
+#define svsubwt_n_s32(...) __builtin_sve_svsubwt_n_s32(__VA_ARGS__)
+#define svsubwt_n_s64(...) __builtin_sve_svsubwt_n_s64(__VA_ARGS__)
+#define svsubwt_n_s16(...) __builtin_sve_svsubwt_n_s16(__VA_ARGS__)
+#define svsubwt_n_u32(...) __builtin_sve_svsubwt_n_u32(__VA_ARGS__)
+#define svsubwt_n_u64(...) __builtin_sve_svsubwt_n_u64(__VA_ARGS__)
+#define svsubwt_n_u16(...) __builtin_sve_svsubwt_n_u16(__VA_ARGS__)
+#define svsubwt_s32(...) __builtin_sve_svsubwt_s32(__VA_ARGS__)
+#define svsubwt_s64(...) __builtin_sve_svsubwt_s64(__VA_ARGS__)
+#define svsubwt_s16(...) __builtin_sve_svsubwt_s16(__VA_ARGS__)
+#define svsubwt_u32(...) __builtin_sve_svsubwt_u32(__VA_ARGS__)
+#define svsubwt_u64(...) __builtin_sve_svsubwt_u64(__VA_ARGS__)
+#define svsubwt_u16(...) __builtin_sve_svsubwt_u16(__VA_ARGS__)
+#define svtbl2_u8(...) __builtin_sve_svtbl2_u8(__VA_ARGS__)
+#define svtbl2_u32(...) __builtin_sve_svtbl2_u32(__VA_ARGS__)
+#define svtbl2_u64(...) __builtin_sve_svtbl2_u64(__VA_ARGS__)
+#define svtbl2_u16(...) __builtin_sve_svtbl2_u16(__VA_ARGS__)
+#define svtbl2_s8(...) __builtin_sve_svtbl2_s8(__VA_ARGS__)
+#define svtbl2_f64(...) __builtin_sve_svtbl2_f64(__VA_ARGS__)
+#define svtbl2_f32(...) __builtin_sve_svtbl2_f32(__VA_ARGS__)
+#define svtbl2_f16(...) __builtin_sve_svtbl2_f16(__VA_ARGS__)
+#define svtbl2_s32(...) __builtin_sve_svtbl2_s32(__VA_ARGS__)
+#define svtbl2_s64(...) __builtin_sve_svtbl2_s64(__VA_ARGS__)
+#define svtbl2_s16(...) __builtin_sve_svtbl2_s16(__VA_ARGS__)
+#define svtbx_u8(...) __builtin_sve_svtbx_u8(__VA_ARGS__)
+#define svtbx_u32(...) __builtin_sve_svtbx_u32(__VA_ARGS__)
+#define svtbx_u64(...) __builtin_sve_svtbx_u64(__VA_ARGS__)
+#define svtbx_u16(...) __builtin_sve_svtbx_u16(__VA_ARGS__)
+#define svtbx_s8(...) __builtin_sve_svtbx_s8(__VA_ARGS__)
+#define svtbx_f64(...) __builtin_sve_svtbx_f64(__VA_ARGS__)
+#define svtbx_f32(...) __builtin_sve_svtbx_f32(__VA_ARGS__)
+#define svtbx_f16(...) __builtin_sve_svtbx_f16(__VA_ARGS__)
+#define svtbx_s32(...) __builtin_sve_svtbx_s32(__VA_ARGS__)
+#define svtbx_s64(...) __builtin_sve_svtbx_s64(__VA_ARGS__)
+#define svtbx_s16(...) __builtin_sve_svtbx_s16(__VA_ARGS__)
+#define svuqadd_n_s8_m(...) __builtin_sve_svuqadd_n_s8_m(__VA_ARGS__)
+#define svuqadd_n_s32_m(...) __builtin_sve_svuqadd_n_s32_m(__VA_ARGS__)
+#define svuqadd_n_s64_m(...) __builtin_sve_svuqadd_n_s64_m(__VA_ARGS__)
+#define svuqadd_n_s16_m(...) __builtin_sve_svuqadd_n_s16_m(__VA_ARGS__)
+#define svuqadd_n_s8_x(...) __builtin_sve_svuqadd_n_s8_x(__VA_ARGS__)
+#define svuqadd_n_s32_x(...) __builtin_sve_svuqadd_n_s32_x(__VA_ARGS__)
+#define svuqadd_n_s64_x(...) __builtin_sve_svuqadd_n_s64_x(__VA_ARGS__)
+#define svuqadd_n_s16_x(...) __builtin_sve_svuqadd_n_s16_x(__VA_ARGS__)
+#define svuqadd_n_s8_z(...) __builtin_sve_svuqadd_n_s8_z(__VA_ARGS__)
+#define svuqadd_n_s32_z(...) __builtin_sve_svuqadd_n_s32_z(__VA_ARGS__)
+#define svuqadd_n_s64_z(...) __builtin_sve_svuqadd_n_s64_z(__VA_ARGS__)
+#define svuqadd_n_s16_z(...) __builtin_sve_svuqadd_n_s16_z(__VA_ARGS__)
+#define svuqadd_s8_m(...) __builtin_sve_svuqadd_s8_m(__VA_ARGS__)
+#define svuqadd_s32_m(...) __builtin_sve_svuqadd_s32_m(__VA_ARGS__)
+#define svuqadd_s64_m(...) __builtin_sve_svuqadd_s64_m(__VA_ARGS__)
+#define svuqadd_s16_m(...) __builtin_sve_svuqadd_s16_m(__VA_ARGS__)
+#define svuqadd_s8_x(...) __builtin_sve_svuqadd_s8_x(__VA_ARGS__)
+#define svuqadd_s32_x(...) __builtin_sve_svuqadd_s32_x(__VA_ARGS__)
+#define svuqadd_s64_x(...) __builtin_sve_svuqadd_s64_x(__VA_ARGS__)
+#define svuqadd_s16_x(...) __builtin_sve_svuqadd_s16_x(__VA_ARGS__)
+#define svuqadd_s8_z(...) __builtin_sve_svuqadd_s8_z(__VA_ARGS__)
+#define svuqadd_s32_z(...) __builtin_sve_svuqadd_s32_z(__VA_ARGS__)
+#define svuqadd_s64_z(...) __builtin_sve_svuqadd_s64_z(__VA_ARGS__)
+#define svuqadd_s16_z(...) __builtin_sve_svuqadd_s16_z(__VA_ARGS__)
+#define svwhilege_b8_s32(...) __builtin_sve_svwhilege_b8_s32(__VA_ARGS__)
+#define svwhilege_b32_s32(...) __builtin_sve_svwhilege_b32_s32(__VA_ARGS__)
+#define svwhilege_b64_s32(...) __builtin_sve_svwhilege_b64_s32(__VA_ARGS__)
+#define svwhilege_b16_s32(...) __builtin_sve_svwhilege_b16_s32(__VA_ARGS__)
+#define svwhilege_b8_s64(...) __builtin_sve_svwhilege_b8_s64(__VA_ARGS__)
+#define svwhilege_b32_s64(...) __builtin_sve_svwhilege_b32_s64(__VA_ARGS__)
+#define svwhilege_b64_s64(...) __builtin_sve_svwhilege_b64_s64(__VA_ARGS__)
+#define svwhilege_b16_s64(...) __builtin_sve_svwhilege_b16_s64(__VA_ARGS__)
+#define svwhilege_b8_u32(...) __builtin_sve_svwhilege_b8_u32(__VA_ARGS__)
+#define svwhilege_b32_u32(...) __builtin_sve_svwhilege_b32_u32(__VA_ARGS__)
+#define svwhilege_b64_u32(...) __builtin_sve_svwhilege_b64_u32(__VA_ARGS__)
+#define svwhilege_b16_u32(...) __builtin_sve_svwhilege_b16_u32(__VA_ARGS__)
+#define svwhilege_b8_u64(...) __builtin_sve_svwhilege_b8_u64(__VA_ARGS__)
+#define svwhilege_b32_u64(...) __builtin_sve_svwhilege_b32_u64(__VA_ARGS__)
+#define svwhilege_b64_u64(...) __builtin_sve_svwhilege_b64_u64(__VA_ARGS__)
+#define svwhilege_b16_u64(...) __builtin_sve_svwhilege_b16_u64(__VA_ARGS__)
+#define svwhilegt_b8_s32(...) __builtin_sve_svwhilegt_b8_s32(__VA_ARGS__)
+#define svwhilegt_b32_s32(...) __builtin_sve_svwhilegt_b32_s32(__VA_ARGS__)
+#define svwhilegt_b64_s32(...) __builtin_sve_svwhilegt_b64_s32(__VA_ARGS__)
+#define svwhilegt_b16_s32(...) __builtin_sve_svwhilegt_b16_s32(__VA_ARGS__)
+#define svwhilegt_b8_s64(...) __builtin_sve_svwhilegt_b8_s64(__VA_ARGS__)
+#define svwhilegt_b32_s64(...) __builtin_sve_svwhilegt_b32_s64(__VA_ARGS__)
+#define svwhilegt_b64_s64(...) __builtin_sve_svwhilegt_b64_s64(__VA_ARGS__)
+#define svwhilegt_b16_s64(...) __builtin_sve_svwhilegt_b16_s64(__VA_ARGS__)
+#define svwhilegt_b8_u32(...) __builtin_sve_svwhilegt_b8_u32(__VA_ARGS__)
+#define svwhilegt_b32_u32(...) __builtin_sve_svwhilegt_b32_u32(__VA_ARGS__)
+#define svwhilegt_b64_u32(...) __builtin_sve_svwhilegt_b64_u32(__VA_ARGS__)
+#define svwhilegt_b16_u32(...) __builtin_sve_svwhilegt_b16_u32(__VA_ARGS__)
+#define svwhilegt_b8_u64(...) __builtin_sve_svwhilegt_b8_u64(__VA_ARGS__)
+#define svwhilegt_b32_u64(...) __builtin_sve_svwhilegt_b32_u64(__VA_ARGS__)
+#define svwhilegt_b64_u64(...) __builtin_sve_svwhilegt_b64_u64(__VA_ARGS__)
+#define svwhilegt_b16_u64(...) __builtin_sve_svwhilegt_b16_u64(__VA_ARGS__)
+#define svwhilerw_u8(...) __builtin_sve_svwhilerw_u8(__VA_ARGS__)
+#define svwhilerw_s8(...) __builtin_sve_svwhilerw_s8(__VA_ARGS__)
+#define svwhilerw_u64(...) __builtin_sve_svwhilerw_u64(__VA_ARGS__)
+#define svwhilerw_f64(...) __builtin_sve_svwhilerw_f64(__VA_ARGS__)
+#define svwhilerw_s64(...) __builtin_sve_svwhilerw_s64(__VA_ARGS__)
+#define svwhilerw_u16(...) __builtin_sve_svwhilerw_u16(__VA_ARGS__)
+#define svwhilerw_f16(...) __builtin_sve_svwhilerw_f16(__VA_ARGS__)
+#define svwhilerw_s16(...) __builtin_sve_svwhilerw_s16(__VA_ARGS__)
+#define svwhilerw_u32(...) __builtin_sve_svwhilerw_u32(__VA_ARGS__)
+#define svwhilerw_f32(...) __builtin_sve_svwhilerw_f32(__VA_ARGS__)
+#define svwhilerw_s32(...) __builtin_sve_svwhilerw_s32(__VA_ARGS__)
+#define svwhilewr_u8(...) __builtin_sve_svwhilewr_u8(__VA_ARGS__)
+#define svwhilewr_s8(...) __builtin_sve_svwhilewr_s8(__VA_ARGS__)
+#define svwhilewr_u64(...) __builtin_sve_svwhilewr_u64(__VA_ARGS__)
+#define svwhilewr_f64(...) __builtin_sve_svwhilewr_f64(__VA_ARGS__)
+#define svwhilewr_s64(...) __builtin_sve_svwhilewr_s64(__VA_ARGS__)
+#define svwhilewr_u16(...) __builtin_sve_svwhilewr_u16(__VA_ARGS__)
+#define svwhilewr_f16(...) __builtin_sve_svwhilewr_f16(__VA_ARGS__)
+#define svwhilewr_s16(...) __builtin_sve_svwhilewr_s16(__VA_ARGS__)
+#define svwhilewr_u32(...) __builtin_sve_svwhilewr_u32(__VA_ARGS__)
+#define svwhilewr_f32(...) __builtin_sve_svwhilewr_f32(__VA_ARGS__)
+#define svwhilewr_s32(...) __builtin_sve_svwhilewr_s32(__VA_ARGS__)
+#define svxar_n_u8(...) __builtin_sve_svxar_n_u8(__VA_ARGS__)
+#define svxar_n_u32(...) __builtin_sve_svxar_n_u32(__VA_ARGS__)
+#define svxar_n_u64(...) __builtin_sve_svxar_n_u64(__VA_ARGS__)
+#define svxar_n_u16(...) __builtin_sve_svxar_n_u16(__VA_ARGS__)
+#define svxar_n_s8(...) __builtin_sve_svxar_n_s8(__VA_ARGS__)
+#define svxar_n_s32(...) __builtin_sve_svxar_n_s32(__VA_ARGS__)
+#define svxar_n_s64(...) __builtin_sve_svxar_n_s64(__VA_ARGS__)
+#define svxar_n_s16(...) __builtin_sve_svxar_n_s16(__VA_ARGS__)
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s8)))
+svint8_t svaba(svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s32)))
+svint32_t svaba(svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s64)))
+svint64_t svaba(svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s16)))
+svint16_t svaba(svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u8)))
+svuint8_t svaba(svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u32)))
+svuint32_t svaba(svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u64)))
+svuint64_t svaba(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u16)))
+svuint16_t svaba(svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s8)))
+svint8_t svaba(svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s32)))
+svint32_t svaba(svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s64)))
+svint64_t svaba(svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s16)))
+svint16_t svaba(svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u8)))
+svuint8_t svaba(svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u32)))
+svuint32_t svaba(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u64)))
+svuint64_t svaba(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u16)))
+svuint16_t svaba(svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_s32)))
+svint32_t svabalb(svint32_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_s64)))
+svint64_t svabalb(svint64_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_s16)))
+svint16_t svabalb(svint16_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_u32)))
+svuint32_t svabalb(svuint32_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_u64)))
+svuint64_t svabalb(svuint64_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_u16)))
+svuint16_t svabalb(svuint16_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_s32)))
+svint32_t svabalb(svint32_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_s64)))
+svint64_t svabalb(svint64_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_s16)))
+svint16_t svabalb(svint16_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_u32)))
+svuint32_t svabalb(svuint32_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_u64)))
+svuint64_t svabalb(svuint64_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_u16)))
+svuint16_t svabalb(svuint16_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_s32)))
+svint32_t svabalt(svint32_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_s64)))
+svint64_t svabalt(svint64_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_s16)))
+svint16_t svabalt(svint16_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_u32)))
+svuint32_t svabalt(svuint32_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_u64)))
+svuint64_t svabalt(svuint64_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_u16)))
+svuint16_t svabalt(svuint16_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_s32)))
+svint32_t svabalt(svint32_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_s64)))
+svint64_t svabalt(svint64_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_s16)))
+svint16_t svabalt(svint16_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_u32)))
+svuint32_t svabalt(svuint32_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_u64)))
+svuint64_t svabalt(svuint64_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_u16)))
+svuint16_t svabalt(svuint16_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_s32)))
+svint32_t svabdlb(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_s64)))
+svint64_t svabdlb(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_s16)))
+svint16_t svabdlb(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_u32)))
+svuint32_t svabdlb(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_u64)))
+svuint64_t svabdlb(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_u16)))
+svuint16_t svabdlb(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_s32)))
+svint32_t svabdlb(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_s64)))
+svint64_t svabdlb(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_s16)))
+svint16_t svabdlb(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_u32)))
+svuint32_t svabdlb(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_u64)))
+svuint64_t svabdlb(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_u16)))
+svuint16_t svabdlb(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_s32)))
+svint32_t svabdlt(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_s64)))
+svint64_t svabdlt(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_s16)))
+svint16_t svabdlt(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_u32)))
+svuint32_t svabdlt(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_u64)))
+svuint64_t svabdlt(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_u16)))
+svuint16_t svabdlt(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_s32)))
+svint32_t svabdlt(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_s64)))
+svint64_t svabdlt(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_s16)))
+svint16_t svabdlt(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_u32)))
+svuint32_t svabdlt(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_u64)))
+svuint64_t svabdlt(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_u16)))
+svuint16_t svabdlt(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s32_m)))
+svint32_t svadalp_m(svbool_t, svint32_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s64_m)))
+svint64_t svadalp_m(svbool_t, svint64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s16_m)))
+svint16_t svadalp_m(svbool_t, svint16_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s32_x)))
+svint32_t svadalp_x(svbool_t, svint32_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s64_x)))
+svint64_t svadalp_x(svbool_t, svint64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s16_x)))
+svint16_t svadalp_x(svbool_t, svint16_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s32_z)))
+svint32_t svadalp_z(svbool_t, svint32_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s64_z)))
+svint64_t svadalp_z(svbool_t, svint64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s16_z)))
+svint16_t svadalp_z(svbool_t, svint16_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u32_m)))
+svuint32_t svadalp_m(svbool_t, svuint32_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u64_m)))
+svuint64_t svadalp_m(svbool_t, svuint64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u16_m)))
+svuint16_t svadalp_m(svbool_t, svuint16_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u32_x)))
+svuint32_t svadalp_x(svbool_t, svuint32_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u64_x)))
+svuint64_t svadalp_x(svbool_t, svuint64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u16_x)))
+svuint16_t svadalp_x(svbool_t, svuint16_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u32_z)))
+svuint32_t svadalp_z(svbool_t, svuint32_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u64_z)))
+svuint64_t svadalp_z(svbool_t, svuint64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u16_z)))
+svuint16_t svadalp_z(svbool_t, svuint16_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_n_u32)))
+svuint32_t svadclb(svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_n_u64)))
+svuint64_t svadclb(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_u32)))
+svuint32_t svadclb(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_u64)))
+svuint64_t svadclb(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_n_u32)))
+svuint32_t svadclt(svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_n_u64)))
+svuint64_t svadclt(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_u32)))
+svuint32_t svadclt(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_u64)))
+svuint64_t svadclt(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_u32)))
+svuint16_t svaddhnb(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_u64)))
+svuint32_t svaddhnb(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_u16)))
+svuint8_t svaddhnb(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_s32)))
+svint16_t svaddhnb(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_s64)))
+svint32_t svaddhnb(svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_s16)))
+svint8_t svaddhnb(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_u32)))
+svuint16_t svaddhnb(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_u64)))
+svuint32_t svaddhnb(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_u16)))
+svuint8_t svaddhnb(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_s32)))
+svint16_t svaddhnb(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_s64)))
+svint32_t svaddhnb(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_s16)))
+svint8_t svaddhnb(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_u32)))
+svuint16_t svaddhnt(svuint16_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_u64)))
+svuint32_t svaddhnt(svuint32_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_u16)))
+svuint8_t svaddhnt(svuint8_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_s32)))
+svint16_t svaddhnt(svint16_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_s64)))
+svint32_t svaddhnt(svint32_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_s16)))
+svint8_t svaddhnt(svint8_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_u32)))
+svuint16_t svaddhnt(svuint16_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_u64)))
+svuint32_t svaddhnt(svuint32_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_u16)))
+svuint8_t svaddhnt(svuint8_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_s32)))
+svint16_t svaddhnt(svint16_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_s64)))
+svint32_t svaddhnt(svint32_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_s16)))
+svint8_t svaddhnt(svint8_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_s32)))
+svint32_t svaddlb(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_s64)))
+svint64_t svaddlb(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_s16)))
+svint16_t svaddlb(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_u32)))
+svuint32_t svaddlb(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_u64)))
+svuint64_t svaddlb(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_u16)))
+svuint16_t svaddlb(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_s32)))
+svint32_t svaddlb(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_s64)))
+svint64_t svaddlb(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_s16)))
+svint16_t svaddlb(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_u32)))
+svuint32_t svaddlb(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_u64)))
+svuint64_t svaddlb(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_u16)))
+svuint16_t svaddlb(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_n_s32)))
+svint32_t svaddlbt(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_n_s64)))
+svint64_t svaddlbt(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_n_s16)))
+svint16_t svaddlbt(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_s32)))
+svint32_t svaddlbt(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_s64)))
+svint64_t svaddlbt(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_s16)))
+svint16_t svaddlbt(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_s32)))
+svint32_t svaddlt(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_s64)))
+svint64_t svaddlt(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_s16)))
+svint16_t svaddlt(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_u32)))
+svuint32_t svaddlt(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_u64)))
+svuint64_t svaddlt(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_u16)))
+svuint16_t svaddlt(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_s32)))
+svint32_t svaddlt(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_s64)))
+svint64_t svaddlt(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_s16)))
+svint16_t svaddlt(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_u32)))
+svuint32_t svaddlt(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_u64)))
+svuint64_t svaddlt(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_u16)))
+svuint16_t svaddlt(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f64_m)))
+svfloat64_t svaddp_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f32_m)))
+svfloat32_t svaddp_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f16_m)))
+svfloat16_t svaddp_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f64_x)))
+svfloat64_t svaddp_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f32_x)))
+svfloat32_t svaddp_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f16_x)))
+svfloat16_t svaddp_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u8_m)))
+svuint8_t svaddp_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u32_m)))
+svuint32_t svaddp_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u64_m)))
+svuint64_t svaddp_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u16_m)))
+svuint16_t svaddp_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s8_m)))
+svint8_t svaddp_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s32_m)))
+svint32_t svaddp_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s64_m)))
+svint64_t svaddp_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s16_m)))
+svint16_t svaddp_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u8_x)))
+svuint8_t svaddp_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u32_x)))
+svuint32_t svaddp_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u64_x)))
+svuint64_t svaddp_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u16_x)))
+svuint16_t svaddp_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s8_x)))
+svint8_t svaddp_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s32_x)))
+svint32_t svaddp_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s64_x)))
+svint64_t svaddp_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s16_x)))
+svint16_t svaddp_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_s32)))
+svint32_t svaddwb(svint32_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_s64)))
+svint64_t svaddwb(svint64_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_s16)))
+svint16_t svaddwb(svint16_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_u32)))
+svuint32_t svaddwb(svuint32_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_u64)))
+svuint64_t svaddwb(svuint64_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_u16)))
+svuint16_t svaddwb(svuint16_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_s32)))
+svint32_t svaddwb(svint32_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_s64)))
+svint64_t svaddwb(svint64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_s16)))
+svint16_t svaddwb(svint16_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_u32)))
+svuint32_t svaddwb(svuint32_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_u64)))
+svuint64_t svaddwb(svuint64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_u16)))
+svuint16_t svaddwb(svuint16_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_s32)))
+svint32_t svaddwt(svint32_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_s64)))
+svint64_t svaddwt(svint64_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_s16)))
+svint16_t svaddwt(svint16_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_u32)))
+svuint32_t svaddwt(svuint32_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_u64)))
+svuint64_t svaddwt(svuint64_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_u16)))
+svuint16_t svaddwt(svuint16_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_s32)))
+svint32_t svaddwt(svint32_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_s64)))
+svint64_t svaddwt(svint64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_s16)))
+svint16_t svaddwt(svint16_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_u32)))
+svuint32_t svaddwt(svuint32_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_u64)))
+svuint64_t svaddwt(svuint64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_u16)))
+svuint16_t svaddwt(svuint16_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u8)))
+svuint8_t svbcax(svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u32)))
+svuint32_t svbcax(svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u64)))
+svuint64_t svbcax(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u16)))
+svuint16_t svbcax(svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s8)))
+svint8_t svbcax(svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s32)))
+svint32_t svbcax(svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s64)))
+svint64_t svbcax(svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s16)))
+svint16_t svbcax(svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u8)))
+svuint8_t svbcax(svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u32)))
+svuint32_t svbcax(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u64)))
+svuint64_t svbcax(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u16)))
+svuint16_t svbcax(svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s8)))
+svint8_t svbcax(svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s32)))
+svint32_t svbcax(svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s64)))
+svint64_t svbcax(svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s16)))
+svint16_t svbcax(svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u8)))
+svuint8_t svbsl1n(svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u32)))
+svuint32_t svbsl1n(svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u64)))
+svuint64_t svbsl1n(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u16)))
+svuint16_t svbsl1n(svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s8)))
+svint8_t svbsl1n(svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s32)))
+svint32_t svbsl1n(svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s64)))
+svint64_t svbsl1n(svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s16)))
+svint16_t svbsl1n(svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u8)))
+svuint8_t svbsl1n(svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u32)))
+svuint32_t svbsl1n(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u64)))
+svuint64_t svbsl1n(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u16)))
+svuint16_t svbsl1n(svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s8)))
+svint8_t svbsl1n(svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s32)))
+svint32_t svbsl1n(svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s64)))
+svint64_t svbsl1n(svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s16)))
+svint16_t svbsl1n(svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u8)))
+svuint8_t svbsl2n(svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u32)))
+svuint32_t svbsl2n(svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u64)))
+svuint64_t svbsl2n(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u16)))
+svuint16_t svbsl2n(svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s8)))
+svint8_t svbsl2n(svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s32)))
+svint32_t svbsl2n(svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s64)))
+svint64_t svbsl2n(svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s16)))
+svint16_t svbsl2n(svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u8)))
+svuint8_t svbsl2n(svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u32)))
+svuint32_t svbsl2n(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u64)))
+svuint64_t svbsl2n(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u16)))
+svuint16_t svbsl2n(svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s8)))
+svint8_t svbsl2n(svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s32)))
+svint32_t svbsl2n(svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s64)))
+svint64_t svbsl2n(svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s16)))
+svint16_t svbsl2n(svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u8)))
+svuint8_t svbsl(svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u32)))
+svuint32_t svbsl(svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u64)))
+svuint64_t svbsl(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u16)))
+svuint16_t svbsl(svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s8)))
+svint8_t svbsl(svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s32)))
+svint32_t svbsl(svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s64)))
+svint64_t svbsl(svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s16)))
+svint16_t svbsl(svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u8)))
+svuint8_t svbsl(svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u32)))
+svuint32_t svbsl(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u64)))
+svuint64_t svbsl(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u16)))
+svuint16_t svbsl(svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s8)))
+svint8_t svbsl(svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s32)))
+svint32_t svbsl(svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s64)))
+svint64_t svbsl(svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s16)))
+svint16_t svbsl(svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u8)))
+svuint8_t svcadd(svuint8_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u32)))
+svuint32_t svcadd(svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u64)))
+svuint64_t svcadd(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u16)))
+svuint16_t svcadd(svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s8)))
+svint8_t svcadd(svint8_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s32)))
+svint32_t svcadd(svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s64)))
+svint64_t svcadd(svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s16)))
+svint16_t svcadd(svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_s32)))
+svint32_t svcdot(svint32_t, svint8_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_s64)))
+svint64_t svcdot(svint64_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_lane_s32)))
+svint32_t svcdot_lane(svint32_t, svint8_t, svint8_t, uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_lane_s64)))
+svint64_t svcdot_lane(svint64_t, svint16_t, svint16_t, uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u8)))
+svuint8_t svcmla(svuint8_t, svuint8_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u32)))
+svuint32_t svcmla(svuint32_t, svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u64)))
+svuint64_t svcmla(svuint64_t, svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u16)))
+svuint16_t svcmla(svuint16_t, svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s8)))
+svint8_t svcmla(svint8_t, svint8_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s32)))
+svint32_t svcmla(svint32_t, svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s64)))
+svint64_t svcmla(svint64_t, svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s16)))
+svint16_t svcmla(svint16_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_u32)))
+svuint32_t svcmla_lane(svuint32_t, svuint32_t, svuint32_t, uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_u16)))
+svuint16_t svcmla_lane(svuint16_t, svuint16_t, svuint16_t, uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_s32)))
+svint32_t svcmla_lane(svint32_t, svint32_t, svint32_t, uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_s16)))
+svint16_t svcmla_lane(svint16_t, svint16_t, svint16_t, uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f32_f16_m)))
+svfloat32_t svcvtlt_f32_m(svfloat32_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f32_f16_x)))
+svfloat32_t svcvtlt_f32_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f64_f32_m)))
+svfloat64_t svcvtlt_f64_m(svfloat64_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f64_f32_x)))
+svfloat64_t svcvtlt_f64_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtnt_f16_f32_m)))
+svfloat16_t svcvtnt_f16_m(svfloat16_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtnt_f32_f64_m)))
+svfloat32_t svcvtnt_f32_m(svfloat32_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtx_f32_f64_m)))
+svfloat32_t svcvtx_f32_m(svfloat32_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtx_f32_f64_x)))
+svfloat32_t svcvtx_f32_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtx_f32_f64_z)))
+svfloat32_t svcvtx_f32_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtxnt_f32_f64_m)))
+svfloat32_t svcvtxnt_f32_m(svfloat32_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u8)))
+svuint8_t sveor3(svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u32)))
+svuint32_t sveor3(svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u64)))
+svuint64_t sveor3(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u16)))
+svuint16_t sveor3(svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s8)))
+svint8_t sveor3(svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s32)))
+svint32_t sveor3(svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s64)))
+svint64_t sveor3(svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s16)))
+svint16_t sveor3(svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u8)))
+svuint8_t sveor3(svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u32)))
+svuint32_t sveor3(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u64)))
+svuint64_t sveor3(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u16)))
+svuint16_t sveor3(svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s8)))
+svint8_t sveor3(svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s32)))
+svint32_t sveor3(svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s64)))
+svint64_t sveor3(svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s16)))
+svint16_t sveor3(svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u8)))
+svuint8_t sveorbt(svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u32)))
+svuint32_t sveorbt(svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u64)))
+svuint64_t sveorbt(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u16)))
+svuint16_t sveorbt(svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s8)))
+svint8_t sveorbt(svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s32)))
+svint32_t sveorbt(svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s64)))
+svint64_t sveorbt(svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s16)))
+svint16_t sveorbt(svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u8)))
+svuint8_t sveorbt(svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u32)))
+svuint32_t sveorbt(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u64)))
+svuint64_t sveorbt(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u16)))
+svuint16_t sveorbt(svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s8)))
+svint8_t sveorbt(svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s32)))
+svint32_t sveorbt(svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s64)))
+svint64_t sveorbt(svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s16)))
+svint16_t sveorbt(svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u8)))
+svuint8_t sveortb(svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u32)))
+svuint32_t sveortb(svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u64)))
+svuint64_t sveortb(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u16)))
+svuint16_t sveortb(svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s8)))
+svint8_t sveortb(svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s32)))
+svint32_t sveortb(svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s64)))
+svint64_t sveortb(svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s16)))
+svint16_t sveortb(svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u8)))
+svuint8_t sveortb(svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u32)))
+svuint32_t sveortb(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u64)))
+svuint64_t sveortb(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u16)))
+svuint16_t sveortb(svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s8)))
+svint8_t sveortb(svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s32)))
+svint32_t sveortb(svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s64)))
+svint64_t sveortb(svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s16)))
+svint16_t sveortb(svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s8_m)))
+svint8_t svhadd_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s32_m)))
+svint32_t svhadd_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s64_m)))
+svint64_t svhadd_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s16_m)))
+svint16_t svhadd_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s8_x)))
+svint8_t svhadd_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s32_x)))
+svint32_t svhadd_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s64_x)))
+svint64_t svhadd_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s16_x)))
+svint16_t svhadd_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s8_z)))
+svint8_t svhadd_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s32_z)))
+svint32_t svhadd_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s64_z)))
+svint64_t svhadd_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s16_z)))
+svint16_t svhadd_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u8_m)))
+svuint8_t svhadd_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u32_m)))
+svuint32_t svhadd_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u64_m)))
+svuint64_t svhadd_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u16_m)))
+svuint16_t svhadd_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u8_x)))
+svuint8_t svhadd_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u32_x)))
+svuint32_t svhadd_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u64_x)))
+svuint64_t svhadd_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u16_x)))
+svuint16_t svhadd_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u8_z)))
+svuint8_t svhadd_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u32_z)))
+svuint32_t svhadd_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u64_z)))
+svuint64_t svhadd_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u16_z)))
+svuint16_t svhadd_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s8_m)))
+svint8_t svhadd_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s32_m)))
+svint32_t svhadd_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s64_m)))
+svint64_t svhadd_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s16_m)))
+svint16_t svhadd_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s8_x)))
+svint8_t svhadd_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s32_x)))
+svint32_t svhadd_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s64_x)))
+svint64_t svhadd_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s16_x)))
+svint16_t svhadd_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s8_z)))
+svint8_t svhadd_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s32_z)))
+svint32_t svhadd_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s64_z)))
+svint64_t svhadd_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s16_z)))
+svint16_t svhadd_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u8_m)))
+svuint8_t svhadd_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u32_m)))
+svuint32_t svhadd_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u64_m)))
+svuint64_t svhadd_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u16_m)))
+svuint16_t svhadd_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u8_x)))
+svuint8_t svhadd_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u32_x)))
+svuint32_t svhadd_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u64_x)))
+svuint64_t svhadd_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u16_x)))
+svuint16_t svhadd_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u8_z)))
+svuint8_t svhadd_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u32_z)))
+svuint32_t svhadd_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u64_z)))
+svuint64_t svhadd_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u16_z)))
+svuint16_t svhadd_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_u32_z)))
+svuint32_t svhistcnt_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_u64_z)))
+svuint64_t svhistcnt_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_s32_z)))
+svuint32_t svhistcnt_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_s64_z)))
+svuint64_t svhistcnt_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistseg_u8)))
+svuint8_t svhistseg(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistseg_s8)))
+svuint8_t svhistseg(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s8_m)))
+svint8_t svhsub_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s32_m)))
+svint32_t svhsub_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s64_m)))
+svint64_t svhsub_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s16_m)))
+svint16_t svhsub_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s8_x)))
+svint8_t svhsub_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s32_x)))
+svint32_t svhsub_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s64_x)))
+svint64_t svhsub_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s16_x)))
+svint16_t svhsub_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s8_z)))
+svint8_t svhsub_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s32_z)))
+svint32_t svhsub_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s64_z)))
+svint64_t svhsub_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s16_z)))
+svint16_t svhsub_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u8_m)))
+svuint8_t svhsub_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u32_m)))
+svuint32_t svhsub_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u64_m)))
+svuint64_t svhsub_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u16_m)))
+svuint16_t svhsub_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u8_x)))
+svuint8_t svhsub_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u32_x)))
+svuint32_t svhsub_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u64_x)))
+svuint64_t svhsub_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u16_x)))
+svuint16_t svhsub_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u8_z)))
+svuint8_t svhsub_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u32_z)))
+svuint32_t svhsub_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u64_z)))
+svuint64_t svhsub_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u16_z)))
+svuint16_t svhsub_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s8_m)))
+svint8_t svhsub_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s32_m)))
+svint32_t svhsub_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s64_m)))
+svint64_t svhsub_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s16_m)))
+svint16_t svhsub_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s8_x)))
+svint8_t svhsub_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s32_x)))
+svint32_t svhsub_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s64_x)))
+svint64_t svhsub_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s16_x)))
+svint16_t svhsub_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s8_z)))
+svint8_t svhsub_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s32_z)))
+svint32_t svhsub_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s64_z)))
+svint64_t svhsub_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s16_z)))
+svint16_t svhsub_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u8_m)))
+svuint8_t svhsub_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u32_m)))
+svuint32_t svhsub_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u64_m)))
+svuint64_t svhsub_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u16_m)))
+svuint16_t svhsub_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u8_x)))
+svuint8_t svhsub_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u32_x)))
+svuint32_t svhsub_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u64_x)))
+svuint64_t svhsub_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u16_x)))
+svuint16_t svhsub_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u8_z)))
+svuint8_t svhsub_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u32_z)))
+svuint32_t svhsub_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u64_z)))
+svuint64_t svhsub_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u16_z)))
+svuint16_t svhsub_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s8_m)))
+svint8_t svhsubr_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s32_m)))
+svint32_t svhsubr_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s64_m)))
+svint64_t svhsubr_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s16_m)))
+svint16_t svhsubr_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s8_x)))
+svint8_t svhsubr_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s32_x)))
+svint32_t svhsubr_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s64_x)))
+svint64_t svhsubr_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s16_x)))
+svint16_t svhsubr_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s8_z)))
+svint8_t svhsubr_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s32_z)))
+svint32_t svhsubr_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s64_z)))
+svint64_t svhsubr_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s16_z)))
+svint16_t svhsubr_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u8_m)))
+svuint8_t svhsubr_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u32_m)))
+svuint32_t svhsubr_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u64_m)))
+svuint64_t svhsubr_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u16_m)))
+svuint16_t svhsubr_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u8_x)))
+svuint8_t svhsubr_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u32_x)))
+svuint32_t svhsubr_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u64_x)))
+svuint64_t svhsubr_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u16_x)))
+svuint16_t svhsubr_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u8_z)))
+svuint8_t svhsubr_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u32_z)))
+svuint32_t svhsubr_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u64_z)))
+svuint64_t svhsubr_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u16_z)))
+svuint16_t svhsubr_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s8_m)))
+svint8_t svhsubr_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s32_m)))
+svint32_t svhsubr_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s64_m)))
+svint64_t svhsubr_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s16_m)))
+svint16_t svhsubr_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s8_x)))
+svint8_t svhsubr_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s32_x)))
+svint32_t svhsubr_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s64_x)))
+svint64_t svhsubr_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s16_x)))
+svint16_t svhsubr_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s8_z)))
+svint8_t svhsubr_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s32_z)))
+svint32_t svhsubr_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s64_z)))
+svint64_t svhsubr_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s16_z)))
+svint16_t svhsubr_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u8_m)))
+svuint8_t svhsubr_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u32_m)))
+svuint32_t svhsubr_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u64_m)))
+svuint64_t svhsubr_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u16_m)))
+svuint16_t svhsubr_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u8_x)))
+svuint8_t svhsubr_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u32_x)))
+svuint32_t svhsubr_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u64_x)))
+svuint64_t svhsubr_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u16_x)))
+svuint16_t svhsubr_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u8_z)))
+svuint8_t svhsubr_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u32_z)))
+svuint32_t svhsubr_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u64_z)))
+svuint64_t svhsubr_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u16_z)))
+svuint16_t svhsubr_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_index_u32)))
+svuint32_t svldnt1_gather_index_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_index_u64)))
+svuint64_t svldnt1_gather_index_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_index_f64)))
+svfloat64_t svldnt1_gather_index_f64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_index_f32)))
+svfloat32_t svldnt1_gather_index_f32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_index_s32)))
+svint32_t svldnt1_gather_index_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_index_s64)))
+svint64_t svldnt1_gather_index_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_offset_u32)))
+svuint32_t svldnt1_gather_offset_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_offset_u64)))
+svuint64_t svldnt1_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_offset_f64)))
+svfloat64_t svldnt1_gather_offset_f64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_offset_f32)))
+svfloat32_t svldnt1_gather_offset_f32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_offset_s32)))
+svint32_t svldnt1_gather_offset_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_offset_s64)))
+svint64_t svldnt1_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_u32)))
+svuint32_t svldnt1_gather_u32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_u64)))
+svuint64_t svldnt1_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_f64)))
+svfloat64_t svldnt1_gather_f64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_f32)))
+svfloat32_t svldnt1_gather_f32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_s32)))
+svint32_t svldnt1_gather_s32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_s64)))
+svint64_t svldnt1_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64index_u64)))
+svuint64_t svldnt1_gather_index(svbool_t, uint64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64index_f64)))
+svfloat64_t svldnt1_gather_index(svbool_t, float64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64index_s64)))
+svint64_t svldnt1_gather_index(svbool_t, int64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64index_u64)))
+svuint64_t svldnt1_gather_index(svbool_t, uint64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64index_f64)))
+svfloat64_t svldnt1_gather_index(svbool_t, float64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64index_s64)))
+svint64_t svldnt1_gather_index(svbool_t, int64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32offset_u32)))
+svuint32_t svldnt1_gather_offset(svbool_t, uint32_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32offset_f32)))
+svfloat32_t svldnt1_gather_offset(svbool_t, float32_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32offset_s32)))
+svint32_t svldnt1_gather_offset(svbool_t, int32_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64offset_u64)))
+svuint64_t svldnt1_gather_offset(svbool_t, uint64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64offset_f64)))
+svfloat64_t svldnt1_gather_offset(svbool_t, float64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64offset_s64)))
+svint64_t svldnt1_gather_offset(svbool_t, int64_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64offset_u64)))
+svuint64_t svldnt1_gather_offset(svbool_t, uint64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64offset_f64)))
+svfloat64_t svldnt1_gather_offset(svbool_t, float64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64offset_s64)))
+svint64_t svldnt1_gather_offset(svbool_t, int64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_offset_u32)))
+svuint32_t svldnt1sb_gather_offset_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_offset_u64)))
+svuint64_t svldnt1sb_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_offset_s32)))
+svint32_t svldnt1sb_gather_offset_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_offset_s64)))
+svint64_t svldnt1sb_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_u32)))
+svuint32_t svldnt1sb_gather_u32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_u64)))
+svuint64_t svldnt1sb_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_s32)))
+svint32_t svldnt1sb_gather_s32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_s64)))
+svint64_t svldnt1sb_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32offset_u32)))
+svuint32_t svldnt1sb_gather_offset_u32(svbool_t, int8_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32offset_s32)))
+svint32_t svldnt1sb_gather_offset_s32(svbool_t, int8_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_s64offset_u64)))
+svuint64_t svldnt1sb_gather_offset_u64(svbool_t, int8_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_s64offset_s64)))
+svint64_t svldnt1sb_gather_offset_s64(svbool_t, int8_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64offset_u64)))
+svuint64_t svldnt1sb_gather_offset_u64(svbool_t, int8_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64offset_s64)))
+svint64_t svldnt1sb_gather_offset_s64(svbool_t, int8_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_index_u32)))
+svuint32_t svldnt1sh_gather_index_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_index_u64)))
+svuint64_t svldnt1sh_gather_index_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_index_s32)))
+svint32_t svldnt1sh_gather_index_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_index_s64)))
+svint64_t svldnt1sh_gather_index_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_offset_u32)))
+svuint32_t svldnt1sh_gather_offset_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_offset_u64)))
+svuint64_t svldnt1sh_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_offset_s32)))
+svint32_t svldnt1sh_gather_offset_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_offset_s64)))
+svint64_t svldnt1sh_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_u32)))
+svuint32_t svldnt1sh_gather_u32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_u64)))
+svuint64_t svldnt1sh_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_s32)))
+svint32_t svldnt1sh_gather_s32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_s64)))
+svint64_t svldnt1sh_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64index_u64)))
+svuint64_t svldnt1sh_gather_index_u64(svbool_t, int16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64index_s64)))
+svint64_t svldnt1sh_gather_index_s64(svbool_t, int16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64index_u64)))
+svuint64_t svldnt1sh_gather_index_u64(svbool_t, int16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64index_s64)))
+svint64_t svldnt1sh_gather_index_s64(svbool_t, int16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32offset_u32)))
+svuint32_t svldnt1sh_gather_offset_u32(svbool_t, int16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32offset_s32)))
+svint32_t svldnt1sh_gather_offset_s32(svbool_t, int16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64offset_u64)))
+svuint64_t svldnt1sh_gather_offset_u64(svbool_t, int16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64offset_s64)))
+svint64_t svldnt1sh_gather_offset_s64(svbool_t, int16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64offset_u64)))
+svuint64_t svldnt1sh_gather_offset_u64(svbool_t, int16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64offset_s64)))
+svint64_t svldnt1sh_gather_offset_s64(svbool_t, int16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_index_u64)))
+svuint64_t svldnt1sw_gather_index_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_index_s64)))
+svint64_t svldnt1sw_gather_index_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_offset_u64)))
+svuint64_t svldnt1sw_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_offset_s64)))
+svint64_t svldnt1sw_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_u64)))
+svuint64_t svldnt1sw_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_s64)))
+svint64_t svldnt1sw_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64index_u64)))
+svuint64_t svldnt1sw_gather_index_u64(svbool_t, int32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64index_s64)))
+svint64_t svldnt1sw_gather_index_s64(svbool_t, int32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64index_u64)))
+svuint64_t svldnt1sw_gather_index_u64(svbool_t, int32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64index_s64)))
+svint64_t svldnt1sw_gather_index_s64(svbool_t, int32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64offset_u64)))
+svuint64_t svldnt1sw_gather_offset_u64(svbool_t, int32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64offset_s64)))
+svint64_t svldnt1sw_gather_offset_s64(svbool_t, int32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64offset_u64)))
+svuint64_t svldnt1sw_gather_offset_u64(svbool_t, int32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64offset_s64)))
+svint64_t svldnt1sw_gather_offset_s64(svbool_t, int32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_offset_u32)))
+svuint32_t svldnt1ub_gather_offset_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_offset_u64)))
+svuint64_t svldnt1ub_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_offset_s32)))
+svint32_t svldnt1ub_gather_offset_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_offset_s64)))
+svint64_t svldnt1ub_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_u32)))
+svuint32_t svldnt1ub_gather_u32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_u64)))
+svuint64_t svldnt1ub_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_s32)))
+svint32_t svldnt1ub_gather_s32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_s64)))
+svint64_t svldnt1ub_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32offset_u32)))
+svuint32_t svldnt1ub_gather_offset_u32(svbool_t, uint8_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32offset_s32)))
+svint32_t svldnt1ub_gather_offset_s32(svbool_t, uint8_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_s64offset_u64)))
+svuint64_t svldnt1ub_gather_offset_u64(svbool_t, uint8_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_s64offset_s64)))
+svint64_t svldnt1ub_gather_offset_s64(svbool_t, uint8_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64offset_u64)))
+svuint64_t svldnt1ub_gather_offset_u64(svbool_t, uint8_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64offset_s64)))
+svint64_t svldnt1ub_gather_offset_s64(svbool_t, uint8_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_index_u32)))
+svuint32_t svldnt1uh_gather_index_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_index_u64)))
+svuint64_t svldnt1uh_gather_index_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_index_s32)))
+svint32_t svldnt1uh_gather_index_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_index_s64)))
+svint64_t svldnt1uh_gather_index_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_offset_u32)))
+svuint32_t svldnt1uh_gather_offset_u32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_offset_u64)))
+svuint64_t svldnt1uh_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_offset_s32)))
+svint32_t svldnt1uh_gather_offset_s32(svbool_t, svuint32_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_offset_s64)))
+svint64_t svldnt1uh_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_u32)))
+svuint32_t svldnt1uh_gather_u32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_u64)))
+svuint64_t svldnt1uh_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_s32)))
+svint32_t svldnt1uh_gather_s32(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_s64)))
+svint64_t svldnt1uh_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64index_u64)))
+svuint64_t svldnt1uh_gather_index_u64(svbool_t, uint16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64index_s64)))
+svint64_t svldnt1uh_gather_index_s64(svbool_t, uint16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64index_u64)))
+svuint64_t svldnt1uh_gather_index_u64(svbool_t, uint16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64index_s64)))
+svint64_t svldnt1uh_gather_index_s64(svbool_t, uint16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32offset_u32)))
+svuint32_t svldnt1uh_gather_offset_u32(svbool_t, uint16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32offset_s32)))
+svint32_t svldnt1uh_gather_offset_s32(svbool_t, uint16_t const *, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64offset_u64)))
+svuint64_t svldnt1uh_gather_offset_u64(svbool_t, uint16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64offset_s64)))
+svint64_t svldnt1uh_gather_offset_s64(svbool_t, uint16_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64offset_u64)))
+svuint64_t svldnt1uh_gather_offset_u64(svbool_t, uint16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64offset_s64)))
+svint64_t svldnt1uh_gather_offset_s64(svbool_t, uint16_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_index_u64)))
+svuint64_t svldnt1uw_gather_index_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_index_s64)))
+svint64_t svldnt1uw_gather_index_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_offset_u64)))
+svuint64_t svldnt1uw_gather_offset_u64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_offset_s64)))
+svint64_t svldnt1uw_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_u64)))
+svuint64_t svldnt1uw_gather_u64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_s64)))
+svint64_t svldnt1uw_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64index_u64)))
+svuint64_t svldnt1uw_gather_index_u64(svbool_t, uint32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64index_s64)))
+svint64_t svldnt1uw_gather_index_s64(svbool_t, uint32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64index_u64)))
+svuint64_t svldnt1uw_gather_index_u64(svbool_t, uint32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64index_s64)))
+svint64_t svldnt1uw_gather_index_s64(svbool_t, uint32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64offset_u64)))
+svuint64_t svldnt1uw_gather_offset_u64(svbool_t, uint32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64offset_s64)))
+svint64_t svldnt1uw_gather_offset_s64(svbool_t, uint32_t const *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64offset_u64)))
+svuint64_t svldnt1uw_gather_offset_u64(svbool_t, uint32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64offset_s64)))
+svint64_t svldnt1uw_gather_offset_s64(svbool_t, uint32_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f64_m)))
+svint64_t svlogb_m(svint64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f32_m)))
+svint32_t svlogb_m(svint32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f16_m)))
+svint16_t svlogb_m(svint16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f64_x)))
+svint64_t svlogb_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f32_x)))
+svint32_t svlogb_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f16_x)))
+svint16_t svlogb_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f64_z)))
+svint64_t svlogb_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f32_z)))
+svint32_t svlogb_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f16_z)))
+svint16_t svlogb_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_u8)))
+svbool_t svmatch(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_u16)))
+svbool_t svmatch(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_s8)))
+svbool_t svmatch(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_s16)))
+svbool_t svmatch(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f64_m)))
+svfloat64_t svmaxnmp_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f32_m)))
+svfloat32_t svmaxnmp_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f16_m)))
+svfloat16_t svmaxnmp_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f64_x)))
+svfloat64_t svmaxnmp_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f32_x)))
+svfloat32_t svmaxnmp_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f16_x)))
+svfloat16_t svmaxnmp_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f64_m)))
+svfloat64_t svmaxp_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f32_m)))
+svfloat32_t svmaxp_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f16_m)))
+svfloat16_t svmaxp_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f64_x)))
+svfloat64_t svmaxp_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f32_x)))
+svfloat32_t svmaxp_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f16_x)))
+svfloat16_t svmaxp_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s8_m)))
+svint8_t svmaxp_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s32_m)))
+svint32_t svmaxp_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s64_m)))
+svint64_t svmaxp_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s16_m)))
+svint16_t svmaxp_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s8_x)))
+svint8_t svmaxp_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s32_x)))
+svint32_t svmaxp_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s64_x)))
+svint64_t svmaxp_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s16_x)))
+svint16_t svmaxp_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u8_m)))
+svuint8_t svmaxp_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u32_m)))
+svuint32_t svmaxp_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u64_m)))
+svuint64_t svmaxp_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u16_m)))
+svuint16_t svmaxp_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u8_x)))
+svuint8_t svmaxp_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u32_x)))
+svuint32_t svmaxp_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u64_x)))
+svuint64_t svmaxp_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u16_x)))
+svuint16_t svmaxp_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f64_m)))
+svfloat64_t svminnmp_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f32_m)))
+svfloat32_t svminnmp_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f16_m)))
+svfloat16_t svminnmp_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f64_x)))
+svfloat64_t svminnmp_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f32_x)))
+svfloat32_t svminnmp_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f16_x)))
+svfloat16_t svminnmp_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f64_m)))
+svfloat64_t svminp_m(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f32_m)))
+svfloat32_t svminp_m(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f16_m)))
+svfloat16_t svminp_m(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f64_x)))
+svfloat64_t svminp_x(svbool_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f32_x)))
+svfloat32_t svminp_x(svbool_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f16_x)))
+svfloat16_t svminp_x(svbool_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s8_m)))
+svint8_t svminp_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s32_m)))
+svint32_t svminp_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s64_m)))
+svint64_t svminp_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s16_m)))
+svint16_t svminp_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s8_x)))
+svint8_t svminp_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s32_x)))
+svint32_t svminp_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s64_x)))
+svint64_t svminp_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s16_x)))
+svint16_t svminp_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u8_m)))
+svuint8_t svminp_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u32_m)))
+svuint32_t svminp_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u64_m)))
+svuint64_t svminp_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u16_m)))
+svuint16_t svminp_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u8_x)))
+svuint8_t svminp_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u32_x)))
+svuint32_t svminp_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u64_x)))
+svuint64_t svminp_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u16_x)))
+svuint16_t svminp_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_u32)))
+svuint32_t svmla_lane(svuint32_t, svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_u64)))
+svuint64_t svmla_lane(svuint64_t, svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_u16)))
+svuint16_t svmla_lane(svuint16_t, svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_s32)))
+svint32_t svmla_lane(svint32_t, svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_s64)))
+svint64_t svmla_lane(svint64_t, svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_s16)))
+svint16_t svmla_lane(svint16_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_f32)))
+svfloat32_t svmlalb(svfloat32_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_s32)))
+svint32_t svmlalb(svint32_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_s64)))
+svint64_t svmlalb(svint64_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_s16)))
+svint16_t svmlalb(svint16_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_u32)))
+svuint32_t svmlalb(svuint32_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_u64)))
+svuint64_t svmlalb(svuint64_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_u16)))
+svuint16_t svmlalb(svuint16_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_f32)))
+svfloat32_t svmlalb(svfloat32_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_s32)))
+svint32_t svmlalb(svint32_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_s64)))
+svint64_t svmlalb(svint64_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_s16)))
+svint16_t svmlalb(svint16_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_u32)))
+svuint32_t svmlalb(svuint32_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_u64)))
+svuint64_t svmlalb(svuint64_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_u16)))
+svuint16_t svmlalb(svuint16_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_f32)))
+svfloat32_t svmlalb_lane(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_s32)))
+svint32_t svmlalb_lane(svint32_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_s64)))
+svint64_t svmlalb_lane(svint64_t, svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_u32)))
+svuint32_t svmlalb_lane(svuint32_t, svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_u64)))
+svuint64_t svmlalb_lane(svuint64_t, svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_f32)))
+svfloat32_t svmlalt(svfloat32_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_s32)))
+svint32_t svmlalt(svint32_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_s64)))
+svint64_t svmlalt(svint64_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_s16)))
+svint16_t svmlalt(svint16_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_u32)))
+svuint32_t svmlalt(svuint32_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_u64)))
+svuint64_t svmlalt(svuint64_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_u16)))
+svuint16_t svmlalt(svuint16_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_f32)))
+svfloat32_t svmlalt(svfloat32_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_s32)))
+svint32_t svmlalt(svint32_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_s64)))
+svint64_t svmlalt(svint64_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_s16)))
+svint16_t svmlalt(svint16_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_u32)))
+svuint32_t svmlalt(svuint32_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_u64)))
+svuint64_t svmlalt(svuint64_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_u16)))
+svuint16_t svmlalt(svuint16_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_f32)))
+svfloat32_t svmlalt_lane(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_s32)))
+svint32_t svmlalt_lane(svint32_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_s64)))
+svint64_t svmlalt_lane(svint64_t, svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_u32)))
+svuint32_t svmlalt_lane(svuint32_t, svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_u64)))
+svuint64_t svmlalt_lane(svuint64_t, svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_u32)))
+svuint32_t svmls_lane(svuint32_t, svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_u64)))
+svuint64_t svmls_lane(svuint64_t, svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_u16)))
+svuint16_t svmls_lane(svuint16_t, svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_s32)))
+svint32_t svmls_lane(svint32_t, svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_s64)))
+svint64_t svmls_lane(svint64_t, svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_s16)))
+svint16_t svmls_lane(svint16_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_f32)))
+svfloat32_t svmlslb(svfloat32_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_s32)))
+svint32_t svmlslb(svint32_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_s64)))
+svint64_t svmlslb(svint64_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_s16)))
+svint16_t svmlslb(svint16_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_u32)))
+svuint32_t svmlslb(svuint32_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_u64)))
+svuint64_t svmlslb(svuint64_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_u16)))
+svuint16_t svmlslb(svuint16_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_f32)))
+svfloat32_t svmlslb(svfloat32_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_s32)))
+svint32_t svmlslb(svint32_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_s64)))
+svint64_t svmlslb(svint64_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_s16)))
+svint16_t svmlslb(svint16_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_u32)))
+svuint32_t svmlslb(svuint32_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_u64)))
+svuint64_t svmlslb(svuint64_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_u16)))
+svuint16_t svmlslb(svuint16_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_f32)))
+svfloat32_t svmlslb_lane(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_s32)))
+svint32_t svmlslb_lane(svint32_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_s64)))
+svint64_t svmlslb_lane(svint64_t, svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_u32)))
+svuint32_t svmlslb_lane(svuint32_t, svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_u64)))
+svuint64_t svmlslb_lane(svuint64_t, svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_f32)))
+svfloat32_t svmlslt(svfloat32_t, svfloat16_t, float16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_s32)))
+svint32_t svmlslt(svint32_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_s64)))
+svint64_t svmlslt(svint64_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_s16)))
+svint16_t svmlslt(svint16_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_u32)))
+svuint32_t svmlslt(svuint32_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_u64)))
+svuint64_t svmlslt(svuint64_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_u16)))
+svuint16_t svmlslt(svuint16_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_f32)))
+svfloat32_t svmlslt(svfloat32_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_s32)))
+svint32_t svmlslt(svint32_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_s64)))
+svint64_t svmlslt(svint64_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_s16)))
+svint16_t svmlslt(svint16_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_u32)))
+svuint32_t svmlslt(svuint32_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_u64)))
+svuint64_t svmlslt(svuint64_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_u16)))
+svuint16_t svmlslt(svuint16_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_f32)))
+svfloat32_t svmlslt_lane(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_s32)))
+svint32_t svmlslt_lane(svint32_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_s64)))
+svint64_t svmlslt_lane(svint64_t, svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_u32)))
+svuint32_t svmlslt_lane(svuint32_t, svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_u64)))
+svuint64_t svmlslt_lane(svuint64_t, svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_s32)))
+svint32_t svmovlb(svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_s64)))
+svint64_t svmovlb(svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_s16)))
+svint16_t svmovlb(svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_u32)))
+svuint32_t svmovlb(svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_u64)))
+svuint64_t svmovlb(svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_u16)))
+svuint16_t svmovlb(svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_s32)))
+svint32_t svmovlt(svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_s64)))
+svint64_t svmovlt(svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_s16)))
+svint16_t svmovlt(svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_u32)))
+svuint32_t svmovlt(svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_u64)))
+svuint64_t svmovlt(svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_u16)))
+svuint16_t svmovlt(svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_u32)))
+svuint32_t svmul_lane(svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_u64)))
+svuint64_t svmul_lane(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_u16)))
+svuint16_t svmul_lane(svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_s32)))
+svint32_t svmul_lane(svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_s64)))
+svint64_t svmul_lane(svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_s16)))
+svint16_t svmul_lane(svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_s32)))
+svint32_t svmullb(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_s64)))
+svint64_t svmullb(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_s16)))
+svint16_t svmullb(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_u32)))
+svuint32_t svmullb(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_u64)))
+svuint64_t svmullb(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_u16)))
+svuint16_t svmullb(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_s32)))
+svint32_t svmullb(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_s64)))
+svint64_t svmullb(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_s16)))
+svint16_t svmullb(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_u32)))
+svuint32_t svmullb(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_u64)))
+svuint64_t svmullb(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_u16)))
+svuint16_t svmullb(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_s32)))
+svint32_t svmullb_lane(svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_s64)))
+svint64_t svmullb_lane(svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_u32)))
+svuint32_t svmullb_lane(svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_u64)))
+svuint64_t svmullb_lane(svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_s32)))
+svint32_t svmullt(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_s64)))
+svint64_t svmullt(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_s16)))
+svint16_t svmullt(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_u32)))
+svuint32_t svmullt(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_u64)))
+svuint64_t svmullt(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_u16)))
+svuint16_t svmullt(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_s32)))
+svint32_t svmullt(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_s64)))
+svint64_t svmullt(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_s16)))
+svint16_t svmullt(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_u32)))
+svuint32_t svmullt(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_u64)))
+svuint64_t svmullt(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_u16)))
+svuint16_t svmullt(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_s32)))
+svint32_t svmullt_lane(svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_s64)))
+svint64_t svmullt_lane(svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_u32)))
+svuint32_t svmullt_lane(svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_u64)))
+svuint64_t svmullt_lane(svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u8)))
+svuint8_t svnbsl(svuint8_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u32)))
+svuint32_t svnbsl(svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u64)))
+svuint64_t svnbsl(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u16)))
+svuint16_t svnbsl(svuint16_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s8)))
+svint8_t svnbsl(svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s32)))
+svint32_t svnbsl(svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s64)))
+svint64_t svnbsl(svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s16)))
+svint16_t svnbsl(svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u8)))
+svuint8_t svnbsl(svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u32)))
+svuint32_t svnbsl(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u64)))
+svuint64_t svnbsl(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u16)))
+svuint16_t svnbsl(svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s8)))
+svint8_t svnbsl(svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s32)))
+svint32_t svnbsl(svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s64)))
+svint64_t svnbsl(svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s16)))
+svint16_t svnbsl(svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_u8)))
+svbool_t svnmatch(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_u16)))
+svbool_t svnmatch(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_s8)))
+svbool_t svnmatch(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_s16)))
+svbool_t svnmatch(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmul_n_u8)))
+svuint8_t svpmul(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmul_u8)))
+svuint8_t svpmul(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_n_u64)))
+svuint64_t svpmullb(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_n_u16)))
+svuint16_t svpmullb(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_u64)))
+svuint64_t svpmullb(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_u16)))
+svuint16_t svpmullb(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_n_u8)))
+svuint8_t svpmullb_pair(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_n_u32)))
+svuint32_t svpmullb_pair(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_u8)))
+svuint8_t svpmullb_pair(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_u32)))
+svuint32_t svpmullb_pair(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_n_u64)))
+svuint64_t svpmullt(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_n_u16)))
+svuint16_t svpmullt(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_u64)))
+svuint64_t svpmullt(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_u16)))
+svuint16_t svpmullt(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_n_u8)))
+svuint8_t svpmullt_pair(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_n_u32)))
+svuint32_t svpmullt_pair(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_u8)))
+svuint8_t svpmullt_pair(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_u32)))
+svuint32_t svpmullt_pair(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s8_m)))
+svint8_t svqabs_m(svint8_t, svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s32_m)))
+svint32_t svqabs_m(svint32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s64_m)))
+svint64_t svqabs_m(svint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s16_m)))
+svint16_t svqabs_m(svint16_t, svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s8_x)))
+svint8_t svqabs_x(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s32_x)))
+svint32_t svqabs_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s64_x)))
+svint64_t svqabs_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s16_x)))
+svint16_t svqabs_x(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s8_z)))
+svint8_t svqabs_z(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s32_z)))
+svint32_t svqabs_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s64_z)))
+svint64_t svqabs_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s16_z)))
+svint16_t svqabs_z(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8_m)))
+svint8_t svqadd_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32_m)))
+svint32_t svqadd_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64_m)))
+svint64_t svqadd_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16_m)))
+svint16_t svqadd_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8_x)))
+svint8_t svqadd_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32_x)))
+svint32_t svqadd_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64_x)))
+svint64_t svqadd_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16_x)))
+svint16_t svqadd_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8_z)))
+svint8_t svqadd_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32_z)))
+svint32_t svqadd_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64_z)))
+svint64_t svqadd_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16_z)))
+svint16_t svqadd_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8_m)))
+svuint8_t svqadd_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32_m)))
+svuint32_t svqadd_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64_m)))
+svuint64_t svqadd_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16_m)))
+svuint16_t svqadd_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8_x)))
+svuint8_t svqadd_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32_x)))
+svuint32_t svqadd_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64_x)))
+svuint64_t svqadd_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16_x)))
+svuint16_t svqadd_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8_z)))
+svuint8_t svqadd_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32_z)))
+svuint32_t svqadd_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64_z)))
+svuint64_t svqadd_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16_z)))
+svuint16_t svqadd_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8_m)))
+svint8_t svqadd_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32_m)))
+svint32_t svqadd_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64_m)))
+svint64_t svqadd_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16_m)))
+svint16_t svqadd_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8_x)))
+svint8_t svqadd_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32_x)))
+svint32_t svqadd_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64_x)))
+svint64_t svqadd_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16_x)))
+svint16_t svqadd_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8_z)))
+svint8_t svqadd_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32_z)))
+svint32_t svqadd_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64_z)))
+svint64_t svqadd_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16_z)))
+svint16_t svqadd_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8_m)))
+svuint8_t svqadd_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32_m)))
+svuint32_t svqadd_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64_m)))
+svuint64_t svqadd_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16_m)))
+svuint16_t svqadd_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8_x)))
+svuint8_t svqadd_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32_x)))
+svuint32_t svqadd_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64_x)))
+svuint64_t svqadd_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16_x)))
+svuint16_t svqadd_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8_z)))
+svuint8_t svqadd_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32_z)))
+svuint32_t svqadd_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64_z)))
+svuint64_t svqadd_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16_z)))
+svuint16_t svqadd_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s8)))
+svint8_t svqcadd(svint8_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s32)))
+svint32_t svqcadd(svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s64)))
+svint64_t svqcadd(svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s16)))
+svint16_t svqcadd(svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_n_s32)))
+svint32_t svqdmlalb(svint32_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_n_s64)))
+svint64_t svqdmlalb(svint64_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_n_s16)))
+svint16_t svqdmlalb(svint16_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_s32)))
+svint32_t svqdmlalb(svint32_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_s64)))
+svint64_t svqdmlalb(svint64_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_s16)))
+svint16_t svqdmlalb(svint16_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_lane_s32)))
+svint32_t svqdmlalb_lane(svint32_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_lane_s64)))
+svint64_t svqdmlalb_lane(svint64_t, svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_n_s32)))
+svint32_t svqdmlalbt(svint32_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_n_s64)))
+svint64_t svqdmlalbt(svint64_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_n_s16)))
+svint16_t svqdmlalbt(svint16_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_s32)))
+svint32_t svqdmlalbt(svint32_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_s64)))
+svint64_t svqdmlalbt(svint64_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_s16)))
+svint16_t svqdmlalbt(svint16_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_n_s32)))
+svint32_t svqdmlalt(svint32_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_n_s64)))
+svint64_t svqdmlalt(svint64_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_n_s16)))
+svint16_t svqdmlalt(svint16_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_s32)))
+svint32_t svqdmlalt(svint32_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_s64)))
+svint64_t svqdmlalt(svint64_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_s16)))
+svint16_t svqdmlalt(svint16_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_lane_s32)))
+svint32_t svqdmlalt_lane(svint32_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_lane_s64)))
+svint64_t svqdmlalt_lane(svint64_t, svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_n_s32)))
+svint32_t svqdmlslb(svint32_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_n_s64)))
+svint64_t svqdmlslb(svint64_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_n_s16)))
+svint16_t svqdmlslb(svint16_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_s32)))
+svint32_t svqdmlslb(svint32_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_s64)))
+svint64_t svqdmlslb(svint64_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_s16)))
+svint16_t svqdmlslb(svint16_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_lane_s32)))
+svint32_t svqdmlslb_lane(svint32_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_lane_s64)))
+svint64_t svqdmlslb_lane(svint64_t, svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_n_s32)))
+svint32_t svqdmlslbt(svint32_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_n_s64)))
+svint64_t svqdmlslbt(svint64_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_n_s16)))
+svint16_t svqdmlslbt(svint16_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_s32)))
+svint32_t svqdmlslbt(svint32_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_s64)))
+svint64_t svqdmlslbt(svint64_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_s16)))
+svint16_t svqdmlslbt(svint16_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_n_s32)))
+svint32_t svqdmlslt(svint32_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_n_s64)))
+svint64_t svqdmlslt(svint64_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_n_s16)))
+svint16_t svqdmlslt(svint16_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_s32)))
+svint32_t svqdmlslt(svint32_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_s64)))
+svint64_t svqdmlslt(svint64_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_s16)))
+svint16_t svqdmlslt(svint16_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_lane_s32)))
+svint32_t svqdmlslt_lane(svint32_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_lane_s64)))
+svint64_t svqdmlslt_lane(svint64_t, svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s8)))
+svint8_t svqdmulh(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s32)))
+svint32_t svqdmulh(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s64)))
+svint64_t svqdmulh(svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s16)))
+svint16_t svqdmulh(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s8)))
+svint8_t svqdmulh(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s32)))
+svint32_t svqdmulh(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s64)))
+svint64_t svqdmulh(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s16)))
+svint16_t svqdmulh(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_lane_s32)))
+svint32_t svqdmulh_lane(svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_lane_s64)))
+svint64_t svqdmulh_lane(svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_lane_s16)))
+svint16_t svqdmulh_lane(svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_n_s32)))
+svint32_t svqdmullb(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_n_s64)))
+svint64_t svqdmullb(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_n_s16)))
+svint16_t svqdmullb(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_s32)))
+svint32_t svqdmullb(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_s64)))
+svint64_t svqdmullb(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_s16)))
+svint16_t svqdmullb(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_lane_s32)))
+svint32_t svqdmullb_lane(svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_lane_s64)))
+svint64_t svqdmullb_lane(svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_n_s32)))
+svint32_t svqdmullt(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_n_s64)))
+svint64_t svqdmullt(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_n_s16)))
+svint16_t svqdmullt(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_s32)))
+svint32_t svqdmullt(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_s64)))
+svint64_t svqdmullt(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_s16)))
+svint16_t svqdmullt(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_lane_s32)))
+svint32_t svqdmullt_lane(svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_lane_s64)))
+svint64_t svqdmullt_lane(svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s8_m)))
+svint8_t svqneg_m(svint8_t, svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s32_m)))
+svint32_t svqneg_m(svint32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s64_m)))
+svint64_t svqneg_m(svint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s16_m)))
+svint16_t svqneg_m(svint16_t, svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s8_x)))
+svint8_t svqneg_x(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s32_x)))
+svint32_t svqneg_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s64_x)))
+svint64_t svqneg_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s16_x)))
+svint16_t svqneg_x(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s8_z)))
+svint8_t svqneg_z(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s32_z)))
+svint32_t svqneg_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s64_z)))
+svint64_t svqneg_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s16_z)))
+svint16_t svqneg_z(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s8)))
+svint8_t svqrdcmlah(svint8_t, svint8_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s32)))
+svint32_t svqrdcmlah(svint32_t, svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s64)))
+svint64_t svqrdcmlah(svint64_t, svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s16)))
+svint16_t svqrdcmlah(svint16_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_lane_s32)))
+svint32_t svqrdcmlah_lane(svint32_t, svint32_t, svint32_t, uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_lane_s16)))
+svint16_t svqrdcmlah_lane(svint16_t, svint16_t, svint16_t, uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s8)))
+svint8_t svqrdmlah(svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s32)))
+svint32_t svqrdmlah(svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s64)))
+svint64_t svqrdmlah(svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s16)))
+svint16_t svqrdmlah(svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s8)))
+svint8_t svqrdmlah(svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s32)))
+svint32_t svqrdmlah(svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s64)))
+svint64_t svqrdmlah(svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s16)))
+svint16_t svqrdmlah(svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_lane_s32)))
+svint32_t svqrdmlah_lane(svint32_t, svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_lane_s64)))
+svint64_t svqrdmlah_lane(svint64_t, svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_lane_s16)))
+svint16_t svqrdmlah_lane(svint16_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s8)))
+svint8_t svqrdmlsh(svint8_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s32)))
+svint32_t svqrdmlsh(svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s64)))
+svint64_t svqrdmlsh(svint64_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s16)))
+svint16_t svqrdmlsh(svint16_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s8)))
+svint8_t svqrdmlsh(svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s32)))
+svint32_t svqrdmlsh(svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s64)))
+svint64_t svqrdmlsh(svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s16)))
+svint16_t svqrdmlsh(svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_lane_s32)))
+svint32_t svqrdmlsh_lane(svint32_t, svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_lane_s64)))
+svint64_t svqrdmlsh_lane(svint64_t, svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_lane_s16)))
+svint16_t svqrdmlsh_lane(svint16_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s8)))
+svint8_t svqrdmulh(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s32)))
+svint32_t svqrdmulh(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s64)))
+svint64_t svqrdmulh(svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s16)))
+svint16_t svqrdmulh(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s8)))
+svint8_t svqrdmulh(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s32)))
+svint32_t svqrdmulh(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s64)))
+svint64_t svqrdmulh(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s16)))
+svint16_t svqrdmulh(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_lane_s32)))
+svint32_t svqrdmulh_lane(svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_lane_s64)))
+svint64_t svqrdmulh_lane(svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_lane_s16)))
+svint16_t svqrdmulh_lane(svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s8_m)))
+svint8_t svqrshl_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s32_m)))
+svint32_t svqrshl_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s64_m)))
+svint64_t svqrshl_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s16_m)))
+svint16_t svqrshl_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s8_x)))
+svint8_t svqrshl_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s32_x)))
+svint32_t svqrshl_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s64_x)))
+svint64_t svqrshl_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s16_x)))
+svint16_t svqrshl_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s8_z)))
+svint8_t svqrshl_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s32_z)))
+svint32_t svqrshl_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s64_z)))
+svint64_t svqrshl_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s16_z)))
+svint16_t svqrshl_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u8_m)))
+svuint8_t svqrshl_m(svbool_t, svuint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u32_m)))
+svuint32_t svqrshl_m(svbool_t, svuint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u64_m)))
+svuint64_t svqrshl_m(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u16_m)))
+svuint16_t svqrshl_m(svbool_t, svuint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u8_x)))
+svuint8_t svqrshl_x(svbool_t, svuint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u32_x)))
+svuint32_t svqrshl_x(svbool_t, svuint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u64_x)))
+svuint64_t svqrshl_x(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u16_x)))
+svuint16_t svqrshl_x(svbool_t, svuint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u8_z)))
+svuint8_t svqrshl_z(svbool_t, svuint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u32_z)))
+svuint32_t svqrshl_z(svbool_t, svuint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u64_z)))
+svuint64_t svqrshl_z(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u16_z)))
+svuint16_t svqrshl_z(svbool_t, svuint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s8_m)))
+svint8_t svqrshl_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s32_m)))
+svint32_t svqrshl_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s64_m)))
+svint64_t svqrshl_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s16_m)))
+svint16_t svqrshl_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s8_x)))
+svint8_t svqrshl_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s32_x)))
+svint32_t svqrshl_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s64_x)))
+svint64_t svqrshl_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s16_x)))
+svint16_t svqrshl_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s8_z)))
+svint8_t svqrshl_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s32_z)))
+svint32_t svqrshl_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s64_z)))
+svint64_t svqrshl_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s16_z)))
+svint16_t svqrshl_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u8_m)))
+svuint8_t svqrshl_m(svbool_t, svuint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u32_m)))
+svuint32_t svqrshl_m(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u64_m)))
+svuint64_t svqrshl_m(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u16_m)))
+svuint16_t svqrshl_m(svbool_t, svuint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u8_x)))
+svuint8_t svqrshl_x(svbool_t, svuint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u32_x)))
+svuint32_t svqrshl_x(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u64_x)))
+svuint64_t svqrshl_x(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u16_x)))
+svuint16_t svqrshl_x(svbool_t, svuint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u8_z)))
+svuint8_t svqrshl_z(svbool_t, svuint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u32_z)))
+svuint32_t svqrshl_z(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u64_z)))
+svuint64_t svqrshl_z(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u16_z)))
+svuint16_t svqrshl_z(svbool_t, svuint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_s32)))
+svint16_t svqrshrnb(svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_s64)))
+svint32_t svqrshrnb(svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_s16)))
+svint8_t svqrshrnb(svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_u32)))
+svuint16_t svqrshrnb(svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_u64)))
+svuint32_t svqrshrnb(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_u16)))
+svuint8_t svqrshrnb(svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_s32)))
+svint16_t svqrshrnt(svint16_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_s64)))
+svint32_t svqrshrnt(svint32_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_s16)))
+svint8_t svqrshrnt(svint8_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_u32)))
+svuint16_t svqrshrnt(svuint16_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_u64)))
+svuint32_t svqrshrnt(svuint32_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_u16)))
+svuint8_t svqrshrnt(svuint8_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunb_n_s32)))
+svuint16_t svqrshrunb(svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunb_n_s64)))
+svuint32_t svqrshrunb(svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunb_n_s16)))
+svuint8_t svqrshrunb(svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunt_n_s32)))
+svuint16_t svqrshrunt(svuint16_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunt_n_s64)))
+svuint32_t svqrshrunt(svuint32_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunt_n_s16)))
+svuint8_t svqrshrunt(svuint8_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s8_m)))
+svint8_t svqshl_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s32_m)))
+svint32_t svqshl_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s64_m)))
+svint64_t svqshl_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s16_m)))
+svint16_t svqshl_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s8_x)))
+svint8_t svqshl_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s32_x)))
+svint32_t svqshl_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s64_x)))
+svint64_t svqshl_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s16_x)))
+svint16_t svqshl_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s8_z)))
+svint8_t svqshl_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s32_z)))
+svint32_t svqshl_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s64_z)))
+svint64_t svqshl_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s16_z)))
+svint16_t svqshl_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u8_m)))
+svuint8_t svqshl_m(svbool_t, svuint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u32_m)))
+svuint32_t svqshl_m(svbool_t, svuint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u64_m)))
+svuint64_t svqshl_m(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u16_m)))
+svuint16_t svqshl_m(svbool_t, svuint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u8_x)))
+svuint8_t svqshl_x(svbool_t, svuint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u32_x)))
+svuint32_t svqshl_x(svbool_t, svuint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u64_x)))
+svuint64_t svqshl_x(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u16_x)))
+svuint16_t svqshl_x(svbool_t, svuint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u8_z)))
+svuint8_t svqshl_z(svbool_t, svuint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u32_z)))
+svuint32_t svqshl_z(svbool_t, svuint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u64_z)))
+svuint64_t svqshl_z(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u16_z)))
+svuint16_t svqshl_z(svbool_t, svuint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s8_m)))
+svint8_t svqshl_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s32_m)))
+svint32_t svqshl_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s64_m)))
+svint64_t svqshl_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s16_m)))
+svint16_t svqshl_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s8_x)))
+svint8_t svqshl_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s32_x)))
+svint32_t svqshl_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s64_x)))
+svint64_t svqshl_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s16_x)))
+svint16_t svqshl_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s8_z)))
+svint8_t svqshl_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s32_z)))
+svint32_t svqshl_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s64_z)))
+svint64_t svqshl_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s16_z)))
+svint16_t svqshl_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u8_m)))
+svuint8_t svqshl_m(svbool_t, svuint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u32_m)))
+svuint32_t svqshl_m(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u64_m)))
+svuint64_t svqshl_m(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u16_m)))
+svuint16_t svqshl_m(svbool_t, svuint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u8_x)))
+svuint8_t svqshl_x(svbool_t, svuint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u32_x)))
+svuint32_t svqshl_x(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u64_x)))
+svuint64_t svqshl_x(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u16_x)))
+svuint16_t svqshl_x(svbool_t, svuint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u8_z)))
+svuint8_t svqshl_z(svbool_t, svuint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u32_z)))
+svuint32_t svqshl_z(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u64_z)))
+svuint64_t svqshl_z(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u16_z)))
+svuint16_t svqshl_z(svbool_t, svuint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s8_m)))
+svuint8_t svqshlu_m(svbool_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s32_m)))
+svuint32_t svqshlu_m(svbool_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s64_m)))
+svuint64_t svqshlu_m(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s16_m)))
+svuint16_t svqshlu_m(svbool_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s8_x)))
+svuint8_t svqshlu_x(svbool_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s32_x)))
+svuint32_t svqshlu_x(svbool_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s64_x)))
+svuint64_t svqshlu_x(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s16_x)))
+svuint16_t svqshlu_x(svbool_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s8_z)))
+svuint8_t svqshlu_z(svbool_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s32_z)))
+svuint32_t svqshlu_z(svbool_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s64_z)))
+svuint64_t svqshlu_z(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s16_z)))
+svuint16_t svqshlu_z(svbool_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_s32)))
+svint16_t svqshrnb(svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_s64)))
+svint32_t svqshrnb(svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_s16)))
+svint8_t svqshrnb(svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_u32)))
+svuint16_t svqshrnb(svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_u64)))
+svuint32_t svqshrnb(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_u16)))
+svuint8_t svqshrnb(svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_s32)))
+svint16_t svqshrnt(svint16_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_s64)))
+svint32_t svqshrnt(svint32_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_s16)))
+svint8_t svqshrnt(svint8_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_u32)))
+svuint16_t svqshrnt(svuint16_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_u64)))
+svuint32_t svqshrnt(svuint32_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_u16)))
+svuint8_t svqshrnt(svuint8_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunb_n_s32)))
+svuint16_t svqshrunb(svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunb_n_s64)))
+svuint32_t svqshrunb(svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunb_n_s16)))
+svuint8_t svqshrunb(svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunt_n_s32)))
+svuint16_t svqshrunt(svuint16_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunt_n_s64)))
+svuint32_t svqshrunt(svuint32_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunt_n_s16)))
+svuint8_t svqshrunt(svuint8_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8_m)))
+svint8_t svqsub_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32_m)))
+svint32_t svqsub_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64_m)))
+svint64_t svqsub_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16_m)))
+svint16_t svqsub_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8_x)))
+svint8_t svqsub_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32_x)))
+svint32_t svqsub_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64_x)))
+svint64_t svqsub_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16_x)))
+svint16_t svqsub_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8_z)))
+svint8_t svqsub_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32_z)))
+svint32_t svqsub_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64_z)))
+svint64_t svqsub_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16_z)))
+svint16_t svqsub_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8_m)))
+svuint8_t svqsub_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32_m)))
+svuint32_t svqsub_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64_m)))
+svuint64_t svqsub_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16_m)))
+svuint16_t svqsub_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8_x)))
+svuint8_t svqsub_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32_x)))
+svuint32_t svqsub_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64_x)))
+svuint64_t svqsub_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16_x)))
+svuint16_t svqsub_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8_z)))
+svuint8_t svqsub_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32_z)))
+svuint32_t svqsub_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64_z)))
+svuint64_t svqsub_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16_z)))
+svuint16_t svqsub_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8_m)))
+svint8_t svqsub_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32_m)))
+svint32_t svqsub_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64_m)))
+svint64_t svqsub_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16_m)))
+svint16_t svqsub_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8_x)))
+svint8_t svqsub_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32_x)))
+svint32_t svqsub_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64_x)))
+svint64_t svqsub_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16_x)))
+svint16_t svqsub_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8_z)))
+svint8_t svqsub_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32_z)))
+svint32_t svqsub_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64_z)))
+svint64_t svqsub_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16_z)))
+svint16_t svqsub_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8_m)))
+svuint8_t svqsub_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32_m)))
+svuint32_t svqsub_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64_m)))
+svuint64_t svqsub_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16_m)))
+svuint16_t svqsub_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8_x)))
+svuint8_t svqsub_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32_x)))
+svuint32_t svqsub_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64_x)))
+svuint64_t svqsub_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16_x)))
+svuint16_t svqsub_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8_z)))
+svuint8_t svqsub_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32_z)))
+svuint32_t svqsub_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64_z)))
+svuint64_t svqsub_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16_z)))
+svuint16_t svqsub_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s8_m)))
+svint8_t svqsubr_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s32_m)))
+svint32_t svqsubr_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s64_m)))
+svint64_t svqsubr_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s16_m)))
+svint16_t svqsubr_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s8_x)))
+svint8_t svqsubr_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s32_x)))
+svint32_t svqsubr_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s64_x)))
+svint64_t svqsubr_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s16_x)))
+svint16_t svqsubr_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s8_z)))
+svint8_t svqsubr_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s32_z)))
+svint32_t svqsubr_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s64_z)))
+svint64_t svqsubr_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s16_z)))
+svint16_t svqsubr_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u8_m)))
+svuint8_t svqsubr_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u32_m)))
+svuint32_t svqsubr_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u64_m)))
+svuint64_t svqsubr_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u16_m)))
+svuint16_t svqsubr_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u8_x)))
+svuint8_t svqsubr_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u32_x)))
+svuint32_t svqsubr_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u64_x)))
+svuint64_t svqsubr_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u16_x)))
+svuint16_t svqsubr_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u8_z)))
+svuint8_t svqsubr_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u32_z)))
+svuint32_t svqsubr_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u64_z)))
+svuint64_t svqsubr_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u16_z)))
+svuint16_t svqsubr_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s8_m)))
+svint8_t svqsubr_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s32_m)))
+svint32_t svqsubr_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s64_m)))
+svint64_t svqsubr_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s16_m)))
+svint16_t svqsubr_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s8_x)))
+svint8_t svqsubr_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s32_x)))
+svint32_t svqsubr_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s64_x)))
+svint64_t svqsubr_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s16_x)))
+svint16_t svqsubr_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s8_z)))
+svint8_t svqsubr_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s32_z)))
+svint32_t svqsubr_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s64_z)))
+svint64_t svqsubr_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s16_z)))
+svint16_t svqsubr_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u8_m)))
+svuint8_t svqsubr_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u32_m)))
+svuint32_t svqsubr_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u64_m)))
+svuint64_t svqsubr_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u16_m)))
+svuint16_t svqsubr_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u8_x)))
+svuint8_t svqsubr_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u32_x)))
+svuint32_t svqsubr_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u64_x)))
+svuint64_t svqsubr_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u16_x)))
+svuint16_t svqsubr_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u8_z)))
+svuint8_t svqsubr_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u32_z)))
+svuint32_t svqsubr_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u64_z)))
+svuint64_t svqsubr_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u16_z)))
+svuint16_t svqsubr_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_s32)))
+svint16_t svqxtnb(svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_s64)))
+svint32_t svqxtnb(svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_s16)))
+svint8_t svqxtnb(svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_u32)))
+svuint16_t svqxtnb(svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_u64)))
+svuint32_t svqxtnb(svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_u16)))
+svuint8_t svqxtnb(svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_s32)))
+svint16_t svqxtnt(svint16_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_s64)))
+svint32_t svqxtnt(svint32_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_s16)))
+svint8_t svqxtnt(svint8_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_u32)))
+svuint16_t svqxtnt(svuint16_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_u64)))
+svuint32_t svqxtnt(svuint32_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_u16)))
+svuint8_t svqxtnt(svuint8_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunb_s32)))
+svuint16_t svqxtunb(svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunb_s64)))
+svuint32_t svqxtunb(svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunb_s16)))
+svuint8_t svqxtunb(svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunt_s32)))
+svuint16_t svqxtunt(svuint16_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunt_s64)))
+svuint32_t svqxtunt(svuint32_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunt_s16)))
+svuint8_t svqxtunt(svuint8_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_u32)))
+svuint16_t svraddhnb(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_u64)))
+svuint32_t svraddhnb(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_u16)))
+svuint8_t svraddhnb(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_s32)))
+svint16_t svraddhnb(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_s64)))
+svint32_t svraddhnb(svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_s16)))
+svint8_t svraddhnb(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_u32)))
+svuint16_t svraddhnb(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_u64)))
+svuint32_t svraddhnb(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_u16)))
+svuint8_t svraddhnb(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_s32)))
+svint16_t svraddhnb(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_s64)))
+svint32_t svraddhnb(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_s16)))
+svint8_t svraddhnb(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_u32)))
+svuint16_t svraddhnt(svuint16_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_u64)))
+svuint32_t svraddhnt(svuint32_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_u16)))
+svuint8_t svraddhnt(svuint8_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_s32)))
+svint16_t svraddhnt(svint16_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_s64)))
+svint32_t svraddhnt(svint32_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_s16)))
+svint8_t svraddhnt(svint8_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_u32)))
+svuint16_t svraddhnt(svuint16_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_u64)))
+svuint32_t svraddhnt(svuint32_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_u16)))
+svuint8_t svraddhnt(svuint8_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_s32)))
+svint16_t svraddhnt(svint16_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_s64)))
+svint32_t svraddhnt(svint32_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_s16)))
+svint8_t svraddhnt(svint8_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_u32_m)))
+svuint32_t svrecpe_m(svuint32_t, svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_u32_x)))
+svuint32_t svrecpe_x(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_u32_z)))
+svuint32_t svrecpe_z(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s8_m)))
+svint8_t svrhadd_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s32_m)))
+svint32_t svrhadd_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s64_m)))
+svint64_t svrhadd_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s16_m)))
+svint16_t svrhadd_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s8_x)))
+svint8_t svrhadd_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s32_x)))
+svint32_t svrhadd_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s64_x)))
+svint64_t svrhadd_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s16_x)))
+svint16_t svrhadd_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s8_z)))
+svint8_t svrhadd_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s32_z)))
+svint32_t svrhadd_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s64_z)))
+svint64_t svrhadd_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s16_z)))
+svint16_t svrhadd_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u8_m)))
+svuint8_t svrhadd_m(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u32_m)))
+svuint32_t svrhadd_m(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u64_m)))
+svuint64_t svrhadd_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u16_m)))
+svuint16_t svrhadd_m(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u8_x)))
+svuint8_t svrhadd_x(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u32_x)))
+svuint32_t svrhadd_x(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u64_x)))
+svuint64_t svrhadd_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u16_x)))
+svuint16_t svrhadd_x(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u8_z)))
+svuint8_t svrhadd_z(svbool_t, svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u32_z)))
+svuint32_t svrhadd_z(svbool_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u64_z)))
+svuint64_t svrhadd_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u16_z)))
+svuint16_t svrhadd_z(svbool_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s8_m)))
+svint8_t svrhadd_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s32_m)))
+svint32_t svrhadd_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s64_m)))
+svint64_t svrhadd_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s16_m)))
+svint16_t svrhadd_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s8_x)))
+svint8_t svrhadd_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s32_x)))
+svint32_t svrhadd_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s64_x)))
+svint64_t svrhadd_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s16_x)))
+svint16_t svrhadd_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s8_z)))
+svint8_t svrhadd_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s32_z)))
+svint32_t svrhadd_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s64_z)))
+svint64_t svrhadd_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s16_z)))
+svint16_t svrhadd_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u8_m)))
+svuint8_t svrhadd_m(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u32_m)))
+svuint32_t svrhadd_m(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u64_m)))
+svuint64_t svrhadd_m(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u16_m)))
+svuint16_t svrhadd_m(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u8_x)))
+svuint8_t svrhadd_x(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u32_x)))
+svuint32_t svrhadd_x(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u64_x)))
+svuint64_t svrhadd_x(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u16_x)))
+svuint16_t svrhadd_x(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u8_z)))
+svuint8_t svrhadd_z(svbool_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u32_z)))
+svuint32_t svrhadd_z(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u64_z)))
+svuint64_t svrhadd_z(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u16_z)))
+svuint16_t svrhadd_z(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s8_m)))
+svint8_t svrshl_m(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s32_m)))
+svint32_t svrshl_m(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s64_m)))
+svint64_t svrshl_m(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s16_m)))
+svint16_t svrshl_m(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s8_x)))
+svint8_t svrshl_x(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s32_x)))
+svint32_t svrshl_x(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s64_x)))
+svint64_t svrshl_x(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s16_x)))
+svint16_t svrshl_x(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s8_z)))
+svint8_t svrshl_z(svbool_t, svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s32_z)))
+svint32_t svrshl_z(svbool_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s64_z)))
+svint64_t svrshl_z(svbool_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s16_z)))
+svint16_t svrshl_z(svbool_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u8_m)))
+svuint8_t svrshl_m(svbool_t, svuint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u32_m)))
+svuint32_t svrshl_m(svbool_t, svuint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u64_m)))
+svuint64_t svrshl_m(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u16_m)))
+svuint16_t svrshl_m(svbool_t, svuint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u8_x)))
+svuint8_t svrshl_x(svbool_t, svuint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u32_x)))
+svuint32_t svrshl_x(svbool_t, svuint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u64_x)))
+svuint64_t svrshl_x(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u16_x)))
+svuint16_t svrshl_x(svbool_t, svuint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u8_z)))
+svuint8_t svrshl_z(svbool_t, svuint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u32_z)))
+svuint32_t svrshl_z(svbool_t, svuint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u64_z)))
+svuint64_t svrshl_z(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u16_z)))
+svuint16_t svrshl_z(svbool_t, svuint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s8_m)))
+svint8_t svrshl_m(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s32_m)))
+svint32_t svrshl_m(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s64_m)))
+svint64_t svrshl_m(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s16_m)))
+svint16_t svrshl_m(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s8_x)))
+svint8_t svrshl_x(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s32_x)))
+svint32_t svrshl_x(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s64_x)))
+svint64_t svrshl_x(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s16_x)))
+svint16_t svrshl_x(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s8_z)))
+svint8_t svrshl_z(svbool_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s32_z)))
+svint32_t svrshl_z(svbool_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s64_z)))
+svint64_t svrshl_z(svbool_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s16_z)))
+svint16_t svrshl_z(svbool_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u8_m)))
+svuint8_t svrshl_m(svbool_t, svuint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u32_m)))
+svuint32_t svrshl_m(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u64_m)))
+svuint64_t svrshl_m(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u16_m)))
+svuint16_t svrshl_m(svbool_t, svuint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u8_x)))
+svuint8_t svrshl_x(svbool_t, svuint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u32_x)))
+svuint32_t svrshl_x(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u64_x)))
+svuint64_t svrshl_x(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u16_x)))
+svuint16_t svrshl_x(svbool_t, svuint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u8_z)))
+svuint8_t svrshl_z(svbool_t, svuint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u32_z)))
+svuint32_t svrshl_z(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u64_z)))
+svuint64_t svrshl_z(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u16_z)))
+svuint16_t svrshl_z(svbool_t, svuint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s8_m)))
+svint8_t svrshr_m(svbool_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s32_m)))
+svint32_t svrshr_m(svbool_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s64_m)))
+svint64_t svrshr_m(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s16_m)))
+svint16_t svrshr_m(svbool_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u8_m)))
+svuint8_t svrshr_m(svbool_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u32_m)))
+svuint32_t svrshr_m(svbool_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u64_m)))
+svuint64_t svrshr_m(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u16_m)))
+svuint16_t svrshr_m(svbool_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s8_x)))
+svint8_t svrshr_x(svbool_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s32_x)))
+svint32_t svrshr_x(svbool_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s64_x)))
+svint64_t svrshr_x(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s16_x)))
+svint16_t svrshr_x(svbool_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u8_x)))
+svuint8_t svrshr_x(svbool_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u32_x)))
+svuint32_t svrshr_x(svbool_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u64_x)))
+svuint64_t svrshr_x(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u16_x)))
+svuint16_t svrshr_x(svbool_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s8_z)))
+svint8_t svrshr_z(svbool_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s32_z)))
+svint32_t svrshr_z(svbool_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s64_z)))
+svint64_t svrshr_z(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s16_z)))
+svint16_t svrshr_z(svbool_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u8_z)))
+svuint8_t svrshr_z(svbool_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u32_z)))
+svuint32_t svrshr_z(svbool_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u64_z)))
+svuint64_t svrshr_z(svbool_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u16_z)))
+svuint16_t svrshr_z(svbool_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_u32)))
+svuint16_t svrshrnb(svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_u64)))
+svuint32_t svrshrnb(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_u16)))
+svuint8_t svrshrnb(svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_s32)))
+svint16_t svrshrnb(svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_s64)))
+svint32_t svrshrnb(svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_s16)))
+svint8_t svrshrnb(svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_u32)))
+svuint16_t svrshrnt(svuint16_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_u64)))
+svuint32_t svrshrnt(svuint32_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_u16)))
+svuint8_t svrshrnt(svuint8_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_s32)))
+svint16_t svrshrnt(svint16_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_s64)))
+svint32_t svrshrnt(svint32_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_s16)))
+svint8_t svrshrnt(svint8_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_u32_m)))
+svuint32_t svrsqrte_m(svuint32_t, svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_u32_x)))
+svuint32_t svrsqrte_x(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_u32_z)))
+svuint32_t svrsqrte_z(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s8)))
+svint8_t svrsra(svint8_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s32)))
+svint32_t svrsra(svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s64)))
+svint64_t svrsra(svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s16)))
+svint16_t svrsra(svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u8)))
+svuint8_t svrsra(svuint8_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u32)))
+svuint32_t svrsra(svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u64)))
+svuint64_t svrsra(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u16)))
+svuint16_t svrsra(svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_u32)))
+svuint16_t svrsubhnb(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_u64)))
+svuint32_t svrsubhnb(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_u16)))
+svuint8_t svrsubhnb(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_s32)))
+svint16_t svrsubhnb(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_s64)))
+svint32_t svrsubhnb(svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_s16)))
+svint8_t svrsubhnb(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_u32)))
+svuint16_t svrsubhnb(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_u64)))
+svuint32_t svrsubhnb(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_u16)))
+svuint8_t svrsubhnb(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_s32)))
+svint16_t svrsubhnb(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_s64)))
+svint32_t svrsubhnb(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_s16)))
+svint8_t svrsubhnb(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_u32)))
+svuint16_t svrsubhnt(svuint16_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_u64)))
+svuint32_t svrsubhnt(svuint32_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_u16)))
+svuint8_t svrsubhnt(svuint8_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_s32)))
+svint16_t svrsubhnt(svint16_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_s64)))
+svint32_t svrsubhnt(svint32_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_s16)))
+svint8_t svrsubhnt(svint8_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_u32)))
+svuint16_t svrsubhnt(svuint16_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_u64)))
+svuint32_t svrsubhnt(svuint32_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_u16)))
+svuint8_t svrsubhnt(svuint8_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_s32)))
+svint16_t svrsubhnt(svint16_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_s64)))
+svint32_t svrsubhnt(svint32_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_s16)))
+svint8_t svrsubhnt(svint8_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_n_u32)))
+svuint32_t svsbclb(svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_n_u64)))
+svuint64_t svsbclb(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_u32)))
+svuint32_t svsbclb(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_u64)))
+svuint64_t svsbclb(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_n_u32)))
+svuint32_t svsbclt(svuint32_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_n_u64)))
+svuint64_t svsbclt(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_u32)))
+svuint32_t svsbclt(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_u64)))
+svuint64_t svsbclt(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_s32)))
+svint32_t svshllb(svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_s64)))
+svint64_t svshllb(svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_s16)))
+svint16_t svshllb(svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_u32)))
+svuint32_t svshllb(svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_u64)))
+svuint64_t svshllb(svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_u16)))
+svuint16_t svshllb(svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_s32)))
+svint32_t svshllt(svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_s64)))
+svint64_t svshllt(svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_s16)))
+svint16_t svshllt(svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_u32)))
+svuint32_t svshllt(svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_u64)))
+svuint64_t svshllt(svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_u16)))
+svuint16_t svshllt(svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_u32)))
+svuint16_t svshrnb(svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_u64)))
+svuint32_t svshrnb(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_u16)))
+svuint8_t svshrnb(svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_s32)))
+svint16_t svshrnb(svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_s64)))
+svint32_t svshrnb(svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_s16)))
+svint8_t svshrnb(svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_u32)))
+svuint16_t svshrnt(svuint16_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_u64)))
+svuint32_t svshrnt(svuint32_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_u16)))
+svuint8_t svshrnt(svuint8_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_s32)))
+svint16_t svshrnt(svint16_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_s64)))
+svint32_t svshrnt(svint32_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_s16)))
+svint8_t svshrnt(svint8_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u8)))
+svuint8_t svsli(svuint8_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u32)))
+svuint32_t svsli(svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u64)))
+svuint64_t svsli(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u16)))
+svuint16_t svsli(svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s8)))
+svint8_t svsli(svint8_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s32)))
+svint32_t svsli(svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s64)))
+svint64_t svsli(svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s16)))
+svint16_t svsli(svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u8_m)))
+svuint8_t svsqadd_m(svbool_t, svuint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u32_m)))
+svuint32_t svsqadd_m(svbool_t, svuint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u64_m)))
+svuint64_t svsqadd_m(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u16_m)))
+svuint16_t svsqadd_m(svbool_t, svuint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u8_x)))
+svuint8_t svsqadd_x(svbool_t, svuint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u32_x)))
+svuint32_t svsqadd_x(svbool_t, svuint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u64_x)))
+svuint64_t svsqadd_x(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u16_x)))
+svuint16_t svsqadd_x(svbool_t, svuint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u8_z)))
+svuint8_t svsqadd_z(svbool_t, svuint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u32_z)))
+svuint32_t svsqadd_z(svbool_t, svuint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u64_z)))
+svuint64_t svsqadd_z(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u16_z)))
+svuint16_t svsqadd_z(svbool_t, svuint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u8_m)))
+svuint8_t svsqadd_m(svbool_t, svuint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u32_m)))
+svuint32_t svsqadd_m(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u64_m)))
+svuint64_t svsqadd_m(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u16_m)))
+svuint16_t svsqadd_m(svbool_t, svuint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u8_x)))
+svuint8_t svsqadd_x(svbool_t, svuint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u32_x)))
+svuint32_t svsqadd_x(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u64_x)))
+svuint64_t svsqadd_x(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u16_x)))
+svuint16_t svsqadd_x(svbool_t, svuint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u8_z)))
+svuint8_t svsqadd_z(svbool_t, svuint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u32_z)))
+svuint32_t svsqadd_z(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u64_z)))
+svuint64_t svsqadd_z(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u16_z)))
+svuint16_t svsqadd_z(svbool_t, svuint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s8)))
+svint8_t svsra(svint8_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s32)))
+svint32_t svsra(svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s64)))
+svint64_t svsra(svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s16)))
+svint16_t svsra(svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u8)))
+svuint8_t svsra(svuint8_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u32)))
+svuint32_t svsra(svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u64)))
+svuint64_t svsra(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u16)))
+svuint16_t svsra(svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u8)))
+svuint8_t svsri(svuint8_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u32)))
+svuint32_t svsri(svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u64)))
+svuint64_t svsri(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u16)))
+svuint16_t svsri(svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s8)))
+svint8_t svsri(svint8_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s32)))
+svint32_t svsri(svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s64)))
+svint64_t svsri(svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s16)))
+svint16_t svsri(svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_index_u32)))
+void svstnt1_scatter_index(svbool_t, svuint32_t, int64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_index_u64)))
+void svstnt1_scatter_index(svbool_t, svuint64_t, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_index_f64)))
+void svstnt1_scatter_index(svbool_t, svuint64_t, int64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_index_f32)))
+void svstnt1_scatter_index(svbool_t, svuint32_t, int64_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_index_s32)))
+void svstnt1_scatter_index(svbool_t, svuint32_t, int64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_index_s64)))
+void svstnt1_scatter_index(svbool_t, svuint64_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_offset_u32)))
+void svstnt1_scatter_offset(svbool_t, svuint32_t, int64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_offset_u64)))
+void svstnt1_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_offset_f64)))
+void svstnt1_scatter_offset(svbool_t, svuint64_t, int64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_offset_f32)))
+void svstnt1_scatter_offset(svbool_t, svuint32_t, int64_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_offset_s32)))
+void svstnt1_scatter_offset(svbool_t, svuint32_t, int64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_offset_s64)))
+void svstnt1_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_u32)))
+void svstnt1_scatter(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_u64)))
+void svstnt1_scatter(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_f64)))
+void svstnt1_scatter(svbool_t, svuint64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_f32)))
+void svstnt1_scatter(svbool_t, svuint32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_s32)))
+void svstnt1_scatter(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_s64)))
+void svstnt1_scatter(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64index_u64)))
+void svstnt1_scatter_index(svbool_t, uint64_t *, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64index_f64)))
+void svstnt1_scatter_index(svbool_t, float64_t *, svint64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64index_s64)))
+void svstnt1_scatter_index(svbool_t, int64_t *, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64index_u64)))
+void svstnt1_scatter_index(svbool_t, uint64_t *, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64index_f64)))
+void svstnt1_scatter_index(svbool_t, float64_t *, svuint64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64index_s64)))
+void svstnt1_scatter_index(svbool_t, int64_t *, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32offset_u32)))
+void svstnt1_scatter_offset(svbool_t, uint32_t *, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32offset_f32)))
+void svstnt1_scatter_offset(svbool_t, float32_t *, svuint32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32offset_s32)))
+void svstnt1_scatter_offset(svbool_t, int32_t *, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64offset_u64)))
+void svstnt1_scatter_offset(svbool_t, uint64_t *, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64offset_f64)))
+void svstnt1_scatter_offset(svbool_t, float64_t *, svint64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64offset_s64)))
+void svstnt1_scatter_offset(svbool_t, int64_t *, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64offset_u64)))
+void svstnt1_scatter_offset(svbool_t, uint64_t *, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64offset_f64)))
+void svstnt1_scatter_offset(svbool_t, float64_t *, svuint64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64offset_s64)))
+void svstnt1_scatter_offset(svbool_t, int64_t *, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_offset_u32)))
+void svstnt1b_scatter_offset(svbool_t, svuint32_t, int64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_offset_u64)))
+void svstnt1b_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_offset_s32)))
+void svstnt1b_scatter_offset(svbool_t, svuint32_t, int64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_offset_s64)))
+void svstnt1b_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_u32)))
+void svstnt1b_scatter(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_u64)))
+void svstnt1b_scatter(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_s32)))
+void svstnt1b_scatter(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_s64)))
+void svstnt1b_scatter(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32offset_s32)))
+void svstnt1b_scatter_offset(svbool_t, int8_t *, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32offset_u32)))
+void svstnt1b_scatter_offset(svbool_t, uint8_t *, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_s64offset_s64)))
+void svstnt1b_scatter_offset(svbool_t, int8_t *, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_s64offset_u64)))
+void svstnt1b_scatter_offset(svbool_t, uint8_t *, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64offset_s64)))
+void svstnt1b_scatter_offset(svbool_t, int8_t *, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64offset_u64)))
+void svstnt1b_scatter_offset(svbool_t, uint8_t *, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_index_u32)))
+void svstnt1h_scatter_index(svbool_t, svuint32_t, int64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_index_u64)))
+void svstnt1h_scatter_index(svbool_t, svuint64_t, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_index_s32)))
+void svstnt1h_scatter_index(svbool_t, svuint32_t, int64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_index_s64)))
+void svstnt1h_scatter_index(svbool_t, svuint64_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_offset_u32)))
+void svstnt1h_scatter_offset(svbool_t, svuint32_t, int64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_offset_u64)))
+void svstnt1h_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_offset_s32)))
+void svstnt1h_scatter_offset(svbool_t, svuint32_t, int64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_offset_s64)))
+void svstnt1h_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_u32)))
+void svstnt1h_scatter(svbool_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_u64)))
+void svstnt1h_scatter(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_s32)))
+void svstnt1h_scatter(svbool_t, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_s64)))
+void svstnt1h_scatter(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64index_s64)))
+void svstnt1h_scatter_index(svbool_t, int16_t *, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64index_u64)))
+void svstnt1h_scatter_index(svbool_t, uint16_t *, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64index_s64)))
+void svstnt1h_scatter_index(svbool_t, int16_t *, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64index_u64)))
+void svstnt1h_scatter_index(svbool_t, uint16_t *, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32offset_s32)))
+void svstnt1h_scatter_offset(svbool_t, int16_t *, svuint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32offset_u32)))
+void svstnt1h_scatter_offset(svbool_t, uint16_t *, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64offset_s64)))
+void svstnt1h_scatter_offset(svbool_t, int16_t *, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64offset_u64)))
+void svstnt1h_scatter_offset(svbool_t, uint16_t *, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64offset_s64)))
+void svstnt1h_scatter_offset(svbool_t, int16_t *, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64offset_u64)))
+void svstnt1h_scatter_offset(svbool_t, uint16_t *, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_index_u64)))
+void svstnt1w_scatter_index(svbool_t, svuint64_t, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_index_s64)))
+void svstnt1w_scatter_index(svbool_t, svuint64_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_offset_u64)))
+void svstnt1w_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_offset_s64)))
+void svstnt1w_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_u64)))
+void svstnt1w_scatter(svbool_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_s64)))
+void svstnt1w_scatter(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64index_s64)))
+void svstnt1w_scatter_index(svbool_t, int32_t *, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64index_u64)))
+void svstnt1w_scatter_index(svbool_t, uint32_t *, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64index_s64)))
+void svstnt1w_scatter_index(svbool_t, int32_t *, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64index_u64)))
+void svstnt1w_scatter_index(svbool_t, uint32_t *, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64offset_s64)))
+void svstnt1w_scatter_offset(svbool_t, int32_t *, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64offset_u64)))
+void svstnt1w_scatter_offset(svbool_t, uint32_t *, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64offset_s64)))
+void svstnt1w_scatter_offset(svbool_t, int32_t *, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64offset_u64)))
+void svstnt1w_scatter_offset(svbool_t, uint32_t *, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_u32)))
+svuint16_t svsubhnb(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_u64)))
+svuint32_t svsubhnb(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_u16)))
+svuint8_t svsubhnb(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_s32)))
+svint16_t svsubhnb(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_s64)))
+svint32_t svsubhnb(svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_s16)))
+svint8_t svsubhnb(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_u32)))
+svuint16_t svsubhnb(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_u64)))
+svuint32_t svsubhnb(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_u16)))
+svuint8_t svsubhnb(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_s32)))
+svint16_t svsubhnb(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_s64)))
+svint32_t svsubhnb(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_s16)))
+svint8_t svsubhnb(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_u32)))
+svuint16_t svsubhnt(svuint16_t, svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_u64)))
+svuint32_t svsubhnt(svuint32_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_u16)))
+svuint8_t svsubhnt(svuint8_t, svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_s32)))
+svint16_t svsubhnt(svint16_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_s64)))
+svint32_t svsubhnt(svint32_t, svint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_s16)))
+svint8_t svsubhnt(svint8_t, svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_u32)))
+svuint16_t svsubhnt(svuint16_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_u64)))
+svuint32_t svsubhnt(svuint32_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_u16)))
+svuint8_t svsubhnt(svuint8_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_s32)))
+svint16_t svsubhnt(svint16_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_s64)))
+svint32_t svsubhnt(svint32_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_s16)))
+svint8_t svsubhnt(svint8_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_s32)))
+svint32_t svsublb(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_s64)))
+svint64_t svsublb(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_s16)))
+svint16_t svsublb(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_u32)))
+svuint32_t svsublb(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_u64)))
+svuint64_t svsublb(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_u16)))
+svuint16_t svsublb(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_s32)))
+svint32_t svsublb(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_s64)))
+svint64_t svsublb(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_s16)))
+svint16_t svsublb(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_u32)))
+svuint32_t svsublb(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_u64)))
+svuint64_t svsublb(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_u16)))
+svuint16_t svsublb(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_n_s32)))
+svint32_t svsublbt(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_n_s64)))
+svint64_t svsublbt(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_n_s16)))
+svint16_t svsublbt(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_s32)))
+svint32_t svsublbt(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_s64)))
+svint64_t svsublbt(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_s16)))
+svint16_t svsublbt(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_s32)))
+svint32_t svsublt(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_s64)))
+svint64_t svsublt(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_s16)))
+svint16_t svsublt(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_u32)))
+svuint32_t svsublt(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_u64)))
+svuint64_t svsublt(svuint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_u16)))
+svuint16_t svsublt(svuint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_s32)))
+svint32_t svsublt(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_s64)))
+svint64_t svsublt(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_s16)))
+svint16_t svsublt(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_u32)))
+svuint32_t svsublt(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_u64)))
+svuint64_t svsublt(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_u16)))
+svuint16_t svsublt(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_n_s32)))
+svint32_t svsubltb(svint16_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_n_s64)))
+svint64_t svsubltb(svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_n_s16)))
+svint16_t svsubltb(svint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_s32)))
+svint32_t svsubltb(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_s64)))
+svint64_t svsubltb(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_s16)))
+svint16_t svsubltb(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_s32)))
+svint32_t svsubwb(svint32_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_s64)))
+svint64_t svsubwb(svint64_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_s16)))
+svint16_t svsubwb(svint16_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_u32)))
+svuint32_t svsubwb(svuint32_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_u64)))
+svuint64_t svsubwb(svuint64_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_u16)))
+svuint16_t svsubwb(svuint16_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_s32)))
+svint32_t svsubwb(svint32_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_s64)))
+svint64_t svsubwb(svint64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_s16)))
+svint16_t svsubwb(svint16_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_u32)))
+svuint32_t svsubwb(svuint32_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_u64)))
+svuint64_t svsubwb(svuint64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_u16)))
+svuint16_t svsubwb(svuint16_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_s32)))
+svint32_t svsubwt(svint32_t, int16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_s64)))
+svint64_t svsubwt(svint64_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_s16)))
+svint16_t svsubwt(svint16_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_u32)))
+svuint32_t svsubwt(svuint32_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_u64)))
+svuint64_t svsubwt(svuint64_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_u16)))
+svuint16_t svsubwt(svuint16_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_s32)))
+svint32_t svsubwt(svint32_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_s64)))
+svint64_t svsubwt(svint64_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_s16)))
+svint16_t svsubwt(svint16_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_u32)))
+svuint32_t svsubwt(svuint32_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_u64)))
+svuint64_t svsubwt(svuint64_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_u16)))
+svuint16_t svsubwt(svuint16_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u8)))
+svuint8_t svtbl2(svuint8x2_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u32)))
+svuint32_t svtbl2(svuint32x2_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u64)))
+svuint64_t svtbl2(svuint64x2_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u16)))
+svuint16_t svtbl2(svuint16x2_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s8)))
+svint8_t svtbl2(svint8x2_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f64)))
+svfloat64_t svtbl2(svfloat64x2_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f32)))
+svfloat32_t svtbl2(svfloat32x2_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f16)))
+svfloat16_t svtbl2(svfloat16x2_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s32)))
+svint32_t svtbl2(svint32x2_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s64)))
+svint64_t svtbl2(svint64x2_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s16)))
+svint16_t svtbl2(svint16x2_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u8)))
+svuint8_t svtbx(svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u32)))
+svuint32_t svtbx(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u64)))
+svuint64_t svtbx(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u16)))
+svuint16_t svtbx(svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s8)))
+svint8_t svtbx(svint8_t, svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f64)))
+svfloat64_t svtbx(svfloat64_t, svfloat64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f32)))
+svfloat32_t svtbx(svfloat32_t, svfloat32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f16)))
+svfloat16_t svtbx(svfloat16_t, svfloat16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s32)))
+svint32_t svtbx(svint32_t, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s64)))
+svint64_t svtbx(svint64_t, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s16)))
+svint16_t svtbx(svint16_t, svint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s8_m)))
+svint8_t svuqadd_m(svbool_t, svint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s32_m)))
+svint32_t svuqadd_m(svbool_t, svint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s64_m)))
+svint64_t svuqadd_m(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s16_m)))
+svint16_t svuqadd_m(svbool_t, svint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s8_x)))
+svint8_t svuqadd_x(svbool_t, svint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s32_x)))
+svint32_t svuqadd_x(svbool_t, svint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s64_x)))
+svint64_t svuqadd_x(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s16_x)))
+svint16_t svuqadd_x(svbool_t, svint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s8_z)))
+svint8_t svuqadd_z(svbool_t, svint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s32_z)))
+svint32_t svuqadd_z(svbool_t, svint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s64_z)))
+svint64_t svuqadd_z(svbool_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s16_z)))
+svint16_t svuqadd_z(svbool_t, svint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s8_m)))
+svint8_t svuqadd_m(svbool_t, svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s32_m)))
+svint32_t svuqadd_m(svbool_t, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s64_m)))
+svint64_t svuqadd_m(svbool_t, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s16_m)))
+svint16_t svuqadd_m(svbool_t, svint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s8_x)))
+svint8_t svuqadd_x(svbool_t, svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s32_x)))
+svint32_t svuqadd_x(svbool_t, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s64_x)))
+svint64_t svuqadd_x(svbool_t, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s16_x)))
+svint16_t svuqadd_x(svbool_t, svint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s8_z)))
+svint8_t svuqadd_z(svbool_t, svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s32_z)))
+svint32_t svuqadd_z(svbool_t, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s64_z)))
+svint64_t svuqadd_z(svbool_t, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s16_z)))
+svint16_t svuqadd_z(svbool_t, svint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_s32)))
+svbool_t svwhilege_b8(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_s32)))
+svbool_t svwhilege_b32(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_s32)))
+svbool_t svwhilege_b64(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_s32)))
+svbool_t svwhilege_b16(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_s64)))
+svbool_t svwhilege_b8(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_s64)))
+svbool_t svwhilege_b32(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_s64)))
+svbool_t svwhilege_b64(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_s64)))
+svbool_t svwhilege_b16(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_u32)))
+svbool_t svwhilege_b8(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_u32)))
+svbool_t svwhilege_b32(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_u32)))
+svbool_t svwhilege_b64(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_u32)))
+svbool_t svwhilege_b16(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_u64)))
+svbool_t svwhilege_b8(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_u64)))
+svbool_t svwhilege_b32(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_u64)))
+svbool_t svwhilege_b64(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_u64)))
+svbool_t svwhilege_b16(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_s32)))
+svbool_t svwhilegt_b8(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_s32)))
+svbool_t svwhilegt_b32(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_s32)))
+svbool_t svwhilegt_b64(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_s32)))
+svbool_t svwhilegt_b16(int32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_s64)))
+svbool_t svwhilegt_b8(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_s64)))
+svbool_t svwhilegt_b32(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_s64)))
+svbool_t svwhilegt_b64(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_s64)))
+svbool_t svwhilegt_b16(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_u32)))
+svbool_t svwhilegt_b8(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_u32)))
+svbool_t svwhilegt_b32(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_u32)))
+svbool_t svwhilegt_b64(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_u32)))
+svbool_t svwhilegt_b16(uint32_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_u64)))
+svbool_t svwhilegt_b8(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_u64)))
+svbool_t svwhilegt_b32(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_u64)))
+svbool_t svwhilegt_b64(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_u64)))
+svbool_t svwhilegt_b16(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u8)))
+svbool_t svwhilerw(uint8_t const *, uint8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s8)))
+svbool_t svwhilerw(int8_t const *, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u64)))
+svbool_t svwhilerw(uint64_t const *, uint64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f64)))
+svbool_t svwhilerw(float64_t const *, float64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s64)))
+svbool_t svwhilerw(int64_t const *, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u16)))
+svbool_t svwhilerw(uint16_t const *, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f16)))
+svbool_t svwhilerw(float16_t const *, float16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s16)))
+svbool_t svwhilerw(int16_t const *, int16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u32)))
+svbool_t svwhilerw(uint32_t const *, uint32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f32)))
+svbool_t svwhilerw(float32_t const *, float32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s32)))
+svbool_t svwhilerw(int32_t const *, int32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u8)))
+svbool_t svwhilewr(uint8_t const *, uint8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s8)))
+svbool_t svwhilewr(int8_t const *, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u64)))
+svbool_t svwhilewr(uint64_t const *, uint64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f64)))
+svbool_t svwhilewr(float64_t const *, float64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s64)))
+svbool_t svwhilewr(int64_t const *, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u16)))
+svbool_t svwhilewr(uint16_t const *, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f16)))
+svbool_t svwhilewr(float16_t const *, float16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s16)))
+svbool_t svwhilewr(int16_t const *, int16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u32)))
+svbool_t svwhilewr(uint32_t const *, uint32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f32)))
+svbool_t svwhilewr(float32_t const *, float32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s32)))
+svbool_t svwhilewr(int32_t const *, int32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u8)))
+svuint8_t svxar(svuint8_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u32)))
+svuint32_t svxar(svuint32_t, svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u64)))
+svuint64_t svxar(svuint64_t, svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u16)))
+svuint16_t svxar(svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s8)))
+svint8_t svxar(svint8_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s32)))
+svint32_t svxar(svint32_t, svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s64)))
+svint64_t svxar(svint64_t, svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s16)))
+svint16_t svxar(svint16_t, svint16_t, uint64_t);
+#endif //defined(__ARM_FEATURE_SVE2)
+
+#if defined(__ARM_FEATURE_SVE2) && defined(__ARM_FEATURE_BF16_SCALAR_ARITHMETIC)
+#define svwhilerw_bf16(...) __builtin_sve_svwhilerw_bf16(__VA_ARGS__)
+#define svwhilewr_bf16(...) __builtin_sve_svwhilewr_bf16(__VA_ARGS__)
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_bf16)))
+svbool_t svwhilerw(bfloat16_t const *, bfloat16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_bf16)))
+svbool_t svwhilewr(bfloat16_t const *, bfloat16_t const *);
+#endif //defined(__ARM_FEATURE_SVE2) && defined(__ARM_FEATURE_BF16_SCALAR_ARITHMETIC)
+
+#if defined(__ARM_FEATURE_SVE2) && defined(__ARM_FEATURE_SVE_BF16)
+#define svtbl2_bf16(...) __builtin_sve_svtbl2_bf16(__VA_ARGS__)
+#define svtbx_bf16(...) __builtin_sve_svtbx_bf16(__VA_ARGS__)
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_bf16)))
+svbfloat16_t svtbl2(svbfloat16x2_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_bf16)))
+svbfloat16_t svtbx(svbfloat16_t, svbfloat16_t, svuint16_t);
+#endif //defined(__ARM_FEATURE_SVE2) && defined(__ARM_FEATURE_SVE_BF16)
+
+#if defined(__ARM_FEATURE_SVE2_AES)
+#define svaesd_u8(...) __builtin_sve_svaesd_u8(__VA_ARGS__)
+#define svaese_u8(...) __builtin_sve_svaese_u8(__VA_ARGS__)
+#define svaesimc_u8(...) __builtin_sve_svaesimc_u8(__VA_ARGS__)
+#define svaesmc_u8(...) __builtin_sve_svaesmc_u8(__VA_ARGS__)
+#define svpmullb_pair_n_u64(...) __builtin_sve_svpmullb_pair_n_u64(__VA_ARGS__)
+#define svpmullb_pair_u64(...) __builtin_sve_svpmullb_pair_u64(__VA_ARGS__)
+#define svpmullt_pair_n_u64(...) __builtin_sve_svpmullt_pair_n_u64(__VA_ARGS__)
+#define svpmullt_pair_u64(...) __builtin_sve_svpmullt_pair_u64(__VA_ARGS__)
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaesd_u8)))
+svuint8_t svaesd(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaese_u8)))
+svuint8_t svaese(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaesimc_u8)))
+svuint8_t svaesimc(svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaesmc_u8)))
+svuint8_t svaesmc(svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_n_u64)))
+svuint64_t svpmullb_pair(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_u64)))
+svuint64_t svpmullb_pair(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_n_u64)))
+svuint64_t svpmullt_pair(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_u64)))
+svuint64_t svpmullt_pair(svuint64_t, svuint64_t);
+#endif //defined(__ARM_FEATURE_SVE2_AES)
+
+#if defined(__ARM_FEATURE_SVE2_SHA3)
+#define svrax1_u64(...) __builtin_sve_svrax1_u64(__VA_ARGS__)
+#define svrax1_s64(...) __builtin_sve_svrax1_s64(__VA_ARGS__)
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_u64)))
+svuint64_t svrax1(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_s64)))
+svint64_t svrax1(svint64_t, svint64_t);
+#endif //defined(__ARM_FEATURE_SVE2_SHA3)
+
+#if defined(__ARM_FEATURE_SVE2_SM4)
+#define svsm4e_u32(...) __builtin_sve_svsm4e_u32(__VA_ARGS__)
+#define svsm4ekey_u32(...) __builtin_sve_svsm4ekey_u32(__VA_ARGS__)
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsm4e_u32)))
+svuint32_t svsm4e(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsm4ekey_u32)))
+svuint32_t svsm4ekey(svuint32_t, svuint32_t);
+#endif //defined(__ARM_FEATURE_SVE2_SM4)
+
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svbfdot_n_f32(...) __builtin_sve_svbfdot_n_f32(__VA_ARGS__)
+#define svbfdot_f32(...) __builtin_sve_svbfdot_f32(__VA_ARGS__)
+#define svbfdot_lane_f32(...) __builtin_sve_svbfdot_lane_f32(__VA_ARGS__)
+#define svbfmlalb_n_f32(...) __builtin_sve_svbfmlalb_n_f32(__VA_ARGS__)
+#define svbfmlalb_f32(...) __builtin_sve_svbfmlalb_f32(__VA_ARGS__)
+#define svbfmlalb_lane_f32(...) __builtin_sve_svbfmlalb_lane_f32(__VA_ARGS__)
+#define svbfmlalt_n_f32(...) __builtin_sve_svbfmlalt_n_f32(__VA_ARGS__)
+#define svbfmlalt_f32(...) __builtin_sve_svbfmlalt_f32(__VA_ARGS__)
+#define svbfmlalt_lane_f32(...) __builtin_sve_svbfmlalt_lane_f32(__VA_ARGS__)
+#define svbfmmla_f32(...) __builtin_sve_svbfmmla_f32(__VA_ARGS__)
+#define svclasta_n_bf16(...) __builtin_sve_svclasta_n_bf16(__VA_ARGS__)
+#define svclasta_bf16(...) __builtin_sve_svclasta_bf16(__VA_ARGS__)
+#define svclastb_n_bf16(...) __builtin_sve_svclastb_n_bf16(__VA_ARGS__)
+#define svclastb_bf16(...) __builtin_sve_svclastb_bf16(__VA_ARGS__)
+#define svcnt_bf16_m(...) __builtin_sve_svcnt_bf16_m(__VA_ARGS__)
+#define svcnt_bf16_x(...) __builtin_sve_svcnt_bf16_x(__VA_ARGS__)
+#define svcnt_bf16_z(...) __builtin_sve_svcnt_bf16_z(__VA_ARGS__)
+#define svcreate2_bf16(...) __builtin_sve_svcreate2_bf16(__VA_ARGS__)
+#define svcreate3_bf16(...) __builtin_sve_svcreate3_bf16(__VA_ARGS__)
+#define svcreate4_bf16(...) __builtin_sve_svcreate4_bf16(__VA_ARGS__)
+#define svcvt_bf16_f32_m(...) __builtin_sve_svcvt_bf16_f32_m(__VA_ARGS__)
+#define svcvt_bf16_f32_x(...) __builtin_sve_svcvt_bf16_f32_x(__VA_ARGS__)
+#define svcvt_bf16_f32_z(...) __builtin_sve_svcvt_bf16_f32_z(__VA_ARGS__)
+#define svcvtnt_bf16_f32_m(...) __builtin_sve_svcvtnt_bf16_f32_m(__VA_ARGS__)
+#define svdup_n_bf16(...) __builtin_sve_svdup_n_bf16(__VA_ARGS__)
+#define svdup_n_bf16_m(...) __builtin_sve_svdup_n_bf16_m(__VA_ARGS__)
+#define svdup_n_bf16_x(...) __builtin_sve_svdup_n_bf16_x(__VA_ARGS__)
+#define svdup_n_bf16_z(...) __builtin_sve_svdup_n_bf16_z(__VA_ARGS__)
+#define svdup_lane_bf16(...) __builtin_sve_svdup_lane_bf16(__VA_ARGS__)
+#define svdupq_n_bf16(...) __builtin_sve_svdupq_n_bf16(__VA_ARGS__)
+#define svdupq_lane_bf16(...) __builtin_sve_svdupq_lane_bf16(__VA_ARGS__)
+#define svext_bf16(...) __builtin_sve_svext_bf16(__VA_ARGS__)
+#define svget2_bf16(...) __builtin_sve_svget2_bf16(__VA_ARGS__)
+#define svget3_bf16(...) __builtin_sve_svget3_bf16(__VA_ARGS__)
+#define svget4_bf16(...) __builtin_sve_svget4_bf16(__VA_ARGS__)
+#define svinsr_n_bf16(...) __builtin_sve_svinsr_n_bf16(__VA_ARGS__)
+#define svlasta_bf16(...) __builtin_sve_svlasta_bf16(__VA_ARGS__)
+#define svlastb_bf16(...) __builtin_sve_svlastb_bf16(__VA_ARGS__)
+#define svld1_bf16(...) __builtin_sve_svld1_bf16(__VA_ARGS__)
+#define svld1_vnum_bf16(...) __builtin_sve_svld1_vnum_bf16(__VA_ARGS__)
+#define svld1rq_bf16(...) __builtin_sve_svld1rq_bf16(__VA_ARGS__)
+#define svld2_bf16(...) __builtin_sve_svld2_bf16(__VA_ARGS__)
+#define svld2_vnum_bf16(...) __builtin_sve_svld2_vnum_bf16(__VA_ARGS__)
+#define svld3_bf16(...) __builtin_sve_svld3_bf16(__VA_ARGS__)
+#define svld3_vnum_bf16(...) __builtin_sve_svld3_vnum_bf16(__VA_ARGS__)
+#define svld4_bf16(...) __builtin_sve_svld4_bf16(__VA_ARGS__)
+#define svld4_vnum_bf16(...) __builtin_sve_svld4_vnum_bf16(__VA_ARGS__)
+#define svldff1_bf16(...) __builtin_sve_svldff1_bf16(__VA_ARGS__)
+#define svldff1_vnum_bf16(...) __builtin_sve_svldff1_vnum_bf16(__VA_ARGS__)
+#define svldnf1_bf16(...) __builtin_sve_svldnf1_bf16(__VA_ARGS__)
+#define svldnf1_vnum_bf16(...) __builtin_sve_svldnf1_vnum_bf16(__VA_ARGS__)
+#define svldnt1_bf16(...) __builtin_sve_svldnt1_bf16(__VA_ARGS__)
+#define svldnt1_vnum_bf16(...) __builtin_sve_svldnt1_vnum_bf16(__VA_ARGS__)
+#define svlen_bf16(...) __builtin_sve_svlen_bf16(__VA_ARGS__)
+#define svrev_bf16(...) __builtin_sve_svrev_bf16(__VA_ARGS__)
+#define svsel_bf16(...) __builtin_sve_svsel_bf16(__VA_ARGS__)
+#define svset2_bf16(...) __builtin_sve_svset2_bf16(__VA_ARGS__)
+#define svset3_bf16(...) __builtin_sve_svset3_bf16(__VA_ARGS__)
+#define svset4_bf16(...) __builtin_sve_svset4_bf16(__VA_ARGS__)
+#define svsplice_bf16(...) __builtin_sve_svsplice_bf16(__VA_ARGS__)
+#define svst1_bf16(...) __builtin_sve_svst1_bf16(__VA_ARGS__)
+#define svst1_vnum_bf16(...) __builtin_sve_svst1_vnum_bf16(__VA_ARGS__)
+#define svst2_bf16(...) __builtin_sve_svst2_bf16(__VA_ARGS__)
+#define svst2_vnum_bf16(...) __builtin_sve_svst2_vnum_bf16(__VA_ARGS__)
+#define svst3_bf16(...) __builtin_sve_svst3_bf16(__VA_ARGS__)
+#define svst3_vnum_bf16(...) __builtin_sve_svst3_vnum_bf16(__VA_ARGS__)
+#define svst4_bf16(...) __builtin_sve_svst4_bf16(__VA_ARGS__)
+#define svst4_vnum_bf16(...) __builtin_sve_svst4_vnum_bf16(__VA_ARGS__)
+#define svstnt1_bf16(...) __builtin_sve_svstnt1_bf16(__VA_ARGS__)
+#define svstnt1_vnum_bf16(...) __builtin_sve_svstnt1_vnum_bf16(__VA_ARGS__)
+#define svtbl_bf16(...) __builtin_sve_svtbl_bf16(__VA_ARGS__)
+#define svtrn1_bf16(...) __builtin_sve_svtrn1_bf16(__VA_ARGS__)
+#define svtrn2_bf16(...) __builtin_sve_svtrn2_bf16(__VA_ARGS__)
+#define svundef2_bf16(...) __builtin_sve_svundef2_bf16(__VA_ARGS__)
+#define svundef3_bf16(...) __builtin_sve_svundef3_bf16(__VA_ARGS__)
+#define svundef4_bf16(...) __builtin_sve_svundef4_bf16(__VA_ARGS__)
+#define svundef_bf16(...) __builtin_sve_svundef_bf16(__VA_ARGS__)
+#define svuzp1_bf16(...) __builtin_sve_svuzp1_bf16(__VA_ARGS__)
+#define svuzp2_bf16(...) __builtin_sve_svuzp2_bf16(__VA_ARGS__)
+#define svzip1_bf16(...) __builtin_sve_svzip1_bf16(__VA_ARGS__)
+#define svzip2_bf16(...) __builtin_sve_svzip2_bf16(__VA_ARGS__)
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfdot_n_f32)))
+svfloat32_t svbfdot(svfloat32_t, svbfloat16_t, bfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfdot_f32)))
+svfloat32_t svbfdot(svfloat32_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfdot_lane_f32)))
+svfloat32_t svbfdot_lane(svfloat32_t, svbfloat16_t, svbfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlalb_n_f32)))
+svfloat32_t svbfmlalb(svfloat32_t, svbfloat16_t, bfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlalb_f32)))
+svfloat32_t svbfmlalb(svfloat32_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlalb_lane_f32)))
+svfloat32_t svbfmlalb_lane(svfloat32_t, svbfloat16_t, svbfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlalt_n_f32)))
+svfloat32_t svbfmlalt(svfloat32_t, svbfloat16_t, bfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlalt_f32)))
+svfloat32_t svbfmlalt(svfloat32_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlalt_lane_f32)))
+svfloat32_t svbfmlalt_lane(svfloat32_t, svbfloat16_t, svbfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmmla_f32)))
+svfloat32_t svbfmmla(svfloat32_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_bf16)))
+bfloat16_t svclasta(svbool_t, bfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_bf16)))
+svbfloat16_t svclasta(svbool_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_bf16)))
+bfloat16_t svclastb(svbool_t, bfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_bf16)))
+svbfloat16_t svclastb(svbool_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_bf16_m)))
+svuint16_t svcnt_m(svuint16_t, svbool_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_bf16_x)))
+svuint16_t svcnt_x(svbool_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_bf16_z)))
+svuint16_t svcnt_z(svbool_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_bf16)))
+svbfloat16x2_t svcreate2(svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_bf16)))
+svbfloat16x3_t svcreate3(svbfloat16_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_bf16)))
+svbfloat16x4_t svcreate4(svbfloat16_t, svbfloat16_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_bf16_f32_m)))
+svbfloat16_t svcvt_bf16_m(svbfloat16_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_bf16_f32_x)))
+svbfloat16_t svcvt_bf16_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_bf16_f32_z)))
+svbfloat16_t svcvt_bf16_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtnt_bf16_f32_m)))
+svbfloat16_t svcvtnt_bf16_m(svbfloat16_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16)))
+svbfloat16_t svdup_bf16(bfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16_m)))
+svbfloat16_t svdup_bf16_m(svbfloat16_t, svbool_t, bfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16_x)))
+svbfloat16_t svdup_bf16_x(svbool_t, bfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16_z)))
+svbfloat16_t svdup_bf16_z(svbool_t, bfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_bf16)))
+svbfloat16_t svdup_lane(svbfloat16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_bf16)))
+svbfloat16_t svdupq_bf16(bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_bf16)))
+svbfloat16_t svdupq_lane(svbfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_bf16)))
+svbfloat16_t svext(svbfloat16_t, svbfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_bf16)))
+svbfloat16_t svget2(svbfloat16x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_bf16)))
+svbfloat16_t svget3(svbfloat16x3_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_bf16)))
+svbfloat16_t svget4(svbfloat16x4_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_bf16)))
+svbfloat16_t svinsr(svbfloat16_t, bfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_bf16)))
+bfloat16_t svlasta(svbool_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_bf16)))
+bfloat16_t svlastb(svbool_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_bf16)))
+svbfloat16_t svld1(svbool_t, bfloat16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_bf16)))
+svbfloat16_t svld1_vnum(svbool_t, bfloat16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_bf16)))
+svbfloat16_t svld1rq(svbool_t, bfloat16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_bf16)))
+svbfloat16x2_t svld2(svbool_t, bfloat16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_bf16)))
+svbfloat16x2_t svld2_vnum(svbool_t, bfloat16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_bf16)))
+svbfloat16x3_t svld3(svbool_t, bfloat16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_bf16)))
+svbfloat16x3_t svld3_vnum(svbool_t, bfloat16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_bf16)))
+svbfloat16x4_t svld4(svbool_t, bfloat16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_bf16)))
+svbfloat16x4_t svld4_vnum(svbool_t, bfloat16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_bf16)))
+svbfloat16_t svldff1(svbool_t, bfloat16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_bf16)))
+svbfloat16_t svldff1_vnum(svbool_t, bfloat16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_bf16)))
+svbfloat16_t svldnf1(svbool_t, bfloat16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_bf16)))
+svbfloat16_t svldnf1_vnum(svbool_t, bfloat16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_bf16)))
+svbfloat16_t svldnt1(svbool_t, bfloat16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_bf16)))
+svbfloat16_t svldnt1_vnum(svbool_t, bfloat16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_bf16)))
+uint64_t svlen(svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_bf16)))
+svbfloat16_t svrev(svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_bf16)))
+svbfloat16_t svsel(svbool_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_bf16)))
+svbfloat16x2_t svset2(svbfloat16x2_t, uint64_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_bf16)))
+svbfloat16x3_t svset3(svbfloat16x3_t, uint64_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_bf16)))
+svbfloat16x4_t svset4(svbfloat16x4_t, uint64_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_bf16)))
+svbfloat16_t svsplice(svbool_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_bf16)))
+void svst1(svbool_t, bfloat16_t *, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_bf16)))
+void svst1_vnum(svbool_t, bfloat16_t *, int64_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_bf16)))
+void svst2(svbool_t, bfloat16_t *, svbfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_bf16)))
+void svst2_vnum(svbool_t, bfloat16_t *, int64_t, svbfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_bf16)))
+void svst3(svbool_t, bfloat16_t *, svbfloat16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_bf16)))
+void svst3_vnum(svbool_t, bfloat16_t *, int64_t, svbfloat16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_bf16)))
+void svst4(svbool_t, bfloat16_t *, svbfloat16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_bf16)))
+void svst4_vnum(svbool_t, bfloat16_t *, int64_t, svbfloat16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_bf16)))
+void svstnt1(svbool_t, bfloat16_t *, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_bf16)))
+void svstnt1_vnum(svbool_t, bfloat16_t *, int64_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_bf16)))
+svbfloat16_t svtbl(svbfloat16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_bf16)))
+svbfloat16_t svtrn1(svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_bf16)))
+svbfloat16_t svtrn2(svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_bf16)))
+svbfloat16_t svuzp1(svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_bf16)))
+svbfloat16_t svuzp2(svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_bf16)))
+svbfloat16_t svzip1(svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_bf16)))
+svbfloat16_t svzip2(svbfloat16_t, svbfloat16_t);
+#endif //defined(__ARM_FEATURE_SVE_BF16)
+
+#if defined(__ARM_FEATURE_SVE_MATMUL_FP32)
+#define svmmla_f32(...) __builtin_sve_svmmla_f32(__VA_ARGS__)
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmmla_f32)))
+svfloat32_t svmmla(svfloat32_t, svfloat32_t, svfloat32_t);
+#endif //defined(__ARM_FEATURE_SVE_MATMUL_FP32)
+
+#if defined(__ARM_FEATURE_SVE_MATMUL_FP64)
+#define svld1ro_u8(...) __builtin_sve_svld1ro_u8(__VA_ARGS__)
+#define svld1ro_u32(...) __builtin_sve_svld1ro_u32(__VA_ARGS__)
+#define svld1ro_u64(...) __builtin_sve_svld1ro_u64(__VA_ARGS__)
+#define svld1ro_u16(...) __builtin_sve_svld1ro_u16(__VA_ARGS__)
+#define svld1ro_s8(...) __builtin_sve_svld1ro_s8(__VA_ARGS__)
+#define svld1ro_f64(...) __builtin_sve_svld1ro_f64(__VA_ARGS__)
+#define svld1ro_f32(...) __builtin_sve_svld1ro_f32(__VA_ARGS__)
+#define svld1ro_f16(...) __builtin_sve_svld1ro_f16(__VA_ARGS__)
+#define svld1ro_s32(...) __builtin_sve_svld1ro_s32(__VA_ARGS__)
+#define svld1ro_s64(...) __builtin_sve_svld1ro_s64(__VA_ARGS__)
+#define svld1ro_s16(...) __builtin_sve_svld1ro_s16(__VA_ARGS__)
+#define svmmla_f64(...) __builtin_sve_svmmla_f64(__VA_ARGS__)
+#define svtrn1q_u8(...) __builtin_sve_svtrn1q_u8(__VA_ARGS__)
+#define svtrn1q_u32(...) __builtin_sve_svtrn1q_u32(__VA_ARGS__)
+#define svtrn1q_u64(...) __builtin_sve_svtrn1q_u64(__VA_ARGS__)
+#define svtrn1q_u16(...) __builtin_sve_svtrn1q_u16(__VA_ARGS__)
+#define svtrn1q_s8(...) __builtin_sve_svtrn1q_s8(__VA_ARGS__)
+#define svtrn1q_f64(...) __builtin_sve_svtrn1q_f64(__VA_ARGS__)
+#define svtrn1q_f32(...) __builtin_sve_svtrn1q_f32(__VA_ARGS__)
+#define svtrn1q_f16(...) __builtin_sve_svtrn1q_f16(__VA_ARGS__)
+#define svtrn1q_s32(...) __builtin_sve_svtrn1q_s32(__VA_ARGS__)
+#define svtrn1q_s64(...) __builtin_sve_svtrn1q_s64(__VA_ARGS__)
+#define svtrn1q_s16(...) __builtin_sve_svtrn1q_s16(__VA_ARGS__)
+#define svtrn2q_u8(...) __builtin_sve_svtrn2q_u8(__VA_ARGS__)
+#define svtrn2q_u32(...) __builtin_sve_svtrn2q_u32(__VA_ARGS__)
+#define svtrn2q_u64(...) __builtin_sve_svtrn2q_u64(__VA_ARGS__)
+#define svtrn2q_u16(...) __builtin_sve_svtrn2q_u16(__VA_ARGS__)
+#define svtrn2q_s8(...) __builtin_sve_svtrn2q_s8(__VA_ARGS__)
+#define svtrn2q_f64(...) __builtin_sve_svtrn2q_f64(__VA_ARGS__)
+#define svtrn2q_f32(...) __builtin_sve_svtrn2q_f32(__VA_ARGS__)
+#define svtrn2q_f16(...) __builtin_sve_svtrn2q_f16(__VA_ARGS__)
+#define svtrn2q_s32(...) __builtin_sve_svtrn2q_s32(__VA_ARGS__)
+#define svtrn2q_s64(...) __builtin_sve_svtrn2q_s64(__VA_ARGS__)
+#define svtrn2q_s16(...) __builtin_sve_svtrn2q_s16(__VA_ARGS__)
+#define svuzp1q_u8(...) __builtin_sve_svuzp1q_u8(__VA_ARGS__)
+#define svuzp1q_u32(...) __builtin_sve_svuzp1q_u32(__VA_ARGS__)
+#define svuzp1q_u64(...) __builtin_sve_svuzp1q_u64(__VA_ARGS__)
+#define svuzp1q_u16(...) __builtin_sve_svuzp1q_u16(__VA_ARGS__)
+#define svuzp1q_s8(...) __builtin_sve_svuzp1q_s8(__VA_ARGS__)
+#define svuzp1q_f64(...) __builtin_sve_svuzp1q_f64(__VA_ARGS__)
+#define svuzp1q_f32(...) __builtin_sve_svuzp1q_f32(__VA_ARGS__)
+#define svuzp1q_f16(...) __builtin_sve_svuzp1q_f16(__VA_ARGS__)
+#define svuzp1q_s32(...) __builtin_sve_svuzp1q_s32(__VA_ARGS__)
+#define svuzp1q_s64(...) __builtin_sve_svuzp1q_s64(__VA_ARGS__)
+#define svuzp1q_s16(...) __builtin_sve_svuzp1q_s16(__VA_ARGS__)
+#define svuzp2q_u8(...) __builtin_sve_svuzp2q_u8(__VA_ARGS__)
+#define svuzp2q_u32(...) __builtin_sve_svuzp2q_u32(__VA_ARGS__)
+#define svuzp2q_u64(...) __builtin_sve_svuzp2q_u64(__VA_ARGS__)
+#define svuzp2q_u16(...) __builtin_sve_svuzp2q_u16(__VA_ARGS__)
+#define svuzp2q_s8(...) __builtin_sve_svuzp2q_s8(__VA_ARGS__)
+#define svuzp2q_f64(...) __builtin_sve_svuzp2q_f64(__VA_ARGS__)
+#define svuzp2q_f32(...) __builtin_sve_svuzp2q_f32(__VA_ARGS__)
+#define svuzp2q_f16(...) __builtin_sve_svuzp2q_f16(__VA_ARGS__)
+#define svuzp2q_s32(...) __builtin_sve_svuzp2q_s32(__VA_ARGS__)
+#define svuzp2q_s64(...) __builtin_sve_svuzp2q_s64(__VA_ARGS__)
+#define svuzp2q_s16(...) __builtin_sve_svuzp2q_s16(__VA_ARGS__)
+#define svzip1q_u8(...) __builtin_sve_svzip1q_u8(__VA_ARGS__)
+#define svzip1q_u32(...) __builtin_sve_svzip1q_u32(__VA_ARGS__)
+#define svzip1q_u64(...) __builtin_sve_svzip1q_u64(__VA_ARGS__)
+#define svzip1q_u16(...) __builtin_sve_svzip1q_u16(__VA_ARGS__)
+#define svzip1q_s8(...) __builtin_sve_svzip1q_s8(__VA_ARGS__)
+#define svzip1q_f64(...) __builtin_sve_svzip1q_f64(__VA_ARGS__)
+#define svzip1q_f32(...) __builtin_sve_svzip1q_f32(__VA_ARGS__)
+#define svzip1q_f16(...) __builtin_sve_svzip1q_f16(__VA_ARGS__)
+#define svzip1q_s32(...) __builtin_sve_svzip1q_s32(__VA_ARGS__)
+#define svzip1q_s64(...) __builtin_sve_svzip1q_s64(__VA_ARGS__)
+#define svzip1q_s16(...) __builtin_sve_svzip1q_s16(__VA_ARGS__)
+#define svzip2q_u8(...) __builtin_sve_svzip2q_u8(__VA_ARGS__)
+#define svzip2q_u32(...) __builtin_sve_svzip2q_u32(__VA_ARGS__)
+#define svzip2q_u64(...) __builtin_sve_svzip2q_u64(__VA_ARGS__)
+#define svzip2q_u16(...) __builtin_sve_svzip2q_u16(__VA_ARGS__)
+#define svzip2q_s8(...) __builtin_sve_svzip2q_s8(__VA_ARGS__)
+#define svzip2q_f64(...) __builtin_sve_svzip2q_f64(__VA_ARGS__)
+#define svzip2q_f32(...) __builtin_sve_svzip2q_f32(__VA_ARGS__)
+#define svzip2q_f16(...) __builtin_sve_svzip2q_f16(__VA_ARGS__)
+#define svzip2q_s32(...) __builtin_sve_svzip2q_s32(__VA_ARGS__)
+#define svzip2q_s64(...) __builtin_sve_svzip2q_s64(__VA_ARGS__)
+#define svzip2q_s16(...) __builtin_sve_svzip2q_s16(__VA_ARGS__)
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_u8)))
+svuint8_t svld1ro(svbool_t, uint8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_u32)))
+svuint32_t svld1ro(svbool_t, uint32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_u64)))
+svuint64_t svld1ro(svbool_t, uint64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_u16)))
+svuint16_t svld1ro(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_s8)))
+svint8_t svld1ro(svbool_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_f64)))
+svfloat64_t svld1ro(svbool_t, float64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_f32)))
+svfloat32_t svld1ro(svbool_t, float32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_f16)))
+svfloat16_t svld1ro(svbool_t, float16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_s32)))
+svint32_t svld1ro(svbool_t, int32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_s64)))
+svint64_t svld1ro(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_s16)))
+svint16_t svld1ro(svbool_t, int16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmmla_f64)))
+svfloat64_t svmmla(svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_u8)))
+svuint8_t svtrn1q(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_u32)))
+svuint32_t svtrn1q(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_u64)))
+svuint64_t svtrn1q(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_u16)))
+svuint16_t svtrn1q(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_s8)))
+svint8_t svtrn1q(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_f64)))
+svfloat64_t svtrn1q(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_f32)))
+svfloat32_t svtrn1q(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_f16)))
+svfloat16_t svtrn1q(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_s32)))
+svint32_t svtrn1q(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_s64)))
+svint64_t svtrn1q(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_s16)))
+svint16_t svtrn1q(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_u8)))
+svuint8_t svtrn2q(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_u32)))
+svuint32_t svtrn2q(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_u64)))
+svuint64_t svtrn2q(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_u16)))
+svuint16_t svtrn2q(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_s8)))
+svint8_t svtrn2q(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_f64)))
+svfloat64_t svtrn2q(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_f32)))
+svfloat32_t svtrn2q(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_f16)))
+svfloat16_t svtrn2q(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_s32)))
+svint32_t svtrn2q(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_s64)))
+svint64_t svtrn2q(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_s16)))
+svint16_t svtrn2q(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_u8)))
+svuint8_t svuzp1q(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_u32)))
+svuint32_t svuzp1q(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_u64)))
+svuint64_t svuzp1q(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_u16)))
+svuint16_t svuzp1q(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_s8)))
+svint8_t svuzp1q(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_f64)))
+svfloat64_t svuzp1q(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_f32)))
+svfloat32_t svuzp1q(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_f16)))
+svfloat16_t svuzp1q(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_s32)))
+svint32_t svuzp1q(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_s64)))
+svint64_t svuzp1q(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_s16)))
+svint16_t svuzp1q(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_u8)))
+svuint8_t svuzp2q(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_u32)))
+svuint32_t svuzp2q(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_u64)))
+svuint64_t svuzp2q(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_u16)))
+svuint16_t svuzp2q(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_s8)))
+svint8_t svuzp2q(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_f64)))
+svfloat64_t svuzp2q(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_f32)))
+svfloat32_t svuzp2q(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_f16)))
+svfloat16_t svuzp2q(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_s32)))
+svint32_t svuzp2q(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_s64)))
+svint64_t svuzp2q(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_s16)))
+svint16_t svuzp2q(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_u8)))
+svuint8_t svzip1q(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_u32)))
+svuint32_t svzip1q(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_u64)))
+svuint64_t svzip1q(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_u16)))
+svuint16_t svzip1q(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_s8)))
+svint8_t svzip1q(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_f64)))
+svfloat64_t svzip1q(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_f32)))
+svfloat32_t svzip1q(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_f16)))
+svfloat16_t svzip1q(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_s32)))
+svint32_t svzip1q(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_s64)))
+svint64_t svzip1q(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_s16)))
+svint16_t svzip1q(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_u8)))
+svuint8_t svzip2q(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_u32)))
+svuint32_t svzip2q(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_u64)))
+svuint64_t svzip2q(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_u16)))
+svuint16_t svzip2q(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_s8)))
+svint8_t svzip2q(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_f64)))
+svfloat64_t svzip2q(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_f32)))
+svfloat32_t svzip2q(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_f16)))
+svfloat16_t svzip2q(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_s32)))
+svint32_t svzip2q(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_s64)))
+svint64_t svzip2q(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_s16)))
+svint16_t svzip2q(svint16_t, svint16_t);
+#endif //defined(__ARM_FEATURE_SVE_MATMUL_FP64)
+
+#if defined(__ARM_FEATURE_SVE_MATMUL_FP64) && defined(__ARM_FEATURE_SVE_BF16)
+#define svld1ro_bf16(...) __builtin_sve_svld1ro_bf16(__VA_ARGS__)
+#define svtrn1q_bf16(...) __builtin_sve_svtrn1q_bf16(__VA_ARGS__)
+#define svtrn2q_bf16(...) __builtin_sve_svtrn2q_bf16(__VA_ARGS__)
+#define svuzp1q_bf16(...) __builtin_sve_svuzp1q_bf16(__VA_ARGS__)
+#define svuzp2q_bf16(...) __builtin_sve_svuzp2q_bf16(__VA_ARGS__)
+#define svzip1q_bf16(...) __builtin_sve_svzip1q_bf16(__VA_ARGS__)
+#define svzip2q_bf16(...) __builtin_sve_svzip2q_bf16(__VA_ARGS__)
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_bf16)))
+svbfloat16_t svld1ro(svbool_t, bfloat16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_bf16)))
+svbfloat16_t svtrn1q(svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_bf16)))
+svbfloat16_t svtrn2q(svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_bf16)))
+svbfloat16_t svuzp1q(svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_bf16)))
+svbfloat16_t svuzp2q(svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_bf16)))
+svbfloat16_t svzip1q(svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_bf16)))
+svbfloat16_t svzip2q(svbfloat16_t, svbfloat16_t);
+#endif //defined(__ARM_FEATURE_SVE_MATMUL_FP64) && defined(__ARM_FEATURE_SVE_BF16)
+
+#if defined(__ARM_FEATURE_SVE_MATMUL_INT8)
+#define svmmla_s32(...) __builtin_sve_svmmla_s32(__VA_ARGS__)
+#define svmmla_u32(...) __builtin_sve_svmmla_u32(__VA_ARGS__)
+#define svsudot_n_s32(...) __builtin_sve_svsudot_n_s32(__VA_ARGS__)
+#define svsudot_s32(...) __builtin_sve_svsudot_s32(__VA_ARGS__)
+#define svsudot_lane_s32(...) __builtin_sve_svsudot_lane_s32(__VA_ARGS__)
+#define svusdot_n_s32(...) __builtin_sve_svusdot_n_s32(__VA_ARGS__)
+#define svusdot_s32(...) __builtin_sve_svusdot_s32(__VA_ARGS__)
+#define svusdot_lane_s32(...) __builtin_sve_svusdot_lane_s32(__VA_ARGS__)
+#define svusmmla_s32(...) __builtin_sve_svusmmla_s32(__VA_ARGS__)
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmmla_s32)))
+svint32_t svmmla(svint32_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmmla_u32)))
+svuint32_t svmmla(svuint32_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsudot_n_s32)))
+svint32_t svsudot(svint32_t, svint8_t, uint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsudot_s32)))
+svint32_t svsudot(svint32_t, svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsudot_lane_s32)))
+svint32_t svsudot_lane(svint32_t, svint8_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svusdot_n_s32)))
+svint32_t svusdot(svint32_t, svuint8_t, int8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svusdot_s32)))
+svint32_t svusdot(svint32_t, svuint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svusdot_lane_s32)))
+svint32_t svusdot_lane(svint32_t, svuint8_t, svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svusmmla_s32)))
+svint32_t svusmmla(svint32_t, svuint8_t, svint8_t);
+#endif //defined(__ARM_FEATURE_SVE_MATMUL_INT8)
+#if defined(__ARM_FEATURE_SVE_BF16)
+#define svcvtnt_bf16_x svcvtnt_bf16_m
+#define svcvtnt_bf16_f32_x svcvtnt_bf16_f32_m
+#endif /*__ARM_FEATURE_SVE_BF16 */
+
+#if defined(__ARM_FEATURE_SVE2)
+#define svcvtnt_f16_x svcvtnt_f16_m
+#define svcvtnt_f16_f32_x svcvtnt_f16_f32_m
+#define svcvtnt_f32_x svcvtnt_f32_m
+#define svcvtnt_f32_f64_x svcvtnt_f32_f64_m
+
+#define svcvtxnt_f32_x svcvtxnt_f32_m
+#define svcvtxnt_f32_f64_x svcvtxnt_f32_f64_m
+
+#endif /*__ARM_FEATURE_SVE2 */
+
+#ifdef __cplusplus
+} // extern "C"
+#endif
+
+#endif /*__ARM_FEATURE_SVE */
+
+#endif /* __ARM_SVE_H */
--- /dev/null
+/*===---- armintr.h - ARM Windows intrinsics -------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+/* Only include this if we're compiling for the windows platform. */
+#ifndef _MSC_VER
+#include_next <armintr.h>
+#else
+
+#ifndef __ARMINTR_H
+#define __ARMINTR_H
+
+typedef enum
+{
+ _ARM_BARRIER_SY = 0xF,
+ _ARM_BARRIER_ST = 0xE,
+ _ARM_BARRIER_ISH = 0xB,
+ _ARM_BARRIER_ISHST = 0xA,
+ _ARM_BARRIER_NSH = 0x7,
+ _ARM_BARRIER_NSHST = 0x6,
+ _ARM_BARRIER_OSH = 0x3,
+ _ARM_BARRIER_OSHST = 0x2
+} _ARMINTR_BARRIER_TYPE;
+
+#endif /* __ARMINTR_H */
+#endif /* _MSC_VER */
/*===---- avx2intrin.h - AVX2 intrinsics -----------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#define __AVX2INTRIN_H
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx2")))
+#define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("avx2"), __min_vector_width__(256)))
+#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx2"), __min_vector_width__(128)))
/* SSE4 Multiple Packed Sums of Absolute Difference. */
#define _mm256_mpsadbw_epu8(X, Y, M) \
(__m256i)__builtin_ia32_mpsadbw256((__v32qi)(__m256i)(X), \
(__v32qi)(__m256i)(Y), (int)(M))
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_abs_epi8(__m256i __a)
{
return (__m256i)__builtin_ia32_pabsb256((__v32qi)__a);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_abs_epi16(__m256i __a)
{
return (__m256i)__builtin_ia32_pabsw256((__v16hi)__a);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_abs_epi32(__m256i __a)
{
return (__m256i)__builtin_ia32_pabsd256((__v8si)__a);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_packs_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_packsswb256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_packs_epi32(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_packssdw256((__v8si)__a, (__v8si)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_packus_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_packuswb256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_packus_epi32(__m256i __V1, __m256i __V2)
{
return (__m256i) __builtin_ia32_packusdw256((__v8si)__V1, (__v8si)__V2);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_add_epi8(__m256i __a, __m256i __b)
{
return (__m256i)((__v32qu)__a + (__v32qu)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_add_epi16(__m256i __a, __m256i __b)
{
return (__m256i)((__v16hu)__a + (__v16hu)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_add_epi32(__m256i __a, __m256i __b)
{
return (__m256i)((__v8su)__a + (__v8su)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_add_epi64(__m256i __a, __m256i __b)
{
return (__m256i)((__v4du)__a + (__v4du)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_adds_epi8(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_paddsb256((__v32qi)__a, (__v32qi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_adds_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_paddsw256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_adds_epu8(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_paddusb256((__v32qi)__a, (__v32qi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_adds_epu16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_paddusw256((__v16hi)__a, (__v16hi)__b);
}
-#define _mm256_alignr_epi8(a, b, n) __extension__ ({ \
+#define _mm256_alignr_epi8(a, b, n) \
(__m256i)__builtin_ia32_palignr256((__v32qi)(__m256i)(a), \
- (__v32qi)(__m256i)(b), (n)); })
+ (__v32qi)(__m256i)(b), (n))
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_and_si256(__m256i __a, __m256i __b)
{
return (__m256i)((__v4du)__a & (__v4du)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_andnot_si256(__m256i __a, __m256i __b)
{
return (__m256i)(~(__v4du)__a & (__v4du)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_avg_epu8(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pavgb256((__v32qi)__a, (__v32qi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_avg_epu16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pavgw256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_blendv_epi8(__m256i __V1, __m256i __V2, __m256i __M)
{
return (__m256i)__builtin_ia32_pblendvb256((__v32qi)__V1, (__v32qi)__V2,
(__v32qi)__M);
}
-#define _mm256_blend_epi16(V1, V2, M) __extension__ ({ \
- (__m256i)__builtin_shufflevector((__v16hi)(__m256i)(V1), \
- (__v16hi)(__m256i)(V2), \
- (((M) & 0x01) ? 16 : 0), \
- (((M) & 0x02) ? 17 : 1), \
- (((M) & 0x04) ? 18 : 2), \
- (((M) & 0x08) ? 19 : 3), \
- (((M) & 0x10) ? 20 : 4), \
- (((M) & 0x20) ? 21 : 5), \
- (((M) & 0x40) ? 22 : 6), \
- (((M) & 0x80) ? 23 : 7), \
- (((M) & 0x01) ? 24 : 8), \
- (((M) & 0x02) ? 25 : 9), \
- (((M) & 0x04) ? 26 : 10), \
- (((M) & 0x08) ? 27 : 11), \
- (((M) & 0x10) ? 28 : 12), \
- (((M) & 0x20) ? 29 : 13), \
- (((M) & 0x40) ? 30 : 14), \
- (((M) & 0x80) ? 31 : 15)); })
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+#define _mm256_blend_epi16(V1, V2, M) \
+ (__m256i)__builtin_ia32_pblendw256((__v16hi)(__m256i)(V1), \
+ (__v16hi)(__m256i)(V2), (int)(M))
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cmpeq_epi8(__m256i __a, __m256i __b)
{
return (__m256i)((__v32qi)__a == (__v32qi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cmpeq_epi16(__m256i __a, __m256i __b)
{
return (__m256i)((__v16hi)__a == (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cmpeq_epi32(__m256i __a, __m256i __b)
{
return (__m256i)((__v8si)__a == (__v8si)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cmpeq_epi64(__m256i __a, __m256i __b)
{
return (__m256i)((__v4di)__a == (__v4di)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cmpgt_epi8(__m256i __a, __m256i __b)
{
/* This function always performs a signed comparison, but __v32qi is a char
return (__m256i)((__v32qs)__a > (__v32qs)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cmpgt_epi16(__m256i __a, __m256i __b)
{
return (__m256i)((__v16hi)__a > (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cmpgt_epi32(__m256i __a, __m256i __b)
{
return (__m256i)((__v8si)__a > (__v8si)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cmpgt_epi64(__m256i __a, __m256i __b)
{
return (__m256i)((__v4di)__a > (__v4di)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_hadd_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_phaddw256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_hadd_epi32(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_phaddd256((__v8si)__a, (__v8si)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_hadds_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_phaddsw256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_hsub_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_phsubw256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_hsub_epi32(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_phsubd256((__v8si)__a, (__v8si)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_hsubs_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_phsubsw256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maddubs_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pmaddubsw256((__v32qi)__a, (__v32qi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_madd_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pmaddwd256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_max_epi8(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pmaxsb256((__v32qi)__a, (__v32qi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_max_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pmaxsw256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_max_epi32(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pmaxsd256((__v8si)__a, (__v8si)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_max_epu8(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pmaxub256((__v32qi)__a, (__v32qi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_max_epu16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pmaxuw256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_max_epu32(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pmaxud256((__v8si)__a, (__v8si)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_min_epi8(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pminsb256((__v32qi)__a, (__v32qi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_min_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pminsw256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_min_epi32(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pminsd256((__v8si)__a, (__v8si)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_min_epu8(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pminub256((__v32qi)__a, (__v32qi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_min_epu16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pminuw256 ((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_min_epu32(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pminud256((__v8si)__a, (__v8si)__b);
}
-static __inline__ int __DEFAULT_FN_ATTRS
+static __inline__ int __DEFAULT_FN_ATTRS256
_mm256_movemask_epi8(__m256i __a)
{
return __builtin_ia32_pmovmskb256((__v32qi)__a);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtepi8_epi16(__m128i __V)
{
/* This function always performs a signed extension, but __v16qi is a char
return (__m256i)__builtin_convertvector((__v16qs)__V, __v16hi);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtepi8_epi32(__m128i __V)
{
/* This function always performs a signed extension, but __v16qi is a char
return (__m256i)__builtin_convertvector(__builtin_shufflevector((__v16qs)__V, (__v16qs)__V, 0, 1, 2, 3, 4, 5, 6, 7), __v8si);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtepi8_epi64(__m128i __V)
{
/* This function always performs a signed extension, but __v16qi is a char
return (__m256i)__builtin_convertvector(__builtin_shufflevector((__v16qs)__V, (__v16qs)__V, 0, 1, 2, 3), __v4di);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtepi16_epi32(__m128i __V)
{
return (__m256i)__builtin_convertvector((__v8hi)__V, __v8si);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtepi16_epi64(__m128i __V)
{
return (__m256i)__builtin_convertvector(__builtin_shufflevector((__v8hi)__V, (__v8hi)__V, 0, 1, 2, 3), __v4di);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtepi32_epi64(__m128i __V)
{
return (__m256i)__builtin_convertvector((__v4si)__V, __v4di);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtepu8_epi16(__m128i __V)
{
return (__m256i)__builtin_convertvector((__v16qu)__V, __v16hi);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtepu8_epi32(__m128i __V)
{
return (__m256i)__builtin_convertvector(__builtin_shufflevector((__v16qu)__V, (__v16qu)__V, 0, 1, 2, 3, 4, 5, 6, 7), __v8si);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtepu8_epi64(__m128i __V)
{
return (__m256i)__builtin_convertvector(__builtin_shufflevector((__v16qu)__V, (__v16qu)__V, 0, 1, 2, 3), __v4di);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtepu16_epi32(__m128i __V)
{
return (__m256i)__builtin_convertvector((__v8hu)__V, __v8si);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtepu16_epi64(__m128i __V)
{
return (__m256i)__builtin_convertvector(__builtin_shufflevector((__v8hu)__V, (__v8hu)__V, 0, 1, 2, 3), __v4di);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtepu32_epi64(__m128i __V)
{
return (__m256i)__builtin_convertvector((__v4su)__V, __v4di);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mul_epi32(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pmuldq256((__v8si)__a, (__v8si)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mulhrs_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pmulhrsw256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mulhi_epu16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pmulhuw256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mulhi_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pmulhw256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mullo_epi16(__m256i __a, __m256i __b)
{
return (__m256i)((__v16hu)__a * (__v16hu)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mullo_epi32 (__m256i __a, __m256i __b)
{
return (__m256i)((__v8su)__a * (__v8su)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mul_epu32(__m256i __a, __m256i __b)
{
return __builtin_ia32_pmuludq256((__v8si)__a, (__v8si)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_or_si256(__m256i __a, __m256i __b)
{
return (__m256i)((__v4du)__a | (__v4du)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sad_epu8(__m256i __a, __m256i __b)
{
return __builtin_ia32_psadbw256((__v32qi)__a, (__v32qi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_shuffle_epi8(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_pshufb256((__v32qi)__a, (__v32qi)__b);
}
-#define _mm256_shuffle_epi32(a, imm) __extension__ ({ \
- (__m256i)__builtin_shufflevector((__v8si)(__m256i)(a), \
- (__v8si)_mm256_undefined_si256(), \
- 0 + (((imm) >> 0) & 0x3), \
- 0 + (((imm) >> 2) & 0x3), \
- 0 + (((imm) >> 4) & 0x3), \
- 0 + (((imm) >> 6) & 0x3), \
- 4 + (((imm) >> 0) & 0x3), \
- 4 + (((imm) >> 2) & 0x3), \
- 4 + (((imm) >> 4) & 0x3), \
- 4 + (((imm) >> 6) & 0x3)); })
-
-#define _mm256_shufflehi_epi16(a, imm) __extension__ ({ \
- (__m256i)__builtin_shufflevector((__v16hi)(__m256i)(a), \
- (__v16hi)_mm256_undefined_si256(), \
- 0, 1, 2, 3, \
- 4 + (((imm) >> 0) & 0x3), \
- 4 + (((imm) >> 2) & 0x3), \
- 4 + (((imm) >> 4) & 0x3), \
- 4 + (((imm) >> 6) & 0x3), \
- 8, 9, 10, 11, \
- 12 + (((imm) >> 0) & 0x3), \
- 12 + (((imm) >> 2) & 0x3), \
- 12 + (((imm) >> 4) & 0x3), \
- 12 + (((imm) >> 6) & 0x3)); })
-
-#define _mm256_shufflelo_epi16(a, imm) __extension__ ({ \
- (__m256i)__builtin_shufflevector((__v16hi)(__m256i)(a), \
- (__v16hi)_mm256_undefined_si256(), \
- 0 + (((imm) >> 0) & 0x3), \
- 0 + (((imm) >> 2) & 0x3), \
- 0 + (((imm) >> 4) & 0x3), \
- 0 + (((imm) >> 6) & 0x3), \
- 4, 5, 6, 7, \
- 8 + (((imm) >> 0) & 0x3), \
- 8 + (((imm) >> 2) & 0x3), \
- 8 + (((imm) >> 4) & 0x3), \
- 8 + (((imm) >> 6) & 0x3), \
- 12, 13, 14, 15); })
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+#define _mm256_shuffle_epi32(a, imm) \
+ (__m256i)__builtin_ia32_pshufd256((__v8si)(__m256i)(a), (int)(imm))
+
+#define _mm256_shufflehi_epi16(a, imm) \
+ (__m256i)__builtin_ia32_pshufhw256((__v16hi)(__m256i)(a), (int)(imm))
+
+#define _mm256_shufflelo_epi16(a, imm) \
+ (__m256i)__builtin_ia32_pshuflw256((__v16hi)(__m256i)(a), (int)(imm))
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sign_epi8(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_psignb256((__v32qi)__a, (__v32qi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sign_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_psignw256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sign_epi32(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_psignd256((__v8si)__a, (__v8si)__b);
}
-#define _mm256_slli_si256(a, imm) __extension__ ({ \
- (__m256i)__builtin_shufflevector( \
- (__v32qi)_mm256_setzero_si256(), \
- (__v32qi)(__m256i)(a), \
- ((char)(imm)&0xF0) ? 0 : ((char)(imm)>0x0 ? 16 : 32) - (char)(imm), \
- ((char)(imm)&0xF0) ? 1 : ((char)(imm)>0x1 ? 17 : 33) - (char)(imm), \
- ((char)(imm)&0xF0) ? 2 : ((char)(imm)>0x2 ? 18 : 34) - (char)(imm), \
- ((char)(imm)&0xF0) ? 3 : ((char)(imm)>0x3 ? 19 : 35) - (char)(imm), \
- ((char)(imm)&0xF0) ? 4 : ((char)(imm)>0x4 ? 20 : 36) - (char)(imm), \
- ((char)(imm)&0xF0) ? 5 : ((char)(imm)>0x5 ? 21 : 37) - (char)(imm), \
- ((char)(imm)&0xF0) ? 6 : ((char)(imm)>0x6 ? 22 : 38) - (char)(imm), \
- ((char)(imm)&0xF0) ? 7 : ((char)(imm)>0x7 ? 23 : 39) - (char)(imm), \
- ((char)(imm)&0xF0) ? 8 : ((char)(imm)>0x8 ? 24 : 40) - (char)(imm), \
- ((char)(imm)&0xF0) ? 9 : ((char)(imm)>0x9 ? 25 : 41) - (char)(imm), \
- ((char)(imm)&0xF0) ? 10 : ((char)(imm)>0xA ? 26 : 42) - (char)(imm), \
- ((char)(imm)&0xF0) ? 11 : ((char)(imm)>0xB ? 27 : 43) - (char)(imm), \
- ((char)(imm)&0xF0) ? 12 : ((char)(imm)>0xC ? 28 : 44) - (char)(imm), \
- ((char)(imm)&0xF0) ? 13 : ((char)(imm)>0xD ? 29 : 45) - (char)(imm), \
- ((char)(imm)&0xF0) ? 14 : ((char)(imm)>0xE ? 30 : 46) - (char)(imm), \
- ((char)(imm)&0xF0) ? 15 : ((char)(imm)>0xF ? 31 : 47) - (char)(imm), \
- ((char)(imm)&0xF0) ? 16 : ((char)(imm)>0x0 ? 32 : 48) - (char)(imm), \
- ((char)(imm)&0xF0) ? 17 : ((char)(imm)>0x1 ? 33 : 49) - (char)(imm), \
- ((char)(imm)&0xF0) ? 18 : ((char)(imm)>0x2 ? 34 : 50) - (char)(imm), \
- ((char)(imm)&0xF0) ? 19 : ((char)(imm)>0x3 ? 35 : 51) - (char)(imm), \
- ((char)(imm)&0xF0) ? 20 : ((char)(imm)>0x4 ? 36 : 52) - (char)(imm), \
- ((char)(imm)&0xF0) ? 21 : ((char)(imm)>0x5 ? 37 : 53) - (char)(imm), \
- ((char)(imm)&0xF0) ? 22 : ((char)(imm)>0x6 ? 38 : 54) - (char)(imm), \
- ((char)(imm)&0xF0) ? 23 : ((char)(imm)>0x7 ? 39 : 55) - (char)(imm), \
- ((char)(imm)&0xF0) ? 24 : ((char)(imm)>0x8 ? 40 : 56) - (char)(imm), \
- ((char)(imm)&0xF0) ? 25 : ((char)(imm)>0x9 ? 41 : 57) - (char)(imm), \
- ((char)(imm)&0xF0) ? 26 : ((char)(imm)>0xA ? 42 : 58) - (char)(imm), \
- ((char)(imm)&0xF0) ? 27 : ((char)(imm)>0xB ? 43 : 59) - (char)(imm), \
- ((char)(imm)&0xF0) ? 28 : ((char)(imm)>0xC ? 44 : 60) - (char)(imm), \
- ((char)(imm)&0xF0) ? 29 : ((char)(imm)>0xD ? 45 : 61) - (char)(imm), \
- ((char)(imm)&0xF0) ? 30 : ((char)(imm)>0xE ? 46 : 62) - (char)(imm), \
- ((char)(imm)&0xF0) ? 31 : ((char)(imm)>0xF ? 47 : 63) - (char)(imm)); })
-
-#define _mm256_bslli_epi128(a, count) _mm256_slli_si256((a), (count))
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+#define _mm256_slli_si256(a, imm) \
+ (__m256i)__builtin_ia32_pslldqi256_byteshift((__v4di)(__m256i)(a), (int)(imm))
+
+#define _mm256_bslli_epi128(a, imm) \
+ (__m256i)__builtin_ia32_pslldqi256_byteshift((__v4di)(__m256i)(a), (int)(imm))
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_slli_epi16(__m256i __a, int __count)
{
return (__m256i)__builtin_ia32_psllwi256((__v16hi)__a, __count);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sll_epi16(__m256i __a, __m128i __count)
{
return (__m256i)__builtin_ia32_psllw256((__v16hi)__a, (__v8hi)__count);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_slli_epi32(__m256i __a, int __count)
{
return (__m256i)__builtin_ia32_pslldi256((__v8si)__a, __count);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sll_epi32(__m256i __a, __m128i __count)
{
return (__m256i)__builtin_ia32_pslld256((__v8si)__a, (__v4si)__count);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_slli_epi64(__m256i __a, int __count)
{
return __builtin_ia32_psllqi256((__v4di)__a, __count);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sll_epi64(__m256i __a, __m128i __count)
{
return __builtin_ia32_psllq256((__v4di)__a, __count);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_srai_epi16(__m256i __a, int __count)
{
return (__m256i)__builtin_ia32_psrawi256((__v16hi)__a, __count);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sra_epi16(__m256i __a, __m128i __count)
{
return (__m256i)__builtin_ia32_psraw256((__v16hi)__a, (__v8hi)__count);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_srai_epi32(__m256i __a, int __count)
{
return (__m256i)__builtin_ia32_psradi256((__v8si)__a, __count);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sra_epi32(__m256i __a, __m128i __count)
{
return (__m256i)__builtin_ia32_psrad256((__v8si)__a, (__v4si)__count);
}
-#define _mm256_srli_si256(a, imm) __extension__ ({ \
- (__m256i)__builtin_shufflevector( \
- (__v32qi)(__m256i)(a), \
- (__v32qi)_mm256_setzero_si256(), \
- ((char)(imm)&0xF0) ? 32 : (char)(imm) + ((char)(imm)>0xF ? 16 : 0), \
- ((char)(imm)&0xF0) ? 33 : (char)(imm) + ((char)(imm)>0xE ? 17 : 1), \
- ((char)(imm)&0xF0) ? 34 : (char)(imm) + ((char)(imm)>0xD ? 18 : 2), \
- ((char)(imm)&0xF0) ? 35 : (char)(imm) + ((char)(imm)>0xC ? 19 : 3), \
- ((char)(imm)&0xF0) ? 36 : (char)(imm) + ((char)(imm)>0xB ? 20 : 4), \
- ((char)(imm)&0xF0) ? 37 : (char)(imm) + ((char)(imm)>0xA ? 21 : 5), \
- ((char)(imm)&0xF0) ? 38 : (char)(imm) + ((char)(imm)>0x9 ? 22 : 6), \
- ((char)(imm)&0xF0) ? 39 : (char)(imm) + ((char)(imm)>0x8 ? 23 : 7), \
- ((char)(imm)&0xF0) ? 40 : (char)(imm) + ((char)(imm)>0x7 ? 24 : 8), \
- ((char)(imm)&0xF0) ? 41 : (char)(imm) + ((char)(imm)>0x6 ? 25 : 9), \
- ((char)(imm)&0xF0) ? 42 : (char)(imm) + ((char)(imm)>0x5 ? 26 : 10), \
- ((char)(imm)&0xF0) ? 43 : (char)(imm) + ((char)(imm)>0x4 ? 27 : 11), \
- ((char)(imm)&0xF0) ? 44 : (char)(imm) + ((char)(imm)>0x3 ? 28 : 12), \
- ((char)(imm)&0xF0) ? 45 : (char)(imm) + ((char)(imm)>0x2 ? 29 : 13), \
- ((char)(imm)&0xF0) ? 46 : (char)(imm) + ((char)(imm)>0x1 ? 30 : 14), \
- ((char)(imm)&0xF0) ? 47 : (char)(imm) + ((char)(imm)>0x0 ? 31 : 15), \
- ((char)(imm)&0xF0) ? 48 : (char)(imm) + ((char)(imm)>0xF ? 32 : 16), \
- ((char)(imm)&0xF0) ? 49 : (char)(imm) + ((char)(imm)>0xE ? 33 : 17), \
- ((char)(imm)&0xF0) ? 50 : (char)(imm) + ((char)(imm)>0xD ? 34 : 18), \
- ((char)(imm)&0xF0) ? 51 : (char)(imm) + ((char)(imm)>0xC ? 35 : 19), \
- ((char)(imm)&0xF0) ? 52 : (char)(imm) + ((char)(imm)>0xB ? 36 : 20), \
- ((char)(imm)&0xF0) ? 53 : (char)(imm) + ((char)(imm)>0xA ? 37 : 21), \
- ((char)(imm)&0xF0) ? 54 : (char)(imm) + ((char)(imm)>0x9 ? 38 : 22), \
- ((char)(imm)&0xF0) ? 55 : (char)(imm) + ((char)(imm)>0x8 ? 39 : 23), \
- ((char)(imm)&0xF0) ? 56 : (char)(imm) + ((char)(imm)>0x7 ? 40 : 24), \
- ((char)(imm)&0xF0) ? 57 : (char)(imm) + ((char)(imm)>0x6 ? 41 : 25), \
- ((char)(imm)&0xF0) ? 58 : (char)(imm) + ((char)(imm)>0x5 ? 42 : 26), \
- ((char)(imm)&0xF0) ? 59 : (char)(imm) + ((char)(imm)>0x4 ? 43 : 27), \
- ((char)(imm)&0xF0) ? 60 : (char)(imm) + ((char)(imm)>0x3 ? 44 : 28), \
- ((char)(imm)&0xF0) ? 61 : (char)(imm) + ((char)(imm)>0x2 ? 45 : 29), \
- ((char)(imm)&0xF0) ? 62 : (char)(imm) + ((char)(imm)>0x1 ? 46 : 30), \
- ((char)(imm)&0xF0) ? 63 : (char)(imm) + ((char)(imm)>0x0 ? 47 : 31)); })
-
-#define _mm256_bsrli_epi128(a, count) _mm256_srli_si256((a), (count))
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+#define _mm256_srli_si256(a, imm) \
+ (__m256i)__builtin_ia32_psrldqi256_byteshift((__m256i)(a), (int)(imm))
+
+#define _mm256_bsrli_epi128(a, imm) \
+ (__m256i)__builtin_ia32_psrldqi256_byteshift((__m256i)(a), (int)(imm))
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_srli_epi16(__m256i __a, int __count)
{
return (__m256i)__builtin_ia32_psrlwi256((__v16hi)__a, __count);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_srl_epi16(__m256i __a, __m128i __count)
{
return (__m256i)__builtin_ia32_psrlw256((__v16hi)__a, (__v8hi)__count);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_srli_epi32(__m256i __a, int __count)
{
return (__m256i)__builtin_ia32_psrldi256((__v8si)__a, __count);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_srl_epi32(__m256i __a, __m128i __count)
{
return (__m256i)__builtin_ia32_psrld256((__v8si)__a, (__v4si)__count);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_srli_epi64(__m256i __a, int __count)
{
return __builtin_ia32_psrlqi256((__v4di)__a, __count);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_srl_epi64(__m256i __a, __m128i __count)
{
return __builtin_ia32_psrlq256((__v4di)__a, __count);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sub_epi8(__m256i __a, __m256i __b)
{
return (__m256i)((__v32qu)__a - (__v32qu)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sub_epi16(__m256i __a, __m256i __b)
{
return (__m256i)((__v16hu)__a - (__v16hu)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sub_epi32(__m256i __a, __m256i __b)
{
return (__m256i)((__v8su)__a - (__v8su)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sub_epi64(__m256i __a, __m256i __b)
{
return (__m256i)((__v4du)__a - (__v4du)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_subs_epi8(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_psubsb256((__v32qi)__a, (__v32qi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_subs_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_psubsw256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_subs_epu8(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_psubusb256((__v32qi)__a, (__v32qi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_subs_epu16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_psubusw256((__v16hi)__a, (__v16hi)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_unpackhi_epi8(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_shufflevector((__v32qi)__a, (__v32qi)__b, 8, 32+8, 9, 32+9, 10, 32+10, 11, 32+11, 12, 32+12, 13, 32+13, 14, 32+14, 15, 32+15, 24, 32+24, 25, 32+25, 26, 32+26, 27, 32+27, 28, 32+28, 29, 32+29, 30, 32+30, 31, 32+31);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_unpackhi_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_shufflevector((__v16hi)__a, (__v16hi)__b, 4, 16+4, 5, 16+5, 6, 16+6, 7, 16+7, 12, 16+12, 13, 16+13, 14, 16+14, 15, 16+15);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_unpackhi_epi32(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_shufflevector((__v8si)__a, (__v8si)__b, 2, 8+2, 3, 8+3, 6, 8+6, 7, 8+7);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_unpackhi_epi64(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_shufflevector((__v4di)__a, (__v4di)__b, 1, 4+1, 3, 4+3);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_unpacklo_epi8(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_shufflevector((__v32qi)__a, (__v32qi)__b, 0, 32+0, 1, 32+1, 2, 32+2, 3, 32+3, 4, 32+4, 5, 32+5, 6, 32+6, 7, 32+7, 16, 32+16, 17, 32+17, 18, 32+18, 19, 32+19, 20, 32+20, 21, 32+21, 22, 32+22, 23, 32+23);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_unpacklo_epi16(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_shufflevector((__v16hi)__a, (__v16hi)__b, 0, 16+0, 1, 16+1, 2, 16+2, 3, 16+3, 8, 16+8, 9, 16+9, 10, 16+10, 11, 16+11);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_unpacklo_epi32(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_shufflevector((__v8si)__a, (__v8si)__b, 0, 8+0, 1, 8+1, 4, 8+4, 5, 8+5);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_unpacklo_epi64(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_shufflevector((__v4di)__a, (__v4di)__b, 0, 4+0, 2, 4+2);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_xor_si256(__m256i __a, __m256i __b)
{
return (__m256i)((__v4du)__a ^ (__v4du)__b);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_stream_load_si256(__m256i const *__V)
{
typedef __v4di __v4di_aligned __attribute__((aligned(32)));
return (__m256i)__builtin_nontemporal_load((const __v4di_aligned *)__V);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_broadcastss_ps(__m128 __X)
{
return (__m128)__builtin_shufflevector((__v4sf)__X, (__v4sf)__X, 0, 0, 0, 0);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_broadcastsd_pd(__m128d __a)
{
return __builtin_shufflevector((__v2df)__a, (__v2df)__a, 0, 0);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_broadcastss_ps(__m128 __X)
{
return (__m256)__builtin_shufflevector((__v4sf)__X, (__v4sf)__X, 0, 0, 0, 0, 0, 0, 0, 0);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_broadcastsd_pd(__m128d __X)
{
return (__m256d)__builtin_shufflevector((__v2df)__X, (__v2df)__X, 0, 0, 0, 0);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_broadcastsi128_si256(__m128i __X)
{
return (__m256i)__builtin_shufflevector((__v2di)__X, (__v2di)__X, 0, 1, 0, 1);
}
-#define _mm_blend_epi32(V1, V2, M) __extension__ ({ \
- (__m128i)__builtin_shufflevector((__v4si)(__m128i)(V1), \
- (__v4si)(__m128i)(V2), \
- (((M) & 0x01) ? 4 : 0), \
- (((M) & 0x02) ? 5 : 1), \
- (((M) & 0x04) ? 6 : 2), \
- (((M) & 0x08) ? 7 : 3)); })
-
-#define _mm256_blend_epi32(V1, V2, M) __extension__ ({ \
- (__m256i)__builtin_shufflevector((__v8si)(__m256i)(V1), \
- (__v8si)(__m256i)(V2), \
- (((M) & 0x01) ? 8 : 0), \
- (((M) & 0x02) ? 9 : 1), \
- (((M) & 0x04) ? 10 : 2), \
- (((M) & 0x08) ? 11 : 3), \
- (((M) & 0x10) ? 12 : 4), \
- (((M) & 0x20) ? 13 : 5), \
- (((M) & 0x40) ? 14 : 6), \
- (((M) & 0x80) ? 15 : 7)); })
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+#define _mm_broadcastsi128_si256(X) _mm256_broadcastsi128_si256(X)
+
+#define _mm_blend_epi32(V1, V2, M) \
+ (__m128i)__builtin_ia32_pblendd128((__v4si)(__m128i)(V1), \
+ (__v4si)(__m128i)(V2), (int)(M))
+
+#define _mm256_blend_epi32(V1, V2, M) \
+ (__m256i)__builtin_ia32_pblendd256((__v8si)(__m256i)(V1), \
+ (__v8si)(__m256i)(V2), (int)(M))
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_broadcastb_epi8(__m128i __X)
{
return (__m256i)__builtin_shufflevector((__v16qi)__X, (__v16qi)__X, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_broadcastw_epi16(__m128i __X)
{
return (__m256i)__builtin_shufflevector((__v8hi)__X, (__v8hi)__X, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_broadcastd_epi32(__m128i __X)
{
return (__m256i)__builtin_shufflevector((__v4si)__X, (__v4si)__X, 0, 0, 0, 0, 0, 0, 0, 0);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_broadcastq_epi64(__m128i __X)
{
return (__m256i)__builtin_shufflevector((__v2di)__X, (__v2di)__X, 0, 0, 0, 0);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_broadcastb_epi8(__m128i __X)
{
return (__m128i)__builtin_shufflevector((__v16qi)__X, (__v16qi)__X, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_broadcastw_epi16(__m128i __X)
{
return (__m128i)__builtin_shufflevector((__v8hi)__X, (__v8hi)__X, 0, 0, 0, 0, 0, 0, 0, 0);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_broadcastd_epi32(__m128i __X)
{
return (__m128i)__builtin_shufflevector((__v4si)__X, (__v4si)__X, 0, 0, 0, 0);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_broadcastq_epi64(__m128i __X)
{
return (__m128i)__builtin_shufflevector((__v2di)__X, (__v2di)__X, 0, 0);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_permutevar8x32_epi32(__m256i __a, __m256i __b)
{
return (__m256i)__builtin_ia32_permvarsi256((__v8si)__a, (__v8si)__b);
}
-#define _mm256_permute4x64_pd(V, M) __extension__ ({ \
- (__m256d)__builtin_shufflevector((__v4df)(__m256d)(V), \
- (__v4df)_mm256_undefined_pd(), \
- ((M) >> 0) & 0x3, \
- ((M) >> 2) & 0x3, \
- ((M) >> 4) & 0x3, \
- ((M) >> 6) & 0x3); })
+#define _mm256_permute4x64_pd(V, M) \
+ (__m256d)__builtin_ia32_permdf256((__v4df)(__m256d)(V), (int)(M))
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_permutevar8x32_ps(__m256 __a, __m256i __b)
{
return (__m256)__builtin_ia32_permvarsf256((__v8sf)__a, (__v8si)__b);
}
-#define _mm256_permute4x64_epi64(V, M) __extension__ ({ \
- (__m256i)__builtin_shufflevector((__v4di)(__m256i)(V), \
- (__v4di)_mm256_undefined_si256(), \
- ((M) >> 0) & 0x3, \
- ((M) >> 2) & 0x3, \
- ((M) >> 4) & 0x3, \
- ((M) >> 6) & 0x3); })
-
-#define _mm256_permute2x128_si256(V1, V2, M) __extension__ ({ \
- (__m256i)__builtin_ia32_permti256((__m256i)(V1), (__m256i)(V2), (M)); })
-
-#define _mm256_extracti128_si256(V, M) __extension__ ({ \
- (__m128i)__builtin_shufflevector((__v4di)(__m256i)(V), \
- (__v4di)_mm256_undefined_si256(), \
- (((M) & 1) ? 2 : 0), \
- (((M) & 1) ? 3 : 1) ); })
-
-#define _mm256_inserti128_si256(V1, V2, M) __extension__ ({ \
- (__m256i)__builtin_shufflevector((__v4di)(__m256i)(V1), \
- (__v4di)_mm256_castsi128_si256((__m128i)(V2)), \
- (((M) & 1) ? 0 : 4), \
- (((M) & 1) ? 1 : 5), \
- (((M) & 1) ? 4 : 2), \
- (((M) & 1) ? 5 : 3) ); })
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+#define _mm256_permute4x64_epi64(V, M) \
+ (__m256i)__builtin_ia32_permdi256((__v4di)(__m256i)(V), (int)(M))
+
+#define _mm256_permute2x128_si256(V1, V2, M) \
+ (__m256i)__builtin_ia32_permti256((__m256i)(V1), (__m256i)(V2), (int)(M))
+
+#define _mm256_extracti128_si256(V, M) \
+ (__m128i)__builtin_ia32_extract128i256((__v4di)(__m256i)(V), (int)(M))
+
+#define _mm256_inserti128_si256(V1, V2, M) \
+ (__m256i)__builtin_ia32_insert128i256((__v4di)(__m256i)(V1), \
+ (__v2di)(__m128i)(V2), (int)(M))
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskload_epi32(int const *__X, __m256i __M)
{
return (__m256i)__builtin_ia32_maskloadd256((const __v8si *)__X, (__v8si)__M);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskload_epi64(long long const *__X, __m256i __M)
{
return (__m256i)__builtin_ia32_maskloadq256((const __v4di *)__X, (__v4di)__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskload_epi32(int const *__X, __m128i __M)
{
return (__m128i)__builtin_ia32_maskloadd((const __v4si *)__X, (__v4si)__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskload_epi64(long long const *__X, __m128i __M)
{
return (__m128i)__builtin_ia32_maskloadq((const __v2di *)__X, (__v2di)__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_maskstore_epi32(int *__X, __m256i __M, __m256i __Y)
{
__builtin_ia32_maskstored256((__v8si *)__X, (__v8si)__M, (__v8si)__Y);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_maskstore_epi64(long long *__X, __m256i __M, __m256i __Y)
{
__builtin_ia32_maskstoreq256((__v4di *)__X, (__v4di)__M, (__v4di)__Y);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_maskstore_epi32(int *__X, __m128i __M, __m128i __Y)
{
__builtin_ia32_maskstored((__v4si *)__X, (__v4si)__M, (__v4si)__Y);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_maskstore_epi64(long long *__X, __m128i __M, __m128i __Y)
{
__builtin_ia32_maskstoreq(( __v2di *)__X, (__v2di)__M, (__v2di)__Y);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sllv_epi32(__m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_psllv8si((__v8si)__X, (__v8si)__Y);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_sllv_epi32(__m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_psllv4si((__v4si)__X, (__v4si)__Y);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sllv_epi64(__m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_psllv4di((__v4di)__X, (__v4di)__Y);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_sllv_epi64(__m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_psllv2di((__v2di)__X, (__v2di)__Y);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_srav_epi32(__m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_psrav8si((__v8si)__X, (__v8si)__Y);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_srav_epi32(__m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_psrav4si((__v4si)__X, (__v4si)__Y);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_srlv_epi32(__m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_psrlv8si((__v8si)__X, (__v8si)__Y);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_srlv_epi32(__m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_psrlv4si((__v4si)__X, (__v4si)__Y);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_srlv_epi64(__m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_psrlv4di((__v4di)__X, (__v4di)__Y);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_srlv_epi64(__m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_psrlv2di((__v2di)__X, (__v2di)__Y);
}
-#define _mm_mask_i32gather_pd(a, m, i, mask, s) __extension__ ({ \
+#define _mm_mask_i32gather_pd(a, m, i, mask, s) \
(__m128d)__builtin_ia32_gatherd_pd((__v2df)(__m128i)(a), \
(double const *)(m), \
(__v4si)(__m128i)(i), \
- (__v2df)(__m128d)(mask), (s)); })
+ (__v2df)(__m128d)(mask), (s))
-#define _mm256_mask_i32gather_pd(a, m, i, mask, s) __extension__ ({ \
+#define _mm256_mask_i32gather_pd(a, m, i, mask, s) \
(__m256d)__builtin_ia32_gatherd_pd256((__v4df)(__m256d)(a), \
(double const *)(m), \
(__v4si)(__m128i)(i), \
- (__v4df)(__m256d)(mask), (s)); })
+ (__v4df)(__m256d)(mask), (s))
-#define _mm_mask_i64gather_pd(a, m, i, mask, s) __extension__ ({ \
+#define _mm_mask_i64gather_pd(a, m, i, mask, s) \
(__m128d)__builtin_ia32_gatherq_pd((__v2df)(__m128d)(a), \
(double const *)(m), \
(__v2di)(__m128i)(i), \
- (__v2df)(__m128d)(mask), (s)); })
+ (__v2df)(__m128d)(mask), (s))
-#define _mm256_mask_i64gather_pd(a, m, i, mask, s) __extension__ ({ \
+#define _mm256_mask_i64gather_pd(a, m, i, mask, s) \
(__m256d)__builtin_ia32_gatherq_pd256((__v4df)(__m256d)(a), \
(double const *)(m), \
(__v4di)(__m256i)(i), \
- (__v4df)(__m256d)(mask), (s)); })
+ (__v4df)(__m256d)(mask), (s))
-#define _mm_mask_i32gather_ps(a, m, i, mask, s) __extension__ ({ \
+#define _mm_mask_i32gather_ps(a, m, i, mask, s) \
(__m128)__builtin_ia32_gatherd_ps((__v4sf)(__m128)(a), \
(float const *)(m), \
(__v4si)(__m128i)(i), \
- (__v4sf)(__m128)(mask), (s)); })
+ (__v4sf)(__m128)(mask), (s))
-#define _mm256_mask_i32gather_ps(a, m, i, mask, s) __extension__ ({ \
+#define _mm256_mask_i32gather_ps(a, m, i, mask, s) \
(__m256)__builtin_ia32_gatherd_ps256((__v8sf)(__m256)(a), \
(float const *)(m), \
(__v8si)(__m256i)(i), \
- (__v8sf)(__m256)(mask), (s)); })
+ (__v8sf)(__m256)(mask), (s))
-#define _mm_mask_i64gather_ps(a, m, i, mask, s) __extension__ ({ \
+#define _mm_mask_i64gather_ps(a, m, i, mask, s) \
(__m128)__builtin_ia32_gatherq_ps((__v4sf)(__m128)(a), \
(float const *)(m), \
(__v2di)(__m128i)(i), \
- (__v4sf)(__m128)(mask), (s)); })
+ (__v4sf)(__m128)(mask), (s))
-#define _mm256_mask_i64gather_ps(a, m, i, mask, s) __extension__ ({ \
+#define _mm256_mask_i64gather_ps(a, m, i, mask, s) \
(__m128)__builtin_ia32_gatherq_ps256((__v4sf)(__m128)(a), \
(float const *)(m), \
(__v4di)(__m256i)(i), \
- (__v4sf)(__m128)(mask), (s)); })
+ (__v4sf)(__m128)(mask), (s))
-#define _mm_mask_i32gather_epi32(a, m, i, mask, s) __extension__ ({ \
+#define _mm_mask_i32gather_epi32(a, m, i, mask, s) \
(__m128i)__builtin_ia32_gatherd_d((__v4si)(__m128i)(a), \
(int const *)(m), \
(__v4si)(__m128i)(i), \
- (__v4si)(__m128i)(mask), (s)); })
+ (__v4si)(__m128i)(mask), (s))
-#define _mm256_mask_i32gather_epi32(a, m, i, mask, s) __extension__ ({ \
+#define _mm256_mask_i32gather_epi32(a, m, i, mask, s) \
(__m256i)__builtin_ia32_gatherd_d256((__v8si)(__m256i)(a), \
(int const *)(m), \
(__v8si)(__m256i)(i), \
- (__v8si)(__m256i)(mask), (s)); })
+ (__v8si)(__m256i)(mask), (s))
-#define _mm_mask_i64gather_epi32(a, m, i, mask, s) __extension__ ({ \
+#define _mm_mask_i64gather_epi32(a, m, i, mask, s) \
(__m128i)__builtin_ia32_gatherq_d((__v4si)(__m128i)(a), \
(int const *)(m), \
(__v2di)(__m128i)(i), \
- (__v4si)(__m128i)(mask), (s)); })
+ (__v4si)(__m128i)(mask), (s))
-#define _mm256_mask_i64gather_epi32(a, m, i, mask, s) __extension__ ({ \
+#define _mm256_mask_i64gather_epi32(a, m, i, mask, s) \
(__m128i)__builtin_ia32_gatherq_d256((__v4si)(__m128i)(a), \
(int const *)(m), \
(__v4di)(__m256i)(i), \
- (__v4si)(__m128i)(mask), (s)); })
+ (__v4si)(__m128i)(mask), (s))
-#define _mm_mask_i32gather_epi64(a, m, i, mask, s) __extension__ ({ \
+#define _mm_mask_i32gather_epi64(a, m, i, mask, s) \
(__m128i)__builtin_ia32_gatherd_q((__v2di)(__m128i)(a), \
(long long const *)(m), \
(__v4si)(__m128i)(i), \
- (__v2di)(__m128i)(mask), (s)); })
+ (__v2di)(__m128i)(mask), (s))
-#define _mm256_mask_i32gather_epi64(a, m, i, mask, s) __extension__ ({ \
+#define _mm256_mask_i32gather_epi64(a, m, i, mask, s) \
(__m256i)__builtin_ia32_gatherd_q256((__v4di)(__m256i)(a), \
(long long const *)(m), \
(__v4si)(__m128i)(i), \
- (__v4di)(__m256i)(mask), (s)); })
+ (__v4di)(__m256i)(mask), (s))
-#define _mm_mask_i64gather_epi64(a, m, i, mask, s) __extension__ ({ \
+#define _mm_mask_i64gather_epi64(a, m, i, mask, s) \
(__m128i)__builtin_ia32_gatherq_q((__v2di)(__m128i)(a), \
(long long const *)(m), \
(__v2di)(__m128i)(i), \
- (__v2di)(__m128i)(mask), (s)); })
+ (__v2di)(__m128i)(mask), (s))
-#define _mm256_mask_i64gather_epi64(a, m, i, mask, s) __extension__ ({ \
+#define _mm256_mask_i64gather_epi64(a, m, i, mask, s) \
(__m256i)__builtin_ia32_gatherq_q256((__v4di)(__m256i)(a), \
(long long const *)(m), \
(__v4di)(__m256i)(i), \
- (__v4di)(__m256i)(mask), (s)); })
+ (__v4di)(__m256i)(mask), (s))
-#define _mm_i32gather_pd(m, i, s) __extension__ ({ \
+#define _mm_i32gather_pd(m, i, s) \
(__m128d)__builtin_ia32_gatherd_pd((__v2df)_mm_undefined_pd(), \
(double const *)(m), \
(__v4si)(__m128i)(i), \
(__v2df)_mm_cmpeq_pd(_mm_setzero_pd(), \
_mm_setzero_pd()), \
- (s)); })
+ (s))
-#define _mm256_i32gather_pd(m, i, s) __extension__ ({ \
+#define _mm256_i32gather_pd(m, i, s) \
(__m256d)__builtin_ia32_gatherd_pd256((__v4df)_mm256_undefined_pd(), \
(double const *)(m), \
(__v4si)(__m128i)(i), \
(__v4df)_mm256_cmp_pd(_mm256_setzero_pd(), \
_mm256_setzero_pd(), \
_CMP_EQ_OQ), \
- (s)); })
+ (s))
-#define _mm_i64gather_pd(m, i, s) __extension__ ({ \
+#define _mm_i64gather_pd(m, i, s) \
(__m128d)__builtin_ia32_gatherq_pd((__v2df)_mm_undefined_pd(), \
(double const *)(m), \
(__v2di)(__m128i)(i), \
(__v2df)_mm_cmpeq_pd(_mm_setzero_pd(), \
_mm_setzero_pd()), \
- (s)); })
+ (s))
-#define _mm256_i64gather_pd(m, i, s) __extension__ ({ \
+#define _mm256_i64gather_pd(m, i, s) \
(__m256d)__builtin_ia32_gatherq_pd256((__v4df)_mm256_undefined_pd(), \
(double const *)(m), \
(__v4di)(__m256i)(i), \
(__v4df)_mm256_cmp_pd(_mm256_setzero_pd(), \
_mm256_setzero_pd(), \
_CMP_EQ_OQ), \
- (s)); })
+ (s))
-#define _mm_i32gather_ps(m, i, s) __extension__ ({ \
+#define _mm_i32gather_ps(m, i, s) \
(__m128)__builtin_ia32_gatherd_ps((__v4sf)_mm_undefined_ps(), \
(float const *)(m), \
(__v4si)(__m128i)(i), \
(__v4sf)_mm_cmpeq_ps(_mm_setzero_ps(), \
_mm_setzero_ps()), \
- (s)); })
+ (s))
-#define _mm256_i32gather_ps(m, i, s) __extension__ ({ \
+#define _mm256_i32gather_ps(m, i, s) \
(__m256)__builtin_ia32_gatherd_ps256((__v8sf)_mm256_undefined_ps(), \
(float const *)(m), \
(__v8si)(__m256i)(i), \
(__v8sf)_mm256_cmp_ps(_mm256_setzero_ps(), \
_mm256_setzero_ps(), \
_CMP_EQ_OQ), \
- (s)); })
+ (s))
-#define _mm_i64gather_ps(m, i, s) __extension__ ({ \
+#define _mm_i64gather_ps(m, i, s) \
(__m128)__builtin_ia32_gatherq_ps((__v4sf)_mm_undefined_ps(), \
(float const *)(m), \
(__v2di)(__m128i)(i), \
(__v4sf)_mm_cmpeq_ps(_mm_setzero_ps(), \
_mm_setzero_ps()), \
- (s)); })
+ (s))
-#define _mm256_i64gather_ps(m, i, s) __extension__ ({ \
+#define _mm256_i64gather_ps(m, i, s) \
(__m128)__builtin_ia32_gatherq_ps256((__v4sf)_mm_undefined_ps(), \
(float const *)(m), \
(__v4di)(__m256i)(i), \
(__v4sf)_mm_cmpeq_ps(_mm_setzero_ps(), \
_mm_setzero_ps()), \
- (s)); })
+ (s))
-#define _mm_i32gather_epi32(m, i, s) __extension__ ({ \
+#define _mm_i32gather_epi32(m, i, s) \
(__m128i)__builtin_ia32_gatherd_d((__v4si)_mm_undefined_si128(), \
(int const *)(m), (__v4si)(__m128i)(i), \
- (__v4si)_mm_set1_epi32(-1), (s)); })
+ (__v4si)_mm_set1_epi32(-1), (s))
-#define _mm256_i32gather_epi32(m, i, s) __extension__ ({ \
+#define _mm256_i32gather_epi32(m, i, s) \
(__m256i)__builtin_ia32_gatherd_d256((__v8si)_mm256_undefined_si256(), \
(int const *)(m), (__v8si)(__m256i)(i), \
- (__v8si)_mm256_set1_epi32(-1), (s)); })
+ (__v8si)_mm256_set1_epi32(-1), (s))
-#define _mm_i64gather_epi32(m, i, s) __extension__ ({ \
+#define _mm_i64gather_epi32(m, i, s) \
(__m128i)__builtin_ia32_gatherq_d((__v4si)_mm_undefined_si128(), \
(int const *)(m), (__v2di)(__m128i)(i), \
- (__v4si)_mm_set1_epi32(-1), (s)); })
+ (__v4si)_mm_set1_epi32(-1), (s))
-#define _mm256_i64gather_epi32(m, i, s) __extension__ ({ \
+#define _mm256_i64gather_epi32(m, i, s) \
(__m128i)__builtin_ia32_gatherq_d256((__v4si)_mm_undefined_si128(), \
(int const *)(m), (__v4di)(__m256i)(i), \
- (__v4si)_mm_set1_epi32(-1), (s)); })
+ (__v4si)_mm_set1_epi32(-1), (s))
-#define _mm_i32gather_epi64(m, i, s) __extension__ ({ \
+#define _mm_i32gather_epi64(m, i, s) \
(__m128i)__builtin_ia32_gatherd_q((__v2di)_mm_undefined_si128(), \
(long long const *)(m), \
(__v4si)(__m128i)(i), \
- (__v2di)_mm_set1_epi64x(-1), (s)); })
+ (__v2di)_mm_set1_epi64x(-1), (s))
-#define _mm256_i32gather_epi64(m, i, s) __extension__ ({ \
+#define _mm256_i32gather_epi64(m, i, s) \
(__m256i)__builtin_ia32_gatherd_q256((__v4di)_mm256_undefined_si256(), \
(long long const *)(m), \
(__v4si)(__m128i)(i), \
- (__v4di)_mm256_set1_epi64x(-1), (s)); })
+ (__v4di)_mm256_set1_epi64x(-1), (s))
-#define _mm_i64gather_epi64(m, i, s) __extension__ ({ \
+#define _mm_i64gather_epi64(m, i, s) \
(__m128i)__builtin_ia32_gatherq_q((__v2di)_mm_undefined_si128(), \
(long long const *)(m), \
(__v2di)(__m128i)(i), \
- (__v2di)_mm_set1_epi64x(-1), (s)); })
+ (__v2di)_mm_set1_epi64x(-1), (s))
-#define _mm256_i64gather_epi64(m, i, s) __extension__ ({ \
+#define _mm256_i64gather_epi64(m, i, s) \
(__m256i)__builtin_ia32_gatherq_q256((__v4di)_mm256_undefined_si256(), \
(long long const *)(m), \
(__v4di)(__m256i)(i), \
- (__v4di)_mm256_set1_epi64x(-1), (s)); })
+ (__v4di)_mm256_set1_epi64x(-1), (s))
-#undef __DEFAULT_FN_ATTRS
+#undef __DEFAULT_FN_ATTRS256
+#undef __DEFAULT_FN_ATTRS128
#endif /* __AVX2INTRIN_H */
--- /dev/null
+/*===------------ avx512bf16intrin.h - AVX512_BF16 intrinsics --------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512bf16intrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __AVX512BF16INTRIN_H
+#define __AVX512BF16INTRIN_H
+
+typedef short __m512bh __attribute__((__vector_size__(64), __aligned__(64)));
+typedef short __m256bh __attribute__((__vector_size__(32), __aligned__(32)));
+typedef unsigned short __bfloat16;
+
+#define __DEFAULT_FN_ATTRS512 \
+ __attribute__((__always_inline__, __nodebug__, __target__("avx512bf16"), \
+ __min_vector_width__(512)))
+#define __DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("avx512bf16")))
+
+/// Convert One BF16 Data to One Single Float Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic does not correspond to a specific instruction.
+///
+/// \param __A
+/// A bfloat data.
+/// \returns A float data whose sign field and exponent field keep unchanged,
+/// and fraction field is extended to 23 bits.
+static __inline__ float __DEFAULT_FN_ATTRS _mm_cvtsbh_ss(__bfloat16 __A) {
+ return __builtin_ia32_cvtsbf162ss_32(__A);
+}
+
+/// Convert Two Packed Single Data to One Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNE2PS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 512-bit vector of [16 x float].
+/// \param __B
+/// A 512-bit vector of [16 x float].
+/// \returns A 512-bit vector of [32 x bfloat] whose lower 256 bits come from
+/// conversion of __B, and higher 256 bits come from conversion of __A.
+static __inline__ __m512bh __DEFAULT_FN_ATTRS512
+_mm512_cvtne2ps_pbh(__m512 __A, __m512 __B) {
+ return (__m512bh)__builtin_ia32_cvtne2ps2bf16_512((__v16sf) __A,
+ (__v16sf) __B);
+}
+
+/// Convert Two Packed Single Data to One Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNE2PS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 512-bit vector of [16 x float].
+/// \param __B
+/// A 512-bit vector of [16 x float].
+/// \param __W
+/// A 512-bit vector of [32 x bfloat].
+/// \param __U
+/// A 32-bit mask value specifying what is chosen for each element.
+/// A 1 means conversion of __A or __B. A 0 means element from __W.
+/// \returns A 512-bit vector of [32 x bfloat] whose lower 256 bits come from
+/// conversion of __B, and higher 256 bits come from conversion of __A.
+static __inline__ __m512bh __DEFAULT_FN_ATTRS512
+_mm512_mask_cvtne2ps_pbh(__m512bh __W, __mmask32 __U, __m512 __A, __m512 __B) {
+ return (__m512bh)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_cvtne2ps_pbh(__A, __B),
+ (__v32hi)__W);
+}
+
+/// Convert Two Packed Single Data to One Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNE2PS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 512-bit vector of [16 x float].
+/// \param __B
+/// A 512-bit vector of [16 x float].
+/// \param __U
+/// A 32-bit mask value specifying what is chosen for each element.
+/// A 1 means conversion of __A or __B. A 0 means element is zero.
+/// \returns A 512-bit vector of [32 x bfloat] whose lower 256 bits come from
+/// conversion of __B, and higher 256 bits come from conversion of __A.
+static __inline__ __m512bh __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvtne2ps_pbh(__mmask32 __U, __m512 __A, __m512 __B) {
+ return (__m512bh)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_cvtne2ps_pbh(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+/// Convert Packed Single Data to Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNEPS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 512-bit vector of [16 x float].
+/// \returns A 256-bit vector of [16 x bfloat] come from conversion of __A.
+static __inline__ __m256bh __DEFAULT_FN_ATTRS512
+_mm512_cvtneps_pbh(__m512 __A) {
+ return (__m256bh)__builtin_ia32_cvtneps2bf16_512_mask((__v16sf)__A,
+ (__v16hi)_mm256_undefined_si256(),
+ (__mmask16)-1);
+}
+
+/// Convert Packed Single Data to Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNEPS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 512-bit vector of [16 x float].
+/// \param __W
+/// A 256-bit vector of [16 x bfloat].
+/// \param __U
+/// A 16-bit mask value specifying what is chosen for each element.
+/// A 1 means conversion of __A. A 0 means element from __W.
+/// \returns A 256-bit vector of [16 x bfloat] come from conversion of __A.
+static __inline__ __m256bh __DEFAULT_FN_ATTRS512
+_mm512_mask_cvtneps_pbh(__m256bh __W, __mmask16 __U, __m512 __A) {
+ return (__m256bh)__builtin_ia32_cvtneps2bf16_512_mask((__v16sf)__A,
+ (__v16hi)__W,
+ (__mmask16)__U);
+}
+
+/// Convert Packed Single Data to Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNEPS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 512-bit vector of [16 x float].
+/// \param __U
+/// A 16-bit mask value specifying what is chosen for each element.
+/// A 1 means conversion of __A. A 0 means element is zero.
+/// \returns A 256-bit vector of [16 x bfloat] come from conversion of __A.
+static __inline__ __m256bh __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvtneps_pbh(__mmask16 __U, __m512 __A) {
+ return (__m256bh)__builtin_ia32_cvtneps2bf16_512_mask((__v16sf)__A,
+ (__v16hi)_mm256_setzero_si256(),
+ (__mmask16)__U);
+}
+
+/// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VDPBF16PS </c> instructions.
+///
+/// \param __A
+/// A 512-bit vector of [32 x bfloat].
+/// \param __B
+/// A 512-bit vector of [32 x bfloat].
+/// \param __D
+/// A 512-bit vector of [16 x float].
+/// \returns A 512-bit vector of [16 x float] comes from Dot Product of
+/// __A, __B and __D
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
+_mm512_dpbf16_ps(__m512 __D, __m512bh __A, __m512bh __B) {
+ return (__m512)__builtin_ia32_dpbf16ps_512((__v16sf) __D,
+ (__v16si) __A,
+ (__v16si) __B);
+}
+
+/// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VDPBF16PS </c> instructions.
+///
+/// \param __A
+/// A 512-bit vector of [32 x bfloat].
+/// \param __B
+/// A 512-bit vector of [32 x bfloat].
+/// \param __D
+/// A 512-bit vector of [16 x float].
+/// \param __U
+/// A 16-bit mask value specifying what is chosen for each element.
+/// A 1 means __A and __B's dot product accumulated with __D. A 0 means __D.
+/// \returns A 512-bit vector of [16 x float] comes from Dot Product of
+/// __A, __B and __D
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
+_mm512_mask_dpbf16_ps(__m512 __D, __mmask16 __U, __m512bh __A, __m512bh __B) {
+ return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
+ (__v16sf)_mm512_dpbf16_ps(__D, __A, __B),
+ (__v16sf)__D);
+}
+
+/// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VDPBF16PS </c> instructions.
+///
+/// \param __A
+/// A 512-bit vector of [32 x bfloat].
+/// \param __B
+/// A 512-bit vector of [32 x bfloat].
+/// \param __D
+/// A 512-bit vector of [16 x float].
+/// \param __U
+/// A 16-bit mask value specifying what is chosen for each element.
+/// A 1 means __A and __B's dot product accumulated with __D. A 0 means 0.
+/// \returns A 512-bit vector of [16 x float] comes from Dot Product of
+/// __A, __B and __D
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
+_mm512_maskz_dpbf16_ps(__mmask16 __U, __m512 __D, __m512bh __A, __m512bh __B) {
+ return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
+ (__v16sf)_mm512_dpbf16_ps(__D, __A, __B),
+ (__v16sf)_mm512_setzero_si512());
+}
+
+/// Convert Packed BF16 Data to Packed float Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// \param __A
+/// A 256-bit vector of [16 x bfloat].
+/// \returns A 512-bit vector of [16 x float] come from convertion of __A
+static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_cvtpbh_ps(__m256bh __A) {
+ return _mm512_castsi512_ps((__m512i)_mm512_slli_epi32(
+ (__m512i)_mm512_cvtepi16_epi32((__m256i)__A), 16));
+}
+
+/// Convert Packed BF16 Data to Packed float Data using zeroing mask.
+///
+/// \headerfile <x86intrin.h>
+///
+/// \param __U
+/// A 16-bit mask. Elements are zeroed out when the corresponding mask
+/// bit is not set.
+/// \param __A
+/// A 256-bit vector of [16 x bfloat].
+/// \returns A 512-bit vector of [16 x float] come from convertion of __A
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvtpbh_ps(__mmask16 __U, __m256bh __A) {
+ return _mm512_castsi512_ps((__m512i)_mm512_slli_epi32(
+ (__m512i)_mm512_maskz_cvtepi16_epi32((__mmask16)__U, (__m256i)__A), 16));
+}
+
+/// Convert Packed BF16 Data to Packed float Data using merging mask.
+///
+/// \headerfile <x86intrin.h>
+///
+/// \param __S
+/// A 512-bit vector of [16 x float]. Elements are copied from __S when
+/// the corresponding mask bit is not set.
+/// \param __U
+/// A 16-bit mask.
+/// \param __A
+/// A 256-bit vector of [16 x bfloat].
+/// \returns A 512-bit vector of [16 x float] come from convertion of __A
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
+_mm512_mask_cvtpbh_ps(__m512 __S, __mmask16 __U, __m256bh __A) {
+ return _mm512_castsi512_ps((__m512i)_mm512_mask_slli_epi32(
+ (__m512i)__S, (__mmask16)__U,
+ (__m512i)_mm512_cvtepi16_epi32((__m256i)__A), 16));
+}
+
+#undef __DEFAULT_FN_ATTRS
+#undef __DEFAULT_FN_ATTRS512
+
+#endif
--- /dev/null
+/*===------------- avx512bitalgintrin.h - BITALG intrinsics ------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512bitalgintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __AVX512BITALGINTRIN_H
+#define __AVX512BITALGINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512bitalg"), __min_vector_width__(512)))
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_popcnt_epi16(__m512i __A)
+{
+ return (__m512i) __builtin_ia32_vpopcntw_512((__v32hi) __A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_popcnt_epi16(__m512i __A, __mmask32 __U, __m512i __B)
+{
+ return (__m512i) __builtin_ia32_selectw_512((__mmask32) __U,
+ (__v32hi) _mm512_popcnt_epi16(__B),
+ (__v32hi) __A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_popcnt_epi16(__mmask32 __U, __m512i __B)
+{
+ return _mm512_mask_popcnt_epi16((__m512i) _mm512_setzero_si512(),
+ __U,
+ __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_popcnt_epi8(__m512i __A)
+{
+ return (__m512i) __builtin_ia32_vpopcntb_512((__v64qi) __A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_popcnt_epi8(__m512i __A, __mmask64 __U, __m512i __B)
+{
+ return (__m512i) __builtin_ia32_selectb_512((__mmask64) __U,
+ (__v64qi) _mm512_popcnt_epi8(__B),
+ (__v64qi) __A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_popcnt_epi8(__mmask64 __U, __m512i __B)
+{
+ return _mm512_mask_popcnt_epi8((__m512i) _mm512_setzero_si512(),
+ __U,
+ __B);
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS
+_mm512_mask_bitshuffle_epi64_mask(__mmask64 __U, __m512i __A, __m512i __B)
+{
+ return (__mmask64) __builtin_ia32_vpshufbitqmb512_mask((__v64qi) __A,
+ (__v64qi) __B,
+ __U);
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS
+_mm512_bitshuffle_epi64_mask(__m512i __A, __m512i __B)
+{
+ return _mm512_mask_bitshuffle_epi64_mask((__mmask64) -1,
+ __A,
+ __B);
+}
+
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif
--- /dev/null
+/*===------------- avx512bwintrin.h - AVX512BW intrinsics ------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512bwintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __AVX512BWINTRIN_H
+#define __AVX512BWINTRIN_H
+
+typedef unsigned int __mmask32;
+typedef unsigned long long __mmask64;
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS512 __attribute__((__always_inline__, __nodebug__, __target__("avx512bw"), __min_vector_width__(512)))
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512bw")))
+
+static __inline __mmask32 __DEFAULT_FN_ATTRS
+_knot_mask32(__mmask32 __M)
+{
+ return __builtin_ia32_knotsi(__M);
+}
+
+static __inline __mmask64 __DEFAULT_FN_ATTRS
+_knot_mask64(__mmask64 __M)
+{
+ return __builtin_ia32_knotdi(__M);
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS
+_kand_mask32(__mmask32 __A, __mmask32 __B)
+{
+ return (__mmask32)__builtin_ia32_kandsi((__mmask32)__A, (__mmask32)__B);
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS
+_kand_mask64(__mmask64 __A, __mmask64 __B)
+{
+ return (__mmask64)__builtin_ia32_kanddi((__mmask64)__A, (__mmask64)__B);
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS
+_kandn_mask32(__mmask32 __A, __mmask32 __B)
+{
+ return (__mmask32)__builtin_ia32_kandnsi((__mmask32)__A, (__mmask32)__B);
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS
+_kandn_mask64(__mmask64 __A, __mmask64 __B)
+{
+ return (__mmask64)__builtin_ia32_kandndi((__mmask64)__A, (__mmask64)__B);
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS
+_kor_mask32(__mmask32 __A, __mmask32 __B)
+{
+ return (__mmask32)__builtin_ia32_korsi((__mmask32)__A, (__mmask32)__B);
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS
+_kor_mask64(__mmask64 __A, __mmask64 __B)
+{
+ return (__mmask64)__builtin_ia32_kordi((__mmask64)__A, (__mmask64)__B);
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS
+_kxnor_mask32(__mmask32 __A, __mmask32 __B)
+{
+ return (__mmask32)__builtin_ia32_kxnorsi((__mmask32)__A, (__mmask32)__B);
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS
+_kxnor_mask64(__mmask64 __A, __mmask64 __B)
+{
+ return (__mmask64)__builtin_ia32_kxnordi((__mmask64)__A, (__mmask64)__B);
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS
+_kxor_mask32(__mmask32 __A, __mmask32 __B)
+{
+ return (__mmask32)__builtin_ia32_kxorsi((__mmask32)__A, (__mmask32)__B);
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS
+_kxor_mask64(__mmask64 __A, __mmask64 __B)
+{
+ return (__mmask64)__builtin_ia32_kxordi((__mmask64)__A, (__mmask64)__B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_kortestc_mask32_u8(__mmask32 __A, __mmask32 __B)
+{
+ return (unsigned char)__builtin_ia32_kortestcsi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_kortestz_mask32_u8(__mmask32 __A, __mmask32 __B)
+{
+ return (unsigned char)__builtin_ia32_kortestzsi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_kortest_mask32_u8(__mmask32 __A, __mmask32 __B, unsigned char *__C) {
+ *__C = (unsigned char)__builtin_ia32_kortestcsi(__A, __B);
+ return (unsigned char)__builtin_ia32_kortestzsi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_kortestc_mask64_u8(__mmask64 __A, __mmask64 __B)
+{
+ return (unsigned char)__builtin_ia32_kortestcdi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_kortestz_mask64_u8(__mmask64 __A, __mmask64 __B)
+{
+ return (unsigned char)__builtin_ia32_kortestzdi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_kortest_mask64_u8(__mmask64 __A, __mmask64 __B, unsigned char *__C) {
+ *__C = (unsigned char)__builtin_ia32_kortestcdi(__A, __B);
+ return (unsigned char)__builtin_ia32_kortestzdi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_ktestc_mask32_u8(__mmask32 __A, __mmask32 __B)
+{
+ return (unsigned char)__builtin_ia32_ktestcsi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_ktestz_mask32_u8(__mmask32 __A, __mmask32 __B)
+{
+ return (unsigned char)__builtin_ia32_ktestzsi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_ktest_mask32_u8(__mmask32 __A, __mmask32 __B, unsigned char *__C) {
+ *__C = (unsigned char)__builtin_ia32_ktestcsi(__A, __B);
+ return (unsigned char)__builtin_ia32_ktestzsi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_ktestc_mask64_u8(__mmask64 __A, __mmask64 __B)
+{
+ return (unsigned char)__builtin_ia32_ktestcdi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_ktestz_mask64_u8(__mmask64 __A, __mmask64 __B)
+{
+ return (unsigned char)__builtin_ia32_ktestzdi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_ktest_mask64_u8(__mmask64 __A, __mmask64 __B, unsigned char *__C) {
+ *__C = (unsigned char)__builtin_ia32_ktestcdi(__A, __B);
+ return (unsigned char)__builtin_ia32_ktestzdi(__A, __B);
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS
+_kadd_mask32(__mmask32 __A, __mmask32 __B)
+{
+ return (__mmask32)__builtin_ia32_kaddsi((__mmask32)__A, (__mmask32)__B);
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS
+_kadd_mask64(__mmask64 __A, __mmask64 __B)
+{
+ return (__mmask64)__builtin_ia32_kadddi((__mmask64)__A, (__mmask64)__B);
+}
+
+#define _kshiftli_mask32(A, I) \
+ (__mmask32)__builtin_ia32_kshiftlisi((__mmask32)(A), (unsigned int)(I))
+
+#define _kshiftri_mask32(A, I) \
+ (__mmask32)__builtin_ia32_kshiftrisi((__mmask32)(A), (unsigned int)(I))
+
+#define _kshiftli_mask64(A, I) \
+ (__mmask64)__builtin_ia32_kshiftlidi((__mmask64)(A), (unsigned int)(I))
+
+#define _kshiftri_mask64(A, I) \
+ (__mmask64)__builtin_ia32_kshiftridi((__mmask64)(A), (unsigned int)(I))
+
+static __inline__ unsigned int __DEFAULT_FN_ATTRS
+_cvtmask32_u32(__mmask32 __A) {
+ return (unsigned int)__builtin_ia32_kmovd((__mmask32)__A);
+}
+
+static __inline__ unsigned long long __DEFAULT_FN_ATTRS
+_cvtmask64_u64(__mmask64 __A) {
+ return (unsigned long long)__builtin_ia32_kmovq((__mmask64)__A);
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS
+_cvtu32_mask32(unsigned int __A) {
+ return (__mmask32)__builtin_ia32_kmovd((__mmask32)__A);
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS
+_cvtu64_mask64(unsigned long long __A) {
+ return (__mmask64)__builtin_ia32_kmovq((__mmask64)__A);
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS
+_load_mask32(__mmask32 *__A) {
+ return (__mmask32)__builtin_ia32_kmovd(*(__mmask32 *)__A);
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS
+_load_mask64(__mmask64 *__A) {
+ return (__mmask64)__builtin_ia32_kmovq(*(__mmask64 *)__A);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_store_mask32(__mmask32 *__A, __mmask32 __B) {
+ *(__mmask32 *)__A = __builtin_ia32_kmovd((__mmask32)__B);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_store_mask64(__mmask64 *__A, __mmask64 __B) {
+ *(__mmask64 *)__A = __builtin_ia32_kmovq((__mmask64)__B);
+}
+
+/* Integer compare */
+
+#define _mm512_cmp_epi8_mask(a, b, p) \
+ (__mmask64)__builtin_ia32_cmpb512_mask((__v64qi)(__m512i)(a), \
+ (__v64qi)(__m512i)(b), (int)(p), \
+ (__mmask64)-1)
+
+#define _mm512_mask_cmp_epi8_mask(m, a, b, p) \
+ (__mmask64)__builtin_ia32_cmpb512_mask((__v64qi)(__m512i)(a), \
+ (__v64qi)(__m512i)(b), (int)(p), \
+ (__mmask64)(m))
+
+#define _mm512_cmp_epu8_mask(a, b, p) \
+ (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)(__m512i)(a), \
+ (__v64qi)(__m512i)(b), (int)(p), \
+ (__mmask64)-1)
+
+#define _mm512_mask_cmp_epu8_mask(m, a, b, p) \
+ (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)(__m512i)(a), \
+ (__v64qi)(__m512i)(b), (int)(p), \
+ (__mmask64)(m))
+
+#define _mm512_cmp_epi16_mask(a, b, p) \
+ (__mmask32)__builtin_ia32_cmpw512_mask((__v32hi)(__m512i)(a), \
+ (__v32hi)(__m512i)(b), (int)(p), \
+ (__mmask32)-1)
+
+#define _mm512_mask_cmp_epi16_mask(m, a, b, p) \
+ (__mmask32)__builtin_ia32_cmpw512_mask((__v32hi)(__m512i)(a), \
+ (__v32hi)(__m512i)(b), (int)(p), \
+ (__mmask32)(m))
+
+#define _mm512_cmp_epu16_mask(a, b, p) \
+ (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)(__m512i)(a), \
+ (__v32hi)(__m512i)(b), (int)(p), \
+ (__mmask32)-1)
+
+#define _mm512_mask_cmp_epu16_mask(m, a, b, p) \
+ (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)(__m512i)(a), \
+ (__v32hi)(__m512i)(b), (int)(p), \
+ (__mmask32)(m))
+
+#define _mm512_cmpeq_epi8_mask(A, B) \
+ _mm512_cmp_epi8_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm512_mask_cmpeq_epi8_mask(k, A, B) \
+ _mm512_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm512_cmpge_epi8_mask(A, B) \
+ _mm512_cmp_epi8_mask((A), (B), _MM_CMPINT_GE)
+#define _mm512_mask_cmpge_epi8_mask(k, A, B) \
+ _mm512_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm512_cmpgt_epi8_mask(A, B) \
+ _mm512_cmp_epi8_mask((A), (B), _MM_CMPINT_GT)
+#define _mm512_mask_cmpgt_epi8_mask(k, A, B) \
+ _mm512_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm512_cmple_epi8_mask(A, B) \
+ _mm512_cmp_epi8_mask((A), (B), _MM_CMPINT_LE)
+#define _mm512_mask_cmple_epi8_mask(k, A, B) \
+ _mm512_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm512_cmplt_epi8_mask(A, B) \
+ _mm512_cmp_epi8_mask((A), (B), _MM_CMPINT_LT)
+#define _mm512_mask_cmplt_epi8_mask(k, A, B) \
+ _mm512_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm512_cmpneq_epi8_mask(A, B) \
+ _mm512_cmp_epi8_mask((A), (B), _MM_CMPINT_NE)
+#define _mm512_mask_cmpneq_epi8_mask(k, A, B) \
+ _mm512_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm512_cmpeq_epu8_mask(A, B) \
+ _mm512_cmp_epu8_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm512_mask_cmpeq_epu8_mask(k, A, B) \
+ _mm512_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm512_cmpge_epu8_mask(A, B) \
+ _mm512_cmp_epu8_mask((A), (B), _MM_CMPINT_GE)
+#define _mm512_mask_cmpge_epu8_mask(k, A, B) \
+ _mm512_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm512_cmpgt_epu8_mask(A, B) \
+ _mm512_cmp_epu8_mask((A), (B), _MM_CMPINT_GT)
+#define _mm512_mask_cmpgt_epu8_mask(k, A, B) \
+ _mm512_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm512_cmple_epu8_mask(A, B) \
+ _mm512_cmp_epu8_mask((A), (B), _MM_CMPINT_LE)
+#define _mm512_mask_cmple_epu8_mask(k, A, B) \
+ _mm512_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm512_cmplt_epu8_mask(A, B) \
+ _mm512_cmp_epu8_mask((A), (B), _MM_CMPINT_LT)
+#define _mm512_mask_cmplt_epu8_mask(k, A, B) \
+ _mm512_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm512_cmpneq_epu8_mask(A, B) \
+ _mm512_cmp_epu8_mask((A), (B), _MM_CMPINT_NE)
+#define _mm512_mask_cmpneq_epu8_mask(k, A, B) \
+ _mm512_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm512_cmpeq_epi16_mask(A, B) \
+ _mm512_cmp_epi16_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm512_mask_cmpeq_epi16_mask(k, A, B) \
+ _mm512_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm512_cmpge_epi16_mask(A, B) \
+ _mm512_cmp_epi16_mask((A), (B), _MM_CMPINT_GE)
+#define _mm512_mask_cmpge_epi16_mask(k, A, B) \
+ _mm512_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm512_cmpgt_epi16_mask(A, B) \
+ _mm512_cmp_epi16_mask((A), (B), _MM_CMPINT_GT)
+#define _mm512_mask_cmpgt_epi16_mask(k, A, B) \
+ _mm512_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm512_cmple_epi16_mask(A, B) \
+ _mm512_cmp_epi16_mask((A), (B), _MM_CMPINT_LE)
+#define _mm512_mask_cmple_epi16_mask(k, A, B) \
+ _mm512_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm512_cmplt_epi16_mask(A, B) \
+ _mm512_cmp_epi16_mask((A), (B), _MM_CMPINT_LT)
+#define _mm512_mask_cmplt_epi16_mask(k, A, B) \
+ _mm512_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm512_cmpneq_epi16_mask(A, B) \
+ _mm512_cmp_epi16_mask((A), (B), _MM_CMPINT_NE)
+#define _mm512_mask_cmpneq_epi16_mask(k, A, B) \
+ _mm512_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm512_cmpeq_epu16_mask(A, B) \
+ _mm512_cmp_epu16_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm512_mask_cmpeq_epu16_mask(k, A, B) \
+ _mm512_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm512_cmpge_epu16_mask(A, B) \
+ _mm512_cmp_epu16_mask((A), (B), _MM_CMPINT_GE)
+#define _mm512_mask_cmpge_epu16_mask(k, A, B) \
+ _mm512_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm512_cmpgt_epu16_mask(A, B) \
+ _mm512_cmp_epu16_mask((A), (B), _MM_CMPINT_GT)
+#define _mm512_mask_cmpgt_epu16_mask(k, A, B) \
+ _mm512_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm512_cmple_epu16_mask(A, B) \
+ _mm512_cmp_epu16_mask((A), (B), _MM_CMPINT_LE)
+#define _mm512_mask_cmple_epu16_mask(k, A, B) \
+ _mm512_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm512_cmplt_epu16_mask(A, B) \
+ _mm512_cmp_epu16_mask((A), (B), _MM_CMPINT_LT)
+#define _mm512_mask_cmplt_epu16_mask(k, A, B) \
+ _mm512_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm512_cmpneq_epu16_mask(A, B) \
+ _mm512_cmp_epu16_mask((A), (B), _MM_CMPINT_NE)
+#define _mm512_mask_cmpneq_epu16_mask(k, A, B) \
+ _mm512_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_NE)
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_add_epi8 (__m512i __A, __m512i __B) {
+ return (__m512i) ((__v64qu) __A + (__v64qu) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_add_epi8(__m512i __W, __mmask64 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_add_epi8(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_add_epi8(__mmask64 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_add_epi8(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_sub_epi8 (__m512i __A, __m512i __B) {
+ return (__m512i) ((__v64qu) __A - (__v64qu) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_sub_epi8(__m512i __W, __mmask64 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_sub_epi8(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_sub_epi8(__mmask64 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_sub_epi8(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_add_epi16 (__m512i __A, __m512i __B) {
+ return (__m512i) ((__v32hu) __A + (__v32hu) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_add_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_add_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_add_epi16(__mmask32 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_add_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_sub_epi16 (__m512i __A, __m512i __B) {
+ return (__m512i) ((__v32hu) __A - (__v32hu) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_sub_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_sub_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_sub_epi16(__mmask32 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_sub_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mullo_epi16 (__m512i __A, __m512i __B) {
+ return (__m512i) ((__v32hu) __A * (__v32hu) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_mullo_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_mullo_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_mullo_epi16(__mmask32 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_mullo_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_blend_epi8 (__mmask64 __U, __m512i __A, __m512i __W)
+{
+ return (__m512i) __builtin_ia32_selectb_512 ((__mmask64) __U,
+ (__v64qi) __W,
+ (__v64qi) __A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_blend_epi16 (__mmask32 __U, __m512i __A, __m512i __W)
+{
+ return (__m512i) __builtin_ia32_selectw_512 ((__mmask32) __U,
+ (__v32hi) __W,
+ (__v32hi) __A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_abs_epi8 (__m512i __A)
+{
+ return (__m512i)__builtin_ia32_pabsb512((__v64qi)__A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_abs_epi8 (__m512i __W, __mmask64 __U, __m512i __A)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_abs_epi8(__A),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_abs_epi8 (__mmask64 __U, __m512i __A)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_abs_epi8(__A),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_abs_epi16 (__m512i __A)
+{
+ return (__m512i)__builtin_ia32_pabsw512((__v32hi)__A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_abs_epi16 (__m512i __W, __mmask32 __U, __m512i __A)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_abs_epi16(__A),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_abs_epi16 (__mmask32 __U, __m512i __A)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_abs_epi16(__A),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_packs_epi32(__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_packssdw512((__v16si)__A, (__v16si)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_packs_epi32(__mmask32 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
+ (__v32hi)_mm512_packs_epi32(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_packs_epi32(__m512i __W, __mmask32 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
+ (__v32hi)_mm512_packs_epi32(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_packs_epi16(__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_packsswb512((__v32hi)__A, (__v32hi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_packs_epi16(__m512i __W, __mmask64 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_packs_epi16(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_packs_epi16(__mmask64 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_packs_epi16(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_packus_epi32(__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_packusdw512((__v16si) __A, (__v16si) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_packus_epi32(__mmask32 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
+ (__v32hi)_mm512_packus_epi32(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_packus_epi32(__m512i __W, __mmask32 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
+ (__v32hi)_mm512_packus_epi32(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_packus_epi16(__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_packuswb512((__v32hi) __A, (__v32hi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_packus_epi16(__m512i __W, __mmask64 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_packus_epi16(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_packus_epi16(__mmask64 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_packus_epi16(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_adds_epi8 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_paddsb512((__v64qi)__A, (__v64qi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_adds_epi8 (__m512i __W, __mmask64 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_adds_epi8(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_adds_epi8 (__mmask64 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_adds_epi8(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_adds_epi16 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_paddsw512((__v32hi)__A, (__v32hi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_adds_epi16 (__m512i __W, __mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_adds_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_adds_epi16 (__mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_adds_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_adds_epu8 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_paddusb512((__v64qi) __A, (__v64qi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_adds_epu8 (__m512i __W, __mmask64 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_adds_epu8(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_adds_epu8 (__mmask64 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_adds_epu8(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_adds_epu16 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_paddusw512((__v32hi) __A, (__v32hi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_adds_epu16 (__m512i __W, __mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_adds_epu16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_adds_epu16 (__mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_adds_epu16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_avg_epu8 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_pavgb512((__v64qi)__A, (__v64qi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_avg_epu8 (__m512i __W, __mmask64 __U, __m512i __A,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_avg_epu8(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_avg_epu8 (__mmask64 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_avg_epu8(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_avg_epu16 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_pavgw512((__v32hi)__A, (__v32hi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_avg_epu16 (__m512i __W, __mmask32 __U, __m512i __A,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_avg_epu16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_avg_epu16 (__mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_avg_epu16(__A, __B),
+ (__v32hi) _mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_max_epi8 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_pmaxsb512((__v64qi) __A, (__v64qi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_max_epi8 (__mmask64 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_max_epi8(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_max_epi8 (__m512i __W, __mmask64 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_max_epi8(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_max_epi16 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_pmaxsw512((__v32hi) __A, (__v32hi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_max_epi16 (__mmask32 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
+ (__v32hi)_mm512_max_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_max_epi16 (__m512i __W, __mmask32 __M, __m512i __A,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
+ (__v32hi)_mm512_max_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_max_epu8 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_pmaxub512((__v64qi)__A, (__v64qi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_max_epu8 (__mmask64 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_max_epu8(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_max_epu8 (__m512i __W, __mmask64 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_max_epu8(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_max_epu16 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_pmaxuw512((__v32hi)__A, (__v32hi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_max_epu16 (__mmask32 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
+ (__v32hi)_mm512_max_epu16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_max_epu16 (__m512i __W, __mmask32 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
+ (__v32hi)_mm512_max_epu16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_min_epi8 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_pminsb512((__v64qi) __A, (__v64qi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_min_epi8 (__mmask64 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_min_epi8(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_min_epi8 (__m512i __W, __mmask64 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_min_epi8(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_min_epi16 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_pminsw512((__v32hi) __A, (__v32hi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_min_epi16 (__mmask32 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
+ (__v32hi)_mm512_min_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_min_epi16 (__m512i __W, __mmask32 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
+ (__v32hi)_mm512_min_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_min_epu8 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_pminub512((__v64qi)__A, (__v64qi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_min_epu8 (__mmask64 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_min_epu8(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_min_epu8 (__m512i __W, __mmask64 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_min_epu8(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_min_epu16 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_pminuw512((__v32hi)__A, (__v32hi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_min_epu16 (__mmask32 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
+ (__v32hi)_mm512_min_epu16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_min_epu16 (__m512i __W, __mmask32 __M, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
+ (__v32hi)_mm512_min_epu16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_shuffle_epi8(__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_pshufb512((__v64qi)__A,(__v64qi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_shuffle_epi8(__m512i __W, __mmask64 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_shuffle_epi8(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_shuffle_epi8(__mmask64 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_shuffle_epi8(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_subs_epi8 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_psubsb512((__v64qi)__A, (__v64qi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_subs_epi8 (__m512i __W, __mmask64 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_subs_epi8(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_subs_epi8 (__mmask64 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_subs_epi8(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_subs_epi16 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_psubsw512((__v32hi)__A, (__v32hi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_subs_epi16 (__m512i __W, __mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_subs_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_subs_epi16 (__mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_subs_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_subs_epu8 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_psubusb512((__v64qi) __A, (__v64qi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_subs_epu8 (__m512i __W, __mmask64 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_subs_epu8(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_subs_epu8 (__mmask64 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_subs_epu8(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_subs_epu16 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_psubusw512((__v32hi) __A, (__v32hi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_subs_epu16 (__m512i __W, __mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_subs_epu16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_subs_epu16 (__mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_subs_epu16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_permutex2var_epi16(__m512i __A, __m512i __I, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_vpermi2varhi512((__v32hi)__A, (__v32hi)__I,
+ (__v32hi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_permutex2var_epi16(__m512i __A, __mmask32 __U, __m512i __I,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512(__U,
+ (__v32hi)_mm512_permutex2var_epi16(__A, __I, __B),
+ (__v32hi)__A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask2_permutex2var_epi16(__m512i __A, __m512i __I, __mmask32 __U,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512(__U,
+ (__v32hi)_mm512_permutex2var_epi16(__A, __I, __B),
+ (__v32hi)__I);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_permutex2var_epi16(__mmask32 __U, __m512i __A, __m512i __I,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512(__U,
+ (__v32hi)_mm512_permutex2var_epi16(__A, __I, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mulhrs_epi16(__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_pmulhrsw512((__v32hi)__A, (__v32hi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_mulhrs_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_mulhrs_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_mulhrs_epi16(__mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_mulhrs_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mulhi_epi16(__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_pmulhw512((__v32hi) __A, (__v32hi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_mulhi_epi16(__m512i __W, __mmask32 __U, __m512i __A,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_mulhi_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_mulhi_epi16(__mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_mulhi_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mulhi_epu16(__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_pmulhuw512((__v32hi) __A, (__v32hi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_mulhi_epu16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_mulhi_epu16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_mulhi_epu16 (__mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_mulhi_epu16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maddubs_epi16(__m512i __X, __m512i __Y) {
+ return (__m512i)__builtin_ia32_pmaddubsw512((__v64qi)__X, (__v64qi)__Y);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_maddubs_epi16(__m512i __W, __mmask32 __U, __m512i __X,
+ __m512i __Y) {
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32) __U,
+ (__v32hi)_mm512_maddubs_epi16(__X, __Y),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_maddubs_epi16(__mmask32 __U, __m512i __X, __m512i __Y) {
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32) __U,
+ (__v32hi)_mm512_maddubs_epi16(__X, __Y),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_madd_epi16(__m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_pmaddwd512((__v32hi)__A, (__v32hi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_madd_epi16(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
+ (__v16si)_mm512_madd_epi16(__A, __B),
+ (__v16si)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_madd_epi16(__mmask16 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
+ (__v16si)_mm512_madd_epi16(__A, __B),
+ (__v16si)_mm512_setzero_si512());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_cvtsepi16_epi8 (__m512i __A) {
+ return (__m256i) __builtin_ia32_pmovswb512_mask ((__v32hi) __A,
+ (__v32qi)_mm256_setzero_si256(),
+ (__mmask32) -1);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_mask_cvtsepi16_epi8 (__m256i __O, __mmask32 __M, __m512i __A) {
+ return (__m256i) __builtin_ia32_pmovswb512_mask ((__v32hi) __A,
+ (__v32qi)__O,
+ __M);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvtsepi16_epi8 (__mmask32 __M, __m512i __A) {
+ return (__m256i) __builtin_ia32_pmovswb512_mask ((__v32hi) __A,
+ (__v32qi) _mm256_setzero_si256(),
+ __M);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_cvtusepi16_epi8 (__m512i __A) {
+ return (__m256i) __builtin_ia32_pmovuswb512_mask ((__v32hi) __A,
+ (__v32qi) _mm256_setzero_si256(),
+ (__mmask32) -1);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_mask_cvtusepi16_epi8 (__m256i __O, __mmask32 __M, __m512i __A) {
+ return (__m256i) __builtin_ia32_pmovuswb512_mask ((__v32hi) __A,
+ (__v32qi) __O,
+ __M);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvtusepi16_epi8 (__mmask32 __M, __m512i __A) {
+ return (__m256i) __builtin_ia32_pmovuswb512_mask ((__v32hi) __A,
+ (__v32qi) _mm256_setzero_si256(),
+ __M);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_cvtepi16_epi8 (__m512i __A) {
+ return (__m256i) __builtin_ia32_pmovwb512_mask ((__v32hi) __A,
+ (__v32qi) _mm256_undefined_si256(),
+ (__mmask32) -1);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_mask_cvtepi16_epi8 (__m256i __O, __mmask32 __M, __m512i __A) {
+ return (__m256i) __builtin_ia32_pmovwb512_mask ((__v32hi) __A,
+ (__v32qi) __O,
+ __M);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvtepi16_epi8 (__mmask32 __M, __m512i __A) {
+ return (__m256i) __builtin_ia32_pmovwb512_mask ((__v32hi) __A,
+ (__v32qi) _mm256_setzero_si256(),
+ __M);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS512
+_mm512_mask_cvtepi16_storeu_epi8 (void * __P, __mmask32 __M, __m512i __A)
+{
+ __builtin_ia32_pmovwb512mem_mask ((__v32qi *) __P, (__v32hi) __A, __M);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS512
+_mm512_mask_cvtsepi16_storeu_epi8 (void * __P, __mmask32 __M, __m512i __A)
+{
+ __builtin_ia32_pmovswb512mem_mask ((__v32qi *) __P, (__v32hi) __A, __M);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS512
+_mm512_mask_cvtusepi16_storeu_epi8 (void * __P, __mmask32 __M, __m512i __A)
+{
+ __builtin_ia32_pmovuswb512mem_mask ((__v32qi *) __P, (__v32hi) __A, __M);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_unpackhi_epi8(__m512i __A, __m512i __B) {
+ return (__m512i)__builtin_shufflevector((__v64qi)__A, (__v64qi)__B,
+ 8, 64+8, 9, 64+9,
+ 10, 64+10, 11, 64+11,
+ 12, 64+12, 13, 64+13,
+ 14, 64+14, 15, 64+15,
+ 24, 64+24, 25, 64+25,
+ 26, 64+26, 27, 64+27,
+ 28, 64+28, 29, 64+29,
+ 30, 64+30, 31, 64+31,
+ 40, 64+40, 41, 64+41,
+ 42, 64+42, 43, 64+43,
+ 44, 64+44, 45, 64+45,
+ 46, 64+46, 47, 64+47,
+ 56, 64+56, 57, 64+57,
+ 58, 64+58, 59, 64+59,
+ 60, 64+60, 61, 64+61,
+ 62, 64+62, 63, 64+63);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_unpackhi_epi8(__m512i __W, __mmask64 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_unpackhi_epi8(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_unpackhi_epi8(__mmask64 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_unpackhi_epi8(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_unpackhi_epi16(__m512i __A, __m512i __B) {
+ return (__m512i)__builtin_shufflevector((__v32hi)__A, (__v32hi)__B,
+ 4, 32+4, 5, 32+5,
+ 6, 32+6, 7, 32+7,
+ 12, 32+12, 13, 32+13,
+ 14, 32+14, 15, 32+15,
+ 20, 32+20, 21, 32+21,
+ 22, 32+22, 23, 32+23,
+ 28, 32+28, 29, 32+29,
+ 30, 32+30, 31, 32+31);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_unpackhi_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_unpackhi_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_unpackhi_epi16(__mmask32 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_unpackhi_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_unpacklo_epi8(__m512i __A, __m512i __B) {
+ return (__m512i)__builtin_shufflevector((__v64qi)__A, (__v64qi)__B,
+ 0, 64+0, 1, 64+1,
+ 2, 64+2, 3, 64+3,
+ 4, 64+4, 5, 64+5,
+ 6, 64+6, 7, 64+7,
+ 16, 64+16, 17, 64+17,
+ 18, 64+18, 19, 64+19,
+ 20, 64+20, 21, 64+21,
+ 22, 64+22, 23, 64+23,
+ 32, 64+32, 33, 64+33,
+ 34, 64+34, 35, 64+35,
+ 36, 64+36, 37, 64+37,
+ 38, 64+38, 39, 64+39,
+ 48, 64+48, 49, 64+49,
+ 50, 64+50, 51, 64+51,
+ 52, 64+52, 53, 64+53,
+ 54, 64+54, 55, 64+55);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_unpacklo_epi8(__m512i __W, __mmask64 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_unpacklo_epi8(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_unpacklo_epi8(__mmask64 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
+ (__v64qi)_mm512_unpacklo_epi8(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_unpacklo_epi16(__m512i __A, __m512i __B) {
+ return (__m512i)__builtin_shufflevector((__v32hi)__A, (__v32hi)__B,
+ 0, 32+0, 1, 32+1,
+ 2, 32+2, 3, 32+3,
+ 8, 32+8, 9, 32+9,
+ 10, 32+10, 11, 32+11,
+ 16, 32+16, 17, 32+17,
+ 18, 32+18, 19, 32+19,
+ 24, 32+24, 25, 32+25,
+ 26, 32+26, 27, 32+27);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_unpacklo_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_unpacklo_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_unpacklo_epi16(__mmask32 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_unpacklo_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_cvtepi8_epi16(__m256i __A)
+{
+ /* This function always performs a signed extension, but __v32qi is a char
+ which may be signed or unsigned, so use __v32qs. */
+ return (__m512i)__builtin_convertvector((__v32qs)__A, __v32hi);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_cvtepi8_epi16(__m512i __W, __mmask32 __U, __m256i __A)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_cvtepi8_epi16(__A),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvtepi8_epi16(__mmask32 __U, __m256i __A)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_cvtepi8_epi16(__A),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_cvtepu8_epi16(__m256i __A)
+{
+ return (__m512i)__builtin_convertvector((__v32qu)__A, __v32hi);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_cvtepu8_epi16(__m512i __W, __mmask32 __U, __m256i __A)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_cvtepu8_epi16(__A),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvtepu8_epi16(__mmask32 __U, __m256i __A)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_cvtepu8_epi16(__A),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+
+#define _mm512_shufflehi_epi16(A, imm) \
+ (__m512i)__builtin_ia32_pshufhw512((__v32hi)(__m512i)(A), (int)(imm))
+
+#define _mm512_mask_shufflehi_epi16(W, U, A, imm) \
+ (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
+ (__v32hi)_mm512_shufflehi_epi16((A), \
+ (imm)), \
+ (__v32hi)(__m512i)(W))
+
+#define _mm512_maskz_shufflehi_epi16(U, A, imm) \
+ (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
+ (__v32hi)_mm512_shufflehi_epi16((A), \
+ (imm)), \
+ (__v32hi)_mm512_setzero_si512())
+
+#define _mm512_shufflelo_epi16(A, imm) \
+ (__m512i)__builtin_ia32_pshuflw512((__v32hi)(__m512i)(A), (int)(imm))
+
+
+#define _mm512_mask_shufflelo_epi16(W, U, A, imm) \
+ (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
+ (__v32hi)_mm512_shufflelo_epi16((A), \
+ (imm)), \
+ (__v32hi)(__m512i)(W))
+
+
+#define _mm512_maskz_shufflelo_epi16(U, A, imm) \
+ (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
+ (__v32hi)_mm512_shufflelo_epi16((A), \
+ (imm)), \
+ (__v32hi)_mm512_setzero_si512())
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_sllv_epi16(__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_psllv32hi((__v32hi) __A, (__v32hi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_sllv_epi16 (__m512i __W, __mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_sllv_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_sllv_epi16(__mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_sllv_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_sll_epi16(__m512i __A, __m128i __B)
+{
+ return (__m512i)__builtin_ia32_psllw512((__v32hi) __A, (__v8hi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_sll_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m128i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_sll_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_sll_epi16(__mmask32 __U, __m512i __A, __m128i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_sll_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_slli_epi16(__m512i __A, unsigned int __B)
+{
+ return (__m512i)__builtin_ia32_psllwi512((__v32hi)__A, __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_slli_epi16(__m512i __W, __mmask32 __U, __m512i __A,
+ unsigned int __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_slli_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_slli_epi16(__mmask32 __U, __m512i __A, unsigned int __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_slli_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+#define _mm512_bslli_epi128(a, imm) \
+ (__m512i)__builtin_ia32_pslldqi512_byteshift((__v8di)(__m512i)(a), (int)(imm))
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_srlv_epi16(__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_psrlv32hi((__v32hi)__A, (__v32hi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_srlv_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_srlv_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_srlv_epi16(__mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_srlv_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_srav_epi16(__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_psrav32hi((__v32hi)__A, (__v32hi)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_srav_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_srav_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_srav_epi16(__mmask32 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_srav_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_sra_epi16(__m512i __A, __m128i __B)
+{
+ return (__m512i)__builtin_ia32_psraw512((__v32hi) __A, (__v8hi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_sra_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m128i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_sra_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_sra_epi16(__mmask32 __U, __m512i __A, __m128i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_sra_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_srai_epi16(__m512i __A, unsigned int __B)
+{
+ return (__m512i)__builtin_ia32_psrawi512((__v32hi)__A, __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_srai_epi16(__m512i __W, __mmask32 __U, __m512i __A,
+ unsigned int __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_srai_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_srai_epi16(__mmask32 __U, __m512i __A, unsigned int __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_srai_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_srl_epi16(__m512i __A, __m128i __B)
+{
+ return (__m512i)__builtin_ia32_psrlw512((__v32hi) __A, (__v8hi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_srl_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m128i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_srl_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_srl_epi16(__mmask32 __U, __m512i __A, __m128i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_srl_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_srli_epi16(__m512i __A, unsigned int __B)
+{
+ return (__m512i)__builtin_ia32_psrlwi512((__v32hi)__A, __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_srli_epi16(__m512i __W, __mmask32 __U, __m512i __A,
+ unsigned int __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_srli_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_srli_epi16(__mmask32 __U, __m512i __A, int __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_srli_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+#define _mm512_bsrli_epi128(a, imm) \
+ (__m512i)__builtin_ia32_psrldqi512_byteshift((__v8di)(__m512i)(a), (int)(imm))
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_mov_epi16 (__m512i __W, __mmask32 __U, __m512i __A)
+{
+ return (__m512i) __builtin_ia32_selectw_512 ((__mmask32) __U,
+ (__v32hi) __A,
+ (__v32hi) __W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_mov_epi16 (__mmask32 __U, __m512i __A)
+{
+ return (__m512i) __builtin_ia32_selectw_512 ((__mmask32) __U,
+ (__v32hi) __A,
+ (__v32hi) _mm512_setzero_si512 ());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_mov_epi8 (__m512i __W, __mmask64 __U, __m512i __A)
+{
+ return (__m512i) __builtin_ia32_selectb_512 ((__mmask64) __U,
+ (__v64qi) __A,
+ (__v64qi) __W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_mov_epi8 (__mmask64 __U, __m512i __A)
+{
+ return (__m512i) __builtin_ia32_selectb_512 ((__mmask64) __U,
+ (__v64qi) __A,
+ (__v64qi) _mm512_setzero_si512 ());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_set1_epi8 (__m512i __O, __mmask64 __M, char __A)
+{
+ return (__m512i) __builtin_ia32_selectb_512(__M,
+ (__v64qi)_mm512_set1_epi8(__A),
+ (__v64qi) __O);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_set1_epi8 (__mmask64 __M, char __A)
+{
+ return (__m512i) __builtin_ia32_selectb_512(__M,
+ (__v64qi) _mm512_set1_epi8(__A),
+ (__v64qi) _mm512_setzero_si512());
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS
+_mm512_kunpackd (__mmask64 __A, __mmask64 __B)
+{
+ return (__mmask64) __builtin_ia32_kunpckdi ((__mmask64) __A,
+ (__mmask64) __B);
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS
+_mm512_kunpackw (__mmask32 __A, __mmask32 __B)
+{
+ return (__mmask32) __builtin_ia32_kunpcksi ((__mmask32) __A,
+ (__mmask32) __B);
+}
+
+static __inline __m512i __DEFAULT_FN_ATTRS512
+_mm512_loadu_epi16 (void const *__P)
+{
+ struct __loadu_epi16 {
+ __m512i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ return ((const struct __loadu_epi16*)__P)->__v;
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_loadu_epi16 (__m512i __W, __mmask32 __U, void const *__P)
+{
+ return (__m512i) __builtin_ia32_loaddquhi512_mask ((const __v32hi *) __P,
+ (__v32hi) __W,
+ (__mmask32) __U);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_loadu_epi16 (__mmask32 __U, void const *__P)
+{
+ return (__m512i) __builtin_ia32_loaddquhi512_mask ((const __v32hi *) __P,
+ (__v32hi)
+ _mm512_setzero_si512 (),
+ (__mmask32) __U);
+}
+
+static __inline __m512i __DEFAULT_FN_ATTRS512
+_mm512_loadu_epi8 (void const *__P)
+{
+ struct __loadu_epi8 {
+ __m512i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ return ((const struct __loadu_epi8*)__P)->__v;
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_loadu_epi8 (__m512i __W, __mmask64 __U, void const *__P)
+{
+ return (__m512i) __builtin_ia32_loaddquqi512_mask ((const __v64qi *) __P,
+ (__v64qi) __W,
+ (__mmask64) __U);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_loadu_epi8 (__mmask64 __U, void const *__P)
+{
+ return (__m512i) __builtin_ia32_loaddquqi512_mask ((const __v64qi *) __P,
+ (__v64qi)
+ _mm512_setzero_si512 (),
+ (__mmask64) __U);
+}
+
+static __inline void __DEFAULT_FN_ATTRS512
+_mm512_storeu_epi16 (void *__P, __m512i __A)
+{
+ struct __storeu_epi16 {
+ __m512i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_epi16*)__P)->__v = __A;
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS512
+_mm512_mask_storeu_epi16 (void *__P, __mmask32 __U, __m512i __A)
+{
+ __builtin_ia32_storedquhi512_mask ((__v32hi *) __P,
+ (__v32hi) __A,
+ (__mmask32) __U);
+}
+
+static __inline void __DEFAULT_FN_ATTRS512
+_mm512_storeu_epi8 (void *__P, __m512i __A)
+{
+ struct __storeu_epi8 {
+ __m512i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_epi8*)__P)->__v = __A;
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS512
+_mm512_mask_storeu_epi8 (void *__P, __mmask64 __U, __m512i __A)
+{
+ __builtin_ia32_storedquqi512_mask ((__v64qi *) __P,
+ (__v64qi) __A,
+ (__mmask64) __U);
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS512
+_mm512_test_epi8_mask (__m512i __A, __m512i __B)
+{
+ return _mm512_cmpneq_epi8_mask (_mm512_and_epi32 (__A, __B),
+ _mm512_setzero_si512());
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS512
+_mm512_mask_test_epi8_mask (__mmask64 __U, __m512i __A, __m512i __B)
+{
+ return _mm512_mask_cmpneq_epi8_mask (__U, _mm512_and_epi32 (__A, __B),
+ _mm512_setzero_si512());
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS512
+_mm512_test_epi16_mask (__m512i __A, __m512i __B)
+{
+ return _mm512_cmpneq_epi16_mask (_mm512_and_epi32 (__A, __B),
+ _mm512_setzero_si512());
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS512
+_mm512_mask_test_epi16_mask (__mmask32 __U, __m512i __A, __m512i __B)
+{
+ return _mm512_mask_cmpneq_epi16_mask (__U, _mm512_and_epi32 (__A, __B),
+ _mm512_setzero_si512());
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS512
+_mm512_testn_epi8_mask (__m512i __A, __m512i __B)
+{
+ return _mm512_cmpeq_epi8_mask (_mm512_and_epi32 (__A, __B), _mm512_setzero_si512());
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS512
+_mm512_mask_testn_epi8_mask (__mmask64 __U, __m512i __A, __m512i __B)
+{
+ return _mm512_mask_cmpeq_epi8_mask (__U, _mm512_and_epi32 (__A, __B),
+ _mm512_setzero_si512());
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS512
+_mm512_testn_epi16_mask (__m512i __A, __m512i __B)
+{
+ return _mm512_cmpeq_epi16_mask (_mm512_and_epi32 (__A, __B),
+ _mm512_setzero_si512());
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS512
+_mm512_mask_testn_epi16_mask (__mmask32 __U, __m512i __A, __m512i __B)
+{
+ return _mm512_mask_cmpeq_epi16_mask (__U, _mm512_and_epi32 (__A, __B),
+ _mm512_setzero_si512());
+}
+
+static __inline__ __mmask64 __DEFAULT_FN_ATTRS512
+_mm512_movepi8_mask (__m512i __A)
+{
+ return (__mmask64) __builtin_ia32_cvtb2mask512 ((__v64qi) __A);
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS512
+_mm512_movepi16_mask (__m512i __A)
+{
+ return (__mmask32) __builtin_ia32_cvtw2mask512 ((__v32hi) __A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_movm_epi8 (__mmask64 __A)
+{
+ return (__m512i) __builtin_ia32_cvtmask2b512 (__A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_movm_epi16 (__mmask32 __A)
+{
+ return (__m512i) __builtin_ia32_cvtmask2w512 (__A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_broadcastb_epi8 (__m128i __A)
+{
+ return (__m512i)__builtin_shufflevector((__v16qi) __A, (__v16qi) __A,
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_broadcastb_epi8 (__m512i __O, __mmask64 __M, __m128i __A)
+{
+ return (__m512i)__builtin_ia32_selectb_512(__M,
+ (__v64qi) _mm512_broadcastb_epi8(__A),
+ (__v64qi) __O);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_broadcastb_epi8 (__mmask64 __M, __m128i __A)
+{
+ return (__m512i)__builtin_ia32_selectb_512(__M,
+ (__v64qi) _mm512_broadcastb_epi8(__A),
+ (__v64qi) _mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_set1_epi16 (__m512i __O, __mmask32 __M, short __A)
+{
+ return (__m512i) __builtin_ia32_selectw_512(__M,
+ (__v32hi) _mm512_set1_epi16(__A),
+ (__v32hi) __O);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_set1_epi16 (__mmask32 __M, short __A)
+{
+ return (__m512i) __builtin_ia32_selectw_512(__M,
+ (__v32hi) _mm512_set1_epi16(__A),
+ (__v32hi) _mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_broadcastw_epi16 (__m128i __A)
+{
+ return (__m512i)__builtin_shufflevector((__v8hi) __A, (__v8hi) __A,
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_broadcastw_epi16 (__m512i __O, __mmask32 __M, __m128i __A)
+{
+ return (__m512i)__builtin_ia32_selectw_512(__M,
+ (__v32hi) _mm512_broadcastw_epi16(__A),
+ (__v32hi) __O);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_broadcastw_epi16 (__mmask32 __M, __m128i __A)
+{
+ return (__m512i)__builtin_ia32_selectw_512(__M,
+ (__v32hi) _mm512_broadcastw_epi16(__A),
+ (__v32hi) _mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_permutexvar_epi16 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_permvarhi512((__v32hi)__B, (__v32hi)__A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_permutexvar_epi16 (__mmask32 __M, __m512i __A,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
+ (__v32hi)_mm512_permutexvar_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_permutexvar_epi16 (__m512i __W, __mmask32 __M, __m512i __A,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
+ (__v32hi)_mm512_permutexvar_epi16(__A, __B),
+ (__v32hi)__W);
+}
+
+#define _mm512_alignr_epi8(A, B, N) \
+ (__m512i)__builtin_ia32_palignr512((__v64qi)(__m512i)(A), \
+ (__v64qi)(__m512i)(B), (int)(N))
+
+#define _mm512_mask_alignr_epi8(W, U, A, B, N) \
+ (__m512i)__builtin_ia32_selectb_512((__mmask64)(U), \
+ (__v64qi)_mm512_alignr_epi8((A), (B), (int)(N)), \
+ (__v64qi)(__m512i)(W))
+
+#define _mm512_maskz_alignr_epi8(U, A, B, N) \
+ (__m512i)__builtin_ia32_selectb_512((__mmask64)(U), \
+ (__v64qi)_mm512_alignr_epi8((A), (B), (int)(N)), \
+ (__v64qi)(__m512i)_mm512_setzero_si512())
+
+#define _mm512_dbsad_epu8(A, B, imm) \
+ (__m512i)__builtin_ia32_dbpsadbw512((__v64qi)(__m512i)(A), \
+ (__v64qi)(__m512i)(B), (int)(imm))
+
+#define _mm512_mask_dbsad_epu8(W, U, A, B, imm) \
+ (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
+ (__v32hi)_mm512_dbsad_epu8((A), (B), (imm)), \
+ (__v32hi)(__m512i)(W))
+
+#define _mm512_maskz_dbsad_epu8(U, A, B, imm) \
+ (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
+ (__v32hi)_mm512_dbsad_epu8((A), (B), (imm)), \
+ (__v32hi)_mm512_setzero_si512())
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_sad_epu8 (__m512i __A, __m512i __B)
+{
+ return (__m512i) __builtin_ia32_psadbw512 ((__v64qi) __A,
+ (__v64qi) __B);
+}
+
+#undef __DEFAULT_FN_ATTRS512
+#undef __DEFAULT_FN_ATTRS
+
+#endif
--- /dev/null
+/*===------------- avx512cdintrin.h - AVX512CD intrinsics ------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512cdintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __AVX512CDINTRIN_H
+#define __AVX512CDINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512cd"), __min_vector_width__(512)))
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_conflict_epi64 (__m512i __A)
+{
+ return (__m512i) __builtin_ia32_vpconflictdi_512 ((__v8di) __A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_conflict_epi64 (__m512i __W, __mmask8 __U, __m512i __A)
+{
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
+ (__v8di)_mm512_conflict_epi64(__A),
+ (__v8di)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_conflict_epi64 (__mmask8 __U, __m512i __A)
+{
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
+ (__v8di)_mm512_conflict_epi64(__A),
+ (__v8di)_mm512_setzero_si512 ());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_conflict_epi32 (__m512i __A)
+{
+ return (__m512i) __builtin_ia32_vpconflictsi_512 ((__v16si) __A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_conflict_epi32 (__m512i __W, __mmask16 __U, __m512i __A)
+{
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
+ (__v16si)_mm512_conflict_epi32(__A),
+ (__v16si)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_conflict_epi32 (__mmask16 __U, __m512i __A)
+{
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
+ (__v16si)_mm512_conflict_epi32(__A),
+ (__v16si)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_lzcnt_epi32 (__m512i __A)
+{
+ return (__m512i) __builtin_ia32_vplzcntd_512 ((__v16si) __A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_lzcnt_epi32 (__m512i __W, __mmask16 __U, __m512i __A)
+{
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
+ (__v16si)_mm512_lzcnt_epi32(__A),
+ (__v16si)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_lzcnt_epi32 (__mmask16 __U, __m512i __A)
+{
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
+ (__v16si)_mm512_lzcnt_epi32(__A),
+ (__v16si)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_lzcnt_epi64 (__m512i __A)
+{
+ return (__m512i) __builtin_ia32_vplzcntq_512 ((__v8di) __A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_lzcnt_epi64 (__m512i __W, __mmask8 __U, __m512i __A)
+{
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
+ (__v8di)_mm512_lzcnt_epi64(__A),
+ (__v8di)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_lzcnt_epi64 (__mmask8 __U, __m512i __A)
+{
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
+ (__v8di)_mm512_lzcnt_epi64(__A),
+ (__v8di)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_broadcastmb_epi64 (__mmask8 __A)
+{
+ return (__m512i) _mm512_set1_epi64((long long) __A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_broadcastmw_epi32 (__mmask16 __A)
+{
+ return (__m512i) _mm512_set1_epi32((int) __A);
+
+}
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif
/*===---- avx512dqintrin.h - AVX512DQ intrinsics ---------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#define __AVX512DQINTRIN_H
/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS512 __attribute__((__always_inline__, __nodebug__, __target__("avx512dq"), __min_vector_width__(512)))
#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512dq")))
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline __mmask8 __DEFAULT_FN_ATTRS
+_knot_mask8(__mmask8 __M)
+{
+ return __builtin_ia32_knotqi(__M);
+}
+
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+_kand_mask8(__mmask8 __A, __mmask8 __B)
+{
+ return (__mmask8)__builtin_ia32_kandqi((__mmask8)__A, (__mmask8)__B);
+}
+
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+_kandn_mask8(__mmask8 __A, __mmask8 __B)
+{
+ return (__mmask8)__builtin_ia32_kandnqi((__mmask8)__A, (__mmask8)__B);
+}
+
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+_kor_mask8(__mmask8 __A, __mmask8 __B)
+{
+ return (__mmask8)__builtin_ia32_korqi((__mmask8)__A, (__mmask8)__B);
+}
+
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+_kxnor_mask8(__mmask8 __A, __mmask8 __B)
+{
+ return (__mmask8)__builtin_ia32_kxnorqi((__mmask8)__A, (__mmask8)__B);
+}
+
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+_kxor_mask8(__mmask8 __A, __mmask8 __B)
+{
+ return (__mmask8)__builtin_ia32_kxorqi((__mmask8)__A, (__mmask8)__B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_kortestc_mask8_u8(__mmask8 __A, __mmask8 __B)
+{
+ return (unsigned char)__builtin_ia32_kortestcqi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_kortestz_mask8_u8(__mmask8 __A, __mmask8 __B)
+{
+ return (unsigned char)__builtin_ia32_kortestzqi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_kortest_mask8_u8(__mmask8 __A, __mmask8 __B, unsigned char *__C) {
+ *__C = (unsigned char)__builtin_ia32_kortestcqi(__A, __B);
+ return (unsigned char)__builtin_ia32_kortestzqi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_ktestc_mask8_u8(__mmask8 __A, __mmask8 __B)
+{
+ return (unsigned char)__builtin_ia32_ktestcqi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_ktestz_mask8_u8(__mmask8 __A, __mmask8 __B)
+{
+ return (unsigned char)__builtin_ia32_ktestzqi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_ktest_mask8_u8(__mmask8 __A, __mmask8 __B, unsigned char *__C) {
+ *__C = (unsigned char)__builtin_ia32_ktestcqi(__A, __B);
+ return (unsigned char)__builtin_ia32_ktestzqi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_ktestc_mask16_u8(__mmask16 __A, __mmask16 __B)
+{
+ return (unsigned char)__builtin_ia32_ktestchi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_ktestz_mask16_u8(__mmask16 __A, __mmask16 __B)
+{
+ return (unsigned char)__builtin_ia32_ktestzhi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_ktest_mask16_u8(__mmask16 __A, __mmask16 __B, unsigned char *__C) {
+ *__C = (unsigned char)__builtin_ia32_ktestchi(__A, __B);
+ return (unsigned char)__builtin_ia32_ktestzhi(__A, __B);
+}
+
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+_kadd_mask8(__mmask8 __A, __mmask8 __B)
+{
+ return (__mmask8)__builtin_ia32_kaddqi((__mmask8)__A, (__mmask8)__B);
+}
+
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS
+_kadd_mask16(__mmask16 __A, __mmask16 __B)
+{
+ return (__mmask16)__builtin_ia32_kaddhi((__mmask16)__A, (__mmask16)__B);
+}
+
+#define _kshiftli_mask8(A, I) \
+ (__mmask8)__builtin_ia32_kshiftliqi((__mmask8)(A), (unsigned int)(I))
+
+#define _kshiftri_mask8(A, I) \
+ (__mmask8)__builtin_ia32_kshiftriqi((__mmask8)(A), (unsigned int)(I))
+
+static __inline__ unsigned int __DEFAULT_FN_ATTRS
+_cvtmask8_u32(__mmask8 __A) {
+ return (unsigned int)__builtin_ia32_kmovb((__mmask8)__A);
+}
+
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+_cvtu32_mask8(unsigned int __A) {
+ return (__mmask8)__builtin_ia32_kmovb((__mmask8)__A);
+}
+
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+_load_mask8(__mmask8 *__A) {
+ return (__mmask8)__builtin_ia32_kmovb(*(__mmask8 *)__A);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_store_mask8(__mmask8 *__A, __mmask8 __B) {
+ *(__mmask8 *)__A = __builtin_ia32_kmovb((__mmask8)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mullo_epi64 (__m512i __A, __m512i __B) {
return (__m512i) ((__v8du) __A * (__v8du) __B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_mullo_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B) {
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_mullo_epi64(__A, __B),
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_mullo_epi64(__mmask8 __U, __m512i __A, __m512i __B) {
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_mullo_epi64(__A, __B),
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_xor_pd(__m512d __A, __m512d __B) {
return (__m512d)((__v8du)__A ^ (__v8du)__B);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_xor_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_xor_pd(__A, __B),
(__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_xor_pd(__mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_xor_pd(__A, __B),
(__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_xor_ps (__m512 __A, __m512 __B) {
return (__m512)((__v16su)__A ^ (__v16su)__B);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_xor_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_xor_ps(__A, __B),
(__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_xor_ps(__mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_xor_ps(__A, __B),
(__v16sf)_mm512_setzero_ps());
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_or_pd(__m512d __A, __m512d __B) {
return (__m512d)((__v8du)__A | (__v8du)__B);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_or_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_or_pd(__A, __B),
(__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_or_pd(__mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_or_pd(__A, __B),
(__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_or_ps(__m512 __A, __m512 __B) {
return (__m512)((__v16su)__A | (__v16su)__B);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_or_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_or_ps(__A, __B),
(__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_or_ps(__mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_or_ps(__A, __B),
(__v16sf)_mm512_setzero_ps());
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_and_pd(__m512d __A, __m512d __B) {
return (__m512d)((__v8du)__A & (__v8du)__B);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_and_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_and_pd(__A, __B),
(__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_and_pd(__mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_and_pd(__A, __B),
(__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_and_ps(__m512 __A, __m512 __B) {
return (__m512)((__v16su)__A & (__v16su)__B);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_and_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_and_ps(__A, __B),
(__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_and_ps(__mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_and_ps(__A, __B),
(__v16sf)_mm512_setzero_ps());
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_andnot_pd(__m512d __A, __m512d __B) {
return (__m512d)(~(__v8du)__A & (__v8du)__B);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_andnot_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_andnot_pd(__A, __B),
(__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_andnot_pd(__mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_andnot_pd(__A, __B),
(__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_andnot_ps(__m512 __A, __m512 __B) {
return (__m512)(~(__v16su)__A & (__v16su)__B);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_andnot_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_andnot_ps(__A, __B),
(__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_andnot_ps(__mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_andnot_ps(__A, __B),
(__v16sf)_mm512_setzero_ps());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtpd_epi64 (__m512d __A) {
return (__m512i) __builtin_ia32_cvtpd2qq512_mask ((__v8df) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtpd_epi64 (__m512i __W, __mmask8 __U, __m512d __A) {
return (__m512i) __builtin_ia32_cvtpd2qq512_mask ((__v8df) __A,
(__v8di) __W,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtpd_epi64 (__mmask8 __U, __m512d __A) {
return (__m512i) __builtin_ia32_cvtpd2qq512_mask ((__v8df) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvt_roundpd_epi64(A, R) __extension__ ({ \
+#define _mm512_cvt_roundpd_epi64(A, R) \
(__m512i)__builtin_ia32_cvtpd2qq512_mask((__v8df)(__m512d)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvt_roundpd_epi64(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundpd_epi64(W, U, A, R) \
(__m512i)__builtin_ia32_cvtpd2qq512_mask((__v8df)(__m512d)(A), \
(__v8di)(__m512i)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_cvt_roundpd_epi64(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundpd_epi64(U, A, R) \
(__m512i)__builtin_ia32_cvtpd2qq512_mask((__v8df)(__m512d)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtpd_epu64 (__m512d __A) {
return (__m512i) __builtin_ia32_cvtpd2uqq512_mask ((__v8df) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtpd_epu64 (__m512i __W, __mmask8 __U, __m512d __A) {
return (__m512i) __builtin_ia32_cvtpd2uqq512_mask ((__v8df) __A,
(__v8di) __W,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtpd_epu64 (__mmask8 __U, __m512d __A) {
return (__m512i) __builtin_ia32_cvtpd2uqq512_mask ((__v8df) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvt_roundpd_epu64(A, R) __extension__ ({ \
+#define _mm512_cvt_roundpd_epu64(A, R) \
(__m512i)__builtin_ia32_cvtpd2uqq512_mask((__v8df)(__m512d)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvt_roundpd_epu64(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundpd_epu64(W, U, A, R) \
(__m512i)__builtin_ia32_cvtpd2uqq512_mask((__v8df)(__m512d)(A), \
(__v8di)(__m512i)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_cvt_roundpd_epu64(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundpd_epu64(U, A, R) \
(__m512i)__builtin_ia32_cvtpd2uqq512_mask((__v8df)(__m512d)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtps_epi64 (__m256 __A) {
return (__m512i) __builtin_ia32_cvtps2qq512_mask ((__v8sf) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtps_epi64 (__m512i __W, __mmask8 __U, __m256 __A) {
return (__m512i) __builtin_ia32_cvtps2qq512_mask ((__v8sf) __A,
(__v8di) __W,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtps_epi64 (__mmask8 __U, __m256 __A) {
return (__m512i) __builtin_ia32_cvtps2qq512_mask ((__v8sf) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvt_roundps_epi64(A, R) __extension__ ({ \
+#define _mm512_cvt_roundps_epi64(A, R) \
(__m512i)__builtin_ia32_cvtps2qq512_mask((__v8sf)(__m256)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvt_roundps_epi64(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundps_epi64(W, U, A, R) \
(__m512i)__builtin_ia32_cvtps2qq512_mask((__v8sf)(__m256)(A), \
(__v8di)(__m512i)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_cvt_roundps_epi64(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundps_epi64(U, A, R) \
(__m512i)__builtin_ia32_cvtps2qq512_mask((__v8sf)(__m256)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtps_epu64 (__m256 __A) {
return (__m512i) __builtin_ia32_cvtps2uqq512_mask ((__v8sf) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtps_epu64 (__m512i __W, __mmask8 __U, __m256 __A) {
return (__m512i) __builtin_ia32_cvtps2uqq512_mask ((__v8sf) __A,
(__v8di) __W,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtps_epu64 (__mmask8 __U, __m256 __A) {
return (__m512i) __builtin_ia32_cvtps2uqq512_mask ((__v8sf) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvt_roundps_epu64(A, R) __extension__ ({ \
+#define _mm512_cvt_roundps_epu64(A, R) \
(__m512i)__builtin_ia32_cvtps2uqq512_mask((__v8sf)(__m256)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvt_roundps_epu64(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundps_epu64(W, U, A, R) \
(__m512i)__builtin_ia32_cvtps2uqq512_mask((__v8sf)(__m256)(A), \
(__v8di)(__m512i)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_cvt_roundps_epu64(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundps_epu64(U, A, R) \
(__m512i)__builtin_ia32_cvtps2uqq512_mask((__v8sf)(__m256)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_cvtepi64_pd (__m512i __A) {
- return (__m512d) __builtin_ia32_cvtqq2pd512_mask ((__v8di) __A,
- (__v8df) _mm512_setzero_pd(),
- (__mmask8) -1,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_convertvector((__v8di)__A, __v8df);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi64_pd (__m512d __W, __mmask8 __U, __m512i __A) {
- return (__m512d) __builtin_ia32_cvtqq2pd512_mask ((__v8di) __A,
- (__v8df) __W,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
+ (__v8df)_mm512_cvtepi64_pd(__A),
+ (__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi64_pd (__mmask8 __U, __m512i __A) {
- return (__m512d) __builtin_ia32_cvtqq2pd512_mask ((__v8di) __A,
- (__v8df) _mm512_setzero_pd(),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
+ (__v8df)_mm512_cvtepi64_pd(__A),
+ (__v8df)_mm512_setzero_pd());
}
-#define _mm512_cvt_roundepi64_pd(A, R) __extension__ ({ \
+#define _mm512_cvt_roundepi64_pd(A, R) \
(__m512d)__builtin_ia32_cvtqq2pd512_mask((__v8di)(__m512i)(A), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvt_roundepi64_pd(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundepi64_pd(W, U, A, R) \
(__m512d)__builtin_ia32_cvtqq2pd512_mask((__v8di)(__m512i)(A), \
(__v8df)(__m512d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_cvt_roundepi64_pd(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundepi64_pd(U, A, R) \
(__m512d)__builtin_ia32_cvtqq2pd512_mask((__v8di)(__m512i)(A), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS512
_mm512_cvtepi64_ps (__m512i __A) {
return (__m256) __builtin_ia32_cvtqq2ps512_mask ((__v8di) __A,
(__v8sf) _mm256_setzero_ps(),
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi64_ps (__m256 __W, __mmask8 __U, __m512i __A) {
return (__m256) __builtin_ia32_cvtqq2ps512_mask ((__v8di) __A,
(__v8sf) __W,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi64_ps (__mmask8 __U, __m512i __A) {
return (__m256) __builtin_ia32_cvtqq2ps512_mask ((__v8di) __A,
(__v8sf) _mm256_setzero_ps(),
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvt_roundepi64_ps(A, R) __extension__ ({ \
+#define _mm512_cvt_roundepi64_ps(A, R) \
(__m256)__builtin_ia32_cvtqq2ps512_mask((__v8di)(__m512i)(A), \
(__v8sf)_mm256_setzero_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvt_roundepi64_ps(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundepi64_ps(W, U, A, R) \
(__m256)__builtin_ia32_cvtqq2ps512_mask((__v8di)(__m512i)(A), \
(__v8sf)(__m256)(W), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_maskz_cvt_roundepi64_ps(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundepi64_ps(U, A, R) \
(__m256)__builtin_ia32_cvtqq2ps512_mask((__v8di)(__m512i)(A), \
(__v8sf)_mm256_setzero_ps(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvttpd_epi64 (__m512d __A) {
return (__m512i) __builtin_ia32_cvttpd2qq512_mask ((__v8df) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvttpd_epi64 (__m512i __W, __mmask8 __U, __m512d __A) {
return (__m512i) __builtin_ia32_cvttpd2qq512_mask ((__v8df) __A,
(__v8di) __W,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvttpd_epi64 (__mmask8 __U, __m512d __A) {
return (__m512i) __builtin_ia32_cvttpd2qq512_mask ((__v8df) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvtt_roundpd_epi64(A, R) __extension__ ({ \
+#define _mm512_cvtt_roundpd_epi64(A, R) \
(__m512i)__builtin_ia32_cvttpd2qq512_mask((__v8df)(__m512d)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvtt_roundpd_epi64(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvtt_roundpd_epi64(W, U, A, R) \
(__m512i)__builtin_ia32_cvttpd2qq512_mask((__v8df)(__m512d)(A), \
(__v8di)(__m512i)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_cvtt_roundpd_epi64(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvtt_roundpd_epi64(U, A, R) \
(__m512i)__builtin_ia32_cvttpd2qq512_mask((__v8df)(__m512d)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvttpd_epu64 (__m512d __A) {
return (__m512i) __builtin_ia32_cvttpd2uqq512_mask ((__v8df) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvttpd_epu64 (__m512i __W, __mmask8 __U, __m512d __A) {
return (__m512i) __builtin_ia32_cvttpd2uqq512_mask ((__v8df) __A,
(__v8di) __W,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvttpd_epu64 (__mmask8 __U, __m512d __A) {
return (__m512i) __builtin_ia32_cvttpd2uqq512_mask ((__v8df) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvtt_roundpd_epu64(A, R) __extension__ ({ \
+#define _mm512_cvtt_roundpd_epu64(A, R) \
(__m512i)__builtin_ia32_cvttpd2uqq512_mask((__v8df)(__m512d)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvtt_roundpd_epu64(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvtt_roundpd_epu64(W, U, A, R) \
(__m512i)__builtin_ia32_cvttpd2uqq512_mask((__v8df)(__m512d)(A), \
(__v8di)(__m512i)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_cvtt_roundpd_epu64(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvtt_roundpd_epu64(U, A, R) \
(__m512i)__builtin_ia32_cvttpd2uqq512_mask((__v8df)(__m512d)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvttps_epi64 (__m256 __A) {
return (__m512i) __builtin_ia32_cvttps2qq512_mask ((__v8sf) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvttps_epi64 (__m512i __W, __mmask8 __U, __m256 __A) {
return (__m512i) __builtin_ia32_cvttps2qq512_mask ((__v8sf) __A,
(__v8di) __W,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvttps_epi64 (__mmask8 __U, __m256 __A) {
return (__m512i) __builtin_ia32_cvttps2qq512_mask ((__v8sf) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvtt_roundps_epi64(A, R) __extension__ ({ \
+#define _mm512_cvtt_roundps_epi64(A, R) \
(__m512i)__builtin_ia32_cvttps2qq512_mask((__v8sf)(__m256)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvtt_roundps_epi64(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvtt_roundps_epi64(W, U, A, R) \
(__m512i)__builtin_ia32_cvttps2qq512_mask((__v8sf)(__m256)(A), \
(__v8di)(__m512i)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_cvtt_roundps_epi64(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvtt_roundps_epi64(U, A, R) \
(__m512i)__builtin_ia32_cvttps2qq512_mask((__v8sf)(__m256)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvttps_epu64 (__m256 __A) {
return (__m512i) __builtin_ia32_cvttps2uqq512_mask ((__v8sf) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvttps_epu64 (__m512i __W, __mmask8 __U, __m256 __A) {
return (__m512i) __builtin_ia32_cvttps2uqq512_mask ((__v8sf) __A,
(__v8di) __W,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvttps_epu64 (__mmask8 __U, __m256 __A) {
return (__m512i) __builtin_ia32_cvttps2uqq512_mask ((__v8sf) __A,
(__v8di) _mm512_setzero_si512(),
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvtt_roundps_epu64(A, R) __extension__ ({ \
+#define _mm512_cvtt_roundps_epu64(A, R) \
(__m512i)__builtin_ia32_cvttps2uqq512_mask((__v8sf)(__m256)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvtt_roundps_epu64(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvtt_roundps_epu64(W, U, A, R) \
(__m512i)__builtin_ia32_cvttps2uqq512_mask((__v8sf)(__m256)(A), \
(__v8di)(__m512i)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_cvtt_roundps_epu64(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvtt_roundps_epu64(U, A, R) \
(__m512i)__builtin_ia32_cvttps2uqq512_mask((__v8sf)(__m256)(A), \
(__v8di)_mm512_setzero_si512(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_cvtepu64_pd (__m512i __A) {
- return (__m512d) __builtin_ia32_cvtuqq2pd512_mask ((__v8di) __A,
- (__v8df) _mm512_setzero_pd(),
- (__mmask8) -1,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_convertvector((__v8du)__A, __v8df);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu64_pd (__m512d __W, __mmask8 __U, __m512i __A) {
- return (__m512d) __builtin_ia32_cvtuqq2pd512_mask ((__v8di) __A,
- (__v8df) __W,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
+ (__v8df)_mm512_cvtepu64_pd(__A),
+ (__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu64_pd (__mmask8 __U, __m512i __A) {
- return (__m512d) __builtin_ia32_cvtuqq2pd512_mask ((__v8di) __A,
- (__v8df) _mm512_setzero_pd(),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
+ (__v8df)_mm512_cvtepu64_pd(__A),
+ (__v8df)_mm512_setzero_pd());
}
-#define _mm512_cvt_roundepu64_pd(A, R) __extension__ ({ \
+#define _mm512_cvt_roundepu64_pd(A, R) \
(__m512d)__builtin_ia32_cvtuqq2pd512_mask((__v8di)(__m512i)(A), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvt_roundepu64_pd(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundepu64_pd(W, U, A, R) \
(__m512d)__builtin_ia32_cvtuqq2pd512_mask((__v8di)(__m512i)(A), \
(__v8df)(__m512d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_cvt_roundepu64_pd(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundepu64_pd(U, A, R) \
(__m512d)__builtin_ia32_cvtuqq2pd512_mask((__v8di)(__m512i)(A), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS512
_mm512_cvtepu64_ps (__m512i __A) {
return (__m256) __builtin_ia32_cvtuqq2ps512_mask ((__v8di) __A,
(__v8sf) _mm256_setzero_ps(),
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu64_ps (__m256 __W, __mmask8 __U, __m512i __A) {
return (__m256) __builtin_ia32_cvtuqq2ps512_mask ((__v8di) __A,
(__v8sf) __W,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu64_ps (__mmask8 __U, __m512i __A) {
return (__m256) __builtin_ia32_cvtuqq2ps512_mask ((__v8di) __A,
(__v8sf) _mm256_setzero_ps(),
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvt_roundepu64_ps(A, R) __extension__ ({ \
+#define _mm512_cvt_roundepu64_ps(A, R) \
(__m256)__builtin_ia32_cvtuqq2ps512_mask((__v8di)(__m512i)(A), \
(__v8sf)_mm256_setzero_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvt_roundepu64_ps(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundepu64_ps(W, U, A, R) \
(__m256)__builtin_ia32_cvtuqq2ps512_mask((__v8di)(__m512i)(A), \
(__v8sf)(__m256)(W), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_maskz_cvt_roundepu64_ps(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundepu64_ps(U, A, R) \
(__m256)__builtin_ia32_cvtuqq2ps512_mask((__v8di)(__m512i)(A), \
(__v8sf)_mm256_setzero_ps(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_range_pd(A, B, C) __extension__ ({ \
+#define _mm512_range_pd(A, B, C) \
(__m512d)__builtin_ia32_rangepd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), (int)(C), \
(__v8df)_mm512_setzero_pd(), \
(__mmask8)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_mask_range_pd(W, U, A, B, C) __extension__ ({ \
+#define _mm512_mask_range_pd(W, U, A, B, C) \
(__m512d)__builtin_ia32_rangepd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), (int)(C), \
(__v8df)(__m512d)(W), (__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_maskz_range_pd(U, A, B, C) __extension__ ({ \
+#define _mm512_maskz_range_pd(U, A, B, C) \
(__m512d)__builtin_ia32_rangepd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), (int)(C), \
(__v8df)_mm512_setzero_pd(), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_range_round_pd(A, B, C, R) __extension__ ({ \
+#define _mm512_range_round_pd(A, B, C, R) \
(__m512d)__builtin_ia32_rangepd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), (int)(C), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_range_round_pd(W, U, A, B, C, R) __extension__ ({ \
+#define _mm512_mask_range_round_pd(W, U, A, B, C, R) \
(__m512d)__builtin_ia32_rangepd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), (int)(C), \
(__v8df)(__m512d)(W), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_maskz_range_round_pd(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_range_round_pd(U, A, B, C, R) \
(__m512d)__builtin_ia32_rangepd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), (int)(C), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_range_ps(A, B, C) __extension__ ({ \
+#define _mm512_range_ps(A, B, C) \
(__m512)__builtin_ia32_rangeps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), (int)(C), \
(__v16sf)_mm512_setzero_ps(), \
(__mmask16)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_mask_range_ps(W, U, A, B, C) __extension__ ({ \
+#define _mm512_mask_range_ps(W, U, A, B, C) \
(__m512)__builtin_ia32_rangeps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), (int)(C), \
(__v16sf)(__m512)(W), (__mmask16)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_maskz_range_ps(U, A, B, C) __extension__ ({ \
+#define _mm512_maskz_range_ps(U, A, B, C) \
(__m512)__builtin_ia32_rangeps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), (int)(C), \
(__v16sf)_mm512_setzero_ps(), \
(__mmask16)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_range_round_ps(A, B, C, R) __extension__ ({ \
+#define _mm512_range_round_ps(A, B, C, R) \
(__m512)__builtin_ia32_rangeps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), (int)(C), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_range_round_ps(W, U, A, B, C, R) __extension__ ({ \
+#define _mm512_mask_range_round_ps(W, U, A, B, C, R) \
(__m512)__builtin_ia32_rangeps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), (int)(C), \
(__v16sf)(__m512)(W), (__mmask16)(U), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_maskz_range_round_ps(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_range_round_ps(U, A, B, C, R) \
(__m512)__builtin_ia32_rangeps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), (int)(C), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm_range_round_ss(A, B, C, R) __extension__ ({ \
+#define _mm_range_round_ss(A, B, C, R) \
(__m128)__builtin_ia32_rangess128_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
(__mmask8) -1, (int)(C),\
- (int)(R)); })
+ (int)(R))
#define _mm_range_ss(A ,B , C) _mm_range_round_ss(A, B, C ,_MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_range_round_ss(W, U, A, B, C, R) __extension__ ({ \
+#define _mm_mask_range_round_ss(W, U, A, B, C, R) \
(__m128)__builtin_ia32_rangess128_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(W),\
(__mmask8)(U), (int)(C),\
- (int)(R)); })
+ (int)(R))
#define _mm_mask_range_ss(W , U, A, B, C) _mm_mask_range_round_ss(W, U, A, B, C , _MM_FROUND_CUR_DIRECTION)
-#define _mm_maskz_range_round_ss(U, A, B, C, R) __extension__ ({ \
+#define _mm_maskz_range_round_ss(U, A, B, C, R) \
(__m128)__builtin_ia32_rangess128_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
(__mmask8)(U), (int)(C),\
- (int)(R)); })
+ (int)(R))
#define _mm_maskz_range_ss(U, A ,B , C) _mm_maskz_range_round_ss(U, A, B, C ,_MM_FROUND_CUR_DIRECTION)
-#define _mm_range_round_sd(A, B, C, R) __extension__ ({ \
+#define _mm_range_round_sd(A, B, C, R) \
(__m128d)__builtin_ia32_rangesd128_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
(__mmask8) -1, (int)(C),\
- (int)(R)); })
+ (int)(R))
#define _mm_range_sd(A ,B , C) _mm_range_round_sd(A, B, C ,_MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_range_round_sd(W, U, A, B, C, R) __extension__ ({ \
+#define _mm_mask_range_round_sd(W, U, A, B, C, R) \
(__m128d)__builtin_ia32_rangesd128_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(W),\
(__mmask8)(U), (int)(C),\
- (int)(R)); })
+ (int)(R))
#define _mm_mask_range_sd(W, U, A, B, C) _mm_mask_range_round_sd(W, U, A, B, C ,_MM_FROUND_CUR_DIRECTION)
-#define _mm_maskz_range_round_sd(U, A, B, C, R) __extension__ ({ \
+#define _mm_maskz_range_round_sd(U, A, B, C, R) \
(__m128d)__builtin_ia32_rangesd128_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
(__mmask8)(U), (int)(C),\
- (int)(R)); })
+ (int)(R))
#define _mm_maskz_range_sd(U, A, B, C) _mm_maskz_range_round_sd(U, A, B, C ,_MM_FROUND_CUR_DIRECTION)
-#define _mm512_reduce_pd(A, B) __extension__ ({ \
+#define _mm512_reduce_pd(A, B) \
(__m512d)__builtin_ia32_reducepd512_mask((__v8df)(__m512d)(A), (int)(B), \
(__v8df)_mm512_setzero_pd(), \
(__mmask8)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_mask_reduce_pd(W, U, A, B) __extension__ ({ \
+#define _mm512_mask_reduce_pd(W, U, A, B) \
(__m512d)__builtin_ia32_reducepd512_mask((__v8df)(__m512d)(A), (int)(B), \
(__v8df)(__m512d)(W), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_maskz_reduce_pd(U, A, B) __extension__ ({ \
+#define _mm512_maskz_reduce_pd(U, A, B) \
(__m512d)__builtin_ia32_reducepd512_mask((__v8df)(__m512d)(A), (int)(B), \
(__v8df)_mm512_setzero_pd(), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_reduce_ps(A, B) __extension__ ({ \
+#define _mm512_reduce_ps(A, B) \
(__m512)__builtin_ia32_reduceps512_mask((__v16sf)(__m512)(A), (int)(B), \
(__v16sf)_mm512_setzero_ps(), \
(__mmask16)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_mask_reduce_ps(W, U, A, B) __extension__ ({ \
+#define _mm512_mask_reduce_ps(W, U, A, B) \
(__m512)__builtin_ia32_reduceps512_mask((__v16sf)(__m512)(A), (int)(B), \
(__v16sf)(__m512)(W), \
(__mmask16)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_maskz_reduce_ps(U, A, B) __extension__ ({ \
+#define _mm512_maskz_reduce_ps(U, A, B) \
(__m512)__builtin_ia32_reduceps512_mask((__v16sf)(__m512)(A), (int)(B), \
(__v16sf)_mm512_setzero_ps(), \
(__mmask16)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_reduce_round_pd(A, B, R) __extension__ ({\
+#define _mm512_reduce_round_pd(A, B, R) \
(__m512d)__builtin_ia32_reducepd512_mask((__v8df)(__m512d)(A), (int)(B), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_reduce_round_pd(W, U, A, B, R) __extension__ ({\
+#define _mm512_mask_reduce_round_pd(W, U, A, B, R) \
(__m512d)__builtin_ia32_reducepd512_mask((__v8df)(__m512d)(A), (int)(B), \
(__v8df)(__m512d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_reduce_round_pd(U, A, B, R) __extension__ ({\
+#define _mm512_maskz_reduce_round_pd(U, A, B, R) \
(__m512d)__builtin_ia32_reducepd512_mask((__v8df)(__m512d)(A), (int)(B), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_reduce_round_ps(A, B, R) __extension__ ({\
+#define _mm512_reduce_round_ps(A, B, R) \
(__m512)__builtin_ia32_reduceps512_mask((__v16sf)(__m512)(A), (int)(B), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_reduce_round_ps(W, U, A, B, R) __extension__ ({\
+#define _mm512_mask_reduce_round_ps(W, U, A, B, R) \
(__m512)__builtin_ia32_reduceps512_mask((__v16sf)(__m512)(A), (int)(B), \
(__v16sf)(__m512)(W), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_reduce_round_ps(U, A, B, R) __extension__ ({\
+#define _mm512_maskz_reduce_round_ps(U, A, B, R) \
(__m512)__builtin_ia32_reduceps512_mask((__v16sf)(__m512)(A), (int)(B), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm_reduce_ss(A, B, C) __extension__ ({ \
+#define _mm_reduce_ss(A, B, C) \
(__m128)__builtin_ia32_reducess_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), (__mmask8)-1, \
- (int)(C), _MM_FROUND_CUR_DIRECTION); })
+ (int)(C), _MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_reduce_ss(W, U, A, B, C) __extension__ ({ \
+#define _mm_mask_reduce_ss(W, U, A, B, C) \
(__m128)__builtin_ia32_reducess_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(W), (__mmask8)(U), \
- (int)(C), _MM_FROUND_CUR_DIRECTION); })
+ (int)(C), _MM_FROUND_CUR_DIRECTION)
-#define _mm_maskz_reduce_ss(U, A, B, C) __extension__ ({ \
+#define _mm_maskz_reduce_ss(U, A, B, C) \
(__m128)__builtin_ia32_reducess_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
(__mmask8)(U), (int)(C), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_reduce_round_ss(A, B, C, R) __extension__ ({ \
+#define _mm_reduce_round_ss(A, B, C, R) \
(__m128)__builtin_ia32_reducess_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), (__mmask8)-1, \
- (int)(C), (int)(R)); })
+ (int)(C), (int)(R))
-#define _mm_mask_reduce_round_ss(W, U, A, B, C, R) __extension__ ({ \
+#define _mm_mask_reduce_round_ss(W, U, A, B, C, R) \
(__m128)__builtin_ia32_reducess_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(W), (__mmask8)(U), \
- (int)(C), (int)(R)); })
+ (int)(C), (int)(R))
-#define _mm_maskz_reduce_round_ss(U, A, B, C, R) __extension__ ({ \
+#define _mm_maskz_reduce_round_ss(U, A, B, C, R) \
(__m128)__builtin_ia32_reducess_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(U), (int)(C), (int)(R)); })
+ (__mmask8)(U), (int)(C), (int)(R))
-#define _mm_reduce_sd(A, B, C) __extension__ ({ \
+#define _mm_reduce_sd(A, B, C) \
(__m128d)__builtin_ia32_reducesd_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
(__mmask8)-1, (int)(C), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_reduce_sd(W, U, A, B, C) __extension__ ({ \
+#define _mm_mask_reduce_sd(W, U, A, B, C) \
(__m128d)__builtin_ia32_reducesd_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(W), (__mmask8)(U), \
- (int)(C), _MM_FROUND_CUR_DIRECTION); })
+ (int)(C), _MM_FROUND_CUR_DIRECTION)
-#define _mm_maskz_reduce_sd(U, A, B, C) __extension__ ({ \
+#define _mm_maskz_reduce_sd(U, A, B, C) \
(__m128d)__builtin_ia32_reducesd_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
(__mmask8)(U), (int)(C), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_reduce_round_sd(A, B, C, R) __extension__ ({ \
+#define _mm_reduce_round_sd(A, B, C, R) \
(__m128d)__builtin_ia32_reducesd_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1, (int)(C), (int)(R)); })
+ (__mmask8)-1, (int)(C), (int)(R))
-#define _mm_mask_reduce_round_sd(W, U, A, B, C, R) __extension__ ({ \
+#define _mm_mask_reduce_round_sd(W, U, A, B, C, R) \
(__m128d)__builtin_ia32_reducesd_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(W), (__mmask8)(U), \
- (int)(C), (int)(R)); })
+ (int)(C), (int)(R))
-#define _mm_maskz_reduce_round_sd(U, A, B, C, R) __extension__ ({ \
+#define _mm_maskz_reduce_round_sd(U, A, B, C, R) \
(__m128d)__builtin_ia32_reducesd_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U), (int)(C), (int)(R)); })
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
+ (__mmask8)(U), (int)(C), (int)(R))
+
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS512
_mm512_movepi32_mask (__m512i __A)
{
return (__mmask16) __builtin_ia32_cvtd2mask512 ((__v16si) __A);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_movm_epi32 (__mmask16 __A)
{
return (__m512i) __builtin_ia32_cvtmask2d512 (__A);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_movm_epi64 (__mmask8 __A)
{
return (__m512i) __builtin_ia32_cvtmask2q512 (__A);
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS512
_mm512_movepi64_mask (__m512i __A)
{
return (__mmask8) __builtin_ia32_cvtq2mask512 ((__v8di) __A);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_broadcast_f32x2 (__m128 __A)
{
- return (__m512) __builtin_ia32_broadcastf32x2_512_mask ((__v4sf) __A,
- (__v16sf)_mm512_undefined_ps(),
- (__mmask16) -1);
+ return (__m512)__builtin_shufflevector((__v4sf)__A, (__v4sf)__A,
+ 0, 1, 0, 1, 0, 1, 0, 1,
+ 0, 1, 0, 1, 0, 1, 0, 1);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_broadcast_f32x2 (__m512 __O, __mmask16 __M, __m128 __A)
{
- return (__m512) __builtin_ia32_broadcastf32x2_512_mask ((__v4sf) __A,
- (__v16sf)
- __O, __M);
+ return (__m512)__builtin_ia32_selectps_512((__mmask16)__M,
+ (__v16sf)_mm512_broadcast_f32x2(__A),
+ (__v16sf)__O);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcast_f32x2 (__mmask16 __M, __m128 __A)
{
- return (__m512) __builtin_ia32_broadcastf32x2_512_mask ((__v4sf) __A,
- (__v16sf)_mm512_setzero_ps (),
- __M);
+ return (__m512)__builtin_ia32_selectps_512((__mmask16)__M,
+ (__v16sf)_mm512_broadcast_f32x2(__A),
+ (__v16sf)_mm512_setzero_ps());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_broadcast_f32x8(__m256 __A)
{
return (__m512)__builtin_shufflevector((__v8sf)__A, (__v8sf)__A,
0, 1, 2, 3, 4, 5, 6, 7);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_broadcast_f32x8(__m512 __O, __mmask16 __M, __m256 __A)
{
- return (__m512)__builtin_ia32_selectps_512((__mmask8)__M,
+ return (__m512)__builtin_ia32_selectps_512((__mmask16)__M,
(__v16sf)_mm512_broadcast_f32x8(__A),
(__v16sf)__O);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcast_f32x8(__mmask16 __M, __m256 __A)
{
- return (__m512)__builtin_ia32_selectps_512((__mmask8)__M,
+ return (__m512)__builtin_ia32_selectps_512((__mmask16)__M,
(__v16sf)_mm512_broadcast_f32x8(__A),
(__v16sf)_mm512_setzero_ps());
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_broadcast_f64x2(__m128d __A)
{
return (__m512d)__builtin_shufflevector((__v2df)__A, (__v2df)__A,
0, 1, 0, 1, 0, 1, 0, 1);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_broadcast_f64x2(__m512d __O, __mmask8 __M, __m128d __A)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__M,
(__v8df)__O);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcast_f64x2(__mmask8 __M, __m128d __A)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__M,
(__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_broadcast_i32x2 (__m128i __A)
{
- return (__m512i) __builtin_ia32_broadcasti32x2_512_mask ((__v4si) __A,
- (__v16si)_mm512_setzero_si512(),
- (__mmask16) -1);
+ return (__m512i)__builtin_shufflevector((__v4si)__A, (__v4si)__A,
+ 0, 1, 0, 1, 0, 1, 0, 1,
+ 0, 1, 0, 1, 0, 1, 0, 1);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_broadcast_i32x2 (__m512i __O, __mmask16 __M, __m128i __A)
{
- return (__m512i) __builtin_ia32_broadcasti32x2_512_mask ((__v4si) __A,
- (__v16si)
- __O, __M);
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
+ (__v16si)_mm512_broadcast_i32x2(__A),
+ (__v16si)__O);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcast_i32x2 (__mmask16 __M, __m128i __A)
{
- return (__m512i) __builtin_ia32_broadcasti32x2_512_mask ((__v4si) __A,
- (__v16si)_mm512_setzero_si512 (),
- __M);
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
+ (__v16si)_mm512_broadcast_i32x2(__A),
+ (__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_broadcast_i32x8(__m256i __A)
{
return (__m512i)__builtin_shufflevector((__v8si)__A, (__v8si)__A,
0, 1, 2, 3, 4, 5, 6, 7);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_broadcast_i32x8(__m512i __O, __mmask16 __M, __m256i __A)
{
- return (__m512i)__builtin_ia32_selectd_512((__mmask8)__M,
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
(__v16si)_mm512_broadcast_i32x8(__A),
(__v16si)__O);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcast_i32x8(__mmask16 __M, __m256i __A)
{
- return (__m512i)__builtin_ia32_selectd_512((__mmask8)__M,
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
(__v16si)_mm512_broadcast_i32x8(__A),
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_broadcast_i64x2(__m128i __A)
{
return (__m512i)__builtin_shufflevector((__v2di)__A, (__v2di)__A,
0, 1, 0, 1, 0, 1, 0, 1);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_broadcast_i64x2(__m512i __O, __mmask8 __M, __m128i __A)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
(__v8di)__O);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcast_i64x2(__mmask8 __M, __m128i __A)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
(__v8di)_mm512_setzero_si512());
}
-#define _mm512_extractf32x8_ps(A, imm) __extension__ ({ \
- (__m256)__builtin_shufflevector((__v16sf)(__m512)(A), \
- (__v16sf)_mm512_undefined_ps(), \
- ((imm) & 1) ? 8 : 0, \
- ((imm) & 1) ? 9 : 1, \
- ((imm) & 1) ? 10 : 2, \
- ((imm) & 1) ? 11 : 3, \
- ((imm) & 1) ? 12 : 4, \
- ((imm) & 1) ? 13 : 5, \
- ((imm) & 1) ? 14 : 6, \
- ((imm) & 1) ? 15 : 7); })
-
-#define _mm512_mask_extractf32x8_ps(W, U, A, imm) __extension__ ({ \
- (__m256)__builtin_ia32_selectps_256((__mmask8)(U), \
- (__v8sf)_mm512_extractf32x8_ps((A), (imm)), \
- (__v8sf)(W)); })
-
-#define _mm512_maskz_extractf32x8_ps(U, A, imm) __extension__ ({ \
- (__m256)__builtin_ia32_selectps_256((__mmask8)(U), \
- (__v8sf)_mm512_extractf32x8_ps((A), (imm)), \
- (__v8sf)_mm256_setzero_ps()); })
-
-#define _mm512_extractf64x2_pd(A, imm) __extension__ ({ \
- (__m128d)__builtin_shufflevector((__v8df)(__m512d)(A), \
- (__v8df)_mm512_undefined_pd(), \
- 0 + ((imm) & 0x3) * 2, \
- 1 + ((imm) & 0x3) * 2); })
-
-#define _mm512_mask_extractf64x2_pd(W, U, A, imm) __extension__ ({ \
- (__m128d)__builtin_ia32_selectpd_128((__mmask8)(U), \
- (__v2df)_mm512_extractf64x2_pd((A), (imm)), \
- (__v2df)(W)); })
-
-#define _mm512_maskz_extractf64x2_pd(U, A, imm) __extension__ ({ \
- (__m128d)__builtin_ia32_selectpd_128((__mmask8)(U), \
- (__v2df)_mm512_extractf64x2_pd((A), (imm)), \
- (__v2df)_mm_setzero_pd()); })
-
-#define _mm512_extracti32x8_epi32(A, imm) __extension__ ({ \
- (__m256i)__builtin_shufflevector((__v16si)(__m512i)(A), \
- (__v16si)_mm512_undefined_epi32(), \
- ((imm) & 1) ? 8 : 0, \
- ((imm) & 1) ? 9 : 1, \
- ((imm) & 1) ? 10 : 2, \
- ((imm) & 1) ? 11 : 3, \
- ((imm) & 1) ? 12 : 4, \
- ((imm) & 1) ? 13 : 5, \
- ((imm) & 1) ? 14 : 6, \
- ((imm) & 1) ? 15 : 7); })
-
-#define _mm512_mask_extracti32x8_epi32(W, U, A, imm) __extension__ ({ \
- (__m256i)__builtin_ia32_selectd_256((__mmask8)(U), \
- (__v8si)_mm512_extracti32x8_epi32((A), (imm)), \
- (__v8si)(W)); })
-
-#define _mm512_maskz_extracti32x8_epi32(U, A, imm) __extension__ ({ \
- (__m256i)__builtin_ia32_selectd_256((__mmask8)(U), \
- (__v8si)_mm512_extracti32x8_epi32((A), (imm)), \
- (__v8si)_mm256_setzero_si256()); })
-
-#define _mm512_extracti64x2_epi64(A, imm) __extension__ ({ \
- (__m128i)__builtin_shufflevector((__v8di)(__m512i)(A), \
- (__v8di)_mm512_undefined_epi32(), \
- 0 + ((imm) & 0x3) * 2, \
- 1 + ((imm) & 0x3) * 2); })
-
-#define _mm512_mask_extracti64x2_epi64(W, U, A, imm) __extension__ ({ \
- (__m128d)__builtin_ia32_selectq_128((__mmask8)(U), \
- (__v2di)_mm512_extracti64x2_epi64((A), (imm)), \
- (__v2di)(W)); })
-
-#define _mm512_maskz_extracti64x2_epi64(U, A, imm) __extension__ ({ \
- (__m128d)__builtin_ia32_selectq_128((__mmask8)(U), \
- (__v2di)_mm512_extracti64x2_epi64((A), (imm)), \
- (__v2di)_mm_setzero_di()); })
-
-#define _mm512_insertf32x8(A, B, imm) __extension__ ({ \
- (__m512)__builtin_shufflevector((__v16sf)(__m512)(A), \
- (__v16sf)_mm512_castps256_ps512((__m256)(B)),\
- ((imm) & 0x1) ? 0 : 16, \
- ((imm) & 0x1) ? 1 : 17, \
- ((imm) & 0x1) ? 2 : 18, \
- ((imm) & 0x1) ? 3 : 19, \
- ((imm) & 0x1) ? 4 : 20, \
- ((imm) & 0x1) ? 5 : 21, \
- ((imm) & 0x1) ? 6 : 22, \
- ((imm) & 0x1) ? 7 : 23, \
- ((imm) & 0x1) ? 16 : 8, \
- ((imm) & 0x1) ? 17 : 9, \
- ((imm) & 0x1) ? 18 : 10, \
- ((imm) & 0x1) ? 19 : 11, \
- ((imm) & 0x1) ? 20 : 12, \
- ((imm) & 0x1) ? 21 : 13, \
- ((imm) & 0x1) ? 22 : 14, \
- ((imm) & 0x1) ? 23 : 15); })
-
-#define _mm512_mask_insertf32x8(W, U, A, B, imm) __extension__ ({ \
+#define _mm512_extractf32x8_ps(A, imm) \
+ (__m256)__builtin_ia32_extractf32x8_mask((__v16sf)(__m512)(A), (int)(imm), \
+ (__v8sf)_mm256_undefined_ps(), \
+ (__mmask8)-1)
+
+#define _mm512_mask_extractf32x8_ps(W, U, A, imm) \
+ (__m256)__builtin_ia32_extractf32x8_mask((__v16sf)(__m512)(A), (int)(imm), \
+ (__v8sf)(__m256)(W), \
+ (__mmask8)(U))
+
+#define _mm512_maskz_extractf32x8_ps(U, A, imm) \
+ (__m256)__builtin_ia32_extractf32x8_mask((__v16sf)(__m512)(A), (int)(imm), \
+ (__v8sf)_mm256_setzero_ps(), \
+ (__mmask8)(U))
+
+#define _mm512_extractf64x2_pd(A, imm) \
+ (__m128d)__builtin_ia32_extractf64x2_512_mask((__v8df)(__m512d)(A), \
+ (int)(imm), \
+ (__v2df)_mm_undefined_pd(), \
+ (__mmask8)-1)
+
+#define _mm512_mask_extractf64x2_pd(W, U, A, imm) \
+ (__m128d)__builtin_ia32_extractf64x2_512_mask((__v8df)(__m512d)(A), \
+ (int)(imm), \
+ (__v2df)(__m128d)(W), \
+ (__mmask8)(U))
+
+#define _mm512_maskz_extractf64x2_pd(U, A, imm) \
+ (__m128d)__builtin_ia32_extractf64x2_512_mask((__v8df)(__m512d)(A), \
+ (int)(imm), \
+ (__v2df)_mm_setzero_pd(), \
+ (__mmask8)(U))
+
+#define _mm512_extracti32x8_epi32(A, imm) \
+ (__m256i)__builtin_ia32_extracti32x8_mask((__v16si)(__m512i)(A), (int)(imm), \
+ (__v8si)_mm256_undefined_si256(), \
+ (__mmask8)-1)
+
+#define _mm512_mask_extracti32x8_epi32(W, U, A, imm) \
+ (__m256i)__builtin_ia32_extracti32x8_mask((__v16si)(__m512i)(A), (int)(imm), \
+ (__v8si)(__m256i)(W), \
+ (__mmask8)(U))
+
+#define _mm512_maskz_extracti32x8_epi32(U, A, imm) \
+ (__m256i)__builtin_ia32_extracti32x8_mask((__v16si)(__m512i)(A), (int)(imm), \
+ (__v8si)_mm256_setzero_si256(), \
+ (__mmask8)(U))
+
+#define _mm512_extracti64x2_epi64(A, imm) \
+ (__m128i)__builtin_ia32_extracti64x2_512_mask((__v8di)(__m512i)(A), \
+ (int)(imm), \
+ (__v2di)_mm_undefined_si128(), \
+ (__mmask8)-1)
+
+#define _mm512_mask_extracti64x2_epi64(W, U, A, imm) \
+ (__m128i)__builtin_ia32_extracti64x2_512_mask((__v8di)(__m512i)(A), \
+ (int)(imm), \
+ (__v2di)(__m128i)(W), \
+ (__mmask8)(U))
+
+#define _mm512_maskz_extracti64x2_epi64(U, A, imm) \
+ (__m128i)__builtin_ia32_extracti64x2_512_mask((__v8di)(__m512i)(A), \
+ (int)(imm), \
+ (__v2di)_mm_setzero_si128(), \
+ (__mmask8)(U))
+
+#define _mm512_insertf32x8(A, B, imm) \
+ (__m512)__builtin_ia32_insertf32x8((__v16sf)(__m512)(A), \
+ (__v8sf)(__m256)(B), (int)(imm))
+
+#define _mm512_mask_insertf32x8(W, U, A, B, imm) \
(__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
(__v16sf)_mm512_insertf32x8((A), (B), (imm)), \
- (__v16sf)(W)); })
+ (__v16sf)(__m512)(W))
-#define _mm512_maskz_insertf32x8(U, A, B, imm) __extension__ ({ \
+#define _mm512_maskz_insertf32x8(U, A, B, imm) \
(__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
(__v16sf)_mm512_insertf32x8((A), (B), (imm)), \
- (__v16sf)_mm512_setzero_ps()); })
-
-#define _mm512_insertf64x2(A, B, imm) __extension__ ({ \
- (__m512d)__builtin_shufflevector((__v8df)(__m512d)(A), \
- (__v8df)_mm512_castpd128_pd512((__m128d)(B)),\
- (((imm) & 0x3) == 0) ? 8 : 0, \
- (((imm) & 0x3) == 0) ? 9 : 1, \
- (((imm) & 0x3) == 1) ? 8 : 2, \
- (((imm) & 0x3) == 1) ? 9 : 3, \
- (((imm) & 0x3) == 2) ? 8 : 4, \
- (((imm) & 0x3) == 2) ? 9 : 5, \
- (((imm) & 0x3) == 3) ? 8 : 6, \
- (((imm) & 0x3) == 3) ? 9 : 7); })
-
-#define _mm512_mask_insertf64x2(W, U, A, B, imm) __extension__ ({ \
+ (__v16sf)_mm512_setzero_ps())
+
+#define _mm512_insertf64x2(A, B, imm) \
+ (__m512d)__builtin_ia32_insertf64x2_512((__v8df)(__m512d)(A), \
+ (__v2df)(__m128d)(B), (int)(imm))
+
+#define _mm512_mask_insertf64x2(W, U, A, B, imm) \
(__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
(__v8df)_mm512_insertf64x2((A), (B), (imm)), \
- (__v8df)(W)); })
+ (__v8df)(__m512d)(W))
-#define _mm512_maskz_insertf64x2(U, A, B, imm) __extension__ ({ \
+#define _mm512_maskz_insertf64x2(U, A, B, imm) \
(__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
(__v8df)_mm512_insertf64x2((A), (B), (imm)), \
- (__v8df)_mm512_setzero_pd()); })
-
-#define _mm512_inserti32x8(A, B, imm) __extension__ ({ \
- (__m512i)__builtin_shufflevector((__v16si)(__m512i)(A), \
- (__v16si)_mm512_castsi256_si512((__m256i)(B)),\
- ((imm) & 0x1) ? 0 : 16, \
- ((imm) & 0x1) ? 1 : 17, \
- ((imm) & 0x1) ? 2 : 18, \
- ((imm) & 0x1) ? 3 : 19, \
- ((imm) & 0x1) ? 4 : 20, \
- ((imm) & 0x1) ? 5 : 21, \
- ((imm) & 0x1) ? 6 : 22, \
- ((imm) & 0x1) ? 7 : 23, \
- ((imm) & 0x1) ? 16 : 8, \
- ((imm) & 0x1) ? 17 : 9, \
- ((imm) & 0x1) ? 18 : 10, \
- ((imm) & 0x1) ? 19 : 11, \
- ((imm) & 0x1) ? 20 : 12, \
- ((imm) & 0x1) ? 21 : 13, \
- ((imm) & 0x1) ? 22 : 14, \
- ((imm) & 0x1) ? 23 : 15); })
-
-#define _mm512_mask_inserti32x8(W, U, A, B, imm) __extension__ ({ \
+ (__v8df)_mm512_setzero_pd())
+
+#define _mm512_inserti32x8(A, B, imm) \
+ (__m512i)__builtin_ia32_inserti32x8((__v16si)(__m512i)(A), \
+ (__v8si)(__m256i)(B), (int)(imm))
+
+#define _mm512_mask_inserti32x8(W, U, A, B, imm) \
(__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
(__v16si)_mm512_inserti32x8((A), (B), (imm)), \
- (__v16si)(W)); })
+ (__v16si)(__m512i)(W))
-#define _mm512_maskz_inserti32x8(U, A, B, imm) __extension__ ({ \
+#define _mm512_maskz_inserti32x8(U, A, B, imm) \
(__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
(__v16si)_mm512_inserti32x8((A), (B), (imm)), \
- (__v16si)_mm512_setzero_si512()); })
-
-#define _mm512_inserti64x2(A, B, imm) __extension__ ({ \
- (__m512i)__builtin_shufflevector((__v8di)(__m512i)(A), \
- (__v8di)_mm512_castsi128_si512((__m128i)(B)),\
- (((imm) & 0x3) == 0) ? 8 : 0, \
- (((imm) & 0x3) == 0) ? 9 : 1, \
- (((imm) & 0x3) == 1) ? 8 : 2, \
- (((imm) & 0x3) == 1) ? 9 : 3, \
- (((imm) & 0x3) == 2) ? 8 : 4, \
- (((imm) & 0x3) == 2) ? 9 : 5, \
- (((imm) & 0x3) == 3) ? 8 : 6, \
- (((imm) & 0x3) == 3) ? 9 : 7); })
-
-#define _mm512_mask_inserti64x2(W, U, A, B, imm) __extension__ ({ \
+ (__v16si)_mm512_setzero_si512())
+
+#define _mm512_inserti64x2(A, B, imm) \
+ (__m512i)__builtin_ia32_inserti64x2_512((__v8di)(__m512i)(A), \
+ (__v2di)(__m128i)(B), (int)(imm))
+
+#define _mm512_mask_inserti64x2(W, U, A, B, imm) \
(__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
(__v8di)_mm512_inserti64x2((A), (B), (imm)), \
- (__v8di)(W)); })
+ (__v8di)(__m512i)(W))
-#define _mm512_maskz_inserti64x2(U, A, B, imm) __extension__ ({ \
+#define _mm512_maskz_inserti64x2(U, A, B, imm) \
(__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
(__v8di)_mm512_inserti64x2((A), (B), (imm)), \
- (__v8di)_mm512_setzero_si512()); })
+ (__v8di)_mm512_setzero_si512())
-#define _mm512_mask_fpclass_ps_mask(U, A, imm) __extension__ ({ \
+#define _mm512_mask_fpclass_ps_mask(U, A, imm) \
(__mmask16)__builtin_ia32_fpclassps512_mask((__v16sf)(__m512)(A), \
- (int)(imm), (__mmask16)(U)); })
+ (int)(imm), (__mmask16)(U))
-#define _mm512_fpclass_ps_mask(A, imm) __extension__ ({ \
+#define _mm512_fpclass_ps_mask(A, imm) \
(__mmask16)__builtin_ia32_fpclassps512_mask((__v16sf)(__m512)(A), \
- (int)(imm), (__mmask16)-1); })
+ (int)(imm), (__mmask16)-1)
-#define _mm512_mask_fpclass_pd_mask(U, A, imm) __extension__ ({ \
+#define _mm512_mask_fpclass_pd_mask(U, A, imm) \
(__mmask8)__builtin_ia32_fpclasspd512_mask((__v8df)(__m512d)(A), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm512_fpclass_pd_mask(A, imm) __extension__ ({ \
+#define _mm512_fpclass_pd_mask(A, imm) \
(__mmask8)__builtin_ia32_fpclasspd512_mask((__v8df)(__m512d)(A), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_fpclass_sd_mask(A, imm) __extension__ ({ \
+#define _mm_fpclass_sd_mask(A, imm) \
(__mmask8)__builtin_ia32_fpclasssd_mask((__v2df)(__m128d)(A), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_fpclass_sd_mask(U, A, imm) __extension__ ({ \
+#define _mm_mask_fpclass_sd_mask(U, A, imm) \
(__mmask8)__builtin_ia32_fpclasssd_mask((__v2df)(__m128d)(A), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_fpclass_ss_mask(A, imm) __extension__ ({ \
+#define _mm_fpclass_ss_mask(A, imm) \
(__mmask8)__builtin_ia32_fpclassss_mask((__v4sf)(__m128)(A), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_fpclass_ss_mask(U, A, imm) __extension__ ({ \
+#define _mm_mask_fpclass_ss_mask(U, A, imm) \
(__mmask8)__builtin_ia32_fpclassss_mask((__v4sf)(__m128)(A), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
+#undef __DEFAULT_FN_ATTRS512
#undef __DEFAULT_FN_ATTRS
#endif
/*===---- avx512erintrin.h - AVX512ER intrinsics ---------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#ifndef __AVX512ERINTRIN_H
#define __AVX512ERINTRIN_H
-// exp2a23
-#define _mm512_exp2a23_round_pd(A, R) __extension__ ({ \
+/* exp2a23 */
+#define _mm512_exp2a23_round_pd(A, R) \
(__m512d)__builtin_ia32_exp2pd_mask((__v8df)(__m512d)(A), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_exp2a23_round_pd(S, M, A, R) __extension__ ({ \
+#define _mm512_mask_exp2a23_round_pd(S, M, A, R) \
(__m512d)__builtin_ia32_exp2pd_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(S), (__mmask8)(M), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_maskz_exp2a23_round_pd(M, A, R) __extension__ ({ \
+#define _mm512_maskz_exp2a23_round_pd(M, A, R) \
(__m512d)__builtin_ia32_exp2pd_mask((__v8df)(__m512d)(A), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)(M), (int)(R)); })
+ (__mmask8)(M), (int)(R))
#define _mm512_exp2a23_pd(A) \
_mm512_exp2a23_round_pd((A), _MM_FROUND_CUR_DIRECTION)
#define _mm512_maskz_exp2a23_pd(M, A) \
_mm512_maskz_exp2a23_round_pd((M), (A), _MM_FROUND_CUR_DIRECTION)
-#define _mm512_exp2a23_round_ps(A, R) __extension__ ({ \
+#define _mm512_exp2a23_round_ps(A, R) \
(__m512)__builtin_ia32_exp2ps_mask((__v16sf)(__m512)(A), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_exp2a23_round_ps(S, M, A, R) __extension__ ({ \
+#define _mm512_mask_exp2a23_round_ps(S, M, A, R) \
(__m512)__builtin_ia32_exp2ps_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(S), (__mmask16)(M), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_maskz_exp2a23_round_ps(M, A, R) __extension__ ({ \
+#define _mm512_maskz_exp2a23_round_ps(M, A, R) \
(__m512)__builtin_ia32_exp2ps_mask((__v16sf)(__m512)(A), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(M), (int)(R)); })
+ (__mmask16)(M), (int)(R))
#define _mm512_exp2a23_ps(A) \
_mm512_exp2a23_round_ps((A), _MM_FROUND_CUR_DIRECTION)
#define _mm512_maskz_exp2a23_ps(M, A) \
_mm512_maskz_exp2a23_round_ps((M), (A), _MM_FROUND_CUR_DIRECTION)
-// rsqrt28
-#define _mm512_rsqrt28_round_pd(A, R) __extension__ ({ \
+/* rsqrt28 */
+#define _mm512_rsqrt28_round_pd(A, R) \
(__m512d)__builtin_ia32_rsqrt28pd_mask((__v8df)(__m512d)(A), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_rsqrt28_round_pd(S, M, A, R) __extension__ ({ \
+#define _mm512_mask_rsqrt28_round_pd(S, M, A, R) \
(__m512d)__builtin_ia32_rsqrt28pd_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(S), (__mmask8)(M), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_maskz_rsqrt28_round_pd(M, A, R) __extension__ ({ \
+#define _mm512_maskz_rsqrt28_round_pd(M, A, R) \
(__m512d)__builtin_ia32_rsqrt28pd_mask((__v8df)(__m512d)(A), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)(M), (int)(R)); })
+ (__mmask8)(M), (int)(R))
#define _mm512_rsqrt28_pd(A) \
_mm512_rsqrt28_round_pd((A), _MM_FROUND_CUR_DIRECTION)
#define _mm512_maskz_rsqrt28_pd(M, A) \
_mm512_maskz_rsqrt28_round_pd((M), (A), _MM_FROUND_CUR_DIRECTION)
-#define _mm512_rsqrt28_round_ps(A, R) __extension__ ({ \
+#define _mm512_rsqrt28_round_ps(A, R) \
(__m512)__builtin_ia32_rsqrt28ps_mask((__v16sf)(__m512)(A), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_rsqrt28_round_ps(S, M, A, R) __extension__ ({ \
+#define _mm512_mask_rsqrt28_round_ps(S, M, A, R) \
(__m512)__builtin_ia32_rsqrt28ps_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(S), (__mmask16)(M), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_maskz_rsqrt28_round_ps(M, A, R) __extension__ ({ \
+#define _mm512_maskz_rsqrt28_round_ps(M, A, R) \
(__m512)__builtin_ia32_rsqrt28ps_mask((__v16sf)(__m512)(A), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(M), (int)(R)); })
+ (__mmask16)(M), (int)(R))
#define _mm512_rsqrt28_ps(A) \
_mm512_rsqrt28_round_ps((A), _MM_FROUND_CUR_DIRECTION)
#define _mm512_maskz_rsqrt28_ps(M, A) \
_mm512_maskz_rsqrt28_round_ps((M), (A), _MM_FROUND_CUR_DIRECTION)
-#define _mm_rsqrt28_round_ss(A, B, R) __extension__ ({ \
+#define _mm_rsqrt28_round_ss(A, B, R) \
(__m128)__builtin_ia32_rsqrt28ss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_rsqrt28_round_ss(S, M, A, B, R) __extension__ ({ \
+#define _mm_mask_rsqrt28_round_ss(S, M, A, B, R) \
(__m128)__builtin_ia32_rsqrt28ss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(S), \
- (__mmask8)(M), (int)(R)); })
+ (__mmask8)(M), (int)(R))
-#define _mm_maskz_rsqrt28_round_ss(M, A, B, R) __extension__ ({ \
+#define _mm_maskz_rsqrt28_round_ss(M, A, B, R) \
(__m128)__builtin_ia32_rsqrt28ss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(M), (int)(R)); })
+ (__mmask8)(M), (int)(R))
#define _mm_rsqrt28_ss(A, B) \
_mm_rsqrt28_round_ss((A), (B), _MM_FROUND_CUR_DIRECTION)
#define _mm_maskz_rsqrt28_ss(M, A, B) \
_mm_maskz_rsqrt28_round_ss((M), (A), (B), _MM_FROUND_CUR_DIRECTION)
-#define _mm_rsqrt28_round_sd(A, B, R) __extension__ ({ \
+#define _mm_rsqrt28_round_sd(A, B, R) \
(__m128d)__builtin_ia32_rsqrt28sd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_rsqrt28_round_sd(S, M, A, B, R) __extension__ ({ \
+#define _mm_mask_rsqrt28_round_sd(S, M, A, B, R) \
(__m128d)__builtin_ia32_rsqrt28sd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(S), \
- (__mmask8)(M), (int)(R)); })
+ (__mmask8)(M), (int)(R))
-#define _mm_maskz_rsqrt28_round_sd(M, A, B, R) __extension__ ({ \
+#define _mm_maskz_rsqrt28_round_sd(M, A, B, R) \
(__m128d)__builtin_ia32_rsqrt28sd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(M), (int)(R)); })
+ (__mmask8)(M), (int)(R))
#define _mm_rsqrt28_sd(A, B) \
_mm_rsqrt28_round_sd((A), (B), _MM_FROUND_CUR_DIRECTION)
#define _mm_maskz_rsqrt28_sd(M, A, B) \
_mm_maskz_rsqrt28_round_sd((M), (A), (B), _MM_FROUND_CUR_DIRECTION)
-// rcp28
-#define _mm512_rcp28_round_pd(A, R) __extension__ ({ \
+/* rcp28 */
+#define _mm512_rcp28_round_pd(A, R) \
(__m512d)__builtin_ia32_rcp28pd_mask((__v8df)(__m512d)(A), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_rcp28_round_pd(S, M, A, R) __extension__ ({ \
+#define _mm512_mask_rcp28_round_pd(S, M, A, R) \
(__m512d)__builtin_ia32_rcp28pd_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(S), (__mmask8)(M), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_maskz_rcp28_round_pd(M, A, R) __extension__ ({ \
+#define _mm512_maskz_rcp28_round_pd(M, A, R) \
(__m512d)__builtin_ia32_rcp28pd_mask((__v8df)(__m512d)(A), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)(M), (int)(R)); })
+ (__mmask8)(M), (int)(R))
#define _mm512_rcp28_pd(A) \
_mm512_rcp28_round_pd((A), _MM_FROUND_CUR_DIRECTION)
#define _mm512_maskz_rcp28_pd(M, A) \
_mm512_maskz_rcp28_round_pd((M), (A), _MM_FROUND_CUR_DIRECTION)
-#define _mm512_rcp28_round_ps(A, R) __extension__ ({ \
+#define _mm512_rcp28_round_ps(A, R) \
(__m512)__builtin_ia32_rcp28ps_mask((__v16sf)(__m512)(A), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_rcp28_round_ps(S, M, A, R) __extension__ ({ \
+#define _mm512_mask_rcp28_round_ps(S, M, A, R) \
(__m512)__builtin_ia32_rcp28ps_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(S), (__mmask16)(M), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_maskz_rcp28_round_ps(M, A, R) __extension__ ({ \
+#define _mm512_maskz_rcp28_round_ps(M, A, R) \
(__m512)__builtin_ia32_rcp28ps_mask((__v16sf)(__m512)(A), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(M), (int)(R)); })
+ (__mmask16)(M), (int)(R))
#define _mm512_rcp28_ps(A) \
_mm512_rcp28_round_ps((A), _MM_FROUND_CUR_DIRECTION)
#define _mm512_maskz_rcp28_ps(M, A) \
_mm512_maskz_rcp28_round_ps((M), (A), _MM_FROUND_CUR_DIRECTION)
-#define _mm_rcp28_round_ss(A, B, R) __extension__ ({ \
+#define _mm_rcp28_round_ss(A, B, R) \
(__m128)__builtin_ia32_rcp28ss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_rcp28_round_ss(S, M, A, B, R) __extension__ ({ \
+#define _mm_mask_rcp28_round_ss(S, M, A, B, R) \
(__m128)__builtin_ia32_rcp28ss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(S), \
- (__mmask8)(M), (int)(R)); })
+ (__mmask8)(M), (int)(R))
-#define _mm_maskz_rcp28_round_ss(M, A, B, R) __extension__ ({ \
+#define _mm_maskz_rcp28_round_ss(M, A, B, R) \
(__m128)__builtin_ia32_rcp28ss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(M), (int)(R)); })
+ (__mmask8)(M), (int)(R))
#define _mm_rcp28_ss(A, B) \
_mm_rcp28_round_ss((A), (B), _MM_FROUND_CUR_DIRECTION)
#define _mm_maskz_rcp28_ss(M, A, B) \
_mm_maskz_rcp28_round_ss((M), (A), (B), _MM_FROUND_CUR_DIRECTION)
-#define _mm_rcp28_round_sd(A, B, R) __extension__ ({ \
+#define _mm_rcp28_round_sd(A, B, R) \
(__m128d)__builtin_ia32_rcp28sd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_rcp28_round_sd(S, M, A, B, R) __extension__ ({ \
+#define _mm_mask_rcp28_round_sd(S, M, A, B, R) \
(__m128d)__builtin_ia32_rcp28sd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(S), \
- (__mmask8)(M), (int)(R)); })
+ (__mmask8)(M), (int)(R))
-#define _mm_maskz_rcp28_round_sd(M, A, B, R) __extension__ ({ \
+#define _mm_maskz_rcp28_round_sd(M, A, B, R) \
(__m128d)__builtin_ia32_rcp28sd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(M), (int)(R)); })
+ (__mmask8)(M), (int)(R))
#define _mm_rcp28_sd(A, B) \
_mm_rcp28_round_sd((A), (B), _MM_FROUND_CUR_DIRECTION)
#define _mm_maskz_rcp28_sd(M, A, B) \
_mm_maskz_rcp28_round_sd((M), (A), (B), _MM_FROUND_CUR_DIRECTION)
-#endif // __AVX512ERINTRIN_H
+#endif /* __AVX512ERINTRIN_H */
/*===---- avx512fintrin.h - AVX512F intrinsics -----------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
typedef unsigned long long __v8du __attribute__((__vector_size__(64)));
typedef unsigned int __v16su __attribute__((__vector_size__(64)));
-typedef float __m512 __attribute__((__vector_size__(64)));
-typedef double __m512d __attribute__((__vector_size__(64)));
-typedef long long __m512i __attribute__((__vector_size__(64)));
+typedef float __m512 __attribute__((__vector_size__(64), __aligned__(64)));
+typedef double __m512d __attribute__((__vector_size__(64), __aligned__(64)));
+typedef long long __m512i __attribute__((__vector_size__(64), __aligned__(64)));
+
+typedef float __m512_u __attribute__((__vector_size__(64), __aligned__(1)));
+typedef double __m512d_u __attribute__((__vector_size__(64), __aligned__(1)));
+typedef long long __m512i_u __attribute__((__vector_size__(64), __aligned__(1)));
typedef unsigned char __mmask8;
typedef unsigned short __mmask16;
} _MM_MANTISSA_SIGN_ENUM;
/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS512 __attribute__((__always_inline__, __nodebug__, __target__("avx512f"), __min_vector_width__(512)))
+#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx512f"), __min_vector_width__(128)))
#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512f")))
/* Create vectors with repeated elements */
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_setzero_si512(void)
{
- return (__m512i)(__v8di){ 0, 0, 0, 0, 0, 0, 0, 0 };
+ return __extension__ (__m512i)(__v8di){ 0, 0, 0, 0, 0, 0, 0, 0 };
}
#define _mm512_setzero_epi32 _mm512_setzero_si512
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_undefined_pd(void)
{
return (__m512d)__builtin_ia32_undef512();
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_undefined(void)
{
return (__m512)__builtin_ia32_undef512();
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_undefined_ps(void)
{
return (__m512)__builtin_ia32_undef512();
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_undefined_epi32(void)
{
return (__m512i)__builtin_ia32_undef512();
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_broadcastd_epi32 (__m128i __A)
{
- return (__m512i)__builtin_shufflevector((__v4si) __A,
- (__v4si)_mm_undefined_si128(),
+ return (__m512i)__builtin_shufflevector((__v4si) __A, (__v4si) __A,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_broadcastd_epi32 (__m512i __O, __mmask16 __M, __m128i __A)
{
return (__m512i)__builtin_ia32_selectd_512(__M,
(__v16si) __O);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcastd_epi32 (__mmask16 __M, __m128i __A)
{
return (__m512i)__builtin_ia32_selectd_512(__M,
(__v16si) _mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_broadcastq_epi64 (__m128i __A)
{
- return (__m512i)__builtin_shufflevector((__v2di) __A,
- (__v2di) _mm_undefined_si128(),
+ return (__m512i)__builtin_shufflevector((__v2di) __A, (__v2di) __A,
0, 0, 0, 0, 0, 0, 0, 0);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_broadcastq_epi64 (__m512i __O, __mmask8 __M, __m128i __A)
{
return (__m512i)__builtin_ia32_selectq_512(__M,
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcastq_epi64 (__mmask8 __M, __m128i __A)
{
return (__m512i)__builtin_ia32_selectq_512(__M,
(__v8di) _mm512_setzero_si512());
}
-static __inline __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_set1_epi32(__mmask16 __M, int __A)
-{
- return (__m512i) __builtin_ia32_pbroadcastd512_gpr_mask (__A,
- (__v16si)
- _mm512_setzero_si512 (),
- __M);
-}
-
-#ifdef __x86_64__
-static __inline __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_set1_epi64(__mmask8 __M, long long __A)
-{
- return (__m512i) __builtin_ia32_pbroadcastq512_gpr_mask (__A,
- (__v8di)
- _mm512_setzero_si512 (),
- __M);
-}
-#endif
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_setzero_ps(void)
{
- return (__m512){ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,
- 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 };
+ return __extension__ (__m512){ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,
+ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 };
}
#define _mm512_setzero _mm512_setzero_ps
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_setzero_pd(void)
{
- return (__m512d){ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 };
+ return __extension__ (__m512d){ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 };
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_set1_ps(float __w)
{
- return (__m512){ __w, __w, __w, __w, __w, __w, __w, __w,
- __w, __w, __w, __w, __w, __w, __w, __w };
+ return __extension__ (__m512){ __w, __w, __w, __w, __w, __w, __w, __w,
+ __w, __w, __w, __w, __w, __w, __w, __w };
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_set1_pd(double __w)
{
- return (__m512d){ __w, __w, __w, __w, __w, __w, __w, __w };
+ return __extension__ (__m512d){ __w, __w, __w, __w, __w, __w, __w, __w };
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set1_epi8(char __w)
{
- return (__m512i)(__v64qi){ __w, __w, __w, __w, __w, __w, __w, __w,
- __w, __w, __w, __w, __w, __w, __w, __w,
- __w, __w, __w, __w, __w, __w, __w, __w,
- __w, __w, __w, __w, __w, __w, __w, __w,
- __w, __w, __w, __w, __w, __w, __w, __w,
- __w, __w, __w, __w, __w, __w, __w, __w,
- __w, __w, __w, __w, __w, __w, __w, __w,
- __w, __w, __w, __w, __w, __w, __w, __w };
+ return __extension__ (__m512i)(__v64qi){
+ __w, __w, __w, __w, __w, __w, __w, __w,
+ __w, __w, __w, __w, __w, __w, __w, __w,
+ __w, __w, __w, __w, __w, __w, __w, __w,
+ __w, __w, __w, __w, __w, __w, __w, __w,
+ __w, __w, __w, __w, __w, __w, __w, __w,
+ __w, __w, __w, __w, __w, __w, __w, __w,
+ __w, __w, __w, __w, __w, __w, __w, __w,
+ __w, __w, __w, __w, __w, __w, __w, __w };
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set1_epi16(short __w)
{
- return (__m512i)(__v32hi){ __w, __w, __w, __w, __w, __w, __w, __w,
- __w, __w, __w, __w, __w, __w, __w, __w,
- __w, __w, __w, __w, __w, __w, __w, __w,
- __w, __w, __w, __w, __w, __w, __w, __w };
+ return __extension__ (__m512i)(__v32hi){
+ __w, __w, __w, __w, __w, __w, __w, __w,
+ __w, __w, __w, __w, __w, __w, __w, __w,
+ __w, __w, __w, __w, __w, __w, __w, __w,
+ __w, __w, __w, __w, __w, __w, __w, __w };
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set1_epi32(int __s)
{
- return (__m512i)(__v16si){ __s, __s, __s, __s, __s, __s, __s, __s,
- __s, __s, __s, __s, __s, __s, __s, __s };
+ return __extension__ (__m512i)(__v16si){
+ __s, __s, __s, __s, __s, __s, __s, __s,
+ __s, __s, __s, __s, __s, __s, __s, __s };
+}
+
+static __inline __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_set1_epi32(__mmask16 __M, int __A)
+{
+ return (__m512i)__builtin_ia32_selectd_512(__M,
+ (__v16si)_mm512_set1_epi32(__A),
+ (__v16si)_mm512_setzero_si512());
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set1_epi64(long long __d)
{
- return (__m512i)(__v8di){ __d, __d, __d, __d, __d, __d, __d, __d };
+ return __extension__(__m512i)(__v8di){ __d, __d, __d, __d, __d, __d, __d, __d };
+}
+
+static __inline __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_set1_epi64(__mmask8 __M, long long __A)
+{
+ return (__m512i)__builtin_ia32_selectq_512(__M,
+ (__v8di)_mm512_set1_epi64(__A),
+ (__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_broadcastss_ps(__m128 __A)
{
- return (__m512)__builtin_shufflevector((__v4sf) __A,
- (__v4sf)_mm_undefined_ps(),
+ return (__m512)__builtin_shufflevector((__v4sf) __A, (__v4sf) __A,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set4_epi32 (int __A, int __B, int __C, int __D)
{
- return (__m512i)(__v16si)
+ return __extension__ (__m512i)(__v16si)
{ __D, __C, __B, __A, __D, __C, __B, __A,
__D, __C, __B, __A, __D, __C, __B, __A };
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set4_epi64 (long long __A, long long __B, long long __C,
long long __D)
{
- return (__m512i) (__v8di)
+ return __extension__ (__m512i) (__v8di)
{ __D, __C, __B, __A, __D, __C, __B, __A };
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_set4_pd (double __A, double __B, double __C, double __D)
{
- return (__m512d)
+ return __extension__ (__m512d)
{ __D, __C, __B, __A, __D, __C, __B, __A };
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_set4_ps (float __A, float __B, float __C, float __D)
{
- return (__m512)
+ return __extension__ (__m512)
{ __D, __C, __B, __A, __D, __C, __B, __A,
__D, __C, __B, __A, __D, __C, __B, __A };
}
#define _mm512_setr4_ps(e0,e1,e2,e3) \
_mm512_set4_ps((e3),(e2),(e1),(e0))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_broadcastsd_pd(__m128d __A)
{
- return (__m512d)__builtin_shufflevector((__v2df) __A,
- (__v2df) _mm_undefined_pd(),
+ return (__m512d)__builtin_shufflevector((__v2df) __A, (__v2df) __A,
0, 0, 0, 0, 0, 0, 0, 0);
}
/* Cast between vector types */
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_castpd256_pd512(__m256d __a)
{
return __builtin_shufflevector(__a, __a, 0, 1, 2, 3, -1, -1, -1, -1);
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_castps256_ps512(__m256 __a)
{
return __builtin_shufflevector(__a, __a, 0, 1, 2, 3, 4, 5, 6, 7,
-1, -1, -1, -1, -1, -1, -1, -1);
}
-static __inline __m128d __DEFAULT_FN_ATTRS
+static __inline __m128d __DEFAULT_FN_ATTRS512
_mm512_castpd512_pd128(__m512d __a)
{
return __builtin_shufflevector(__a, __a, 0, 1);
}
-static __inline __m256d __DEFAULT_FN_ATTRS
+static __inline __m256d __DEFAULT_FN_ATTRS512
_mm512_castpd512_pd256 (__m512d __A)
{
return __builtin_shufflevector(__A, __A, 0, 1, 2, 3);
}
-static __inline __m128 __DEFAULT_FN_ATTRS
+static __inline __m128 __DEFAULT_FN_ATTRS512
_mm512_castps512_ps128(__m512 __a)
{
return __builtin_shufflevector(__a, __a, 0, 1, 2, 3);
}
-static __inline __m256 __DEFAULT_FN_ATTRS
+static __inline __m256 __DEFAULT_FN_ATTRS512
_mm512_castps512_ps256 (__m512 __A)
{
return __builtin_shufflevector(__A, __A, 0, 1, 2, 3, 4, 5, 6, 7);
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_castpd_ps (__m512d __A)
{
return (__m512) (__A);
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_castpd_si512 (__m512d __A)
{
return (__m512i) (__A);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_castpd128_pd512 (__m128d __A)
{
return __builtin_shufflevector( __A, __A, 0, 1, -1, -1, -1, -1, -1, -1);
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_castps_pd (__m512 __A)
{
return (__m512d) (__A);
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_castps_si512 (__m512 __A)
{
return (__m512i) (__A);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_castps128_ps512 (__m128 __A)
{
return __builtin_shufflevector( __A, __A, 0, 1, 2, 3, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_castsi128_si512 (__m128i __A)
{
return __builtin_shufflevector( __A, __A, 0, 1, -1, -1, -1, -1, -1, -1);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_castsi256_si512 (__m256i __A)
{
return __builtin_shufflevector( __A, __A, 0, 1, 2, 3, -1, -1, -1, -1);
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_castsi512_ps (__m512i __A)
{
return (__m512) (__A);
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_castsi512_pd (__m512i __A)
{
return (__m512d) (__A);
}
-static __inline __m128i __DEFAULT_FN_ATTRS
+static __inline __m128i __DEFAULT_FN_ATTRS512
_mm512_castsi512_si128 (__m512i __A)
{
return (__m128i)__builtin_shufflevector(__A, __A , 0, 1);
}
-static __inline __m256i __DEFAULT_FN_ATTRS
+static __inline __m256i __DEFAULT_FN_ATTRS512
_mm512_castsi512_si256 (__m512i __A)
{
return (__m256i)__builtin_shufflevector(__A, __A , 0, 1, 2, 3);
return (int)__a;
}
-/// \brief Constructs a 512-bit floating-point vector of [8 x double] from a
+/// Constructs a 512-bit floating-point vector of [8 x double] from a
/// 128-bit floating-point vector of [2 x double]. The lower 128 bits
/// contain the value of the source vector. The upper 384 bits are set
/// to zero.
/// A 128-bit vector of [2 x double].
/// \returns A 512-bit floating-point vector of [8 x double]. The lower 128 bits
/// contain the value of the parameter. The upper 384 bits are set to zero.
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_zextpd128_pd512(__m128d __a)
{
return __builtin_shufflevector((__v2df)__a, (__v2df)_mm_setzero_pd(), 0, 1, 2, 3, 2, 3, 2, 3);
}
-/// \brief Constructs a 512-bit floating-point vector of [8 x double] from a
+/// Constructs a 512-bit floating-point vector of [8 x double] from a
/// 256-bit floating-point vector of [4 x double]. The lower 256 bits
/// contain the value of the source vector. The upper 256 bits are set
/// to zero.
/// A 256-bit vector of [4 x double].
/// \returns A 512-bit floating-point vector of [8 x double]. The lower 256 bits
/// contain the value of the parameter. The upper 256 bits are set to zero.
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_zextpd256_pd512(__m256d __a)
{
return __builtin_shufflevector((__v4df)__a, (__v4df)_mm256_setzero_pd(), 0, 1, 2, 3, 4, 5, 6, 7);
}
-/// \brief Constructs a 512-bit floating-point vector of [16 x float] from a
+/// Constructs a 512-bit floating-point vector of [16 x float] from a
/// 128-bit floating-point vector of [4 x float]. The lower 128 bits contain
/// the value of the source vector. The upper 384 bits are set to zero.
///
/// A 128-bit vector of [4 x float].
/// \returns A 512-bit floating-point vector of [16 x float]. The lower 128 bits
/// contain the value of the parameter. The upper 384 bits are set to zero.
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_zextps128_ps512(__m128 __a)
{
return __builtin_shufflevector((__v4sf)__a, (__v4sf)_mm_setzero_ps(), 0, 1, 2, 3, 4, 5, 6, 7, 4, 5, 6, 7, 4, 5, 6, 7);
}
-/// \brief Constructs a 512-bit floating-point vector of [16 x float] from a
+/// Constructs a 512-bit floating-point vector of [16 x float] from a
/// 256-bit floating-point vector of [8 x float]. The lower 256 bits contain
/// the value of the source vector. The upper 256 bits are set to zero.
///
/// A 256-bit vector of [8 x float].
/// \returns A 512-bit floating-point vector of [16 x float]. The lower 256 bits
/// contain the value of the parameter. The upper 256 bits are set to zero.
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_zextps256_ps512(__m256 __a)
{
return __builtin_shufflevector((__v8sf)__a, (__v8sf)_mm256_setzero_ps(), 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
}
-/// \brief Constructs a 512-bit integer vector from a 128-bit integer vector.
+/// Constructs a 512-bit integer vector from a 128-bit integer vector.
/// The lower 128 bits contain the value of the source vector. The upper
/// 384 bits are set to zero.
///
/// A 128-bit integer vector.
/// \returns A 512-bit integer vector. The lower 128 bits contain the value of
/// the parameter. The upper 384 bits are set to zero.
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_zextsi128_si512(__m128i __a)
{
return __builtin_shufflevector((__v2di)__a, (__v2di)_mm_setzero_si128(), 0, 1, 2, 3, 2, 3, 2, 3);
}
-/// \brief Constructs a 512-bit integer vector from a 256-bit integer vector.
+/// Constructs a 512-bit integer vector from a 256-bit integer vector.
/// The lower 256 bits contain the value of the source vector. The upper
/// 256 bits are set to zero.
///
/// A 256-bit integer vector.
/// \returns A 512-bit integer vector. The lower 256 bits contain the value of
/// the parameter. The upper 256 bits are set to zero.
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_zextsi256_si512(__m256i __a)
{
return __builtin_shufflevector((__v4di)__a, (__v4di)_mm256_setzero_si256(), 0, 1, 2, 3, 4, 5, 6, 7);
}
/* Bitwise operators */
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_and_epi32(__m512i __a, __m512i __b)
{
return (__m512i)((__v16su)__a & (__v16su)__b);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_and_epi32(__m512i __src, __mmask16 __k, __m512i __a, __m512i __b)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__k,
(__v16si) __src);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_and_epi32(__mmask16 __k, __m512i __a, __m512i __b)
{
return (__m512i) _mm512_mask_and_epi32(_mm512_setzero_si512 (),
__k, __a, __b);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_and_epi64(__m512i __a, __m512i __b)
{
return (__m512i)((__v8du)__a & (__v8du)__b);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_and_epi64(__m512i __src, __mmask8 __k, __m512i __a, __m512i __b)
{
return (__m512i) __builtin_ia32_selectq_512 ((__mmask8) __k,
(__v8di) __src);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_and_epi64(__mmask8 __k, __m512i __a, __m512i __b)
{
return (__m512i) _mm512_mask_and_epi64(_mm512_setzero_si512 (),
__k, __a, __b);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_andnot_si512 (__m512i __A, __m512i __B)
{
- return (__m512i)(~(__v8du)(__A) & (__v8du)__B);
+ return (__m512i)(~(__v8du)__A & (__v8du)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_andnot_epi32 (__m512i __A, __m512i __B)
{
- return (__m512i)(~(__v16su)(__A) & (__v16su)__B);
+ return (__m512i)(~(__v16su)__A & (__v16su)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_andnot_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_andnot_epi32(__mmask16 __U, __m512i __A, __m512i __B)
{
return (__m512i)_mm512_mask_andnot_epi32(_mm512_setzero_si512(),
__U, __A, __B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_andnot_epi64(__m512i __A, __m512i __B)
{
- return (__m512i)(~(__v8du)(__A) & (__v8du)__B);
+ return (__m512i)(~(__v8du)__A & (__v8du)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_andnot_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_andnot_epi64(__mmask8 __U, __m512i __A, __m512i __B)
{
return (__m512i)_mm512_mask_andnot_epi64(_mm512_setzero_si512(),
__U, __A, __B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_or_epi32(__m512i __a, __m512i __b)
{
return (__m512i)((__v16su)__a | (__v16su)__b);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_or_epi32(__m512i __src, __mmask16 __k, __m512i __a, __m512i __b)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__k,
(__v16si)__src);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_or_epi32(__mmask16 __k, __m512i __a, __m512i __b)
{
return (__m512i)_mm512_mask_or_epi32(_mm512_setzero_si512(), __k, __a, __b);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_or_epi64(__m512i __a, __m512i __b)
{
return (__m512i)((__v8du)__a | (__v8du)__b);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_or_epi64(__m512i __src, __mmask8 __k, __m512i __a, __m512i __b)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__k,
(__v8di)__src);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_or_epi64(__mmask8 __k, __m512i __a, __m512i __b)
{
return (__m512i)_mm512_mask_or_epi64(_mm512_setzero_si512(), __k, __a, __b);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_xor_epi32(__m512i __a, __m512i __b)
{
return (__m512i)((__v16su)__a ^ (__v16su)__b);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_xor_epi32(__m512i __src, __mmask16 __k, __m512i __a, __m512i __b)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__k,
(__v16si)__src);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_xor_epi32(__mmask16 __k, __m512i __a, __m512i __b)
{
return (__m512i)_mm512_mask_xor_epi32(_mm512_setzero_si512(), __k, __a, __b);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_xor_epi64(__m512i __a, __m512i __b)
{
return (__m512i)((__v8du)__a ^ (__v8du)__b);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_xor_epi64(__m512i __src, __mmask8 __k, __m512i __a, __m512i __b)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__k,
(__v8di)__src);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_xor_epi64(__mmask8 __k, __m512i __a, __m512i __b)
{
return (__m512i)_mm512_mask_xor_epi64(_mm512_setzero_si512(), __k, __a, __b);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_and_si512(__m512i __a, __m512i __b)
{
return (__m512i)((__v8du)__a & (__v8du)__b);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_or_si512(__m512i __a, __m512i __b)
{
return (__m512i)((__v8du)__a | (__v8du)__b);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_xor_si512(__m512i __a, __m512i __b)
{
return (__m512i)((__v8du)__a ^ (__v8du)__b);
/* Arithmetic */
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_add_pd(__m512d __a, __m512d __b)
{
return (__m512d)((__v8df)__a + (__v8df)__b);
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_add_ps(__m512 __a, __m512 __b)
{
return (__m512)((__v16sf)__a + (__v16sf)__b);
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_mul_pd(__m512d __a, __m512d __b)
{
return (__m512d)((__v8df)__a * (__v8df)__b);
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_mul_ps(__m512 __a, __m512 __b)
{
return (__m512)((__v16sf)__a * (__v16sf)__b);
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_sub_pd(__m512d __a, __m512d __b)
{
return (__m512d)((__v8df)__a - (__v8df)__b);
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_sub_ps(__m512 __a, __m512 __b)
{
return (__m512)((__v16sf)__a - (__v16sf)__b);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_add_epi64 (__m512i __A, __m512i __B)
{
return (__m512i) ((__v8du) __A + (__v8du) __B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_add_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_add_epi64(__mmask8 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sub_epi64 (__m512i __A, __m512i __B)
{
return (__m512i) ((__v8du) __A - (__v8du) __B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sub_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sub_epi64(__mmask8 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_add_epi32 (__m512i __A, __m512i __B)
{
return (__m512i) ((__v16su) __A + (__v16su) __B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_add_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_add_epi32 (__mmask16 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sub_epi32 (__m512i __A, __m512i __B)
{
return (__m512i) ((__v16su) __A - (__v16su) __B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sub_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sub_epi32(__mmask16 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_setzero_si512());
}
-#define _mm512_mask_max_round_pd(W, U, A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_maxpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)(__m512d)(W), (__mmask8)(U), \
- (int)(R)); })
+#define _mm512_max_round_pd(A, B, R) \
+ (__m512d)__builtin_ia32_maxpd512((__v8df)(__m512d)(A), \
+ (__v8df)(__m512d)(B), (int)(R))
-#define _mm512_maskz_max_round_pd(U, A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_maxpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+#define _mm512_mask_max_round_pd(W, U, A, B, R) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_max_round_pd((A), (B), (R)), \
+ (__v8df)(W))
-#define _mm512_max_round_pd(A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_maxpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)_mm512_undefined_pd(), \
- (__mmask8)-1, (int)(R)); })
+#define _mm512_maskz_max_round_pd(U, A, B, R) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_max_round_pd((A), (B), (R)), \
+ (__v8df)_mm512_setzero_pd())
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_max_pd(__m512d __A, __m512d __B)
{
- return (__m512d) __builtin_ia32_maxpd512_mask ((__v8df) __A,
- (__v8df) __B,
- (__v8df)
- _mm512_setzero_pd (),
- (__mmask8) -1,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d) __builtin_ia32_maxpd512((__v8df) __A, (__v8df) __B,
+ _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_max_pd (__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
{
- return (__m512d) __builtin_ia32_maxpd512_mask ((__v8df) __A,
- (__v8df) __B,
- (__v8df) __W,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_ia32_selectpd_512(__U,
+ (__v8df)_mm512_max_pd(__A, __B),
+ (__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_max_pd (__mmask8 __U, __m512d __A, __m512d __B)
{
- return (__m512d) __builtin_ia32_maxpd512_mask ((__v8df) __A,
- (__v8df) __B,
- (__v8df)
- _mm512_setzero_pd (),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_ia32_selectpd_512(__U,
+ (__v8df)_mm512_max_pd(__A, __B),
+ (__v8df)_mm512_setzero_pd());
}
-#define _mm512_mask_max_round_ps(W, U, A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_maxps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)(__m512)(W), (__mmask16)(U), \
- (int)(R)); })
+#define _mm512_max_round_ps(A, B, R) \
+ (__m512)__builtin_ia32_maxps512((__v16sf)(__m512)(A), \
+ (__v16sf)(__m512)(B), (int)(R))
-#define _mm512_maskz_max_round_ps(U, A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_maxps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U), (int)(R)); })
+#define _mm512_mask_max_round_ps(W, U, A, B, R) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_max_round_ps((A), (B), (R)), \
+ (__v16sf)(W))
-#define _mm512_max_round_ps(A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_maxps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)_mm512_undefined_ps(), \
- (__mmask16)-1, (int)(R)); })
+#define _mm512_maskz_max_round_ps(U, A, B, R) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_max_round_ps((A), (B), (R)), \
+ (__v16sf)_mm512_setzero_ps())
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_max_ps(__m512 __A, __m512 __B)
{
- return (__m512) __builtin_ia32_maxps512_mask ((__v16sf) __A,
- (__v16sf) __B,
- (__v16sf)
- _mm512_setzero_ps (),
- (__mmask16) -1,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512) __builtin_ia32_maxps512((__v16sf) __A, (__v16sf) __B,
+ _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_max_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
{
- return (__m512) __builtin_ia32_maxps512_mask ((__v16sf) __A,
- (__v16sf) __B,
- (__v16sf) __W,
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512)__builtin_ia32_selectps_512(__U,
+ (__v16sf)_mm512_max_ps(__A, __B),
+ (__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_max_ps (__mmask16 __U, __m512 __A, __m512 __B)
{
- return (__m512) __builtin_ia32_maxps512_mask ((__v16sf) __A,
- (__v16sf) __B,
- (__v16sf)
- _mm512_setzero_ps (),
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512)__builtin_ia32_selectps_512(__U,
+ (__v16sf)_mm512_max_ps(__A, __B),
+ (__v16sf)_mm512_setzero_ps());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_max_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
return (__m128) __builtin_ia32_maxss_round_mask ((__v4sf) __A,
(__v4sf) __B,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_max_ss(__mmask8 __U,__m128 __A, __m128 __B) {
return (__m128) __builtin_ia32_maxss_round_mask ((__v4sf) __A,
(__v4sf) __B,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_max_round_ss(A, B, R) __extension__ ({ \
+#define _mm_max_round_ss(A, B, R) \
(__m128)__builtin_ia32_maxss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_max_round_ss(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_max_round_ss(W, U, A, B, R) \
(__m128)__builtin_ia32_maxss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(W), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-#define _mm_maskz_max_round_ss(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_max_round_ss(U, A, B, R) \
(__m128)__builtin_ia32_maxss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_max_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
return (__m128d) __builtin_ia32_maxsd_round_mask ((__v2df) __A,
(__v2df) __B,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_max_sd(__mmask8 __U,__m128d __A, __m128d __B) {
return (__m128d) __builtin_ia32_maxsd_round_mask ((__v2df) __A,
(__v2df) __B,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_max_round_sd(A, B, R) __extension__ ({ \
+#define _mm_max_round_sd(A, B, R) \
(__m128d)__builtin_ia32_maxsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_max_round_sd(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_max_round_sd(W, U, A, B, R) \
(__m128d)__builtin_ia32_maxsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_maskz_max_round_sd(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_max_round_sd(U, A, B, R) \
(__m128d)__builtin_ia32_maxsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
static __inline __m512i
-__DEFAULT_FN_ATTRS
+__DEFAULT_FN_ATTRS512
_mm512_max_epi32(__m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pmaxsd512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) -1);
+ return (__m512i)__builtin_ia32_pmaxsd512((__v16si)__A, (__v16si)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_max_epi32 (__m512i __W, __mmask16 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pmaxsd512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si) __W, __M);
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
+ (__v16si)_mm512_max_epi32(__A, __B),
+ (__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_max_epi32 (__mmask16 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pmaxsd512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si)
- _mm512_setzero_si512 (),
- __M);
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
+ (__v16si)_mm512_max_epi32(__A, __B),
+ (__v16si)_mm512_setzero_si512());
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_max_epu32(__m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pmaxud512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) -1);
+ return (__m512i)__builtin_ia32_pmaxud512((__v16si)__A, (__v16si)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_max_epu32 (__m512i __W, __mmask16 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pmaxud512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si) __W, __M);
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
+ (__v16si)_mm512_max_epu32(__A, __B),
+ (__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_max_epu32 (__mmask16 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pmaxud512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si)
- _mm512_setzero_si512 (),
- __M);
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
+ (__v16si)_mm512_max_epu32(__A, __B),
+ (__v16si)_mm512_setzero_si512());
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_max_epi64(__m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pmaxsq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) -1);
+ return (__m512i)__builtin_ia32_pmaxsq512((__v8di)__A, (__v8di)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_max_epi64 (__m512i __W, __mmask8 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pmaxsq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di) __W, __M);
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
+ (__v8di)_mm512_max_epi64(__A, __B),
+ (__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_max_epi64 (__mmask8 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pmaxsq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di)
- _mm512_setzero_si512 (),
- __M);
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
+ (__v8di)_mm512_max_epi64(__A, __B),
+ (__v8di)_mm512_setzero_si512());
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_max_epu64(__m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pmaxuq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) -1);
+ return (__m512i)__builtin_ia32_pmaxuq512((__v8di)__A, (__v8di)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_max_epu64 (__m512i __W, __mmask8 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pmaxuq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di) __W, __M);
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
+ (__v8di)_mm512_max_epu64(__A, __B),
+ (__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_max_epu64 (__mmask8 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pmaxuq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di)
- _mm512_setzero_si512 (),
- __M);
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
+ (__v8di)_mm512_max_epu64(__A, __B),
+ (__v8di)_mm512_setzero_si512());
}
-#define _mm512_mask_min_round_pd(W, U, A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_minpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)(__m512d)(W), (__mmask8)(U), \
- (int)(R)); })
+#define _mm512_min_round_pd(A, B, R) \
+ (__m512d)__builtin_ia32_minpd512((__v8df)(__m512d)(A), \
+ (__v8df)(__m512d)(B), (int)(R))
-#define _mm512_maskz_min_round_pd(U, A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_minpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+#define _mm512_mask_min_round_pd(W, U, A, B, R) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_min_round_pd((A), (B), (R)), \
+ (__v8df)(W))
-#define _mm512_min_round_pd(A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_minpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)_mm512_undefined_pd(), \
- (__mmask8)-1, (int)(R)); })
+#define _mm512_maskz_min_round_pd(U, A, B, R) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_min_round_pd((A), (B), (R)), \
+ (__v8df)_mm512_setzero_pd())
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_min_pd(__m512d __A, __m512d __B)
{
- return (__m512d) __builtin_ia32_minpd512_mask ((__v8df) __A,
- (__v8df) __B,
- (__v8df)
- _mm512_setzero_pd (),
- (__mmask8) -1,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d) __builtin_ia32_minpd512((__v8df) __A, (__v8df) __B,
+ _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_min_pd (__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
{
- return (__m512d) __builtin_ia32_minpd512_mask ((__v8df) __A,
- (__v8df) __B,
- (__v8df) __W,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_ia32_selectpd_512(__U,
+ (__v8df)_mm512_min_pd(__A, __B),
+ (__v8df)__W);
}
-#define _mm512_mask_min_round_ps(W, U, A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_minps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)(__m512)(W), (__mmask16)(U), \
- (int)(R)); })
-
-#define _mm512_maskz_min_round_ps(U, A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_minps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U), (int)(R)); })
-
-#define _mm512_min_round_ps(A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_minps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)_mm512_undefined_ps(), \
- (__mmask16)-1, (int)(R)); })
-
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_min_pd (__mmask8 __U, __m512d __A, __m512d __B)
{
- return (__m512d) __builtin_ia32_minpd512_mask ((__v8df) __A,
- (__v8df) __B,
- (__v8df)
- _mm512_setzero_pd (),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_ia32_selectpd_512(__U,
+ (__v8df)_mm512_min_pd(__A, __B),
+ (__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+#define _mm512_min_round_ps(A, B, R) \
+ (__m512)__builtin_ia32_minps512((__v16sf)(__m512)(A), \
+ (__v16sf)(__m512)(B), (int)(R))
+
+#define _mm512_mask_min_round_ps(W, U, A, B, R) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_min_round_ps((A), (B), (R)), \
+ (__v16sf)(W))
+
+#define _mm512_maskz_min_round_ps(U, A, B, R) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_min_round_ps((A), (B), (R)), \
+ (__v16sf)_mm512_setzero_ps())
+
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_min_ps(__m512 __A, __m512 __B)
{
- return (__m512) __builtin_ia32_minps512_mask ((__v16sf) __A,
- (__v16sf) __B,
- (__v16sf)
- _mm512_setzero_ps (),
- (__mmask16) -1,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512) __builtin_ia32_minps512((__v16sf) __A, (__v16sf) __B,
+ _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_min_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
{
- return (__m512) __builtin_ia32_minps512_mask ((__v16sf) __A,
- (__v16sf) __B,
- (__v16sf) __W,
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512)__builtin_ia32_selectps_512(__U,
+ (__v16sf)_mm512_min_ps(__A, __B),
+ (__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_min_ps (__mmask16 __U, __m512 __A, __m512 __B)
{
- return (__m512) __builtin_ia32_minps512_mask ((__v16sf) __A,
- (__v16sf) __B,
- (__v16sf)
- _mm512_setzero_ps (),
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512)__builtin_ia32_selectps_512(__U,
+ (__v16sf)_mm512_min_ps(__A, __B),
+ (__v16sf)_mm512_setzero_ps());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_min_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
return (__m128) __builtin_ia32_minss_round_mask ((__v4sf) __A,
(__v4sf) __B,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_min_ss(__mmask8 __U,__m128 __A, __m128 __B) {
return (__m128) __builtin_ia32_minss_round_mask ((__v4sf) __A,
(__v4sf) __B,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_min_round_ss(A, B, R) __extension__ ({ \
+#define _mm_min_round_ss(A, B, R) \
(__m128)__builtin_ia32_minss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_min_round_ss(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_min_round_ss(W, U, A, B, R) \
(__m128)__builtin_ia32_minss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(W), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-#define _mm_maskz_min_round_ss(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_min_round_ss(U, A, B, R) \
(__m128)__builtin_ia32_minss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_min_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
return (__m128d) __builtin_ia32_minsd_round_mask ((__v2df) __A,
(__v2df) __B,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_min_sd(__mmask8 __U,__m128d __A, __m128d __B) {
return (__m128d) __builtin_ia32_minsd_round_mask ((__v2df) __A,
(__v2df) __B,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_min_round_sd(A, B, R) __extension__ ({ \
+#define _mm_min_round_sd(A, B, R) \
(__m128d)__builtin_ia32_minsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_min_round_sd(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_min_round_sd(W, U, A, B, R) \
(__m128d)__builtin_ia32_minsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_maskz_min_round_sd(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_min_round_sd(U, A, B, R) \
(__m128d)__builtin_ia32_minsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
static __inline __m512i
-__DEFAULT_FN_ATTRS
+__DEFAULT_FN_ATTRS512
_mm512_min_epi32(__m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pminsd512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) -1);
+ return (__m512i)__builtin_ia32_pminsd512((__v16si)__A, (__v16si)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_min_epi32 (__m512i __W, __mmask16 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pminsd512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si) __W, __M);
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
+ (__v16si)_mm512_min_epi32(__A, __B),
+ (__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_min_epi32 (__mmask16 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pminsd512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si)
- _mm512_setzero_si512 (),
- __M);
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
+ (__v16si)_mm512_min_epi32(__A, __B),
+ (__v16si)_mm512_setzero_si512());
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_min_epu32(__m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pminud512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) -1);
+ return (__m512i)__builtin_ia32_pminud512((__v16si)__A, (__v16si)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_min_epu32 (__m512i __W, __mmask16 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pminud512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si) __W, __M);
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
+ (__v16si)_mm512_min_epu32(__A, __B),
+ (__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_min_epu32 (__mmask16 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pminud512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si)
- _mm512_setzero_si512 (),
- __M);
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
+ (__v16si)_mm512_min_epu32(__A, __B),
+ (__v16si)_mm512_setzero_si512());
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_min_epi64(__m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pminsq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) -1);
+ return (__m512i)__builtin_ia32_pminsq512((__v8di)__A, (__v8di)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_min_epi64 (__m512i __W, __mmask8 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pminsq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di) __W, __M);
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
+ (__v8di)_mm512_min_epi64(__A, __B),
+ (__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_min_epi64 (__mmask8 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pminsq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di)
- _mm512_setzero_si512 (),
- __M);
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
+ (__v8di)_mm512_min_epi64(__A, __B),
+ (__v8di)_mm512_setzero_si512());
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_min_epu64(__m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pminuq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) -1);
+ return (__m512i)__builtin_ia32_pminuq512((__v8di)__A, (__v8di)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_min_epu64 (__m512i __W, __mmask8 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pminuq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di) __W, __M);
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
+ (__v8di)_mm512_min_epu64(__A, __B),
+ (__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_min_epu64 (__mmask8 __M, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_pminuq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di)
- _mm512_setzero_si512 (),
- __M);
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
+ (__v8di)_mm512_min_epu64(__A, __B),
+ (__v8di)_mm512_setzero_si512());
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mul_epi32(__m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_pmuldq512((__v16si)__X, (__v16si) __Y);
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_mul_epi32(__m512i __W, __mmask8 __M, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
(__v8di)__W);
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_mul_epi32(__mmask8 __M, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
(__v8di)_mm512_setzero_si512 ());
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mul_epu32(__m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_pmuludq512((__v16si)__X, (__v16si)__Y);
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_mul_epu32(__m512i __W, __mmask8 __M, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
(__v8di)__W);
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_mul_epu32(__mmask8 __M, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
(__v8di)_mm512_setzero_si512 ());
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mullo_epi32 (__m512i __A, __m512i __B)
{
return (__m512i) ((__v16su) __A * (__v16su) __B);
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_mullo_epi32(__mmask16 __M, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
(__v16si)_mm512_setzero_si512());
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_mullo_epi32(__m512i __W, __mmask16 __M, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
(__v16si)__W);
}
-#define _mm512_mask_sqrt_round_pd(W, U, A, R) __extension__ ({ \
- (__m512d)__builtin_ia32_sqrtpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(W), (__mmask8)(U), \
- (int)(R)); })
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mullox_epi64 (__m512i __A, __m512i __B) {
+ return (__m512i) ((__v8du) __A * (__v8du) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_mullox_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B) {
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
+ (__v8di)_mm512_mullox_epi64(__A, __B),
+ (__v8di)__W);
+}
+
+#define _mm512_sqrt_round_pd(A, R) \
+ (__m512d)__builtin_ia32_sqrtpd512((__v8df)(__m512d)(A), (int)(R))
-#define _mm512_maskz_sqrt_round_pd(U, A, R) __extension__ ({ \
- (__m512d)__builtin_ia32_sqrtpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+#define _mm512_mask_sqrt_round_pd(W, U, A, R) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_sqrt_round_pd((A), (R)), \
+ (__v8df)(__m512d)(W))
-#define _mm512_sqrt_round_pd(A, R) __extension__ ({ \
- (__m512d)__builtin_ia32_sqrtpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)_mm512_undefined_pd(), \
- (__mmask8)-1, (int)(R)); })
+#define _mm512_maskz_sqrt_round_pd(U, A, R) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_sqrt_round_pd((A), (R)), \
+ (__v8df)_mm512_setzero_pd())
-static __inline__ __m512d __DEFAULT_FN_ATTRS
-_mm512_sqrt_pd(__m512d __a)
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
+_mm512_sqrt_pd(__m512d __A)
{
- return (__m512d)__builtin_ia32_sqrtpd512_mask((__v8df)__a,
- (__v8df) _mm512_setzero_pd (),
- (__mmask8) -1,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_ia32_sqrtpd512((__v8df)__A,
+ _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_sqrt_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
- return (__m512d) __builtin_ia32_sqrtpd512_mask ((__v8df) __A,
- (__v8df) __W,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_ia32_selectpd_512(__U,
+ (__v8df)_mm512_sqrt_pd(__A),
+ (__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_sqrt_pd (__mmask8 __U, __m512d __A)
{
- return (__m512d) __builtin_ia32_sqrtpd512_mask ((__v8df) __A,
- (__v8df)
- _mm512_setzero_pd (),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_ia32_selectpd_512(__U,
+ (__v8df)_mm512_sqrt_pd(__A),
+ (__v8df)_mm512_setzero_pd());
}
-#define _mm512_mask_sqrt_round_ps(W, U, A, R) __extension__ ({ \
- (__m512)__builtin_ia32_sqrtps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(W), (__mmask16)(U), \
- (int)(R)); })
+#define _mm512_sqrt_round_ps(A, R) \
+ (__m512)__builtin_ia32_sqrtps512((__v16sf)(__m512)(A), (int)(R))
-#define _mm512_maskz_sqrt_round_ps(U, A, R) __extension__ ({ \
- (__m512)__builtin_ia32_sqrtps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U), (int)(R)); })
+#define _mm512_mask_sqrt_round_ps(W, U, A, R) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_sqrt_round_ps((A), (R)), \
+ (__v16sf)(__m512)(W))
-#define _mm512_sqrt_round_ps(A, R) __extension__ ({ \
- (__m512)__builtin_ia32_sqrtps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)_mm512_undefined_ps(), \
- (__mmask16)-1, (int)(R)); })
+#define _mm512_maskz_sqrt_round_ps(U, A, R) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_sqrt_round_ps((A), (R)), \
+ (__v16sf)_mm512_setzero_ps())
-static __inline__ __m512 __DEFAULT_FN_ATTRS
-_mm512_sqrt_ps(__m512 __a)
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
+_mm512_sqrt_ps(__m512 __A)
{
- return (__m512)__builtin_ia32_sqrtps512_mask((__v16sf)__a,
- (__v16sf) _mm512_setzero_ps (),
- (__mmask16) -1,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512)__builtin_ia32_sqrtps512((__v16sf)__A,
+ _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_sqrt_ps(__m512 __W, __mmask16 __U, __m512 __A)
{
- return (__m512)__builtin_ia32_sqrtps512_mask((__v16sf)__A,
- (__v16sf) __W,
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512)__builtin_ia32_selectps_512(__U,
+ (__v16sf)_mm512_sqrt_ps(__A),
+ (__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_sqrt_ps( __mmask16 __U, __m512 __A)
{
- return (__m512)__builtin_ia32_sqrtps512_mask((__v16sf)__A,
- (__v16sf) _mm512_setzero_ps (),
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512)__builtin_ia32_selectps_512(__U,
+ (__v16sf)_mm512_sqrt_ps(__A),
+ (__v16sf)_mm512_setzero_ps());
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_rsqrt14_pd(__m512d __A)
{
return (__m512d) __builtin_ia32_rsqrt14pd512_mask ((__v8df) __A,
_mm512_setzero_pd (),
(__mmask8) -1);}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_rsqrt14_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
return (__m512d) __builtin_ia32_rsqrt14pd512_mask ((__v8df) __A,
(__mmask8) __U);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_rsqrt14_pd (__mmask8 __U, __m512d __A)
{
return (__m512d) __builtin_ia32_rsqrt14pd512_mask ((__v8df) __A,
(__mmask8) __U);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_rsqrt14_ps(__m512 __A)
{
return (__m512) __builtin_ia32_rsqrt14ps512_mask ((__v16sf) __A,
(__mmask16) -1);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_rsqrt14_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
return (__m512) __builtin_ia32_rsqrt14ps512_mask ((__v16sf) __A,
(__mmask16) __U);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_rsqrt14_ps (__mmask16 __U, __m512 __A)
{
return (__m512) __builtin_ia32_rsqrt14ps512_mask ((__v16sf) __A,
(__mmask16) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_rsqrt14_ss(__m128 __A, __m128 __B)
{
return (__m128) __builtin_ia32_rsqrt14ss_mask ((__v4sf) __A,
(__mmask8) -1);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_rsqrt14_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
return (__m128) __builtin_ia32_rsqrt14ss_mask ((__v4sf) __A,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_rsqrt14_ss (__mmask8 __U, __m128 __A, __m128 __B)
{
return (__m128) __builtin_ia32_rsqrt14ss_mask ((__v4sf) __A,
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_rsqrt14_sd(__m128d __A, __m128d __B)
{
return (__m128d) __builtin_ia32_rsqrt14sd_mask ((__v2df) __A,
(__mmask8) -1);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_rsqrt14_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
return (__m128d) __builtin_ia32_rsqrt14sd_mask ( (__v2df) __A,
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_rsqrt14_sd (__mmask8 __U, __m128d __A, __m128d __B)
{
return (__m128d) __builtin_ia32_rsqrt14sd_mask ( (__v2df) __A,
(__mmask8) __U);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_rcp14_pd(__m512d __A)
{
return (__m512d) __builtin_ia32_rcp14pd512_mask ((__v8df) __A,
(__mmask8) -1);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_rcp14_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
return (__m512d) __builtin_ia32_rcp14pd512_mask ((__v8df) __A,
(__mmask8) __U);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_rcp14_pd (__mmask8 __U, __m512d __A)
{
return (__m512d) __builtin_ia32_rcp14pd512_mask ((__v8df) __A,
(__mmask8) __U);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_rcp14_ps(__m512 __A)
{
return (__m512) __builtin_ia32_rcp14ps512_mask ((__v16sf) __A,
(__mmask16) -1);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_rcp14_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
return (__m512) __builtin_ia32_rcp14ps512_mask ((__v16sf) __A,
(__mmask16) __U);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_rcp14_ps (__mmask16 __U, __m512 __A)
{
return (__m512) __builtin_ia32_rcp14ps512_mask ((__v16sf) __A,
(__mmask16) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_rcp14_ss(__m128 __A, __m128 __B)
{
return (__m128) __builtin_ia32_rcp14ss_mask ((__v4sf) __A,
(__mmask8) -1);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_rcp14_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
return (__m128) __builtin_ia32_rcp14ss_mask ((__v4sf) __A,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_rcp14_ss (__mmask8 __U, __m128 __A, __m128 __B)
{
return (__m128) __builtin_ia32_rcp14ss_mask ((__v4sf) __A,
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_rcp14_sd(__m128d __A, __m128d __B)
{
return (__m128d) __builtin_ia32_rcp14sd_mask ((__v2df) __A,
(__mmask8) -1);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_rcp14_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
return (__m128d) __builtin_ia32_rcp14sd_mask ( (__v2df) __A,
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_rcp14_sd (__mmask8 __U, __m128d __A, __m128d __B)
{
return (__m128d) __builtin_ia32_rcp14sd_mask ( (__v2df) __A,
(__mmask8) __U);
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_floor_ps(__m512 __A)
{
return (__m512) __builtin_ia32_rndscaleps_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_floor_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
return (__m512) __builtin_ia32_rndscaleps_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_floor_pd(__m512d __A)
{
return (__m512d) __builtin_ia32_rndscalepd_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_floor_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
return (__m512d) __builtin_ia32_rndscalepd_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_ceil_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
return (__m512) __builtin_ia32_rndscaleps_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_ceil_ps(__m512 __A)
{
return (__m512) __builtin_ia32_rndscaleps_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_ceil_pd(__m512d __A)
{
return (__m512d) __builtin_ia32_rndscalepd_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_ceil_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
return (__m512d) __builtin_ia32_rndscalepd_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_abs_epi64(__m512i __A)
{
- return (__m512i) __builtin_ia32_pabsq512_mask ((__v8di) __A,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) -1);
+ return (__m512i)__builtin_ia32_pabsq512((__v8di)__A);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_abs_epi64 (__m512i __W, __mmask8 __U, __m512i __A)
{
- return (__m512i) __builtin_ia32_pabsq512_mask ((__v8di) __A,
- (__v8di) __W,
- (__mmask8) __U);
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
+ (__v8di)_mm512_abs_epi64(__A),
+ (__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_abs_epi64 (__mmask8 __U, __m512i __A)
{
- return (__m512i) __builtin_ia32_pabsq512_mask ((__v8di) __A,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) __U);
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
+ (__v8di)_mm512_abs_epi64(__A),
+ (__v8di)_mm512_setzero_si512());
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_abs_epi32(__m512i __A)
{
- return (__m512i) __builtin_ia32_pabsd512_mask ((__v16si) __A,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) -1);
+ return (__m512i)__builtin_ia32_pabsd512((__v16si) __A);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_abs_epi32 (__m512i __W, __mmask16 __U, __m512i __A)
{
- return (__m512i) __builtin_ia32_pabsd512_mask ((__v16si) __A,
- (__v16si) __W,
- (__mmask16) __U);
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_abs_epi32(__A),
+ (__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_abs_epi32 (__mmask16 __U, __m512i __A)
{
- return (__m512i) __builtin_ia32_pabsd512_mask ((__v16si) __A,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) __U);
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_abs_epi32(__A),
+ (__v16si)_mm512_setzero_si512());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_add_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
- return (__m128) __builtin_ia32_addss_round_mask ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __W,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_add_ss(__A, __B);
+ return __builtin_ia32_selectss_128(__U, __A, __W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_add_ss(__mmask8 __U,__m128 __A, __m128 __B) {
- return (__m128) __builtin_ia32_addss_round_mask ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) _mm_setzero_ps (),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_add_ss(__A, __B);
+ return __builtin_ia32_selectss_128(__U, __A, _mm_setzero_ps());
}
-#define _mm_add_round_ss(A, B, R) __extension__ ({ \
+#define _mm_add_round_ss(A, B, R) \
(__m128)__builtin_ia32_addss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_add_round_ss(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_add_round_ss(W, U, A, B, R) \
(__m128)__builtin_ia32_addss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(W), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-#define _mm_maskz_add_round_ss(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_add_round_ss(U, A, B, R) \
(__m128)__builtin_ia32_addss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_add_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
- return (__m128d) __builtin_ia32_addsd_round_mask ((__v2df) __A,
- (__v2df) __B,
- (__v2df) __W,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_add_sd(__A, __B);
+ return __builtin_ia32_selectsd_128(__U, __A, __W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_add_sd(__mmask8 __U,__m128d __A, __m128d __B) {
- return (__m128d) __builtin_ia32_addsd_round_mask ((__v2df) __A,
- (__v2df) __B,
- (__v2df) _mm_setzero_pd (),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_add_sd(__A, __B);
+ return __builtin_ia32_selectsd_128(__U, __A, _mm_setzero_pd());
}
-#define _mm_add_round_sd(A, B, R) __extension__ ({ \
+#define _mm_add_round_sd(A, B, R) \
(__m128d)__builtin_ia32_addsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_add_round_sd(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_add_round_sd(W, U, A, B, R) \
(__m128d)__builtin_ia32_addsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_maskz_add_round_sd(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_add_round_sd(U, A, B, R) \
(__m128d)__builtin_ia32_addsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_add_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_add_pd(__A, __B),
(__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_add_pd(__mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_add_pd(__A, __B),
(__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_add_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_add_ps(__A, __B),
(__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_add_ps(__mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_add_ps(__A, __B),
(__v16sf)_mm512_setzero_ps());
}
-#define _mm512_add_round_pd(A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_addpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)_mm512_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
-
-#define _mm512_mask_add_round_pd(W, U, A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_addpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)(__m512d)(W), (__mmask8)(U), \
- (int)(R)); })
-
-#define _mm512_maskz_add_round_pd(U, A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_addpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
-
-#define _mm512_add_round_ps(A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_addps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)_mm512_setzero_ps(), \
- (__mmask16)-1, (int)(R)); })
-
-#define _mm512_mask_add_round_ps(W, U, A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_addps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)(__m512)(W), (__mmask16)(U), \
- (int)(R)); })
-
-#define _mm512_maskz_add_round_ps(U, A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_addps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U), (int)(R)); })
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+#define _mm512_add_round_pd(A, B, R) \
+ (__m512d)__builtin_ia32_addpd512((__v8df)(__m512d)(A), \
+ (__v8df)(__m512d)(B), (int)(R))
+
+#define _mm512_mask_add_round_pd(W, U, A, B, R) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_add_round_pd((A), (B), (R)), \
+ (__v8df)(__m512d)(W))
+
+#define _mm512_maskz_add_round_pd(U, A, B, R) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_add_round_pd((A), (B), (R)), \
+ (__v8df)_mm512_setzero_pd())
+
+#define _mm512_add_round_ps(A, B, R) \
+ (__m512)__builtin_ia32_addps512((__v16sf)(__m512)(A), \
+ (__v16sf)(__m512)(B), (int)(R))
+
+#define _mm512_mask_add_round_ps(W, U, A, B, R) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_add_round_ps((A), (B), (R)), \
+ (__v16sf)(__m512)(W))
+
+#define _mm512_maskz_add_round_ps(U, A, B, R) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_add_round_ps((A), (B), (R)), \
+ (__v16sf)_mm512_setzero_ps())
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_sub_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
- return (__m128) __builtin_ia32_subss_round_mask ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __W,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_sub_ss(__A, __B);
+ return __builtin_ia32_selectss_128(__U, __A, __W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_sub_ss(__mmask8 __U,__m128 __A, __m128 __B) {
- return (__m128) __builtin_ia32_subss_round_mask ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) _mm_setzero_ps (),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_sub_ss(__A, __B);
+ return __builtin_ia32_selectss_128(__U, __A, _mm_setzero_ps());
}
-#define _mm_sub_round_ss(A, B, R) __extension__ ({ \
+#define _mm_sub_round_ss(A, B, R) \
(__m128)__builtin_ia32_subss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_sub_round_ss(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_sub_round_ss(W, U, A, B, R) \
(__m128)__builtin_ia32_subss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(W), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-#define _mm_maskz_sub_round_ss(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_sub_round_ss(U, A, B, R) \
(__m128)__builtin_ia32_subss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_sub_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
- return (__m128d) __builtin_ia32_subsd_round_mask ((__v2df) __A,
- (__v2df) __B,
- (__v2df) __W,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_sub_sd(__A, __B);
+ return __builtin_ia32_selectsd_128(__U, __A, __W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_sub_sd(__mmask8 __U,__m128d __A, __m128d __B) {
- return (__m128d) __builtin_ia32_subsd_round_mask ((__v2df) __A,
- (__v2df) __B,
- (__v2df) _mm_setzero_pd (),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_sub_sd(__A, __B);
+ return __builtin_ia32_selectsd_128(__U, __A, _mm_setzero_pd());
}
-#define _mm_sub_round_sd(A, B, R) __extension__ ({ \
+#define _mm_sub_round_sd(A, B, R) \
(__m128d)__builtin_ia32_subsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_sub_round_sd(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_sub_round_sd(W, U, A, B, R) \
(__m128d)__builtin_ia32_subsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_maskz_sub_round_sd(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_sub_round_sd(U, A, B, R) \
(__m128d)__builtin_ia32_subsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_sub_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_sub_pd(__A, __B),
(__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_sub_pd(__mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_sub_pd(__A, __B),
(__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_sub_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_sub_ps(__A, __B),
(__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_sub_ps(__mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_sub_ps(__A, __B),
(__v16sf)_mm512_setzero_ps());
}
-#define _mm512_sub_round_pd(A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_subpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)_mm512_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
-
-#define _mm512_mask_sub_round_pd(W, U, A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_subpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)(__m512d)(W), (__mmask8)(U), \
- (int)(R)); })
-
-#define _mm512_maskz_sub_round_pd(U, A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_subpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
-
-#define _mm512_sub_round_ps(A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_subps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)_mm512_setzero_ps(), \
- (__mmask16)-1, (int)(R)); })
-
-#define _mm512_mask_sub_round_ps(W, U, A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_subps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)(__m512)(W), (__mmask16)(U), \
- (int)(R)); });
-
-#define _mm512_maskz_sub_round_ps(U, A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_subps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U), (int)(R)); });
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+#define _mm512_sub_round_pd(A, B, R) \
+ (__m512d)__builtin_ia32_subpd512((__v8df)(__m512d)(A), \
+ (__v8df)(__m512d)(B), (int)(R))
+
+#define _mm512_mask_sub_round_pd(W, U, A, B, R) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_sub_round_pd((A), (B), (R)), \
+ (__v8df)(__m512d)(W))
+
+#define _mm512_maskz_sub_round_pd(U, A, B, R) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_sub_round_pd((A), (B), (R)), \
+ (__v8df)_mm512_setzero_pd())
+
+#define _mm512_sub_round_ps(A, B, R) \
+ (__m512)__builtin_ia32_subps512((__v16sf)(__m512)(A), \
+ (__v16sf)(__m512)(B), (int)(R))
+
+#define _mm512_mask_sub_round_ps(W, U, A, B, R) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_sub_round_ps((A), (B), (R)), \
+ (__v16sf)(__m512)(W))
+
+#define _mm512_maskz_sub_round_ps(U, A, B, R) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_sub_round_ps((A), (B), (R)), \
+ (__v16sf)_mm512_setzero_ps())
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_mul_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
- return (__m128) __builtin_ia32_mulss_round_mask ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __W,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_mul_ss(__A, __B);
+ return __builtin_ia32_selectss_128(__U, __A, __W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_mul_ss(__mmask8 __U,__m128 __A, __m128 __B) {
- return (__m128) __builtin_ia32_mulss_round_mask ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) _mm_setzero_ps (),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_mul_ss(__A, __B);
+ return __builtin_ia32_selectss_128(__U, __A, _mm_setzero_ps());
}
-#define _mm_mul_round_ss(A, B, R) __extension__ ({ \
+#define _mm_mul_round_ss(A, B, R) \
(__m128)__builtin_ia32_mulss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_mul_round_ss(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_mul_round_ss(W, U, A, B, R) \
(__m128)__builtin_ia32_mulss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(W), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-#define _mm_maskz_mul_round_ss(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_mul_round_ss(U, A, B, R) \
(__m128)__builtin_ia32_mulss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_mul_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
- return (__m128d) __builtin_ia32_mulsd_round_mask ((__v2df) __A,
- (__v2df) __B,
- (__v2df) __W,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_mul_sd(__A, __B);
+ return __builtin_ia32_selectsd_128(__U, __A, __W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_mul_sd(__mmask8 __U,__m128d __A, __m128d __B) {
- return (__m128d) __builtin_ia32_mulsd_round_mask ((__v2df) __A,
- (__v2df) __B,
- (__v2df) _mm_setzero_pd (),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_mul_sd(__A, __B);
+ return __builtin_ia32_selectsd_128(__U, __A, _mm_setzero_pd());
}
-#define _mm_mul_round_sd(A, B, R) __extension__ ({ \
+#define _mm_mul_round_sd(A, B, R) \
(__m128d)__builtin_ia32_mulsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_mul_round_sd(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_mul_round_sd(W, U, A, B, R) \
(__m128d)__builtin_ia32_mulsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_maskz_mul_round_sd(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_mul_round_sd(U, A, B, R) \
(__m128d)__builtin_ia32_mulsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_mul_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_mul_pd(__A, __B),
(__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_mul_pd(__mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_mul_pd(__A, __B),
(__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_mul_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_mul_ps(__A, __B),
(__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_mul_ps(__mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_mul_ps(__A, __B),
(__v16sf)_mm512_setzero_ps());
}
-#define _mm512_mul_round_pd(A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_mulpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)_mm512_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
-
-#define _mm512_mask_mul_round_pd(W, U, A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_mulpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)(__m512d)(W), (__mmask8)(U), \
- (int)(R)); })
-
-#define _mm512_maskz_mul_round_pd(U, A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_mulpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
-
-#define _mm512_mul_round_ps(A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_mulps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)_mm512_setzero_ps(), \
- (__mmask16)-1, (int)(R)); })
-
-#define _mm512_mask_mul_round_ps(W, U, A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_mulps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)(__m512)(W), (__mmask16)(U), \
- (int)(R)); });
-
-#define _mm512_maskz_mul_round_ps(U, A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_mulps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U), (int)(R)); });
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+#define _mm512_mul_round_pd(A, B, R) \
+ (__m512d)__builtin_ia32_mulpd512((__v8df)(__m512d)(A), \
+ (__v8df)(__m512d)(B), (int)(R))
+
+#define _mm512_mask_mul_round_pd(W, U, A, B, R) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_mul_round_pd((A), (B), (R)), \
+ (__v8df)(__m512d)(W))
+
+#define _mm512_maskz_mul_round_pd(U, A, B, R) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_mul_round_pd((A), (B), (R)), \
+ (__v8df)_mm512_setzero_pd())
+
+#define _mm512_mul_round_ps(A, B, R) \
+ (__m512)__builtin_ia32_mulps512((__v16sf)(__m512)(A), \
+ (__v16sf)(__m512)(B), (int)(R))
+
+#define _mm512_mask_mul_round_ps(W, U, A, B, R) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_mul_round_ps((A), (B), (R)), \
+ (__v16sf)(__m512)(W))
+
+#define _mm512_maskz_mul_round_ps(U, A, B, R) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_mul_round_ps((A), (B), (R)), \
+ (__v16sf)_mm512_setzero_ps())
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_div_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
- return (__m128) __builtin_ia32_divss_round_mask ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __W,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_div_ss(__A, __B);
+ return __builtin_ia32_selectss_128(__U, __A, __W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_div_ss(__mmask8 __U,__m128 __A, __m128 __B) {
- return (__m128) __builtin_ia32_divss_round_mask ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) _mm_setzero_ps (),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_div_ss(__A, __B);
+ return __builtin_ia32_selectss_128(__U, __A, _mm_setzero_ps());
}
-#define _mm_div_round_ss(A, B, R) __extension__ ({ \
+#define _mm_div_round_ss(A, B, R) \
(__m128)__builtin_ia32_divss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_div_round_ss(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_div_round_ss(W, U, A, B, R) \
(__m128)__builtin_ia32_divss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(W), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-#define _mm_maskz_div_round_ss(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_div_round_ss(U, A, B, R) \
(__m128)__builtin_ia32_divss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_div_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
- return (__m128d) __builtin_ia32_divsd_round_mask ((__v2df) __A,
- (__v2df) __B,
- (__v2df) __W,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_div_sd(__A, __B);
+ return __builtin_ia32_selectsd_128(__U, __A, __W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_div_sd(__mmask8 __U,__m128d __A, __m128d __B) {
- return (__m128d) __builtin_ia32_divsd_round_mask ((__v2df) __A,
- (__v2df) __B,
- (__v2df) _mm_setzero_pd (),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ __A = _mm_div_sd(__A, __B);
+ return __builtin_ia32_selectsd_128(__U, __A, _mm_setzero_pd());
}
-#define _mm_div_round_sd(A, B, R) __extension__ ({ \
+#define _mm_div_round_sd(A, B, R) \
(__m128d)__builtin_ia32_divsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_div_round_sd(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_div_round_sd(W, U, A, B, R) \
(__m128d)__builtin_ia32_divsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_maskz_div_round_sd(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_div_round_sd(U, A, B, R) \
(__m128d)__builtin_ia32_divsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_div_pd(__m512d __a, __m512d __b)
{
return (__m512d)((__v8df)__a/(__v8df)__b);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_div_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_div_pd(__A, __B),
(__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_div_pd(__mmask8 __U, __m512d __A, __m512d __B) {
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_div_pd(__A, __B),
(__v8df)_mm512_setzero_pd());
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_div_ps(__m512 __a, __m512 __b)
{
return (__m512)((__v16sf)__a/(__v16sf)__b);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_div_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_div_ps(__A, __B),
(__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_div_ps(__mmask16 __U, __m512 __A, __m512 __B) {
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_div_ps(__A, __B),
(__v16sf)_mm512_setzero_ps());
}
-#define _mm512_div_round_pd(A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_divpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)_mm512_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
-
-#define _mm512_mask_div_round_pd(W, U, A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_divpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)(__m512d)(W), (__mmask8)(U), \
- (int)(R)); })
-
-#define _mm512_maskz_div_round_pd(U, A, B, R) __extension__ ({ \
- (__m512d)__builtin_ia32_divpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
-
-#define _mm512_div_round_ps(A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_divps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)_mm512_setzero_ps(), \
- (__mmask16)-1, (int)(R)); })
-
-#define _mm512_mask_div_round_ps(W, U, A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_divps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)(__m512)(W), (__mmask16)(U), \
- (int)(R)); });
-
-#define _mm512_maskz_div_round_ps(U, A, B, R) __extension__ ({ \
- (__m512)__builtin_ia32_divps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U), (int)(R)); });
-
-#define _mm512_roundscale_ps(A, B) __extension__ ({ \
+#define _mm512_div_round_pd(A, B, R) \
+ (__m512d)__builtin_ia32_divpd512((__v8df)(__m512d)(A), \
+ (__v8df)(__m512d)(B), (int)(R))
+
+#define _mm512_mask_div_round_pd(W, U, A, B, R) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_div_round_pd((A), (B), (R)), \
+ (__v8df)(__m512d)(W))
+
+#define _mm512_maskz_div_round_pd(U, A, B, R) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_div_round_pd((A), (B), (R)), \
+ (__v8df)_mm512_setzero_pd())
+
+#define _mm512_div_round_ps(A, B, R) \
+ (__m512)__builtin_ia32_divps512((__v16sf)(__m512)(A), \
+ (__v16sf)(__m512)(B), (int)(R))
+
+#define _mm512_mask_div_round_ps(W, U, A, B, R) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_div_round_ps((A), (B), (R)), \
+ (__v16sf)(__m512)(W))
+
+#define _mm512_maskz_div_round_ps(U, A, B, R) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_div_round_ps((A), (B), (R)), \
+ (__v16sf)_mm512_setzero_ps())
+
+#define _mm512_roundscale_ps(A, B) \
(__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(A), (int)(B), \
- (__v16sf)(__m512)(A), (__mmask16)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ (__v16sf)_mm512_undefined_ps(), \
+ (__mmask16)-1, \
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_mask_roundscale_ps(A, B, C, imm) __extension__ ({\
+#define _mm512_mask_roundscale_ps(A, B, C, imm) \
(__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(C), (int)(imm), \
(__v16sf)(__m512)(A), (__mmask16)(B), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_maskz_roundscale_ps(A, B, imm) __extension__ ({\
+#define _mm512_maskz_roundscale_ps(A, B, imm) \
(__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(B), (int)(imm), \
(__v16sf)_mm512_setzero_ps(), \
(__mmask16)(A), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_mask_roundscale_round_ps(A, B, C, imm, R) __extension__ ({ \
+#define _mm512_mask_roundscale_round_ps(A, B, C, imm, R) \
(__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(C), (int)(imm), \
(__v16sf)(__m512)(A), (__mmask16)(B), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_maskz_roundscale_round_ps(A, B, imm, R) __extension__ ({ \
+#define _mm512_maskz_roundscale_round_ps(A, B, imm, R) \
(__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(B), (int)(imm), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(A), (int)(R)); })
+ (__mmask16)(A), (int)(R))
-#define _mm512_roundscale_round_ps(A, imm, R) __extension__ ({ \
+#define _mm512_roundscale_round_ps(A, imm, R) \
(__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(A), (int)(imm), \
(__v16sf)_mm512_undefined_ps(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_roundscale_pd(A, B) __extension__ ({ \
+#define _mm512_roundscale_pd(A, B) \
(__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(A), (int)(B), \
- (__v8df)(__m512d)(A), (__mmask8)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ (__v8df)_mm512_undefined_pd(), \
+ (__mmask8)-1, \
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_mask_roundscale_pd(A, B, C, imm) __extension__ ({\
+#define _mm512_mask_roundscale_pd(A, B, C, imm) \
(__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(C), (int)(imm), \
(__v8df)(__m512d)(A), (__mmask8)(B), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_maskz_roundscale_pd(A, B, imm) __extension__ ({\
+#define _mm512_maskz_roundscale_pd(A, B, imm) \
(__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(B), (int)(imm), \
(__v8df)_mm512_setzero_pd(), \
(__mmask8)(A), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_mask_roundscale_round_pd(A, B, C, imm, R) __extension__ ({ \
+#define _mm512_mask_roundscale_round_pd(A, B, C, imm, R) \
(__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(C), (int)(imm), \
(__v8df)(__m512d)(A), (__mmask8)(B), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_maskz_roundscale_round_pd(A, B, imm, R) __extension__ ({ \
+#define _mm512_maskz_roundscale_round_pd(A, B, imm, R) \
(__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(B), (int)(imm), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)(A), (int)(R)); })
+ (__mmask8)(A), (int)(R))
-#define _mm512_roundscale_round_pd(A, imm, R) __extension__ ({ \
+#define _mm512_roundscale_round_pd(A, imm, R) \
(__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(A), (int)(imm), \
(__v8df)_mm512_undefined_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_fmadd_round_pd(A, B, C, R) __extension__ ({ \
+#define _mm512_fmadd_round_pd(A, B, C, R) \
(__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
- (__v8df)(__m512d)(C), (__mmask8)-1, \
- (int)(R)); })
+ (__v8df)(__m512d)(C), \
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_fmadd_round_pd(A, U, B, C, R) __extension__ ({ \
+#define _mm512_mask_fmadd_round_pd(A, U, B, C, R) \
(__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_mask3_fmadd_round_pd(A, B, C, U, R) __extension__ ({ \
+#define _mm512_mask3_fmadd_round_pd(A, B, C, U, R) \
(__m512d)__builtin_ia32_vfmaddpd512_mask3((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_fmadd_round_pd(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_fmadd_round_pd(U, A, B, C, R) \
(__m512d)__builtin_ia32_vfmaddpd512_maskz((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_fmsub_round_pd(A, B, C, R) __extension__ ({ \
+#define _mm512_fmsub_round_pd(A, B, C, R) \
(__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
-(__v8df)(__m512d)(C), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_fmsub_round_pd(A, U, B, C, R) __extension__ ({ \
+#define _mm512_mask_fmsub_round_pd(A, U, B, C, R) \
(__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
-(__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_fmsub_round_pd(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_fmsub_round_pd(U, A, B, C, R) \
(__m512d)__builtin_ia32_vfmaddpd512_maskz((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
-(__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_fnmadd_round_pd(A, B, C, R) __extension__ ({ \
+#define _mm512_fnmadd_round_pd(A, B, C, R) \
(__m512d)__builtin_ia32_vfmaddpd512_mask(-(__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
- (__v8df)(__m512d)(C), (__mmask8)-1, \
- (int)(R)); })
+ (__v8df)(__m512d)(C), \
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask3_fnmadd_round_pd(A, B, C, U, R) __extension__ ({ \
+#define _mm512_mask3_fnmadd_round_pd(A, B, C, U, R) \
(__m512d)__builtin_ia32_vfmaddpd512_mask3(-(__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_fnmadd_round_pd(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_fnmadd_round_pd(U, A, B, C, R) \
(__m512d)__builtin_ia32_vfmaddpd512_maskz(-(__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_fnmsub_round_pd(A, B, C, R) __extension__ ({ \
+#define _mm512_fnmsub_round_pd(A, B, C, R) \
(__m512d)__builtin_ia32_vfmaddpd512_mask(-(__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
-(__v8df)(__m512d)(C), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_maskz_fnmsub_round_pd(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_fnmsub_round_pd(U, A, B, C, R) \
(__m512d)__builtin_ia32_vfmaddpd512_maskz(-(__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
-(__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_fmadd_pd(__m512d __A, __m512d __B, __m512d __C)
{
return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_fmadd_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
{
return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask3_fmadd_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
{
return (__m512d) __builtin_ia32_vfmaddpd512_mask3 ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_fmadd_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
{
return (__m512d) __builtin_ia32_vfmaddpd512_maskz ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_fmsub_pd(__m512d __A, __m512d __B, __m512d __C)
{
return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_fmsub_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
{
return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_fmsub_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
{
return (__m512d) __builtin_ia32_vfmaddpd512_maskz ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_fnmadd_pd(__m512d __A, __m512d __B, __m512d __C)
{
- return (__m512d) __builtin_ia32_vfmaddpd512_mask (-(__v8df) __A,
- (__v8df) __B,
+ return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
+ -(__v8df) __B,
(__v8df) __C,
(__mmask8) -1,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask3_fnmadd_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
{
return (__m512d) __builtin_ia32_vfmaddpd512_mask3 (-(__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_fnmadd_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
{
return (__m512d) __builtin_ia32_vfmaddpd512_maskz (-(__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_fnmsub_pd(__m512d __A, __m512d __B, __m512d __C)
{
- return (__m512d) __builtin_ia32_vfmaddpd512_mask (-(__v8df) __A,
- (__v8df) __B,
+ return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
+ -(__v8df) __B,
-(__v8df) __C,
(__mmask8) -1,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_fnmsub_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
{
return (__m512d) __builtin_ia32_vfmaddpd512_maskz (-(__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_fmadd_round_ps(A, B, C, R) __extension__ ({ \
+#define _mm512_fmadd_round_ps(A, B, C, R) \
(__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
- (__v16sf)(__m512)(C), (__mmask16)-1, \
- (int)(R)); })
+ (__v16sf)(__m512)(C), \
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_fmadd_round_ps(A, U, B, C, R) __extension__ ({ \
+#define _mm512_mask_fmadd_round_ps(A, U, B, C, R) \
(__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_mask3_fmadd_round_ps(A, B, C, U, R) __extension__ ({ \
+#define _mm512_mask3_fmadd_round_ps(A, B, C, U, R) \
(__m512)__builtin_ia32_vfmaddps512_mask3((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_fmadd_round_ps(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_fmadd_round_ps(U, A, B, C, R) \
(__m512)__builtin_ia32_vfmaddps512_maskz((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_fmsub_round_ps(A, B, C, R) __extension__ ({ \
+#define _mm512_fmsub_round_ps(A, B, C, R) \
(__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
-(__v16sf)(__m512)(C), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_fmsub_round_ps(A, U, B, C, R) __extension__ ({ \
+#define _mm512_mask_fmsub_round_ps(A, U, B, C, R) \
(__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
-(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_fmsub_round_ps(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_fmsub_round_ps(U, A, B, C, R) \
(__m512)__builtin_ia32_vfmaddps512_maskz((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
-(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_fnmadd_round_ps(A, B, C, R) __extension__ ({ \
- (__m512)__builtin_ia32_vfmaddps512_mask(-(__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)(__m512)(C), (__mmask16)-1, \
- (int)(R)); })
+#define _mm512_fnmadd_round_ps(A, B, C, R) \
+ (__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
+ -(__v16sf)(__m512)(B), \
+ (__v16sf)(__m512)(C), \
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask3_fnmadd_round_ps(A, B, C, U, R) __extension__ ({ \
+#define _mm512_mask3_fnmadd_round_ps(A, B, C, U, R) \
(__m512)__builtin_ia32_vfmaddps512_mask3(-(__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_fnmadd_round_ps(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_fnmadd_round_ps(U, A, B, C, R) \
(__m512)__builtin_ia32_vfmaddps512_maskz(-(__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_fnmsub_round_ps(A, B, C, R) __extension__ ({ \
- (__m512)__builtin_ia32_vfmaddps512_mask(-(__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
+#define _mm512_fnmsub_round_ps(A, B, C, R) \
+ (__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
+ -(__v16sf)(__m512)(B), \
-(__v16sf)(__m512)(C), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_maskz_fnmsub_round_ps(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_fnmsub_round_ps(U, A, B, C, R) \
(__m512)__builtin_ia32_vfmaddps512_maskz(-(__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
-(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_fmadd_ps(__m512 __A, __m512 __B, __m512 __C)
{
return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_fmadd_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
{
return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask3_fmadd_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
{
return (__m512) __builtin_ia32_vfmaddps512_mask3 ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_fmadd_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
{
return (__m512) __builtin_ia32_vfmaddps512_maskz ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_fmsub_ps(__m512 __A, __m512 __B, __m512 __C)
{
return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_fmsub_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
{
return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_fmsub_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
{
return (__m512) __builtin_ia32_vfmaddps512_maskz ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_fnmadd_ps(__m512 __A, __m512 __B, __m512 __C)
{
- return (__m512) __builtin_ia32_vfmaddps512_mask (-(__v16sf) __A,
- (__v16sf) __B,
+ return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
+ -(__v16sf) __B,
(__v16sf) __C,
(__mmask16) -1,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask3_fnmadd_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
{
return (__m512) __builtin_ia32_vfmaddps512_mask3 (-(__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_fnmadd_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
{
return (__m512) __builtin_ia32_vfmaddps512_maskz (-(__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_fnmsub_ps(__m512 __A, __m512 __B, __m512 __C)
{
- return (__m512) __builtin_ia32_vfmaddps512_mask (-(__v16sf) __A,
- (__v16sf) __B,
+ return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
+ -(__v16sf) __B,
-(__v16sf) __C,
(__mmask16) -1,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_fnmsub_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
{
return (__m512) __builtin_ia32_vfmaddps512_maskz (-(__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_fmaddsub_round_pd(A, B, C, R) __extension__ ({ \
+#define _mm512_fmaddsub_round_pd(A, B, C, R) \
(__m512d)__builtin_ia32_vfmaddsubpd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8df)(__m512d)(C), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_fmaddsub_round_pd(A, U, B, C, R) __extension__ ({ \
+#define _mm512_mask_fmaddsub_round_pd(A, U, B, C, R) \
(__m512d)__builtin_ia32_vfmaddsubpd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_mask3_fmaddsub_round_pd(A, B, C, U, R) __extension__ ({ \
+#define _mm512_mask3_fmaddsub_round_pd(A, B, C, U, R) \
(__m512d)__builtin_ia32_vfmaddsubpd512_mask3((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_fmaddsub_round_pd(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_fmaddsub_round_pd(U, A, B, C, R) \
(__m512d)__builtin_ia32_vfmaddsubpd512_maskz((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_fmsubadd_round_pd(A, B, C, R) __extension__ ({ \
+#define _mm512_fmsubadd_round_pd(A, B, C, R) \
(__m512d)__builtin_ia32_vfmaddsubpd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
-(__v8df)(__m512d)(C), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_fmsubadd_round_pd(A, U, B, C, R) __extension__ ({ \
+#define _mm512_mask_fmsubadd_round_pd(A, U, B, C, R) \
(__m512d)__builtin_ia32_vfmaddsubpd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
-(__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_fmsubadd_round_pd(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_fmsubadd_round_pd(U, A, B, C, R) \
(__m512d)__builtin_ia32_vfmaddsubpd512_maskz((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
-(__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_fmaddsub_pd(__m512d __A, __m512d __B, __m512d __C)
{
return (__m512d) __builtin_ia32_vfmaddsubpd512_mask ((__v8df) __A,
- (__v8df) __B,
- (__v8df) __C,
- (__mmask8) -1,
- _MM_FROUND_CUR_DIRECTION);
+ (__v8df) __B,
+ (__v8df) __C,
+ (__mmask8) -1,
+ _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_fmaddsub_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
{
return (__m512d) __builtin_ia32_vfmaddsubpd512_mask ((__v8df) __A,
- (__v8df) __B,
- (__v8df) __C,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ (__v8df) __B,
+ (__v8df) __C,
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask3_fmaddsub_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
{
return (__m512d) __builtin_ia32_vfmaddsubpd512_mask3 ((__v8df) __A,
- (__v8df) __B,
- (__v8df) __C,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ (__v8df) __B,
+ (__v8df) __C,
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_fmaddsub_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
{
return (__m512d) __builtin_ia32_vfmaddsubpd512_maskz ((__v8df) __A,
- (__v8df) __B,
- (__v8df) __C,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ (__v8df) __B,
+ (__v8df) __C,
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_fmsubadd_pd(__m512d __A, __m512d __B, __m512d __C)
{
return (__m512d) __builtin_ia32_vfmaddsubpd512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_fmsubadd_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
{
return (__m512d) __builtin_ia32_vfmaddsubpd512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_fmsubadd_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
{
return (__m512d) __builtin_ia32_vfmaddsubpd512_maskz ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_fmaddsub_round_ps(A, B, C, R) __extension__ ({ \
+#define _mm512_fmaddsub_round_ps(A, B, C, R) \
(__m512)__builtin_ia32_vfmaddsubps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16sf)(__m512)(C), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_fmaddsub_round_ps(A, U, B, C, R) __extension__ ({ \
+#define _mm512_mask_fmaddsub_round_ps(A, U, B, C, R) \
(__m512)__builtin_ia32_vfmaddsubps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_mask3_fmaddsub_round_ps(A, B, C, U, R) __extension__ ({ \
+#define _mm512_mask3_fmaddsub_round_ps(A, B, C, U, R) \
(__m512)__builtin_ia32_vfmaddsubps512_mask3((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_fmaddsub_round_ps(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_fmaddsub_round_ps(U, A, B, C, R) \
(__m512)__builtin_ia32_vfmaddsubps512_maskz((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_fmsubadd_round_ps(A, B, C, R) __extension__ ({ \
+#define _mm512_fmsubadd_round_ps(A, B, C, R) \
(__m512)__builtin_ia32_vfmaddsubps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
-(__v16sf)(__m512)(C), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_fmsubadd_round_ps(A, U, B, C, R) __extension__ ({ \
+#define _mm512_mask_fmsubadd_round_ps(A, U, B, C, R) \
(__m512)__builtin_ia32_vfmaddsubps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
-(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_fmsubadd_round_ps(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_fmsubadd_round_ps(U, A, B, C, R) \
(__m512)__builtin_ia32_vfmaddsubps512_maskz((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
-(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_fmaddsub_ps(__m512 __A, __m512 __B, __m512 __C)
{
return (__m512) __builtin_ia32_vfmaddsubps512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_fmaddsub_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
{
return (__m512) __builtin_ia32_vfmaddsubps512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask3_fmaddsub_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
{
return (__m512) __builtin_ia32_vfmaddsubps512_mask3 ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_fmaddsub_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
{
return (__m512) __builtin_ia32_vfmaddsubps512_maskz ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_fmsubadd_ps(__m512 __A, __m512 __B, __m512 __C)
{
return (__m512) __builtin_ia32_vfmaddsubps512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_fmsubadd_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
{
return (__m512) __builtin_ia32_vfmaddsubps512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_fmsubadd_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
{
return (__m512) __builtin_ia32_vfmaddsubps512_maskz ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_mask3_fmsub_round_pd(A, B, C, U, R) __extension__ ({ \
+#define _mm512_mask3_fmsub_round_pd(A, B, C, U, R) \
(__m512d)__builtin_ia32_vfmsubpd512_mask3((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask3_fmsub_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
{
- return (__m512d) __builtin_ia32_vfmsubpd512_mask3 ((__v8df) __A,
- (__v8df) __B,
- (__v8df) __C,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_ia32_vfmsubpd512_mask3 ((__v8df) __A,
+ (__v8df) __B,
+ (__v8df) __C,
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_mask3_fmsub_round_ps(A, B, C, U, R) __extension__ ({ \
+#define _mm512_mask3_fmsub_round_ps(A, B, C, U, R) \
(__m512)__builtin_ia32_vfmsubps512_mask3((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask3_fmsub_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
{
- return (__m512) __builtin_ia32_vfmsubps512_mask3 ((__v16sf) __A,
- (__v16sf) __B,
- (__v16sf) __C,
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512)__builtin_ia32_vfmsubps512_mask3 ((__v16sf) __A,
+ (__v16sf) __B,
+ (__v16sf) __C,
+ (__mmask16) __U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_mask3_fmsubadd_round_pd(A, B, C, U, R) __extension__ ({ \
+#define _mm512_mask3_fmsubadd_round_pd(A, B, C, U, R) \
(__m512d)__builtin_ia32_vfmsubaddpd512_mask3((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask3_fmsubadd_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
{
- return (__m512d) __builtin_ia32_vfmsubaddpd512_mask3 ((__v8df) __A,
- (__v8df) __B,
- (__v8df) __C,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_ia32_vfmsubaddpd512_mask3 ((__v8df) __A,
+ (__v8df) __B,
+ (__v8df) __C,
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_mask3_fmsubadd_round_ps(A, B, C, U, R) __extension__ ({ \
+#define _mm512_mask3_fmsubadd_round_ps(A, B, C, U, R) \
(__m512)__builtin_ia32_vfmsubaddps512_mask3((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask3_fmsubadd_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
{
- return (__m512) __builtin_ia32_vfmsubaddps512_mask3 ((__v16sf) __A,
- (__v16sf) __B,
- (__v16sf) __C,
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512)__builtin_ia32_vfmsubaddps512_mask3 ((__v16sf) __A,
+ (__v16sf) __B,
+ (__v16sf) __C,
+ (__mmask16) __U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_mask_fnmadd_round_pd(A, U, B, C, R) __extension__ ({ \
- (__m512d)__builtin_ia32_vfnmaddpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+#define _mm512_mask_fnmadd_round_pd(A, U, B, C, R) \
+ (__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
+ -(__v8df)(__m512d)(B), \
+ (__v8df)(__m512d)(C), \
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_fnmadd_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
{
- return (__m512d) __builtin_ia32_vfnmaddpd512_mask ((__v8df) __A,
- (__v8df) __B,
- (__v8df) __C,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
+ -(__v8df) __B,
+ (__v8df) __C,
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_mask_fnmadd_round_ps(A, U, B, C, R) __extension__ ({ \
- (__m512)__builtin_ia32_vfnmaddps512_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+#define _mm512_mask_fnmadd_round_ps(A, U, B, C, R) \
+ (__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
+ -(__v16sf)(__m512)(B), \
+ (__v16sf)(__m512)(C), \
+ (__mmask16)(U), (int)(R))
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_fnmadd_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
{
- return (__m512) __builtin_ia32_vfnmaddps512_mask ((__v16sf) __A,
- (__v16sf) __B,
- (__v16sf) __C,
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
+ -(__v16sf) __B,
+ (__v16sf) __C,
+ (__mmask16) __U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_mask_fnmsub_round_pd(A, U, B, C, R) __extension__ ({ \
- (__m512d)__builtin_ia32_vfnmsubpd512_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+#define _mm512_mask_fnmsub_round_pd(A, U, B, C, R) \
+ (__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
+ -(__v8df)(__m512d)(B), \
+ -(__v8df)(__m512d)(C), \
+ (__mmask8)(U), (int)(R))
-#define _mm512_mask3_fnmsub_round_pd(A, B, C, U, R) __extension__ ({ \
- (__m512d)__builtin_ia32_vfnmsubpd512_mask3((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- (__v8df)(__m512d)(C), \
- (__mmask8)(U), (int)(R)); })
+#define _mm512_mask3_fnmsub_round_pd(A, B, C, U, R) \
+ (__m512d)__builtin_ia32_vfmsubpd512_mask3(-(__v8df)(__m512d)(A), \
+ (__v8df)(__m512d)(B), \
+ (__v8df)(__m512d)(C), \
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_fnmsub_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
{
- return (__m512d) __builtin_ia32_vfnmsubpd512_mask ((__v8df) __A,
+ return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
+ -(__v8df) __B,
+ -(__v8df) __C,
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
+_mm512_mask3_fnmsub_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
+{
+ return (__m512d) __builtin_ia32_vfmsubpd512_mask3 (-(__v8df) __A,
(__v8df) __B,
(__v8df) __C,
(__mmask8) __U,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
-_mm512_mask3_fnmsub_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
-{
- return (__m512d) __builtin_ia32_vfnmsubpd512_mask3 ((__v8df) __A,
- (__v8df) __B,
- (__v8df) __C,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
-}
+#define _mm512_mask_fnmsub_round_ps(A, U, B, C, R) \
+ (__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
+ -(__v16sf)(__m512)(B), \
+ -(__v16sf)(__m512)(C), \
+ (__mmask16)(U), (int)(R))
-#define _mm512_mask_fnmsub_round_ps(A, U, B, C, R) __extension__ ({ \
- (__m512)__builtin_ia32_vfnmsubps512_mask((__v16sf)(__m512)(A), \
+
+#define _mm512_mask3_fnmsub_round_ps(A, B, C, U, R) \
+ (__m512)__builtin_ia32_vfmsubps512_mask3(-(__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
-
-
-#define _mm512_mask3_fnmsub_round_ps(A, B, C, U, R) __extension__ ({ \
- (__m512)__builtin_ia32_vfnmsubps512_mask3((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- (__v16sf)(__m512)(C), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_fnmsub_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
{
- return (__m512) __builtin_ia32_vfnmsubps512_mask ((__v16sf) __A,
- (__v16sf) __B,
- (__v16sf) __C,
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
+ -(__v16sf) __B,
+ -(__v16sf) __C,
+ (__mmask16) __U,
+ _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask3_fnmsub_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
{
- return (__m512) __builtin_ia32_vfnmsubps512_mask3 ((__v16sf) __A,
- (__v16sf) __B,
- (__v16sf) __C,
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512) __builtin_ia32_vfmsubps512_mask3 (-(__v16sf) __A,
+ (__v16sf) __B,
+ (__v16sf) __C,
+ (__mmask16) __U,
+ _MM_FROUND_CUR_DIRECTION);
}
/* Vector permutations */
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_permutex2var_epi32(__m512i __A, __m512i __I, __m512i __B)
{
- return (__m512i) __builtin_ia32_vpermt2vard512_mask ((__v16si) __I
- /* idx */ ,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) -1);
+ return (__m512i)__builtin_ia32_vpermi2vard512((__v16si)__A, (__v16si) __I,
+ (__v16si) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_permutex2var_epi32(__m512i __A, __mmask16 __U, __m512i __I,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_permutex2var_epi32(__A, __I, __B),
+ (__v16si)__A);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_permutex2var_epi32 (__m512i __A, __mmask16 __U,
- __m512i __I, __m512i __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask2_permutex2var_epi32(__m512i __A, __m512i __I, __mmask16 __U,
+ __m512i __B)
{
- return (__m512i) __builtin_ia32_vpermt2vard512_mask ((__v16si) __I
- /* idx */ ,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_permutex2var_epi32(__A, __I, __B),
+ (__v16si)__I);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_permutex2var_epi32 (__mmask16 __U, __m512i __A,
- __m512i __I, __m512i __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_permutex2var_epi32(__mmask16 __U, __m512i __A, __m512i __I,
+ __m512i __B)
{
- return (__m512i) __builtin_ia32_vpermt2vard512_maskz ((__v16si) __I
- /* idx */ ,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_permutex2var_epi32(__A, __I, __B),
+ (__v16si)_mm512_setzero_si512());
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_permutex2var_epi64(__m512i __A, __m512i __I, __m512i __B)
{
- return (__m512i) __builtin_ia32_vpermt2varq512_mask ((__v8di) __I
- /* idx */ ,
- (__v8di) __A,
- (__v8di) __B,
- (__mmask8) -1);
+ return (__m512i)__builtin_ia32_vpermi2varq512((__v8di)__A, (__v8di) __I,
+ (__v8di) __B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_permutex2var_epi64 (__m512i __A, __mmask8 __U, __m512i __I,
- __m512i __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_permutex2var_epi64(__m512i __A, __mmask8 __U, __m512i __I,
+ __m512i __B)
{
- return (__m512i) __builtin_ia32_vpermt2varq512_mask ((__v8di) __I
- /* idx */ ,
- (__v8di) __A,
- (__v8di) __B,
- (__mmask8) __U);
+ return (__m512i)__builtin_ia32_selectq_512(__U,
+ (__v8di)_mm512_permutex2var_epi64(__A, __I, __B),
+ (__v8di)__A);
}
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask2_permutex2var_epi64(__m512i __A, __m512i __I, __mmask8 __U,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectq_512(__U,
+ (__v8di)_mm512_permutex2var_epi64(__A, __I, __B),
+ (__v8di)__I);
+}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_permutex2var_epi64 (__mmask8 __U, __m512i __A,
- __m512i __I, __m512i __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_permutex2var_epi64(__mmask8 __U, __m512i __A, __m512i __I,
+ __m512i __B)
{
- return (__m512i) __builtin_ia32_vpermt2varq512_maskz ((__v8di) __I
- /* idx */ ,
- (__v8di) __A,
- (__v8di) __B,
- (__mmask8) __U);
+ return (__m512i)__builtin_ia32_selectq_512(__U,
+ (__v8di)_mm512_permutex2var_epi64(__A, __I, __B),
+ (__v8di)_mm512_setzero_si512());
}
-#define _mm512_alignr_epi64(A, B, I) __extension__ ({ \
- (__m512i)__builtin_shufflevector((__v8di)(__m512i)(B), \
- (__v8di)(__m512i)(A), \
- ((int)(I) & 0x7) + 0, \
- ((int)(I) & 0x7) + 1, \
- ((int)(I) & 0x7) + 2, \
- ((int)(I) & 0x7) + 3, \
- ((int)(I) & 0x7) + 4, \
- ((int)(I) & 0x7) + 5, \
- ((int)(I) & 0x7) + 6, \
- ((int)(I) & 0x7) + 7); })
+#define _mm512_alignr_epi64(A, B, I) \
+ (__m512i)__builtin_ia32_alignq512((__v8di)(__m512i)(A), \
+ (__v8di)(__m512i)(B), (int)(I))
-#define _mm512_mask_alignr_epi64(W, U, A, B, imm) __extension__({\
+#define _mm512_mask_alignr_epi64(W, U, A, B, imm) \
(__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
(__v8di)_mm512_alignr_epi64((A), (B), (imm)), \
- (__v8di)(__m512i)(W)); })
+ (__v8di)(__m512i)(W))
-#define _mm512_maskz_alignr_epi64(U, A, B, imm) __extension__({\
+#define _mm512_maskz_alignr_epi64(U, A, B, imm) \
(__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
(__v8di)_mm512_alignr_epi64((A), (B), (imm)), \
- (__v8di)_mm512_setzero_si512()); })
-
-#define _mm512_alignr_epi32(A, B, I) __extension__ ({ \
- (__m512i)__builtin_shufflevector((__v16si)(__m512i)(B), \
- (__v16si)(__m512i)(A), \
- ((int)(I) & 0xf) + 0, \
- ((int)(I) & 0xf) + 1, \
- ((int)(I) & 0xf) + 2, \
- ((int)(I) & 0xf) + 3, \
- ((int)(I) & 0xf) + 4, \
- ((int)(I) & 0xf) + 5, \
- ((int)(I) & 0xf) + 6, \
- ((int)(I) & 0xf) + 7, \
- ((int)(I) & 0xf) + 8, \
- ((int)(I) & 0xf) + 9, \
- ((int)(I) & 0xf) + 10, \
- ((int)(I) & 0xf) + 11, \
- ((int)(I) & 0xf) + 12, \
- ((int)(I) & 0xf) + 13, \
- ((int)(I) & 0xf) + 14, \
- ((int)(I) & 0xf) + 15); })
-
-#define _mm512_mask_alignr_epi32(W, U, A, B, imm) __extension__ ({\
+ (__v8di)_mm512_setzero_si512())
+
+#define _mm512_alignr_epi32(A, B, I) \
+ (__m512i)__builtin_ia32_alignd512((__v16si)(__m512i)(A), \
+ (__v16si)(__m512i)(B), (int)(I))
+
+#define _mm512_mask_alignr_epi32(W, U, A, B, imm) \
(__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
(__v16si)_mm512_alignr_epi32((A), (B), (imm)), \
- (__v16si)(__m512i)(W)); })
+ (__v16si)(__m512i)(W))
-#define _mm512_maskz_alignr_epi32(U, A, B, imm) __extension__({\
+#define _mm512_maskz_alignr_epi32(U, A, B, imm) \
(__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
(__v16si)_mm512_alignr_epi32((A), (B), (imm)), \
- (__v16si)_mm512_setzero_si512()); })
+ (__v16si)_mm512_setzero_si512())
/* Vector Extract */
-#define _mm512_extractf64x4_pd(A, I) __extension__ ({ \
- (__m256d)__builtin_shufflevector((__v8df)(__m512d)(A), \
- (__v8df)_mm512_undefined_pd(), \
- ((I) & 1) ? 4 : 0, \
- ((I) & 1) ? 5 : 1, \
- ((I) & 1) ? 6 : 2, \
- ((I) & 1) ? 7 : 3); })
-
-#define _mm512_mask_extractf64x4_pd(W, U, A, imm) __extension__ ({\
- (__m256d)__builtin_ia32_selectpd_256((__mmask8)(U), \
- (__v4df)_mm512_extractf64x4_pd((A), (imm)), \
- (__v4df)(W)); })
-
-#define _mm512_maskz_extractf64x4_pd(U, A, imm) __extension__ ({\
- (__m256d)__builtin_ia32_selectpd_256((__mmask8)(U), \
- (__v4df)_mm512_extractf64x4_pd((A), (imm)), \
- (__v4df)_mm256_setzero_pd()); })
-
-#define _mm512_extractf32x4_ps(A, I) __extension__ ({ \
- (__m128)__builtin_shufflevector((__v16sf)(__m512)(A), \
- (__v16sf)_mm512_undefined_ps(), \
- 0 + ((I) & 0x3) * 4, \
- 1 + ((I) & 0x3) * 4, \
- 2 + ((I) & 0x3) * 4, \
- 3 + ((I) & 0x3) * 4); })
-
-#define _mm512_mask_extractf32x4_ps(W, U, A, imm) __extension__ ({\
- (__m128)__builtin_ia32_selectps_128((__mmask8)(U), \
- (__v4sf)_mm512_extractf32x4_ps((A), (imm)), \
- (__v4sf)(W)); })
-
-#define _mm512_maskz_extractf32x4_ps(U, A, imm) __extension__ ({\
- (__m128)__builtin_ia32_selectps_128((__mmask8)(U), \
- (__v4sf)_mm512_extractf32x4_ps((A), (imm)), \
- (__v4sf)_mm_setzero_ps()); })
+#define _mm512_extractf64x4_pd(A, I) \
+ (__m256d)__builtin_ia32_extractf64x4_mask((__v8df)(__m512d)(A), (int)(I), \
+ (__v4df)_mm256_undefined_pd(), \
+ (__mmask8)-1)
+
+#define _mm512_mask_extractf64x4_pd(W, U, A, imm) \
+ (__m256d)__builtin_ia32_extractf64x4_mask((__v8df)(__m512d)(A), (int)(imm), \
+ (__v4df)(__m256d)(W), \
+ (__mmask8)(U))
+
+#define _mm512_maskz_extractf64x4_pd(U, A, imm) \
+ (__m256d)__builtin_ia32_extractf64x4_mask((__v8df)(__m512d)(A), (int)(imm), \
+ (__v4df)_mm256_setzero_pd(), \
+ (__mmask8)(U))
+
+#define _mm512_extractf32x4_ps(A, I) \
+ (__m128)__builtin_ia32_extractf32x4_mask((__v16sf)(__m512)(A), (int)(I), \
+ (__v4sf)_mm_undefined_ps(), \
+ (__mmask8)-1)
+
+#define _mm512_mask_extractf32x4_ps(W, U, A, imm) \
+ (__m128)__builtin_ia32_extractf32x4_mask((__v16sf)(__m512)(A), (int)(imm), \
+ (__v4sf)(__m128)(W), \
+ (__mmask8)(U))
+
+#define _mm512_maskz_extractf32x4_ps(U, A, imm) \
+ (__m128)__builtin_ia32_extractf32x4_mask((__v16sf)(__m512)(A), (int)(imm), \
+ (__v4sf)_mm_setzero_ps(), \
+ (__mmask8)(U))
/* Vector Blend */
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_blend_pd(__mmask8 __U, __m512d __A, __m512d __W)
{
return (__m512d) __builtin_ia32_selectpd_512 ((__mmask8) __U,
(__v8df) __A);
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_blend_ps(__mmask16 __U, __m512 __A, __m512 __W)
{
return (__m512) __builtin_ia32_selectps_512 ((__mmask16) __U,
(__v16sf) __A);
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_blend_epi64(__mmask8 __U, __m512i __A, __m512i __W)
{
return (__m512i) __builtin_ia32_selectq_512 ((__mmask8) __U,
(__v8di) __A);
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_blend_epi32(__mmask16 __U, __m512i __A, __m512i __W)
{
return (__m512i) __builtin_ia32_selectd_512 ((__mmask16) __U,
/* Compare */
-#define _mm512_cmp_round_ps_mask(A, B, P, R) __extension__ ({ \
+#define _mm512_cmp_round_ps_mask(A, B, P, R) \
(__mmask16)__builtin_ia32_cmpps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), (int)(P), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_cmp_round_ps_mask(U, A, B, P, R) __extension__ ({ \
+#define _mm512_mask_cmp_round_ps_mask(U, A, B, P, R) \
(__mmask16)__builtin_ia32_cmpps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), (int)(P), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
#define _mm512_cmp_ps_mask(A, B, P) \
_mm512_cmp_round_ps_mask((A), (B), (P), _MM_FROUND_CUR_DIRECTION)
#define _mm512_mask_cmpord_ps_mask(k, A, B) \
_mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_ORD_Q)
-#define _mm512_cmp_round_pd_mask(A, B, P, R) __extension__ ({ \
+#define _mm512_cmp_round_pd_mask(A, B, P, R) \
(__mmask8)__builtin_ia32_cmppd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), (int)(P), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cmp_round_pd_mask(U, A, B, P, R) __extension__ ({ \
+#define _mm512_mask_cmp_round_pd_mask(U, A, B, P, R) \
(__mmask8)__builtin_ia32_cmppd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), (int)(P), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
#define _mm512_cmp_pd_mask(A, B, P) \
_mm512_cmp_round_pd_mask((A), (B), (P), _MM_FROUND_CUR_DIRECTION)
/* Conversion */
-#define _mm512_cvtt_roundps_epu32(A, R) __extension__ ({ \
+#define _mm512_cvtt_roundps_epu32(A, R) \
(__m512i)__builtin_ia32_cvttps2udq512_mask((__v16sf)(__m512)(A), \
(__v16si)_mm512_undefined_epi32(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_cvtt_roundps_epu32(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvtt_roundps_epu32(W, U, A, R) \
(__m512i)__builtin_ia32_cvttps2udq512_mask((__v16sf)(__m512)(A), \
(__v16si)(__m512i)(W), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_cvtt_roundps_epu32(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvtt_roundps_epu32(U, A, R) \
(__m512i)__builtin_ia32_cvttps2udq512_mask((__v16sf)(__m512)(A), \
(__v16si)_mm512_setzero_si512(), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_cvttps_epu32(__m512 __A)
{
return (__m512i) __builtin_ia32_cvttps2udq512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvttps_epu32 (__m512i __W, __mmask16 __U, __m512 __A)
{
return (__m512i) __builtin_ia32_cvttps2udq512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvttps_epu32 (__mmask16 __U, __m512 __A)
{
return (__m512i) __builtin_ia32_cvttps2udq512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvt_roundepi32_ps(A, R) __extension__ ({ \
+#define _mm512_cvt_roundepi32_ps(A, R) \
(__m512)__builtin_ia32_cvtdq2ps512_mask((__v16si)(__m512i)(A), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_cvt_roundepi32_ps(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundepi32_ps(W, U, A, R) \
(__m512)__builtin_ia32_cvtdq2ps512_mask((__v16si)(__m512i)(A), \
(__v16sf)(__m512)(W), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_cvt_roundepi32_ps(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundepi32_ps(U, A, R) \
(__m512)__builtin_ia32_cvtdq2ps512_mask((__v16si)(__m512i)(A), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_cvt_roundepu32_ps(A, R) __extension__ ({ \
+#define _mm512_cvt_roundepu32_ps(A, R) \
(__m512)__builtin_ia32_cvtudq2ps512_mask((__v16si)(__m512i)(A), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_cvt_roundepu32_ps(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundepu32_ps(W, U, A, R) \
(__m512)__builtin_ia32_cvtudq2ps512_mask((__v16si)(__m512i)(A), \
(__v16sf)(__m512)(W), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_cvt_roundepu32_ps(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundepu32_ps(U, A, R) \
(__m512)__builtin_ia32_cvtudq2ps512_mask((__v16si)(__m512i)(A), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_cvtepu32_ps (__m512i __A)
{
- return (__m512) __builtin_ia32_cvtudq2ps512_mask ((__v16si) __A,
- (__v16sf) _mm512_undefined_ps (),
- (__mmask16) -1,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512)__builtin_convertvector((__v16su)__A, __v16sf);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu32_ps (__m512 __W, __mmask16 __U, __m512i __A)
{
- return (__m512) __builtin_ia32_cvtudq2ps512_mask ((__v16si) __A,
- (__v16sf) __W,
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
+ (__v16sf)_mm512_cvtepu32_ps(__A),
+ (__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu32_ps (__mmask16 __U, __m512i __A)
{
- return (__m512) __builtin_ia32_cvtudq2ps512_mask ((__v16si) __A,
- (__v16sf) _mm512_setzero_ps (),
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
+ (__v16sf)_mm512_cvtepu32_ps(__A),
+ (__v16sf)_mm512_setzero_ps());
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_cvtepi32_pd(__m256i __A)
{
return (__m512d)__builtin_convertvector((__v8si)__A, __v8df);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32_pd (__m512d __W, __mmask8 __U, __m256i __A)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
(__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi32_pd (__mmask8 __U, __m256i __A)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
(__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_cvtepi32lo_pd(__m512i __A)
{
return (__m512d) _mm512_cvtepi32_pd(_mm512_castsi512_si256(__A));
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32lo_pd(__m512d __W, __mmask8 __U,__m512i __A)
{
return (__m512d) _mm512_mask_cvtepi32_pd(__W, __U, _mm512_castsi512_si256(__A));
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_cvtepi32_ps (__m512i __A)
{
- return (__m512) __builtin_ia32_cvtdq2ps512_mask ((__v16si) __A,
- (__v16sf) _mm512_undefined_ps (),
- (__mmask16) -1,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512)__builtin_convertvector((__v16si)__A, __v16sf);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32_ps (__m512 __W, __mmask16 __U, __m512i __A)
{
- return (__m512) __builtin_ia32_cvtdq2ps512_mask ((__v16si) __A,
- (__v16sf) __W,
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
+ (__v16sf)_mm512_cvtepi32_ps(__A),
+ (__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi32_ps (__mmask16 __U, __m512i __A)
{
- return (__m512) __builtin_ia32_cvtdq2ps512_mask ((__v16si) __A,
- (__v16sf) _mm512_setzero_ps (),
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
+ (__v16sf)_mm512_cvtepi32_ps(__A),
+ (__v16sf)_mm512_setzero_ps());
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_cvtepu32_pd(__m256i __A)
{
return (__m512d)__builtin_convertvector((__v8su)__A, __v8df);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu32_pd (__m512d __W, __mmask8 __U, __m256i __A)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
(__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu32_pd (__mmask8 __U, __m256i __A)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
(__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_cvtepu32lo_pd(__m512i __A)
{
return (__m512d) _mm512_cvtepu32_pd(_mm512_castsi512_si256(__A));
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu32lo_pd(__m512d __W, __mmask8 __U,__m512i __A)
{
return (__m512d) _mm512_mask_cvtepu32_pd(__W, __U, _mm512_castsi512_si256(__A));
}
-#define _mm512_cvt_roundpd_ps(A, R) __extension__ ({ \
+#define _mm512_cvt_roundpd_ps(A, R) \
(__m256)__builtin_ia32_cvtpd2ps512_mask((__v8df)(__m512d)(A), \
(__v8sf)_mm256_setzero_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvt_roundpd_ps(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundpd_ps(W, U, A, R) \
(__m256)__builtin_ia32_cvtpd2ps512_mask((__v8df)(__m512d)(A), \
(__v8sf)(__m256)(W), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_maskz_cvt_roundpd_ps(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundpd_ps(U, A, R) \
(__m256)__builtin_ia32_cvtpd2ps512_mask((__v8df)(__m512d)(A), \
(__v8sf)_mm256_setzero_ps(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS512
_mm512_cvtpd_ps (__m512d __A)
{
return (__m256) __builtin_ia32_cvtpd2ps512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS512
_mm512_mask_cvtpd_ps (__m256 __W, __mmask8 __U, __m512d __A)
{
return (__m256) __builtin_ia32_cvtpd2ps512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtpd_ps (__mmask8 __U, __m512d __A)
{
return (__m256) __builtin_ia32_cvtpd2ps512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_cvtpd_pslo (__m512d __A)
{
return (__m512) __builtin_shufflevector((__v8sf) _mm512_cvtpd_ps(__A),
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_cvtpd_pslo (__m512 __W, __mmask8 __U,__m512d __A)
{
return (__m512) __builtin_shufflevector (
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
}
-#define _mm512_cvt_roundps_ph(A, I) __extension__ ({ \
+#define _mm512_cvt_roundps_ph(A, I) \
(__m256i)__builtin_ia32_vcvtps2ph512_mask((__v16sf)(__m512)(A), (int)(I), \
(__v16hi)_mm256_undefined_si256(), \
- (__mmask16)-1); })
+ (__mmask16)-1)
-#define _mm512_mask_cvt_roundps_ph(U, W, A, I) __extension__ ({ \
+#define _mm512_mask_cvt_roundps_ph(U, W, A, I) \
(__m256i)__builtin_ia32_vcvtps2ph512_mask((__v16sf)(__m512)(A), (int)(I), \
(__v16hi)(__m256i)(U), \
- (__mmask16)(W)); })
-
-#define _mm512_maskz_cvt_roundps_ph(W, A, I) __extension__ ({ \
- (__m256i)__builtin_ia32_vcvtps2ph512_mask((__v16sf)(__m512)(A), (int)(I), \
- (__v16hi)_mm256_setzero_si256(), \
- (__mmask16)(W)); })
+ (__mmask16)(W))
-#define _mm512_cvtps_ph(A, I) __extension__ ({ \
+#define _mm512_maskz_cvt_roundps_ph(W, A, I) \
(__m256i)__builtin_ia32_vcvtps2ph512_mask((__v16sf)(__m512)(A), (int)(I), \
(__v16hi)_mm256_setzero_si256(), \
- (__mmask16)-1); })
+ (__mmask16)(W))
-#define _mm512_mask_cvtps_ph(U, W, A, I) __extension__ ({ \
- (__m256i)__builtin_ia32_vcvtps2ph512_mask((__v16sf)(__m512)(A), (int)(I), \
- (__v16hi)(__m256i)(U), \
- (__mmask16)(W)); })
-
-#define _mm512_maskz_cvtps_ph(W, A, I) __extension__ ({\
- (__m256i)__builtin_ia32_vcvtps2ph512_mask((__v16sf)(__m512)(A), (int)(I), \
- (__v16hi)_mm256_setzero_si256(), \
- (__mmask16)(W)); })
+#define _mm512_cvtps_ph _mm512_cvt_roundps_ph
+#define _mm512_mask_cvtps_ph _mm512_mask_cvt_roundps_ph
+#define _mm512_maskz_cvtps_ph _mm512_maskz_cvt_roundps_ph
-#define _mm512_cvt_roundph_ps(A, R) __extension__ ({ \
+#define _mm512_cvt_roundph_ps(A, R) \
(__m512)__builtin_ia32_vcvtph2ps512_mask((__v16hi)(__m256i)(A), \
(__v16sf)_mm512_undefined_ps(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_cvt_roundph_ps(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundph_ps(W, U, A, R) \
(__m512)__builtin_ia32_vcvtph2ps512_mask((__v16hi)(__m256i)(A), \
(__v16sf)(__m512)(W), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_cvt_roundph_ps(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundph_ps(U, A, R) \
(__m512)__builtin_ia32_vcvtph2ps512_mask((__v16hi)(__m256i)(A), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_cvtph_ps(__m256i __A)
{
return (__m512) __builtin_ia32_vcvtph2ps512_mask ((__v16hi) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_cvtph_ps (__m512 __W, __mmask16 __U, __m256i __A)
{
return (__m512) __builtin_ia32_vcvtph2ps512_mask ((__v16hi) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtph_ps (__mmask16 __U, __m256i __A)
{
return (__m512) __builtin_ia32_vcvtph2ps512_mask ((__v16hi) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvtt_roundpd_epi32(A, R) __extension__ ({ \
+#define _mm512_cvtt_roundpd_epi32(A, R) \
(__m256i)__builtin_ia32_cvttpd2dq512_mask((__v8df)(__m512d)(A), \
(__v8si)_mm256_setzero_si256(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvtt_roundpd_epi32(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvtt_roundpd_epi32(W, U, A, R) \
(__m256i)__builtin_ia32_cvttpd2dq512_mask((__v8df)(__m512d)(A), \
(__v8si)(__m256i)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_cvtt_roundpd_epi32(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvtt_roundpd_epi32(U, A, R) \
(__m256i)__builtin_ia32_cvttpd2dq512_mask((__v8df)(__m512d)(A), \
(__v8si)_mm256_setzero_si256(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline __m256i __DEFAULT_FN_ATTRS
+static __inline __m256i __DEFAULT_FN_ATTRS512
_mm512_cvttpd_epi32(__m512d __a)
{
return (__m256i)__builtin_ia32_cvttpd2dq512_mask((__v8df) __a,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvttpd_epi32 (__m256i __W, __mmask8 __U, __m512d __A)
{
return (__m256i) __builtin_ia32_cvttpd2dq512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvttpd_epi32 (__mmask8 __U, __m512d __A)
{
return (__m256i) __builtin_ia32_cvttpd2dq512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvtt_roundps_epi32(A, R) __extension__ ({ \
+#define _mm512_cvtt_roundps_epi32(A, R) \
(__m512i)__builtin_ia32_cvttps2dq512_mask((__v16sf)(__m512)(A), \
(__v16si)_mm512_setzero_si512(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_cvtt_roundps_epi32(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvtt_roundps_epi32(W, U, A, R) \
(__m512i)__builtin_ia32_cvttps2dq512_mask((__v16sf)(__m512)(A), \
(__v16si)(__m512i)(W), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_cvtt_roundps_epi32(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvtt_roundps_epi32(U, A, R) \
(__m512i)__builtin_ia32_cvttps2dq512_mask((__v16sf)(__m512)(A), \
(__v16si)_mm512_setzero_si512(), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_cvttps_epi32(__m512 __a)
{
return (__m512i)
(__mmask16) -1, _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvttps_epi32 (__m512i __W, __mmask16 __U, __m512 __A)
{
return (__m512i) __builtin_ia32_cvttps2dq512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvttps_epi32 (__mmask16 __U, __m512 __A)
{
return (__m512i) __builtin_ia32_cvttps2dq512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvt_roundps_epi32(A, R) __extension__ ({ \
+#define _mm512_cvt_roundps_epi32(A, R) \
(__m512i)__builtin_ia32_cvtps2dq512_mask((__v16sf)(__m512)(A), \
(__v16si)_mm512_setzero_si512(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_cvt_roundps_epi32(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundps_epi32(W, U, A, R) \
(__m512i)__builtin_ia32_cvtps2dq512_mask((__v16sf)(__m512)(A), \
(__v16si)(__m512i)(W), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_cvt_roundps_epi32(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundps_epi32(U, A, R) \
(__m512i)__builtin_ia32_cvtps2dq512_mask((__v16sf)(__m512)(A), \
(__v16si)_mm512_setzero_si512(), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtps_epi32 (__m512 __A)
{
return (__m512i) __builtin_ia32_cvtps2dq512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtps_epi32 (__m512i __W, __mmask16 __U, __m512 __A)
{
return (__m512i) __builtin_ia32_cvtps2dq512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtps_epi32 (__mmask16 __U, __m512 __A)
{
return (__m512i) __builtin_ia32_cvtps2dq512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvt_roundpd_epi32(A, R) __extension__ ({ \
+#define _mm512_cvt_roundpd_epi32(A, R) \
(__m256i)__builtin_ia32_cvtpd2dq512_mask((__v8df)(__m512d)(A), \
(__v8si)_mm256_setzero_si256(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvt_roundpd_epi32(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundpd_epi32(W, U, A, R) \
(__m256i)__builtin_ia32_cvtpd2dq512_mask((__v8df)(__m512d)(A), \
(__v8si)(__m256i)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_cvt_roundpd_epi32(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundpd_epi32(U, A, R) \
(__m256i)__builtin_ia32_cvtpd2dq512_mask((__v8df)(__m512d)(A), \
(__v8si)_mm256_setzero_si256(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtpd_epi32 (__m512d __A)
{
return (__m256i) __builtin_ia32_cvtpd2dq512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtpd_epi32 (__m256i __W, __mmask8 __U, __m512d __A)
{
return (__m256i) __builtin_ia32_cvtpd2dq512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtpd_epi32 (__mmask8 __U, __m512d __A)
{
return (__m256i) __builtin_ia32_cvtpd2dq512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvt_roundps_epu32(A, R) __extension__ ({ \
+#define _mm512_cvt_roundps_epu32(A, R) \
(__m512i)__builtin_ia32_cvtps2udq512_mask((__v16sf)(__m512)(A), \
(__v16si)_mm512_setzero_si512(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_cvt_roundps_epu32(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundps_epu32(W, U, A, R) \
(__m512i)__builtin_ia32_cvtps2udq512_mask((__v16sf)(__m512)(A), \
(__v16si)(__m512i)(W), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_cvt_roundps_epu32(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundps_epu32(U, A, R) \
(__m512i)__builtin_ia32_cvtps2udq512_mask((__v16sf)(__m512)(A), \
(__v16si)_mm512_setzero_si512(), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtps_epu32 ( __m512 __A)
{
return (__m512i) __builtin_ia32_cvtps2udq512_mask ((__v16sf) __A,\
(__v16si)\
- _mm512_undefined_epi32 (),\
+ _mm512_undefined_epi32 (),
(__mmask16) -1,\
- _MM_FROUND_CUR_DIRECTION);\
+ _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtps_epu32 (__m512i __W, __mmask16 __U, __m512 __A)
{
return (__m512i) __builtin_ia32_cvtps2udq512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtps_epu32 ( __mmask16 __U, __m512 __A)
{
return (__m512i) __builtin_ia32_cvtps2udq512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_cvt_roundpd_epu32(A, R) __extension__ ({ \
+#define _mm512_cvt_roundpd_epu32(A, R) \
(__m256i)__builtin_ia32_cvtpd2udq512_mask((__v8df)(__m512d)(A), \
(__v8si)_mm256_setzero_si256(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvt_roundpd_epu32(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundpd_epu32(W, U, A, R) \
(__m256i)__builtin_ia32_cvtpd2udq512_mask((__v8df)(__m512d)(A), \
- (__v8si)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__v8si)(__m256i)(W), \
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_cvt_roundpd_epu32(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundpd_epu32(U, A, R) \
(__m256i)__builtin_ia32_cvtpd2udq512_mask((__v8df)(__m512d)(A), \
(__v8si)_mm256_setzero_si256(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtpd_epu32 (__m512d __A)
{
return (__m256i) __builtin_ia32_cvtpd2udq512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtpd_epu32 (__m256i __W, __mmask8 __U, __m512d __A)
{
return (__m256i) __builtin_ia32_cvtpd2udq512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtpd_epu32 (__mmask8 __U, __m512d __A)
{
return (__m256i) __builtin_ia32_cvtpd2udq512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ double __DEFAULT_FN_ATTRS
+static __inline__ double __DEFAULT_FN_ATTRS512
_mm512_cvtsd_f64(__m512d __a)
{
return __a[0];
}
-static __inline__ float __DEFAULT_FN_ATTRS
+static __inline__ float __DEFAULT_FN_ATTRS512
_mm512_cvtss_f32(__m512 __a)
{
return __a[0];
/* Unpack and Interleave */
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_unpackhi_pd(__m512d __a, __m512d __b)
{
return (__m512d)__builtin_shufflevector((__v8df)__a, (__v8df)__b,
1, 9, 1+2, 9+2, 1+4, 9+4, 1+6, 9+6);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_unpackhi_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
(__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_unpackhi_pd(__mmask8 __U, __m512d __A, __m512d __B)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
(__v8df)_mm512_setzero_pd());
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_unpacklo_pd(__m512d __a, __m512d __b)
{
return (__m512d)__builtin_shufflevector((__v8df)__a, (__v8df)__b,
0, 8, 0+2, 8+2, 0+4, 8+4, 0+6, 8+6);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_unpacklo_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
(__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_unpacklo_pd (__mmask8 __U, __m512d __A, __m512d __B)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
(__v8df)_mm512_setzero_pd());
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_unpackhi_ps(__m512 __a, __m512 __b)
{
return (__m512)__builtin_shufflevector((__v16sf)__a, (__v16sf)__b,
2+12, 18+12, 3+12, 19+12);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_unpackhi_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
{
return (__m512)__builtin_ia32_selectps_512((__mmask16) __U,
(__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_unpackhi_ps (__mmask16 __U, __m512 __A, __m512 __B)
{
return (__m512)__builtin_ia32_selectps_512((__mmask16) __U,
(__v16sf)_mm512_setzero_ps());
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_unpacklo_ps(__m512 __a, __m512 __b)
{
return (__m512)__builtin_shufflevector((__v16sf)__a, (__v16sf)__b,
0+12, 16+12, 1+12, 17+12);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_unpacklo_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
{
return (__m512)__builtin_ia32_selectps_512((__mmask16) __U,
(__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_unpacklo_ps (__mmask16 __U, __m512 __A, __m512 __B)
{
return (__m512)__builtin_ia32_selectps_512((__mmask16) __U,
(__v16sf)_mm512_setzero_ps());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_unpackhi_epi32(__m512i __A, __m512i __B)
{
return (__m512i)__builtin_shufflevector((__v16si)__A, (__v16si)__B,
2+12, 18+12, 3+12, 19+12);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_unpackhi_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16) __U,
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_unpackhi_epi32(__mmask16 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16) __U,
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_unpacklo_epi32(__m512i __A, __m512i __B)
{
return (__m512i)__builtin_shufflevector((__v16si)__A, (__v16si)__B,
0+12, 16+12, 1+12, 17+12);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_unpacklo_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16) __U,
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_unpacklo_epi32(__mmask16 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16) __U,
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_unpackhi_epi64(__m512i __A, __m512i __B)
{
return (__m512i)__builtin_shufflevector((__v8di)__A, (__v8di)__B,
1, 9, 1+2, 9+2, 1+4, 9+4, 1+6, 9+6);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_unpackhi_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8) __U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_unpackhi_epi64(__mmask8 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8) __U,
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_unpacklo_epi64 (__m512i __A, __m512i __B)
{
return (__m512i)__builtin_shufflevector((__v8di)__A, (__v8di)__B,
0, 8, 0+2, 8+2, 0+4, 8+4, 0+6, 8+6);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_unpacklo_epi64 (__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8) __U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_unpacklo_epi64 (__mmask8 __U, __m512i __A, __m512i __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8) __U,
(__v8di)_mm512_setzero_si512());
}
-/* Bit Test */
-
-static __inline __mmask16 __DEFAULT_FN_ATTRS
-_mm512_test_epi32_mask(__m512i __A, __m512i __B)
-{
- return (__mmask16) __builtin_ia32_ptestmd512 ((__v16si) __A,
- (__v16si) __B,
- (__mmask16) -1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_mask_test_epi32_mask (__mmask16 __U, __m512i __A, __m512i __B)
-{
- return (__mmask16) __builtin_ia32_ptestmd512 ((__v16si) __A,
- (__v16si) __B, __U);
-}
-static __inline __mmask8 __DEFAULT_FN_ATTRS
-_mm512_test_epi64_mask(__m512i __A, __m512i __B)
-{
- return (__mmask8) __builtin_ia32_ptestmq512 ((__v8di) __A,
- (__v8di) __B,
- (__mmask8) -1);
-}
+/* SIMD load ops */
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_mask_test_epi64_mask (__mmask8 __U, __m512i __A, __m512i __B)
+static __inline __m512i __DEFAULT_FN_ATTRS512
+_mm512_loadu_si512 (void const *__P)
{
- return (__mmask8) __builtin_ia32_ptestmq512 ((__v8di) __A, (__v8di) __B, __U);
+ struct __loadu_si512 {
+ __m512i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ return ((const struct __loadu_si512*)__P)->__v;
}
-
-/* SIMD load ops */
-
-static __inline __m512i __DEFAULT_FN_ATTRS
-_mm512_loadu_si512 (void const *__P)
+static __inline __m512i __DEFAULT_FN_ATTRS512
+_mm512_loadu_epi32 (void const *__P)
{
- return (__m512i) __builtin_ia32_loaddqusi512_mask ((const int *) __P,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) -1);
+ struct __loadu_epi32 {
+ __m512i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ return ((const struct __loadu_epi32*)__P)->__v;
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_loadu_epi32 (__m512i __W, __mmask16 __U, void const *__P)
{
return (__m512i) __builtin_ia32_loaddqusi512_mask ((const int *) __P,
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_loadu_epi32(__mmask16 __U, void const *__P)
{
return (__m512i) __builtin_ia32_loaddqusi512_mask ((const int *)__P,
(__mmask16) __U);
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
+_mm512_loadu_epi64 (void const *__P)
+{
+ struct __loadu_epi64 {
+ __m512i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ return ((const struct __loadu_epi64*)__P)->__v;
+}
+
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_loadu_epi64 (__m512i __W, __mmask8 __U, void const *__P)
{
return (__m512i) __builtin_ia32_loaddqudi512_mask ((const long long *) __P,
(__mmask8) __U);
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_loadu_epi64(__mmask8 __U, void const *__P)
{
return (__m512i) __builtin_ia32_loaddqudi512_mask ((const long long *)__P,
(__mmask8) __U);
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_loadu_ps (__m512 __W, __mmask16 __U, void const *__P)
{
return (__m512) __builtin_ia32_loadups512_mask ((const float *) __P,
(__mmask16) __U);
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_loadu_ps(__mmask16 __U, void const *__P)
{
return (__m512) __builtin_ia32_loadups512_mask ((const float *)__P,
(__mmask16) __U);
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_loadu_pd (__m512d __W, __mmask8 __U, void const *__P)
{
return (__m512d) __builtin_ia32_loadupd512_mask ((const double *) __P,
(__mmask8) __U);
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_loadu_pd(__mmask8 __U, void const *__P)
{
return (__m512d) __builtin_ia32_loadupd512_mask ((const double *)__P,
(__mmask8) __U);
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_loadu_pd(void const *__p)
{
struct __loadu_pd {
- __m512d __v;
+ __m512d_u __v;
} __attribute__((__packed__, __may_alias__));
- return ((struct __loadu_pd*)__p)->__v;
+ return ((const struct __loadu_pd*)__p)->__v;
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_loadu_ps(void const *__p)
{
struct __loadu_ps {
- __m512 __v;
+ __m512_u __v;
} __attribute__((__packed__, __may_alias__));
- return ((struct __loadu_ps*)__p)->__v;
+ return ((const struct __loadu_ps*)__p)->__v;
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_load_ps(void const *__p)
{
- return (__m512) __builtin_ia32_loadaps512_mask ((const __v16sf *)__p,
- (__v16sf)
- _mm512_setzero_ps (),
- (__mmask16) -1);
+ return *(const __m512*)__p;
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_load_ps (__m512 __W, __mmask16 __U, void const *__P)
{
return (__m512) __builtin_ia32_loadaps512_mask ((const __v16sf *) __P,
(__mmask16) __U);
}
-static __inline __m512 __DEFAULT_FN_ATTRS
+static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_load_ps(__mmask16 __U, void const *__P)
{
return (__m512) __builtin_ia32_loadaps512_mask ((const __v16sf *)__P,
(__mmask16) __U);
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_load_pd(void const *__p)
{
- return (__m512d) __builtin_ia32_loadapd512_mask ((const __v8df *)__p,
- (__v8df)
- _mm512_setzero_pd (),
- (__mmask8) -1);
+ return *(const __m512d*)__p;
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_load_pd (__m512d __W, __mmask8 __U, void const *__P)
{
return (__m512d) __builtin_ia32_loadapd512_mask ((const __v8df *) __P,
(__mmask8) __U);
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_load_pd(__mmask8 __U, void const *__P)
{
return (__m512d) __builtin_ia32_loadapd512_mask ((const __v8df *)__P,
(__mmask8) __U);
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_load_si512 (void const *__P)
{
- return *(__m512i *) __P;
+ return *(const __m512i *) __P;
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_load_epi32 (void const *__P)
{
- return *(__m512i *) __P;
+ return *(const __m512i *) __P;
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_load_epi64 (void const *__P)
{
- return *(__m512i *) __P;
+ return *(const __m512i *) __P;
}
/* SIMD store ops */
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS512
+_mm512_storeu_epi64 (void *__P, __m512i __A)
+{
+ struct __storeu_epi64 {
+ __m512i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_epi64*)__P)->__v = __A;
+}
+
+static __inline void __DEFAULT_FN_ATTRS512
_mm512_mask_storeu_epi64(void *__P, __mmask8 __U, __m512i __A)
{
__builtin_ia32_storedqudi512_mask ((long long *)__P, (__v8di) __A,
(__mmask8) __U);
}
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS512
_mm512_storeu_si512 (void *__P, __m512i __A)
{
- __builtin_ia32_storedqusi512_mask ((int *) __P, (__v16si) __A,
- (__mmask16) -1);
+ struct __storeu_si512 {
+ __m512i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_si512*)__P)->__v = __A;
+}
+
+static __inline void __DEFAULT_FN_ATTRS512
+_mm512_storeu_epi32 (void *__P, __m512i __A)
+{
+ struct __storeu_epi32 {
+ __m512i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_epi32*)__P)->__v = __A;
}
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS512
_mm512_mask_storeu_epi32(void *__P, __mmask16 __U, __m512i __A)
{
__builtin_ia32_storedqusi512_mask ((int *)__P, (__v16si) __A,
(__mmask16) __U);
}
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS512
_mm512_mask_storeu_pd(void *__P, __mmask8 __U, __m512d __A)
{
__builtin_ia32_storeupd512_mask ((double *)__P, (__v8df) __A, (__mmask8) __U);
}
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS512
_mm512_storeu_pd(void *__P, __m512d __A)
{
- __builtin_ia32_storeupd512_mask((double *)__P, (__v8df)__A, (__mmask8)-1);
+ struct __storeu_pd {
+ __m512d_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_pd*)__P)->__v = __A;
}
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS512
_mm512_mask_storeu_ps(void *__P, __mmask16 __U, __m512 __A)
{
__builtin_ia32_storeups512_mask ((float *)__P, (__v16sf) __A,
(__mmask16) __U);
}
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS512
_mm512_storeu_ps(void *__P, __m512 __A)
{
- __builtin_ia32_storeups512_mask((float *)__P, (__v16sf)__A, (__mmask16)-1);
+ struct __storeu_ps {
+ __m512_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_ps*)__P)->__v = __A;
}
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS512
_mm512_mask_store_pd(void *__P, __mmask8 __U, __m512d __A)
{
__builtin_ia32_storeapd512_mask ((__v8df *)__P, (__v8df) __A, (__mmask8) __U);
}
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS512
_mm512_store_pd(void *__P, __m512d __A)
{
*(__m512d*)__P = __A;
}
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS512
_mm512_mask_store_ps(void *__P, __mmask16 __U, __m512 __A)
{
__builtin_ia32_storeaps512_mask ((__v16sf *)__P, (__v16sf) __A,
(__mmask16) __U);
}
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS512
_mm512_store_ps(void *__P, __m512 __A)
{
*(__m512*)__P = __A;
}
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS512
_mm512_store_si512 (void *__P, __m512i __A)
{
*(__m512i *) __P = __A;
}
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS512
_mm512_store_epi32 (void *__P, __m512i __A)
{
*(__m512i *) __P = __A;
}
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS512
_mm512_store_epi64 (void *__P, __m512i __A)
{
*(__m512i *) __P = __A;
/* Integer compare */
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_cmpeq_epi32_mask(__m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_pcmpeqd512_mask((__v16si)__a, (__v16si)__b,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpeq_epi32_mask(__mmask16 __u, __m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_pcmpeqd512_mask((__v16si)__a, (__v16si)__b,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_cmpeq_epu32_mask(__m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)__a, (__v16si)__b, 0,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpeq_epu32_mask(__mmask16 __u, __m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)__a, (__v16si)__b, 0,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpeq_epi64_mask(__mmask8 __u, __m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_pcmpeqq512_mask((__v8di)__a, (__v8di)__b,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_cmpeq_epi64_mask(__m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_pcmpeqq512_mask((__v8di)__a, (__v8di)__b,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_cmpeq_epu64_mask(__m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)__a, (__v8di)__b, 0,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpeq_epu64_mask(__mmask8 __u, __m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)__a, (__v8di)__b, 0,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_cmpge_epi32_mask(__m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_cmpd512_mask((__v16si)__a, (__v16si)__b, 5,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpge_epi32_mask(__mmask16 __u, __m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_cmpd512_mask((__v16si)__a, (__v16si)__b, 5,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_cmpge_epu32_mask(__m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)__a, (__v16si)__b, 5,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpge_epu32_mask(__mmask16 __u, __m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)__a, (__v16si)__b, 5,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_cmpge_epi64_mask(__m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_cmpq512_mask((__v8di)__a, (__v8di)__b, 5,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpge_epi64_mask(__mmask8 __u, __m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_cmpq512_mask((__v8di)__a, (__v8di)__b, 5,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_cmpge_epu64_mask(__m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)__a, (__v8di)__b, 5,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpge_epu64_mask(__mmask8 __u, __m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)__a, (__v8di)__b, 5,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_cmpgt_epi32_mask(__m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_pcmpgtd512_mask((__v16si)__a, (__v16si)__b,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpgt_epi32_mask(__mmask16 __u, __m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_pcmpgtd512_mask((__v16si)__a, (__v16si)__b,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_cmpgt_epu32_mask(__m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)__a, (__v16si)__b, 6,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpgt_epu32_mask(__mmask16 __u, __m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)__a, (__v16si)__b, 6,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpgt_epi64_mask(__mmask8 __u, __m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_pcmpgtq512_mask((__v8di)__a, (__v8di)__b,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_cmpgt_epi64_mask(__m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_pcmpgtq512_mask((__v8di)__a, (__v8di)__b,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_cmpgt_epu64_mask(__m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)__a, (__v8di)__b, 6,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpgt_epu64_mask(__mmask8 __u, __m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)__a, (__v8di)__b, 6,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_cmple_epi32_mask(__m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_cmpd512_mask((__v16si)__a, (__v16si)__b, 2,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_mask_cmple_epi32_mask(__mmask16 __u, __m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_cmpd512_mask((__v16si)__a, (__v16si)__b, 2,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_cmple_epu32_mask(__m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)__a, (__v16si)__b, 2,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_mask_cmple_epu32_mask(__mmask16 __u, __m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)__a, (__v16si)__b, 2,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_cmple_epi64_mask(__m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_cmpq512_mask((__v8di)__a, (__v8di)__b, 2,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_mask_cmple_epi64_mask(__mmask8 __u, __m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_cmpq512_mask((__v8di)__a, (__v8di)__b, 2,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_cmple_epu64_mask(__m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)__a, (__v8di)__b, 2,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_mask_cmple_epu64_mask(__mmask8 __u, __m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)__a, (__v8di)__b, 2,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_cmplt_epi32_mask(__m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_cmpd512_mask((__v16si)__a, (__v16si)__b, 1,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_mask_cmplt_epi32_mask(__mmask16 __u, __m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_cmpd512_mask((__v16si)__a, (__v16si)__b, 1,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_cmplt_epu32_mask(__m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)__a, (__v16si)__b, 1,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_mask_cmplt_epu32_mask(__mmask16 __u, __m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)__a, (__v16si)__b, 1,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_cmplt_epi64_mask(__m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_cmpq512_mask((__v8di)__a, (__v8di)__b, 1,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_mask_cmplt_epi64_mask(__mmask8 __u, __m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_cmpq512_mask((__v8di)__a, (__v8di)__b, 1,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_cmplt_epu64_mask(__m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)__a, (__v8di)__b, 1,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_mask_cmplt_epu64_mask(__mmask8 __u, __m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)__a, (__v8di)__b, 1,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_cmpneq_epi32_mask(__m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_cmpd512_mask((__v16si)__a, (__v16si)__b, 4,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpneq_epi32_mask(__mmask16 __u, __m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_cmpd512_mask((__v16si)__a, (__v16si)__b, 4,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_cmpneq_epu32_mask(__m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)__a, (__v16si)__b, 4,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpneq_epu32_mask(__mmask16 __u, __m512i __a, __m512i __b) {
- return (__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)__a, (__v16si)__b, 4,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_cmpneq_epi64_mask(__m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_cmpq512_mask((__v8di)__a, (__v8di)__b, 4,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpneq_epi64_mask(__mmask8 __u, __m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_cmpq512_mask((__v8di)__a, (__v8di)__b, 4,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_cmpneq_epu64_mask(__m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)__a, (__v8di)__b, 4,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpneq_epu64_mask(__mmask8 __u, __m512i __a, __m512i __b) {
- return (__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)__a, (__v8di)__b, 4,
- __u);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+#define _mm512_cmpeq_epi32_mask(A, B) \
+ _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm512_mask_cmpeq_epi32_mask(k, A, B) \
+ _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm512_cmpge_epi32_mask(A, B) \
+ _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_GE)
+#define _mm512_mask_cmpge_epi32_mask(k, A, B) \
+ _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm512_cmpgt_epi32_mask(A, B) \
+ _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_GT)
+#define _mm512_mask_cmpgt_epi32_mask(k, A, B) \
+ _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm512_cmple_epi32_mask(A, B) \
+ _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_LE)
+#define _mm512_mask_cmple_epi32_mask(k, A, B) \
+ _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm512_cmplt_epi32_mask(A, B) \
+ _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_LT)
+#define _mm512_mask_cmplt_epi32_mask(k, A, B) \
+ _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm512_cmpneq_epi32_mask(A, B) \
+ _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_NE)
+#define _mm512_mask_cmpneq_epi32_mask(k, A, B) \
+ _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm512_cmpeq_epu32_mask(A, B) \
+ _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm512_mask_cmpeq_epu32_mask(k, A, B) \
+ _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm512_cmpge_epu32_mask(A, B) \
+ _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_GE)
+#define _mm512_mask_cmpge_epu32_mask(k, A, B) \
+ _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm512_cmpgt_epu32_mask(A, B) \
+ _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_GT)
+#define _mm512_mask_cmpgt_epu32_mask(k, A, B) \
+ _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm512_cmple_epu32_mask(A, B) \
+ _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_LE)
+#define _mm512_mask_cmple_epu32_mask(k, A, B) \
+ _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm512_cmplt_epu32_mask(A, B) \
+ _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_LT)
+#define _mm512_mask_cmplt_epu32_mask(k, A, B) \
+ _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm512_cmpneq_epu32_mask(A, B) \
+ _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_NE)
+#define _mm512_mask_cmpneq_epu32_mask(k, A, B) \
+ _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm512_cmpeq_epi64_mask(A, B) \
+ _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm512_mask_cmpeq_epi64_mask(k, A, B) \
+ _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm512_cmpge_epi64_mask(A, B) \
+ _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_GE)
+#define _mm512_mask_cmpge_epi64_mask(k, A, B) \
+ _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm512_cmpgt_epi64_mask(A, B) \
+ _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_GT)
+#define _mm512_mask_cmpgt_epi64_mask(k, A, B) \
+ _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm512_cmple_epi64_mask(A, B) \
+ _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_LE)
+#define _mm512_mask_cmple_epi64_mask(k, A, B) \
+ _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm512_cmplt_epi64_mask(A, B) \
+ _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_LT)
+#define _mm512_mask_cmplt_epi64_mask(k, A, B) \
+ _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm512_cmpneq_epi64_mask(A, B) \
+ _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_NE)
+#define _mm512_mask_cmpneq_epi64_mask(k, A, B) \
+ _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm512_cmpeq_epu64_mask(A, B) \
+ _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm512_mask_cmpeq_epu64_mask(k, A, B) \
+ _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm512_cmpge_epu64_mask(A, B) \
+ _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_GE)
+#define _mm512_mask_cmpge_epu64_mask(k, A, B) \
+ _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm512_cmpgt_epu64_mask(A, B) \
+ _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_GT)
+#define _mm512_mask_cmpgt_epu64_mask(k, A, B) \
+ _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm512_cmple_epu64_mask(A, B) \
+ _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_LE)
+#define _mm512_mask_cmple_epu64_mask(k, A, B) \
+ _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm512_cmplt_epu64_mask(A, B) \
+ _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_LT)
+#define _mm512_mask_cmplt_epu64_mask(k, A, B) \
+ _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm512_cmpneq_epu64_mask(A, B) \
+ _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_NE)
+#define _mm512_mask_cmpneq_epu64_mask(k, A, B) \
+ _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_NE)
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepi8_epi32(__m128i __A)
{
/* This function always performs a signed extension, but __v16qi is a char
return (__m512i)__builtin_convertvector((__v16qs)__A, __v16si);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi8_epi32(__m512i __W, __mmask16 __U, __m128i __A)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi8_epi32(__mmask16 __U, __m128i __A)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepi8_epi64(__m128i __A)
{
/* This function always performs a signed extension, but __v16qi is a char
return (__m512i)__builtin_convertvector(__builtin_shufflevector((__v16qs)__A, (__v16qs)__A, 0, 1, 2, 3, 4, 5, 6, 7), __v8di);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi8_epi64(__m512i __W, __mmask8 __U, __m128i __A)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi8_epi64(__mmask8 __U, __m128i __A)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_setzero_si512 ());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepi32_epi64(__m256i __X)
{
return (__m512i)__builtin_convertvector((__v8si)__X, __v8di);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32_epi64(__m512i __W, __mmask8 __U, __m256i __X)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi32_epi64(__mmask8 __U, __m256i __X)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepi16_epi32(__m256i __A)
{
return (__m512i)__builtin_convertvector((__v16hi)__A, __v16si);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi16_epi32(__m512i __W, __mmask16 __U, __m256i __A)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi16_epi32(__mmask16 __U, __m256i __A)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_setzero_si512 ());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepi16_epi64(__m128i __A)
{
return (__m512i)__builtin_convertvector((__v8hi)__A, __v8di);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi16_epi64(__m512i __W, __mmask8 __U, __m128i __A)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi16_epi64(__mmask8 __U, __m128i __A)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepu8_epi32(__m128i __A)
{
return (__m512i)__builtin_convertvector((__v16qu)__A, __v16si);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu8_epi32(__m512i __W, __mmask16 __U, __m128i __A)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu8_epi32(__mmask16 __U, __m128i __A)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepu8_epi64(__m128i __A)
{
return (__m512i)__builtin_convertvector(__builtin_shufflevector((__v16qu)__A, (__v16qu)__A, 0, 1, 2, 3, 4, 5, 6, 7), __v8di);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu8_epi64(__m512i __W, __mmask8 __U, __m128i __A)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu8_epi64(__mmask8 __U, __m128i __A)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepu32_epi64(__m256i __X)
{
return (__m512i)__builtin_convertvector((__v8su)__X, __v8di);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu32_epi64(__m512i __W, __mmask8 __U, __m256i __X)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu32_epi64(__mmask8 __U, __m256i __X)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepu16_epi32(__m256i __A)
{
return (__m512i)__builtin_convertvector((__v16hu)__A, __v16si);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu16_epi32(__m512i __W, __mmask16 __U, __m256i __A)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu16_epi32(__mmask16 __U, __m256i __A)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepu16_epi64(__m128i __A)
{
return (__m512i)__builtin_convertvector((__v8hu)__A, __v8di);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu16_epi64(__m512i __W, __mmask8 __U, __m128i __A)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu16_epi64(__mmask8 __U, __m128i __A)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_rorv_epi32 (__m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_prorvd512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) -1);
+ return (__m512i)__builtin_ia32_prorvd512((__v16si)__A, (__v16si)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_rorv_epi32 (__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_prorvd512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si) __W,
- (__mmask16) __U);
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_rorv_epi32(__A, __B),
+ (__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_rorv_epi32 (__mmask16 __U, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_prorvd512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) __U);
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_rorv_epi32(__A, __B),
+ (__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_rorv_epi64 (__m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_prorvq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) -1);
+ return (__m512i)__builtin_ia32_prorvq512((__v8di)__A, (__v8di)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_rorv_epi64 (__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_prorvq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di) __W,
- (__mmask8) __U);
+ return (__m512i)__builtin_ia32_selectq_512(__U,
+ (__v8di)_mm512_rorv_epi64(__A, __B),
+ (__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_rorv_epi64 (__mmask8 __U, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_prorvq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) __U);
+ return (__m512i)__builtin_ia32_selectq_512(__U,
+ (__v8di)_mm512_rorv_epi64(__A, __B),
+ (__v8di)_mm512_setzero_si512());
}
-#define _mm512_cmp_epi32_mask(a, b, p) __extension__ ({ \
+#define _mm512_cmp_epi32_mask(a, b, p) \
(__mmask16)__builtin_ia32_cmpd512_mask((__v16si)(__m512i)(a), \
(__v16si)(__m512i)(b), (int)(p), \
- (__mmask16)-1); })
+ (__mmask16)-1)
-#define _mm512_cmp_epu32_mask(a, b, p) __extension__ ({ \
+#define _mm512_cmp_epu32_mask(a, b, p) \
(__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)(__m512i)(a), \
(__v16si)(__m512i)(b), (int)(p), \
- (__mmask16)-1); })
+ (__mmask16)-1)
-#define _mm512_cmp_epi64_mask(a, b, p) __extension__ ({ \
+#define _mm512_cmp_epi64_mask(a, b, p) \
(__mmask8)__builtin_ia32_cmpq512_mask((__v8di)(__m512i)(a), \
(__v8di)(__m512i)(b), (int)(p), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm512_cmp_epu64_mask(a, b, p) __extension__ ({ \
+#define _mm512_cmp_epu64_mask(a, b, p) \
(__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)(__m512i)(a), \
(__v8di)(__m512i)(b), (int)(p), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm512_mask_cmp_epi32_mask(m, a, b, p) __extension__ ({ \
+#define _mm512_mask_cmp_epi32_mask(m, a, b, p) \
(__mmask16)__builtin_ia32_cmpd512_mask((__v16si)(__m512i)(a), \
(__v16si)(__m512i)(b), (int)(p), \
- (__mmask16)(m)); })
+ (__mmask16)(m))
-#define _mm512_mask_cmp_epu32_mask(m, a, b, p) __extension__ ({ \
+#define _mm512_mask_cmp_epu32_mask(m, a, b, p) \
(__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)(__m512i)(a), \
(__v16si)(__m512i)(b), (int)(p), \
- (__mmask16)(m)); })
+ (__mmask16)(m))
-#define _mm512_mask_cmp_epi64_mask(m, a, b, p) __extension__ ({ \
+#define _mm512_mask_cmp_epi64_mask(m, a, b, p) \
(__mmask8)__builtin_ia32_cmpq512_mask((__v8di)(__m512i)(a), \
(__v8di)(__m512i)(b), (int)(p), \
- (__mmask8)(m)); })
+ (__mmask8)(m))
-#define _mm512_mask_cmp_epu64_mask(m, a, b, p) __extension__ ({ \
+#define _mm512_mask_cmp_epu64_mask(m, a, b, p) \
(__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)(__m512i)(a), \
(__v8di)(__m512i)(b), (int)(p), \
- (__mmask8)(m)); })
-
-#define _mm512_rol_epi32(a, b) __extension__ ({ \
- (__m512i)__builtin_ia32_prold512_mask((__v16si)(__m512i)(a), (int)(b), \
- (__v16si)_mm512_setzero_si512(), \
- (__mmask16)-1); })
-
-#define _mm512_mask_rol_epi32(W, U, a, b) __extension__ ({ \
- (__m512i)__builtin_ia32_prold512_mask((__v16si)(__m512i)(a), (int)(b), \
- (__v16si)(__m512i)(W), \
- (__mmask16)(U)); })
-
-#define _mm512_maskz_rol_epi32(U, a, b) __extension__ ({ \
- (__m512i)__builtin_ia32_prold512_mask((__v16si)(__m512i)(a), (int)(b), \
- (__v16si)_mm512_setzero_si512(), \
- (__mmask16)(U)); })
-
-#define _mm512_rol_epi64(a, b) __extension__ ({ \
- (__m512i)__builtin_ia32_prolq512_mask((__v8di)(__m512i)(a), (int)(b), \
- (__v8di)_mm512_setzero_si512(), \
- (__mmask8)-1); })
-
-#define _mm512_mask_rol_epi64(W, U, a, b) __extension__ ({ \
- (__m512i)__builtin_ia32_prolq512_mask((__v8di)(__m512i)(a), (int)(b), \
- (__v8di)(__m512i)(W), (__mmask8)(U)); })
-
-#define _mm512_maskz_rol_epi64(U, a, b) __extension__ ({ \
- (__m512i)__builtin_ia32_prolq512_mask((__v8di)(__m512i)(a), (int)(b), \
- (__v8di)_mm512_setzero_si512(), \
- (__mmask8)(U)); })
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+ (__mmask8)(m))
+
+#define _mm512_rol_epi32(a, b) \
+ (__m512i)__builtin_ia32_prold512((__v16si)(__m512i)(a), (int)(b))
+
+#define _mm512_mask_rol_epi32(W, U, a, b) \
+ (__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
+ (__v16si)_mm512_rol_epi32((a), (b)), \
+ (__v16si)(__m512i)(W))
+
+#define _mm512_maskz_rol_epi32(U, a, b) \
+ (__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
+ (__v16si)_mm512_rol_epi32((a), (b)), \
+ (__v16si)_mm512_setzero_si512())
+
+#define _mm512_rol_epi64(a, b) \
+ (__m512i)__builtin_ia32_prolq512((__v8di)(__m512i)(a), (int)(b))
+
+#define _mm512_mask_rol_epi64(W, U, a, b) \
+ (__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
+ (__v8di)_mm512_rol_epi64((a), (b)), \
+ (__v8di)(__m512i)(W))
+
+#define _mm512_maskz_rol_epi64(U, a, b) \
+ (__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
+ (__v8di)_mm512_rol_epi64((a), (b)), \
+ (__v8di)_mm512_setzero_si512())
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_rolv_epi32 (__m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_prolvd512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) -1);
+ return (__m512i)__builtin_ia32_prolvd512((__v16si)__A, (__v16si)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_rolv_epi32 (__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_prolvd512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si) __W,
- (__mmask16) __U);
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_rolv_epi32(__A, __B),
+ (__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_rolv_epi32 (__mmask16 __U, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_prolvd512_mask ((__v16si) __A,
- (__v16si) __B,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) __U);
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_rolv_epi32(__A, __B),
+ (__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_rolv_epi64 (__m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_prolvq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) -1);
+ return (__m512i)__builtin_ia32_prolvq512((__v8di)__A, (__v8di)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_rolv_epi64 (__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_prolvq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di) __W,
- (__mmask8) __U);
+ return (__m512i)__builtin_ia32_selectq_512(__U,
+ (__v8di)_mm512_rolv_epi64(__A, __B),
+ (__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_rolv_epi64 (__mmask8 __U, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_prolvq512_mask ((__v8di) __A,
- (__v8di) __B,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) __U);
+ return (__m512i)__builtin_ia32_selectq_512(__U,
+ (__v8di)_mm512_rolv_epi64(__A, __B),
+ (__v8di)_mm512_setzero_si512());
}
-#define _mm512_ror_epi32(A, B) __extension__ ({ \
- (__m512i)__builtin_ia32_prord512_mask((__v16si)(__m512i)(A), (int)(B), \
- (__v16si)_mm512_setzero_si512(), \
- (__mmask16)-1); })
+#define _mm512_ror_epi32(A, B) \
+ (__m512i)__builtin_ia32_prord512((__v16si)(__m512i)(A), (int)(B))
-#define _mm512_mask_ror_epi32(W, U, A, B) __extension__ ({ \
- (__m512i)__builtin_ia32_prord512_mask((__v16si)(__m512i)(A), (int)(B), \
- (__v16si)(__m512i)(W), \
- (__mmask16)(U)); })
+#define _mm512_mask_ror_epi32(W, U, A, B) \
+ (__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
+ (__v16si)_mm512_ror_epi32((A), (B)), \
+ (__v16si)(__m512i)(W))
-#define _mm512_maskz_ror_epi32(U, A, B) __extension__ ({ \
- (__m512i)__builtin_ia32_prord512_mask((__v16si)(__m512i)(A), (int)(B), \
- (__v16si)_mm512_setzero_si512(), \
- (__mmask16)(U)); })
+#define _mm512_maskz_ror_epi32(U, A, B) \
+ (__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
+ (__v16si)_mm512_ror_epi32((A), (B)), \
+ (__v16si)_mm512_setzero_si512())
-#define _mm512_ror_epi64(A, B) __extension__ ({ \
- (__m512i)__builtin_ia32_prorq512_mask((__v8di)(__m512i)(A), (int)(B), \
- (__v8di)_mm512_setzero_si512(), \
- (__mmask8)-1); })
+#define _mm512_ror_epi64(A, B) \
+ (__m512i)__builtin_ia32_prorq512((__v8di)(__m512i)(A), (int)(B))
-#define _mm512_mask_ror_epi64(W, U, A, B) __extension__ ({ \
- (__m512i)__builtin_ia32_prorq512_mask((__v8di)(__m512i)(A), (int)(B), \
- (__v8di)(__m512i)(W), (__mmask8)(U)); })
+#define _mm512_mask_ror_epi64(W, U, A, B) \
+ (__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
+ (__v8di)_mm512_ror_epi64((A), (B)), \
+ (__v8di)(__m512i)(W))
-#define _mm512_maskz_ror_epi64(U, A, B) __extension__ ({ \
- (__m512i)__builtin_ia32_prorq512_mask((__v8di)(__m512i)(A), (int)(B), \
- (__v8di)_mm512_setzero_si512(), \
- (__mmask8)(U)); })
+#define _mm512_maskz_ror_epi64(U, A, B) \
+ (__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
+ (__v8di)_mm512_ror_epi64((A), (B)), \
+ (__v8di)_mm512_setzero_si512())
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_slli_epi32(__m512i __A, int __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_slli_epi32(__m512i __A, unsigned int __B)
{
return (__m512i)__builtin_ia32_pslldi512((__v16si)__A, __B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_slli_epi32(__m512i __W, __mmask16 __U, __m512i __A, int __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_slli_epi32(__m512i __W, __mmask16 __U, __m512i __A,
+ unsigned int __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_slli_epi32(__A, __B),
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_slli_epi32(__mmask16 __U, __m512i __A, int __B) {
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_slli_epi32(__mmask16 __U, __m512i __A, unsigned int __B) {
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_slli_epi32(__A, __B),
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_slli_epi64(__m512i __A, int __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_slli_epi64(__m512i __A, unsigned int __B)
{
return (__m512i)__builtin_ia32_psllqi512((__v8di)__A, __B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_slli_epi64(__m512i __W, __mmask8 __U, __m512i __A, int __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_slli_epi64(__m512i __W, __mmask8 __U, __m512i __A, unsigned int __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_slli_epi64(__A, __B),
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_slli_epi64(__mmask8 __U, __m512i __A, int __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_slli_epi64(__mmask8 __U, __m512i __A, unsigned int __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_slli_epi64(__A, __B),
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_srli_epi32(__m512i __A, int __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_srli_epi32(__m512i __A, unsigned int __B)
{
return (__m512i)__builtin_ia32_psrldi512((__v16si)__A, __B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_srli_epi32(__m512i __W, __mmask16 __U, __m512i __A, int __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_srli_epi32(__m512i __W, __mmask16 __U, __m512i __A,
+ unsigned int __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_srli_epi32(__A, __B),
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_srli_epi32(__mmask16 __U, __m512i __A, int __B) {
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_srli_epi32(__mmask16 __U, __m512i __A, unsigned int __B) {
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_srli_epi32(__A, __B),
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_srli_epi64(__m512i __A, int __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_srli_epi64(__m512i __A, unsigned int __B)
{
return (__m512i)__builtin_ia32_psrlqi512((__v8di)__A, __B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_srli_epi64(__m512i __W, __mmask8 __U, __m512i __A, int __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_srli_epi64(__m512i __W, __mmask8 __U, __m512i __A,
+ unsigned int __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_srli_epi64(__A, __B),
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_srli_epi64(__mmask8 __U, __m512i __A, int __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_srli_epi64(__mmask8 __U, __m512i __A,
+ unsigned int __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_srli_epi64(__A, __B),
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_load_epi32 (__m512i __W, __mmask16 __U, void const *__P)
{
return (__m512i) __builtin_ia32_movdqa32load512_mask ((const __v16si *) __P,
(__mmask16) __U);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_load_epi32 (__mmask16 __U, void const *__P)
{
return (__m512i) __builtin_ia32_movdqa32load512_mask ((const __v16si *) __P,
(__mmask16) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_store_epi32 (void *__P, __mmask16 __U, __m512i __A)
{
__builtin_ia32_movdqa32store512_mask ((__v16si *) __P, (__v16si) __A,
(__mmask16) __U);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_mov_epi32 (__m512i __W, __mmask16 __U, __m512i __A)
{
return (__m512i) __builtin_ia32_selectd_512 ((__mmask16) __U,
(__v16si) __W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_mov_epi32 (__mmask16 __U, __m512i __A)
{
return (__m512i) __builtin_ia32_selectd_512 ((__mmask16) __U,
(__v16si) _mm512_setzero_si512 ());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_mov_epi64 (__m512i __W, __mmask8 __U, __m512i __A)
{
return (__m512i) __builtin_ia32_selectq_512 ((__mmask8) __U,
(__v8di) __W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_mov_epi64 (__mmask8 __U, __m512i __A)
{
return (__m512i) __builtin_ia32_selectq_512 ((__mmask8) __U,
(__v8di) _mm512_setzero_si512 ());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_load_epi64 (__m512i __W, __mmask8 __U, void const *__P)
{
return (__m512i) __builtin_ia32_movdqa64load512_mask ((const __v8di *) __P,
(__mmask8) __U);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_load_epi64 (__mmask8 __U, void const *__P)
{
return (__m512i) __builtin_ia32_movdqa64load512_mask ((const __v8di *) __P,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_store_epi64 (void *__P, __mmask8 __U, __m512i __A)
{
__builtin_ia32_movdqa64store512_mask ((__v8di *) __P, (__v8di) __A,
(__mmask8) __U);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_movedup_pd (__m512d __A)
{
return (__m512d)__builtin_shufflevector((__v8df)__A, (__v8df)__A,
0, 0, 2, 2, 4, 4, 6, 6);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_movedup_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_movedup_pd (__mmask8 __U, __m512d __A)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_setzero_pd());
}
-#define _mm512_fixupimm_round_pd(A, B, C, imm, R) __extension__ ({ \
+#define _mm512_fixupimm_round_pd(A, B, C, imm, R) \
(__m512d)__builtin_ia32_fixupimmpd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8di)(__m512i)(C), (int)(imm), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_fixupimm_round_pd(A, U, B, C, imm, R) __extension__ ({ \
+#define _mm512_mask_fixupimm_round_pd(A, U, B, C, imm, R) \
(__m512d)__builtin_ia32_fixupimmpd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8di)(__m512i)(C), (int)(imm), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_fixupimm_pd(A, B, C, imm) __extension__ ({ \
+#define _mm512_fixupimm_pd(A, B, C, imm) \
(__m512d)__builtin_ia32_fixupimmpd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8di)(__m512i)(C), (int)(imm), \
(__mmask8)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_mask_fixupimm_pd(A, U, B, C, imm) __extension__ ({ \
+#define _mm512_mask_fixupimm_pd(A, U, B, C, imm) \
(__m512d)__builtin_ia32_fixupimmpd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8di)(__m512i)(C), (int)(imm), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_maskz_fixupimm_round_pd(U, A, B, C, imm, R) __extension__ ({ \
+#define _mm512_maskz_fixupimm_round_pd(U, A, B, C, imm, R) \
(__m512d)__builtin_ia32_fixupimmpd512_maskz((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8di)(__m512i)(C), \
(int)(imm), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_maskz_fixupimm_pd(U, A, B, C, imm) __extension__ ({ \
+#define _mm512_maskz_fixupimm_pd(U, A, B, C, imm) \
(__m512d)__builtin_ia32_fixupimmpd512_maskz((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8di)(__m512i)(C), \
(int)(imm), (__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_fixupimm_round_ps(A, B, C, imm, R) __extension__ ({ \
+#define _mm512_fixupimm_round_ps(A, B, C, imm, R) \
(__m512)__builtin_ia32_fixupimmps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16si)(__m512i)(C), (int)(imm), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_fixupimm_round_ps(A, U, B, C, imm, R) __extension__ ({ \
+#define _mm512_mask_fixupimm_round_ps(A, U, B, C, imm, R) \
(__m512)__builtin_ia32_fixupimmps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16si)(__m512i)(C), (int)(imm), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_fixupimm_ps(A, B, C, imm) __extension__ ({ \
+#define _mm512_fixupimm_ps(A, B, C, imm) \
(__m512)__builtin_ia32_fixupimmps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16si)(__m512i)(C), (int)(imm), \
(__mmask16)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_mask_fixupimm_ps(A, U, B, C, imm) __extension__ ({ \
+#define _mm512_mask_fixupimm_ps(A, U, B, C, imm) \
(__m512)__builtin_ia32_fixupimmps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16si)(__m512i)(C), (int)(imm), \
(__mmask16)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_maskz_fixupimm_round_ps(U, A, B, C, imm, R) __extension__ ({ \
+#define _mm512_maskz_fixupimm_round_ps(U, A, B, C, imm, R) \
(__m512)__builtin_ia32_fixupimmps512_maskz((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16si)(__m512i)(C), \
(int)(imm), (__mmask16)(U), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_maskz_fixupimm_ps(U, A, B, C, imm) __extension__ ({ \
+#define _mm512_maskz_fixupimm_ps(U, A, B, C, imm) \
(__m512)__builtin_ia32_fixupimmps512_maskz((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16si)(__m512i)(C), \
(int)(imm), (__mmask16)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_fixupimm_round_sd(A, B, C, imm, R) __extension__ ({ \
+#define _mm_fixupimm_round_sd(A, B, C, imm, R) \
(__m128d)__builtin_ia32_fixupimmsd_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2di)(__m128i)(C), (int)(imm), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_fixupimm_round_sd(A, U, B, C, imm, R) __extension__ ({ \
+#define _mm_mask_fixupimm_round_sd(A, U, B, C, imm, R) \
(__m128d)__builtin_ia32_fixupimmsd_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2di)(__m128i)(C), (int)(imm), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_fixupimm_sd(A, B, C, imm) __extension__ ({ \
+#define _mm_fixupimm_sd(A, B, C, imm) \
(__m128d)__builtin_ia32_fixupimmsd_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2di)(__m128i)(C), (int)(imm), \
(__mmask8)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_fixupimm_sd(A, U, B, C, imm) __extension__ ({ \
+#define _mm_mask_fixupimm_sd(A, U, B, C, imm) \
(__m128d)__builtin_ia32_fixupimmsd_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2di)(__m128i)(C), (int)(imm), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_maskz_fixupimm_round_sd(U, A, B, C, imm, R) __extension__ ({ \
+#define _mm_maskz_fixupimm_round_sd(U, A, B, C, imm, R) \
(__m128d)__builtin_ia32_fixupimmsd_maskz((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2di)(__m128i)(C), (int)(imm), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_maskz_fixupimm_sd(U, A, B, C, imm) __extension__ ({ \
+#define _mm_maskz_fixupimm_sd(U, A, B, C, imm) \
(__m128d)__builtin_ia32_fixupimmsd_maskz((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2di)(__m128i)(C), (int)(imm), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_fixupimm_round_ss(A, B, C, imm, R) __extension__ ({ \
+#define _mm_fixupimm_round_ss(A, B, C, imm, R) \
(__m128)__builtin_ia32_fixupimmss_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4si)(__m128i)(C), (int)(imm), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_fixupimm_round_ss(A, U, B, C, imm, R) __extension__ ({ \
+#define _mm_mask_fixupimm_round_ss(A, U, B, C, imm, R) \
(__m128)__builtin_ia32_fixupimmss_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4si)(__m128i)(C), (int)(imm), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_fixupimm_ss(A, B, C, imm) __extension__ ({ \
+#define _mm_fixupimm_ss(A, B, C, imm) \
(__m128)__builtin_ia32_fixupimmss_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4si)(__m128i)(C), (int)(imm), \
(__mmask8)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_fixupimm_ss(A, U, B, C, imm) __extension__ ({ \
+#define _mm_mask_fixupimm_ss(A, U, B, C, imm) \
(__m128)__builtin_ia32_fixupimmss_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4si)(__m128i)(C), (int)(imm), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_maskz_fixupimm_round_ss(U, A, B, C, imm, R) __extension__ ({ \
+#define _mm_maskz_fixupimm_round_ss(U, A, B, C, imm, R) \
(__m128)__builtin_ia32_fixupimmss_maskz((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4si)(__m128i)(C), (int)(imm), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_maskz_fixupimm_ss(U, A, B, C, imm) __extension__ ({ \
+#define _mm_maskz_fixupimm_ss(U, A, B, C, imm) \
(__m128)__builtin_ia32_fixupimmss_maskz((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4si)(__m128i)(C), (int)(imm), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_getexp_round_sd(A, B, R) __extension__ ({ \
+#define _mm_getexp_round_sd(A, B, R) \
(__m128d)__builtin_ia32_getexpsd128_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_getexp_sd (__m128d __A, __m128d __B)
{
return (__m128d) __builtin_ia32_getexpsd128_round_mask ((__v2df) __A,
(__v2df) __B, (__v2df) _mm_setzero_pd(), (__mmask8) -1, _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_getexp_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
return (__m128d) __builtin_ia32_getexpsd128_round_mask ( (__v2df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask_getexp_round_sd(W, U, A, B, R) __extension__ ({\
+#define _mm_mask_getexp_round_sd(W, U, A, B, R) \
(__m128d)__builtin_ia32_getexpsd128_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_getexp_sd (__mmask8 __U, __m128d __A, __m128d __B)
{
return (__m128d) __builtin_ia32_getexpsd128_round_mask ( (__v2df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_maskz_getexp_round_sd(U, A, B, R) __extension__ ({\
+#define _mm_maskz_getexp_round_sd(U, A, B, R) \
(__m128d)__builtin_ia32_getexpsd128_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_getexp_round_ss(A, B, R) __extension__ ({ \
+#define _mm_getexp_round_ss(A, B, R) \
(__m128)__builtin_ia32_getexpss128_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_getexp_ss (__m128 __A, __m128 __B)
{
return (__m128) __builtin_ia32_getexpss128_round_mask ((__v4sf) __A,
(__v4sf) __B, (__v4sf) _mm_setzero_ps(), (__mmask8) -1, _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_getexp_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
return (__m128) __builtin_ia32_getexpss128_round_mask ((__v4sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask_getexp_round_ss(W, U, A, B, R) __extension__ ({\
+#define _mm_mask_getexp_round_ss(W, U, A, B, R) \
(__m128)__builtin_ia32_getexpss128_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_getexp_ss (__mmask8 __U, __m128 __A, __m128 __B)
{
return (__m128) __builtin_ia32_getexpss128_round_mask ((__v4sf) __A,
(__v4sf) __B,
- (__v4sf) _mm_setzero_pd (),
+ (__v4sf) _mm_setzero_ps (),
(__mmask8) __U,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_maskz_getexp_round_ss(U, A, B, R) __extension__ ({\
+#define _mm_maskz_getexp_round_ss(U, A, B, R) \
(__m128)__builtin_ia32_getexpss128_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_getmant_round_sd(A, B, C, D, R) __extension__ ({ \
+#define _mm_getmant_round_sd(A, B, C, D, R) \
(__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(int)(((D)<<2) | (C)), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_getmant_sd(A, B, C, D) __extension__ ({ \
+#define _mm_getmant_sd(A, B, C, D) \
(__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(int)(((D)<<2) | (C)), \
(__v2df)_mm_setzero_pd(), \
(__mmask8)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_getmant_sd(W, U, A, B, C, D) __extension__ ({\
+#define _mm_mask_getmant_sd(W, U, A, B, C, D) \
(__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(int)(((D)<<2) | (C)), \
(__v2df)(__m128d)(W), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_getmant_round_sd(W, U, A, B, C, D, R)({\
+#define _mm_mask_getmant_round_sd(W, U, A, B, C, D, R) \
(__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(int)(((D)<<2) | (C)), \
(__v2df)(__m128d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_maskz_getmant_sd(U, A, B, C, D) __extension__ ({\
+#define _mm_maskz_getmant_sd(U, A, B, C, D) \
(__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(int)(((D)<<2) | (C)), \
(__v2df)_mm_setzero_pd(), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_maskz_getmant_round_sd(U, A, B, C, D, R) __extension__ ({\
+#define _mm_maskz_getmant_round_sd(U, A, B, C, D, R) \
(__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(int)(((D)<<2) | (C)), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_getmant_round_ss(A, B, C, D, R) __extension__ ({ \
+#define _mm_getmant_round_ss(A, B, C, D, R) \
(__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(int)(((D)<<2) | (C)), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_getmant_ss(A, B, C, D) __extension__ ({ \
+#define _mm_getmant_ss(A, B, C, D) \
(__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(int)(((D)<<2) | (C)), \
(__v4sf)_mm_setzero_ps(), \
(__mmask8)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_getmant_ss(W, U, A, B, C, D) __extension__ ({\
+#define _mm_mask_getmant_ss(W, U, A, B, C, D) \
(__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(int)(((D)<<2) | (C)), \
(__v4sf)(__m128)(W), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_getmant_round_ss(W, U, A, B, C, D, R)({\
+#define _mm_mask_getmant_round_ss(W, U, A, B, C, D, R) \
(__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(int)(((D)<<2) | (C)), \
(__v4sf)(__m128)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_maskz_getmant_ss(U, A, B, C, D) __extension__ ({\
+#define _mm_maskz_getmant_ss(U, A, B, C, D) \
(__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(int)(((D)<<2) | (C)), \
- (__v4sf)_mm_setzero_pd(), \
+ (__v4sf)_mm_setzero_ps(), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_maskz_getmant_round_ss(U, A, B, C, D, R) __extension__ ({\
+#define _mm_maskz_getmant_round_ss(U, A, B, C, D, R) \
(__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(int)(((D)<<2) | (C)), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
static __inline__ __mmask16 __DEFAULT_FN_ATTRS
_mm512_kmov (__mmask16 __A)
return __A;
}
-#define _mm_comi_round_sd(A, B, P, R) __extension__ ({\
+#define _mm_comi_round_sd(A, B, P, R) \
(int)__builtin_ia32_vcomisd((__v2df)(__m128d)(A), (__v2df)(__m128d)(B), \
- (int)(P), (int)(R)); })
+ (int)(P), (int)(R))
-#define _mm_comi_round_ss(A, B, P, R) __extension__ ({\
+#define _mm_comi_round_ss(A, B, P, R) \
(int)__builtin_ia32_vcomiss((__v4sf)(__m128)(A), (__v4sf)(__m128)(B), \
- (int)(P), (int)(R)); })
+ (int)(P), (int)(R))
#ifdef __x86_64__
-#define _mm_cvt_roundsd_si64(A, R) __extension__ ({ \
- (long long)__builtin_ia32_vcvtsd2si64((__v2df)(__m128d)(A), (int)(R)); })
+#define _mm_cvt_roundsd_si64(A, R) \
+ (long long)__builtin_ia32_vcvtsd2si64((__v2df)(__m128d)(A), (int)(R))
#endif
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask2_permutex2var_epi32 (__m512i __A, __m512i __I,
- __mmask16 __U, __m512i __B)
-{
- return (__m512i) __builtin_ia32_vpermi2vard512_mask ((__v16si) __A,
- (__v16si) __I
- /* idx */ ,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sll_epi32(__m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_pslld512((__v16si) __A, (__v4si)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sll_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sll_epi32(__mmask16 __U, __m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sll_epi64(__m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_psllq512((__v8di)__A, (__v2di)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sll_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sll_epi64(__mmask8 __U, __m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sllv_epi32(__m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_psllv16si((__v16si)__X, (__v16si)__Y);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sllv_epi32(__m512i __W, __mmask16 __U, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sllv_epi32(__mmask16 __U, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sllv_epi64(__m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_psllv8di((__v8di)__X, (__v8di)__Y);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sllv_epi64(__m512i __W, __mmask8 __U, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sllv_epi64(__mmask8 __U, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sra_epi32(__m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_psrad512((__v16si) __A, (__v4si)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sra_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sra_epi32(__mmask16 __U, __m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sra_epi64(__m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_psraq512((__v8di)__A, (__v2di)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sra_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sra_epi64(__mmask8 __U, __m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srav_epi32(__m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_psrav16si((__v16si)__X, (__v16si)__Y);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srav_epi32(__m512i __W, __mmask16 __U, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srav_epi32(__mmask16 __U, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srav_epi64(__m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_psrav8di((__v8di)__X, (__v8di)__Y);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srav_epi64(__m512i __W, __mmask8 __U, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srav_epi64(__mmask8 __U, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srl_epi32(__m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_psrld512((__v16si) __A, (__v4si)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srl_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srl_epi32(__mmask16 __U, __m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srl_epi64(__m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_psrlq512((__v8di)__A, (__v2di)__B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srl_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srl_epi64(__mmask8 __U, __m512i __A, __m128i __B)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srlv_epi32(__m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_psrlv16si((__v16si)__X, (__v16si)__Y);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srlv_epi32(__m512i __W, __mmask16 __U, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srlv_epi32(__mmask16 __U, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srlv_epi64 (__m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_psrlv8di((__v8di)__X, (__v8di)__Y);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srlv_epi64(__m512i __W, __mmask8 __U, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srlv_epi64(__mmask8 __U, __m512i __X, __m512i __Y)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
(__v8di)_mm512_setzero_si512());
}
-#define _mm512_ternarylogic_epi32(A, B, C, imm) __extension__ ({ \
+#define _mm512_ternarylogic_epi32(A, B, C, imm) \
(__m512i)__builtin_ia32_pternlogd512_mask((__v16si)(__m512i)(A), \
(__v16si)(__m512i)(B), \
(__v16si)(__m512i)(C), (int)(imm), \
- (__mmask16)-1); })
+ (__mmask16)-1)
-#define _mm512_mask_ternarylogic_epi32(A, U, B, C, imm) __extension__ ({ \
+#define _mm512_mask_ternarylogic_epi32(A, U, B, C, imm) \
(__m512i)__builtin_ia32_pternlogd512_mask((__v16si)(__m512i)(A), \
(__v16si)(__m512i)(B), \
(__v16si)(__m512i)(C), (int)(imm), \
- (__mmask16)(U)); })
+ (__mmask16)(U))
-#define _mm512_maskz_ternarylogic_epi32(U, A, B, C, imm) __extension__ ({ \
+#define _mm512_maskz_ternarylogic_epi32(U, A, B, C, imm) \
(__m512i)__builtin_ia32_pternlogd512_maskz((__v16si)(__m512i)(A), \
(__v16si)(__m512i)(B), \
(__v16si)(__m512i)(C), \
- (int)(imm), (__mmask16)(U)); })
+ (int)(imm), (__mmask16)(U))
-#define _mm512_ternarylogic_epi64(A, B, C, imm) __extension__ ({ \
+#define _mm512_ternarylogic_epi64(A, B, C, imm) \
(__m512i)__builtin_ia32_pternlogq512_mask((__v8di)(__m512i)(A), \
(__v8di)(__m512i)(B), \
(__v8di)(__m512i)(C), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm512_mask_ternarylogic_epi64(A, U, B, C, imm) __extension__ ({ \
+#define _mm512_mask_ternarylogic_epi64(A, U, B, C, imm) \
(__m512i)__builtin_ia32_pternlogq512_mask((__v8di)(__m512i)(A), \
(__v8di)(__m512i)(B), \
(__v8di)(__m512i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm512_maskz_ternarylogic_epi64(U, A, B, C, imm) __extension__ ({ \
+#define _mm512_maskz_ternarylogic_epi64(U, A, B, C, imm) \
(__m512i)__builtin_ia32_pternlogq512_maskz((__v8di)(__m512i)(A), \
(__v8di)(__m512i)(B), \
(__v8di)(__m512i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
#ifdef __x86_64__
-#define _mm_cvt_roundsd_i64(A, R) __extension__ ({ \
- (long long)__builtin_ia32_vcvtsd2si64((__v2df)(__m128d)(A), (int)(R)); })
+#define _mm_cvt_roundsd_i64(A, R) \
+ (long long)__builtin_ia32_vcvtsd2si64((__v2df)(__m128d)(A), (int)(R))
#endif
-#define _mm_cvt_roundsd_si32(A, R) __extension__ ({ \
- (int)__builtin_ia32_vcvtsd2si32((__v2df)(__m128d)(A), (int)(R)); })
+#define _mm_cvt_roundsd_si32(A, R) \
+ (int)__builtin_ia32_vcvtsd2si32((__v2df)(__m128d)(A), (int)(R))
-#define _mm_cvt_roundsd_i32(A, R) __extension__ ({ \
- (int)__builtin_ia32_vcvtsd2si32((__v2df)(__m128d)(A), (int)(R)); })
+#define _mm_cvt_roundsd_i32(A, R) \
+ (int)__builtin_ia32_vcvtsd2si32((__v2df)(__m128d)(A), (int)(R))
-#define _mm_cvt_roundsd_u32(A, R) __extension__ ({ \
- (unsigned int)__builtin_ia32_vcvtsd2usi32((__v2df)(__m128d)(A), (int)(R)); })
+#define _mm_cvt_roundsd_u32(A, R) \
+ (unsigned int)__builtin_ia32_vcvtsd2usi32((__v2df)(__m128d)(A), (int)(R))
-static __inline__ unsigned __DEFAULT_FN_ATTRS
+static __inline__ unsigned __DEFAULT_FN_ATTRS128
_mm_cvtsd_u32 (__m128d __A)
{
return (unsigned) __builtin_ia32_vcvtsd2usi32 ((__v2df) __A,
}
#ifdef __x86_64__
-#define _mm_cvt_roundsd_u64(A, R) __extension__ ({ \
+#define _mm_cvt_roundsd_u64(A, R) \
(unsigned long long)__builtin_ia32_vcvtsd2usi64((__v2df)(__m128d)(A), \
- (int)(R)); })
+ (int)(R))
-static __inline__ unsigned long long __DEFAULT_FN_ATTRS
+static __inline__ unsigned long long __DEFAULT_FN_ATTRS128
_mm_cvtsd_u64 (__m128d __A)
{
return (unsigned long long) __builtin_ia32_vcvtsd2usi64 ((__v2df)
}
#endif
-#define _mm_cvt_roundss_si32(A, R) __extension__ ({ \
- (int)__builtin_ia32_vcvtss2si32((__v4sf)(__m128)(A), (int)(R)); })
+#define _mm_cvt_roundss_si32(A, R) \
+ (int)__builtin_ia32_vcvtss2si32((__v4sf)(__m128)(A), (int)(R))
-#define _mm_cvt_roundss_i32(A, R) __extension__ ({ \
- (int)__builtin_ia32_vcvtss2si32((__v4sf)(__m128)(A), (int)(R)); })
+#define _mm_cvt_roundss_i32(A, R) \
+ (int)__builtin_ia32_vcvtss2si32((__v4sf)(__m128)(A), (int)(R))
#ifdef __x86_64__
-#define _mm_cvt_roundss_si64(A, R) __extension__ ({ \
- (long long)__builtin_ia32_vcvtss2si64((__v4sf)(__m128)(A), (int)(R)); })
+#define _mm_cvt_roundss_si64(A, R) \
+ (long long)__builtin_ia32_vcvtss2si64((__v4sf)(__m128)(A), (int)(R))
-#define _mm_cvt_roundss_i64(A, R) __extension__ ({ \
- (long long)__builtin_ia32_vcvtss2si64((__v4sf)(__m128)(A), (int)(R)); })
+#define _mm_cvt_roundss_i64(A, R) \
+ (long long)__builtin_ia32_vcvtss2si64((__v4sf)(__m128)(A), (int)(R))
#endif
-#define _mm_cvt_roundss_u32(A, R) __extension__ ({ \
- (unsigned int)__builtin_ia32_vcvtss2usi32((__v4sf)(__m128)(A), (int)(R)); })
+#define _mm_cvt_roundss_u32(A, R) \
+ (unsigned int)__builtin_ia32_vcvtss2usi32((__v4sf)(__m128)(A), (int)(R))
-static __inline__ unsigned __DEFAULT_FN_ATTRS
+static __inline__ unsigned __DEFAULT_FN_ATTRS128
_mm_cvtss_u32 (__m128 __A)
{
return (unsigned) __builtin_ia32_vcvtss2usi32 ((__v4sf) __A,
}
#ifdef __x86_64__
-#define _mm_cvt_roundss_u64(A, R) __extension__ ({ \
+#define _mm_cvt_roundss_u64(A, R) \
(unsigned long long)__builtin_ia32_vcvtss2usi64((__v4sf)(__m128)(A), \
- (int)(R)); })
+ (int)(R))
-static __inline__ unsigned long long __DEFAULT_FN_ATTRS
+static __inline__ unsigned long long __DEFAULT_FN_ATTRS128
_mm_cvtss_u64 (__m128 __A)
{
return (unsigned long long) __builtin_ia32_vcvtss2usi64 ((__v4sf)
}
#endif
-#define _mm_cvtt_roundsd_i32(A, R) __extension__ ({ \
- (int)__builtin_ia32_vcvttsd2si32((__v2df)(__m128d)(A), (int)(R)); })
+#define _mm_cvtt_roundsd_i32(A, R) \
+ (int)__builtin_ia32_vcvttsd2si32((__v2df)(__m128d)(A), (int)(R))
-#define _mm_cvtt_roundsd_si32(A, R) __extension__ ({ \
- (int)__builtin_ia32_vcvttsd2si32((__v2df)(__m128d)(A), (int)(R)); })
+#define _mm_cvtt_roundsd_si32(A, R) \
+ (int)__builtin_ia32_vcvttsd2si32((__v2df)(__m128d)(A), (int)(R))
-static __inline__ int __DEFAULT_FN_ATTRS
+static __inline__ int __DEFAULT_FN_ATTRS128
_mm_cvttsd_i32 (__m128d __A)
{
return (int) __builtin_ia32_vcvttsd2si32 ((__v2df) __A,
}
#ifdef __x86_64__
-#define _mm_cvtt_roundsd_si64(A, R) __extension__ ({ \
- (long long)__builtin_ia32_vcvttsd2si64((__v2df)(__m128d)(A), (int)(R)); })
+#define _mm_cvtt_roundsd_si64(A, R) \
+ (long long)__builtin_ia32_vcvttsd2si64((__v2df)(__m128d)(A), (int)(R))
-#define _mm_cvtt_roundsd_i64(A, R) __extension__ ({ \
- (long long)__builtin_ia32_vcvttsd2si64((__v2df)(__m128d)(A), (int)(R)); })
+#define _mm_cvtt_roundsd_i64(A, R) \
+ (long long)__builtin_ia32_vcvttsd2si64((__v2df)(__m128d)(A), (int)(R))
-static __inline__ long long __DEFAULT_FN_ATTRS
+static __inline__ long long __DEFAULT_FN_ATTRS128
_mm_cvttsd_i64 (__m128d __A)
{
return (long long) __builtin_ia32_vcvttsd2si64 ((__v2df) __A,
}
#endif
-#define _mm_cvtt_roundsd_u32(A, R) __extension__ ({ \
- (unsigned int)__builtin_ia32_vcvttsd2usi32((__v2df)(__m128d)(A), (int)(R)); })
+#define _mm_cvtt_roundsd_u32(A, R) \
+ (unsigned int)__builtin_ia32_vcvttsd2usi32((__v2df)(__m128d)(A), (int)(R))
-static __inline__ unsigned __DEFAULT_FN_ATTRS
+static __inline__ unsigned __DEFAULT_FN_ATTRS128
_mm_cvttsd_u32 (__m128d __A)
{
return (unsigned) __builtin_ia32_vcvttsd2usi32 ((__v2df) __A,
}
#ifdef __x86_64__
-#define _mm_cvtt_roundsd_u64(A, R) __extension__ ({ \
+#define _mm_cvtt_roundsd_u64(A, R) \
(unsigned long long)__builtin_ia32_vcvttsd2usi64((__v2df)(__m128d)(A), \
- (int)(R)); })
+ (int)(R))
-static __inline__ unsigned long long __DEFAULT_FN_ATTRS
+static __inline__ unsigned long long __DEFAULT_FN_ATTRS128
_mm_cvttsd_u64 (__m128d __A)
{
return (unsigned long long) __builtin_ia32_vcvttsd2usi64 ((__v2df)
}
#endif
-#define _mm_cvtt_roundss_i32(A, R) __extension__ ({ \
- (int)__builtin_ia32_vcvttss2si32((__v4sf)(__m128)(A), (int)(R)); })
+#define _mm_cvtt_roundss_i32(A, R) \
+ (int)__builtin_ia32_vcvttss2si32((__v4sf)(__m128)(A), (int)(R))
-#define _mm_cvtt_roundss_si32(A, R) __extension__ ({ \
- (int)__builtin_ia32_vcvttss2si32((__v4sf)(__m128)(A), (int)(R)); })
+#define _mm_cvtt_roundss_si32(A, R) \
+ (int)__builtin_ia32_vcvttss2si32((__v4sf)(__m128)(A), (int)(R))
-static __inline__ int __DEFAULT_FN_ATTRS
+static __inline__ int __DEFAULT_FN_ATTRS128
_mm_cvttss_i32 (__m128 __A)
{
return (int) __builtin_ia32_vcvttss2si32 ((__v4sf) __A,
}
#ifdef __x86_64__
-#define _mm_cvtt_roundss_i64(A, R) __extension__ ({ \
- (long long)__builtin_ia32_vcvttss2si64((__v4sf)(__m128)(A), (int)(R)); })
+#define _mm_cvtt_roundss_i64(A, R) \
+ (long long)__builtin_ia32_vcvttss2si64((__v4sf)(__m128)(A), (int)(R))
-#define _mm_cvtt_roundss_si64(A, R) __extension__ ({ \
- (long long)__builtin_ia32_vcvttss2si64((__v4sf)(__m128)(A), (int)(R)); })
+#define _mm_cvtt_roundss_si64(A, R) \
+ (long long)__builtin_ia32_vcvttss2si64((__v4sf)(__m128)(A), (int)(R))
-static __inline__ long long __DEFAULT_FN_ATTRS
+static __inline__ long long __DEFAULT_FN_ATTRS128
_mm_cvttss_i64 (__m128 __A)
{
return (long long) __builtin_ia32_vcvttss2si64 ((__v4sf) __A,
}
#endif
-#define _mm_cvtt_roundss_u32(A, R) __extension__ ({ \
- (unsigned int)__builtin_ia32_vcvttss2usi32((__v4sf)(__m128)(A), (int)(R)); })
+#define _mm_cvtt_roundss_u32(A, R) \
+ (unsigned int)__builtin_ia32_vcvttss2usi32((__v4sf)(__m128)(A), (int)(R))
-static __inline__ unsigned __DEFAULT_FN_ATTRS
+static __inline__ unsigned __DEFAULT_FN_ATTRS128
_mm_cvttss_u32 (__m128 __A)
{
return (unsigned) __builtin_ia32_vcvttss2usi32 ((__v4sf) __A,
}
#ifdef __x86_64__
-#define _mm_cvtt_roundss_u64(A, R) __extension__ ({ \
+#define _mm_cvtt_roundss_u64(A, R) \
(unsigned long long)__builtin_ia32_vcvttss2usi64((__v4sf)(__m128)(A), \
- (int)(R)); })
+ (int)(R))
-static __inline__ unsigned long long __DEFAULT_FN_ATTRS
+static __inline__ unsigned long long __DEFAULT_FN_ATTRS128
_mm_cvttss_u64 (__m128 __A)
{
return (unsigned long long) __builtin_ia32_vcvttss2usi64 ((__v4sf)
}
#endif
-static __inline__ __m512d __DEFAULT_FN_ATTRS
-_mm512_mask2_permutex2var_pd (__m512d __A, __m512i __I, __mmask8 __U,
- __m512d __B)
-{
- return (__m512d) __builtin_ia32_vpermi2varpd512_mask ((__v8df) __A,
- (__v8di) __I
- /* idx */ ,
- (__v8df) __B,
- (__mmask8) __U);
-}
-
-static __inline__ __m512 __DEFAULT_FN_ATTRS
-_mm512_mask2_permutex2var_ps (__m512 __A, __m512i __I, __mmask16 __U,
- __m512 __B)
-{
- return (__m512) __builtin_ia32_vpermi2varps512_mask ((__v16sf) __A,
- (__v16si) __I
- /* idx */ ,
- (__v16sf) __B,
- (__mmask16) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask2_permutex2var_epi64 (__m512i __A, __m512i __I,
- __mmask8 __U, __m512i __B)
-{
- return (__m512i) __builtin_ia32_vpermi2varq512_mask ((__v8di) __A,
- (__v8di) __I
- /* idx */ ,
- (__v8di) __B,
- (__mmask8) __U);
-}
+#define _mm512_permute_pd(X, C) \
+ (__m512d)__builtin_ia32_vpermilpd512((__v8df)(__m512d)(X), (int)(C))
-#define _mm512_permute_pd(X, C) __extension__ ({ \
- (__m512d)__builtin_shufflevector((__v8df)(__m512d)(X), \
- (__v8df)_mm512_undefined_pd(), \
- 0 + (((C) >> 0) & 0x1), \
- 0 + (((C) >> 1) & 0x1), \
- 2 + (((C) >> 2) & 0x1), \
- 2 + (((C) >> 3) & 0x1), \
- 4 + (((C) >> 4) & 0x1), \
- 4 + (((C) >> 5) & 0x1), \
- 6 + (((C) >> 6) & 0x1), \
- 6 + (((C) >> 7) & 0x1)); })
-
-#define _mm512_mask_permute_pd(W, U, X, C) __extension__ ({ \
+#define _mm512_mask_permute_pd(W, U, X, C) \
(__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
(__v8df)_mm512_permute_pd((X), (C)), \
- (__v8df)(__m512d)(W)); })
+ (__v8df)(__m512d)(W))
-#define _mm512_maskz_permute_pd(U, X, C) __extension__ ({ \
+#define _mm512_maskz_permute_pd(U, X, C) \
(__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
(__v8df)_mm512_permute_pd((X), (C)), \
- (__v8df)_mm512_setzero_pd()); })
-
-#define _mm512_permute_ps(X, C) __extension__ ({ \
- (__m512)__builtin_shufflevector((__v16sf)(__m512)(X), \
- (__v16sf)_mm512_undefined_ps(), \
- 0 + (((C) >> 0) & 0x3), \
- 0 + (((C) >> 2) & 0x3), \
- 0 + (((C) >> 4) & 0x3), \
- 0 + (((C) >> 6) & 0x3), \
- 4 + (((C) >> 0) & 0x3), \
- 4 + (((C) >> 2) & 0x3), \
- 4 + (((C) >> 4) & 0x3), \
- 4 + (((C) >> 6) & 0x3), \
- 8 + (((C) >> 0) & 0x3), \
- 8 + (((C) >> 2) & 0x3), \
- 8 + (((C) >> 4) & 0x3), \
- 8 + (((C) >> 6) & 0x3), \
- 12 + (((C) >> 0) & 0x3), \
- 12 + (((C) >> 2) & 0x3), \
- 12 + (((C) >> 4) & 0x3), \
- 12 + (((C) >> 6) & 0x3)); })
-
-#define _mm512_mask_permute_ps(W, U, X, C) __extension__ ({ \
+ (__v8df)_mm512_setzero_pd())
+
+#define _mm512_permute_ps(X, C) \
+ (__m512)__builtin_ia32_vpermilps512((__v16sf)(__m512)(X), (int)(C))
+
+#define _mm512_mask_permute_ps(W, U, X, C) \
(__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
(__v16sf)_mm512_permute_ps((X), (C)), \
- (__v16sf)(__m512)(W)); })
+ (__v16sf)(__m512)(W))
-#define _mm512_maskz_permute_ps(U, X, C) __extension__ ({ \
+#define _mm512_maskz_permute_ps(U, X, C) \
(__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
(__v16sf)_mm512_permute_ps((X), (C)), \
- (__v16sf)_mm512_setzero_ps()); })
+ (__v16sf)_mm512_setzero_ps())
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_permutevar_pd(__m512d __A, __m512i __C)
{
return (__m512d)__builtin_ia32_vpermilvarpd512((__v8df)__A, (__v8di)__C);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_permutevar_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512i __C)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_permutevar_pd(__mmask8 __U, __m512d __A, __m512i __C)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
(__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_permutevar_ps(__m512 __A, __m512i __C)
{
return (__m512)__builtin_ia32_vpermilvarps512((__v16sf)__A, (__v16si)__C);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_permutevar_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512i __C)
{
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_permutevar_ps(__mmask16 __U, __m512 __A, __m512i __C)
{
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_setzero_ps());
}
-static __inline __m512d __DEFAULT_FN_ATTRS
+static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_permutex2var_pd(__m512d __A, __m512i __I, __m512d __B)
{
- return (__m512d) __builtin_ia32_vpermt2varpd512_mask ((__v8di) __I
- /* idx */ ,
- (__v8df) __A,
- (__v8df) __B,
- (__mmask8) -1);
-}
-
-static __inline__ __m512d __DEFAULT_FN_ATTRS
-_mm512_mask_permutex2var_pd (__m512d __A, __mmask8 __U, __m512i __I, __m512d __B)
-{
- return (__m512d) __builtin_ia32_vpermt2varpd512_mask ((__v8di) __I
- /* idx */ ,
- (__v8df) __A,
- (__v8df) __B,
- (__mmask8) __U);
+ return (__m512d)__builtin_ia32_vpermi2varpd512((__v8df)__A, (__v8di)__I,
+ (__v8df)__B);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
-_mm512_maskz_permutex2var_pd (__mmask8 __U, __m512d __A, __m512i __I,
- __m512d __B)
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
+_mm512_mask_permutex2var_pd(__m512d __A, __mmask8 __U, __m512i __I, __m512d __B)
{
- return (__m512d) __builtin_ia32_vpermt2varpd512_maskz ((__v8di) __I
- /* idx */ ,
- (__v8df) __A,
- (__v8df) __B,
- (__mmask8) __U);
+ return (__m512d)__builtin_ia32_selectpd_512(__U,
+ (__v8df)_mm512_permutex2var_pd(__A, __I, __B),
+ (__v8df)__A);
}
-static __inline __m512 __DEFAULT_FN_ATTRS
-_mm512_permutex2var_ps(__m512 __A, __m512i __I, __m512 __B)
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
+_mm512_mask2_permutex2var_pd(__m512d __A, __m512i __I, __mmask8 __U,
+ __m512d __B)
{
- return (__m512) __builtin_ia32_vpermt2varps512_mask ((__v16si) __I
- /* idx */ ,
- (__v16sf) __A,
- (__v16sf) __B,
- (__mmask16) -1);
+ return (__m512d)__builtin_ia32_selectpd_512(__U,
+ (__v8df)_mm512_permutex2var_pd(__A, __I, __B),
+ (__v8df)(__m512d)__I);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
-_mm512_mask_permutex2var_ps (__m512 __A, __mmask16 __U, __m512i __I, __m512 __B)
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
+_mm512_maskz_permutex2var_pd(__mmask8 __U, __m512d __A, __m512i __I,
+ __m512d __B)
{
- return (__m512) __builtin_ia32_vpermt2varps512_mask ((__v16si) __I
- /* idx */ ,
- (__v16sf) __A,
- (__v16sf) __B,
- (__mmask16) __U);
+ return (__m512d)__builtin_ia32_selectpd_512(__U,
+ (__v8df)_mm512_permutex2var_pd(__A, __I, __B),
+ (__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
-_mm512_maskz_permutex2var_ps (__mmask16 __U, __m512 __A, __m512i __I,
- __m512 __B)
+static __inline __m512 __DEFAULT_FN_ATTRS512
+_mm512_permutex2var_ps(__m512 __A, __m512i __I, __m512 __B)
{
- return (__m512) __builtin_ia32_vpermt2varps512_maskz ((__v16si) __I
- /* idx */ ,
- (__v16sf) __A,
- (__v16sf) __B,
- (__mmask16) __U);
+ return (__m512)__builtin_ia32_vpermi2varps512((__v16sf)__A, (__v16si)__I,
+ (__v16sf) __B);
}
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_testn_epi32_mask (__m512i __A, __m512i __B)
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
+_mm512_mask_permutex2var_ps(__m512 __A, __mmask16 __U, __m512i __I, __m512 __B)
{
- return (__mmask16) __builtin_ia32_ptestnmd512 ((__v16si) __A,
- (__v16si) __B,
- (__mmask16) -1);
+ return (__m512)__builtin_ia32_selectps_512(__U,
+ (__v16sf)_mm512_permutex2var_ps(__A, __I, __B),
+ (__v16sf)__A);
}
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm512_mask_testn_epi32_mask (__mmask16 __U, __m512i __A, __m512i __B)
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
+_mm512_mask2_permutex2var_ps(__m512 __A, __m512i __I, __mmask16 __U, __m512 __B)
{
- return (__mmask16) __builtin_ia32_ptestnmd512 ((__v16si) __A,
- (__v16si) __B, __U);
+ return (__m512)__builtin_ia32_selectps_512(__U,
+ (__v16sf)_mm512_permutex2var_ps(__A, __I, __B),
+ (__v16sf)(__m512)__I);
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_testn_epi64_mask (__m512i __A, __m512i __B)
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
+_mm512_maskz_permutex2var_ps(__mmask16 __U, __m512 __A, __m512i __I, __m512 __B)
{
- return (__mmask8) __builtin_ia32_ptestnmq512 ((__v8di) __A,
- (__v8di) __B,
- (__mmask8) -1);
+ return (__m512)__builtin_ia32_selectps_512(__U,
+ (__v16sf)_mm512_permutex2var_ps(__A, __I, __B),
+ (__v16sf)_mm512_setzero_ps());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm512_mask_testn_epi64_mask (__mmask8 __U, __m512i __A, __m512i __B)
-{
- return (__mmask8) __builtin_ia32_ptestnmq512 ((__v8di) __A,
- (__v8di) __B, __U);
-}
-#define _mm512_cvtt_roundpd_epu32(A, R) __extension__ ({ \
+#define _mm512_cvtt_roundpd_epu32(A, R) \
(__m256i)__builtin_ia32_cvttpd2udq512_mask((__v8df)(__m512d)(A), \
(__v8si)_mm256_undefined_si256(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvtt_roundpd_epu32(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvtt_roundpd_epu32(W, U, A, R) \
(__m256i)__builtin_ia32_cvttpd2udq512_mask((__v8df)(__m512d)(A), \
(__v8si)(__m256i)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_cvtt_roundpd_epu32(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvtt_roundpd_epu32(U, A, R) \
(__m256i)__builtin_ia32_cvttpd2udq512_mask((__v8df)(__m512d)(A), \
(__v8si)_mm256_setzero_si256(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvttpd_epu32 (__m512d __A)
{
return (__m256i) __builtin_ia32_cvttpd2udq512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvttpd_epu32 (__m256i __W, __mmask8 __U, __m512d __A)
{
return (__m256i) __builtin_ia32_cvttpd2udq512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvttpd_epu32 (__mmask8 __U, __m512d __A)
{
return (__m256i) __builtin_ia32_cvttpd2udq512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_roundscale_round_sd(A, B, imm, R) __extension__ ({ \
+#define _mm_roundscale_round_sd(A, B, imm, R) \
(__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
(__mmask8)-1, (int)(imm), \
- (int)(R)); })
+ (int)(R))
-#define _mm_roundscale_sd(A, B, imm) __extension__ ({ \
+#define _mm_roundscale_sd(A, B, imm) \
(__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
(__mmask8)-1, (int)(imm), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_roundscale_sd(W, U, A, B, imm) __extension__ ({ \
+#define _mm_mask_roundscale_sd(W, U, A, B, imm) \
(__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(W), \
(__mmask8)(U), (int)(imm), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_roundscale_round_sd(W, U, A, B, I, R) __extension__ ({ \
+#define _mm_mask_roundscale_round_sd(W, U, A, B, I, R) \
(__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(W), \
(__mmask8)(U), (int)(I), \
- (int)(R)); })
+ (int)(R))
-#define _mm_maskz_roundscale_sd(U, A, B, I) __extension__ ({ \
+#define _mm_maskz_roundscale_sd(U, A, B, I) \
(__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
(__mmask8)(U), (int)(I), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_maskz_roundscale_round_sd(U, A, B, I, R) __extension__ ({ \
+#define _mm_maskz_roundscale_round_sd(U, A, B, I, R) \
(__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
(__mmask8)(U), (int)(I), \
- (int)(R)); })
+ (int)(R))
-#define _mm_roundscale_round_ss(A, B, imm, R) __extension__ ({ \
+#define _mm_roundscale_round_ss(A, B, imm, R) \
(__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
(__mmask8)-1, (int)(imm), \
- (int)(R)); })
+ (int)(R))
-#define _mm_roundscale_ss(A, B, imm) __extension__ ({ \
+#define _mm_roundscale_ss(A, B, imm) \
(__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
(__mmask8)-1, (int)(imm), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_roundscale_ss(W, U, A, B, I) __extension__ ({ \
+#define _mm_mask_roundscale_ss(W, U, A, B, I) \
(__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(W), \
(__mmask8)(U), (int)(I), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_roundscale_round_ss(W, U, A, B, I, R) __extension__ ({ \
+#define _mm_mask_roundscale_round_ss(W, U, A, B, I, R) \
(__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(W), \
(__mmask8)(U), (int)(I), \
- (int)(R)); })
+ (int)(R))
-#define _mm_maskz_roundscale_ss(U, A, B, I) __extension__ ({ \
+#define _mm_maskz_roundscale_ss(U, A, B, I) \
(__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
(__mmask8)(U), (int)(I), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_maskz_roundscale_round_ss(U, A, B, I, R) __extension__ ({ \
+#define _mm_maskz_roundscale_round_ss(U, A, B, I, R) \
(__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
(__mmask8)(U), (int)(I), \
- (int)(R)); })
+ (int)(R))
-#define _mm512_scalef_round_pd(A, B, R) __extension__ ({ \
+#define _mm512_scalef_round_pd(A, B, R) \
(__m512d)__builtin_ia32_scalefpd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8df)_mm512_undefined_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_scalef_round_pd(W, U, A, B, R) __extension__ ({ \
+#define _mm512_mask_scalef_round_pd(W, U, A, B, R) \
(__m512d)__builtin_ia32_scalefpd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8df)(__m512d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_scalef_round_pd(U, A, B, R) __extension__ ({ \
+#define _mm512_maskz_scalef_round_pd(U, A, B, R) \
(__m512d)__builtin_ia32_scalefpd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(B), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_scalef_pd (__m512d __A, __m512d __B)
{
return (__m512d) __builtin_ia32_scalefpd512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_scalef_pd (__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
{
return (__m512d) __builtin_ia32_scalefpd512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_scalef_pd (__mmask8 __U, __m512d __A, __m512d __B)
{
return (__m512d) __builtin_ia32_scalefpd512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_scalef_round_ps(A, B, R) __extension__ ({ \
+#define _mm512_scalef_round_ps(A, B, R) \
(__m512)__builtin_ia32_scalefps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16sf)_mm512_undefined_ps(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_scalef_round_ps(W, U, A, B, R) __extension__ ({ \
+#define _mm512_mask_scalef_round_ps(W, U, A, B, R) \
(__m512)__builtin_ia32_scalefps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16sf)(__m512)(W), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_scalef_round_ps(U, A, B, R) __extension__ ({ \
+#define _mm512_maskz_scalef_round_ps(U, A, B, R) \
(__m512)__builtin_ia32_scalefps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(B), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_scalef_ps (__m512 __A, __m512 __B)
{
return (__m512) __builtin_ia32_scalefps512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_scalef_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
{
return (__m512) __builtin_ia32_scalefps512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_scalef_ps (__mmask16 __U, __m512 __A, __m512 __B)
{
return (__m512) __builtin_ia32_scalefps512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_scalef_round_sd(A, B, R) __extension__ ({ \
+#define _mm_scalef_round_sd(A, B, R) \
(__m128d)__builtin_ia32_scalefsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_scalef_sd (__m128d __A, __m128d __B)
{
return (__m128d) __builtin_ia32_scalefsd_round_mask ((__v2df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_scalef_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
return (__m128d) __builtin_ia32_scalefsd_round_mask ( (__v2df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask_scalef_round_sd(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_scalef_round_sd(W, U, A, B, R) \
(__m128d)__builtin_ia32_scalefsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_scalef_sd (__mmask8 __U, __m128d __A, __m128d __B)
{
return (__m128d) __builtin_ia32_scalefsd_round_mask ( (__v2df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_maskz_scalef_round_sd(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_scalef_round_sd(U, A, B, R) \
(__m128d)__builtin_ia32_scalefsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_scalef_round_ss(A, B, R) __extension__ ({ \
+#define _mm_scalef_round_ss(A, B, R) \
(__m128)__builtin_ia32_scalefss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_scalef_ss (__m128 __A, __m128 __B)
{
return (__m128) __builtin_ia32_scalefss_round_mask ((__v4sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_scalef_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
return (__m128) __builtin_ia32_scalefss_round_mask ( (__v4sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask_scalef_round_ss(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_scalef_round_ss(W, U, A, B, R) \
(__m128)__builtin_ia32_scalefss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_scalef_ss (__mmask8 __U, __m128 __A, __m128 __B)
{
return (__m128) __builtin_ia32_scalefss_round_mask ( (__v4sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_maskz_scalef_round_ss(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_scalef_round_ss(U, A, B, R) \
(__m128)__builtin_ia32_scalefss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ (int)(R))
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_srai_epi32(__m512i __A, int __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_srai_epi32(__m512i __A, unsigned int __B)
{
return (__m512i)__builtin_ia32_psradi512((__v16si)__A, __B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_srai_epi32(__m512i __W, __mmask16 __U, __m512i __A, int __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_srai_epi32(__m512i __W, __mmask16 __U, __m512i __A,
+ unsigned int __B)
{
- return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U, \
- (__v16si)_mm512_srai_epi32(__A, __B), \
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
+ (__v16si)_mm512_srai_epi32(__A, __B),
(__v16si)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_srai_epi32(__mmask16 __U, __m512i __A, int __B) {
- return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U, \
- (__v16si)_mm512_srai_epi32(__A, __B), \
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_srai_epi32(__mmask16 __U, __m512i __A,
+ unsigned int __B) {
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
+ (__v16si)_mm512_srai_epi32(__A, __B),
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_srai_epi64(__m512i __A, int __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_srai_epi64(__m512i __A, unsigned int __B)
{
return (__m512i)__builtin_ia32_psraqi512((__v8di)__A, __B);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_srai_epi64(__m512i __W, __mmask8 __U, __m512i __A, int __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_srai_epi64(__m512i __W, __mmask8 __U, __m512i __A, unsigned int __B)
{
- return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U, \
- (__v8di)_mm512_srai_epi64(__A, __B), \
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
+ (__v8di)_mm512_srai_epi64(__A, __B),
(__v8di)__W);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_srai_epi64(__mmask8 __U, __m512i __A, int __B)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_srai_epi64(__mmask8 __U, __m512i __A, unsigned int __B)
{
- return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U, \
- (__v8di)_mm512_srai_epi64(__A, __B), \
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
+ (__v8di)_mm512_srai_epi64(__A, __B),
(__v8di)_mm512_setzero_si512());
}
-#define _mm512_shuffle_f32x4(A, B, imm) __extension__ ({ \
- (__m512)__builtin_ia32_shuf_f32x4_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), (int)(imm), \
- (__v16sf)_mm512_undefined_ps(), \
- (__mmask16)-1); })
+#define _mm512_shuffle_f32x4(A, B, imm) \
+ (__m512)__builtin_ia32_shuf_f32x4((__v16sf)(__m512)(A), \
+ (__v16sf)(__m512)(B), (int)(imm))
-#define _mm512_mask_shuffle_f32x4(W, U, A, B, imm) __extension__ ({ \
- (__m512)__builtin_ia32_shuf_f32x4_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), (int)(imm), \
- (__v16sf)(__m512)(W), \
- (__mmask16)(U)); })
+#define _mm512_mask_shuffle_f32x4(W, U, A, B, imm) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_shuffle_f32x4((A), (B), (imm)), \
+ (__v16sf)(__m512)(W))
-#define _mm512_maskz_shuffle_f32x4(U, A, B, imm) __extension__ ({ \
- (__m512)__builtin_ia32_shuf_f32x4_mask((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), (int)(imm), \
- (__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U)); })
+#define _mm512_maskz_shuffle_f32x4(U, A, B, imm) \
+ (__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
+ (__v16sf)_mm512_shuffle_f32x4((A), (B), (imm)), \
+ (__v16sf)_mm512_setzero_ps())
-#define _mm512_shuffle_f64x2(A, B, imm) __extension__ ({ \
- (__m512d)__builtin_ia32_shuf_f64x2_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), (int)(imm), \
- (__v8df)_mm512_undefined_pd(), \
- (__mmask8)-1); })
+#define _mm512_shuffle_f64x2(A, B, imm) \
+ (__m512d)__builtin_ia32_shuf_f64x2((__v8df)(__m512d)(A), \
+ (__v8df)(__m512d)(B), (int)(imm))
-#define _mm512_mask_shuffle_f64x2(W, U, A, B, imm) __extension__ ({ \
- (__m512d)__builtin_ia32_shuf_f64x2_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), (int)(imm), \
- (__v8df)(__m512d)(W), \
- (__mmask8)(U)); })
+#define _mm512_mask_shuffle_f64x2(W, U, A, B, imm) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_shuffle_f64x2((A), (B), (imm)), \
+ (__v8df)(__m512d)(W))
-#define _mm512_maskz_shuffle_f64x2(U, A, B, imm) __extension__ ({ \
- (__m512d)__builtin_ia32_shuf_f64x2_mask((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), (int)(imm), \
- (__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U)); })
-
-#define _mm512_shuffle_i32x4(A, B, imm) __extension__ ({ \
- (__m512i)__builtin_ia32_shuf_i32x4_mask((__v16si)(__m512i)(A), \
- (__v16si)(__m512i)(B), (int)(imm), \
- (__v16si)_mm512_setzero_si512(), \
- (__mmask16)-1); })
-
-#define _mm512_mask_shuffle_i32x4(W, U, A, B, imm) __extension__ ({ \
- (__m512i)__builtin_ia32_shuf_i32x4_mask((__v16si)(__m512i)(A), \
- (__v16si)(__m512i)(B), (int)(imm), \
- (__v16si)(__m512i)(W), \
- (__mmask16)(U)); })
-
-#define _mm512_maskz_shuffle_i32x4(U, A, B, imm) __extension__ ({ \
- (__m512i)__builtin_ia32_shuf_i32x4_mask((__v16si)(__m512i)(A), \
- (__v16si)(__m512i)(B), (int)(imm), \
- (__v16si)_mm512_setzero_si512(), \
- (__mmask16)(U)); })
-
-#define _mm512_shuffle_i64x2(A, B, imm) __extension__ ({ \
- (__m512i)__builtin_ia32_shuf_i64x2_mask((__v8di)(__m512i)(A), \
- (__v8di)(__m512i)(B), (int)(imm), \
- (__v8di)_mm512_setzero_si512(), \
- (__mmask8)-1); })
-
-#define _mm512_mask_shuffle_i64x2(W, U, A, B, imm) __extension__ ({ \
- (__m512i)__builtin_ia32_shuf_i64x2_mask((__v8di)(__m512i)(A), \
- (__v8di)(__m512i)(B), (int)(imm), \
- (__v8di)(__m512i)(W), \
- (__mmask8)(U)); })
-
-#define _mm512_maskz_shuffle_i64x2(U, A, B, imm) __extension__ ({ \
- (__m512i)__builtin_ia32_shuf_i64x2_mask((__v8di)(__m512i)(A), \
- (__v8di)(__m512i)(B), (int)(imm), \
- (__v8di)_mm512_setzero_si512(), \
- (__mmask8)(U)); })
-
-#define _mm512_shuffle_pd(A, B, M) __extension__ ({ \
- (__m512d)__builtin_shufflevector((__v8df)(__m512d)(A), \
- (__v8df)(__m512d)(B), \
- 0 + (((M) >> 0) & 0x1), \
- 8 + (((M) >> 1) & 0x1), \
- 2 + (((M) >> 2) & 0x1), \
- 10 + (((M) >> 3) & 0x1), \
- 4 + (((M) >> 4) & 0x1), \
- 12 + (((M) >> 5) & 0x1), \
- 6 + (((M) >> 6) & 0x1), \
- 14 + (((M) >> 7) & 0x1)); })
-
-#define _mm512_mask_shuffle_pd(W, U, A, B, M) __extension__ ({ \
+#define _mm512_maskz_shuffle_f64x2(U, A, B, imm) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_shuffle_f64x2((A), (B), (imm)), \
+ (__v8df)_mm512_setzero_pd())
+
+#define _mm512_shuffle_i32x4(A, B, imm) \
+ (__m512i)__builtin_ia32_shuf_i32x4((__v16si)(__m512i)(A), \
+ (__v16si)(__m512i)(B), (int)(imm))
+
+#define _mm512_mask_shuffle_i32x4(W, U, A, B, imm) \
+ (__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
+ (__v16si)_mm512_shuffle_i32x4((A), (B), (imm)), \
+ (__v16si)(__m512i)(W))
+
+#define _mm512_maskz_shuffle_i32x4(U, A, B, imm) \
+ (__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
+ (__v16si)_mm512_shuffle_i32x4((A), (B), (imm)), \
+ (__v16si)_mm512_setzero_si512())
+
+#define _mm512_shuffle_i64x2(A, B, imm) \
+ (__m512i)__builtin_ia32_shuf_i64x2((__v8di)(__m512i)(A), \
+ (__v8di)(__m512i)(B), (int)(imm))
+
+#define _mm512_mask_shuffle_i64x2(W, U, A, B, imm) \
+ (__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
+ (__v8di)_mm512_shuffle_i64x2((A), (B), (imm)), \
+ (__v8di)(__m512i)(W))
+
+#define _mm512_maskz_shuffle_i64x2(U, A, B, imm) \
+ (__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
+ (__v8di)_mm512_shuffle_i64x2((A), (B), (imm)), \
+ (__v8di)_mm512_setzero_si512())
+
+#define _mm512_shuffle_pd(A, B, M) \
+ (__m512d)__builtin_ia32_shufpd512((__v8df)(__m512d)(A), \
+ (__v8df)(__m512d)(B), (int)(M))
+
+#define _mm512_mask_shuffle_pd(W, U, A, B, M) \
(__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
(__v8df)_mm512_shuffle_pd((A), (B), (M)), \
- (__v8df)(__m512d)(W)); })
+ (__v8df)(__m512d)(W))
-#define _mm512_maskz_shuffle_pd(U, A, B, M) __extension__ ({ \
+#define _mm512_maskz_shuffle_pd(U, A, B, M) \
(__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
(__v8df)_mm512_shuffle_pd((A), (B), (M)), \
- (__v8df)_mm512_setzero_pd()); })
-
-#define _mm512_shuffle_ps(A, B, M) __extension__ ({ \
- (__m512d)__builtin_shufflevector((__v16sf)(__m512)(A), \
- (__v16sf)(__m512)(B), \
- 0 + (((M) >> 0) & 0x3), \
- 0 + (((M) >> 2) & 0x3), \
- 16 + (((M) >> 4) & 0x3), \
- 16 + (((M) >> 6) & 0x3), \
- 4 + (((M) >> 0) & 0x3), \
- 4 + (((M) >> 2) & 0x3), \
- 20 + (((M) >> 4) & 0x3), \
- 20 + (((M) >> 6) & 0x3), \
- 8 + (((M) >> 0) & 0x3), \
- 8 + (((M) >> 2) & 0x3), \
- 24 + (((M) >> 4) & 0x3), \
- 24 + (((M) >> 6) & 0x3), \
- 12 + (((M) >> 0) & 0x3), \
- 12 + (((M) >> 2) & 0x3), \
- 28 + (((M) >> 4) & 0x3), \
- 28 + (((M) >> 6) & 0x3)); })
-
-#define _mm512_mask_shuffle_ps(W, U, A, B, M) __extension__ ({ \
+ (__v8df)_mm512_setzero_pd())
+
+#define _mm512_shuffle_ps(A, B, M) \
+ (__m512)__builtin_ia32_shufps512((__v16sf)(__m512)(A), \
+ (__v16sf)(__m512)(B), (int)(M))
+
+#define _mm512_mask_shuffle_ps(W, U, A, B, M) \
(__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
(__v16sf)_mm512_shuffle_ps((A), (B), (M)), \
- (__v16sf)(__m512)(W)); })
+ (__v16sf)(__m512)(W))
-#define _mm512_maskz_shuffle_ps(U, A, B, M) __extension__ ({ \
+#define _mm512_maskz_shuffle_ps(U, A, B, M) \
(__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
(__v16sf)_mm512_shuffle_ps((A), (B), (M)), \
- (__v16sf)_mm512_setzero_ps()); })
+ (__v16sf)_mm512_setzero_ps())
-#define _mm_sqrt_round_sd(A, B, R) __extension__ ({ \
+#define _mm_sqrt_round_sd(A, B, R) \
(__m128d)__builtin_ia32_sqrtsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_sqrt_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
return (__m128d) __builtin_ia32_sqrtsd_round_mask ( (__v2df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask_sqrt_round_sd(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_sqrt_round_sd(W, U, A, B, R) \
(__m128d)__builtin_ia32_sqrtsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_sqrt_sd (__mmask8 __U, __m128d __A, __m128d __B)
{
return (__m128d) __builtin_ia32_sqrtsd_round_mask ( (__v2df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_maskz_sqrt_round_sd(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_sqrt_round_sd(U, A, B, R) \
(__m128d)__builtin_ia32_sqrtsd_round_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_sqrt_round_ss(A, B, R) __extension__ ({ \
+#define _mm_sqrt_round_ss(A, B, R) \
(__m128)__builtin_ia32_sqrtss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_sqrt_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
return (__m128) __builtin_ia32_sqrtss_round_mask ( (__v4sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask_sqrt_round_ss(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_sqrt_round_ss(W, U, A, B, R) \
(__m128)__builtin_ia32_sqrtss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(W), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_sqrt_ss (__mmask8 __U, __m128 __A, __m128 __B)
{
return (__m128) __builtin_ia32_sqrtss_round_mask ( (__v4sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm_maskz_sqrt_round_ss(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_sqrt_round_ss(U, A, B, R) \
(__m128)__builtin_ia32_sqrtss_round_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_broadcast_f32x4(__m128 __A)
{
return (__m512)__builtin_shufflevector((__v4sf)__A, (__v4sf)__A,
0, 1, 2, 3, 0, 1, 2, 3);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_broadcast_f32x4(__m512 __O, __mmask16 __M, __m128 __A)
{
return (__m512)__builtin_ia32_selectps_512((__mmask16)__M,
(__v16sf)__O);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcast_f32x4(__mmask16 __M, __m128 __A)
{
return (__m512)__builtin_ia32_selectps_512((__mmask16)__M,
(__v16sf)_mm512_setzero_ps());
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_broadcast_f64x4(__m256d __A)
{
return (__m512d)__builtin_shufflevector((__v4df)__A, (__v4df)__A,
0, 1, 2, 3, 0, 1, 2, 3);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_broadcast_f64x4(__m512d __O, __mmask8 __M, __m256d __A)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__M,
(__v8df)__O);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcast_f64x4(__mmask8 __M, __m256d __A)
{
return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__M,
(__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_broadcast_i32x4(__m128i __A)
{
return (__m512i)__builtin_shufflevector((__v4si)__A, (__v4si)__A,
0, 1, 2, 3, 0, 1, 2, 3);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_broadcast_i32x4(__m512i __O, __mmask16 __M, __m128i __A)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
(__v16si)__O);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcast_i32x4(__mmask16 __M, __m128i __A)
{
return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
(__v16si)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_broadcast_i64x4(__m256i __A)
{
return (__m512i)__builtin_shufflevector((__v4di)__A, (__v4di)__A,
0, 1, 2, 3, 0, 1, 2, 3);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_broadcast_i64x4(__m512i __O, __mmask8 __M, __m256i __A)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
(__v8di)__O);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcast_i64x4(__mmask8 __M, __m256i __A)
{
return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
(__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_broadcastsd_pd (__m512d __O, __mmask8 __M, __m128d __A)
{
return (__m512d)__builtin_ia32_selectpd_512(__M,
(__v8df) __O);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcastsd_pd (__mmask8 __M, __m128d __A)
{
return (__m512d)__builtin_ia32_selectpd_512(__M,
(__v8df) _mm512_setzero_pd());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_broadcastss_ps (__m512 __O, __mmask16 __M, __m128 __A)
{
return (__m512)__builtin_ia32_selectps_512(__M,
(__v16sf) __O);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcastss_ps (__mmask16 __M, __m128 __A)
{
return (__m512)__builtin_ia32_selectps_512(__M,
(__v16sf) _mm512_setzero_ps());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtsepi32_epi8 (__m512i __A)
{
return (__m128i) __builtin_ia32_pmovsdb512_mask ((__v16si) __A,
(__mmask16) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi32_epi8 (__m128i __O, __mmask16 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovsdb512_mask ((__v16si) __A,
(__v16qi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtsepi32_epi8 (__mmask16 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovsdb512_mask ((__v16si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi32_storeu_epi8 (void * __P, __mmask16 __M, __m512i __A)
{
__builtin_ia32_pmovsdb512mem_mask ((__v16qi *) __P, (__v16si) __A, __M);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtsepi32_epi16 (__m512i __A)
{
return (__m256i) __builtin_ia32_pmovsdw512_mask ((__v16si) __A,
(__mmask16) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi32_epi16 (__m256i __O, __mmask16 __M, __m512i __A)
{
return (__m256i) __builtin_ia32_pmovsdw512_mask ((__v16si) __A,
(__v16hi) __O, __M);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtsepi32_epi16 (__mmask16 __M, __m512i __A)
{
return (__m256i) __builtin_ia32_pmovsdw512_mask ((__v16si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi32_storeu_epi16 (void *__P, __mmask16 __M, __m512i __A)
{
__builtin_ia32_pmovsdw512mem_mask ((__v16hi*) __P, (__v16si) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtsepi64_epi8 (__m512i __A)
{
return (__m128i) __builtin_ia32_pmovsqb512_mask ((__v8di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi64_epi8 (__m128i __O, __mmask8 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovsqb512_mask ((__v8di) __A,
(__v16qi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtsepi64_epi8 (__mmask8 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovsqb512_mask ((__v8di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi64_storeu_epi8 (void * __P, __mmask8 __M, __m512i __A)
{
__builtin_ia32_pmovsqb512mem_mask ((__v16qi *) __P, (__v8di) __A, __M);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtsepi64_epi32 (__m512i __A)
{
return (__m256i) __builtin_ia32_pmovsqd512_mask ((__v8di) __A,
(__mmask8) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi64_epi32 (__m256i __O, __mmask8 __M, __m512i __A)
{
return (__m256i) __builtin_ia32_pmovsqd512_mask ((__v8di) __A,
(__v8si) __O, __M);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtsepi64_epi32 (__mmask8 __M, __m512i __A)
{
return (__m256i) __builtin_ia32_pmovsqd512_mask ((__v8di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi64_storeu_epi32 (void *__P, __mmask8 __M, __m512i __A)
{
__builtin_ia32_pmovsqd512mem_mask ((__v8si *) __P, (__v8di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtsepi64_epi16 (__m512i __A)
{
return (__m128i) __builtin_ia32_pmovsqw512_mask ((__v8di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi64_epi16 (__m128i __O, __mmask8 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovsqw512_mask ((__v8di) __A,
(__v8hi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtsepi64_epi16 (__mmask8 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovsqw512_mask ((__v8di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi64_storeu_epi16 (void * __P, __mmask8 __M, __m512i __A)
{
__builtin_ia32_pmovsqw512mem_mask ((__v8hi *) __P, (__v8di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtusepi32_epi8 (__m512i __A)
{
return (__m128i) __builtin_ia32_pmovusdb512_mask ((__v16si) __A,
(__mmask16) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi32_epi8 (__m128i __O, __mmask16 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovusdb512_mask ((__v16si) __A,
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtusepi32_epi8 (__mmask16 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovusdb512_mask ((__v16si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi32_storeu_epi8 (void * __P, __mmask16 __M, __m512i __A)
{
__builtin_ia32_pmovusdb512mem_mask ((__v16qi *) __P, (__v16si) __A, __M);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtusepi32_epi16 (__m512i __A)
{
return (__m256i) __builtin_ia32_pmovusdw512_mask ((__v16si) __A,
(__mmask16) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi32_epi16 (__m256i __O, __mmask16 __M, __m512i __A)
{
return (__m256i) __builtin_ia32_pmovusdw512_mask ((__v16si) __A,
__M);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtusepi32_epi16 (__mmask16 __M, __m512i __A)
{
return (__m256i) __builtin_ia32_pmovusdw512_mask ((__v16si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi32_storeu_epi16 (void *__P, __mmask16 __M, __m512i __A)
{
__builtin_ia32_pmovusdw512mem_mask ((__v16hi*) __P, (__v16si) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtusepi64_epi8 (__m512i __A)
{
return (__m128i) __builtin_ia32_pmovusqb512_mask ((__v8di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi64_epi8 (__m128i __O, __mmask8 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovusqb512_mask ((__v8di) __A,
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtusepi64_epi8 (__mmask8 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovusqb512_mask ((__v8di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi64_storeu_epi8 (void * __P, __mmask8 __M, __m512i __A)
{
__builtin_ia32_pmovusqb512mem_mask ((__v16qi *) __P, (__v8di) __A, __M);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtusepi64_epi32 (__m512i __A)
{
return (__m256i) __builtin_ia32_pmovusqd512_mask ((__v8di) __A,
(__mmask8) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi64_epi32 (__m256i __O, __mmask8 __M, __m512i __A)
{
return (__m256i) __builtin_ia32_pmovusqd512_mask ((__v8di) __A,
(__v8si) __O, __M);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtusepi64_epi32 (__mmask8 __M, __m512i __A)
{
return (__m256i) __builtin_ia32_pmovusqd512_mask ((__v8di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi64_storeu_epi32 (void* __P, __mmask8 __M, __m512i __A)
{
__builtin_ia32_pmovusqd512mem_mask ((__v8si*) __P, (__v8di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtusepi64_epi16 (__m512i __A)
{
return (__m128i) __builtin_ia32_pmovusqw512_mask ((__v8di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi64_epi16 (__m128i __O, __mmask8 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovusqw512_mask ((__v8di) __A,
(__v8hi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtusepi64_epi16 (__mmask8 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovusqw512_mask ((__v8di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi64_storeu_epi16 (void *__P, __mmask8 __M, __m512i __A)
{
__builtin_ia32_pmovusqw512mem_mask ((__v8hi*) __P, (__v8di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtepi32_epi8 (__m512i __A)
{
return (__m128i) __builtin_ia32_pmovdb512_mask ((__v16si) __A,
(__mmask16) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32_epi8 (__m128i __O, __mmask16 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovdb512_mask ((__v16si) __A,
(__v16qi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi32_epi8 (__mmask16 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovdb512_mask ((__v16si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32_storeu_epi8 (void * __P, __mmask16 __M, __m512i __A)
{
__builtin_ia32_pmovdb512mem_mask ((__v16qi *) __P, (__v16si) __A, __M);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtepi32_epi16 (__m512i __A)
{
return (__m256i) __builtin_ia32_pmovdw512_mask ((__v16si) __A,
(__mmask16) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32_epi16 (__m256i __O, __mmask16 __M, __m512i __A)
{
return (__m256i) __builtin_ia32_pmovdw512_mask ((__v16si) __A,
(__v16hi) __O, __M);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi32_epi16 (__mmask16 __M, __m512i __A)
{
return (__m256i) __builtin_ia32_pmovdw512_mask ((__v16si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32_storeu_epi16 (void * __P, __mmask16 __M, __m512i __A)
{
__builtin_ia32_pmovdw512mem_mask ((__v16hi *) __P, (__v16si) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtepi64_epi8 (__m512i __A)
{
return (__m128i) __builtin_ia32_pmovqb512_mask ((__v8di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi64_epi8 (__m128i __O, __mmask8 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovqb512_mask ((__v8di) __A,
(__v16qi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi64_epi8 (__mmask8 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovqb512_mask ((__v8di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi64_storeu_epi8 (void * __P, __mmask8 __M, __m512i __A)
{
__builtin_ia32_pmovqb512mem_mask ((__v16qi *) __P, (__v8di) __A, __M);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtepi64_epi32 (__m512i __A)
{
return (__m256i) __builtin_ia32_pmovqd512_mask ((__v8di) __A,
(__mmask8) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi64_epi32 (__m256i __O, __mmask8 __M, __m512i __A)
{
return (__m256i) __builtin_ia32_pmovqd512_mask ((__v8di) __A,
(__v8si) __O, __M);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi64_epi32 (__mmask8 __M, __m512i __A)
{
return (__m256i) __builtin_ia32_pmovqd512_mask ((__v8di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi64_storeu_epi32 (void* __P, __mmask8 __M, __m512i __A)
{
__builtin_ia32_pmovqd512mem_mask ((__v8si *) __P, (__v8di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtepi64_epi16 (__m512i __A)
{
return (__m128i) __builtin_ia32_pmovqw512_mask ((__v8di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi64_epi16 (__m128i __O, __mmask8 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovqw512_mask ((__v8di) __A,
(__v8hi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi64_epi16 (__mmask8 __M, __m512i __A)
{
return (__m128i) __builtin_ia32_pmovqw512_mask ((__v8di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi64_storeu_epi16 (void *__P, __mmask8 __M, __m512i __A)
{
__builtin_ia32_pmovqw512mem_mask ((__v8hi *) __P, (__v8di) __A, __M);
}
-#define _mm512_extracti32x4_epi32(A, imm) __extension__ ({ \
- (__m128i)__builtin_shufflevector((__v16si)(__m512i)(A), \
- (__v16si)_mm512_undefined_epi32(), \
- 0 + ((imm) & 0x3) * 4, \
- 1 + ((imm) & 0x3) * 4, \
- 2 + ((imm) & 0x3) * 4, \
- 3 + ((imm) & 0x3) * 4); })
-
-#define _mm512_mask_extracti32x4_epi32(W, U, A, imm) __extension__ ({ \
- (__m128i)__builtin_ia32_selectd_128((__mmask8)(U), \
- (__v4si)_mm512_extracti32x4_epi32((A), (imm)), \
- (__v4si)(W)); })
-
-#define _mm512_maskz_extracti32x4_epi32(U, A, imm) __extension__ ({ \
- (__m128i)__builtin_ia32_selectd_128((__mmask8)(U), \
- (__v4si)_mm512_extracti32x4_epi32((A), (imm)), \
- (__v4si)_mm_setzero_si128()); })
-
-#define _mm512_extracti64x4_epi64(A, imm) __extension__ ({ \
- (__m256i)__builtin_shufflevector((__v8di)(__m512i)(A), \
- (__v8di)_mm512_undefined_epi32(), \
- ((imm) & 1) ? 4 : 0, \
- ((imm) & 1) ? 5 : 1, \
- ((imm) & 1) ? 6 : 2, \
- ((imm) & 1) ? 7 : 3); })
-
-#define _mm512_mask_extracti64x4_epi64(W, U, A, imm) __extension__ ({ \
- (__m256i)__builtin_ia32_selectq_256((__mmask8)(U), \
- (__v4di)_mm512_extracti64x4_epi64((A), (imm)), \
- (__v4di)(W)); })
-
-#define _mm512_maskz_extracti64x4_epi64(U, A, imm) __extension__ ({ \
- (__m256i)__builtin_ia32_selectq_256((__mmask8)(U), \
- (__v4di)_mm512_extracti64x4_epi64((A), (imm)), \
- (__v4di)_mm256_setzero_si256()); })
-
-#define _mm512_insertf64x4(A, B, imm) __extension__ ({ \
- (__m512d)__builtin_shufflevector((__v8df)(__m512d)(A), \
- (__v8df)_mm512_castpd256_pd512((__m256d)(B)), \
- ((imm) & 0x1) ? 0 : 8, \
- ((imm) & 0x1) ? 1 : 9, \
- ((imm) & 0x1) ? 2 : 10, \
- ((imm) & 0x1) ? 3 : 11, \
- ((imm) & 0x1) ? 8 : 4, \
- ((imm) & 0x1) ? 9 : 5, \
- ((imm) & 0x1) ? 10 : 6, \
- ((imm) & 0x1) ? 11 : 7); })
-
-#define _mm512_mask_insertf64x4(W, U, A, B, imm) __extension__ ({ \
- (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
- (__v8df)_mm512_insertf64x4((A), (B), (imm)), \
- (__v8df)(W)); })
+#define _mm512_extracti32x4_epi32(A, imm) \
+ (__m128i)__builtin_ia32_extracti32x4_mask((__v16si)(__m512i)(A), (int)(imm), \
+ (__v4si)_mm_undefined_si128(), \
+ (__mmask8)-1)
-#define _mm512_maskz_insertf64x4(U, A, B, imm) __extension__ ({ \
- (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
- (__v8df)_mm512_insertf64x4((A), (B), (imm)), \
- (__v8df)_mm512_setzero_pd()); })
-
-#define _mm512_inserti64x4(A, B, imm) __extension__ ({ \
- (__m512i)__builtin_shufflevector((__v8di)(__m512i)(A), \
- (__v8di)_mm512_castsi256_si512((__m256i)(B)), \
- ((imm) & 0x1) ? 0 : 8, \
- ((imm) & 0x1) ? 1 : 9, \
- ((imm) & 0x1) ? 2 : 10, \
- ((imm) & 0x1) ? 3 : 11, \
- ((imm) & 0x1) ? 8 : 4, \
- ((imm) & 0x1) ? 9 : 5, \
- ((imm) & 0x1) ? 10 : 6, \
- ((imm) & 0x1) ? 11 : 7); })
-
-#define _mm512_mask_inserti64x4(W, U, A, B, imm) __extension__ ({ \
+#define _mm512_mask_extracti32x4_epi32(W, U, A, imm) \
+ (__m128i)__builtin_ia32_extracti32x4_mask((__v16si)(__m512i)(A), (int)(imm), \
+ (__v4si)(__m128i)(W), \
+ (__mmask8)(U))
+
+#define _mm512_maskz_extracti32x4_epi32(U, A, imm) \
+ (__m128i)__builtin_ia32_extracti32x4_mask((__v16si)(__m512i)(A), (int)(imm), \
+ (__v4si)_mm_setzero_si128(), \
+ (__mmask8)(U))
+
+#define _mm512_extracti64x4_epi64(A, imm) \
+ (__m256i)__builtin_ia32_extracti64x4_mask((__v8di)(__m512i)(A), (int)(imm), \
+ (__v4di)_mm256_undefined_si256(), \
+ (__mmask8)-1)
+
+#define _mm512_mask_extracti64x4_epi64(W, U, A, imm) \
+ (__m256i)__builtin_ia32_extracti64x4_mask((__v8di)(__m512i)(A), (int)(imm), \
+ (__v4di)(__m256i)(W), \
+ (__mmask8)(U))
+
+#define _mm512_maskz_extracti64x4_epi64(U, A, imm) \
+ (__m256i)__builtin_ia32_extracti64x4_mask((__v8di)(__m512i)(A), (int)(imm), \
+ (__v4di)_mm256_setzero_si256(), \
+ (__mmask8)(U))
+
+#define _mm512_insertf64x4(A, B, imm) \
+ (__m512d)__builtin_ia32_insertf64x4((__v8df)(__m512d)(A), \
+ (__v4df)(__m256d)(B), (int)(imm))
+
+#define _mm512_mask_insertf64x4(W, U, A, B, imm) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_insertf64x4((A), (B), (imm)), \
+ (__v8df)(__m512d)(W))
+
+#define _mm512_maskz_insertf64x4(U, A, B, imm) \
+ (__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
+ (__v8df)_mm512_insertf64x4((A), (B), (imm)), \
+ (__v8df)_mm512_setzero_pd())
+
+#define _mm512_inserti64x4(A, B, imm) \
+ (__m512i)__builtin_ia32_inserti64x4((__v8di)(__m512i)(A), \
+ (__v4di)(__m256i)(B), (int)(imm))
+
+#define _mm512_mask_inserti64x4(W, U, A, B, imm) \
(__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
(__v8di)_mm512_inserti64x4((A), (B), (imm)), \
- (__v8di)(W)); })
+ (__v8di)(__m512i)(W))
-#define _mm512_maskz_inserti64x4(U, A, B, imm) __extension__ ({ \
+#define _mm512_maskz_inserti64x4(U, A, B, imm) \
(__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
(__v8di)_mm512_inserti64x4((A), (B), (imm)), \
- (__v8di)_mm512_setzero_si512()); })
-
-#define _mm512_insertf32x4(A, B, imm) __extension__ ({ \
- (__m512)__builtin_shufflevector((__v16sf)(__m512)(A), \
- (__v16sf)_mm512_castps128_ps512((__m128)(B)),\
- (((imm) & 0x3) == 0) ? 16 : 0, \
- (((imm) & 0x3) == 0) ? 17 : 1, \
- (((imm) & 0x3) == 0) ? 18 : 2, \
- (((imm) & 0x3) == 0) ? 19 : 3, \
- (((imm) & 0x3) == 1) ? 16 : 4, \
- (((imm) & 0x3) == 1) ? 17 : 5, \
- (((imm) & 0x3) == 1) ? 18 : 6, \
- (((imm) & 0x3) == 1) ? 19 : 7, \
- (((imm) & 0x3) == 2) ? 16 : 8, \
- (((imm) & 0x3) == 2) ? 17 : 9, \
- (((imm) & 0x3) == 2) ? 18 : 10, \
- (((imm) & 0x3) == 2) ? 19 : 11, \
- (((imm) & 0x3) == 3) ? 16 : 12, \
- (((imm) & 0x3) == 3) ? 17 : 13, \
- (((imm) & 0x3) == 3) ? 18 : 14, \
- (((imm) & 0x3) == 3) ? 19 : 15); })
-
-#define _mm512_mask_insertf32x4(W, U, A, B, imm) __extension__ ({ \
+ (__v8di)_mm512_setzero_si512())
+
+#define _mm512_insertf32x4(A, B, imm) \
+ (__m512)__builtin_ia32_insertf32x4((__v16sf)(__m512)(A), \
+ (__v4sf)(__m128)(B), (int)(imm))
+
+#define _mm512_mask_insertf32x4(W, U, A, B, imm) \
(__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
(__v16sf)_mm512_insertf32x4((A), (B), (imm)), \
- (__v16sf)(W)); })
+ (__v16sf)(__m512)(W))
-#define _mm512_maskz_insertf32x4(U, A, B, imm) __extension__ ({ \
+#define _mm512_maskz_insertf32x4(U, A, B, imm) \
(__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
(__v16sf)_mm512_insertf32x4((A), (B), (imm)), \
- (__v16sf)_mm512_setzero_ps()); })
-
-#define _mm512_inserti32x4(A, B, imm) __extension__ ({ \
- (__m512i)__builtin_shufflevector((__v16si)(__m512i)(A), \
- (__v16si)_mm512_castsi128_si512((__m128i)(B)),\
- (((imm) & 0x3) == 0) ? 16 : 0, \
- (((imm) & 0x3) == 0) ? 17 : 1, \
- (((imm) & 0x3) == 0) ? 18 : 2, \
- (((imm) & 0x3) == 0) ? 19 : 3, \
- (((imm) & 0x3) == 1) ? 16 : 4, \
- (((imm) & 0x3) == 1) ? 17 : 5, \
- (((imm) & 0x3) == 1) ? 18 : 6, \
- (((imm) & 0x3) == 1) ? 19 : 7, \
- (((imm) & 0x3) == 2) ? 16 : 8, \
- (((imm) & 0x3) == 2) ? 17 : 9, \
- (((imm) & 0x3) == 2) ? 18 : 10, \
- (((imm) & 0x3) == 2) ? 19 : 11, \
- (((imm) & 0x3) == 3) ? 16 : 12, \
- (((imm) & 0x3) == 3) ? 17 : 13, \
- (((imm) & 0x3) == 3) ? 18 : 14, \
- (((imm) & 0x3) == 3) ? 19 : 15); })
-
-#define _mm512_mask_inserti32x4(W, U, A, B, imm) __extension__ ({ \
+ (__v16sf)_mm512_setzero_ps())
+
+#define _mm512_inserti32x4(A, B, imm) \
+ (__m512i)__builtin_ia32_inserti32x4((__v16si)(__m512i)(A), \
+ (__v4si)(__m128i)(B), (int)(imm))
+
+#define _mm512_mask_inserti32x4(W, U, A, B, imm) \
(__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
(__v16si)_mm512_inserti32x4((A), (B), (imm)), \
- (__v16si)(W)); })
+ (__v16si)(__m512i)(W))
-#define _mm512_maskz_inserti32x4(U, A, B, imm) __extension__ ({ \
+#define _mm512_maskz_inserti32x4(U, A, B, imm) \
(__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
(__v16si)_mm512_inserti32x4((A), (B), (imm)), \
- (__v16si)_mm512_setzero_si512()); })
+ (__v16si)_mm512_setzero_si512())
-#define _mm512_getmant_round_pd(A, B, C, R) __extension__ ({ \
+#define _mm512_getmant_round_pd(A, B, C, R) \
(__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
(int)(((C)<<2) | (B)), \
(__v8df)_mm512_undefined_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_getmant_round_pd(W, U, A, B, C, R) __extension__ ({ \
+#define _mm512_mask_getmant_round_pd(W, U, A, B, C, R) \
(__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
(int)(((C)<<2) | (B)), \
(__v8df)(__m512d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_getmant_round_pd(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_getmant_round_pd(U, A, B, C, R) \
(__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
(int)(((C)<<2) | (B)), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_getmant_pd(A, B, C) __extension__ ({ \
+#define _mm512_getmant_pd(A, B, C) \
(__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
(int)(((C)<<2) | (B)), \
(__v8df)_mm512_setzero_pd(), \
(__mmask8)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_mask_getmant_pd(W, U, A, B, C) __extension__ ({ \
+#define _mm512_mask_getmant_pd(W, U, A, B, C) \
(__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
(int)(((C)<<2) | (B)), \
(__v8df)(__m512d)(W), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_maskz_getmant_pd(U, A, B, C) __extension__ ({ \
+#define _mm512_maskz_getmant_pd(U, A, B, C) \
(__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
(int)(((C)<<2) | (B)), \
(__v8df)_mm512_setzero_pd(), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_getmant_round_ps(A, B, C, R) __extension__ ({ \
+#define _mm512_getmant_round_ps(A, B, C, R) \
(__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
(int)(((C)<<2) | (B)), \
(__v16sf)_mm512_undefined_ps(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_getmant_round_ps(W, U, A, B, C, R) __extension__ ({ \
+#define _mm512_mask_getmant_round_ps(W, U, A, B, C, R) \
(__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
(int)(((C)<<2) | (B)), \
(__v16sf)(__m512)(W), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_getmant_round_ps(U, A, B, C, R) __extension__ ({ \
+#define _mm512_maskz_getmant_round_ps(U, A, B, C, R) \
(__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
(int)(((C)<<2) | (B)), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_getmant_ps(A, B, C) __extension__ ({ \
+#define _mm512_getmant_ps(A, B, C) \
(__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
(int)(((C)<<2)|(B)), \
(__v16sf)_mm512_undefined_ps(), \
(__mmask16)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_mask_getmant_ps(W, U, A, B, C) __extension__ ({ \
+#define _mm512_mask_getmant_ps(W, U, A, B, C) \
(__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
(int)(((C)<<2)|(B)), \
(__v16sf)(__m512)(W), \
(__mmask16)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_maskz_getmant_ps(U, A, B, C) __extension__ ({ \
+#define _mm512_maskz_getmant_ps(U, A, B, C) \
(__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
(int)(((C)<<2)|(B)), \
(__v16sf)_mm512_setzero_ps(), \
(__mmask16)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm512_getexp_round_pd(A, R) __extension__ ({ \
+#define _mm512_getexp_round_pd(A, R) \
(__m512d)__builtin_ia32_getexppd512_mask((__v8df)(__m512d)(A), \
(__v8df)_mm512_undefined_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_getexp_round_pd(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_getexp_round_pd(W, U, A, R) \
(__m512d)__builtin_ia32_getexppd512_mask((__v8df)(__m512d)(A), \
(__v8df)(__m512d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_getexp_round_pd(U, A, R) __extension__ ({ \
+#define _mm512_maskz_getexp_round_pd(U, A, R) \
(__m512d)__builtin_ia32_getexppd512_mask((__v8df)(__m512d)(A), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_getexp_pd (__m512d __A)
{
return (__m512d) __builtin_ia32_getexppd512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_getexp_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
return (__m512d) __builtin_ia32_getexppd512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_getexp_pd (__mmask8 __U, __m512d __A)
{
return (__m512d) __builtin_ia32_getexppd512_mask ((__v8df) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_getexp_round_ps(A, R) __extension__ ({ \
+#define _mm512_getexp_round_ps(A, R) \
(__m512)__builtin_ia32_getexpps512_mask((__v16sf)(__m512)(A), \
(__v16sf)_mm512_undefined_ps(), \
- (__mmask16)-1, (int)(R)); })
+ (__mmask16)-1, (int)(R))
-#define _mm512_mask_getexp_round_ps(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_getexp_round_ps(W, U, A, R) \
(__m512)__builtin_ia32_getexpps512_mask((__v16sf)(__m512)(A), \
(__v16sf)(__m512)(W), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-#define _mm512_maskz_getexp_round_ps(U, A, R) __extension__ ({ \
+#define _mm512_maskz_getexp_round_ps(U, A, R) \
(__m512)__builtin_ia32_getexpps512_mask((__v16sf)(__m512)(A), \
(__v16sf)_mm512_setzero_ps(), \
- (__mmask16)(U), (int)(R)); })
+ (__mmask16)(U), (int)(R))
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_getexp_ps (__m512 __A)
{
return (__m512) __builtin_ia32_getexpps512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_getexp_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
return (__m512) __builtin_ia32_getexpps512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_getexp_ps (__mmask16 __U, __m512 __A)
{
return (__m512) __builtin_ia32_getexpps512_mask ((__v16sf) __A,
_MM_FROUND_CUR_DIRECTION);
}
-#define _mm512_i64gather_ps(index, addr, scale) __extension__ ({ \
+#define _mm512_i64gather_ps(index, addr, scale) \
(__m256)__builtin_ia32_gatherdiv16sf((__v8sf)_mm256_undefined_ps(), \
- (float const *)(addr), \
+ (void const *)(addr), \
(__v8di)(__m512i)(index), (__mmask8)-1, \
- (int)(scale)); })
+ (int)(scale))
-#define _mm512_mask_i64gather_ps(v1_old, mask, index, addr, scale) __extension__({\
+#define _mm512_mask_i64gather_ps(v1_old, mask, index, addr, scale) \
(__m256)__builtin_ia32_gatherdiv16sf((__v8sf)(__m256)(v1_old),\
- (float const *)(addr), \
+ (void const *)(addr), \
(__v8di)(__m512i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm512_i64gather_epi32(index, addr, scale) __extension__ ({\
- (__m256i)__builtin_ia32_gatherdiv16si((__v8si)_mm256_undefined_ps(), \
- (int const *)(addr), \
+#define _mm512_i64gather_epi32(index, addr, scale) \
+ (__m256i)__builtin_ia32_gatherdiv16si((__v8si)_mm256_undefined_si256(), \
+ (void const *)(addr), \
(__v8di)(__m512i)(index), \
- (__mmask8)-1, (int)(scale)); })
+ (__mmask8)-1, (int)(scale))
-#define _mm512_mask_i64gather_epi32(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm512_mask_i64gather_epi32(v1_old, mask, index, addr, scale) \
(__m256i)__builtin_ia32_gatherdiv16si((__v8si)(__m256i)(v1_old), \
- (int const *)(addr), \
+ (void const *)(addr), \
(__v8di)(__m512i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm512_i64gather_pd(index, addr, scale) __extension__ ({\
+#define _mm512_i64gather_pd(index, addr, scale) \
(__m512d)__builtin_ia32_gatherdiv8df((__v8df)_mm512_undefined_pd(), \
- (double const *)(addr), \
+ (void const *)(addr), \
(__v8di)(__m512i)(index), (__mmask8)-1, \
- (int)(scale)); })
+ (int)(scale))
-#define _mm512_mask_i64gather_pd(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm512_mask_i64gather_pd(v1_old, mask, index, addr, scale) \
(__m512d)__builtin_ia32_gatherdiv8df((__v8df)(__m512d)(v1_old), \
- (double const *)(addr), \
+ (void const *)(addr), \
(__v8di)(__m512i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm512_i64gather_epi64(index, addr, scale) __extension__ ({\
- (__m512i)__builtin_ia32_gatherdiv8di((__v8di)_mm512_undefined_pd(), \
- (long long const *)(addr), \
+#define _mm512_i64gather_epi64(index, addr, scale) \
+ (__m512i)__builtin_ia32_gatherdiv8di((__v8di)_mm512_undefined_epi32(), \
+ (void const *)(addr), \
(__v8di)(__m512i)(index), (__mmask8)-1, \
- (int)(scale)); })
+ (int)(scale))
-#define _mm512_mask_i64gather_epi64(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm512_mask_i64gather_epi64(v1_old, mask, index, addr, scale) \
(__m512i)__builtin_ia32_gatherdiv8di((__v8di)(__m512i)(v1_old), \
- (long long const *)(addr), \
+ (void const *)(addr), \
(__v8di)(__m512i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm512_i32gather_ps(index, addr, scale) __extension__ ({\
+#define _mm512_i32gather_ps(index, addr, scale) \
(__m512)__builtin_ia32_gathersiv16sf((__v16sf)_mm512_undefined_ps(), \
- (float const *)(addr), \
- (__v16sf)(__m512)(index), \
- (__mmask16)-1, (int)(scale)); })
+ (void const *)(addr), \
+ (__v16si)(__m512)(index), \
+ (__mmask16)-1, (int)(scale))
-#define _mm512_mask_i32gather_ps(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm512_mask_i32gather_ps(v1_old, mask, index, addr, scale) \
(__m512)__builtin_ia32_gathersiv16sf((__v16sf)(__m512)(v1_old), \
- (float const *)(addr), \
- (__v16sf)(__m512)(index), \
- (__mmask16)(mask), (int)(scale)); })
+ (void const *)(addr), \
+ (__v16si)(__m512)(index), \
+ (__mmask16)(mask), (int)(scale))
-#define _mm512_i32gather_epi32(index, addr, scale) __extension__ ({\
+#define _mm512_i32gather_epi32(index, addr, scale) \
(__m512i)__builtin_ia32_gathersiv16si((__v16si)_mm512_undefined_epi32(), \
- (int const *)(addr), \
+ (void const *)(addr), \
(__v16si)(__m512i)(index), \
- (__mmask16)-1, (int)(scale)); })
+ (__mmask16)-1, (int)(scale))
-#define _mm512_mask_i32gather_epi32(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm512_mask_i32gather_epi32(v1_old, mask, index, addr, scale) \
(__m512i)__builtin_ia32_gathersiv16si((__v16si)(__m512i)(v1_old), \
- (int const *)(addr), \
+ (void const *)(addr), \
(__v16si)(__m512i)(index), \
- (__mmask16)(mask), (int)(scale)); })
+ (__mmask16)(mask), (int)(scale))
-#define _mm512_i32gather_pd(index, addr, scale) __extension__ ({\
+#define _mm512_i32gather_pd(index, addr, scale) \
(__m512d)__builtin_ia32_gathersiv8df((__v8df)_mm512_undefined_pd(), \
- (double const *)(addr), \
+ (void const *)(addr), \
(__v8si)(__m256i)(index), (__mmask8)-1, \
- (int)(scale)); })
+ (int)(scale))
-#define _mm512_mask_i32gather_pd(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm512_mask_i32gather_pd(v1_old, mask, index, addr, scale) \
(__m512d)__builtin_ia32_gathersiv8df((__v8df)(__m512d)(v1_old), \
- (double const *)(addr), \
+ (void const *)(addr), \
(__v8si)(__m256i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm512_i32gather_epi64(index, addr, scale) __extension__ ({\
+#define _mm512_i32gather_epi64(index, addr, scale) \
(__m512i)__builtin_ia32_gathersiv8di((__v8di)_mm512_undefined_epi32(), \
- (long long const *)(addr), \
+ (void const *)(addr), \
(__v8si)(__m256i)(index), (__mmask8)-1, \
- (int)(scale)); })
+ (int)(scale))
-#define _mm512_mask_i32gather_epi64(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm512_mask_i32gather_epi64(v1_old, mask, index, addr, scale) \
(__m512i)__builtin_ia32_gathersiv8di((__v8di)(__m512i)(v1_old), \
- (long long const *)(addr), \
+ (void const *)(addr), \
(__v8si)(__m256i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm512_i64scatter_ps(addr, index, v1, scale) __extension__ ({\
- __builtin_ia32_scatterdiv16sf((float *)(addr), (__mmask8)-1, \
+#define _mm512_i64scatter_ps(addr, index, v1, scale) \
+ __builtin_ia32_scatterdiv16sf((void *)(addr), (__mmask8)-1, \
(__v8di)(__m512i)(index), \
- (__v8sf)(__m256)(v1), (int)(scale)); })
+ (__v8sf)(__m256)(v1), (int)(scale))
-#define _mm512_mask_i64scatter_ps(addr, mask, index, v1, scale) __extension__ ({\
- __builtin_ia32_scatterdiv16sf((float *)(addr), (__mmask8)(mask), \
+#define _mm512_mask_i64scatter_ps(addr, mask, index, v1, scale) \
+ __builtin_ia32_scatterdiv16sf((void *)(addr), (__mmask8)(mask), \
(__v8di)(__m512i)(index), \
- (__v8sf)(__m256)(v1), (int)(scale)); })
+ (__v8sf)(__m256)(v1), (int)(scale))
-#define _mm512_i64scatter_epi32(addr, index, v1, scale) __extension__ ({\
- __builtin_ia32_scatterdiv16si((int *)(addr), (__mmask8)-1, \
+#define _mm512_i64scatter_epi32(addr, index, v1, scale) \
+ __builtin_ia32_scatterdiv16si((void *)(addr), (__mmask8)-1, \
(__v8di)(__m512i)(index), \
- (__v8si)(__m256i)(v1), (int)(scale)); })
+ (__v8si)(__m256i)(v1), (int)(scale))
-#define _mm512_mask_i64scatter_epi32(addr, mask, index, v1, scale) __extension__ ({\
- __builtin_ia32_scatterdiv16si((int *)(addr), (__mmask8)(mask), \
+#define _mm512_mask_i64scatter_epi32(addr, mask, index, v1, scale) \
+ __builtin_ia32_scatterdiv16si((void *)(addr), (__mmask8)(mask), \
(__v8di)(__m512i)(index), \
- (__v8si)(__m256i)(v1), (int)(scale)); })
+ (__v8si)(__m256i)(v1), (int)(scale))
-#define _mm512_i64scatter_pd(addr, index, v1, scale) __extension__ ({\
- __builtin_ia32_scatterdiv8df((double *)(addr), (__mmask8)-1, \
+#define _mm512_i64scatter_pd(addr, index, v1, scale) \
+ __builtin_ia32_scatterdiv8df((void *)(addr), (__mmask8)-1, \
(__v8di)(__m512i)(index), \
- (__v8df)(__m512d)(v1), (int)(scale)); })
+ (__v8df)(__m512d)(v1), (int)(scale))
-#define _mm512_mask_i64scatter_pd(addr, mask, index, v1, scale) __extension__ ({\
- __builtin_ia32_scatterdiv8df((double *)(addr), (__mmask8)(mask), \
+#define _mm512_mask_i64scatter_pd(addr, mask, index, v1, scale) \
+ __builtin_ia32_scatterdiv8df((void *)(addr), (__mmask8)(mask), \
(__v8di)(__m512i)(index), \
- (__v8df)(__m512d)(v1), (int)(scale)); })
+ (__v8df)(__m512d)(v1), (int)(scale))
-#define _mm512_i64scatter_epi64(addr, index, v1, scale) __extension__ ({\
- __builtin_ia32_scatterdiv8di((long long *)(addr), (__mmask8)-1, \
+#define _mm512_i64scatter_epi64(addr, index, v1, scale) \
+ __builtin_ia32_scatterdiv8di((void *)(addr), (__mmask8)-1, \
(__v8di)(__m512i)(index), \
- (__v8di)(__m512i)(v1), (int)(scale)); })
+ (__v8di)(__m512i)(v1), (int)(scale))
-#define _mm512_mask_i64scatter_epi64(addr, mask, index, v1, scale) __extension__ ({\
- __builtin_ia32_scatterdiv8di((long long *)(addr), (__mmask8)(mask), \
+#define _mm512_mask_i64scatter_epi64(addr, mask, index, v1, scale) \
+ __builtin_ia32_scatterdiv8di((void *)(addr), (__mmask8)(mask), \
(__v8di)(__m512i)(index), \
- (__v8di)(__m512i)(v1), (int)(scale)); })
+ (__v8di)(__m512i)(v1), (int)(scale))
-#define _mm512_i32scatter_ps(addr, index, v1, scale) __extension__ ({\
- __builtin_ia32_scattersiv16sf((float *)(addr), (__mmask16)-1, \
+#define _mm512_i32scatter_ps(addr, index, v1, scale) \
+ __builtin_ia32_scattersiv16sf((void *)(addr), (__mmask16)-1, \
(__v16si)(__m512i)(index), \
- (__v16sf)(__m512)(v1), (int)(scale)); })
+ (__v16sf)(__m512)(v1), (int)(scale))
-#define _mm512_mask_i32scatter_ps(addr, mask, index, v1, scale) __extension__ ({\
- __builtin_ia32_scattersiv16sf((float *)(addr), (__mmask16)(mask), \
+#define _mm512_mask_i32scatter_ps(addr, mask, index, v1, scale) \
+ __builtin_ia32_scattersiv16sf((void *)(addr), (__mmask16)(mask), \
(__v16si)(__m512i)(index), \
- (__v16sf)(__m512)(v1), (int)(scale)); })
+ (__v16sf)(__m512)(v1), (int)(scale))
-#define _mm512_i32scatter_epi32(addr, index, v1, scale) __extension__ ({\
- __builtin_ia32_scattersiv16si((int *)(addr), (__mmask16)-1, \
+#define _mm512_i32scatter_epi32(addr, index, v1, scale) \
+ __builtin_ia32_scattersiv16si((void *)(addr), (__mmask16)-1, \
(__v16si)(__m512i)(index), \
- (__v16si)(__m512i)(v1), (int)(scale)); })
+ (__v16si)(__m512i)(v1), (int)(scale))
-#define _mm512_mask_i32scatter_epi32(addr, mask, index, v1, scale) __extension__ ({\
- __builtin_ia32_scattersiv16si((int *)(addr), (__mmask16)(mask), \
+#define _mm512_mask_i32scatter_epi32(addr, mask, index, v1, scale) \
+ __builtin_ia32_scattersiv16si((void *)(addr), (__mmask16)(mask), \
(__v16si)(__m512i)(index), \
- (__v16si)(__m512i)(v1), (int)(scale)); })
+ (__v16si)(__m512i)(v1), (int)(scale))
-#define _mm512_i32scatter_pd(addr, index, v1, scale) __extension__ ({\
- __builtin_ia32_scattersiv8df((double *)(addr), (__mmask8)-1, \
+#define _mm512_i32scatter_pd(addr, index, v1, scale) \
+ __builtin_ia32_scattersiv8df((void *)(addr), (__mmask8)-1, \
(__v8si)(__m256i)(index), \
- (__v8df)(__m512d)(v1), (int)(scale)); })
+ (__v8df)(__m512d)(v1), (int)(scale))
-#define _mm512_mask_i32scatter_pd(addr, mask, index, v1, scale) __extension__ ({\
- __builtin_ia32_scattersiv8df((double *)(addr), (__mmask8)(mask), \
+#define _mm512_mask_i32scatter_pd(addr, mask, index, v1, scale) \
+ __builtin_ia32_scattersiv8df((void *)(addr), (__mmask8)(mask), \
(__v8si)(__m256i)(index), \
- (__v8df)(__m512d)(v1), (int)(scale)); })
+ (__v8df)(__m512d)(v1), (int)(scale))
-#define _mm512_i32scatter_epi64(addr, index, v1, scale) __extension__ ({\
- __builtin_ia32_scattersiv8di((long long *)(addr), (__mmask8)-1, \
+#define _mm512_i32scatter_epi64(addr, index, v1, scale) \
+ __builtin_ia32_scattersiv8di((void *)(addr), (__mmask8)-1, \
(__v8si)(__m256i)(index), \
- (__v8di)(__m512i)(v1), (int)(scale)); })
+ (__v8di)(__m512i)(v1), (int)(scale))
-#define _mm512_mask_i32scatter_epi64(addr, mask, index, v1, scale) __extension__ ({\
- __builtin_ia32_scattersiv8di((long long *)(addr), (__mmask8)(mask), \
+#define _mm512_mask_i32scatter_epi64(addr, mask, index, v1, scale) \
+ __builtin_ia32_scattersiv8di((void *)(addr), (__mmask8)(mask), \
(__v8si)(__m256i)(index), \
- (__v8di)(__m512i)(v1), (int)(scale)); })
+ (__v8di)(__m512i)(v1), (int)(scale))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_fmadd_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
- return (__m128) __builtin_ia32_vfmaddss3_mask ((__v4sf) __W,
- (__v4sf) __A,
- (__v4sf) __B,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddss3_mask((__v4sf)__W,
+ (__v4sf)__A,
+ (__v4sf)__B,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask_fmadd_round_ss(W, U, A, B, R) __extension__({\
+#define _mm_fmadd_round_ss(A, B, C, R) \
+ (__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(A), \
+ (__v4sf)(__m128)(B), \
+ (__v4sf)(__m128)(C), (__mmask8)-1, \
+ (int)(R))
+
+#define _mm_mask_fmadd_round_ss(W, U, A, B, R) \
(__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(W), \
(__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_fmadd_ss (__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfmaddss3_maskz ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __C,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddss3_maskz((__v4sf)__A,
+ (__v4sf)__B,
+ (__v4sf)__C,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_maskz_fmadd_round_ss(U, A, B, C, R) __extension__ ({\
+#define _mm_maskz_fmadd_round_ss(U, A, B, C, R) \
(__m128)__builtin_ia32_vfmaddss3_maskz((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(C), (__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask3_fmadd_ss (__m128 __W, __m128 __X, __m128 __Y, __mmask8 __U)
{
- return (__m128) __builtin_ia32_vfmaddss3_mask3 ((__v4sf) __W,
- (__v4sf) __X,
- (__v4sf) __Y,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddss3_mask3((__v4sf)__W,
+ (__v4sf)__X,
+ (__v4sf)__Y,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask3_fmadd_round_ss(W, X, Y, U, R) __extension__ ({\
+#define _mm_mask3_fmadd_round_ss(W, X, Y, U, R) \
(__m128)__builtin_ia32_vfmaddss3_mask3((__v4sf)(__m128)(W), \
(__v4sf)(__m128)(X), \
(__v4sf)(__m128)(Y), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_fmsub_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
- return (__m128) __builtin_ia32_vfmaddss3_mask ((__v4sf) __W,
- (__v4sf) __A,
- -(__v4sf) __B,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddss3_mask((__v4sf)__W,
+ (__v4sf)__A,
+ -(__v4sf)__B,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask_fmsub_round_ss(W, U, A, B, R) __extension__ ({\
+#define _mm_fmsub_round_ss(A, B, C, R) \
+ (__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(A), \
+ (__v4sf)(__m128)(B), \
+ -(__v4sf)(__m128)(C), (__mmask8)-1, \
+ (int)(R))
+
+#define _mm_mask_fmsub_round_ss(W, U, A, B, R) \
(__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(W), \
(__v4sf)(__m128)(A), \
- (__v4sf)(__m128)(B), (__mmask8)(U), \
- (int)(R)); })
+ -(__v4sf)(__m128)(B), (__mmask8)(U), \
+ (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_fmsub_ss (__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfmaddss3_maskz ((__v4sf) __A,
- (__v4sf) __B,
- -(__v4sf) __C,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddss3_maskz((__v4sf)__A,
+ (__v4sf)__B,
+ -(__v4sf)__C,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_maskz_fmsub_round_ss(U, A, B, C, R) __extension__ ({\
+#define _mm_maskz_fmsub_round_ss(U, A, B, C, R) \
(__m128)__builtin_ia32_vfmaddss3_maskz((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
-(__v4sf)(__m128)(C), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask3_fmsub_ss (__m128 __W, __m128 __X, __m128 __Y, __mmask8 __U)
{
- return (__m128) __builtin_ia32_vfmsubss3_mask3 ((__v4sf) __W,
- (__v4sf) __X,
- (__v4sf) __Y,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmsubss3_mask3((__v4sf)__W,
+ (__v4sf)__X,
+ (__v4sf)__Y,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask3_fmsub_round_ss(W, X, Y, U, R) __extension__ ({\
+#define _mm_mask3_fmsub_round_ss(W, X, Y, U, R) \
(__m128)__builtin_ia32_vfmsubss3_mask3((__v4sf)(__m128)(W), \
(__v4sf)(__m128)(X), \
(__v4sf)(__m128)(Y), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_fnmadd_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
- return (__m128) __builtin_ia32_vfmaddss3_mask ((__v4sf) __W,
- -(__v4sf) __A,
- (__v4sf) __B,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddss3_mask((__v4sf)__W,
+ -(__v4sf)__A,
+ (__v4sf)__B,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask_fnmadd_round_ss(W, U, A, B, R) __extension__ ({\
+#define _mm_fnmadd_round_ss(A, B, C, R) \
+ (__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(A), \
+ -(__v4sf)(__m128)(B), \
+ (__v4sf)(__m128)(C), (__mmask8)-1, \
+ (int)(R))
+
+#define _mm_mask_fnmadd_round_ss(W, U, A, B, R) \
(__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(W), \
-(__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_fnmadd_ss (__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfmaddss3_maskz (-(__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __C,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddss3_maskz((__v4sf)__A,
+ -(__v4sf)__B,
+ (__v4sf)__C,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_maskz_fnmadd_round_ss(U, A, B, C, R) __extension__ ({\
- (__m128)__builtin_ia32_vfmaddss3_maskz(-(__v4sf)(__m128)(A), \
- (__v4sf)(__m128)(B), \
+#define _mm_maskz_fnmadd_round_ss(U, A, B, C, R) \
+ (__m128)__builtin_ia32_vfmaddss3_maskz((__v4sf)(__m128)(A), \
+ -(__v4sf)(__m128)(B), \
(__v4sf)(__m128)(C), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask3_fnmadd_ss (__m128 __W, __m128 __X, __m128 __Y, __mmask8 __U)
{
- return (__m128) __builtin_ia32_vfmaddss3_mask3 (-(__v4sf) __W,
- (__v4sf) __X,
- (__v4sf) __Y,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddss3_mask3((__v4sf)__W,
+ -(__v4sf)__X,
+ (__v4sf)__Y,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask3_fnmadd_round_ss(W, X, Y, U, R) __extension__({\
- (__m128)__builtin_ia32_vfmaddss3_mask3(-(__v4sf)(__m128)(W), \
- (__v4sf)(__m128)(X), \
+#define _mm_mask3_fnmadd_round_ss(W, X, Y, U, R) \
+ (__m128)__builtin_ia32_vfmaddss3_mask3((__v4sf)(__m128)(W), \
+ -(__v4sf)(__m128)(X), \
(__v4sf)(__m128)(Y), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_fnmsub_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
- return (__m128) __builtin_ia32_vfmaddss3_mask ((__v4sf) __W,
- -(__v4sf) __A,
- -(__v4sf) __B,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddss3_mask((__v4sf)__W,
+ -(__v4sf)__A,
+ -(__v4sf)__B,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask_fnmsub_round_ss(W, U, A, B, R) __extension__ ({\
+#define _mm_fnmsub_round_ss(A, B, C, R) \
+ (__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(A), \
+ -(__v4sf)(__m128)(B), \
+ -(__v4sf)(__m128)(C), (__mmask8)-1, \
+ (int)(R))
+
+#define _mm_mask_fnmsub_round_ss(W, U, A, B, R) \
(__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(W), \
-(__v4sf)(__m128)(A), \
-(__v4sf)(__m128)(B), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_fnmsub_ss (__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfmaddss3_maskz (-(__v4sf) __A,
- (__v4sf) __B,
- -(__v4sf) __C,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddss3_maskz((__v4sf)__A,
+ -(__v4sf)__B,
+ -(__v4sf)__C,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_maskz_fnmsub_round_ss(U, A, B, C, R) __extension__ ({\
- (__m128)__builtin_ia32_vfmaddss3_maskz(-(__v4sf)(__m128)(A), \
- (__v4sf)(__m128)(B), \
+#define _mm_maskz_fnmsub_round_ss(U, A, B, C, R) \
+ (__m128)__builtin_ia32_vfmaddss3_maskz((__v4sf)(__m128)(A), \
+ -(__v4sf)(__m128)(B), \
-(__v4sf)(__m128)(C), (__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask3_fnmsub_ss (__m128 __W, __m128 __X, __m128 __Y, __mmask8 __U)
{
- return (__m128) __builtin_ia32_vfnmsubss3_mask3 ((__v4sf) __W,
- (__v4sf) __X,
- (__v4sf) __Y,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmsubss3_mask3((__v4sf)__W,
+ -(__v4sf)__X,
+ (__v4sf)__Y,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask3_fnmsub_round_ss(W, X, Y, U, R) __extension__({\
- (__m128)__builtin_ia32_vfnmsubss3_mask3((__v4sf)(__m128)(W), \
- (__v4sf)(__m128)(X), \
+#define _mm_mask3_fnmsub_round_ss(W, X, Y, U, R) \
+ (__m128)__builtin_ia32_vfmsubss3_mask3((__v4sf)(__m128)(W), \
+ -(__v4sf)(__m128)(X), \
(__v4sf)(__m128)(Y), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_fmadd_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
- return (__m128d) __builtin_ia32_vfmaddsd3_mask ( (__v2df) __W,
- (__v2df) __A,
- (__v2df) __B,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddsd3_mask((__v2df)__W,
+ (__v2df)__A,
+ (__v2df)__B,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask_fmadd_round_sd(W, U, A, B, R) __extension__({\
+#define _mm_fmadd_round_sd(A, B, C, R) \
+ (__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(A), \
+ (__v2df)(__m128d)(B), \
+ (__v2df)(__m128d)(C), (__mmask8)-1, \
+ (int)(R))
+
+#define _mm_mask_fmadd_round_sd(W, U, A, B, R) \
(__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(W), \
(__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_fmadd_sd (__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfmaddsd3_maskz ( (__v2df) __A,
- (__v2df) __B,
- (__v2df) __C,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddsd3_maskz((__v2df)__A,
+ (__v2df)__B,
+ (__v2df)__C,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_maskz_fmadd_round_sd(U, A, B, C, R) __extension__ ({\
+#define _mm_maskz_fmadd_round_sd(U, A, B, C, R) \
(__m128d)__builtin_ia32_vfmaddsd3_maskz((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(C), (__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask3_fmadd_sd (__m128d __W, __m128d __X, __m128d __Y, __mmask8 __U)
{
- return (__m128d) __builtin_ia32_vfmaddsd3_mask3 ((__v2df) __W,
- (__v2df) __X,
- (__v2df) __Y,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddsd3_mask3((__v2df)__W,
+ (__v2df)__X,
+ (__v2df)__Y,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask3_fmadd_round_sd(W, X, Y, U, R) __extension__ ({\
+#define _mm_mask3_fmadd_round_sd(W, X, Y, U, R) \
(__m128d)__builtin_ia32_vfmaddsd3_mask3((__v2df)(__m128d)(W), \
(__v2df)(__m128d)(X), \
(__v2df)(__m128d)(Y), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_fmsub_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
- return (__m128d) __builtin_ia32_vfmaddsd3_mask ( (__v2df) __W,
- (__v2df) __A,
- -(__v2df) __B,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddsd3_mask((__v2df)__W,
+ (__v2df)__A,
+ -(__v2df)__B,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask_fmsub_round_sd(W, U, A, B, R) __extension__ ({\
+#define _mm_fmsub_round_sd(A, B, C, R) \
+ (__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(A), \
+ (__v2df)(__m128d)(B), \
+ -(__v2df)(__m128d)(C), (__mmask8)-1, \
+ (int)(R))
+
+#define _mm_mask_fmsub_round_sd(W, U, A, B, R) \
(__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(W), \
(__v2df)(__m128d)(A), \
-(__v2df)(__m128d)(B), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_fmsub_sd (__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfmaddsd3_maskz ( (__v2df) __A,
- (__v2df) __B,
- -(__v2df) __C,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddsd3_maskz((__v2df)__A,
+ (__v2df)__B,
+ -(__v2df)__C,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_maskz_fmsub_round_sd(U, A, B, C, R) __extension__ ({\
+#define _mm_maskz_fmsub_round_sd(U, A, B, C, R) \
(__m128d)__builtin_ia32_vfmaddsd3_maskz((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
-(__v2df)(__m128d)(C), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask3_fmsub_sd (__m128d __W, __m128d __X, __m128d __Y, __mmask8 __U)
{
- return (__m128d) __builtin_ia32_vfmsubsd3_mask3 ((__v2df) __W,
- (__v2df) __X,
- (__v2df) __Y,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmsubsd3_mask3((__v2df)__W,
+ (__v2df)__X,
+ (__v2df)__Y,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask3_fmsub_round_sd(W, X, Y, U, R) __extension__ ({\
+#define _mm_mask3_fmsub_round_sd(W, X, Y, U, R) \
(__m128d)__builtin_ia32_vfmsubsd3_mask3((__v2df)(__m128d)(W), \
(__v2df)(__m128d)(X), \
(__v2df)(__m128d)(Y), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_fnmadd_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
- return (__m128d) __builtin_ia32_vfmaddsd3_mask ( (__v2df) __W,
- -(__v2df) __A,
- (__v2df) __B,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddsd3_mask((__v2df)__W,
+ -(__v2df)__A,
+ (__v2df)__B,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask_fnmadd_round_sd(W, U, A, B, R) __extension__ ({\
+#define _mm_fnmadd_round_sd(A, B, C, R) \
+ (__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(A), \
+ -(__v2df)(__m128d)(B), \
+ (__v2df)(__m128d)(C), (__mmask8)-1, \
+ (int)(R))
+
+#define _mm_mask_fnmadd_round_sd(W, U, A, B, R) \
(__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(W), \
-(__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_fnmadd_sd (__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfmaddsd3_maskz ( -(__v2df) __A,
- (__v2df) __B,
- (__v2df) __C,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddsd3_maskz((__v2df)__A,
+ -(__v2df)__B,
+ (__v2df)__C,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_maskz_fnmadd_round_sd(U, A, B, C, R) __extension__ ({\
- (__m128d)__builtin_ia32_vfmaddsd3_maskz(-(__v2df)(__m128d)(A), \
- (__v2df)(__m128d)(B), \
+#define _mm_maskz_fnmadd_round_sd(U, A, B, C, R) \
+ (__m128d)__builtin_ia32_vfmaddsd3_maskz((__v2df)(__m128d)(A), \
+ -(__v2df)(__m128d)(B), \
(__v2df)(__m128d)(C), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask3_fnmadd_sd (__m128d __W, __m128d __X, __m128d __Y, __mmask8 __U)
{
- return (__m128d) __builtin_ia32_vfmaddsd3_mask3 (-(__v2df) __W,
- (__v2df) __X,
- (__v2df) __Y,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddsd3_mask3((__v2df)__W,
+ -(__v2df)__X,
+ (__v2df)__Y,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask3_fnmadd_round_sd(W, X, Y, U, R) __extension__({\
- (__m128d)__builtin_ia32_vfmaddsd3_mask3(-(__v2df)(__m128d)(W), \
- (__v2df)(__m128d)(X), \
+#define _mm_mask3_fnmadd_round_sd(W, X, Y, U, R) \
+ (__m128d)__builtin_ia32_vfmaddsd3_mask3((__v2df)(__m128d)(W), \
+ -(__v2df)(__m128d)(X), \
(__v2df)(__m128d)(Y), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_fnmsub_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
- return (__m128d) __builtin_ia32_vfmaddsd3_mask ( (__v2df) __W,
- -(__v2df) __A,
- -(__v2df) __B,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddsd3_mask((__v2df)__W,
+ -(__v2df)__A,
+ -(__v2df)__B,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask_fnmsub_round_sd(W, U, A, B, R) __extension__ ({\
+#define _mm_fnmsub_round_sd(A, B, C, R) \
+ (__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(A), \
+ -(__v2df)(__m128d)(B), \
+ -(__v2df)(__m128d)(C), (__mmask8)-1, \
+ (int)(R))
+
+#define _mm_mask_fnmsub_round_sd(W, U, A, B, R) \
(__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(W), \
-(__v2df)(__m128d)(A), \
-(__v2df)(__m128d)(B), (__mmask8)(U), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_fnmsub_sd (__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfmaddsd3_maskz ( -(__v2df) __A,
- (__v2df) __B,
- -(__v2df) __C,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmaddsd3_maskz((__v2df)__A,
+ -(__v2df)__B,
+ -(__v2df)__C,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_maskz_fnmsub_round_sd(U, A, B, C, R) __extension__ ({\
- (__m128d)__builtin_ia32_vfmaddsd3_maskz(-(__v2df)(__m128d)(A), \
- (__v2df)(__m128d)(B), \
+#define _mm_maskz_fnmsub_round_sd(U, A, B, C, R) \
+ (__m128d)__builtin_ia32_vfmaddsd3_maskz((__v2df)(__m128d)(A), \
+ -(__v2df)(__m128d)(B), \
-(__v2df)(__m128d)(C), \
(__mmask8)(U), \
- _MM_FROUND_CUR_DIRECTION); })
+ (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask3_fnmsub_sd (__m128d __W, __m128d __X, __m128d __Y, __mmask8 __U)
{
- return (__m128d) __builtin_ia32_vfnmsubsd3_mask3 ((__v2df) (__W),
- (__v2df) __X,
- (__v2df) (__Y),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_vfmsubsd3_mask3((__v2df)__W,
+ -(__v2df)__X,
+ (__v2df)__Y,
+ (__mmask8)__U,
+ _MM_FROUND_CUR_DIRECTION);
}
-#define _mm_mask3_fnmsub_round_sd(W, X, Y, U, R) __extension__({\
- (__m128d)__builtin_ia32_vfnmsubsd3_mask3((__v2df)(__m128d)(W), \
- (__v2df)(__m128d)(X), \
+#define _mm_mask3_fnmsub_round_sd(W, X, Y, U, R) \
+ (__m128d)__builtin_ia32_vfmsubsd3_mask3((__v2df)(__m128d)(W), \
+ -(__v2df)(__m128d)(X), \
(__v2df)(__m128d)(Y), \
- (__mmask8)(U), (int)(R)); })
-
-#define _mm512_permutex_pd(X, C) __extension__ ({ \
- (__m512d)__builtin_shufflevector((__v8df)(__m512d)(X), \
- (__v8df)_mm512_undefined_pd(), \
- 0 + (((C) >> 0) & 0x3), \
- 0 + (((C) >> 2) & 0x3), \
- 0 + (((C) >> 4) & 0x3), \
- 0 + (((C) >> 6) & 0x3), \
- 4 + (((C) >> 0) & 0x3), \
- 4 + (((C) >> 2) & 0x3), \
- 4 + (((C) >> 4) & 0x3), \
- 4 + (((C) >> 6) & 0x3)); })
-
-#define _mm512_mask_permutex_pd(W, U, X, C) __extension__ ({ \
+ (__mmask8)(U), (int)(R))
+
+#define _mm512_permutex_pd(X, C) \
+ (__m512d)__builtin_ia32_permdf512((__v8df)(__m512d)(X), (int)(C))
+
+#define _mm512_mask_permutex_pd(W, U, X, C) \
(__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
(__v8df)_mm512_permutex_pd((X), (C)), \
- (__v8df)(__m512d)(W)); })
+ (__v8df)(__m512d)(W))
-#define _mm512_maskz_permutex_pd(U, X, C) __extension__ ({ \
+#define _mm512_maskz_permutex_pd(U, X, C) \
(__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
(__v8df)_mm512_permutex_pd((X), (C)), \
- (__v8df)_mm512_setzero_pd()); })
-
-#define _mm512_permutex_epi64(X, C) __extension__ ({ \
- (__m512i)__builtin_shufflevector((__v8di)(__m512i)(X), \
- (__v8di)_mm512_undefined_epi32(), \
- 0 + (((C) >> 0) & 0x3), \
- 0 + (((C) >> 2) & 0x3), \
- 0 + (((C) >> 4) & 0x3), \
- 0 + (((C) >> 6) & 0x3), \
- 4 + (((C) >> 0) & 0x3), \
- 4 + (((C) >> 2) & 0x3), \
- 4 + (((C) >> 4) & 0x3), \
- 4 + (((C) >> 6) & 0x3)); })
-
-#define _mm512_mask_permutex_epi64(W, U, X, C) __extension__ ({ \
+ (__v8df)_mm512_setzero_pd())
+
+#define _mm512_permutex_epi64(X, C) \
+ (__m512i)__builtin_ia32_permdi512((__v8di)(__m512i)(X), (int)(C))
+
+#define _mm512_mask_permutex_epi64(W, U, X, C) \
(__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
(__v8di)_mm512_permutex_epi64((X), (C)), \
- (__v8di)(__m512i)(W)); })
+ (__v8di)(__m512i)(W))
-#define _mm512_maskz_permutex_epi64(U, X, C) __extension__ ({ \
+#define _mm512_maskz_permutex_epi64(U, X, C) \
(__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
(__v8di)_mm512_permutex_epi64((X), (C)), \
- (__v8di)_mm512_setzero_si512()); })
+ (__v8di)_mm512_setzero_si512())
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_permutexvar_pd (__m512i __X, __m512d __Y)
{
- return (__m512d) __builtin_ia32_permvardf512_mask ((__v8df) __Y,
- (__v8di) __X,
- (__v8df) _mm512_undefined_pd (),
- (__mmask8) -1);
+ return (__m512d)__builtin_ia32_permvardf512((__v8df) __Y, (__v8di) __X);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_permutexvar_pd (__m512d __W, __mmask8 __U, __m512i __X, __m512d __Y)
{
- return (__m512d) __builtin_ia32_permvardf512_mask ((__v8df) __Y,
- (__v8di) __X,
- (__v8df) __W,
- (__mmask8) __U);
+ return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
+ (__v8df)_mm512_permutexvar_pd(__X, __Y),
+ (__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_permutexvar_pd (__mmask8 __U, __m512i __X, __m512d __Y)
{
- return (__m512d) __builtin_ia32_permvardf512_mask ((__v8df) __Y,
- (__v8di) __X,
- (__v8df) _mm512_setzero_pd (),
- (__mmask8) __U);
+ return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
+ (__v8df)_mm512_permutexvar_pd(__X, __Y),
+ (__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_permutexvar_epi64 (__mmask8 __M, __m512i __X, __m512i __Y)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_permutexvar_epi64 (__m512i __X, __m512i __Y)
{
- return (__m512i) __builtin_ia32_permvardi512_mask ((__v8di) __Y,
- (__v8di) __X,
- (__v8di) _mm512_setzero_si512 (),
- __M);
+ return (__m512i)__builtin_ia32_permvardi512((__v8di)__Y, (__v8di)__X);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_permutexvar_epi64 (__m512i __X, __m512i __Y)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_permutexvar_epi64 (__mmask8 __M, __m512i __X, __m512i __Y)
{
- return (__m512i) __builtin_ia32_permvardi512_mask ((__v8di) __Y,
- (__v8di) __X,
- (__v8di) _mm512_undefined_epi32 (),
- (__mmask8) -1);
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
+ (__v8di)_mm512_permutexvar_epi64(__X, __Y),
+ (__v8di)_mm512_setzero_si512());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_permutexvar_epi64 (__m512i __W, __mmask8 __M, __m512i __X,
__m512i __Y)
{
- return (__m512i) __builtin_ia32_permvardi512_mask ((__v8di) __Y,
- (__v8di) __X,
- (__v8di) __W,
- __M);
+ return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
+ (__v8di)_mm512_permutexvar_epi64(__X, __Y),
+ (__v8di)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_permutexvar_ps (__m512i __X, __m512 __Y)
{
- return (__m512) __builtin_ia32_permvarsf512_mask ((__v16sf) __Y,
- (__v16si) __X,
- (__v16sf) _mm512_undefined_ps (),
- (__mmask16) -1);
+ return (__m512)__builtin_ia32_permvarsf512((__v16sf)__Y, (__v16si)__X);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_permutexvar_ps (__m512 __W, __mmask16 __U, __m512i __X, __m512 __Y)
{
- return (__m512) __builtin_ia32_permvarsf512_mask ((__v16sf) __Y,
- (__v16si) __X,
- (__v16sf) __W,
- (__mmask16) __U);
+ return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
+ (__v16sf)_mm512_permutexvar_ps(__X, __Y),
+ (__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_permutexvar_ps (__mmask16 __U, __m512i __X, __m512 __Y)
{
- return (__m512) __builtin_ia32_permvarsf512_mask ((__v16sf) __Y,
- (__v16si) __X,
- (__v16sf) _mm512_setzero_ps (),
- (__mmask16) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_permutexvar_epi32 (__mmask16 __M, __m512i __X, __m512i __Y)
-{
- return (__m512i) __builtin_ia32_permvarsi512_mask ((__v16si) __Y,
- (__v16si) __X,
- (__v16si) _mm512_setzero_si512 (),
- __M);
+ return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
+ (__v16sf)_mm512_permutexvar_ps(__X, __Y),
+ (__v16sf)_mm512_setzero_ps());
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_permutexvar_epi32 (__m512i __X, __m512i __Y)
{
- return (__m512i) __builtin_ia32_permvarsi512_mask ((__v16si) __Y,
- (__v16si) __X,
- (__v16si) _mm512_undefined_epi32 (),
- (__mmask16) -1);
+ return (__m512i)__builtin_ia32_permvarsi512((__v16si)__Y, (__v16si)__X);
}
#define _mm512_permutevar_epi32 _mm512_permutexvar_epi32
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_permutexvar_epi32 (__mmask16 __M, __m512i __X, __m512i __Y)
+{
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
+ (__v16si)_mm512_permutexvar_epi32(__X, __Y),
+ (__v16si)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_permutexvar_epi32 (__m512i __W, __mmask16 __M, __m512i __X,
__m512i __Y)
{
- return (__m512i) __builtin_ia32_permvarsi512_mask ((__v16si) __Y,
- (__v16si) __X,
- (__v16si) __W,
- __M);
+ return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
+ (__v16si)_mm512_permutexvar_epi32(__X, __Y),
+ (__v16si)__W);
}
#define _mm512_mask_permutevar_epi32 _mm512_mask_permutexvar_epi32
return __builtin_ia32_kortestzhi ((__mmask16) __A, (__mmask16) __B);
}
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_kortestc_mask16_u8(__mmask16 __A, __mmask16 __B)
+{
+ return (unsigned char)__builtin_ia32_kortestchi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_kortestz_mask16_u8(__mmask16 __A, __mmask16 __B)
+{
+ return (unsigned char)__builtin_ia32_kortestzhi(__A, __B);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_kortest_mask16_u8(__mmask16 __A, __mmask16 __B, unsigned char *__C) {
+ *__C = (unsigned char)__builtin_ia32_kortestchi(__A, __B);
+ return (unsigned char)__builtin_ia32_kortestzhi(__A, __B);
+}
+
static __inline__ __mmask16 __DEFAULT_FN_ATTRS
_mm512_kunpackb (__mmask16 __A, __mmask16 __B)
{
return (__mmask16) __builtin_ia32_kxorhi ((__mmask16) __A, (__mmask16) __B);
}
+#define _kand_mask16 _mm512_kand
+#define _kandn_mask16 _mm512_kandn
+#define _knot_mask16 _mm512_knot
+#define _kor_mask16 _mm512_kor
+#define _kxnor_mask16 _mm512_kxnor
+#define _kxor_mask16 _mm512_kxor
+
+#define _kshiftli_mask16(A, I) \
+ (__mmask16)__builtin_ia32_kshiftlihi((__mmask16)(A), (unsigned int)(I))
+
+#define _kshiftri_mask16(A, I) \
+ (__mmask16)__builtin_ia32_kshiftrihi((__mmask16)(A), (unsigned int)(I))
+
+static __inline__ unsigned int __DEFAULT_FN_ATTRS
+_cvtmask16_u32(__mmask16 __A) {
+ return (unsigned int)__builtin_ia32_kmovw((__mmask16)__A);
+}
+
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS
+_cvtu32_mask16(unsigned int __A) {
+ return (__mmask16)__builtin_ia32_kmovw((__mmask16)__A);
+}
+
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS
+_load_mask16(__mmask16 *__A) {
+ return (__mmask16)__builtin_ia32_kmovw(*(__mmask16 *)__A);
+}
+
static __inline__ void __DEFAULT_FN_ATTRS
-_mm512_stream_si512 (__m512i * __P, __m512i __A)
+_store_mask16(__mmask16 *__A, __mmask16 __B) {
+ *(__mmask16 *)__A = __builtin_ia32_kmovw((__mmask16)__B);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS512
+_mm512_stream_si512 (void * __P, __m512i __A)
{
typedef __v8di __v8di_aligned __attribute__((aligned(64)));
__builtin_nontemporal_store((__v8di_aligned)__A, (__v8di_aligned*)__P);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_stream_load_si512 (void *__P)
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_stream_load_si512 (void const *__P)
{
typedef __v8di __v8di_aligned __attribute__((aligned(64)));
return (__m512i) __builtin_nontemporal_load((const __v8di_aligned *)__P);
}
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm512_stream_pd (double *__P, __m512d __A)
+static __inline__ void __DEFAULT_FN_ATTRS512
+_mm512_stream_pd (void *__P, __m512d __A)
{
typedef __v8df __v8df_aligned __attribute__((aligned(64)));
__builtin_nontemporal_store((__v8df_aligned)__A, (__v8df_aligned*)__P);
}
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm512_stream_ps (float *__P, __m512 __A)
+static __inline__ void __DEFAULT_FN_ATTRS512
+_mm512_stream_ps (void *__P, __m512 __A)
{
typedef __v16sf __v16sf_aligned __attribute__((aligned(64)));
__builtin_nontemporal_store((__v16sf_aligned)__A, (__v16sf_aligned*)__P);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_compress_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
return (__m512d) __builtin_ia32_compressdf512_mask ((__v8df) __A,
(__mmask8) __U);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_compress_pd (__mmask8 __U, __m512d __A)
{
return (__m512d) __builtin_ia32_compressdf512_mask ((__v8df) __A,
(__mmask8) __U);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_compress_epi64 (__m512i __W, __mmask8 __U, __m512i __A)
{
return (__m512i) __builtin_ia32_compressdi512_mask ((__v8di) __A,
(__mmask8) __U);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_compress_epi64 (__mmask8 __U, __m512i __A)
{
return (__m512i) __builtin_ia32_compressdi512_mask ((__v8di) __A,
(__mmask8) __U);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_compress_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
return (__m512) __builtin_ia32_compresssf512_mask ((__v16sf) __A,
(__mmask16) __U);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_compress_ps (__mmask16 __U, __m512 __A)
{
return (__m512) __builtin_ia32_compresssf512_mask ((__v16sf) __A,
(__mmask16) __U);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_compress_epi32 (__m512i __W, __mmask16 __U, __m512i __A)
{
return (__m512i) __builtin_ia32_compresssi512_mask ((__v16si) __A,
(__mmask16) __U);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_compress_epi32 (__mmask16 __U, __m512i __A)
{
return (__m512i) __builtin_ia32_compresssi512_mask ((__v16si) __A,
(__mmask16) __U);
}
-#define _mm_cmp_round_ss_mask(X, Y, P, R) __extension__ ({ \
+#define _mm_cmp_round_ss_mask(X, Y, P, R) \
(__mmask8)__builtin_ia32_cmpss_mask((__v4sf)(__m128)(X), \
(__v4sf)(__m128)(Y), (int)(P), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_cmp_round_ss_mask(M, X, Y, P, R) __extension__ ({ \
+#define _mm_mask_cmp_round_ss_mask(M, X, Y, P, R) \
(__mmask8)__builtin_ia32_cmpss_mask((__v4sf)(__m128)(X), \
(__v4sf)(__m128)(Y), (int)(P), \
- (__mmask8)(M), (int)(R)); })
+ (__mmask8)(M), (int)(R))
-#define _mm_cmp_ss_mask(X, Y, P) __extension__ ({ \
+#define _mm_cmp_ss_mask(X, Y, P) \
(__mmask8)__builtin_ia32_cmpss_mask((__v4sf)(__m128)(X), \
(__v4sf)(__m128)(Y), (int)(P), \
(__mmask8)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_cmp_ss_mask(M, X, Y, P) __extension__ ({ \
+#define _mm_mask_cmp_ss_mask(M, X, Y, P) \
(__mmask8)__builtin_ia32_cmpss_mask((__v4sf)(__m128)(X), \
(__v4sf)(__m128)(Y), (int)(P), \
(__mmask8)(M), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_cmp_round_sd_mask(X, Y, P, R) __extension__ ({ \
+#define _mm_cmp_round_sd_mask(X, Y, P, R) \
(__mmask8)__builtin_ia32_cmpsd_mask((__v2df)(__m128d)(X), \
(__v2df)(__m128d)(Y), (int)(P), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_cmp_round_sd_mask(M, X, Y, P, R) __extension__ ({ \
+#define _mm_mask_cmp_round_sd_mask(M, X, Y, P, R) \
(__mmask8)__builtin_ia32_cmpsd_mask((__v2df)(__m128d)(X), \
(__v2df)(__m128d)(Y), (int)(P), \
- (__mmask8)(M), (int)(R)); })
+ (__mmask8)(M), (int)(R))
-#define _mm_cmp_sd_mask(X, Y, P) __extension__ ({ \
+#define _mm_cmp_sd_mask(X, Y, P) \
(__mmask8)__builtin_ia32_cmpsd_mask((__v2df)(__m128d)(X), \
(__v2df)(__m128d)(Y), (int)(P), \
(__mmask8)-1, \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
-#define _mm_mask_cmp_sd_mask(M, X, Y, P) __extension__ ({ \
+#define _mm_mask_cmp_sd_mask(M, X, Y, P) \
(__mmask8)__builtin_ia32_cmpsd_mask((__v2df)(__m128d)(X), \
(__v2df)(__m128d)(Y), (int)(P), \
(__mmask8)(M), \
- _MM_FROUND_CUR_DIRECTION); })
+ _MM_FROUND_CUR_DIRECTION)
+
+/* Bit Test */
+
+static __inline __mmask16 __DEFAULT_FN_ATTRS512
+_mm512_test_epi32_mask (__m512i __A, __m512i __B)
+{
+ return _mm512_cmpneq_epi32_mask (_mm512_and_epi32(__A, __B),
+ _mm512_setzero_si512());
+}
+
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS512
+_mm512_mask_test_epi32_mask (__mmask16 __U, __m512i __A, __m512i __B)
+{
+ return _mm512_mask_cmpneq_epi32_mask (__U, _mm512_and_epi32 (__A, __B),
+ _mm512_setzero_si512());
+}
+
+static __inline __mmask8 __DEFAULT_FN_ATTRS512
+_mm512_test_epi64_mask (__m512i __A, __m512i __B)
+{
+ return _mm512_cmpneq_epi64_mask (_mm512_and_epi32 (__A, __B),
+ _mm512_setzero_si512());
+}
+
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS512
+_mm512_mask_test_epi64_mask (__mmask8 __U, __m512i __A, __m512i __B)
+{
+ return _mm512_mask_cmpneq_epi64_mask (__U, _mm512_and_epi32 (__A, __B),
+ _mm512_setzero_si512());
+}
+
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS512
+_mm512_testn_epi32_mask (__m512i __A, __m512i __B)
+{
+ return _mm512_cmpeq_epi32_mask (_mm512_and_epi32 (__A, __B),
+ _mm512_setzero_si512());
+}
+
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS512
+_mm512_mask_testn_epi32_mask (__mmask16 __U, __m512i __A, __m512i __B)
+{
+ return _mm512_mask_cmpeq_epi32_mask (__U, _mm512_and_epi32 (__A, __B),
+ _mm512_setzero_si512());
+}
+
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS512
+_mm512_testn_epi64_mask (__m512i __A, __m512i __B)
+{
+ return _mm512_cmpeq_epi64_mask (_mm512_and_epi32 (__A, __B),
+ _mm512_setzero_si512());
+}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS512
+_mm512_mask_testn_epi64_mask (__mmask8 __U, __m512i __A, __m512i __B)
+{
+ return _mm512_mask_cmpeq_epi64_mask (__U, _mm512_and_epi32 (__A, __B),
+ _mm512_setzero_si512());
+}
+
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_movehdup_ps (__m512 __A)
{
return (__m512)__builtin_shufflevector((__v16sf)__A, (__v16sf)__A,
1, 1, 3, 3, 5, 5, 7, 7, 9, 9, 11, 11, 13, 13, 15, 15);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_movehdup_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_movehdup_ps (__mmask16 __U, __m512 __A)
{
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_setzero_ps());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_moveldup_ps (__m512 __A)
{
return (__m512)__builtin_shufflevector((__v16sf)__A, (__v16sf)__A,
0, 0, 2, 2, 4, 4, 6, 6, 8, 8, 10, 10, 12, 12, 14, 14);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_moveldup_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)__W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_moveldup_ps (__mmask16 __U, __m512 __A)
{
return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
(__v16sf)_mm512_setzero_ps());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_move_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
- __m128 res = __A;
- res[0] = (__U & 1) ? __B[0] : __W[0];
- return res;
+ return __builtin_ia32_selectss_128(__U, _mm_move_ss(__A, __B), __W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_move_ss (__mmask8 __U, __m128 __A, __m128 __B)
{
- __m128 res = __A;
- res[0] = (__U & 1) ? __B[0] : 0;
- return res;
+ return __builtin_ia32_selectss_128(__U, _mm_move_ss(__A, __B),
+ _mm_setzero_ps());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_move_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
- __m128d res = __A;
- res[0] = (__U & 1) ? __B[0] : __W[0];
- return res;
+ return __builtin_ia32_selectsd_128(__U, _mm_move_sd(__A, __B), __W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_move_sd (__mmask8 __U, __m128d __A, __m128d __B)
{
- __m128d res = __A;
- res[0] = (__U & 1) ? __B[0] : 0;
- return res;
+ return __builtin_ia32_selectsd_128(__U, _mm_move_sd(__A, __B),
+ _mm_setzero_pd());
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_store_ss (float * __W, __mmask8 __U, __m128 __A)
{
- __builtin_ia32_storess128_mask ((__v16sf *)__W,
- (__v16sf) _mm512_castps128_ps512(__A),
- (__mmask16) __U & (__mmask16)1);
+ __builtin_ia32_storess128_mask ((__v4sf *)__W, __A, __U & 1);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_store_sd (double * __W, __mmask8 __U, __m128d __A)
{
- __builtin_ia32_storesd128_mask ((__v8df *)__W,
- (__v8df) _mm512_castpd128_pd512(__A),
- (__mmask8) __U & 1);
+ __builtin_ia32_storesd128_mask ((__v2df *)__W, __A, __U & 1);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_load_ss (__m128 __W, __mmask8 __U, const float* __A)
{
__m128 src = (__v4sf) __builtin_shufflevector((__v4sf) __W,
- (__v4sf) {0.0, 0.0, 0.0, 0.0},
+ (__v4sf)_mm_setzero_ps(),
0, 4, 4, 4);
- return (__m128) __builtin_shufflevector(
- __builtin_ia32_loadss128_mask ((__v16sf *) __A,
- (__v16sf) _mm512_castps128_ps512(src),
- (__mmask16) __U & 1),
- _mm512_undefined_ps(), 0, 1, 2, 3);
+ return (__m128) __builtin_ia32_loadss128_mask ((const __v4sf *) __A, src, __U & 1);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_load_ss (__mmask8 __U, const float* __A)
{
- return (__m128) __builtin_shufflevector(
- __builtin_ia32_loadss128_mask ((__v16sf *) __A,
- (__v16sf) _mm512_setzero_ps(),
- (__mmask16) __U & 1),
- _mm512_undefined_ps(), 0, 1, 2, 3);
+ return (__m128)__builtin_ia32_loadss128_mask ((const __v4sf *) __A,
+ (__v4sf) _mm_setzero_ps(),
+ __U & 1);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_load_sd (__m128d __W, __mmask8 __U, const double* __A)
{
__m128d src = (__v2df) __builtin_shufflevector((__v2df) __W,
- (__v2df) {0.0, 0.0}, 0, 2);
+ (__v2df)_mm_setzero_pd(),
+ 0, 2);
- return (__m128d) __builtin_shufflevector(
- __builtin_ia32_loadsd128_mask ((__v8df *) __A,
- (__v8df) _mm512_castpd128_pd512(src),
- (__mmask8) __U & 1),
- _mm512_undefined_pd(), 0, 1);
+ return (__m128d) __builtin_ia32_loadsd128_mask ((const __v2df *) __A, src, __U & 1);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_load_sd (__mmask8 __U, const double* __A)
{
- return (__m128d) __builtin_shufflevector(
- __builtin_ia32_loadsd128_mask ((__v8df *) __A,
- (__v8df) _mm512_setzero_pd(),
- (__mmask8) __U & 1),
- _mm512_undefined_pd(), 0, 1);
-}
-
-#define _mm512_shuffle_epi32(A, I) __extension__ ({ \
- (__m512i)__builtin_shufflevector((__v16si)(__m512i)(A), \
- (__v16si)_mm512_undefined_epi32(), \
- 0 + (((I) >> 0) & 0x3), \
- 0 + (((I) >> 2) & 0x3), \
- 0 + (((I) >> 4) & 0x3), \
- 0 + (((I) >> 6) & 0x3), \
- 4 + (((I) >> 0) & 0x3), \
- 4 + (((I) >> 2) & 0x3), \
- 4 + (((I) >> 4) & 0x3), \
- 4 + (((I) >> 6) & 0x3), \
- 8 + (((I) >> 0) & 0x3), \
- 8 + (((I) >> 2) & 0x3), \
- 8 + (((I) >> 4) & 0x3), \
- 8 + (((I) >> 6) & 0x3), \
- 12 + (((I) >> 0) & 0x3), \
- 12 + (((I) >> 2) & 0x3), \
- 12 + (((I) >> 4) & 0x3), \
- 12 + (((I) >> 6) & 0x3)); })
-
-#define _mm512_mask_shuffle_epi32(W, U, A, I) __extension__ ({ \
+ return (__m128d) __builtin_ia32_loadsd128_mask ((const __v2df *) __A,
+ (__v2df) _mm_setzero_pd(),
+ __U & 1);
+}
+
+#define _mm512_shuffle_epi32(A, I) \
+ (__m512i)__builtin_ia32_pshufd512((__v16si)(__m512i)(A), (int)(I))
+
+#define _mm512_mask_shuffle_epi32(W, U, A, I) \
(__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
(__v16si)_mm512_shuffle_epi32((A), (I)), \
- (__v16si)(__m512i)(W)); })
+ (__v16si)(__m512i)(W))
-#define _mm512_maskz_shuffle_epi32(U, A, I) __extension__ ({ \
+#define _mm512_maskz_shuffle_epi32(U, A, I) \
(__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
(__v16si)_mm512_shuffle_epi32((A), (I)), \
- (__v16si)_mm512_setzero_si512()); })
+ (__v16si)_mm512_setzero_si512())
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_expand_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
return (__m512d) __builtin_ia32_expanddf512_mask ((__v8df) __A,
(__mmask8) __U);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_expand_pd (__mmask8 __U, __m512d __A)
{
return (__m512d) __builtin_ia32_expanddf512_mask ((__v8df) __A,
(__mmask8) __U);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_expand_epi64 (__m512i __W, __mmask8 __U, __m512i __A)
{
return (__m512i) __builtin_ia32_expanddi512_mask ((__v8di) __A,
(__mmask8) __U);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_expand_epi64 ( __mmask8 __U, __m512i __A)
{
return (__m512i) __builtin_ia32_expanddi512_mask ((__v8di) __A,
- (__v8di) _mm512_setzero_pd (),
+ (__v8di) _mm512_setzero_si512 (),
(__mmask8) __U);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_expandloadu_pd(__m512d __W, __mmask8 __U, void const *__P)
{
return (__m512d) __builtin_ia32_expandloaddf512_mask ((const __v8df *)__P,
(__mmask8) __U);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_expandloadu_pd(__mmask8 __U, void const *__P)
{
return (__m512d) __builtin_ia32_expandloaddf512_mask ((const __v8df *)__P,
(__mmask8) __U);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_expandloadu_epi64(__m512i __W, __mmask8 __U, void const *__P)
{
return (__m512i) __builtin_ia32_expandloaddi512_mask ((const __v8di *)__P,
(__mmask8) __U);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_expandloadu_epi64(__mmask8 __U, void const *__P)
{
return (__m512i) __builtin_ia32_expandloaddi512_mask ((const __v8di *)__P,
- (__v8di) _mm512_setzero_pd(),
+ (__v8di) _mm512_setzero_si512(),
(__mmask8) __U);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_expandloadu_ps(__m512 __W, __mmask16 __U, void const *__P)
{
return (__m512) __builtin_ia32_expandloadsf512_mask ((const __v16sf *)__P,
(__mmask16) __U);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_expandloadu_ps(__mmask16 __U, void const *__P)
{
return (__m512) __builtin_ia32_expandloadsf512_mask ((const __v16sf *)__P,
(__mmask16) __U);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_expandloadu_epi32(__m512i __W, __mmask16 __U, void const *__P)
{
return (__m512i) __builtin_ia32_expandloadsi512_mask ((const __v16si *)__P,
(__mmask16) __U);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_expandloadu_epi32(__mmask16 __U, void const *__P)
{
return (__m512i) __builtin_ia32_expandloadsi512_mask ((const __v16si *)__P,
- (__v16si) _mm512_setzero_ps(),
+ (__v16si) _mm512_setzero_si512(),
(__mmask16) __U);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_expand_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
return (__m512) __builtin_ia32_expandsf512_mask ((__v16sf) __A,
(__mmask16) __U);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_expand_ps (__mmask16 __U, __m512 __A)
{
return (__m512) __builtin_ia32_expandsf512_mask ((__v16sf) __A,
(__mmask16) __U);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_expand_epi32 (__m512i __W, __mmask16 __U, __m512i __A)
{
return (__m512i) __builtin_ia32_expandsi512_mask ((__v16si) __A,
(__mmask16) __U);
}
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_expand_epi32 (__mmask16 __U, __m512i __A)
{
return (__m512i) __builtin_ia32_expandsi512_mask ((__v16si) __A,
- (__v16si) _mm512_setzero_ps(),
+ (__v16si) _mm512_setzero_si512(),
(__mmask16) __U);
}
-#define _mm512_cvt_roundps_pd(A, R) __extension__ ({ \
+#define _mm512_cvt_roundps_pd(A, R) \
(__m512d)__builtin_ia32_cvtps2pd512_mask((__v8sf)(__m256)(A), \
(__v8df)_mm512_undefined_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm512_mask_cvt_roundps_pd(W, U, A, R) __extension__ ({ \
+#define _mm512_mask_cvt_roundps_pd(W, U, A, R) \
(__m512d)__builtin_ia32_cvtps2pd512_mask((__v8sf)(__m256)(A), \
(__v8df)(__m512d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm512_maskz_cvt_roundps_pd(U, A, R) __extension__ ({ \
+#define _mm512_maskz_cvt_roundps_pd(U, A, R) \
(__m512d)__builtin_ia32_cvtps2pd512_mask((__v8sf)(__m256)(A), \
(__v8df)_mm512_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_cvtps_pd (__m256 __A)
{
- return (__m512d) __builtin_ia32_cvtps2pd512_mask ((__v8sf) __A,
- (__v8df)
- _mm512_undefined_pd (),
- (__mmask8) -1,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d) __builtin_convertvector((__v8sf)__A, __v8df);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_cvtps_pd (__m512d __W, __mmask8 __U, __m256 __A)
{
- return (__m512d) __builtin_ia32_cvtps2pd512_mask ((__v8sf) __A,
- (__v8df) __W,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
+ (__v8df)_mm512_cvtps_pd(__A),
+ (__v8df)__W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtps_pd (__mmask8 __U, __m256 __A)
{
- return (__m512d) __builtin_ia32_cvtps2pd512_mask ((__v8sf) __A,
- (__v8df)
- _mm512_setzero_pd (),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
+ return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
+ (__v8df)_mm512_cvtps_pd(__A),
+ (__v8df)_mm512_setzero_pd());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_cvtpslo_pd (__m512 __A)
{
- return (__m512) _mm512_cvtps_pd(_mm512_castps512_ps256(__A));
+ return (__m512d) _mm512_cvtps_pd(_mm512_castps512_ps256(__A));
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_cvtpslo_pd (__m512d __W, __mmask8 __U, __m512 __A)
{
- return (__m512) _mm512_mask_cvtps_pd(__W, __U, _mm512_castps512_ps256(__A));
+ return (__m512d) _mm512_mask_cvtps_pd(__W, __U, _mm512_castps512_ps256(__A));
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_mov_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
return (__m512d) __builtin_ia32_selectpd_512 ((__mmask8) __U,
(__v8df) __W);
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_mov_pd (__mmask8 __U, __m512d __A)
{
return (__m512d) __builtin_ia32_selectpd_512 ((__mmask8) __U,
(__v8df) _mm512_setzero_pd ());
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_mov_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
return (__m512) __builtin_ia32_selectps_512 ((__mmask16) __U,
(__v16sf) __W);
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_mov_ps (__mmask16 __U, __m512 __A)
{
return (__m512) __builtin_ia32_selectps_512 ((__mmask16) __U,
(__v16sf) _mm512_setzero_ps ());
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_compressstoreu_pd (void *__P, __mmask8 __U, __m512d __A)
{
__builtin_ia32_compressstoredf512_mask ((__v8df *) __P, (__v8df) __A,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_compressstoreu_epi64 (void *__P, __mmask8 __U, __m512i __A)
{
__builtin_ia32_compressstoredi512_mask ((__v8di *) __P, (__v8di) __A,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_compressstoreu_ps (void *__P, __mmask16 __U, __m512 __A)
{
__builtin_ia32_compressstoresf512_mask ((__v16sf *) __P, (__v16sf) __A,
(__mmask16) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_compressstoreu_epi32 (void *__P, __mmask16 __U, __m512i __A)
{
__builtin_ia32_compressstoresi512_mask ((__v16si *) __P, (__v16si) __A,
(__mmask16) __U);
}
-#define _mm_cvt_roundsd_ss(A, B, R) __extension__ ({ \
+#define _mm_cvt_roundsd_ss(A, B, R) \
(__m128)__builtin_ia32_cvtsd2ss_round_mask((__v4sf)(__m128)(A), \
(__v2df)(__m128d)(B), \
(__v4sf)_mm_undefined_ps(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_cvt_roundsd_ss(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_cvt_roundsd_ss(W, U, A, B, R) \
(__m128)__builtin_ia32_cvtsd2ss_round_mask((__v4sf)(__m128)(A), \
(__v2df)(__m128d)(B), \
(__v4sf)(__m128)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_maskz_cvt_roundsd_ss(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_cvt_roundsd_ss(U, A, B, R) \
(__m128)__builtin_ia32_cvtsd2ss_round_mask((__v4sf)(__m128)(A), \
(__v2df)(__m128d)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_cvtsd_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128d __B)
{
- return __builtin_ia32_cvtsd2ss_round_mask ((__v4sf)(__A),
- (__v2df)(__B),
- (__v4sf)(__W),
- (__mmask8)(__U), _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_cvtsd2ss_round_mask ((__v4sf)__A,
+ (__v2df)__B,
+ (__v4sf)__W,
+ (__mmask8)__U, _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_cvtsd_ss (__mmask8 __U, __m128 __A, __m128d __B)
{
- return __builtin_ia32_cvtsd2ss_round_mask ((__v4sf)(__A),
- (__v2df)(__B),
+ return __builtin_ia32_cvtsd2ss_round_mask ((__v4sf)__A,
+ (__v2df)__B,
(__v4sf)_mm_setzero_ps(),
- (__mmask8)(__U), _MM_FROUND_CUR_DIRECTION);
+ (__mmask8)__U, _MM_FROUND_CUR_DIRECTION);
}
#define _mm_cvtss_i32 _mm_cvtss_si32
#endif
#ifdef __x86_64__
-#define _mm_cvt_roundi64_sd(A, B, R) __extension__ ({ \
+#define _mm_cvt_roundi64_sd(A, B, R) \
(__m128d)__builtin_ia32_cvtsi2sd64((__v2df)(__m128d)(A), (long long)(B), \
- (int)(R)); })
+ (int)(R))
-#define _mm_cvt_roundsi64_sd(A, B, R) __extension__ ({ \
+#define _mm_cvt_roundsi64_sd(A, B, R) \
(__m128d)__builtin_ia32_cvtsi2sd64((__v2df)(__m128d)(A), (long long)(B), \
- (int)(R)); })
+ (int)(R))
#endif
-#define _mm_cvt_roundsi32_ss(A, B, R) __extension__ ({ \
- (__m128)__builtin_ia32_cvtsi2ss32((__v4sf)(__m128)(A), (int)(B), (int)(R)); })
+#define _mm_cvt_roundsi32_ss(A, B, R) \
+ (__m128)__builtin_ia32_cvtsi2ss32((__v4sf)(__m128)(A), (int)(B), (int)(R))
-#define _mm_cvt_roundi32_ss(A, B, R) __extension__ ({ \
- (__m128)__builtin_ia32_cvtsi2ss32((__v4sf)(__m128)(A), (int)(B), (int)(R)); })
+#define _mm_cvt_roundi32_ss(A, B, R) \
+ (__m128)__builtin_ia32_cvtsi2ss32((__v4sf)(__m128)(A), (int)(B), (int)(R))
#ifdef __x86_64__
-#define _mm_cvt_roundsi64_ss(A, B, R) __extension__ ({ \
+#define _mm_cvt_roundsi64_ss(A, B, R) \
(__m128)__builtin_ia32_cvtsi2ss64((__v4sf)(__m128)(A), (long long)(B), \
- (int)(R)); })
+ (int)(R))
-#define _mm_cvt_roundi64_ss(A, B, R) __extension__ ({ \
+#define _mm_cvt_roundi64_ss(A, B, R) \
(__m128)__builtin_ia32_cvtsi2ss64((__v4sf)(__m128)(A), (long long)(B), \
- (int)(R)); })
+ (int)(R))
#endif
-#define _mm_cvt_roundss_sd(A, B, R) __extension__ ({ \
+#define _mm_cvt_roundss_sd(A, B, R) \
(__m128d)__builtin_ia32_cvtss2sd_round_mask((__v2df)(__m128d)(A), \
(__v4sf)(__m128)(B), \
(__v2df)_mm_undefined_pd(), \
- (__mmask8)-1, (int)(R)); })
+ (__mmask8)-1, (int)(R))
-#define _mm_mask_cvt_roundss_sd(W, U, A, B, R) __extension__ ({ \
+#define _mm_mask_cvt_roundss_sd(W, U, A, B, R) \
(__m128d)__builtin_ia32_cvtss2sd_round_mask((__v2df)(__m128d)(A), \
(__v4sf)(__m128)(B), \
(__v2df)(__m128d)(W), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-#define _mm_maskz_cvt_roundss_sd(U, A, B, R) __extension__ ({ \
+#define _mm_maskz_cvt_roundss_sd(U, A, B, R) \
(__m128d)__builtin_ia32_cvtss2sd_round_mask((__v2df)(__m128d)(A), \
(__v4sf)(__m128)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U), (int)(R)); })
+ (__mmask8)(U), (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_cvtss_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128 __B)
{
- return __builtin_ia32_cvtss2sd_round_mask((__v2df)(__A),
- (__v4sf)(__B),
- (__v2df)(__W),
- (__mmask8)(__U), _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_cvtss2sd_round_mask((__v2df)__A,
+ (__v4sf)__B,
+ (__v2df)__W,
+ (__mmask8)__U, _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_cvtss_sd (__mmask8 __U, __m128d __A, __m128 __B)
{
- return __builtin_ia32_cvtss2sd_round_mask((__v2df)(__A),
- (__v4sf)(__B),
- (__v2df)_mm_setzero_pd(),
- (__mmask8)(__U), _MM_FROUND_CUR_DIRECTION);
+ return __builtin_ia32_cvtss2sd_round_mask((__v2df)__A,
+ (__v4sf)__B,
+ (__v2df)_mm_setzero_pd(),
+ (__mmask8)__U, _MM_FROUND_CUR_DIRECTION);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_cvtu32_sd (__m128d __A, unsigned __B)
{
- return (__m128d) __builtin_ia32_cvtusi2sd32 ((__v2df) __A, __B);
+ __A[0] = __B;
+ return __A;
}
#ifdef __x86_64__
-#define _mm_cvt_roundu64_sd(A, B, R) __extension__ ({ \
+#define _mm_cvt_roundu64_sd(A, B, R) \
(__m128d)__builtin_ia32_cvtusi2sd64((__v2df)(__m128d)(A), \
- (unsigned long long)(B), (int)(R)); })
+ (unsigned long long)(B), (int)(R))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_cvtu64_sd (__m128d __A, unsigned long long __B)
{
- return (__m128d) __builtin_ia32_cvtusi2sd64 ((__v2df) __A, __B,
- _MM_FROUND_CUR_DIRECTION);
+ __A[0] = __B;
+ return __A;
}
#endif
-#define _mm_cvt_roundu32_ss(A, B, R) __extension__ ({ \
+#define _mm_cvt_roundu32_ss(A, B, R) \
(__m128)__builtin_ia32_cvtusi2ss32((__v4sf)(__m128)(A), (unsigned int)(B), \
- (int)(R)); })
+ (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_cvtu32_ss (__m128 __A, unsigned __B)
{
- return (__m128) __builtin_ia32_cvtusi2ss32 ((__v4sf) __A, __B,
- _MM_FROUND_CUR_DIRECTION);
+ __A[0] = __B;
+ return __A;
}
#ifdef __x86_64__
-#define _mm_cvt_roundu64_ss(A, B, R) __extension__ ({ \
+#define _mm_cvt_roundu64_ss(A, B, R) \
(__m128)__builtin_ia32_cvtusi2ss64((__v4sf)(__m128)(A), \
- (unsigned long long)(B), (int)(R)); })
+ (unsigned long long)(B), (int)(R))
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_cvtu64_ss (__m128 __A, unsigned long long __B)
{
- return (__m128) __builtin_ia32_cvtusi2ss64 ((__v4sf) __A, __B,
- _MM_FROUND_CUR_DIRECTION);
+ __A[0] = __B;
+ return __A;
}
#endif
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_set1_epi32 (__m512i __O, __mmask16 __M, int __A)
{
- return (__m512i) __builtin_ia32_pbroadcastd512_gpr_mask (__A, (__v16si) __O,
- __M);
+ return (__m512i) __builtin_ia32_selectd_512(__M,
+ (__v16si) _mm512_set1_epi32(__A),
+ (__v16si) __O);
}
-#ifdef __x86_64__
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_set1_epi64 (__m512i __O, __mmask8 __M, long long __A)
{
- return (__m512i) __builtin_ia32_pbroadcastq512_gpr_mask (__A, (__v8di) __O,
- __M);
+ return (__m512i) __builtin_ia32_selectq_512(__M,
+ (__v8di) _mm512_set1_epi64(__A),
+ (__v8di) __O);
}
-#endif
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set_epi8 (char __e63, char __e62, char __e61, char __e60, char __e59,
char __e58, char __e57, char __e56, char __e55, char __e54, char __e53,
char __e52, char __e51, char __e50, char __e49, char __e48, char __e47,
__e56, __e57, __e58, __e59, __e60, __e61, __e62, __e63};
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set_epi16(short __e31, short __e30, short __e29, short __e28,
short __e27, short __e26, short __e25, short __e24, short __e23,
short __e22, short __e21, short __e20, short __e19, short __e18,
__e24, __e25, __e26, __e27, __e28, __e29, __e30, __e31 };
}
-static __inline __m512i __DEFAULT_FN_ATTRS
+static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set_epi32 (int __A, int __B, int __C, int __D,
int __E, int __F, int __G, int __H,
int __I, int __J, int __K, int __L,
_mm512_set_epi32((e15),(e14),(e13),(e12),(e11),(e10),(e9),(e8),(e7),(e6), \
(e5),(e4),(e3),(e2),(e1),(e0))
-static __inline__ __m512i __DEFAULT_FN_ATTRS
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_set_epi64 (long long __A, long long __B, long long __C,
long long __D, long long __E, long long __F,
long long __G, long long __H)
#define _mm512_setr_epi64(e0,e1,e2,e3,e4,e5,e6,e7) \
_mm512_set_epi64((e7),(e6),(e5),(e4),(e3),(e2),(e1),(e0))
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_set_pd (double __A, double __B, double __C, double __D,
double __E, double __F, double __G, double __H)
{
#define _mm512_setr_pd(e0,e1,e2,e3,e4,e5,e6,e7) \
_mm512_set_pd((e7),(e6),(e5),(e4),(e3),(e2),(e1),(e0))
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_set_ps (float __A, float __B, float __C, float __D,
float __E, float __F, float __G, float __H,
float __I, float __J, float __K, float __L,
_mm512_set_ps((e15),(e14),(e13),(e12),(e11),(e10),(e9),(e8),(e7),(e6),(e5), \
(e4),(e3),(e2),(e1),(e0))
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_abs_ps(__m512 __A)
{
return (__m512)_mm512_and_epi32(_mm512_set1_epi32(0x7FFFFFFF),(__m512i)__A) ;
}
-static __inline__ __m512 __DEFAULT_FN_ATTRS
+static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_abs_ps(__m512 __W, __mmask16 __K, __m512 __A)
{
return (__m512)_mm512_mask_and_epi32((__m512i)__W, __K, _mm512_set1_epi32(0x7FFFFFFF),(__m512i)__A) ;
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_abs_pd(__m512d __A)
{
return (__m512d)_mm512_and_epi64(_mm512_set1_epi64(0x7FFFFFFFFFFFFFFF),(__v8di)__A) ;
}
-static __inline__ __m512d __DEFAULT_FN_ATTRS
+static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_abs_pd(__m512d __W, __mmask8 __K, __m512d __A)
{
return (__m512d)_mm512_mask_and_epi64((__v8di)__W, __K, _mm512_set1_epi64(0x7FFFFFFFFFFFFFFF),(__v8di)__A);
}
-// Vector-reduction arithmetic accepts vectors as inputs and produces scalars as
-// outputs. This class of vector operation forms the basis of many scientific
-// computations. In vector-reduction arithmetic, the evaluation off is
-// independent of the order of the input elements of V.
-
-// Used bisection method. At each step, we partition the vector with previous
-// step in half, and the operation is performed on its two halves.
-// This takes log2(n) steps where n is the number of elements in the vector.
-
-// Vec512 - Vector with size 512.
-// Operator - Can be one of following: +,*,&,|
-// T2 - Can get 'i' for int and 'f' for float.
-// T1 - Can get 'i' for int and 'd' for double.
-
-#define _mm512_reduce_operator_64bit(Vec512, Operator, T2, T1) \
- __extension__({ \
- __m256##T1 Vec256 = __builtin_shufflevector( \
- (__v8d##T2)Vec512, \
- (__v8d##T2)Vec512, \
- 0, 1, 2, 3) \
- Operator \
- __builtin_shufflevector( \
- (__v8d##T2)Vec512, \
- (__v8d##T2)Vec512, \
- 4, 5, 6, 7); \
- __m128##T1 Vec128 = __builtin_shufflevector( \
- (__v4d##T2)Vec256, \
- (__v4d##T2)Vec256, \
- 0, 1) \
- Operator \
- __builtin_shufflevector( \
- (__v4d##T2)Vec256, \
- (__v4d##T2)Vec256, \
- 2, 3); \
- Vec128 = __builtin_shufflevector((__v2d##T2)Vec128, \
- (__v2d##T2)Vec128, 0, -1) \
- Operator \
- __builtin_shufflevector((__v2d##T2)Vec128, \
- (__v2d##T2)Vec128, 1, -1); \
- return Vec128[0]; \
- })
-
-static __inline__ long long __DEFAULT_FN_ATTRS _mm512_reduce_add_epi64(__m512i __W) {
- _mm512_reduce_operator_64bit(__W, +, i, i);
-}
-
-static __inline__ long long __DEFAULT_FN_ATTRS _mm512_reduce_mul_epi64(__m512i __W) {
- _mm512_reduce_operator_64bit(__W, *, i, i);
-}
-
-static __inline__ long long __DEFAULT_FN_ATTRS _mm512_reduce_and_epi64(__m512i __W) {
- _mm512_reduce_operator_64bit(__W, &, i, i);
-}
-
-static __inline__ long long __DEFAULT_FN_ATTRS _mm512_reduce_or_epi64(__m512i __W) {
- _mm512_reduce_operator_64bit(__W, |, i, i);
-}
-
-static __inline__ double __DEFAULT_FN_ATTRS _mm512_reduce_add_pd(__m512d __W) {
- _mm512_reduce_operator_64bit(__W, +, f, d);
-}
-
-static __inline__ double __DEFAULT_FN_ATTRS _mm512_reduce_mul_pd(__m512d __W) {
- _mm512_reduce_operator_64bit(__W, *, f, d);
-}
-
-// Vec512 - Vector with size 512.
-// Vec512Neutral - All vector elements set to the identity element.
-// Identity element: {+,0},{*,1},{&,0xFFFFFFFFFFFFFFFF},{|,0}
-// Operator - Can be one of following: +,*,&,|
-// Mask - Intrinsic Mask
-// T2 - Can get 'i' for int and 'f' for float.
-// T1 - Can get 'i' for int and 'd' for packed double-precision.
-// T3 - Can be Pd for packed double or q for q-word.
-
-#define _mm512_mask_reduce_operator_64bit(Vec512, Vec512Neutral, Operator, \
- Mask, T2, T1, T3) \
- __extension__({ \
- Vec512 = __builtin_ia32_select##T3##_512( \
- (__mmask8)Mask, \
- (__v8d##T2)Vec512, \
- (__v8d##T2)Vec512Neutral); \
- _mm512_reduce_operator_64bit(Vec512, Operator, T2, T1); \
- })
-
-static __inline__ long long __DEFAULT_FN_ATTRS
+/* Vector-reduction arithmetic accepts vectors as inputs and produces scalars as
+ * outputs. This class of vector operation forms the basis of many scientific
+ * computations. In vector-reduction arithmetic, the evaluation off is
+ * independent of the order of the input elements of V.
+
+ * Used bisection method. At each step, we partition the vector with previous
+ * step in half, and the operation is performed on its two halves.
+ * This takes log2(n) steps where n is the number of elements in the vector.
+ */
+
+#define _mm512_mask_reduce_operator(op) \
+ __v4du __t1 = (__v4du)_mm512_extracti64x4_epi64(__W, 0); \
+ __v4du __t2 = (__v4du)_mm512_extracti64x4_epi64(__W, 1); \
+ __m256i __t3 = (__m256i)(__t1 op __t2); \
+ __v2du __t4 = (__v2du)_mm256_extracti128_si256(__t3, 0); \
+ __v2du __t5 = (__v2du)_mm256_extracti128_si256(__t3, 1); \
+ __v2du __t6 = __t4 op __t5; \
+ __v2du __t7 = __builtin_shufflevector(__t6, __t6, 1, 0); \
+ __v2du __t8 = __t6 op __t7; \
+ return __t8[0]
+
+static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_reduce_add_epi64(__m512i __W) {
+ _mm512_mask_reduce_operator(+);
+}
+
+static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_reduce_mul_epi64(__m512i __W) {
+ _mm512_mask_reduce_operator(*);
+}
+
+static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_reduce_and_epi64(__m512i __W) {
+ _mm512_mask_reduce_operator(&);
+}
+
+static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_reduce_or_epi64(__m512i __W) {
+ _mm512_mask_reduce_operator(|);
+}
+
+static __inline__ long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_add_epi64(__mmask8 __M, __m512i __W) {
- _mm512_mask_reduce_operator_64bit(__W, _mm512_set1_epi64(0), +, __M, i, i, q);
+ __W = _mm512_maskz_mov_epi64(__M, __W);
+ _mm512_mask_reduce_operator(+);
}
-static __inline__ long long __DEFAULT_FN_ATTRS
+static __inline__ long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_mul_epi64(__mmask8 __M, __m512i __W) {
- _mm512_mask_reduce_operator_64bit(__W, _mm512_set1_epi64(1), *, __M, i, i, q);
+ __W = _mm512_mask_mov_epi64(_mm512_set1_epi64(1), __M, __W);
+ _mm512_mask_reduce_operator(*);
}
-static __inline__ long long __DEFAULT_FN_ATTRS
+static __inline__ long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_and_epi64(__mmask8 __M, __m512i __W) {
- _mm512_mask_reduce_operator_64bit(__W, _mm512_set1_epi64(0xFFFFFFFFFFFFFFFF),
- &, __M, i, i, q);
+ __W = _mm512_mask_mov_epi64(_mm512_set1_epi64(~0ULL), __M, __W);
+ _mm512_mask_reduce_operator(&);
}
-static __inline__ long long __DEFAULT_FN_ATTRS
+static __inline__ long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_or_epi64(__mmask8 __M, __m512i __W) {
- _mm512_mask_reduce_operator_64bit(__W, _mm512_set1_epi64(0), |, __M,
- i, i, q);
+ __W = _mm512_maskz_mov_epi64(__M, __W);
+ _mm512_mask_reduce_operator(|);
}
+#undef _mm512_mask_reduce_operator
-static __inline__ double __DEFAULT_FN_ATTRS
+#define _mm512_mask_reduce_operator(op) \
+ __m256d __t1 = _mm512_extractf64x4_pd(__W, 0); \
+ __m256d __t2 = _mm512_extractf64x4_pd(__W, 1); \
+ __m256d __t3 = __t1 op __t2; \
+ __m128d __t4 = _mm256_extractf128_pd(__t3, 0); \
+ __m128d __t5 = _mm256_extractf128_pd(__t3, 1); \
+ __m128d __t6 = __t4 op __t5; \
+ __m128d __t7 = __builtin_shufflevector(__t6, __t6, 1, 0); \
+ __m128d __t8 = __t6 op __t7; \
+ return __t8[0]
+
+static __inline__ double __DEFAULT_FN_ATTRS512 _mm512_reduce_add_pd(__m512d __W) {
+ _mm512_mask_reduce_operator(+);
+}
+
+static __inline__ double __DEFAULT_FN_ATTRS512 _mm512_reduce_mul_pd(__m512d __W) {
+ _mm512_mask_reduce_operator(*);
+}
+
+static __inline__ double __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_add_pd(__mmask8 __M, __m512d __W) {
- _mm512_mask_reduce_operator_64bit(__W, _mm512_set1_pd(0), +, __M,
- f, d, pd);
+ __W = _mm512_maskz_mov_pd(__M, __W);
+ _mm512_mask_reduce_operator(+);
}
-static __inline__ double __DEFAULT_FN_ATTRS
+static __inline__ double __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_mul_pd(__mmask8 __M, __m512d __W) {
- _mm512_mask_reduce_operator_64bit(__W, _mm512_set1_pd(1), *, __M,
- f, d, pd);
-}
-
-// Vec512 - Vector with size 512.
-// Operator - Can be one of following: +,*,&,|
-// T2 - Can get 'i' for int and ' ' for packed single.
-// T1 - Can get 'i' for int and 'f' for float.
-
-#define _mm512_reduce_operator_32bit(Vec512, Operator, T2, T1) __extension__({ \
- __m256##T1 Vec256 = \
- (__m256##T1)(__builtin_shufflevector( \
- (__v16s##T2)Vec512, \
- (__v16s##T2)Vec512, \
- 0, 1, 2, 3, 4, 5, 6, 7) \
- Operator \
- __builtin_shufflevector( \
- (__v16s##T2)Vec512, \
- (__v16s##T2)Vec512, \
- 8, 9, 10, 11, 12, 13, 14, 15)); \
- __m128##T1 Vec128 = \
- (__m128##T1)(__builtin_shufflevector( \
- (__v8s##T2)Vec256, \
- (__v8s##T2)Vec256, \
- 0, 1, 2, 3) \
- Operator \
- __builtin_shufflevector( \
- (__v8s##T2)Vec256, \
- (__v8s##T2)Vec256, \
- 4, 5, 6, 7)); \
- Vec128 = (__m128##T1)(__builtin_shufflevector( \
- (__v4s##T2)Vec128, \
- (__v4s##T2)Vec128, \
- 0, 1, -1, -1) \
- Operator \
- __builtin_shufflevector( \
- (__v4s##T2)Vec128, \
- (__v4s##T2)Vec128, \
- 2, 3, -1, -1)); \
- Vec128 = (__m128##T1)(__builtin_shufflevector( \
- (__v4s##T2)Vec128, \
- (__v4s##T2)Vec128, \
- 0, -1, -1, -1) \
- Operator \
- __builtin_shufflevector( \
- (__v4s##T2)Vec128, \
- (__v4s##T2)Vec128, \
- 1, -1, -1, -1)); \
- return Vec128[0]; \
- })
-
-static __inline__ int __DEFAULT_FN_ATTRS
+ __W = _mm512_mask_mov_pd(_mm512_set1_pd(1.0), __M, __W);
+ _mm512_mask_reduce_operator(*);
+}
+#undef _mm512_mask_reduce_operator
+
+#define _mm512_mask_reduce_operator(op) \
+ __v8su __t1 = (__v8su)_mm512_extracti64x4_epi64(__W, 0); \
+ __v8su __t2 = (__v8su)_mm512_extracti64x4_epi64(__W, 1); \
+ __m256i __t3 = (__m256i)(__t1 op __t2); \
+ __v4su __t4 = (__v4su)_mm256_extracti128_si256(__t3, 0); \
+ __v4su __t5 = (__v4su)_mm256_extracti128_si256(__t3, 1); \
+ __v4su __t6 = __t4 op __t5; \
+ __v4su __t7 = __builtin_shufflevector(__t6, __t6, 2, 3, 0, 1); \
+ __v4su __t8 = __t6 op __t7; \
+ __v4su __t9 = __builtin_shufflevector(__t8, __t8, 1, 0, 3, 2); \
+ __v4su __t10 = __t8 op __t9; \
+ return __t10[0]
+
+static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_reduce_add_epi32(__m512i __W) {
- _mm512_reduce_operator_32bit(__W, +, i, i);
+ _mm512_mask_reduce_operator(+);
}
-static __inline__ int __DEFAULT_FN_ATTRS
+static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_reduce_mul_epi32(__m512i __W) {
- _mm512_reduce_operator_32bit(__W, *, i, i);
+ _mm512_mask_reduce_operator(*);
}
-static __inline__ int __DEFAULT_FN_ATTRS
+static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_reduce_and_epi32(__m512i __W) {
- _mm512_reduce_operator_32bit(__W, &, i, i);
+ _mm512_mask_reduce_operator(&);
}
-static __inline__ int __DEFAULT_FN_ATTRS
+static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_reduce_or_epi32(__m512i __W) {
- _mm512_reduce_operator_32bit(__W, |, i, i);
-}
-
-static __inline__ float __DEFAULT_FN_ATTRS
-_mm512_reduce_add_ps(__m512 __W) {
- _mm512_reduce_operator_32bit(__W, +, f, );
+ _mm512_mask_reduce_operator(|);
}
-static __inline__ float __DEFAULT_FN_ATTRS
-_mm512_reduce_mul_ps(__m512 __W) {
- _mm512_reduce_operator_32bit(__W, *, f, );
-}
-
-// Vec512 - Vector with size 512.
-// Vec512Neutral - All vector elements set to the identity element.
-// Identity element: {+,0},{*,1},{&,0xFFFFFFFF},{|,0}
-// Operator - Can be one of following: +,*,&,|
-// Mask - Intrinsic Mask
-// T2 - Can get 'i' for int and 'f' for float.
-// T1 - Can get 'i' for int and 'd' for double.
-// T3 - Can be Ps for packed single or d for d-word.
-
-#define _mm512_mask_reduce_operator_32bit(Vec512, Vec512Neutral, Operator, \
- Mask, T2, T1, T3) \
- __extension__({ \
- Vec512 = (__m512##T1)__builtin_ia32_select##T3##_512( \
- (__mmask16)Mask, \
- (__v16s##T2)Vec512, \
- (__v16s##T2)Vec512Neutral); \
- _mm512_reduce_operator_32bit(Vec512, Operator, T2, T1); \
- })
-
-static __inline__ int __DEFAULT_FN_ATTRS
+static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_add_epi32( __mmask16 __M, __m512i __W) {
- _mm512_mask_reduce_operator_32bit(__W, _mm512_set1_epi32(0), +, __M, i, i, d);
+ __W = _mm512_maskz_mov_epi32(__M, __W);
+ _mm512_mask_reduce_operator(+);
}
-static __inline__ int __DEFAULT_FN_ATTRS
+static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_mul_epi32( __mmask16 __M, __m512i __W) {
- _mm512_mask_reduce_operator_32bit(__W, _mm512_set1_epi32(1), *, __M, i, i, d);
+ __W = _mm512_mask_mov_epi32(_mm512_set1_epi32(1), __M, __W);
+ _mm512_mask_reduce_operator(*);
}
-static __inline__ int __DEFAULT_FN_ATTRS
+static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_and_epi32( __mmask16 __M, __m512i __W) {
- _mm512_mask_reduce_operator_32bit(__W, _mm512_set1_epi32(0xFFFFFFFF), &, __M,
- i, i, d);
+ __W = _mm512_mask_mov_epi32(_mm512_set1_epi32(~0U), __M, __W);
+ _mm512_mask_reduce_operator(&);
}
-static __inline__ int __DEFAULT_FN_ATTRS
+static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_or_epi32(__mmask16 __M, __m512i __W) {
- _mm512_mask_reduce_operator_32bit(__W, _mm512_set1_epi32(0), |, __M, i, i, d);
+ __W = _mm512_maskz_mov_epi32(__M, __W);
+ _mm512_mask_reduce_operator(|);
+}
+#undef _mm512_mask_reduce_operator
+
+#define _mm512_mask_reduce_operator(op) \
+ __m256 __t1 = (__m256)_mm512_extractf64x4_pd((__m512d)__W, 0); \
+ __m256 __t2 = (__m256)_mm512_extractf64x4_pd((__m512d)__W, 1); \
+ __m256 __t3 = __t1 op __t2; \
+ __m128 __t4 = _mm256_extractf128_ps(__t3, 0); \
+ __m128 __t5 = _mm256_extractf128_ps(__t3, 1); \
+ __m128 __t6 = __t4 op __t5; \
+ __m128 __t7 = __builtin_shufflevector(__t6, __t6, 2, 3, 0, 1); \
+ __m128 __t8 = __t6 op __t7; \
+ __m128 __t9 = __builtin_shufflevector(__t8, __t8, 1, 0, 3, 2); \
+ __m128 __t10 = __t8 op __t9; \
+ return __t10[0]
+
+static __inline__ float __DEFAULT_FN_ATTRS512
+_mm512_reduce_add_ps(__m512 __W) {
+ _mm512_mask_reduce_operator(+);
}
-static __inline__ float __DEFAULT_FN_ATTRS
+static __inline__ float __DEFAULT_FN_ATTRS512
+_mm512_reduce_mul_ps(__m512 __W) {
+ _mm512_mask_reduce_operator(*);
+}
+
+static __inline__ float __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_add_ps(__mmask16 __M, __m512 __W) {
- _mm512_mask_reduce_operator_32bit(__W, _mm512_set1_ps(0), +, __M, f, , ps);
+ __W = _mm512_maskz_mov_ps(__M, __W);
+ _mm512_mask_reduce_operator(+);
}
-static __inline__ float __DEFAULT_FN_ATTRS
+static __inline__ float __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_mul_ps(__mmask16 __M, __m512 __W) {
- _mm512_mask_reduce_operator_32bit(__W, _mm512_set1_ps(1), *, __M, f, , ps);
-}
-
-// Used bisection method. At each step, we partition the vector with previous
-// step in half, and the operation is performed on its two halves.
-// This takes log2(n) steps where n is the number of elements in the vector.
-// This macro uses only intrinsics from the AVX512F feature.
-
-// Vec512 - Vector with size of 512.
-// IntrinName - Can be one of following: {max|min}_{epi64|epu64|pd} for example:
-// __mm512_max_epi64
-// T1 - Can get 'i' for int and 'd' for double.[__m512{i|d}]
-// T2 - Can get 'i' for int and 'f' for float. [__v8d{i|f}]
-
-#define _mm512_reduce_maxMin_64bit(Vec512, IntrinName, T1, T2) __extension__({ \
- Vec512 = _mm512_##IntrinName( \
- (__m512##T1)__builtin_shufflevector( \
- (__v8d##T2)Vec512, \
- (__v8d##T2)Vec512, \
- 0, 1, 2, 3, -1, -1, -1, -1), \
- (__m512##T1)__builtin_shufflevector( \
- (__v8d##T2)Vec512, \
- (__v8d##T2)Vec512, \
- 4, 5, 6, 7, -1, -1, -1, -1)); \
- Vec512 = _mm512_##IntrinName( \
- (__m512##T1)__builtin_shufflevector( \
- (__v8d##T2)Vec512, \
- (__v8d##T2)Vec512, \
- 0, 1, -1, -1, -1, -1, -1, -1),\
- (__m512##T1)__builtin_shufflevector( \
- (__v8d##T2)Vec512, \
- (__v8d##T2)Vec512, \
- 2, 3, -1, -1, -1, -1, -1, \
- -1)); \
- Vec512 = _mm512_##IntrinName( \
- (__m512##T1)__builtin_shufflevector( \
- (__v8d##T2)Vec512, \
- (__v8d##T2)Vec512, \
- 0, -1, -1, -1, -1, -1, -1, -1),\
- (__m512##T1)__builtin_shufflevector( \
- (__v8d##T2)Vec512, \
- (__v8d##T2)Vec512, \
- 1, -1, -1, -1, -1, -1, -1, -1))\
- ; \
- return Vec512[0]; \
- })
-
-static __inline__ long long __DEFAULT_FN_ATTRS
-_mm512_reduce_max_epi64(__m512i __V) {
- _mm512_reduce_maxMin_64bit(__V, max_epi64, i, i);
+ __W = _mm512_mask_mov_ps(_mm512_set1_ps(1.0f), __M, __W);
+ _mm512_mask_reduce_operator(*);
}
+#undef _mm512_mask_reduce_operator
-static __inline__ unsigned long long __DEFAULT_FN_ATTRS
-_mm512_reduce_max_epu64(__m512i __V) {
- _mm512_reduce_maxMin_64bit(__V, max_epu64, i, i);
+#define _mm512_mask_reduce_operator(op) \
+ __m512i __t1 = (__m512i)__builtin_shufflevector((__v8di)__V, (__v8di)__V, 4, 5, 6, 7, 0, 1, 2, 3); \
+ __m512i __t2 = _mm512_##op(__V, __t1); \
+ __m512i __t3 = (__m512i)__builtin_shufflevector((__v8di)__t2, (__v8di)__t2, 2, 3, 0, 1, 6, 7, 4, 5); \
+ __m512i __t4 = _mm512_##op(__t2, __t3); \
+ __m512i __t5 = (__m512i)__builtin_shufflevector((__v8di)__t4, (__v8di)__t4, 1, 0, 3, 2, 5, 4, 7, 6); \
+ __v8di __t6 = (__v8di)_mm512_##op(__t4, __t5); \
+ return __t6[0]
+
+static __inline__ long long __DEFAULT_FN_ATTRS512
+_mm512_reduce_max_epi64(__m512i __V) {
+ _mm512_mask_reduce_operator(max_epi64);
}
-static __inline__ double __DEFAULT_FN_ATTRS
-_mm512_reduce_max_pd(__m512d __V) {
- _mm512_reduce_maxMin_64bit(__V, max_pd, d, f);
+static __inline__ unsigned long long __DEFAULT_FN_ATTRS512
+_mm512_reduce_max_epu64(__m512i __V) {
+ _mm512_mask_reduce_operator(max_epu64);
}
-static __inline__ long long __DEFAULT_FN_ATTRS _mm512_reduce_min_epi64
-(__m512i __V) {
- _mm512_reduce_maxMin_64bit(__V, min_epi64, i, i);
+static __inline__ long long __DEFAULT_FN_ATTRS512
+_mm512_reduce_min_epi64(__m512i __V) {
+ _mm512_mask_reduce_operator(min_epi64);
}
-static __inline__ unsigned long long __DEFAULT_FN_ATTRS
+static __inline__ unsigned long long __DEFAULT_FN_ATTRS512
_mm512_reduce_min_epu64(__m512i __V) {
- _mm512_reduce_maxMin_64bit(__V, min_epu64, i, i);
+ _mm512_mask_reduce_operator(min_epu64);
}
-static __inline__ double __DEFAULT_FN_ATTRS
-_mm512_reduce_min_pd(__m512d __V) {
- _mm512_reduce_maxMin_64bit(__V, min_pd, d, f);
-}
-
-// Vec512 - Vector with size 512.
-// Vec512Neutral - A 512 length vector with elements set to the identity element
-// Identity element: {max_epi,0x8000000000000000}
-// {max_epu,0x0000000000000000}
-// {max_pd, 0xFFF0000000000000}
-// {min_epi,0x7FFFFFFFFFFFFFFF}
-// {min_epu,0xFFFFFFFFFFFFFFFF}
-// {min_pd, 0x7FF0000000000000}
-//
-// IntrinName - Can be one of following: {max|min}_{epi64|epu64|pd} for example:
-// __mm512_max_epi64
-// T1 - Can get 'i' for int and 'd' for double.[__m512{i|d}]
-// T2 - Can get 'i' for int and 'f' for float. [__v8d{i|f}]
-// T3 - Can get 'q' q word and 'pd' for packed double.
-// [__builtin_ia32_select{q|pd}_512]
-// Mask - Intrinsic Mask
-
-#define _mm512_mask_reduce_maxMin_64bit(Vec512, Vec512Neutral, IntrinName, T1, \
- T2, T3, Mask) \
- __extension__({ \
- Vec512 = (__m512##T1)__builtin_ia32_select##T3##_512( \
- (__mmask8)Mask, \
- (__v8d##T2)Vec512, \
- (__v8d##T2)Vec512Neutral); \
- _mm512_reduce_maxMin_64bit(Vec512, IntrinName, T1, T2); \
- })
-
-static __inline__ long long __DEFAULT_FN_ATTRS
+static __inline__ long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_max_epi64(__mmask8 __M, __m512i __V) {
- _mm512_mask_reduce_maxMin_64bit(__V, _mm512_set1_epi64(0x8000000000000000),
- max_epi64, i, i, q, __M);
+ __V = _mm512_mask_mov_epi64(_mm512_set1_epi64(-__LONG_LONG_MAX__ - 1LL), __M, __V);
+ _mm512_mask_reduce_operator(max_epi64);
}
-static __inline__ unsigned long long __DEFAULT_FN_ATTRS
+static __inline__ unsigned long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_max_epu64(__mmask8 __M, __m512i __V) {
- _mm512_mask_reduce_maxMin_64bit(__V, _mm512_set1_epi64(0x0000000000000000),
- max_epu64, i, i, q, __M);
+ __V = _mm512_maskz_mov_epi64(__M, __V);
+ _mm512_mask_reduce_operator(max_epu64);
}
-static __inline__ double __DEFAULT_FN_ATTRS
-_mm512_mask_reduce_max_pd(__mmask8 __M, __m512d __V) {
- _mm512_mask_reduce_maxMin_64bit(__V, -_mm512_set1_pd(__builtin_inf()),
- max_pd, d, f, pd, __M);
-}
-
-static __inline__ long long __DEFAULT_FN_ATTRS
+static __inline__ long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_min_epi64(__mmask8 __M, __m512i __V) {
- _mm512_mask_reduce_maxMin_64bit(__V, _mm512_set1_epi64(0x7FFFFFFFFFFFFFFF),
- min_epi64, i, i, q, __M);
+ __V = _mm512_mask_mov_epi64(_mm512_set1_epi64(__LONG_LONG_MAX__), __M, __V);
+ _mm512_mask_reduce_operator(min_epi64);
}
-static __inline__ unsigned long long __DEFAULT_FN_ATTRS
+static __inline__ unsigned long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_min_epu64(__mmask8 __M, __m512i __V) {
- _mm512_mask_reduce_maxMin_64bit(__V, _mm512_set1_epi64(0xFFFFFFFFFFFFFFFF),
- min_epu64, i, i, q, __M);
+ __V = _mm512_mask_mov_epi64(_mm512_set1_epi64(~0ULL), __M, __V);
+ _mm512_mask_reduce_operator(min_epu64);
}
+#undef _mm512_mask_reduce_operator
-static __inline__ double __DEFAULT_FN_ATTRS
-_mm512_mask_reduce_min_pd(__mmask8 __M, __m512d __V) {
- _mm512_mask_reduce_maxMin_64bit(__V, _mm512_set1_pd(__builtin_inf()),
- min_pd, d, f, pd, __M);
-}
-
-// Vec512 - Vector with size 512.
-// IntrinName - Can be one of following: {max|min}_{epi32|epu32|ps} for example:
-// __mm512_max_epi32
-// T1 - Can get 'i' for int and ' ' .[__m512{i|}]
-// T2 - Can get 'i' for int and 'f' for float.[__v16s{i|f}]
-
-#define _mm512_reduce_maxMin_32bit(Vec512, IntrinName, T1, T2) __extension__({ \
- Vec512 = _mm512_##IntrinName( \
- (__m512##T1)__builtin_shufflevector( \
- (__v16s##T2)Vec512, \
- (__v16s##T2)Vec512, \
- 0, 1, 2, 3, 4, 5, 6, 7, \
- -1, -1, -1, -1, -1, -1, -1, -1), \
- (__m512##T1)__builtin_shufflevector( \
- (__v16s##T2)Vec512, \
- (__v16s##T2)Vec512, \
- 8, 9, 10, 11, 12, 13, 14, 15, \
- -1, -1, -1, -1, -1, -1, -1, -1)); \
- Vec512 = _mm512_##IntrinName( \
- (__m512##T1)__builtin_shufflevector( \
- (__v16s##T2)Vec512, \
- (__v16s##T2)Vec512, \
- 0, 1, 2, 3, -1, -1, -1, -1, \
- -1, -1, -1, -1, -1, -1, -1, -1), \
- (__m512##T1)__builtin_shufflevector( \
- (__v16s##T2)Vec512, \
- (__v16s##T2)Vec512, \
- 4, 5, 6, 7, -1, -1, -1, -1, \
- -1, -1, -1, -1, -1, -1, -1, -1)); \
- Vec512 = _mm512_##IntrinName( \
- (__m512##T1)__builtin_shufflevector( \
- (__v16s##T2)Vec512, \
- (__v16s##T2)Vec512, \
- 0, 1, -1, -1, -1, -1, -1, -1, \
- -1, -1, -1, -1, -1, -1, -1, -1), \
- (__m512##T1)__builtin_shufflevector( \
- (__v16s##T2)Vec512, \
- (__v16s##T2)Vec512, \
- 2, 3, -1, -1, -1, -1, -1, -1, \
- -1, -1, -1, -1, -1, -1, -1, -1)); \
- Vec512 = _mm512_##IntrinName( \
- (__m512##T1)__builtin_shufflevector( \
- (__v16s##T2)Vec512, \
- (__v16s##T2)Vec512, \
- 0, -1, -1, -1, -1, -1, -1, -1, \
- -1, -1, -1, -1, -1, -1, -1, -1), \
- (__m512##T1)__builtin_shufflevector( \
- (__v16s##T2)Vec512, \
- (__v16s##T2)Vec512, \
- 1, -1, -1, -1, -1, -1, -1, -1, \
- -1, -1, -1, -1, -1, -1, -1, -1)); \
- return Vec512[0]; \
- })
-
-static __inline__ int __DEFAULT_FN_ATTRS _mm512_reduce_max_epi32(__m512i a) {
- _mm512_reduce_maxMin_32bit(a, max_epi32, i, i);
-}
+#define _mm512_mask_reduce_operator(op) \
+ __m256i __t1 = _mm512_extracti64x4_epi64(__V, 0); \
+ __m256i __t2 = _mm512_extracti64x4_epi64(__V, 1); \
+ __m256i __t3 = _mm256_##op(__t1, __t2); \
+ __m128i __t4 = _mm256_extracti128_si256(__t3, 0); \
+ __m128i __t5 = _mm256_extracti128_si256(__t3, 1); \
+ __m128i __t6 = _mm_##op(__t4, __t5); \
+ __m128i __t7 = (__m128i)__builtin_shufflevector((__v4si)__t6, (__v4si)__t6, 2, 3, 0, 1); \
+ __m128i __t8 = _mm_##op(__t6, __t7); \
+ __m128i __t9 = (__m128i)__builtin_shufflevector((__v4si)__t8, (__v4si)__t8, 1, 0, 3, 2); \
+ __v4si __t10 = (__v4si)_mm_##op(__t8, __t9); \
+ return __t10[0]
-static __inline__ unsigned int __DEFAULT_FN_ATTRS
-_mm512_reduce_max_epu32(__m512i a) {
- _mm512_reduce_maxMin_32bit(a, max_epu32, i, i);
+static __inline__ int __DEFAULT_FN_ATTRS512
+_mm512_reduce_max_epi32(__m512i __V) {
+ _mm512_mask_reduce_operator(max_epi32);
}
-static __inline__ float __DEFAULT_FN_ATTRS _mm512_reduce_max_ps(__m512 a) {
- _mm512_reduce_maxMin_32bit(a, max_ps, , f);
+static __inline__ unsigned int __DEFAULT_FN_ATTRS512
+_mm512_reduce_max_epu32(__m512i __V) {
+ _mm512_mask_reduce_operator(max_epu32);
}
-static __inline__ int __DEFAULT_FN_ATTRS _mm512_reduce_min_epi32(__m512i a) {
- _mm512_reduce_maxMin_32bit(a, min_epi32, i, i);
+static __inline__ int __DEFAULT_FN_ATTRS512
+_mm512_reduce_min_epi32(__m512i __V) {
+ _mm512_mask_reduce_operator(min_epi32);
}
-static __inline__ unsigned int __DEFAULT_FN_ATTRS
-_mm512_reduce_min_epu32(__m512i a) {
- _mm512_reduce_maxMin_32bit(a, min_epu32, i, i);
-}
-
-static __inline__ float __DEFAULT_FN_ATTRS _mm512_reduce_min_ps(__m512 a) {
- _mm512_reduce_maxMin_32bit(a, min_ps, , f);
-}
-
-// Vec512 - Vector with size 512.
-// Vec512Neutral - A 512 length vector with elements set to the identity element
-// Identity element: {max_epi,0x80000000}
-// {max_epu,0x00000000}
-// {max_ps, 0xFF800000}
-// {min_epi,0x7FFFFFFF}
-// {min_epu,0xFFFFFFFF}
-// {min_ps, 0x7F800000}
-//
-// IntrinName - Can be one of following: {max|min}_{epi32|epu32|ps} for example:
-// __mm512_max_epi32
-// T1 - Can get 'i' for int and ' ' .[__m512{i|}]
-// T2 - Can get 'i' for int and 'f' for float.[__v16s{i|f}]
-// T3 - Can get 'q' q word and 'pd' for packed double.
-// [__builtin_ia32_select{q|pd}_512]
-// Mask - Intrinsic Mask
-
-#define _mm512_mask_reduce_maxMin_32bit(Vec512, Vec512Neutral, IntrinName, T1, \
- T2, T3, Mask) \
- __extension__({ \
- Vec512 = (__m512##T1)__builtin_ia32_select##T3##_512( \
- (__mmask16)Mask, \
- (__v16s##T2)Vec512, \
- (__v16s##T2)Vec512Neutral); \
- _mm512_reduce_maxMin_32bit(Vec512, IntrinName, T1, T2); \
- })
+static __inline__ unsigned int __DEFAULT_FN_ATTRS512
+_mm512_reduce_min_epu32(__m512i __V) {
+ _mm512_mask_reduce_operator(min_epu32);
+}
-static __inline__ int __DEFAULT_FN_ATTRS
+static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_max_epi32(__mmask16 __M, __m512i __V) {
- _mm512_mask_reduce_maxMin_32bit(__V, _mm512_set1_epi32(0x80000000), max_epi32,
- i, i, d, __M);
+ __V = _mm512_mask_mov_epi32(_mm512_set1_epi32(-__INT_MAX__ - 1), __M, __V);
+ _mm512_mask_reduce_operator(max_epi32);
}
-static __inline__ unsigned int __DEFAULT_FN_ATTRS
+static __inline__ unsigned int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_max_epu32(__mmask16 __M, __m512i __V) {
- _mm512_mask_reduce_maxMin_32bit(__V, _mm512_set1_epi32(0x00000000), max_epu32,
- i, i, d, __M);
+ __V = _mm512_maskz_mov_epi32(__M, __V);
+ _mm512_mask_reduce_operator(max_epu32);
}
-static __inline__ float __DEFAULT_FN_ATTRS
-_mm512_mask_reduce_max_ps(__mmask16 __M, __m512 __V) {
- _mm512_mask_reduce_maxMin_32bit(__V,-_mm512_set1_ps(__builtin_inff()), max_ps, , f,
- ps, __M);
-}
-
-static __inline__ int __DEFAULT_FN_ATTRS
+static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_min_epi32(__mmask16 __M, __m512i __V) {
- _mm512_mask_reduce_maxMin_32bit(__V, _mm512_set1_epi32(0x7FFFFFFF), min_epi32,
- i, i, d, __M);
+ __V = _mm512_mask_mov_epi32(_mm512_set1_epi32(__INT_MAX__), __M, __V);
+ _mm512_mask_reduce_operator(min_epi32);
}
-static __inline__ unsigned int __DEFAULT_FN_ATTRS
+static __inline__ unsigned int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_min_epu32(__mmask16 __M, __m512i __V) {
- _mm512_mask_reduce_maxMin_32bit(__V, _mm512_set1_epi32(0xFFFFFFFF), min_epu32,
- i, i, d, __M);
+ __V = _mm512_mask_mov_epi32(_mm512_set1_epi32(~0U), __M, __V);
+ _mm512_mask_reduce_operator(min_epu32);
+}
+#undef _mm512_mask_reduce_operator
+
+#define _mm512_mask_reduce_operator(op) \
+ __m256d __t1 = _mm512_extractf64x4_pd(__V, 0); \
+ __m256d __t2 = _mm512_extractf64x4_pd(__V, 1); \
+ __m256d __t3 = _mm256_##op(__t1, __t2); \
+ __m128d __t4 = _mm256_extractf128_pd(__t3, 0); \
+ __m128d __t5 = _mm256_extractf128_pd(__t3, 1); \
+ __m128d __t6 = _mm_##op(__t4, __t5); \
+ __m128d __t7 = __builtin_shufflevector(__t6, __t6, 1, 0); \
+ __m128d __t8 = _mm_##op(__t6, __t7); \
+ return __t8[0]
+
+static __inline__ double __DEFAULT_FN_ATTRS512
+_mm512_reduce_max_pd(__m512d __V) {
+ _mm512_mask_reduce_operator(max_pd);
+}
+
+static __inline__ double __DEFAULT_FN_ATTRS512
+_mm512_reduce_min_pd(__m512d __V) {
+ _mm512_mask_reduce_operator(min_pd);
}
-static __inline__ float __DEFAULT_FN_ATTRS
+static __inline__ double __DEFAULT_FN_ATTRS512
+_mm512_mask_reduce_max_pd(__mmask8 __M, __m512d __V) {
+ __V = _mm512_mask_mov_pd(_mm512_set1_pd(-__builtin_inf()), __M, __V);
+ _mm512_mask_reduce_operator(max_pd);
+}
+
+static __inline__ double __DEFAULT_FN_ATTRS512
+_mm512_mask_reduce_min_pd(__mmask8 __M, __m512d __V) {
+ __V = _mm512_mask_mov_pd(_mm512_set1_pd(__builtin_inf()), __M, __V);
+ _mm512_mask_reduce_operator(min_pd);
+}
+#undef _mm512_mask_reduce_operator
+
+#define _mm512_mask_reduce_operator(op) \
+ __m256 __t1 = (__m256)_mm512_extractf64x4_pd((__m512d)__V, 0); \
+ __m256 __t2 = (__m256)_mm512_extractf64x4_pd((__m512d)__V, 1); \
+ __m256 __t3 = _mm256_##op(__t1, __t2); \
+ __m128 __t4 = _mm256_extractf128_ps(__t3, 0); \
+ __m128 __t5 = _mm256_extractf128_ps(__t3, 1); \
+ __m128 __t6 = _mm_##op(__t4, __t5); \
+ __m128 __t7 = __builtin_shufflevector(__t6, __t6, 2, 3, 0, 1); \
+ __m128 __t8 = _mm_##op(__t6, __t7); \
+ __m128 __t9 = __builtin_shufflevector(__t8, __t8, 1, 0, 3, 2); \
+ __m128 __t10 = _mm_##op(__t8, __t9); \
+ return __t10[0]
+
+static __inline__ float __DEFAULT_FN_ATTRS512
+_mm512_reduce_max_ps(__m512 __V) {
+ _mm512_mask_reduce_operator(max_ps);
+}
+
+static __inline__ float __DEFAULT_FN_ATTRS512
+_mm512_reduce_min_ps(__m512 __V) {
+ _mm512_mask_reduce_operator(min_ps);
+}
+
+static __inline__ float __DEFAULT_FN_ATTRS512
+_mm512_mask_reduce_max_ps(__mmask16 __M, __m512 __V) {
+ __V = _mm512_mask_mov_ps(_mm512_set1_ps(-__builtin_inff()), __M, __V);
+ _mm512_mask_reduce_operator(max_ps);
+}
+
+static __inline__ float __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_min_ps(__mmask16 __M, __m512 __V) {
- _mm512_mask_reduce_maxMin_32bit(__V, _mm512_set1_ps(__builtin_inff()), min_ps, , f,
- ps, __M);
+ __V = _mm512_mask_mov_ps(_mm512_set1_ps(__builtin_inff()), __M, __V);
+ _mm512_mask_reduce_operator(min_ps);
}
+#undef _mm512_mask_reduce_operator
+/// Moves the least significant 32 bits of a vector of [16 x i32] to a
+/// 32-bit signed integer value.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VMOVD / MOVD </c> instruction.
+///
+/// \param __A
+/// A vector of [16 x i32]. The least significant 32 bits are moved to the
+/// destination.
+/// \returns A 32-bit signed integer containing the moved value.
+static __inline__ int __DEFAULT_FN_ATTRS512
+_mm512_cvtsi512_si32(__m512i __A) {
+ __v16si __b = (__v16si)__A;
+ return __b[0];
+}
+
+#undef __DEFAULT_FN_ATTRS512
+#undef __DEFAULT_FN_ATTRS128
#undef __DEFAULT_FN_ATTRS
-#endif // __AVX512FINTRIN_H
+#endif /* __AVX512FINTRIN_H */
--- /dev/null
+/*===------------- avx512ifmaintrin.h - IFMA intrinsics ------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512ifmaintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __IFMAINTRIN_H
+#define __IFMAINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512ifma"), __min_vector_width__(512)))
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_madd52hi_epu64 (__m512i __X, __m512i __Y, __m512i __Z)
+{
+ return (__m512i)__builtin_ia32_vpmadd52huq512((__v8di) __X, (__v8di) __Y,
+ (__v8di) __Z);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_madd52hi_epu64 (__m512i __W, __mmask8 __M, __m512i __X, __m512i __Y)
+{
+ return (__m512i)__builtin_ia32_selectq_512(__M,
+ (__v8di)_mm512_madd52hi_epu64(__W, __X, __Y),
+ (__v8di)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_madd52hi_epu64 (__mmask8 __M, __m512i __X, __m512i __Y, __m512i __Z)
+{
+ return (__m512i)__builtin_ia32_selectq_512(__M,
+ (__v8di)_mm512_madd52hi_epu64(__X, __Y, __Z),
+ (__v8di)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_madd52lo_epu64 (__m512i __X, __m512i __Y, __m512i __Z)
+{
+ return (__m512i)__builtin_ia32_vpmadd52luq512((__v8di) __X, (__v8di) __Y,
+ (__v8di) __Z);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_madd52lo_epu64 (__m512i __W, __mmask8 __M, __m512i __X, __m512i __Y)
+{
+ return (__m512i)__builtin_ia32_selectq_512(__M,
+ (__v8di)_mm512_madd52lo_epu64(__W, __X, __Y),
+ (__v8di)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_madd52lo_epu64 (__mmask8 __M, __m512i __X, __m512i __Y, __m512i __Z)
+{
+ return (__m512i)__builtin_ia32_selectq_512(__M,
+ (__v8di)_mm512_madd52lo_epu64(__X, __Y, __Z),
+ (__v8di)_mm512_setzero_si512());
+}
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif
--- /dev/null
+/*===------------- avx512ifmavlintrin.h - IFMA intrinsics ------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512ifmavlintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __IFMAVLINTRIN_H
+#define __IFMAVLINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx512ifma,avx512vl"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("avx512ifma,avx512vl"), __min_vector_width__(256)))
+
+
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_madd52hi_epu64 (__m128i __X, __m128i __Y, __m128i __Z)
+{
+ return (__m128i)__builtin_ia32_vpmadd52huq128((__v2di) __X, (__v2di) __Y,
+ (__v2di) __Z);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_madd52hi_epu64 (__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y)
+{
+ return (__m128i)__builtin_ia32_selectq_128(__M,
+ (__v2di)_mm_madd52hi_epu64(__W, __X, __Y),
+ (__v2di)__W);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_madd52hi_epu64 (__mmask8 __M, __m128i __X, __m128i __Y, __m128i __Z)
+{
+ return (__m128i)__builtin_ia32_selectq_128(__M,
+ (__v2di)_mm_madd52hi_epu64(__X, __Y, __Z),
+ (__v2di)_mm_setzero_si128());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_madd52hi_epu64 (__m256i __X, __m256i __Y, __m256i __Z)
+{
+ return (__m256i)__builtin_ia32_vpmadd52huq256((__v4di)__X, (__v4di)__Y,
+ (__v4di)__Z);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_madd52hi_epu64 (__m256i __W, __mmask8 __M, __m256i __X, __m256i __Y)
+{
+ return (__m256i)__builtin_ia32_selectq_256(__M,
+ (__v4di)_mm256_madd52hi_epu64(__W, __X, __Y),
+ (__v4di)__W);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_madd52hi_epu64 (__mmask8 __M, __m256i __X, __m256i __Y, __m256i __Z)
+{
+ return (__m256i)__builtin_ia32_selectq_256(__M,
+ (__v4di)_mm256_madd52hi_epu64(__X, __Y, __Z),
+ (__v4di)_mm256_setzero_si256());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_madd52lo_epu64 (__m128i __X, __m128i __Y, __m128i __Z)
+{
+ return (__m128i)__builtin_ia32_vpmadd52luq128((__v2di)__X, (__v2di)__Y,
+ (__v2di)__Z);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_madd52lo_epu64 (__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y)
+{
+ return (__m128i)__builtin_ia32_selectq_128(__M,
+ (__v2di)_mm_madd52lo_epu64(__W, __X, __Y),
+ (__v2di)__W);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_madd52lo_epu64 (__mmask8 __M, __m128i __X, __m128i __Y, __m128i __Z)
+{
+ return (__m128i)__builtin_ia32_selectq_128(__M,
+ (__v2di)_mm_madd52lo_epu64(__X, __Y, __Z),
+ (__v2di)_mm_setzero_si128());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_madd52lo_epu64 (__m256i __X, __m256i __Y, __m256i __Z)
+{
+ return (__m256i)__builtin_ia32_vpmadd52luq256((__v4di)__X, (__v4di)__Y,
+ (__v4di)__Z);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_madd52lo_epu64 (__m256i __W, __mmask8 __M, __m256i __X, __m256i __Y)
+{
+ return (__m256i)__builtin_ia32_selectq_256(__M,
+ (__v4di)_mm256_madd52lo_epu64(__W, __X, __Y),
+ (__v4di)__W);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_madd52lo_epu64 (__mmask8 __M, __m256i __X, __m256i __Y, __m256i __Z)
+{
+ return (__m256i)__builtin_ia32_selectq_256(__M,
+ (__v4di)_mm256_madd52lo_epu64(__X, __Y, __Z),
+ (__v4di)_mm256_setzero_si256());
+}
+
+
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
+
+#endif
--- /dev/null
+/*===------------- avx512pfintrin.h - PF intrinsics ------------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512pfintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __AVX512PFINTRIN_H
+#define __AVX512PFINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512pf")))
+
+#define _mm512_mask_prefetch_i32gather_pd(index, mask, addr, scale, hint) \
+ __builtin_ia32_gatherpfdpd((__mmask8)(mask), (__v8si)(__m256i)(index), \
+ (void const *)(addr), (int)(scale), \
+ (int)(hint))
+
+#define _mm512_prefetch_i32gather_pd(index, addr, scale, hint) \
+ __builtin_ia32_gatherpfdpd((__mmask8) -1, (__v8si)(__m256i)(index), \
+ (void const *)(addr), (int)(scale), \
+ (int)(hint))
+
+#define _mm512_mask_prefetch_i32gather_ps(index, mask, addr, scale, hint) \
+ __builtin_ia32_gatherpfdps((__mmask16)(mask), \
+ (__v16si)(__m512i)(index), (void const *)(addr), \
+ (int)(scale), (int)(hint))
+
+#define _mm512_prefetch_i32gather_ps(index, addr, scale, hint) \
+ __builtin_ia32_gatherpfdps((__mmask16) -1, \
+ (__v16si)(__m512i)(index), (void const *)(addr), \
+ (int)(scale), (int)(hint))
+
+#define _mm512_mask_prefetch_i64gather_pd(index, mask, addr, scale, hint) \
+ __builtin_ia32_gatherpfqpd((__mmask8)(mask), (__v8di)(__m512i)(index), \
+ (void const *)(addr), (int)(scale), \
+ (int)(hint))
+
+#define _mm512_prefetch_i64gather_pd(index, addr, scale, hint) \
+ __builtin_ia32_gatherpfqpd((__mmask8) -1, (__v8di)(__m512i)(index), \
+ (void const *)(addr), (int)(scale), \
+ (int)(hint))
+
+#define _mm512_mask_prefetch_i64gather_ps(index, mask, addr, scale, hint) \
+ __builtin_ia32_gatherpfqps((__mmask8)(mask), (__v8di)(__m512i)(index), \
+ (void const *)(addr), (int)(scale), (int)(hint))
+
+#define _mm512_prefetch_i64gather_ps(index, addr, scale, hint) \
+ __builtin_ia32_gatherpfqps((__mmask8) -1, (__v8di)(__m512i)(index), \
+ (void const *)(addr), (int)(scale), (int)(hint))
+
+#define _mm512_prefetch_i32scatter_pd(addr, index, scale, hint) \
+ __builtin_ia32_scatterpfdpd((__mmask8)-1, (__v8si)(__m256i)(index), \
+ (void *)(addr), (int)(scale), \
+ (int)(hint))
+
+#define _mm512_mask_prefetch_i32scatter_pd(addr, mask, index, scale, hint) \
+ __builtin_ia32_scatterpfdpd((__mmask8)(mask), (__v8si)(__m256i)(index), \
+ (void *)(addr), (int)(scale), \
+ (int)(hint))
+
+#define _mm512_prefetch_i32scatter_ps(addr, index, scale, hint) \
+ __builtin_ia32_scatterpfdps((__mmask16)-1, (__v16si)(__m512i)(index), \
+ (void *)(addr), (int)(scale), (int)(hint))
+
+#define _mm512_mask_prefetch_i32scatter_ps(addr, mask, index, scale, hint) \
+ __builtin_ia32_scatterpfdps((__mmask16)(mask), \
+ (__v16si)(__m512i)(index), (void *)(addr), \
+ (int)(scale), (int)(hint))
+
+#define _mm512_prefetch_i64scatter_pd(addr, index, scale, hint) \
+ __builtin_ia32_scatterpfqpd((__mmask8)-1, (__v8di)(__m512i)(index), \
+ (void *)(addr), (int)(scale), \
+ (int)(hint))
+
+#define _mm512_mask_prefetch_i64scatter_pd(addr, mask, index, scale, hint) \
+ __builtin_ia32_scatterpfqpd((__mmask8)(mask), (__v8di)(__m512i)(index), \
+ (void *)(addr), (int)(scale), \
+ (int)(hint))
+
+#define _mm512_prefetch_i64scatter_ps(addr, index, scale, hint) \
+ __builtin_ia32_scatterpfqps((__mmask8)-1, (__v8di)(__m512i)(index), \
+ (void *)(addr), (int)(scale), (int)(hint))
+
+#define _mm512_mask_prefetch_i64scatter_ps(addr, mask, index, scale, hint) \
+ __builtin_ia32_scatterpfqps((__mmask8)(mask), (__v8di)(__m512i)(index), \
+ (void *)(addr), (int)(scale), (int)(hint))
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif
--- /dev/null
+/*===------------- avx512vbmi2intrin.h - VBMI2 intrinsics ------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512vbmi2intrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __AVX512VBMI2INTRIN_H
+#define __AVX512VBMI2INTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512vbmi2"), __min_vector_width__(512)))
+
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_compress_epi16(__m512i __S, __mmask32 __U, __m512i __D)
+{
+ return (__m512i) __builtin_ia32_compresshi512_mask ((__v32hi) __D,
+ (__v32hi) __S,
+ __U);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_compress_epi16(__mmask32 __U, __m512i __D)
+{
+ return (__m512i) __builtin_ia32_compresshi512_mask ((__v32hi) __D,
+ (__v32hi) _mm512_setzero_si512(),
+ __U);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_compress_epi8(__m512i __S, __mmask64 __U, __m512i __D)
+{
+ return (__m512i) __builtin_ia32_compressqi512_mask ((__v64qi) __D,
+ (__v64qi) __S,
+ __U);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_compress_epi8(__mmask64 __U, __m512i __D)
+{
+ return (__m512i) __builtin_ia32_compressqi512_mask ((__v64qi) __D,
+ (__v64qi) _mm512_setzero_si512(),
+ __U);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_mm512_mask_compressstoreu_epi16(void *__P, __mmask32 __U, __m512i __D)
+{
+ __builtin_ia32_compressstorehi512_mask ((__v32hi *) __P, (__v32hi) __D,
+ __U);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_mm512_mask_compressstoreu_epi8(void *__P, __mmask64 __U, __m512i __D)
+{
+ __builtin_ia32_compressstoreqi512_mask ((__v64qi *) __P, (__v64qi) __D,
+ __U);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_expand_epi16(__m512i __S, __mmask32 __U, __m512i __D)
+{
+ return (__m512i) __builtin_ia32_expandhi512_mask ((__v32hi) __D,
+ (__v32hi) __S,
+ __U);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_expand_epi16(__mmask32 __U, __m512i __D)
+{
+ return (__m512i) __builtin_ia32_expandhi512_mask ((__v32hi) __D,
+ (__v32hi) _mm512_setzero_si512(),
+ __U);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_expand_epi8(__m512i __S, __mmask64 __U, __m512i __D)
+{
+ return (__m512i) __builtin_ia32_expandqi512_mask ((__v64qi) __D,
+ (__v64qi) __S,
+ __U);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_expand_epi8(__mmask64 __U, __m512i __D)
+{
+ return (__m512i) __builtin_ia32_expandqi512_mask ((__v64qi) __D,
+ (__v64qi) _mm512_setzero_si512(),
+ __U);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_expandloadu_epi16(__m512i __S, __mmask32 __U, void const *__P)
+{
+ return (__m512i) __builtin_ia32_expandloadhi512_mask ((const __v32hi *)__P,
+ (__v32hi) __S,
+ __U);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_expandloadu_epi16(__mmask32 __U, void const *__P)
+{
+ return (__m512i) __builtin_ia32_expandloadhi512_mask ((const __v32hi *)__P,
+ (__v32hi) _mm512_setzero_si512(),
+ __U);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_expandloadu_epi8(__m512i __S, __mmask64 __U, void const *__P)
+{
+ return (__m512i) __builtin_ia32_expandloadqi512_mask ((const __v64qi *)__P,
+ (__v64qi) __S,
+ __U);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_expandloadu_epi8(__mmask64 __U, void const *__P)
+{
+ return (__m512i) __builtin_ia32_expandloadqi512_mask ((const __v64qi *)__P,
+ (__v64qi) _mm512_setzero_si512(),
+ __U);
+}
+
+#define _mm512_shldi_epi64(A, B, I) \
+ (__m512i)__builtin_ia32_vpshldq512((__v8di)(__m512i)(A), \
+ (__v8di)(__m512i)(B), (int)(I))
+
+#define _mm512_mask_shldi_epi64(S, U, A, B, I) \
+ (__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
+ (__v8di)_mm512_shldi_epi64((A), (B), (I)), \
+ (__v8di)(__m512i)(S))
+
+#define _mm512_maskz_shldi_epi64(U, A, B, I) \
+ (__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
+ (__v8di)_mm512_shldi_epi64((A), (B), (I)), \
+ (__v8di)_mm512_setzero_si512())
+
+#define _mm512_shldi_epi32(A, B, I) \
+ (__m512i)__builtin_ia32_vpshldd512((__v16si)(__m512i)(A), \
+ (__v16si)(__m512i)(B), (int)(I))
+
+#define _mm512_mask_shldi_epi32(S, U, A, B, I) \
+ (__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
+ (__v16si)_mm512_shldi_epi32((A), (B), (I)), \
+ (__v16si)(__m512i)(S))
+
+#define _mm512_maskz_shldi_epi32(U, A, B, I) \
+ (__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
+ (__v16si)_mm512_shldi_epi32((A), (B), (I)), \
+ (__v16si)_mm512_setzero_si512())
+
+#define _mm512_shldi_epi16(A, B, I) \
+ (__m512i)__builtin_ia32_vpshldw512((__v32hi)(__m512i)(A), \
+ (__v32hi)(__m512i)(B), (int)(I))
+
+#define _mm512_mask_shldi_epi16(S, U, A, B, I) \
+ (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
+ (__v32hi)_mm512_shldi_epi16((A), (B), (I)), \
+ (__v32hi)(__m512i)(S))
+
+#define _mm512_maskz_shldi_epi16(U, A, B, I) \
+ (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
+ (__v32hi)_mm512_shldi_epi16((A), (B), (I)), \
+ (__v32hi)_mm512_setzero_si512())
+
+#define _mm512_shrdi_epi64(A, B, I) \
+ (__m512i)__builtin_ia32_vpshrdq512((__v8di)(__m512i)(A), \
+ (__v8di)(__m512i)(B), (int)(I))
+
+#define _mm512_mask_shrdi_epi64(S, U, A, B, I) \
+ (__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
+ (__v8di)_mm512_shrdi_epi64((A), (B), (I)), \
+ (__v8di)(__m512i)(S))
+
+#define _mm512_maskz_shrdi_epi64(U, A, B, I) \
+ (__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
+ (__v8di)_mm512_shrdi_epi64((A), (B), (I)), \
+ (__v8di)_mm512_setzero_si512())
+
+#define _mm512_shrdi_epi32(A, B, I) \
+ (__m512i)__builtin_ia32_vpshrdd512((__v16si)(__m512i)(A), \
+ (__v16si)(__m512i)(B), (int)(I))
+
+#define _mm512_mask_shrdi_epi32(S, U, A, B, I) \
+ (__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
+ (__v16si)_mm512_shrdi_epi32((A), (B), (I)), \
+ (__v16si)(__m512i)(S))
+
+#define _mm512_maskz_shrdi_epi32(U, A, B, I) \
+ (__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
+ (__v16si)_mm512_shrdi_epi32((A), (B), (I)), \
+ (__v16si)_mm512_setzero_si512())
+
+#define _mm512_shrdi_epi16(A, B, I) \
+ (__m512i)__builtin_ia32_vpshrdw512((__v32hi)(__m512i)(A), \
+ (__v32hi)(__m512i)(B), (int)(I))
+
+#define _mm512_mask_shrdi_epi16(S, U, A, B, I) \
+ (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
+ (__v32hi)_mm512_shrdi_epi16((A), (B), (I)), \
+ (__v32hi)(__m512i)(S))
+
+#define _mm512_maskz_shrdi_epi16(U, A, B, I) \
+ (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
+ (__v32hi)_mm512_shrdi_epi16((A), (B), (I)), \
+ (__v32hi)_mm512_setzero_si512())
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_shldv_epi64(__m512i __A, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_vpshldvq512((__v8di)__A, (__v8di)__B,
+ (__v8di)__C);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_shldv_epi64(__m512i __A, __mmask8 __U, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_selectq_512(__U,
+ (__v8di)_mm512_shldv_epi64(__A, __B, __C),
+ (__v8di)__A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_shldv_epi64(__mmask8 __U, __m512i __A, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_selectq_512(__U,
+ (__v8di)_mm512_shldv_epi64(__A, __B, __C),
+ (__v8di)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_shldv_epi32(__m512i __A, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_vpshldvd512((__v16si)__A, (__v16si)__B,
+ (__v16si)__C);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_shldv_epi32(__m512i __A, __mmask16 __U, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_shldv_epi32(__A, __B, __C),
+ (__v16si)__A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_shldv_epi32(__mmask16 __U, __m512i __A, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_shldv_epi32(__A, __B, __C),
+ (__v16si)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_shldv_epi16(__m512i __A, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_vpshldvw512((__v32hi)__A, (__v32hi)__B,
+ (__v32hi)__C);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_shldv_epi16(__m512i __A, __mmask32 __U, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_selectw_512(__U,
+ (__v32hi)_mm512_shldv_epi16(__A, __B, __C),
+ (__v32hi)__A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_shldv_epi16(__mmask32 __U, __m512i __A, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_selectw_512(__U,
+ (__v32hi)_mm512_shldv_epi16(__A, __B, __C),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_shrdv_epi64(__m512i __A, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_vpshrdvq512((__v8di)__A, (__v8di)__B,
+ (__v8di)__C);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_shrdv_epi64(__m512i __A, __mmask8 __U, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_selectq_512(__U,
+ (__v8di)_mm512_shrdv_epi64(__A, __B, __C),
+ (__v8di)__A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_shrdv_epi64(__mmask8 __U, __m512i __A, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_selectq_512(__U,
+ (__v8di)_mm512_shrdv_epi64(__A, __B, __C),
+ (__v8di)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_shrdv_epi32(__m512i __A, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_vpshrdvd512((__v16si)__A, (__v16si)__B,
+ (__v16si)__C);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_shrdv_epi32(__m512i __A, __mmask16 __U, __m512i __B, __m512i __C)
+{
+ return (__m512i) __builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_shrdv_epi32(__A, __B, __C),
+ (__v16si)__A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_shrdv_epi32(__mmask16 __U, __m512i __A, __m512i __B, __m512i __C)
+{
+ return (__m512i) __builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_shrdv_epi32(__A, __B, __C),
+ (__v16si)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_shrdv_epi16(__m512i __A, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_vpshrdvw512((__v32hi)__A, (__v32hi)__B,
+ (__v32hi)__C);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_shrdv_epi16(__m512i __A, __mmask32 __U, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_selectw_512(__U,
+ (__v32hi)_mm512_shrdv_epi16(__A, __B, __C),
+ (__v32hi)__A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_shrdv_epi16(__mmask32 __U, __m512i __A, __m512i __B, __m512i __C)
+{
+ return (__m512i)__builtin_ia32_selectw_512(__U,
+ (__v32hi)_mm512_shrdv_epi16(__A, __B, __C),
+ (__v32hi)_mm512_setzero_si512());
+}
+
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif
+
--- /dev/null
+/*===------------- avx512vbmiintrin.h - VBMI intrinsics ------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512vbmiintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __VBMIINTRIN_H
+#define __VBMIINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512vbmi"), __min_vector_width__(512)))
+
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_permutex2var_epi8(__m512i __A, __m512i __I, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_vpermi2varqi512((__v64qi)__A, (__v64qi)__I,
+ (__v64qi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_permutex2var_epi8(__m512i __A, __mmask64 __U, __m512i __I,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512(__U,
+ (__v64qi)_mm512_permutex2var_epi8(__A, __I, __B),
+ (__v64qi)__A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask2_permutex2var_epi8(__m512i __A, __m512i __I, __mmask64 __U,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512(__U,
+ (__v64qi)_mm512_permutex2var_epi8(__A, __I, __B),
+ (__v64qi)__I);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_permutex2var_epi8(__mmask64 __U, __m512i __A, __m512i __I,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512(__U,
+ (__v64qi)_mm512_permutex2var_epi8(__A, __I, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_permutexvar_epi8 (__m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_permvarqi512((__v64qi) __B, (__v64qi) __A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_permutexvar_epi8 (__mmask64 __M, __m512i __A,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_permutexvar_epi8(__A, __B),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_permutexvar_epi8 (__m512i __W, __mmask64 __M, __m512i __A,
+ __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_permutexvar_epi8(__A, __B),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_multishift_epi64_epi8(__m512i __X, __m512i __Y)
+{
+ return (__m512i)__builtin_ia32_vpmultishiftqb512((__v64qi)__X, (__v64qi) __Y);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_multishift_epi64_epi8(__m512i __W, __mmask64 __M, __m512i __X,
+ __m512i __Y)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_multishift_epi64_epi8(__X, __Y),
+ (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_multishift_epi64_epi8(__mmask64 __M, __m512i __X, __m512i __Y)
+{
+ return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
+ (__v64qi)_mm512_multishift_epi64_epi8(__X, __Y),
+ (__v64qi)_mm512_setzero_si512());
+}
+
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif
--- /dev/null
+/*===------------- avx512vbmivlintrin.h - VBMI intrinsics ------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512vbmivlintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __VBMIVLINTRIN_H
+#define __VBMIVLINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx512vbmi,avx512vl"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("avx512vbmi,avx512vl"), __min_vector_width__(256)))
+
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_permutex2var_epi8(__m128i __A, __m128i __I, __m128i __B)
+{
+ return (__m128i)__builtin_ia32_vpermi2varqi128((__v16qi)__A,
+ (__v16qi)__I,
+ (__v16qi)__B);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_permutex2var_epi8(__m128i __A, __mmask16 __U, __m128i __I,
+ __m128i __B)
+{
+ return (__m128i)__builtin_ia32_selectb_128(__U,
+ (__v16qi)_mm_permutex2var_epi8(__A, __I, __B),
+ (__v16qi)__A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask2_permutex2var_epi8(__m128i __A, __m128i __I, __mmask16 __U,
+ __m128i __B)
+{
+ return (__m128i)__builtin_ia32_selectb_128(__U,
+ (__v16qi)_mm_permutex2var_epi8(__A, __I, __B),
+ (__v16qi)__I);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_permutex2var_epi8(__mmask16 __U, __m128i __A, __m128i __I,
+ __m128i __B)
+{
+ return (__m128i)__builtin_ia32_selectb_128(__U,
+ (__v16qi)_mm_permutex2var_epi8(__A, __I, __B),
+ (__v16qi)_mm_setzero_si128());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_permutex2var_epi8(__m256i __A, __m256i __I, __m256i __B)
+{
+ return (__m256i)__builtin_ia32_vpermi2varqi256((__v32qi)__A, (__v32qi)__I,
+ (__v32qi)__B);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_permutex2var_epi8(__m256i __A, __mmask32 __U, __m256i __I,
+ __m256i __B)
+{
+ return (__m256i)__builtin_ia32_selectb_256(__U,
+ (__v32qi)_mm256_permutex2var_epi8(__A, __I, __B),
+ (__v32qi)__A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask2_permutex2var_epi8(__m256i __A, __m256i __I, __mmask32 __U,
+ __m256i __B)
+{
+ return (__m256i)__builtin_ia32_selectb_256(__U,
+ (__v32qi)_mm256_permutex2var_epi8(__A, __I, __B),
+ (__v32qi)__I);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_permutex2var_epi8(__mmask32 __U, __m256i __A, __m256i __I,
+ __m256i __B)
+{
+ return (__m256i)__builtin_ia32_selectb_256(__U,
+ (__v32qi)_mm256_permutex2var_epi8(__A, __I, __B),
+ (__v32qi)_mm256_setzero_si256());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_permutexvar_epi8 (__m128i __A, __m128i __B)
+{
+ return (__m128i)__builtin_ia32_permvarqi128((__v16qi)__B, (__v16qi)__A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_permutexvar_epi8 (__mmask16 __M, __m128i __A, __m128i __B)
+{
+ return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
+ (__v16qi)_mm_permutexvar_epi8(__A, __B),
+ (__v16qi)_mm_setzero_si128());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_permutexvar_epi8 (__m128i __W, __mmask16 __M, __m128i __A,
+ __m128i __B)
+{
+ return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
+ (__v16qi)_mm_permutexvar_epi8(__A, __B),
+ (__v16qi)__W);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_permutexvar_epi8 (__m256i __A, __m256i __B)
+{
+ return (__m256i)__builtin_ia32_permvarqi256((__v32qi) __B, (__v32qi) __A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_permutexvar_epi8 (__mmask32 __M, __m256i __A,
+ __m256i __B)
+{
+ return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
+ (__v32qi)_mm256_permutexvar_epi8(__A, __B),
+ (__v32qi)_mm256_setzero_si256());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_permutexvar_epi8 (__m256i __W, __mmask32 __M, __m256i __A,
+ __m256i __B)
+{
+ return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
+ (__v32qi)_mm256_permutexvar_epi8(__A, __B),
+ (__v32qi)__W);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_multishift_epi64_epi8(__m128i __X, __m128i __Y)
+{
+ return (__m128i)__builtin_ia32_vpmultishiftqb128((__v16qi)__X, (__v16qi)__Y);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_multishift_epi64_epi8(__m128i __W, __mmask16 __M, __m128i __X,
+ __m128i __Y)
+{
+ return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
+ (__v16qi)_mm_multishift_epi64_epi8(__X, __Y),
+ (__v16qi)__W);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_multishift_epi64_epi8(__mmask16 __M, __m128i __X, __m128i __Y)
+{
+ return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
+ (__v16qi)_mm_multishift_epi64_epi8(__X, __Y),
+ (__v16qi)_mm_setzero_si128());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_multishift_epi64_epi8(__m256i __X, __m256i __Y)
+{
+ return (__m256i)__builtin_ia32_vpmultishiftqb256((__v32qi)__X, (__v32qi)__Y);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_multishift_epi64_epi8(__m256i __W, __mmask32 __M, __m256i __X,
+ __m256i __Y)
+{
+ return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
+ (__v32qi)_mm256_multishift_epi64_epi8(__X, __Y),
+ (__v32qi)__W);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_multishift_epi64_epi8(__mmask32 __M, __m256i __X, __m256i __Y)
+{
+ return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
+ (__v32qi)_mm256_multishift_epi64_epi8(__X, __Y),
+ (__v32qi)_mm256_setzero_si256());
+}
+
+
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
+
+#endif
--- /dev/null
+/*===--------- avx512vlbf16intrin.h - AVX512_BF16 intrinsics ---------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512vlbf16intrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __AVX512VLBF16INTRIN_H
+#define __AVX512VLBF16INTRIN_H
+
+typedef short __m128bh __attribute__((__vector_size__(16), __aligned__(16)));
+
+#define __DEFAULT_FN_ATTRS128 \
+ __attribute__((__always_inline__, __nodebug__, \
+ __target__("avx512vl, avx512bf16"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 \
+ __attribute__((__always_inline__, __nodebug__, \
+ __target__("avx512vl, avx512bf16"), __min_vector_width__(256)))
+
+/// Convert Two Packed Single Data to One Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNE2PS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 128-bit vector of [4 x float].
+/// \param __B
+/// A 128-bit vector of [4 x float].
+/// \returns A 128-bit vector of [8 x bfloat] whose lower 64 bits come from
+/// conversion of __B, and higher 64 bits come from conversion of __A.
+static __inline__ __m128bh __DEFAULT_FN_ATTRS128
+_mm_cvtne2ps_pbh(__m128 __A, __m128 __B) {
+ return (__m128bh)__builtin_ia32_cvtne2ps2bf16_128((__v4sf) __A,
+ (__v4sf) __B);
+}
+
+/// Convert Two Packed Single Data to One Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNE2PS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 128-bit vector of [4 x float].
+/// \param __B
+/// A 128-bit vector of [4 x float].
+/// \param __W
+/// A 128-bit vector of [8 x bfloat].
+/// \param __U
+/// A 8-bit mask value specifying what is chosen for each element.
+/// A 1 means conversion of __A or __B. A 0 means element from __W.
+/// \returns A 128-bit vector of [8 x bfloat] whose lower 64 bits come from
+/// conversion of __B, and higher 64 bits come from conversion of __A.
+static __inline__ __m128bh __DEFAULT_FN_ATTRS128
+_mm_mask_cvtne2ps_pbh(__m128bh __W, __mmask8 __U, __m128 __A, __m128 __B) {
+ return (__m128bh)__builtin_ia32_selectw_128((__mmask8)__U,
+ (__v8hi)_mm_cvtne2ps_pbh(__A, __B),
+ (__v8hi)__W);
+}
+
+/// Convert Two Packed Single Data to One Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNE2PS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 128-bit vector of [4 x float].
+/// \param __B
+/// A 128-bit vector of [4 x float].
+/// \param __U
+/// A 8-bit mask value specifying what is chosen for each element.
+/// A 1 means conversion of __A or __B. A 0 means element is zero.
+/// \returns A 128-bit vector of [8 x bfloat] whose lower 64 bits come from
+/// conversion of __B, and higher 64 bits come from conversion of __A.
+static __inline__ __m128bh __DEFAULT_FN_ATTRS128
+_mm_maskz_cvtne2ps_pbh(__mmask8 __U, __m128 __A, __m128 __B) {
+ return (__m128bh)__builtin_ia32_selectw_128((__mmask8)__U,
+ (__v8hi)_mm_cvtne2ps_pbh(__A, __B),
+ (__v8hi)_mm_setzero_si128());
+}
+
+/// Convert Two Packed Single Data to One Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNE2PS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 256-bit vector of [8 x float].
+/// \param __B
+/// A 256-bit vector of [8 x float].
+/// \returns A 256-bit vector of [16 x bfloat] whose lower 128 bits come from
+/// conversion of __B, and higher 128 bits come from conversion of __A.
+static __inline__ __m256bh __DEFAULT_FN_ATTRS256
+_mm256_cvtne2ps_pbh(__m256 __A, __m256 __B) {
+ return (__m256bh)__builtin_ia32_cvtne2ps2bf16_256((__v8sf) __A,
+ (__v8sf) __B);
+}
+
+/// Convert Two Packed Single Data to One Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNE2PS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 256-bit vector of [8 x float].
+/// \param __B
+/// A 256-bit vector of [8 x float].
+/// \param __W
+/// A 256-bit vector of [16 x bfloat].
+/// \param __U
+/// A 16-bit mask value specifying what is chosen for each element.
+/// A 1 means conversion of __A or __B. A 0 means element from __W.
+/// \returns A 256-bit vector of [16 x bfloat] whose lower 128 bits come from
+/// conversion of __B, and higher 128 bits come from conversion of __A.
+static __inline__ __m256bh __DEFAULT_FN_ATTRS256
+_mm256_mask_cvtne2ps_pbh(__m256bh __W, __mmask16 __U, __m256 __A, __m256 __B) {
+ return (__m256bh)__builtin_ia32_selectw_256((__mmask16)__U,
+ (__v16hi)_mm256_cvtne2ps_pbh(__A, __B),
+ (__v16hi)__W);
+}
+
+/// Convert Two Packed Single Data to One Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNE2PS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 256-bit vector of [8 x float].
+/// \param __B
+/// A 256-bit vector of [8 x float].
+/// \param __U
+/// A 16-bit mask value specifying what is chosen for each element.
+/// A 1 means conversion of __A or __B. A 0 means element is zero.
+/// \returns A 256-bit vector of [16 x bfloat] whose lower 128 bits come from
+/// conversion of __B, and higher 128 bits come from conversion of __A.
+static __inline__ __m256bh __DEFAULT_FN_ATTRS256
+_mm256_maskz_cvtne2ps_pbh(__mmask16 __U, __m256 __A, __m256 __B) {
+ return (__m256bh)__builtin_ia32_selectw_256((__mmask16)__U,
+ (__v16hi)_mm256_cvtne2ps_pbh(__A, __B),
+ (__v16hi)_mm256_setzero_si256());
+}
+
+/// Convert Packed Single Data to Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNEPS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 128-bit vector of [4 x float].
+/// \returns A 128-bit vector of [8 x bfloat] whose lower 64 bits come from
+/// conversion of __A, and higher 64 bits are 0.
+static __inline__ __m128bh __DEFAULT_FN_ATTRS128
+_mm_cvtneps_pbh(__m128 __A) {
+ return (__m128bh)__builtin_ia32_cvtneps2bf16_128_mask((__v4sf) __A,
+ (__v8hi)_mm_undefined_si128(),
+ (__mmask8)-1);
+}
+
+/// Convert Packed Single Data to Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNEPS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 128-bit vector of [4 x float].
+/// \param __W
+/// A 128-bit vector of [8 x bfloat].
+/// \param __U
+/// A 4-bit mask value specifying what is chosen for each element.
+/// A 1 means conversion of __A. A 0 means element from __W.
+/// \returns A 128-bit vector of [8 x bfloat] whose lower 64 bits come from
+/// conversion of __A, and higher 64 bits are 0.
+static __inline__ __m128bh __DEFAULT_FN_ATTRS128
+_mm_mask_cvtneps_pbh(__m128bh __W, __mmask8 __U, __m128 __A) {
+ return (__m128bh)__builtin_ia32_cvtneps2bf16_128_mask((__v4sf) __A,
+ (__v8hi)__W,
+ (__mmask8)__U);
+}
+
+/// Convert Packed Single Data to Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNEPS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 128-bit vector of [4 x float].
+/// \param __U
+/// A 4-bit mask value specifying what is chosen for each element.
+/// A 1 means conversion of __A. A 0 means element is zero.
+/// \returns A 128-bit vector of [8 x bfloat] whose lower 64 bits come from
+/// conversion of __A, and higher 64 bits are 0.
+static __inline__ __m128bh __DEFAULT_FN_ATTRS128
+_mm_maskz_cvtneps_pbh(__mmask8 __U, __m128 __A) {
+ return (__m128bh)__builtin_ia32_cvtneps2bf16_128_mask((__v4sf) __A,
+ (__v8hi)_mm_setzero_si128(),
+ (__mmask8)__U);
+}
+
+/// Convert Packed Single Data to Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNEPS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 256-bit vector of [8 x float].
+/// \returns A 128-bit vector of [8 x bfloat] comes from conversion of __A.
+static __inline__ __m128bh __DEFAULT_FN_ATTRS256
+_mm256_cvtneps_pbh(__m256 __A) {
+ return (__m128bh)__builtin_ia32_cvtneps2bf16_256_mask((__v8sf)__A,
+ (__v8hi)_mm_undefined_si128(),
+ (__mmask8)-1);
+}
+
+/// Convert Packed Single Data to Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNEPS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 256-bit vector of [8 x float].
+/// \param __W
+/// A 256-bit vector of [8 x bfloat].
+/// \param __U
+/// A 8-bit mask value specifying what is chosen for each element.
+/// A 1 means conversion of __A. A 0 means element from __W.
+/// \returns A 128-bit vector of [8 x bfloat] comes from conversion of __A.
+static __inline__ __m128bh __DEFAULT_FN_ATTRS256
+_mm256_mask_cvtneps_pbh(__m128bh __W, __mmask8 __U, __m256 __A) {
+ return (__m128bh)__builtin_ia32_cvtneps2bf16_256_mask((__v8sf)__A,
+ (__v8hi)__W,
+ (__mmask8)__U);
+}
+
+/// Convert Packed Single Data to Packed BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNEPS2BF16 </c> instructions.
+///
+/// \param __A
+/// A 256-bit vector of [8 x float].
+/// \param __U
+/// A 8-bit mask value specifying what is chosen for each element.
+/// A 1 means conversion of __A. A 0 means element is zero.
+/// \returns A 128-bit vector of [8 x bfloat] comes from conversion of __A.
+static __inline__ __m128bh __DEFAULT_FN_ATTRS256
+_mm256_maskz_cvtneps_pbh(__mmask8 __U, __m256 __A) {
+ return (__m128bh)__builtin_ia32_cvtneps2bf16_256_mask((__v8sf)__A,
+ (__v8hi)_mm_setzero_si128(),
+ (__mmask8)__U);
+}
+
+/// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VDPBF16PS </c> instructions.
+///
+/// \param __A
+/// A 128-bit vector of [8 x bfloat].
+/// \param __B
+/// A 128-bit vector of [8 x bfloat].
+/// \param __D
+/// A 128-bit vector of [4 x float].
+/// \returns A 128-bit vector of [4 x float] comes from Dot Product of
+/// __A, __B and __D
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_dpbf16_ps(__m128 __D, __m128bh __A, __m128bh __B) {
+ return (__m128)__builtin_ia32_dpbf16ps_128((__v4sf)__D,
+ (__v4si)__A,
+ (__v4si)__B);
+}
+
+/// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VDPBF16PS </c> instructions.
+///
+/// \param __A
+/// A 128-bit vector of [8 x bfloat].
+/// \param __B
+/// A 128-bit vector of [8 x bfloat].
+/// \param __D
+/// A 128-bit vector of [4 x float].
+/// \param __U
+/// A 8-bit mask value specifying what is chosen for each element.
+/// A 1 means __A and __B's dot product accumulated with __D. A 0 means __D.
+/// \returns A 128-bit vector of [4 x float] comes from Dot Product of
+/// __A, __B and __D
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_mask_dpbf16_ps(__m128 __D, __mmask8 __U, __m128bh __A, __m128bh __B) {
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm_dpbf16_ps(__D, __A, __B),
+ (__v4sf)__D);
+}
+
+/// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VDPBF16PS </c> instructions.
+///
+/// \param __A
+/// A 128-bit vector of [8 x bfloat].
+/// \param __B
+/// A 128-bit vector of [8 x bfloat].
+/// \param __D
+/// A 128-bit vector of [4 x float].
+/// \param __U
+/// A 8-bit mask value specifying what is chosen for each element.
+/// A 1 means __A and __B's dot product accumulated with __D. A 0 means 0.
+/// \returns A 128-bit vector of [4 x float] comes from Dot Product of
+/// __A, __B and __D
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_maskz_dpbf16_ps(__mmask8 __U, __m128 __D, __m128bh __A, __m128bh __B) {
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm_dpbf16_ps(__D, __A, __B),
+ (__v4sf)_mm_setzero_si128());
+}
+
+/// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VDPBF16PS </c> instructions.
+///
+/// \param __A
+/// A 256-bit vector of [16 x bfloat].
+/// \param __B
+/// A 256-bit vector of [16 x bfloat].
+/// \param __D
+/// A 256-bit vector of [8 x float].
+/// \returns A 256-bit vector of [8 x float] comes from Dot Product of
+/// __A, __B and __D
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_dpbf16_ps(__m256 __D, __m256bh __A, __m256bh __B) {
+ return (__m256)__builtin_ia32_dpbf16ps_256((__v8sf)__D,
+ (__v8si)__A,
+ (__v8si)__B);
+}
+
+/// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VDPBF16PS </c> instructions.
+///
+/// \param __A
+/// A 256-bit vector of [16 x bfloat].
+/// \param __B
+/// A 256-bit vector of [16 x bfloat].
+/// \param __D
+/// A 256-bit vector of [8 x float].
+/// \param __U
+/// A 16-bit mask value specifying what is chosen for each element.
+/// A 1 means __A and __B's dot product accumulated with __D. A 0 means __D.
+/// \returns A 256-bit vector of [8 x float] comes from Dot Product of
+/// __A, __B and __D
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_mask_dpbf16_ps(__m256 __D, __mmask8 __U, __m256bh __A, __m256bh __B) {
+ return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
+ (__v8sf)_mm256_dpbf16_ps(__D, __A, __B),
+ (__v8sf)__D);
+}
+
+/// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VDPBF16PS </c> instructions.
+///
+/// \param __A
+/// A 256-bit vector of [16 x bfloat].
+/// \param __B
+/// A 256-bit vector of [16 x bfloat].
+/// \param __D
+/// A 256-bit vector of [8 x float].
+/// \param __U
+/// A 8-bit mask value specifying what is chosen for each element.
+/// A 1 means __A and __B's dot product accumulated with __D. A 0 means 0.
+/// \returns A 256-bit vector of [8 x float] comes from Dot Product of
+/// __A, __B and __D
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_maskz_dpbf16_ps(__mmask8 __U, __m256 __D, __m256bh __A, __m256bh __B) {
+ return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
+ (__v8sf)_mm256_dpbf16_ps(__D, __A, __B),
+ (__v8sf)_mm256_setzero_si256());
+}
+
+/// Convert One Single float Data to One BF16 Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTNEPS2BF16 </c> instructions.
+///
+/// \param __A
+/// A float data.
+/// \returns A bf16 data whose sign field and exponent field keep unchanged,
+/// and fraction field is truncated to 7 bits.
+static __inline__ __bfloat16 __DEFAULT_FN_ATTRS128 _mm_cvtness_sbh(float __A) {
+ __v4sf __V = {__A, 0, 0, 0};
+ __v8hi __R = __builtin_ia32_cvtneps2bf16_128_mask(
+ (__v4sf)__V, (__v8hi)_mm_undefined_si128(), (__mmask8)-1);
+ return __R[0];
+}
+
+/// Convert Packed BF16 Data to Packed float Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// \param __A
+/// A 128-bit vector of [8 x bfloat].
+/// \returns A 256-bit vector of [8 x float] come from convertion of __A
+static __inline__ __m256 __DEFAULT_FN_ATTRS256 _mm256_cvtpbh_ps(__m128bh __A) {
+ return _mm256_castsi256_ps((__m256i)_mm256_slli_epi32(
+ (__m256i)_mm256_cvtepi16_epi32((__m128i)__A), 16));
+}
+
+/// Convert Packed BF16 Data to Packed float Data using zeroing mask.
+///
+/// \headerfile <x86intrin.h>
+///
+/// \param __U
+/// A 8-bit mask. Elements are zeroed out when the corresponding mask
+/// bit is not set.
+/// \param __A
+/// A 128-bit vector of [8 x bfloat].
+/// \returns A 256-bit vector of [8 x float] come from convertion of __A
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_maskz_cvtpbh_ps(__mmask8 __U, __m128bh __A) {
+ return _mm256_castsi256_ps((__m256i)_mm256_slli_epi32(
+ (__m256i)_mm256_maskz_cvtepi16_epi32((__mmask8)__U, (__m128i)__A), 16));
+}
+
+/// Convert Packed BF16 Data to Packed float Data using merging mask.
+///
+/// \headerfile <x86intrin.h>
+///
+/// \param __S
+/// A 256-bit vector of [8 x float]. Elements are copied from __S when
+/// the corresponding mask bit is not set.
+/// \param __U
+/// A 8-bit mask. Elements are zeroed out when the corresponding mask
+/// bit is not set.
+/// \param __A
+/// A 128-bit vector of [8 x bfloat].
+/// \returns A 256-bit vector of [8 x float] come from convertion of __A
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_mask_cvtpbh_ps(__m256 __S, __mmask8 __U, __m128bh __A) {
+ return _mm256_castsi256_ps((__m256i)_mm256_mask_slli_epi32(
+ (__m256i)__S, (__mmask8)__U, (__m256i)_mm256_cvtepi16_epi32((__m128i)__A),
+ 16));
+}
+
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
+
+#endif
--- /dev/null
+/*===---- avx512vlbitalgintrin.h - BITALG intrinsics -----------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512vlbitalgintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __AVX512VLBITALGINTRIN_H
+#define __AVX512VLBITALGINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512bitalg"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512bitalg"), __min_vector_width__(256)))
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_popcnt_epi16(__m256i __A)
+{
+ return (__m256i) __builtin_ia32_vpopcntw_256((__v16hi) __A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_popcnt_epi16(__m256i __A, __mmask16 __U, __m256i __B)
+{
+ return (__m256i) __builtin_ia32_selectw_256((__mmask16) __U,
+ (__v16hi) _mm256_popcnt_epi16(__B),
+ (__v16hi) __A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_popcnt_epi16(__mmask16 __U, __m256i __B)
+{
+ return _mm256_mask_popcnt_epi16((__m256i) _mm256_setzero_si256(),
+ __U,
+ __B);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_popcnt_epi16(__m128i __A)
+{
+ return (__m128i) __builtin_ia32_vpopcntw_128((__v8hi) __A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_popcnt_epi16(__m128i __A, __mmask8 __U, __m128i __B)
+{
+ return (__m128i) __builtin_ia32_selectw_128((__mmask8) __U,
+ (__v8hi) _mm_popcnt_epi16(__B),
+ (__v8hi) __A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_popcnt_epi16(__mmask8 __U, __m128i __B)
+{
+ return _mm_mask_popcnt_epi16((__m128i) _mm_setzero_si128(),
+ __U,
+ __B);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_popcnt_epi8(__m256i __A)
+{
+ return (__m256i) __builtin_ia32_vpopcntb_256((__v32qi) __A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_popcnt_epi8(__m256i __A, __mmask32 __U, __m256i __B)
+{
+ return (__m256i) __builtin_ia32_selectb_256((__mmask32) __U,
+ (__v32qi) _mm256_popcnt_epi8(__B),
+ (__v32qi) __A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_popcnt_epi8(__mmask32 __U, __m256i __B)
+{
+ return _mm256_mask_popcnt_epi8((__m256i) _mm256_setzero_si256(),
+ __U,
+ __B);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_popcnt_epi8(__m128i __A)
+{
+ return (__m128i) __builtin_ia32_vpopcntb_128((__v16qi) __A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_popcnt_epi8(__m128i __A, __mmask16 __U, __m128i __B)
+{
+ return (__m128i) __builtin_ia32_selectb_128((__mmask16) __U,
+ (__v16qi) _mm_popcnt_epi8(__B),
+ (__v16qi) __A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_popcnt_epi8(__mmask16 __U, __m128i __B)
+{
+ return _mm_mask_popcnt_epi8((__m128i) _mm_setzero_si128(),
+ __U,
+ __B);
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS256
+_mm256_mask_bitshuffle_epi64_mask(__mmask32 __U, __m256i __A, __m256i __B)
+{
+ return (__mmask32) __builtin_ia32_vpshufbitqmb256_mask((__v32qi) __A,
+ (__v32qi) __B,
+ __U);
+}
+
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS256
+_mm256_bitshuffle_epi64_mask(__m256i __A, __m256i __B)
+{
+ return _mm256_mask_bitshuffle_epi64_mask((__mmask32) -1,
+ __A,
+ __B);
+}
+
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS128
+_mm_mask_bitshuffle_epi64_mask(__mmask16 __U, __m128i __A, __m128i __B)
+{
+ return (__mmask16) __builtin_ia32_vpshufbitqmb128_mask((__v16qi) __A,
+ (__v16qi) __B,
+ __U);
+}
+
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS128
+_mm_bitshuffle_epi64_mask(__m128i __A, __m128i __B)
+{
+ return _mm_mask_bitshuffle_epi64_mask((__mmask16) -1,
+ __A,
+ __B);
+}
+
+
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
+
+#endif
/*===---- avx512vlbwintrin.h - AVX512VL and AVX512BW intrinsics ------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#define __AVX512VLBWINTRIN_H
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512bw")))
-
-static __inline __m128i __DEFAULT_FN_ATTRS
-_mm_setzero_hi(void){
- return (__m128i)(__v8hi){ 0, 0, 0, 0, 0, 0, 0, 0 };
-}
+#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512bw"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512bw"), __min_vector_width__(256)))
/* Integer compare */
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_cmpeq_epi8_mask(__m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_pcmpeqb128_mask((__v16qi)__a, (__v16qi)__b,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_mask_cmpeq_epi8_mask(__mmask16 __u, __m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_pcmpeqb128_mask((__v16qi)__a, (__v16qi)__b,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_cmpeq_epu8_mask(__m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)__a, (__v16qi)__b, 0,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_mask_cmpeq_epu8_mask(__mmask16 __u, __m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)__a, (__v16qi)__b, 0,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_cmpeq_epi8_mask(__m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_pcmpeqb256_mask((__v32qi)__a, (__v32qi)__b,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpeq_epi8_mask(__mmask32 __u, __m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_pcmpeqb256_mask((__v32qi)__a, (__v32qi)__b,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_cmpeq_epu8_mask(__m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)__a, (__v32qi)__b, 0,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpeq_epu8_mask(__mmask32 __u, __m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)__a, (__v32qi)__b, 0,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpeq_epi16_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_pcmpeqw128_mask((__v8hi)__a, (__v8hi)__b,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpeq_epi16_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_pcmpeqw128_mask((__v8hi)__a, (__v8hi)__b,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpeq_epu16_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)__a, (__v8hi)__b, 0,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpeq_epu16_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)__a, (__v8hi)__b, 0,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_cmpeq_epi16_mask(__m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_pcmpeqw256_mask((__v16hi)__a, (__v16hi)__b,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpeq_epi16_mask(__mmask16 __u, __m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_pcmpeqw256_mask((__v16hi)__a, (__v16hi)__b,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_cmpeq_epu16_mask(__m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)__a, (__v16hi)__b, 0,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpeq_epu16_mask(__mmask16 __u, __m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)__a, (__v16hi)__b, 0,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_cmpge_epi8_mask(__m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_cmpb128_mask((__v16qi)__a, (__v16qi)__b, 5,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_mask_cmpge_epi8_mask(__mmask16 __u, __m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_cmpb128_mask((__v16qi)__a, (__v16qi)__b, 5,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_cmpge_epu8_mask(__m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)__a, (__v16qi)__b, 5,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_mask_cmpge_epu8_mask(__mmask16 __u, __m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)__a, (__v16qi)__b, 5,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_cmpge_epi8_mask(__m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_cmpb256_mask((__v32qi)__a, (__v32qi)__b, 5,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpge_epi8_mask(__mmask32 __u, __m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_cmpb256_mask((__v32qi)__a, (__v32qi)__b, 5,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_cmpge_epu8_mask(__m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)__a, (__v32qi)__b, 5,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpge_epu8_mask(__mmask32 __u, __m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)__a, (__v32qi)__b, 5,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpge_epi16_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpw128_mask((__v8hi)__a, (__v8hi)__b, 5,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpge_epi16_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpw128_mask((__v8hi)__a, (__v8hi)__b, 5,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpge_epu16_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)__a, (__v8hi)__b, 5,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpge_epu16_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)__a, (__v8hi)__b, 5,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_cmpge_epi16_mask(__m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_cmpw256_mask((__v16hi)__a, (__v16hi)__b, 5,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpge_epi16_mask(__mmask16 __u, __m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_cmpw256_mask((__v16hi)__a, (__v16hi)__b, 5,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_cmpge_epu16_mask(__m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)__a, (__v16hi)__b, 5,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpge_epu16_mask(__mmask16 __u, __m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)__a, (__v16hi)__b, 5,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_cmpgt_epi8_mask(__m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_pcmpgtb128_mask((__v16qi)__a, (__v16qi)__b,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_mask_cmpgt_epi8_mask(__mmask16 __u, __m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_pcmpgtb128_mask((__v16qi)__a, (__v16qi)__b,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_cmpgt_epu8_mask(__m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)__a, (__v16qi)__b, 6,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_mask_cmpgt_epu8_mask(__mmask16 __u, __m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)__a, (__v16qi)__b, 6,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_cmpgt_epi8_mask(__m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_pcmpgtb256_mask((__v32qi)__a, (__v32qi)__b,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpgt_epi8_mask(__mmask32 __u, __m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_pcmpgtb256_mask((__v32qi)__a, (__v32qi)__b,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_cmpgt_epu8_mask(__m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)__a, (__v32qi)__b, 6,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpgt_epu8_mask(__mmask32 __u, __m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)__a, (__v32qi)__b, 6,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpgt_epi16_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_pcmpgtw128_mask((__v8hi)__a, (__v8hi)__b,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpgt_epi16_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_pcmpgtw128_mask((__v8hi)__a, (__v8hi)__b,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpgt_epu16_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)__a, (__v8hi)__b, 6,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpgt_epu16_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)__a, (__v8hi)__b, 6,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_cmpgt_epi16_mask(__m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_pcmpgtw256_mask((__v16hi)__a, (__v16hi)__b,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpgt_epi16_mask(__mmask16 __u, __m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_pcmpgtw256_mask((__v16hi)__a, (__v16hi)__b,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_cmpgt_epu16_mask(__m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)__a, (__v16hi)__b, 6,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpgt_epu16_mask(__mmask16 __u, __m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)__a, (__v16hi)__b, 6,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_cmple_epi8_mask(__m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_cmpb128_mask((__v16qi)__a, (__v16qi)__b, 2,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_mask_cmple_epi8_mask(__mmask16 __u, __m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_cmpb128_mask((__v16qi)__a, (__v16qi)__b, 2,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_cmple_epu8_mask(__m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)__a, (__v16qi)__b, 2,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_mask_cmple_epu8_mask(__mmask16 __u, __m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)__a, (__v16qi)__b, 2,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_cmple_epi8_mask(__m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_cmpb256_mask((__v32qi)__a, (__v32qi)__b, 2,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_mask_cmple_epi8_mask(__mmask32 __u, __m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_cmpb256_mask((__v32qi)__a, (__v32qi)__b, 2,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_cmple_epu8_mask(__m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)__a, (__v32qi)__b, 2,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_mask_cmple_epu8_mask(__mmask32 __u, __m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)__a, (__v32qi)__b, 2,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmple_epi16_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpw128_mask((__v8hi)__a, (__v8hi)__b, 2,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmple_epi16_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpw128_mask((__v8hi)__a, (__v8hi)__b, 2,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmple_epu16_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)__a, (__v8hi)__b, 2,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmple_epu16_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)__a, (__v8hi)__b, 2,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_cmple_epi16_mask(__m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_cmpw256_mask((__v16hi)__a, (__v16hi)__b, 2,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_mask_cmple_epi16_mask(__mmask16 __u, __m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_cmpw256_mask((__v16hi)__a, (__v16hi)__b, 2,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_cmple_epu16_mask(__m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)__a, (__v16hi)__b, 2,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_mask_cmple_epu16_mask(__mmask16 __u, __m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)__a, (__v16hi)__b, 2,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_cmplt_epi8_mask(__m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_cmpb128_mask((__v16qi)__a, (__v16qi)__b, 1,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_mask_cmplt_epi8_mask(__mmask16 __u, __m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_cmpb128_mask((__v16qi)__a, (__v16qi)__b, 1,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_cmplt_epu8_mask(__m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)__a, (__v16qi)__b, 1,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_mask_cmplt_epu8_mask(__mmask16 __u, __m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)__a, (__v16qi)__b, 1,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_cmplt_epi8_mask(__m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_cmpb256_mask((__v32qi)__a, (__v32qi)__b, 1,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_mask_cmplt_epi8_mask(__mmask32 __u, __m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_cmpb256_mask((__v32qi)__a, (__v32qi)__b, 1,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_cmplt_epu8_mask(__m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)__a, (__v32qi)__b, 1,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_mask_cmplt_epu8_mask(__mmask32 __u, __m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)__a, (__v32qi)__b, 1,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmplt_epi16_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpw128_mask((__v8hi)__a, (__v8hi)__b, 1,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmplt_epi16_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpw128_mask((__v8hi)__a, (__v8hi)__b, 1,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmplt_epu16_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)__a, (__v8hi)__b, 1,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmplt_epu16_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)__a, (__v8hi)__b, 1,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_cmplt_epi16_mask(__m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_cmpw256_mask((__v16hi)__a, (__v16hi)__b, 1,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_mask_cmplt_epi16_mask(__mmask16 __u, __m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_cmpw256_mask((__v16hi)__a, (__v16hi)__b, 1,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_cmplt_epu16_mask(__m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)__a, (__v16hi)__b, 1,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_mask_cmplt_epu16_mask(__mmask16 __u, __m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)__a, (__v16hi)__b, 1,
- __u);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_cmpneq_epi8_mask(__m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_cmpb128_mask((__v16qi)__a, (__v16qi)__b, 4,
- (__mmask16)-1);
-}
-
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_mask_cmpneq_epi8_mask(__mmask16 __u, __m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_cmpb128_mask((__v16qi)__a, (__v16qi)__b, 4,
- __u);
-}
+#define _mm_cmp_epi8_mask(a, b, p) \
+ (__mmask16)__builtin_ia32_cmpb128_mask((__v16qi)(__m128i)(a), \
+ (__v16qi)(__m128i)(b), (int)(p), \
+ (__mmask16)-1)
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_cmpneq_epu8_mask(__m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)__a, (__v16qi)__b, 4,
- (__mmask16)-1);
-}
+#define _mm_mask_cmp_epi8_mask(m, a, b, p) \
+ (__mmask16)__builtin_ia32_cmpb128_mask((__v16qi)(__m128i)(a), \
+ (__v16qi)(__m128i)(b), (int)(p), \
+ (__mmask16)(m))
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm_mask_cmpneq_epu8_mask(__mmask16 __u, __m128i __a, __m128i __b) {
- return (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)__a, (__v16qi)__b, 4,
- __u);
-}
+#define _mm_cmp_epu8_mask(a, b, p) \
+ (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)(__m128i)(a), \
+ (__v16qi)(__m128i)(b), (int)(p), \
+ (__mmask16)-1)
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_cmpneq_epi8_mask(__m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_cmpb256_mask((__v32qi)__a, (__v32qi)__b, 4,
- (__mmask32)-1);
-}
+#define _mm_mask_cmp_epu8_mask(m, a, b, p) \
+ (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)(__m128i)(a), \
+ (__v16qi)(__m128i)(b), (int)(p), \
+ (__mmask16)(m))
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpneq_epi8_mask(__mmask32 __u, __m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_cmpb256_mask((__v32qi)__a, (__v32qi)__b, 4,
- __u);
-}
+#define _mm256_cmp_epi8_mask(a, b, p) \
+ (__mmask32)__builtin_ia32_cmpb256_mask((__v32qi)(__m256i)(a), \
+ (__v32qi)(__m256i)(b), (int)(p), \
+ (__mmask32)-1)
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_cmpneq_epu8_mask(__m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)__a, (__v32qi)__b, 4,
- (__mmask32)-1);
-}
+#define _mm256_mask_cmp_epi8_mask(m, a, b, p) \
+ (__mmask32)__builtin_ia32_cmpb256_mask((__v32qi)(__m256i)(a), \
+ (__v32qi)(__m256i)(b), (int)(p), \
+ (__mmask32)(m))
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpneq_epu8_mask(__mmask32 __u, __m256i __a, __m256i __b) {
- return (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)__a, (__v32qi)__b, 4,
- __u);
-}
+#define _mm256_cmp_epu8_mask(a, b, p) \
+ (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)(__m256i)(a), \
+ (__v32qi)(__m256i)(b), (int)(p), \
+ (__mmask32)-1)
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpneq_epi16_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpw128_mask((__v8hi)__a, (__v8hi)__b, 4,
- (__mmask8)-1);
-}
+#define _mm256_mask_cmp_epu8_mask(m, a, b, p) \
+ (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)(__m256i)(a), \
+ (__v32qi)(__m256i)(b), (int)(p), \
+ (__mmask32)(m))
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpneq_epi16_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpw128_mask((__v8hi)__a, (__v8hi)__b, 4,
- __u);
-}
+#define _mm_cmp_epi16_mask(a, b, p) \
+ (__mmask8)__builtin_ia32_cmpw128_mask((__v8hi)(__m128i)(a), \
+ (__v8hi)(__m128i)(b), (int)(p), \
+ (__mmask8)-1)
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpneq_epu16_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)__a, (__v8hi)__b, 4,
- (__mmask8)-1);
-}
+#define _mm_mask_cmp_epi16_mask(m, a, b, p) \
+ (__mmask8)__builtin_ia32_cmpw128_mask((__v8hi)(__m128i)(a), \
+ (__v8hi)(__m128i)(b), (int)(p), \
+ (__mmask8)(m))
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpneq_epu16_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)__a, (__v8hi)__b, 4,
- __u);
-}
+#define _mm_cmp_epu16_mask(a, b, p) \
+ (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)(__m128i)(a), \
+ (__v8hi)(__m128i)(b), (int)(p), \
+ (__mmask8)-1)
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_cmpneq_epi16_mask(__m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_cmpw256_mask((__v16hi)__a, (__v16hi)__b, 4,
- (__mmask16)-1);
-}
+#define _mm_mask_cmp_epu16_mask(m, a, b, p) \
+ (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)(__m128i)(a), \
+ (__v8hi)(__m128i)(b), (int)(p), \
+ (__mmask8)(m))
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpneq_epi16_mask(__mmask16 __u, __m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_cmpw256_mask((__v16hi)__a, (__v16hi)__b, 4,
- __u);
-}
+#define _mm256_cmp_epi16_mask(a, b, p) \
+ (__mmask16)__builtin_ia32_cmpw256_mask((__v16hi)(__m256i)(a), \
+ (__v16hi)(__m256i)(b), (int)(p), \
+ (__mmask16)-1)
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_cmpneq_epu16_mask(__m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)__a, (__v16hi)__b, 4,
- (__mmask16)-1);
-}
+#define _mm256_mask_cmp_epi16_mask(m, a, b, p) \
+ (__mmask16)__builtin_ia32_cmpw256_mask((__v16hi)(__m256i)(a), \
+ (__v16hi)(__m256i)(b), (int)(p), \
+ (__mmask16)(m))
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpneq_epu16_mask(__mmask16 __u, __m256i __a, __m256i __b) {
- return (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)__a, (__v16hi)__b, 4,
- __u);
-}
+#define _mm256_cmp_epu16_mask(a, b, p) \
+ (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)(__m256i)(a), \
+ (__v16hi)(__m256i)(b), (int)(p), \
+ (__mmask16)-1)
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+#define _mm256_mask_cmp_epu16_mask(m, a, b, p) \
+ (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)(__m256i)(a), \
+ (__v16hi)(__m256i)(b), (int)(p), \
+ (__mmask16)(m))
+
+#define _mm_cmpeq_epi8_mask(A, B) \
+ _mm_cmp_epi8_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm_mask_cmpeq_epi8_mask(k, A, B) \
+ _mm_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm_cmpge_epi8_mask(A, B) \
+ _mm_cmp_epi8_mask((A), (B), _MM_CMPINT_GE)
+#define _mm_mask_cmpge_epi8_mask(k, A, B) \
+ _mm_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm_cmpgt_epi8_mask(A, B) \
+ _mm_cmp_epi8_mask((A), (B), _MM_CMPINT_GT)
+#define _mm_mask_cmpgt_epi8_mask(k, A, B) \
+ _mm_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm_cmple_epi8_mask(A, B) \
+ _mm_cmp_epi8_mask((A), (B), _MM_CMPINT_LE)
+#define _mm_mask_cmple_epi8_mask(k, A, B) \
+ _mm_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm_cmplt_epi8_mask(A, B) \
+ _mm_cmp_epi8_mask((A), (B), _MM_CMPINT_LT)
+#define _mm_mask_cmplt_epi8_mask(k, A, B) \
+ _mm_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm_cmpneq_epi8_mask(A, B) \
+ _mm_cmp_epi8_mask((A), (B), _MM_CMPINT_NE)
+#define _mm_mask_cmpneq_epi8_mask(k, A, B) \
+ _mm_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm256_cmpeq_epi8_mask(A, B) \
+ _mm256_cmp_epi8_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm256_mask_cmpeq_epi8_mask(k, A, B) \
+ _mm256_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm256_cmpge_epi8_mask(A, B) \
+ _mm256_cmp_epi8_mask((A), (B), _MM_CMPINT_GE)
+#define _mm256_mask_cmpge_epi8_mask(k, A, B) \
+ _mm256_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm256_cmpgt_epi8_mask(A, B) \
+ _mm256_cmp_epi8_mask((A), (B), _MM_CMPINT_GT)
+#define _mm256_mask_cmpgt_epi8_mask(k, A, B) \
+ _mm256_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm256_cmple_epi8_mask(A, B) \
+ _mm256_cmp_epi8_mask((A), (B), _MM_CMPINT_LE)
+#define _mm256_mask_cmple_epi8_mask(k, A, B) \
+ _mm256_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm256_cmplt_epi8_mask(A, B) \
+ _mm256_cmp_epi8_mask((A), (B), _MM_CMPINT_LT)
+#define _mm256_mask_cmplt_epi8_mask(k, A, B) \
+ _mm256_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm256_cmpneq_epi8_mask(A, B) \
+ _mm256_cmp_epi8_mask((A), (B), _MM_CMPINT_NE)
+#define _mm256_mask_cmpneq_epi8_mask(k, A, B) \
+ _mm256_mask_cmp_epi8_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm_cmpeq_epu8_mask(A, B) \
+ _mm_cmp_epu8_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm_mask_cmpeq_epu8_mask(k, A, B) \
+ _mm_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm_cmpge_epu8_mask(A, B) \
+ _mm_cmp_epu8_mask((A), (B), _MM_CMPINT_GE)
+#define _mm_mask_cmpge_epu8_mask(k, A, B) \
+ _mm_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm_cmpgt_epu8_mask(A, B) \
+ _mm_cmp_epu8_mask((A), (B), _MM_CMPINT_GT)
+#define _mm_mask_cmpgt_epu8_mask(k, A, B) \
+ _mm_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm_cmple_epu8_mask(A, B) \
+ _mm_cmp_epu8_mask((A), (B), _MM_CMPINT_LE)
+#define _mm_mask_cmple_epu8_mask(k, A, B) \
+ _mm_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm_cmplt_epu8_mask(A, B) \
+ _mm_cmp_epu8_mask((A), (B), _MM_CMPINT_LT)
+#define _mm_mask_cmplt_epu8_mask(k, A, B) \
+ _mm_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm_cmpneq_epu8_mask(A, B) \
+ _mm_cmp_epu8_mask((A), (B), _MM_CMPINT_NE)
+#define _mm_mask_cmpneq_epu8_mask(k, A, B) \
+ _mm_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm256_cmpeq_epu8_mask(A, B) \
+ _mm256_cmp_epu8_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm256_mask_cmpeq_epu8_mask(k, A, B) \
+ _mm256_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm256_cmpge_epu8_mask(A, B) \
+ _mm256_cmp_epu8_mask((A), (B), _MM_CMPINT_GE)
+#define _mm256_mask_cmpge_epu8_mask(k, A, B) \
+ _mm256_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm256_cmpgt_epu8_mask(A, B) \
+ _mm256_cmp_epu8_mask((A), (B), _MM_CMPINT_GT)
+#define _mm256_mask_cmpgt_epu8_mask(k, A, B) \
+ _mm256_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm256_cmple_epu8_mask(A, B) \
+ _mm256_cmp_epu8_mask((A), (B), _MM_CMPINT_LE)
+#define _mm256_mask_cmple_epu8_mask(k, A, B) \
+ _mm256_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm256_cmplt_epu8_mask(A, B) \
+ _mm256_cmp_epu8_mask((A), (B), _MM_CMPINT_LT)
+#define _mm256_mask_cmplt_epu8_mask(k, A, B) \
+ _mm256_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm256_cmpneq_epu8_mask(A, B) \
+ _mm256_cmp_epu8_mask((A), (B), _MM_CMPINT_NE)
+#define _mm256_mask_cmpneq_epu8_mask(k, A, B) \
+ _mm256_mask_cmp_epu8_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm_cmpeq_epi16_mask(A, B) \
+ _mm_cmp_epi16_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm_mask_cmpeq_epi16_mask(k, A, B) \
+ _mm_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm_cmpge_epi16_mask(A, B) \
+ _mm_cmp_epi16_mask((A), (B), _MM_CMPINT_GE)
+#define _mm_mask_cmpge_epi16_mask(k, A, B) \
+ _mm_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm_cmpgt_epi16_mask(A, B) \
+ _mm_cmp_epi16_mask((A), (B), _MM_CMPINT_GT)
+#define _mm_mask_cmpgt_epi16_mask(k, A, B) \
+ _mm_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm_cmple_epi16_mask(A, B) \
+ _mm_cmp_epi16_mask((A), (B), _MM_CMPINT_LE)
+#define _mm_mask_cmple_epi16_mask(k, A, B) \
+ _mm_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm_cmplt_epi16_mask(A, B) \
+ _mm_cmp_epi16_mask((A), (B), _MM_CMPINT_LT)
+#define _mm_mask_cmplt_epi16_mask(k, A, B) \
+ _mm_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm_cmpneq_epi16_mask(A, B) \
+ _mm_cmp_epi16_mask((A), (B), _MM_CMPINT_NE)
+#define _mm_mask_cmpneq_epi16_mask(k, A, B) \
+ _mm_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm256_cmpeq_epi16_mask(A, B) \
+ _mm256_cmp_epi16_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm256_mask_cmpeq_epi16_mask(k, A, B) \
+ _mm256_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm256_cmpge_epi16_mask(A, B) \
+ _mm256_cmp_epi16_mask((A), (B), _MM_CMPINT_GE)
+#define _mm256_mask_cmpge_epi16_mask(k, A, B) \
+ _mm256_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm256_cmpgt_epi16_mask(A, B) \
+ _mm256_cmp_epi16_mask((A), (B), _MM_CMPINT_GT)
+#define _mm256_mask_cmpgt_epi16_mask(k, A, B) \
+ _mm256_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm256_cmple_epi16_mask(A, B) \
+ _mm256_cmp_epi16_mask((A), (B), _MM_CMPINT_LE)
+#define _mm256_mask_cmple_epi16_mask(k, A, B) \
+ _mm256_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm256_cmplt_epi16_mask(A, B) \
+ _mm256_cmp_epi16_mask((A), (B), _MM_CMPINT_LT)
+#define _mm256_mask_cmplt_epi16_mask(k, A, B) \
+ _mm256_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm256_cmpneq_epi16_mask(A, B) \
+ _mm256_cmp_epi16_mask((A), (B), _MM_CMPINT_NE)
+#define _mm256_mask_cmpneq_epi16_mask(k, A, B) \
+ _mm256_mask_cmp_epi16_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm_cmpeq_epu16_mask(A, B) \
+ _mm_cmp_epu16_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm_mask_cmpeq_epu16_mask(k, A, B) \
+ _mm_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm_cmpge_epu16_mask(A, B) \
+ _mm_cmp_epu16_mask((A), (B), _MM_CMPINT_GE)
+#define _mm_mask_cmpge_epu16_mask(k, A, B) \
+ _mm_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm_cmpgt_epu16_mask(A, B) \
+ _mm_cmp_epu16_mask((A), (B), _MM_CMPINT_GT)
+#define _mm_mask_cmpgt_epu16_mask(k, A, B) \
+ _mm_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm_cmple_epu16_mask(A, B) \
+ _mm_cmp_epu16_mask((A), (B), _MM_CMPINT_LE)
+#define _mm_mask_cmple_epu16_mask(k, A, B) \
+ _mm_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm_cmplt_epu16_mask(A, B) \
+ _mm_cmp_epu16_mask((A), (B), _MM_CMPINT_LT)
+#define _mm_mask_cmplt_epu16_mask(k, A, B) \
+ _mm_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm_cmpneq_epu16_mask(A, B) \
+ _mm_cmp_epu16_mask((A), (B), _MM_CMPINT_NE)
+#define _mm_mask_cmpneq_epu16_mask(k, A, B) \
+ _mm_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm256_cmpeq_epu16_mask(A, B) \
+ _mm256_cmp_epu16_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm256_mask_cmpeq_epu16_mask(k, A, B) \
+ _mm256_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm256_cmpge_epu16_mask(A, B) \
+ _mm256_cmp_epu16_mask((A), (B), _MM_CMPINT_GE)
+#define _mm256_mask_cmpge_epu16_mask(k, A, B) \
+ _mm256_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm256_cmpgt_epu16_mask(A, B) \
+ _mm256_cmp_epu16_mask((A), (B), _MM_CMPINT_GT)
+#define _mm256_mask_cmpgt_epu16_mask(k, A, B) \
+ _mm256_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm256_cmple_epu16_mask(A, B) \
+ _mm256_cmp_epu16_mask((A), (B), _MM_CMPINT_LE)
+#define _mm256_mask_cmple_epu16_mask(k, A, B) \
+ _mm256_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm256_cmplt_epu16_mask(A, B) \
+ _mm256_cmp_epu16_mask((A), (B), _MM_CMPINT_LT)
+#define _mm256_mask_cmplt_epu16_mask(k, A, B) \
+ _mm256_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm256_cmpneq_epu16_mask(A, B) \
+ _mm256_cmp_epu16_mask((A), (B), _MM_CMPINT_NE)
+#define _mm256_mask_cmpneq_epu16_mask(k, A, B) \
+ _mm256_mask_cmp_epu16_mask((k), (A), (B), _MM_CMPINT_NE)
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_add_epi8(__m256i __W, __mmask32 __U, __m256i __A, __m256i __B){
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)_mm256_add_epi8(__A, __B),
(__v32qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_add_epi8(__mmask32 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)_mm256_add_epi8(__A, __B),
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_add_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_add_epi16(__A, __B),
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_add_epi16(__mmask16 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_add_epi16(__A, __B),
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_sub_epi8(__m256i __W, __mmask32 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)_mm256_sub_epi8(__A, __B),
(__v32qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_sub_epi8(__mmask32 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)_mm256_sub_epi8(__A, __B),
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_sub_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_sub_epi16(__A, __B),
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_sub_epi16(__mmask16 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_sub_epi16(__A, __B),
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_add_epi8(__m128i __W, __mmask16 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)_mm_add_epi8(__A, __B),
(__v16qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_add_epi8(__mmask16 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)_mm_add_epi8(__A, __B),
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_add_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_add_epi16(__A, __B),
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_add_epi16(__mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_add_epi16(__A, __B),
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_sub_epi8(__m128i __W, __mmask16 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)_mm_sub_epi8(__A, __B),
(__v16qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_sub_epi8(__mmask16 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)_mm_sub_epi8(__A, __B),
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_sub_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_sub_epi16(__A, __B),
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_sub_epi16(__mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_sub_epi16(__A, __B),
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_mullo_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_mullo_epi16(__A, __B),
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_mullo_epi16(__mmask16 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_mullo_epi16(__A, __B),
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_mullo_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_mullo_epi16(__A, __B),
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_mullo_epi16(__mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_mullo_epi16(__A, __B),
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_blend_epi8 (__mmask16 __U, __m128i __A, __m128i __W)
{
return (__m128i) __builtin_ia32_selectb_128 ((__mmask16) __U,
(__v16qi) __A);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_blend_epi8 (__mmask32 __U, __m256i __A, __m256i __W)
{
return (__m256i) __builtin_ia32_selectb_256 ((__mmask32) __U,
(__v32qi) __A);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_blend_epi16 (__mmask8 __U, __m128i __A, __m128i __W)
{
return (__m128i) __builtin_ia32_selectw_128 ((__mmask8) __U,
(__v8hi) __A);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_blend_epi16 (__mmask16 __U, __m256i __A, __m256i __W)
{
return (__m256i) __builtin_ia32_selectw_256 ((__mmask16) __U,
(__v16hi) __A);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_abs_epi8(__m128i __W, __mmask16 __U, __m128i __A)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_abs_epi8(__mmask16 __U, __m128i __A)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_abs_epi8(__m256i __W, __mmask32 __U, __m256i __A)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_abs_epi8 (__mmask32 __U, __m256i __A)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_abs_epi16(__m128i __W, __mmask8 __U, __m128i __A)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_abs_epi16(__mmask8 __U, __m128i __A)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_abs_epi16(__m256i __W, __mmask16 __U, __m256i __A)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_abs_epi16(__mmask16 __U, __m256i __A)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_packs_epi32(__mmask8 __M, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__M,
(__v8hi)_mm_packs_epi32(__A, __B),
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_packs_epi32(__m128i __W, __mmask16 __M, __m128i __A, __m128i __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_packs_epi32(__m128i __W, __mmask8 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__M,
(__v8hi)_mm_packs_epi32(__A, __B),
(__v8hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_packs_epi32(__mmask16 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__M,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_packs_epi32(__m256i __W, __mmask16 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__M,
(__v16hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_packs_epi16(__mmask16 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_packs_epi16(__m128i __W, __mmask16 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
(__v16qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_packs_epi16(__mmask32 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_packs_epi16(__m256i __W, __mmask32 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
(__v32qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_packus_epi32(__mmask8 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__M,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_packus_epi32(__m128i __W, __mmask16 __M, __m128i __A, __m128i __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_packus_epi32(__m128i __W, __mmask8 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__M,
(__v8hi)_mm_packus_epi32(__A, __B),
(__v8hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_packus_epi32(__mmask16 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__M,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_packus_epi32(__m256i __W, __mmask16 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__M,
(__v16hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_packus_epi16(__mmask16 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_packus_epi16(__m128i __W, __mmask16 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
(__v16qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_packus_epi16(__mmask32 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_packus_epi16(__m256i __W, __mmask32 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
(__v32qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_adds_epi8(__m128i __W, __mmask16 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_adds_epi8(__mmask16 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_adds_epi8(__m256i __W, __mmask32 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_adds_epi8(__mmask32 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_adds_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_adds_epi16(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_adds_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_adds_epi16(__mmask16 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_adds_epu8(__m128i __W, __mmask16 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_adds_epu8(__mmask16 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_adds_epu8(__m256i __W, __mmask32 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_adds_epu8(__mmask32 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_adds_epu16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_adds_epu16(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_adds_epu16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_adds_epu16(__mmask16 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_avg_epu8(__m128i __W, __mmask16 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_avg_epu8(__mmask16 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_avg_epu8(__m256i __W, __mmask32 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_avg_epu8(__mmask32 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_avg_epu16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_avg_epu16(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_avg_epu16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_avg_epu16(__mmask16 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_max_epi8(__mmask16 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_max_epi8(__m128i __W, __mmask16 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
(__v16qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_max_epi8(__mmask32 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_max_epi8(__m256i __W, __mmask32 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
(__v32qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_max_epi16(__mmask8 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__M,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_max_epi16(__m128i __W, __mmask8 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__M,
(__v8hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_max_epi16(__mmask16 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__M,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_max_epi16(__m256i __W, __mmask16 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__M,
(__v16hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_max_epu8(__mmask16 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_max_epu8(__m128i __W, __mmask16 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
(__v16qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_max_epu8 (__mmask32 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_max_epu8(__m256i __W, __mmask32 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
(__v32qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_max_epu16(__mmask8 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__M,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_max_epu16(__m128i __W, __mmask8 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__M,
(__v8hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_max_epu16(__mmask16 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__M,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_max_epu16(__m256i __W, __mmask16 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__M,
(__v16hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_min_epi8(__mmask16 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_min_epi8(__m128i __W, __mmask16 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
(__v16qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_min_epi8(__mmask32 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_min_epi8(__m256i __W, __mmask32 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
(__v32qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_min_epi16(__mmask8 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__M,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_min_epi16(__m128i __W, __mmask8 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__M,
(__v8hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_min_epi16(__mmask16 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__M,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_min_epi16(__m256i __W, __mmask16 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__M,
(__v16hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_min_epu8(__mmask16 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_min_epu8(__m128i __W, __mmask16 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
(__v16qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_min_epu8 (__mmask32 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_min_epu8(__m256i __W, __mmask32 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__M,
(__v32qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_min_epu16(__mmask8 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__M,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_min_epu16(__m128i __W, __mmask8 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__M,
(__v8hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_min_epu16(__mmask16 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__M,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_min_epu16(__m256i __W, __mmask16 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__M,
(__v16hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_shuffle_epi8(__m128i __W, __mmask16 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_shuffle_epi8(__mmask16 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_shuffle_epi8(__m256i __W, __mmask32 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_shuffle_epi8(__mmask32 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_subs_epi8(__m128i __W, __mmask16 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_subs_epi8(__mmask16 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_subs_epi8(__m256i __W, __mmask32 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_subs_epi8(__mmask32 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_subs_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_subs_epi16(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_subs_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_subs_epi16(__mmask16 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_subs_epu8(__m128i __W, __mmask16 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_subs_epu8(__mmask16 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_subs_epu8(__m256i __W, __mmask32 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_subs_epu8(__mmask32 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_subs_epu16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_subs_epu16(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_subs_epu16(__m256i __W, __mmask16 __U, __m256i __A,
__m256i __B) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_subs_epu16(__mmask16 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask2_permutex2var_epi16 (__m128i __A, __m128i __I, __mmask8 __U,
- __m128i __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_permutex2var_epi16(__m128i __A, __m128i __I, __m128i __B)
{
- return (__m128i) __builtin_ia32_vpermi2varhi128_mask ((__v8hi) __A,
- (__v8hi) __I /* idx */ ,
- (__v8hi) __B,
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_vpermi2varhi128((__v8hi)__A, (__v8hi)__I,
+ (__v8hi) __B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask2_permutex2var_epi16 (__m256i __A, __m256i __I,
- __mmask16 __U, __m256i __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_permutex2var_epi16(__m128i __A, __mmask8 __U, __m128i __I,
+ __m128i __B)
{
- return (__m256i) __builtin_ia32_vpermi2varhi256_mask ((__v16hi) __A,
- (__v16hi) __I /* idx */ ,
- (__v16hi) __B,
- (__mmask16) __U);
+ return (__m128i)__builtin_ia32_selectw_128(__U,
+ (__v8hi)_mm_permutex2var_epi16(__A, __I, __B),
+ (__v8hi)__A);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_permutex2var_epi16 (__m128i __A, __m128i __I, __m128i __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask2_permutex2var_epi16(__m128i __A, __m128i __I, __mmask8 __U,
+ __m128i __B)
{
- return (__m128i) __builtin_ia32_vpermt2varhi128_mask ((__v8hi) __I/* idx */,
- (__v8hi) __A,
- (__v8hi) __B,
- (__mmask8) -1);
+ return (__m128i)__builtin_ia32_selectw_128(__U,
+ (__v8hi)_mm_permutex2var_epi16(__A, __I, __B),
+ (__v8hi)__I);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_permutex2var_epi16 (__m128i __A, __mmask8 __U, __m128i __I,
- __m128i __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_permutex2var_epi16 (__mmask8 __U, __m128i __A, __m128i __I,
+ __m128i __B)
{
- return (__m128i) __builtin_ia32_vpermt2varhi128_mask ((__v8hi) __I/* idx */,
- (__v8hi) __A,
- (__v8hi) __B,
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectw_128(__U,
+ (__v8hi)_mm_permutex2var_epi16(__A, __I, __B),
+ (__v8hi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_permutex2var_epi16 (__mmask8 __U, __m128i __A, __m128i __I,
- __m128i __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_permutex2var_epi16(__m256i __A, __m256i __I, __m256i __B)
{
- return (__m128i) __builtin_ia32_vpermt2varhi128_maskz ((__v8hi) __I/* idx */,
- (__v8hi) __A,
- (__v8hi) __B,
- (__mmask8) __U);
+ return (__m256i)__builtin_ia32_vpermi2varhi256((__v16hi)__A, (__v16hi)__I,
+ (__v16hi)__B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_permutex2var_epi16 (__m256i __A, __m256i __I, __m256i __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_permutex2var_epi16(__m256i __A, __mmask16 __U, __m256i __I,
+ __m256i __B)
{
- return (__m256i) __builtin_ia32_vpermt2varhi256_mask ((__v16hi) __I/* idx */,
- (__v16hi) __A,
- (__v16hi) __B,
- (__mmask16) -1);
+ return (__m256i)__builtin_ia32_selectw_256(__U,
+ (__v16hi)_mm256_permutex2var_epi16(__A, __I, __B),
+ (__v16hi)__A);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_permutex2var_epi16 (__m256i __A, __mmask16 __U,
- __m256i __I, __m256i __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask2_permutex2var_epi16(__m256i __A, __m256i __I, __mmask16 __U,
+ __m256i __B)
{
- return (__m256i) __builtin_ia32_vpermt2varhi256_mask ((__v16hi) __I/* idx */,
- (__v16hi) __A,
- (__v16hi) __B,
- (__mmask16) __U);
+ return (__m256i)__builtin_ia32_selectw_256(__U,
+ (__v16hi)_mm256_permutex2var_epi16(__A, __I, __B),
+ (__v16hi)__I);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_permutex2var_epi16 (__mmask16 __U, __m256i __A,
- __m256i __I, __m256i __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_permutex2var_epi16 (__mmask16 __U, __m256i __A, __m256i __I,
+ __m256i __B)
{
- return (__m256i) __builtin_ia32_vpermt2varhi256_maskz ((__v16hi) __I/* idx */,
- (__v16hi) __A,
- (__v16hi) __B,
- (__mmask16) __U);
+ return (__m256i)__builtin_ia32_selectw_256(__U,
+ (__v16hi)_mm256_permutex2var_epi16(__A, __I, __B),
+ (__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_maddubs_epi16(__m128i __W, __mmask8 __U, __m128i __X, __m128i __Y) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_maddubs_epi16(__X, __Y),
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_maddubs_epi16(__mmask8 __U, __m128i __X, __m128i __Y) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_maddubs_epi16(__X, __Y),
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_maddubs_epi16(__m256i __W, __mmask16 __U, __m256i __X,
__m256i __Y) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_maddubs_epi16(__mmask16 __U, __m256i __X, __m256i __Y) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_maddubs_epi16(__X, __Y),
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_madd_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_madd_epi16(__A, __B),
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_madd_epi16(__mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_madd_epi16(__A, __B),
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_madd_epi16(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_madd_epi16(__A, __B),
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_madd_epi16(__mmask8 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_madd_epi16(__A, __B),
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtsepi16_epi8 (__m128i __A) {
return (__m128i) __builtin_ia32_pmovswb128_mask ((__v8hi) __A,
(__v16qi) _mm_setzero_si128(),
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtsepi16_epi8 (__m128i __O, __mmask8 __M, __m128i __A) {
return (__m128i) __builtin_ia32_pmovswb128_mask ((__v8hi) __A,
(__v16qi) __O,
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtsepi16_epi8 (__mmask8 __M, __m128i __A) {
return (__m128i) __builtin_ia32_pmovswb128_mask ((__v8hi) __A,
(__v16qi) _mm_setzero_si128(),
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtsepi16_epi8 (__m256i __A) {
return (__m128i) __builtin_ia32_pmovswb256_mask ((__v16hi) __A,
(__v16qi) _mm_setzero_si128(),
(__mmask16) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtsepi16_epi8 (__m128i __O, __mmask16 __M, __m256i __A) {
return (__m128i) __builtin_ia32_pmovswb256_mask ((__v16hi) __A,
(__v16qi) __O,
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtsepi16_epi8 (__mmask16 __M, __m256i __A) {
return (__m128i) __builtin_ia32_pmovswb256_mask ((__v16hi) __A,
(__v16qi) _mm_setzero_si128(),
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtusepi16_epi8 (__m128i __A) {
return (__m128i) __builtin_ia32_pmovuswb128_mask ((__v8hi) __A,
(__v16qi) _mm_setzero_si128(),
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtusepi16_epi8 (__m128i __O, __mmask8 __M, __m128i __A) {
return (__m128i) __builtin_ia32_pmovuswb128_mask ((__v8hi) __A,
(__v16qi) __O,
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtusepi16_epi8 (__mmask8 __M, __m128i __A) {
return (__m128i) __builtin_ia32_pmovuswb128_mask ((__v8hi) __A,
(__v16qi) _mm_setzero_si128(),
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtusepi16_epi8 (__m256i __A) {
return (__m128i) __builtin_ia32_pmovuswb256_mask ((__v16hi) __A,
(__v16qi) _mm_setzero_si128(),
(__mmask16) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtusepi16_epi8 (__m128i __O, __mmask16 __M, __m256i __A) {
return (__m128i) __builtin_ia32_pmovuswb256_mask ((__v16hi) __A,
(__v16qi) __O,
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtusepi16_epi8 (__mmask16 __M, __m256i __A) {
return (__m128i) __builtin_ia32_pmovuswb256_mask ((__v16hi) __A,
(__v16qi) _mm_setzero_si128(),
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtepi16_epi8 (__m128i __A) {
-
- return (__m128i) __builtin_ia32_pmovwb128_mask ((__v8hi) __A,
- (__v16qi) _mm_setzero_si128(),
- (__mmask8) -1);
+ return (__m128i)__builtin_shufflevector(
+ __builtin_convertvector((__v8hi)__A, __v8qi),
+ (__v8qi){0, 0, 0, 0, 0, 0, 0, 0}, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11,
+ 12, 13, 14, 15);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi16_epi8 (__m128i __O, __mmask8 __M, __m128i __A) {
return (__m128i) __builtin_ia32_pmovwb128_mask ((__v8hi) __A,
(__v16qi) __O,
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtepi16_epi8 (__mmask8 __M, __m128i __A) {
return (__m128i) __builtin_ia32_pmovwb128_mask ((__v8hi) __A,
(__v16qi) _mm_setzero_si128(),
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi16_storeu_epi8 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovwb128mem_mask ((__v16qi *) __P, (__v8hi) __A, __M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtsepi16_storeu_epi8 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovswb128mem_mask ((__v16qi *) __P, (__v8hi) __A, __M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtusepi16_storeu_epi8 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovuswb128mem_mask ((__v16qi *) __P, (__v8hi) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtepi16_epi8 (__m256i __A) {
- return (__m128i) __builtin_ia32_pmovwb256_mask ((__v16hi) __A,
- (__v16qi) _mm_setzero_si128(),
- (__mmask16) -1);
+ return (__m128i)__builtin_convertvector((__v16hi) __A, __v16qi);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi16_epi8 (__m128i __O, __mmask16 __M, __m256i __A) {
- return (__m128i) __builtin_ia32_pmovwb256_mask ((__v16hi) __A,
- (__v16qi) __O,
- __M);
+ return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
+ (__v16qi)_mm256_cvtepi16_epi8(__A),
+ (__v16qi)__O);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtepi16_epi8 (__mmask16 __M, __m256i __A) {
- return (__m128i) __builtin_ia32_pmovwb256_mask ((__v16hi) __A,
- (__v16qi) _mm_setzero_si128(),
- __M);
+ return (__m128i)__builtin_ia32_selectb_128((__mmask16)__M,
+ (__v16qi)_mm256_cvtepi16_epi8(__A),
+ (__v16qi)_mm_setzero_si128());
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi16_storeu_epi8 (void * __P, __mmask16 __M, __m256i __A)
{
__builtin_ia32_pmovwb256mem_mask ((__v16qi *) __P, (__v16hi) __A, __M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtsepi16_storeu_epi8 (void * __P, __mmask16 __M, __m256i __A)
{
__builtin_ia32_pmovswb256mem_mask ((__v16qi *) __P, (__v16hi) __A, __M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm256_mask_cvtusepi16_storeu_epi8 (void * __P, __mmask8 __M, __m256i __A)
+static __inline__ void __DEFAULT_FN_ATTRS256
+_mm256_mask_cvtusepi16_storeu_epi8 (void * __P, __mmask16 __M, __m256i __A)
{
__builtin_ia32_pmovuswb256mem_mask ((__v16qi*) __P, (__v16hi) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_mulhrs_epi16(__m128i __W, __mmask8 __U, __m128i __X, __m128i __Y) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_mulhrs_epi16(__X, __Y),
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_mulhrs_epi16(__mmask8 __U, __m128i __X, __m128i __Y) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_mulhrs_epi16(__X, __Y),
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_mulhrs_epi16(__m256i __W, __mmask16 __U, __m256i __X, __m256i __Y) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_mulhrs_epi16(__X, __Y),
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_mulhrs_epi16(__mmask16 __U, __m256i __X, __m256i __Y) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_mulhrs_epi16(__X, __Y),
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_mulhi_epu16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_mulhi_epu16(__A, __B),
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_mulhi_epu16(__mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_mulhi_epu16(__A, __B),
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_mulhi_epu16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_mulhi_epu16(__A, __B),
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_mulhi_epu16(__mmask16 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_mulhi_epu16(__A, __B),
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_mulhi_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_mulhi_epi16(__A, __B),
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_mulhi_epi16(__mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_mulhi_epi16(__A, __B),
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_mulhi_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_mulhi_epi16(__A, __B),
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_mulhi_epi16(__mmask16 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_mulhi_epi16(__A, __B),
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_unpackhi_epi8(__m128i __W, __mmask16 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)_mm_unpackhi_epi8(__A, __B),
(__v16qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_unpackhi_epi8(__mmask16 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)_mm_unpackhi_epi8(__A, __B),
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_unpackhi_epi8(__m256i __W, __mmask32 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)_mm256_unpackhi_epi8(__A, __B),
(__v32qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_unpackhi_epi8(__mmask32 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)_mm256_unpackhi_epi8(__A, __B),
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_unpackhi_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_unpackhi_epi16(__A, __B),
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_unpackhi_epi16(__mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_unpackhi_epi16(__A, __B),
(__v8hi) _mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_unpackhi_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_unpackhi_epi16(__A, __B),
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_unpackhi_epi16(__mmask16 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_unpackhi_epi16(__A, __B),
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_unpacklo_epi8(__m128i __W, __mmask16 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)_mm_unpacklo_epi8(__A, __B),
(__v16qi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_unpacklo_epi8(__mmask16 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectb_128((__mmask16)__U,
(__v16qi)_mm_unpacklo_epi8(__A, __B),
(__v16qi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_unpacklo_epi8(__m256i __W, __mmask32 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)_mm256_unpacklo_epi8(__A, __B),
(__v32qi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_unpacklo_epi8(__mmask32 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectb_256((__mmask32)__U,
(__v32qi)_mm256_unpacklo_epi8(__A, __B),
(__v32qi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_unpacklo_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_unpacklo_epi16(__A, __B),
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_unpacklo_epi16(__mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_unpacklo_epi16(__A, __B),
(__v8hi) _mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_unpacklo_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_unpacklo_epi16(__A, __B),
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_unpacklo_epi16(__mmask16 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_unpacklo_epi16(__A, __B),
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi8_epi16(__m128i __W, __mmask8 __U, __m128i __A)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtepi8_epi16(__mmask8 __U, __m128i __A)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi8_epi16(__m256i __W, __mmask16 __U, __m128i __A)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtepi8_epi16(__mmask16 __U, __m128i __A)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtepu8_epi16(__m128i __W, __mmask8 __U, __m128i __A)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtepu8_epi16(__mmask8 __U, __m128i __A)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepu8_epi16(__m256i __W, __mmask16 __U, __m128i __A)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtepu8_epi16 (__mmask16 __U, __m128i __A)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
}
-#define _mm_cmp_epi8_mask(a, b, p) __extension__ ({ \
- (__mmask16)__builtin_ia32_cmpb128_mask((__v16qi)(__m128i)(a), \
- (__v16qi)(__m128i)(b), (int)(p), \
- (__mmask16)-1); })
-
-#define _mm_mask_cmp_epi8_mask(m, a, b, p) __extension__ ({ \
- (__mmask16)__builtin_ia32_cmpb128_mask((__v16qi)(__m128i)(a), \
- (__v16qi)(__m128i)(b), (int)(p), \
- (__mmask16)(m)); })
-
-#define _mm_cmp_epu8_mask(a, b, p) __extension__ ({ \
- (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)(__m128i)(a), \
- (__v16qi)(__m128i)(b), (int)(p), \
- (__mmask16)-1); })
-
-#define _mm_mask_cmp_epu8_mask(m, a, b, p) __extension__ ({ \
- (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)(__m128i)(a), \
- (__v16qi)(__m128i)(b), (int)(p), \
- (__mmask16)(m)); })
-
-#define _mm256_cmp_epi8_mask(a, b, p) __extension__ ({ \
- (__mmask32)__builtin_ia32_cmpb256_mask((__v32qi)(__m256i)(a), \
- (__v32qi)(__m256i)(b), (int)(p), \
- (__mmask32)-1); })
-
-#define _mm256_mask_cmp_epi8_mask(m, a, b, p) __extension__ ({ \
- (__mmask32)__builtin_ia32_cmpb256_mask((__v32qi)(__m256i)(a), \
- (__v32qi)(__m256i)(b), (int)(p), \
- (__mmask32)(m)); })
-
-#define _mm256_cmp_epu8_mask(a, b, p) __extension__ ({ \
- (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)(__m256i)(a), \
- (__v32qi)(__m256i)(b), (int)(p), \
- (__mmask32)-1); })
-
-#define _mm256_mask_cmp_epu8_mask(m, a, b, p) __extension__ ({ \
- (__mmask32)__builtin_ia32_ucmpb256_mask((__v32qi)(__m256i)(a), \
- (__v32qi)(__m256i)(b), (int)(p), \
- (__mmask32)(m)); })
-
-#define _mm_cmp_epi16_mask(a, b, p) __extension__ ({ \
- (__mmask8)__builtin_ia32_cmpw128_mask((__v8hi)(__m128i)(a), \
- (__v8hi)(__m128i)(b), (int)(p), \
- (__mmask8)-1); })
-
-#define _mm_mask_cmp_epi16_mask(m, a, b, p) __extension__ ({ \
- (__mmask8)__builtin_ia32_cmpw128_mask((__v8hi)(__m128i)(a), \
- (__v8hi)(__m128i)(b), (int)(p), \
- (__mmask8)(m)); })
-
-#define _mm_cmp_epu16_mask(a, b, p) __extension__ ({ \
- (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)(__m128i)(a), \
- (__v8hi)(__m128i)(b), (int)(p), \
- (__mmask8)-1); })
-
-#define _mm_mask_cmp_epu16_mask(m, a, b, p) __extension__ ({ \
- (__mmask8)__builtin_ia32_ucmpw128_mask((__v8hi)(__m128i)(a), \
- (__v8hi)(__m128i)(b), (int)(p), \
- (__mmask8)(m)); })
-
-#define _mm256_cmp_epi16_mask(a, b, p) __extension__ ({ \
- (__mmask16)__builtin_ia32_cmpw256_mask((__v16hi)(__m256i)(a), \
- (__v16hi)(__m256i)(b), (int)(p), \
- (__mmask16)-1); })
-
-#define _mm256_mask_cmp_epi16_mask(m, a, b, p) __extension__ ({ \
- (__mmask16)__builtin_ia32_cmpw256_mask((__v16hi)(__m256i)(a), \
- (__v16hi)(__m256i)(b), (int)(p), \
- (__mmask16)(m)); })
-
-#define _mm256_cmp_epu16_mask(a, b, p) __extension__ ({ \
- (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)(__m256i)(a), \
- (__v16hi)(__m256i)(b), (int)(p), \
- (__mmask16)-1); })
-
-#define _mm256_mask_cmp_epu16_mask(m, a, b, p) __extension__ ({ \
- (__mmask16)__builtin_ia32_ucmpw256_mask((__v16hi)(__m256i)(a), \
- (__v16hi)(__m256i)(b), (int)(p), \
- (__mmask16)(m)); })
-
-#define _mm_mask_shufflehi_epi16(W, U, A, imm) __extension__ ({ \
+#define _mm_mask_shufflehi_epi16(W, U, A, imm) \
(__m128i)__builtin_ia32_selectw_128((__mmask8)(U), \
(__v8hi)_mm_shufflehi_epi16((A), (imm)), \
- (__v8hi)(__m128i)(W)); })
+ (__v8hi)(__m128i)(W))
-#define _mm_maskz_shufflehi_epi16(U, A, imm) __extension__ ({ \
+#define _mm_maskz_shufflehi_epi16(U, A, imm) \
(__m128i)__builtin_ia32_selectw_128((__mmask8)(U), \
(__v8hi)_mm_shufflehi_epi16((A), (imm)), \
- (__v8hi)_mm_setzero_hi()); })
+ (__v8hi)_mm_setzero_si128())
-#define _mm256_mask_shufflehi_epi16(W, U, A, imm) __extension__ ({ \
+#define _mm256_mask_shufflehi_epi16(W, U, A, imm) \
(__m256i)__builtin_ia32_selectw_256((__mmask16)(U), \
(__v16hi)_mm256_shufflehi_epi16((A), (imm)), \
- (__v16hi)(__m256i)(W)); })
+ (__v16hi)(__m256i)(W))
-#define _mm256_maskz_shufflehi_epi16(U, A, imm) __extension__ ({ \
+#define _mm256_maskz_shufflehi_epi16(U, A, imm) \
(__m256i)__builtin_ia32_selectw_256((__mmask16)(U), \
(__v16hi)_mm256_shufflehi_epi16((A), (imm)), \
- (__v16hi)_mm256_setzero_si256()); })
+ (__v16hi)_mm256_setzero_si256())
-#define _mm_mask_shufflelo_epi16(W, U, A, imm) __extension__ ({ \
+#define _mm_mask_shufflelo_epi16(W, U, A, imm) \
(__m128i)__builtin_ia32_selectw_128((__mmask8)(U), \
(__v8hi)_mm_shufflelo_epi16((A), (imm)), \
- (__v8hi)(__m128i)(W)); })
+ (__v8hi)(__m128i)(W))
-#define _mm_maskz_shufflelo_epi16(U, A, imm) __extension__ ({ \
+#define _mm_maskz_shufflelo_epi16(U, A, imm) \
(__m128i)__builtin_ia32_selectw_128((__mmask8)(U), \
(__v8hi)_mm_shufflelo_epi16((A), (imm)), \
- (__v8hi)_mm_setzero_hi()); })
+ (__v8hi)_mm_setzero_si128())
-#define _mm256_mask_shufflelo_epi16(W, U, A, imm) __extension__ ({ \
+#define _mm256_mask_shufflelo_epi16(W, U, A, imm) \
(__m256i)__builtin_ia32_selectw_256((__mmask16)(U), \
(__v16hi)_mm256_shufflelo_epi16((A), \
(imm)), \
- (__v16hi)(__m256i)(W)); })
+ (__v16hi)(__m256i)(W))
-#define _mm256_maskz_shufflelo_epi16(U, A, imm) __extension__ ({ \
+#define _mm256_maskz_shufflelo_epi16(U, A, imm) \
(__m256i)__builtin_ia32_selectw_256((__mmask16)(U), \
(__v16hi)_mm256_shufflelo_epi16((A), \
(imm)), \
- (__v16hi)_mm256_setzero_si256()); })
+ (__v16hi)_mm256_setzero_si256())
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sllv_epi16(__m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_psllv16hi((__v16hi)__A, (__v16hi)__B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_sllv_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_sllv_epi16(__mmask16 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_sllv_epi16(__m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_psllv8hi((__v8hi)__A, (__v8hi)__B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_sllv_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_sllv_epi16(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_sll_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_sll_epi16 (__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_sll_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_sll_epi16(__mmask16 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_slli_epi16(__m128i __W, __mmask8 __U, __m128i __A, int __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_slli_epi16(__m128i __W, __mmask8 __U, __m128i __A, unsigned int __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_slli_epi16(__A, __B),
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_slli_epi16 (__mmask8 __U, __m128i __A, int __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_slli_epi16 (__mmask8 __U, __m128i __A, unsigned int __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_slli_epi16(__A, __B),
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_slli_epi16(__m256i __W, __mmask16 __U, __m256i __A, int __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_slli_epi16(__m256i __W, __mmask16 __U, __m256i __A,
+ unsigned int __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_slli_epi16(__A, __B),
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_slli_epi16(__mmask16 __U, __m256i __A, int __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_slli_epi16(__mmask16 __U, __m256i __A, unsigned int __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_slli_epi16(__A, __B),
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_srlv_epi16(__m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_psrlv16hi((__v16hi)__A, (__v16hi)__B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_srlv_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_srlv_epi16(__mmask16 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_srlv_epi16(__m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_psrlv8hi((__v8hi)__A, (__v8hi)__B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_srlv_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_srlv_epi16(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_srav_epi16(__m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_psrav16hi((__v16hi)__A, (__v16hi)__B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_srav_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_srav_epi16(__mmask16 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_srav_epi16(__m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_psrav8hi((__v8hi)__A, (__v8hi)__B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_srav_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_srav_epi16(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_sra_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_sra_epi16(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_sra_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_sra_epi16(__mmask16 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_srai_epi16(__m128i __W, __mmask8 __U, __m128i __A, int __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_srai_epi16(__m128i __W, __mmask8 __U, __m128i __A, unsigned int __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_srai_epi16(__A, __B),
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_srai_epi16(__mmask8 __U, __m128i __A, int __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_srai_epi16(__mmask8 __U, __m128i __A, unsigned int __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_srai_epi16(__A, __B),
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_srai_epi16(__m256i __W, __mmask16 __U, __m256i __A, int __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_srai_epi16(__m256i __W, __mmask16 __U, __m256i __A,
+ unsigned int __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_srai_epi16(__A, __B),
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_srai_epi16(__mmask16 __U, __m256i __A, int __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_srai_epi16(__mmask16 __U, __m256i __A, unsigned int __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_srai_epi16(__A, __B),
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_srl_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_srl_epi16 (__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_srl_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_srl_epi16(__mmask16 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_srli_epi16(__m128i __W, __mmask8 __U, __m128i __A, int __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_srli_epi16 (__mmask8 __U, __m128i __A, int __B)
{
return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
(__v8hi)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_srli_epi16(__m256i __W, __mmask16 __U, __m256i __A, int __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_srli_epi16(__mmask16 __U, __m256i __A, int __B)
{
return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
(__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_mov_epi16 (__m128i __W, __mmask8 __U, __m128i __A)
{
return (__m128i) __builtin_ia32_selectw_128 ((__mmask8) __U,
(__v8hi) __W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_mov_epi16 (__mmask8 __U, __m128i __A)
{
return (__m128i) __builtin_ia32_selectw_128 ((__mmask8) __U,
(__v8hi) __A,
- (__v8hi) _mm_setzero_hi ());
+ (__v8hi) _mm_setzero_si128 ());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_mov_epi16 (__m256i __W, __mmask16 __U, __m256i __A)
{
return (__m256i) __builtin_ia32_selectw_256 ((__mmask16) __U,
(__v16hi) __W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_mov_epi16 (__mmask16 __U, __m256i __A)
{
return (__m256i) __builtin_ia32_selectw_256 ((__mmask16) __U,
(__v16hi) _mm256_setzero_si256 ());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_mov_epi8 (__m128i __W, __mmask16 __U, __m128i __A)
{
return (__m128i) __builtin_ia32_selectb_128 ((__mmask16) __U,
(__v16qi) __W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_mov_epi8 (__mmask16 __U, __m128i __A)
{
return (__m128i) __builtin_ia32_selectb_128 ((__mmask16) __U,
(__v16qi) __A,
- (__v16qi) _mm_setzero_hi ());
+ (__v16qi) _mm_setzero_si128 ());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_mov_epi8 (__m256i __W, __mmask32 __U, __m256i __A)
{
return (__m256i) __builtin_ia32_selectb_256 ((__mmask32) __U,
(__v32qi) __W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_mov_epi8 (__mmask32 __U, __m256i __A)
{
return (__m256i) __builtin_ia32_selectb_256 ((__mmask32) __U,
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_set1_epi8 (__m128i __O, __mmask16 __M, char __A)
{
- return (__m128i) __builtin_ia32_pbroadcastb128_gpr_mask (__A,
- (__v16qi) __O,
- __M);
+ return (__m128i) __builtin_ia32_selectb_128(__M,
+ (__v16qi) _mm_set1_epi8(__A),
+ (__v16qi) __O);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_set1_epi8 (__mmask16 __M, char __A)
{
- return (__m128i) __builtin_ia32_pbroadcastb128_gpr_mask (__A,
- (__v16qi)
- _mm_setzero_si128 (),
- __M);
+ return (__m128i) __builtin_ia32_selectb_128(__M,
+ (__v16qi) _mm_set1_epi8(__A),
+ (__v16qi) _mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_set1_epi8 (__m256i __O, __mmask32 __M, char __A)
{
- return (__m256i) __builtin_ia32_pbroadcastb256_gpr_mask (__A,
- (__v32qi) __O,
- __M);
+ return (__m256i) __builtin_ia32_selectb_256(__M,
+ (__v32qi) _mm256_set1_epi8(__A),
+ (__v32qi) __O);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_set1_epi8 (__mmask32 __M, char __A)
{
- return (__m256i) __builtin_ia32_pbroadcastb256_gpr_mask (__A,
- (__v32qi)
- _mm256_setzero_si256 (),
- __M);
+ return (__m256i) __builtin_ia32_selectb_256(__M,
+ (__v32qi) _mm256_set1_epi8(__A),
+ (__v32qi) _mm256_setzero_si256());
+}
+
+static __inline __m128i __DEFAULT_FN_ATTRS128
+_mm_loadu_epi16 (void const *__P)
+{
+ struct __loadu_epi16 {
+ __m128i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ return ((const struct __loadu_epi16*)__P)->__v;
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_loadu_epi16 (__m128i __W, __mmask8 __U, void const *__P)
{
- return (__m128i) __builtin_ia32_loaddquhi128_mask ((__v8hi *) __P,
+ return (__m128i) __builtin_ia32_loaddquhi128_mask ((const __v8hi *) __P,
(__v8hi) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_loadu_epi16 (__mmask8 __U, void const *__P)
{
- return (__m128i) __builtin_ia32_loaddquhi128_mask ((__v8hi *) __P,
+ return (__m128i) __builtin_ia32_loaddquhi128_mask ((const __v8hi *) __P,
(__v8hi)
- _mm_setzero_hi (),
+ _mm_setzero_si128 (),
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline __m256i __DEFAULT_FN_ATTRS256
+_mm256_loadu_epi16 (void const *__P)
+{
+ struct __loadu_epi16 {
+ __m256i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ return ((const struct __loadu_epi16*)__P)->__v;
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_loadu_epi16 (__m256i __W, __mmask16 __U, void const *__P)
{
- return (__m256i) __builtin_ia32_loaddquhi256_mask ((__v16hi *) __P,
+ return (__m256i) __builtin_ia32_loaddquhi256_mask ((const __v16hi *) __P,
(__v16hi) __W,
(__mmask16) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_loadu_epi16 (__mmask16 __U, void const *__P)
{
- return (__m256i) __builtin_ia32_loaddquhi256_mask ((__v16hi *) __P,
+ return (__m256i) __builtin_ia32_loaddquhi256_mask ((const __v16hi *) __P,
(__v16hi)
_mm256_setzero_si256 (),
(__mmask16) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline __m128i __DEFAULT_FN_ATTRS128
+_mm_loadu_epi8 (void const *__P)
+{
+ struct __loadu_epi8 {
+ __m128i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ return ((const struct __loadu_epi8*)__P)->__v;
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_loadu_epi8 (__m128i __W, __mmask16 __U, void const *__P)
{
- return (__m128i) __builtin_ia32_loaddquqi128_mask ((__v16qi *) __P,
+ return (__m128i) __builtin_ia32_loaddquqi128_mask ((const __v16qi *) __P,
(__v16qi) __W,
(__mmask16) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_loadu_epi8 (__mmask16 __U, void const *__P)
{
- return (__m128i) __builtin_ia32_loaddquqi128_mask ((__v16qi *) __P,
+ return (__m128i) __builtin_ia32_loaddquqi128_mask ((const __v16qi *) __P,
(__v16qi)
_mm_setzero_si128 (),
(__mmask16) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline __m256i __DEFAULT_FN_ATTRS256
+_mm256_loadu_epi8 (void const *__P)
+{
+ struct __loadu_epi8 {
+ __m256i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ return ((const struct __loadu_epi8*)__P)->__v;
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_loadu_epi8 (__m256i __W, __mmask32 __U, void const *__P)
{
- return (__m256i) __builtin_ia32_loaddquqi256_mask ((__v32qi *) __P,
+ return (__m256i) __builtin_ia32_loaddquqi256_mask ((const __v32qi *) __P,
(__v32qi) __W,
(__mmask32) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_loadu_epi8 (__mmask32 __U, void const *__P)
{
- return (__m256i) __builtin_ia32_loaddquqi256_mask ((__v32qi *) __P,
+ return (__m256i) __builtin_ia32_loaddquqi256_mask ((const __v32qi *) __P,
(__v32qi)
_mm256_setzero_si256 (),
(__mmask32) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+
+static __inline void __DEFAULT_FN_ATTRS128
+_mm_storeu_epi16 (void *__P, __m128i __A)
+{
+ struct __storeu_epi16 {
+ __m128i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_epi16*)__P)->__v = __A;
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_storeu_epi16 (void *__P, __mmask8 __U, __m128i __A)
{
__builtin_ia32_storedquhi128_mask ((__v8hi *) __P,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS256
+_mm256_storeu_epi16 (void *__P, __m256i __A)
+{
+ struct __storeu_epi16 {
+ __m256i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_epi16*)__P)->__v = __A;
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_storeu_epi16 (void *__P, __mmask16 __U, __m256i __A)
{
__builtin_ia32_storedquhi256_mask ((__v16hi *) __P,
(__mmask16) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS128
+_mm_storeu_epi8 (void *__P, __m128i __A)
+{
+ struct __storeu_epi8 {
+ __m128i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_epi8*)__P)->__v = __A;
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_storeu_epi8 (void *__P, __mmask16 __U, __m128i __A)
{
__builtin_ia32_storedquqi128_mask ((__v16qi *) __P,
(__mmask16) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS256
+_mm256_storeu_epi8 (void *__P, __m256i __A)
+{
+ struct __storeu_epi8 {
+ __m256i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_epi8*)__P)->__v = __A;
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_storeu_epi8 (void *__P, __mmask32 __U, __m256i __A)
{
__builtin_ia32_storedquqi256_mask ((__v32qi *) __P,
(__mmask32) __U);
}
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS128
_mm_test_epi8_mask (__m128i __A, __m128i __B)
{
- return (__mmask16) __builtin_ia32_ptestmb128 ((__v16qi) __A,
- (__v16qi) __B,
- (__mmask16) -1);
+ return _mm_cmpneq_epi8_mask (_mm_and_si128(__A, __B), _mm_setzero_si128());
}
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS128
_mm_mask_test_epi8_mask (__mmask16 __U, __m128i __A, __m128i __B)
{
- return (__mmask16) __builtin_ia32_ptestmb128 ((__v16qi) __A,
- (__v16qi) __B, __U);
+ return _mm_mask_cmpneq_epi8_mask (__U, _mm_and_si128 (__A, __B),
+ _mm_setzero_si128());
}
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS256
_mm256_test_epi8_mask (__m256i __A, __m256i __B)
{
- return (__mmask32) __builtin_ia32_ptestmb256 ((__v32qi) __A,
- (__v32qi) __B,
- (__mmask32) -1);
+ return _mm256_cmpneq_epi8_mask (_mm256_and_si256(__A, __B),
+ _mm256_setzero_si256());
}
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS256
_mm256_mask_test_epi8_mask (__mmask32 __U, __m256i __A, __m256i __B)
{
- return (__mmask32) __builtin_ia32_ptestmb256 ((__v32qi) __A,
- (__v32qi) __B, __U);
+ return _mm256_mask_cmpneq_epi8_mask (__U, _mm256_and_si256(__A, __B),
+ _mm256_setzero_si256());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS128
_mm_test_epi16_mask (__m128i __A, __m128i __B)
{
- return (__mmask8) __builtin_ia32_ptestmw128 ((__v8hi) __A,
- (__v8hi) __B,
- (__mmask8) -1);
+ return _mm_cmpneq_epi16_mask (_mm_and_si128 (__A, __B), _mm_setzero_si128());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS128
_mm_mask_test_epi16_mask (__mmask8 __U, __m128i __A, __m128i __B)
{
- return (__mmask8) __builtin_ia32_ptestmw128 ((__v8hi) __A,
- (__v8hi) __B, __U);
+ return _mm_mask_cmpneq_epi16_mask (__U, _mm_and_si128 (__A, __B),
+ _mm_setzero_si128());
}
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS256
_mm256_test_epi16_mask (__m256i __A, __m256i __B)
{
- return (__mmask16) __builtin_ia32_ptestmw256 ((__v16hi) __A,
- (__v16hi) __B,
- (__mmask16) -1);
+ return _mm256_cmpneq_epi16_mask (_mm256_and_si256 (__A, __B),
+ _mm256_setzero_si256 ());
}
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS256
_mm256_mask_test_epi16_mask (__mmask16 __U, __m256i __A, __m256i __B)
{
- return (__mmask16) __builtin_ia32_ptestmw256 ((__v16hi) __A,
- (__v16hi) __B, __U);
+ return _mm256_mask_cmpneq_epi16_mask (__U, _mm256_and_si256(__A, __B),
+ _mm256_setzero_si256());
}
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS128
_mm_testn_epi8_mask (__m128i __A, __m128i __B)
{
- return (__mmask16) __builtin_ia32_ptestnmb128 ((__v16qi) __A,
- (__v16qi) __B,
- (__mmask16) -1);
+ return _mm_cmpeq_epi8_mask (_mm_and_si128 (__A, __B), _mm_setzero_si128());
}
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS128
_mm_mask_testn_epi8_mask (__mmask16 __U, __m128i __A, __m128i __B)
{
- return (__mmask16) __builtin_ia32_ptestnmb128 ((__v16qi) __A,
- (__v16qi) __B, __U);
+ return _mm_mask_cmpeq_epi8_mask (__U, _mm_and_si128 (__A, __B),
+ _mm_setzero_si128());
}
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS256
_mm256_testn_epi8_mask (__m256i __A, __m256i __B)
{
- return (__mmask32) __builtin_ia32_ptestnmb256 ((__v32qi) __A,
- (__v32qi) __B,
- (__mmask32) -1);
+ return _mm256_cmpeq_epi8_mask (_mm256_and_si256 (__A, __B),
+ _mm256_setzero_si256());
}
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS256
_mm256_mask_testn_epi8_mask (__mmask32 __U, __m256i __A, __m256i __B)
{
- return (__mmask32) __builtin_ia32_ptestnmb256 ((__v32qi) __A,
- (__v32qi) __B, __U);
+ return _mm256_mask_cmpeq_epi8_mask (__U, _mm256_and_si256 (__A, __B),
+ _mm256_setzero_si256());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS128
_mm_testn_epi16_mask (__m128i __A, __m128i __B)
{
- return (__mmask8) __builtin_ia32_ptestnmw128 ((__v8hi) __A,
- (__v8hi) __B,
- (__mmask8) -1);
+ return _mm_cmpeq_epi16_mask (_mm_and_si128 (__A, __B), _mm_setzero_si128());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS128
_mm_mask_testn_epi16_mask (__mmask8 __U, __m128i __A, __m128i __B)
{
- return (__mmask8) __builtin_ia32_ptestnmw128 ((__v8hi) __A,
- (__v8hi) __B, __U);
+ return _mm_mask_cmpeq_epi16_mask (__U, _mm_and_si128(__A, __B), _mm_setzero_si128());
}
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS256
_mm256_testn_epi16_mask (__m256i __A, __m256i __B)
{
- return (__mmask16) __builtin_ia32_ptestnmw256 ((__v16hi) __A,
- (__v16hi) __B,
- (__mmask16) -1);
+ return _mm256_cmpeq_epi16_mask (_mm256_and_si256(__A, __B),
+ _mm256_setzero_si256());
}
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS256
_mm256_mask_testn_epi16_mask (__mmask16 __U, __m256i __A, __m256i __B)
{
- return (__mmask16) __builtin_ia32_ptestnmw256 ((__v16hi) __A,
- (__v16hi) __B, __U);
+ return _mm256_mask_cmpeq_epi16_mask (__U, _mm256_and_si256 (__A, __B),
+ _mm256_setzero_si256());
}
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS128
_mm_movepi8_mask (__m128i __A)
{
return (__mmask16) __builtin_ia32_cvtb2mask128 ((__v16qi) __A);
}
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
+static __inline__ __mmask32 __DEFAULT_FN_ATTRS256
_mm256_movepi8_mask (__m256i __A)
{
return (__mmask32) __builtin_ia32_cvtb2mask256 ((__v32qi) __A);
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS128
_mm_movepi16_mask (__m128i __A)
{
return (__mmask8) __builtin_ia32_cvtw2mask128 ((__v8hi) __A);
}
-static __inline__ __mmask16 __DEFAULT_FN_ATTRS
+static __inline__ __mmask16 __DEFAULT_FN_ATTRS256
_mm256_movepi16_mask (__m256i __A)
{
return (__mmask16) __builtin_ia32_cvtw2mask256 ((__v16hi) __A);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_movm_epi8 (__mmask16 __A)
{
return (__m128i) __builtin_ia32_cvtmask2b128 (__A);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_movm_epi8 (__mmask32 __A)
{
return (__m256i) __builtin_ia32_cvtmask2b256 (__A);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_movm_epi16 (__mmask8 __A)
{
return (__m128i) __builtin_ia32_cvtmask2w128 (__A);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_movm_epi16 (__mmask16 __A)
{
return (__m256i) __builtin_ia32_cvtmask2w256 (__A);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_broadcastb_epi8 (__m128i __O, __mmask16 __M, __m128i __A)
{
return (__m128i)__builtin_ia32_selectb_128(__M,
(__v16qi) __O);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_broadcastb_epi8 (__mmask16 __M, __m128i __A)
{
return (__m128i)__builtin_ia32_selectb_128(__M,
(__v16qi) _mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_broadcastb_epi8 (__m256i __O, __mmask32 __M, __m128i __A)
{
return (__m256i)__builtin_ia32_selectb_256(__M,
(__v32qi) __O);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_broadcastb_epi8 (__mmask32 __M, __m128i __A)
{
return (__m256i)__builtin_ia32_selectb_256(__M,
(__v32qi) _mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_broadcastw_epi16 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i)__builtin_ia32_selectw_128(__M,
(__v8hi) __O);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_broadcastw_epi16 (__mmask8 __M, __m128i __A)
{
return (__m128i)__builtin_ia32_selectw_128(__M,
(__v8hi) _mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_broadcastw_epi16 (__m256i __O, __mmask16 __M, __m128i __A)
{
return (__m256i)__builtin_ia32_selectw_256(__M,
(__v16hi) __O);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_broadcastw_epi16 (__mmask16 __M, __m128i __A)
{
return (__m256i)__builtin_ia32_selectw_256(__M,
(__v16hi) _mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_set1_epi16 (__m256i __O, __mmask16 __M, short __A)
{
- return (__m256i) __builtin_ia32_pbroadcastw256_gpr_mask (__A,
- (__v16hi) __O,
- __M);
+ return (__m256i) __builtin_ia32_selectw_256 (__M,
+ (__v16hi) _mm256_set1_epi16(__A),
+ (__v16hi) __O);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_set1_epi16 (__mmask16 __M, short __A)
{
- return (__m256i) __builtin_ia32_pbroadcastw256_gpr_mask (__A,
- (__v16hi) _mm256_setzero_si256 (),
- __M);
+ return (__m256i) __builtin_ia32_selectw_256(__M,
+ (__v16hi)_mm256_set1_epi16(__A),
+ (__v16hi) _mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_set1_epi16 (__m128i __O, __mmask8 __M, short __A)
{
- return (__m128i) __builtin_ia32_pbroadcastw128_gpr_mask (__A,
- (__v8hi) __O,
- __M);
+ return (__m128i) __builtin_ia32_selectw_128(__M,
+ (__v8hi) _mm_set1_epi16(__A),
+ (__v8hi) __O);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_set1_epi16 (__mmask8 __M, short __A)
{
- return (__m128i) __builtin_ia32_pbroadcastw128_gpr_mask (__A,
- (__v8hi) _mm_setzero_si128 (),
- __M);
+ return (__m128i) __builtin_ia32_selectw_128(__M,
+ (__v8hi) _mm_set1_epi16(__A),
+ (__v8hi) _mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_permutexvar_epi16 (__m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_permvarhi128_mask ((__v8hi) __B,
- (__v8hi) __A,
- (__v8hi) _mm_undefined_si128 (),
- (__mmask8) -1);
+ return (__m128i)__builtin_ia32_permvarhi128((__v8hi) __B, (__v8hi) __A);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_permutexvar_epi16 (__mmask8 __M, __m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_permvarhi128_mask ((__v8hi) __B,
- (__v8hi) __A,
- (__v8hi) _mm_setzero_si128 (),
- (__mmask8) __M);
+ return (__m128i)__builtin_ia32_selectw_128((__mmask8)__M,
+ (__v8hi)_mm_permutexvar_epi16(__A, __B),
+ (__v8hi) _mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_permutexvar_epi16 (__m128i __W, __mmask8 __M, __m128i __A,
__m128i __B)
{
- return (__m128i) __builtin_ia32_permvarhi128_mask ((__v8hi) __B,
- (__v8hi) __A,
- (__v8hi) __W,
- (__mmask8) __M);
+ return (__m128i)__builtin_ia32_selectw_128((__mmask8)__M,
+ (__v8hi)_mm_permutexvar_epi16(__A, __B),
+ (__v8hi)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_permutexvar_epi16 (__m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_permvarhi256_mask ((__v16hi) __B,
- (__v16hi) __A,
- (__v16hi) _mm256_undefined_si256 (),
- (__mmask16) -1);
+ return (__m256i)__builtin_ia32_permvarhi256((__v16hi) __B, (__v16hi) __A);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_permutexvar_epi16 (__mmask16 __M, __m256i __A,
__m256i __B)
{
- return (__m256i) __builtin_ia32_permvarhi256_mask ((__v16hi) __B,
- (__v16hi) __A,
- (__v16hi) _mm256_setzero_si256 (),
- (__mmask16) __M);
+ return (__m256i)__builtin_ia32_selectw_256((__mmask16)__M,
+ (__v16hi)_mm256_permutexvar_epi16(__A, __B),
+ (__v16hi)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_permutexvar_epi16 (__m256i __W, __mmask16 __M, __m256i __A,
__m256i __B)
{
- return (__m256i) __builtin_ia32_permvarhi256_mask ((__v16hi) __B,
- (__v16hi) __A,
- (__v16hi) __W,
- (__mmask16) __M);
+ return (__m256i)__builtin_ia32_selectw_256((__mmask16)__M,
+ (__v16hi)_mm256_permutexvar_epi16(__A, __B),
+ (__v16hi)__W);
}
-#define _mm_mask_alignr_epi8(W, U, A, B, N) __extension__ ({ \
+#define _mm_mask_alignr_epi8(W, U, A, B, N) \
(__m128i)__builtin_ia32_selectb_128((__mmask16)(U), \
(__v16qi)_mm_alignr_epi8((A), (B), (int)(N)), \
- (__v16qi)(__m128i)(W)); })
+ (__v16qi)(__m128i)(W))
-#define _mm_maskz_alignr_epi8(U, A, B, N) __extension__ ({ \
+#define _mm_maskz_alignr_epi8(U, A, B, N) \
(__m128i)__builtin_ia32_selectb_128((__mmask16)(U), \
(__v16qi)_mm_alignr_epi8((A), (B), (int)(N)), \
- (__v16qi)_mm_setzero_si128()); })
+ (__v16qi)_mm_setzero_si128())
-#define _mm256_mask_alignr_epi8(W, U, A, B, N) __extension__ ({ \
+#define _mm256_mask_alignr_epi8(W, U, A, B, N) \
(__m256i)__builtin_ia32_selectb_256((__mmask32)(U), \
(__v32qi)_mm256_alignr_epi8((A), (B), (int)(N)), \
- (__v32qi)(__m256i)(W)); })
+ (__v32qi)(__m256i)(W))
-#define _mm256_maskz_alignr_epi8(U, A, B, N) __extension__ ({ \
+#define _mm256_maskz_alignr_epi8(U, A, B, N) \
(__m256i)__builtin_ia32_selectb_256((__mmask32)(U), \
(__v32qi)_mm256_alignr_epi8((A), (B), (int)(N)), \
- (__v32qi)_mm256_setzero_si256()); })
-
-#define _mm_dbsad_epu8(A, B, imm) __extension__ ({ \
- (__m128i)__builtin_ia32_dbpsadbw128_mask((__v16qi)(__m128i)(A), \
- (__v16qi)(__m128i)(B), (int)(imm), \
- (__v8hi)_mm_setzero_hi(), \
- (__mmask8)-1); })
-
-#define _mm_mask_dbsad_epu8(W, U, A, B, imm) __extension__ ({ \
- (__m128i)__builtin_ia32_dbpsadbw128_mask((__v16qi)(__m128i)(A), \
- (__v16qi)(__m128i)(B), (int)(imm), \
- (__v8hi)(__m128i)(W), \
- (__mmask8)(U)); })
-
-#define _mm_maskz_dbsad_epu8(U, A, B, imm) __extension__ ({ \
- (__m128i)__builtin_ia32_dbpsadbw128_mask((__v16qi)(__m128i)(A), \
- (__v16qi)(__m128i)(B), (int)(imm), \
- (__v8hi)_mm_setzero_si128(), \
- (__mmask8)(U)); })
-
-#define _mm256_dbsad_epu8(A, B, imm) __extension__ ({ \
- (__m256i)__builtin_ia32_dbpsadbw256_mask((__v32qi)(__m256i)(A), \
- (__v32qi)(__m256i)(B), (int)(imm), \
- (__v16hi)_mm256_setzero_si256(), \
- (__mmask16)-1); })
-
-#define _mm256_mask_dbsad_epu8(W, U, A, B, imm) __extension__ ({ \
- (__m256i)__builtin_ia32_dbpsadbw256_mask((__v32qi)(__m256i)(A), \
- (__v32qi)(__m256i)(B), (int)(imm), \
- (__v16hi)(__m256i)(W), \
- (__mmask16)(U)); })
-
-#define _mm256_maskz_dbsad_epu8(U, A, B, imm) __extension__ ({ \
- (__m256i)__builtin_ia32_dbpsadbw256_mask((__v32qi)(__m256i)(A), \
- (__v32qi)(__m256i)(B), (int)(imm), \
- (__v16hi)_mm256_setzero_si256(), \
- (__mmask16)(U)); })
-
-#undef __DEFAULT_FN_ATTRS
+ (__v32qi)_mm256_setzero_si256())
+
+#define _mm_dbsad_epu8(A, B, imm) \
+ (__m128i)__builtin_ia32_dbpsadbw128((__v16qi)(__m128i)(A), \
+ (__v16qi)(__m128i)(B), (int)(imm))
+
+#define _mm_mask_dbsad_epu8(W, U, A, B, imm) \
+ (__m128i)__builtin_ia32_selectw_128((__mmask8)(U), \
+ (__v8hi)_mm_dbsad_epu8((A), (B), (imm)), \
+ (__v8hi)(__m128i)(W))
+
+#define _mm_maskz_dbsad_epu8(U, A, B, imm) \
+ (__m128i)__builtin_ia32_selectw_128((__mmask8)(U), \
+ (__v8hi)_mm_dbsad_epu8((A), (B), (imm)), \
+ (__v8hi)_mm_setzero_si128())
+
+#define _mm256_dbsad_epu8(A, B, imm) \
+ (__m256i)__builtin_ia32_dbpsadbw256((__v32qi)(__m256i)(A), \
+ (__v32qi)(__m256i)(B), (int)(imm))
+
+#define _mm256_mask_dbsad_epu8(W, U, A, B, imm) \
+ (__m256i)__builtin_ia32_selectw_256((__mmask16)(U), \
+ (__v16hi)_mm256_dbsad_epu8((A), (B), (imm)), \
+ (__v16hi)(__m256i)(W))
+
+#define _mm256_maskz_dbsad_epu8(U, A, B, imm) \
+ (__m256i)__builtin_ia32_selectw_256((__mmask16)(U), \
+ (__v16hi)_mm256_dbsad_epu8((A), (B), (imm)), \
+ (__v16hi)_mm256_setzero_si256())
+
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
#endif /* __AVX512VLBWINTRIN_H */
--- /dev/null
+/*===---- avx512vlcdintrin.h - AVX512VL and AVX512CD intrinsics ------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512vlcdintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __AVX512VLCDINTRIN_H
+#define __AVX512VLCDINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512cd"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512cd"), __min_vector_width__(256)))
+
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_broadcastmb_epi64 (__mmask8 __A)
+{
+ return (__m128i) _mm_set1_epi64x((long long) __A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_broadcastmb_epi64 (__mmask8 __A)
+{
+ return (__m256i) _mm256_set1_epi64x((long long)__A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_broadcastmw_epi32 (__mmask16 __A)
+{
+ return (__m128i) _mm_set1_epi32((int)__A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_broadcastmw_epi32 (__mmask16 __A)
+{
+ return (__m256i) _mm256_set1_epi32((int)__A);
+}
+
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_conflict_epi64 (__m128i __A)
+{
+ return (__m128i) __builtin_ia32_vpconflictdi_128 ((__v2di) __A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_conflict_epi64 (__m128i __W, __mmask8 __U, __m128i __A)
+{
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_conflict_epi64(__A),
+ (__v2di)__W);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_conflict_epi64 (__mmask8 __U, __m128i __A)
+{
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_conflict_epi64(__A),
+ (__v2di)_mm_setzero_si128());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_conflict_epi64 (__m256i __A)
+{
+ return (__m256i) __builtin_ia32_vpconflictdi_256 ((__v4di) __A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_conflict_epi64 (__m256i __W, __mmask8 __U, __m256i __A)
+{
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_conflict_epi64(__A),
+ (__v4di)__W);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_conflict_epi64 (__mmask8 __U, __m256i __A)
+{
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_conflict_epi64(__A),
+ (__v4di)_mm256_setzero_si256());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_conflict_epi32 (__m128i __A)
+{
+ return (__m128i) __builtin_ia32_vpconflictsi_128 ((__v4si) __A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_conflict_epi32 (__m128i __W, __mmask8 __U, __m128i __A)
+{
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_conflict_epi32(__A),
+ (__v4si)__W);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_conflict_epi32 (__mmask8 __U, __m128i __A)
+{
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_conflict_epi32(__A),
+ (__v4si)_mm_setzero_si128());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_conflict_epi32 (__m256i __A)
+{
+ return (__m256i) __builtin_ia32_vpconflictsi_256 ((__v8si) __A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_conflict_epi32 (__m256i __W, __mmask8 __U, __m256i __A)
+{
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_conflict_epi32(__A),
+ (__v8si)__W);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_conflict_epi32 (__mmask8 __U, __m256i __A)
+{
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_conflict_epi32(__A),
+ (__v8si)_mm256_setzero_si256());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_lzcnt_epi32 (__m128i __A)
+{
+ return (__m128i) __builtin_ia32_vplzcntd_128 ((__v4si) __A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_lzcnt_epi32 (__m128i __W, __mmask8 __U, __m128i __A)
+{
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_lzcnt_epi32(__A),
+ (__v4si)__W);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_lzcnt_epi32 (__mmask8 __U, __m128i __A)
+{
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_lzcnt_epi32(__A),
+ (__v4si)_mm_setzero_si128());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_lzcnt_epi32 (__m256i __A)
+{
+ return (__m256i) __builtin_ia32_vplzcntd_256 ((__v8si) __A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_lzcnt_epi32 (__m256i __W, __mmask8 __U, __m256i __A)
+{
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_lzcnt_epi32(__A),
+ (__v8si)__W);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_lzcnt_epi32 (__mmask8 __U, __m256i __A)
+{
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_lzcnt_epi32(__A),
+ (__v8si)_mm256_setzero_si256());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_lzcnt_epi64 (__m128i __A)
+{
+ return (__m128i) __builtin_ia32_vplzcntq_128 ((__v2di) __A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_lzcnt_epi64 (__m128i __W, __mmask8 __U, __m128i __A)
+{
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_lzcnt_epi64(__A),
+ (__v2di)__W);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_lzcnt_epi64 (__mmask8 __U, __m128i __A)
+{
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_lzcnt_epi64(__A),
+ (__v2di)_mm_setzero_si128());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_lzcnt_epi64 (__m256i __A)
+{
+ return (__m256i) __builtin_ia32_vplzcntq_256 ((__v4di) __A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_lzcnt_epi64 (__m256i __W, __mmask8 __U, __m256i __A)
+{
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_lzcnt_epi64(__A),
+ (__v4di)__W);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_lzcnt_epi64 (__mmask8 __U, __m256i __A)
+{
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_lzcnt_epi64(__A),
+ (__v4di)_mm256_setzero_si256());
+}
+
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
+
+#endif /* __AVX512VLCDINTRIN_H */
/*===---- avx512vldqintrin.h - AVX512VL and AVX512DQ intrinsics ------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#define __AVX512VLDQINTRIN_H
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512dq")))
+#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512dq"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512dq"), __min_vector_width__(256)))
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mullo_epi64 (__m256i __A, __m256i __B) {
return (__m256i) ((__v4du) __A * (__v4du) __B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_mullo_epi64(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)_mm256_mullo_epi64(__A, __B),
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_mullo_epi64(__mmask8 __U, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)_mm256_mullo_epi64(__A, __B),
(__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mullo_epi64 (__m128i __A, __m128i __B) {
return (__m128i) ((__v2du) __A * (__v2du) __B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_mullo_epi64(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)_mm_mullo_epi64(__A, __B),
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_mullo_epi64(__mmask8 __U, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)_mm_mullo_epi64(__A, __B),
(__v2di)_mm_setzero_si128());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_andnot_pd(__m256d __W, __mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_andnot_pd(__A, __B),
(__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_andnot_pd(__mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_andnot_pd(__A, __B),
(__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_andnot_pd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_andnot_pd(__A, __B),
(__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_andnot_pd(__mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_andnot_pd(__A, __B),
(__v2df)_mm_setzero_pd());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_andnot_ps(__m256 __W, __mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_andnot_ps(__A, __B),
(__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_andnot_ps(__mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_andnot_ps(__A, __B),
(__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_andnot_ps(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_andnot_ps(__A, __B),
(__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_andnot_ps(__mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_andnot_ps(__A, __B),
(__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_and_pd(__m256d __W, __mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_and_pd(__A, __B),
(__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_and_pd(__mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_and_pd(__A, __B),
(__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_and_pd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_and_pd(__A, __B),
(__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_and_pd(__mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_and_pd(__A, __B),
(__v2df)_mm_setzero_pd());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_and_ps(__m256 __W, __mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_and_ps(__A, __B),
(__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_and_ps(__mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_and_ps(__A, __B),
(__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_and_ps(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_and_ps(__A, __B),
(__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_and_ps(__mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_and_ps(__A, __B),
(__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_xor_pd(__m256d __W, __mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_xor_pd(__A, __B),
(__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_xor_pd(__mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_xor_pd(__A, __B),
(__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_xor_pd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_xor_pd(__A, __B),
(__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_xor_pd (__mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_xor_pd(__A, __B),
(__v2df)_mm_setzero_pd());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_xor_ps(__m256 __W, __mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_xor_ps(__A, __B),
(__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_xor_ps(__mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_xor_ps(__A, __B),
(__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_xor_ps(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_xor_ps(__A, __B),
(__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_xor_ps(__mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_xor_ps(__A, __B),
(__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_or_pd(__m256d __W, __mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_or_pd(__A, __B),
(__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_or_pd(__mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_or_pd(__A, __B),
(__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_or_pd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_or_pd(__A, __B),
(__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_or_pd(__mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_or_pd(__A, __B),
(__v2df)_mm_setzero_pd());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_or_ps(__m256 __W, __mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_or_ps(__A, __B),
(__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_or_ps(__mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_or_ps(__A, __B),
(__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_or_ps(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_or_ps(__A, __B),
(__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_or_ps(__mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_or_ps(__A, __B),
(__v4sf)_mm_setzero_ps());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtpd_epi64 (__m128d __A) {
return (__m128i) __builtin_ia32_cvtpd2qq128_mask ((__v2df) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtpd_epi64 (__m128i __W, __mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvtpd2qq128_mask ((__v2df) __A,
(__v2di) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtpd_epi64 (__mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvtpd2qq128_mask ((__v2df) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtpd_epi64 (__m256d __A) {
return (__m256i) __builtin_ia32_cvtpd2qq256_mask ((__v4df) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtpd_epi64 (__m256i __W, __mmask8 __U, __m256d __A) {
return (__m256i) __builtin_ia32_cvtpd2qq256_mask ((__v4df) __A,
(__v4di) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtpd_epi64 (__mmask8 __U, __m256d __A) {
return (__m256i) __builtin_ia32_cvtpd2qq256_mask ((__v4df) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtpd_epu64 (__m128d __A) {
return (__m128i) __builtin_ia32_cvtpd2uqq128_mask ((__v2df) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtpd_epu64 (__m128i __W, __mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvtpd2uqq128_mask ((__v2df) __A,
(__v2di) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtpd_epu64 (__mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvtpd2uqq128_mask ((__v2df) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtpd_epu64 (__m256d __A) {
return (__m256i) __builtin_ia32_cvtpd2uqq256_mask ((__v4df) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtpd_epu64 (__m256i __W, __mmask8 __U, __m256d __A) {
return (__m256i) __builtin_ia32_cvtpd2uqq256_mask ((__v4df) __A,
(__v4di) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtpd_epu64 (__mmask8 __U, __m256d __A) {
return (__m256i) __builtin_ia32_cvtpd2uqq256_mask ((__v4df) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtps_epi64 (__m128 __A) {
return (__m128i) __builtin_ia32_cvtps2qq128_mask ((__v4sf) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtps_epi64 (__m128i __W, __mmask8 __U, __m128 __A) {
return (__m128i) __builtin_ia32_cvtps2qq128_mask ((__v4sf) __A,
(__v2di) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtps_epi64 (__mmask8 __U, __m128 __A) {
return (__m128i) __builtin_ia32_cvtps2qq128_mask ((__v4sf) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtps_epi64 (__m128 __A) {
return (__m256i) __builtin_ia32_cvtps2qq256_mask ((__v4sf) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtps_epi64 (__m256i __W, __mmask8 __U, __m128 __A) {
return (__m256i) __builtin_ia32_cvtps2qq256_mask ((__v4sf) __A,
(__v4di) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtps_epi64 (__mmask8 __U, __m128 __A) {
return (__m256i) __builtin_ia32_cvtps2qq256_mask ((__v4sf) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtps_epu64 (__m128 __A) {
return (__m128i) __builtin_ia32_cvtps2uqq128_mask ((__v4sf) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtps_epu64 (__m128i __W, __mmask8 __U, __m128 __A) {
return (__m128i) __builtin_ia32_cvtps2uqq128_mask ((__v4sf) __A,
(__v2di) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtps_epu64 (__mmask8 __U, __m128 __A) {
return (__m128i) __builtin_ia32_cvtps2uqq128_mask ((__v4sf) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtps_epu64 (__m128 __A) {
return (__m256i) __builtin_ia32_cvtps2uqq256_mask ((__v4sf) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtps_epu64 (__m256i __W, __mmask8 __U, __m128 __A) {
return (__m256i) __builtin_ia32_cvtps2uqq256_mask ((__v4sf) __A,
(__v4di) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtps_epu64 (__mmask8 __U, __m128 __A) {
return (__m256i) __builtin_ia32_cvtps2uqq256_mask ((__v4sf) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_cvtepi64_pd (__m128i __A) {
- return (__m128d) __builtin_ia32_cvtqq2pd128_mask ((__v2di) __A,
- (__v2df) _mm_setzero_pd(),
- (__mmask8) -1);
+ return (__m128d)__builtin_convertvector((__v2di)__A, __v2df);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi64_pd (__m128d __W, __mmask8 __U, __m128i __A) {
- return (__m128d) __builtin_ia32_cvtqq2pd128_mask ((__v2di) __A,
- (__v2df) __W,
- (__mmask8) __U);
+ return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
+ (__v2df)_mm_cvtepi64_pd(__A),
+ (__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_cvtepi64_pd (__mmask8 __U, __m128i __A) {
- return (__m128d) __builtin_ia32_cvtqq2pd128_mask ((__v2di) __A,
- (__v2df) _mm_setzero_pd(),
- (__mmask8) __U);
+ return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
+ (__v2df)_mm_cvtepi64_pd(__A),
+ (__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_cvtepi64_pd (__m256i __A) {
- return (__m256d) __builtin_ia32_cvtqq2pd256_mask ((__v4di) __A,
- (__v4df) _mm256_setzero_pd(),
- (__mmask8) -1);
+ return (__m256d)__builtin_convertvector((__v4di)__A, __v4df);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi64_pd (__m256d __W, __mmask8 __U, __m256i __A) {
- return (__m256d) __builtin_ia32_cvtqq2pd256_mask ((__v4di) __A,
- (__v4df) __W,
- (__mmask8) __U);
+ return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
+ (__v4df)_mm256_cvtepi64_pd(__A),
+ (__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtepi64_pd (__mmask8 __U, __m256i __A) {
- return (__m256d) __builtin_ia32_cvtqq2pd256_mask ((__v4di) __A,
- (__v4df) _mm256_setzero_pd(),
- (__mmask8) __U);
+ return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
+ (__v4df)_mm256_cvtepi64_pd(__A),
+ (__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_cvtepi64_ps (__m128i __A) {
return (__m128) __builtin_ia32_cvtqq2ps128_mask ((__v2di) __A,
(__v4sf) _mm_setzero_ps(),
(__mmask8) -1);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi64_ps (__m128 __W, __mmask8 __U, __m128i __A) {
return (__m128) __builtin_ia32_cvtqq2ps128_mask ((__v2di) __A,
(__v4sf) __W,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_cvtepi64_ps (__mmask8 __U, __m128i __A) {
return (__m128) __builtin_ia32_cvtqq2ps128_mask ((__v2di) __A,
(__v4sf) _mm_setzero_ps(),
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS256
_mm256_cvtepi64_ps (__m256i __A) {
- return (__m128) __builtin_ia32_cvtqq2ps256_mask ((__v4di) __A,
- (__v4sf) _mm_setzero_ps(),
- (__mmask8) -1);
+ return (__m128)__builtin_convertvector((__v4di)__A, __v4sf);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi64_ps (__m128 __W, __mmask8 __U, __m256i __A) {
- return (__m128) __builtin_ia32_cvtqq2ps256_mask ((__v4di) __A,
- (__v4sf) __W,
- (__mmask8) __U);
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm256_cvtepi64_ps(__A),
+ (__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtepi64_ps (__mmask8 __U, __m256i __A) {
- return (__m128) __builtin_ia32_cvtqq2ps256_mask ((__v4di) __A,
- (__v4sf) _mm_setzero_ps(),
- (__mmask8) __U);
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm256_cvtepi64_ps(__A),
+ (__v4sf)_mm_setzero_ps());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvttpd_epi64 (__m128d __A) {
return (__m128i) __builtin_ia32_cvttpd2qq128_mask ((__v2df) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvttpd_epi64 (__m128i __W, __mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvttpd2qq128_mask ((__v2df) __A,
(__v2di) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvttpd_epi64 (__mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvttpd2qq128_mask ((__v2df) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvttpd_epi64 (__m256d __A) {
return (__m256i) __builtin_ia32_cvttpd2qq256_mask ((__v4df) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_cvttpd_epi64 (__m256i __W, __mmask8 __U, __m256d __A) {
return (__m256i) __builtin_ia32_cvttpd2qq256_mask ((__v4df) __A,
(__v4di) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvttpd_epi64 (__mmask8 __U, __m256d __A) {
return (__m256i) __builtin_ia32_cvttpd2qq256_mask ((__v4df) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvttpd_epu64 (__m128d __A) {
return (__m128i) __builtin_ia32_cvttpd2uqq128_mask ((__v2df) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvttpd_epu64 (__m128i __W, __mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvttpd2uqq128_mask ((__v2df) __A,
(__v2di) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvttpd_epu64 (__mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvttpd2uqq128_mask ((__v2df) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvttpd_epu64 (__m256d __A) {
return (__m256i) __builtin_ia32_cvttpd2uqq256_mask ((__v4df) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_cvttpd_epu64 (__m256i __W, __mmask8 __U, __m256d __A) {
return (__m256i) __builtin_ia32_cvttpd2uqq256_mask ((__v4df) __A,
(__v4di) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvttpd_epu64 (__mmask8 __U, __m256d __A) {
return (__m256i) __builtin_ia32_cvttpd2uqq256_mask ((__v4df) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvttps_epi64 (__m128 __A) {
return (__m128i) __builtin_ia32_cvttps2qq128_mask ((__v4sf) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvttps_epi64 (__m128i __W, __mmask8 __U, __m128 __A) {
return (__m128i) __builtin_ia32_cvttps2qq128_mask ((__v4sf) __A,
(__v2di) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvttps_epi64 (__mmask8 __U, __m128 __A) {
return (__m128i) __builtin_ia32_cvttps2qq128_mask ((__v4sf) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvttps_epi64 (__m128 __A) {
return (__m256i) __builtin_ia32_cvttps2qq256_mask ((__v4sf) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_cvttps_epi64 (__m256i __W, __mmask8 __U, __m128 __A) {
return (__m256i) __builtin_ia32_cvttps2qq256_mask ((__v4sf) __A,
(__v4di) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvttps_epi64 (__mmask8 __U, __m128 __A) {
return (__m256i) __builtin_ia32_cvttps2qq256_mask ((__v4sf) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvttps_epu64 (__m128 __A) {
return (__m128i) __builtin_ia32_cvttps2uqq128_mask ((__v4sf) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvttps_epu64 (__m128i __W, __mmask8 __U, __m128 __A) {
return (__m128i) __builtin_ia32_cvttps2uqq128_mask ((__v4sf) __A,
(__v2di) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvttps_epu64 (__mmask8 __U, __m128 __A) {
return (__m128i) __builtin_ia32_cvttps2uqq128_mask ((__v4sf) __A,
(__v2di) _mm_setzero_si128(),
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvttps_epu64 (__m128 __A) {
return (__m256i) __builtin_ia32_cvttps2uqq256_mask ((__v4sf) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_cvttps_epu64 (__m256i __W, __mmask8 __U, __m128 __A) {
return (__m256i) __builtin_ia32_cvttps2uqq256_mask ((__v4sf) __A,
(__v4di) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvttps_epu64 (__mmask8 __U, __m128 __A) {
return (__m256i) __builtin_ia32_cvttps2uqq256_mask ((__v4sf) __A,
(__v4di) _mm256_setzero_si256(),
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_cvtepu64_pd (__m128i __A) {
- return (__m128d) __builtin_ia32_cvtuqq2pd128_mask ((__v2di) __A,
- (__v2df) _mm_setzero_pd(),
- (__mmask8) -1);
+ return (__m128d)__builtin_convertvector((__v2du)__A, __v2df);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_cvtepu64_pd (__m128d __W, __mmask8 __U, __m128i __A) {
- return (__m128d) __builtin_ia32_cvtuqq2pd128_mask ((__v2di) __A,
- (__v2df) __W,
- (__mmask8) __U);
+ return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
+ (__v2df)_mm_cvtepu64_pd(__A),
+ (__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_cvtepu64_pd (__mmask8 __U, __m128i __A) {
- return (__m128d) __builtin_ia32_cvtuqq2pd128_mask ((__v2di) __A,
- (__v2df) _mm_setzero_pd(),
- (__mmask8) __U);
+ return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
+ (__v2df)_mm_cvtepu64_pd(__A),
+ (__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_cvtepu64_pd (__m256i __A) {
- return (__m256d) __builtin_ia32_cvtuqq2pd256_mask ((__v4di) __A,
- (__v4df) _mm256_setzero_pd(),
- (__mmask8) -1);
+ return (__m256d)__builtin_convertvector((__v4du)__A, __v4df);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepu64_pd (__m256d __W, __mmask8 __U, __m256i __A) {
- return (__m256d) __builtin_ia32_cvtuqq2pd256_mask ((__v4di) __A,
- (__v4df) __W,
- (__mmask8) __U);
+ return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
+ (__v4df)_mm256_cvtepu64_pd(__A),
+ (__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtepu64_pd (__mmask8 __U, __m256i __A) {
- return (__m256d) __builtin_ia32_cvtuqq2pd256_mask ((__v4di) __A,
- (__v4df) _mm256_setzero_pd(),
- (__mmask8) __U);
+ return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
+ (__v4df)_mm256_cvtepu64_pd(__A),
+ (__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_cvtepu64_ps (__m128i __A) {
return (__m128) __builtin_ia32_cvtuqq2ps128_mask ((__v2di) __A,
(__v4sf) _mm_setzero_ps(),
(__mmask8) -1);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_cvtepu64_ps (__m128 __W, __mmask8 __U, __m128i __A) {
return (__m128) __builtin_ia32_cvtuqq2ps128_mask ((__v2di) __A,
(__v4sf) __W,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_cvtepu64_ps (__mmask8 __U, __m128i __A) {
return (__m128) __builtin_ia32_cvtuqq2ps128_mask ((__v2di) __A,
(__v4sf) _mm_setzero_ps(),
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS256
_mm256_cvtepu64_ps (__m256i __A) {
- return (__m128) __builtin_ia32_cvtuqq2ps256_mask ((__v4di) __A,
- (__v4sf) _mm_setzero_ps(),
- (__mmask8) -1);
+ return (__m128)__builtin_convertvector((__v4du)__A, __v4sf);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepu64_ps (__m128 __W, __mmask8 __U, __m256i __A) {
- return (__m128) __builtin_ia32_cvtuqq2ps256_mask ((__v4di) __A,
- (__v4sf) __W,
- (__mmask8) __U);
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm256_cvtepu64_ps(__A),
+ (__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtepu64_ps (__mmask8 __U, __m256i __A) {
- return (__m128) __builtin_ia32_cvtuqq2ps256_mask ((__v4di) __A,
- (__v4sf) _mm_setzero_ps(),
- (__mmask8) __U);
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm256_cvtepu64_ps(__A),
+ (__v4sf)_mm_setzero_ps());
}
-#define _mm_range_pd(A, B, C) __extension__ ({ \
+#define _mm_range_pd(A, B, C) \
(__m128d)__builtin_ia32_rangepd128_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), (int)(C), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_range_pd(W, U, A, B, C) __extension__ ({ \
+#define _mm_mask_range_pd(W, U, A, B, C) \
(__m128d)__builtin_ia32_rangepd128_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), (int)(C), \
(__v2df)(__m128d)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_maskz_range_pd(U, A, B, C) __extension__ ({ \
+#define _mm_maskz_range_pd(U, A, B, C) \
(__m128d)__builtin_ia32_rangepd128_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), (int)(C), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_range_pd(A, B, C) __extension__ ({ \
+#define _mm256_range_pd(A, B, C) \
(__m256d)__builtin_ia32_rangepd256_mask((__v4df)(__m256d)(A), \
(__v4df)(__m256d)(B), (int)(C), \
(__v4df)_mm256_setzero_pd(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_range_pd(W, U, A, B, C) __extension__ ({ \
+#define _mm256_mask_range_pd(W, U, A, B, C) \
(__m256d)__builtin_ia32_rangepd256_mask((__v4df)(__m256d)(A), \
(__v4df)(__m256d)(B), (int)(C), \
(__v4df)(__m256d)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_maskz_range_pd(U, A, B, C) __extension__ ({ \
+#define _mm256_maskz_range_pd(U, A, B, C) \
(__m256d)__builtin_ia32_rangepd256_mask((__v4df)(__m256d)(A), \
(__v4df)(__m256d)(B), (int)(C), \
(__v4df)_mm256_setzero_pd(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_range_ps(A, B, C) __extension__ ({ \
+#define _mm_range_ps(A, B, C) \
(__m128)__builtin_ia32_rangeps128_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), (int)(C), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_range_ps(W, U, A, B, C) __extension__ ({ \
+#define _mm_mask_range_ps(W, U, A, B, C) \
(__m128)__builtin_ia32_rangeps128_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), (int)(C), \
- (__v4sf)(__m128)(W), (__mmask8)(U)); })
+ (__v4sf)(__m128)(W), (__mmask8)(U))
-#define _mm_maskz_range_ps(U, A, B, C) __extension__ ({ \
+#define _mm_maskz_range_ps(U, A, B, C) \
(__m128)__builtin_ia32_rangeps128_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), (int)(C), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_range_ps(A, B, C) __extension__ ({ \
+#define _mm256_range_ps(A, B, C) \
(__m256)__builtin_ia32_rangeps256_mask((__v8sf)(__m256)(A), \
(__v8sf)(__m256)(B), (int)(C), \
(__v8sf)_mm256_setzero_ps(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_range_ps(W, U, A, B, C) __extension__ ({ \
+#define _mm256_mask_range_ps(W, U, A, B, C) \
(__m256)__builtin_ia32_rangeps256_mask((__v8sf)(__m256)(A), \
(__v8sf)(__m256)(B), (int)(C), \
- (__v8sf)(__m256)(W), (__mmask8)(U)); })
+ (__v8sf)(__m256)(W), (__mmask8)(U))
-#define _mm256_maskz_range_ps(U, A, B, C) __extension__ ({ \
+#define _mm256_maskz_range_ps(U, A, B, C) \
(__m256)__builtin_ia32_rangeps256_mask((__v8sf)(__m256)(A), \
(__v8sf)(__m256)(B), (int)(C), \
(__v8sf)_mm256_setzero_ps(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_reduce_pd(A, B) __extension__ ({ \
+#define _mm_reduce_pd(A, B) \
(__m128d)__builtin_ia32_reducepd128_mask((__v2df)(__m128d)(A), (int)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_reduce_pd(W, U, A, B) __extension__ ({ \
+#define _mm_mask_reduce_pd(W, U, A, B) \
(__m128d)__builtin_ia32_reducepd128_mask((__v2df)(__m128d)(A), (int)(B), \
(__v2df)(__m128d)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_maskz_reduce_pd(U, A, B) __extension__ ({ \
+#define _mm_maskz_reduce_pd(U, A, B) \
(__m128d)__builtin_ia32_reducepd128_mask((__v2df)(__m128d)(A), (int)(B), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_reduce_pd(A, B) __extension__ ({ \
+#define _mm256_reduce_pd(A, B) \
(__m256d)__builtin_ia32_reducepd256_mask((__v4df)(__m256d)(A), (int)(B), \
(__v4df)_mm256_setzero_pd(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_reduce_pd(W, U, A, B) __extension__ ({ \
+#define _mm256_mask_reduce_pd(W, U, A, B) \
(__m256d)__builtin_ia32_reducepd256_mask((__v4df)(__m256d)(A), (int)(B), \
(__v4df)(__m256d)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_maskz_reduce_pd(U, A, B) __extension__ ({ \
+#define _mm256_maskz_reduce_pd(U, A, B) \
(__m256d)__builtin_ia32_reducepd256_mask((__v4df)(__m256d)(A), (int)(B), \
(__v4df)_mm256_setzero_pd(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_reduce_ps(A, B) __extension__ ({ \
+#define _mm_reduce_ps(A, B) \
(__m128)__builtin_ia32_reduceps128_mask((__v4sf)(__m128)(A), (int)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_reduce_ps(W, U, A, B) __extension__ ({ \
+#define _mm_mask_reduce_ps(W, U, A, B) \
(__m128)__builtin_ia32_reduceps128_mask((__v4sf)(__m128)(A), (int)(B), \
(__v4sf)(__m128)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_maskz_reduce_ps(U, A, B) __extension__ ({ \
+#define _mm_maskz_reduce_ps(U, A, B) \
(__m128)__builtin_ia32_reduceps128_mask((__v4sf)(__m128)(A), (int)(B), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_reduce_ps(A, B) __extension__ ({ \
+#define _mm256_reduce_ps(A, B) \
(__m256)__builtin_ia32_reduceps256_mask((__v8sf)(__m256)(A), (int)(B), \
(__v8sf)_mm256_setzero_ps(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_reduce_ps(W, U, A, B) __extension__ ({ \
+#define _mm256_mask_reduce_ps(W, U, A, B) \
(__m256)__builtin_ia32_reduceps256_mask((__v8sf)(__m256)(A), (int)(B), \
(__v8sf)(__m256)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_maskz_reduce_ps(U, A, B) __extension__ ({ \
+#define _mm256_maskz_reduce_ps(U, A, B) \
(__m256)__builtin_ia32_reduceps256_mask((__v8sf)(__m256)(A), (int)(B), \
(__v8sf)_mm256_setzero_ps(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS128
_mm_movepi32_mask (__m128i __A)
{
return (__mmask8) __builtin_ia32_cvtd2mask128 ((__v4si) __A);
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS256
_mm256_movepi32_mask (__m256i __A)
{
return (__mmask8) __builtin_ia32_cvtd2mask256 ((__v8si) __A);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_movm_epi32 (__mmask8 __A)
{
return (__m128i) __builtin_ia32_cvtmask2d128 (__A);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_movm_epi32 (__mmask8 __A)
{
return (__m256i) __builtin_ia32_cvtmask2d256 (__A);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_movm_epi64 (__mmask8 __A)
{
return (__m128i) __builtin_ia32_cvtmask2q128 (__A);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_movm_epi64 (__mmask8 __A)
{
return (__m256i) __builtin_ia32_cvtmask2q256 (__A);
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS128
_mm_movepi64_mask (__m128i __A)
{
return (__mmask8) __builtin_ia32_cvtq2mask128 ((__v2di) __A);
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS256
_mm256_movepi64_mask (__m256i __A)
{
return (__mmask8) __builtin_ia32_cvtq2mask256 ((__v4di) __A);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_broadcast_f32x2 (__m128 __A)
{
- return (__m256) __builtin_ia32_broadcastf32x2_256_mask ((__v4sf) __A,
- (__v8sf)_mm256_undefined_ps(),
- (__mmask8) -1);
+ return (__m256)__builtin_shufflevector((__v4sf)__A, (__v4sf)__A,
+ 0, 1, 0, 1, 0, 1, 0, 1);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_broadcast_f32x2 (__m256 __O, __mmask8 __M, __m128 __A)
{
- return (__m256) __builtin_ia32_broadcastf32x2_256_mask ((__v4sf) __A,
- (__v8sf) __O,
- __M);
+ return (__m256)__builtin_ia32_selectps_256((__mmask8)__M,
+ (__v8sf)_mm256_broadcast_f32x2(__A),
+ (__v8sf)__O);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_broadcast_f32x2 (__mmask8 __M, __m128 __A)
{
- return (__m256) __builtin_ia32_broadcastf32x2_256_mask ((__v4sf) __A,
- (__v8sf) _mm256_setzero_ps (),
- __M);
+ return (__m256)__builtin_ia32_selectps_256((__mmask8)__M,
+ (__v8sf)_mm256_broadcast_f32x2(__A),
+ (__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_broadcast_f64x2(__m128d __A)
{
return (__m256d)__builtin_shufflevector((__v2df)__A, (__v2df)__A,
0, 1, 0, 1);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_broadcast_f64x2(__m256d __O, __mmask8 __M, __m128d __A)
{
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__M,
(__v4df)__O);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_broadcast_f64x2 (__mmask8 __M, __m128d __A)
{
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__M,
(__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_broadcast_i32x2 (__m128i __A)
{
- return (__m128i) __builtin_ia32_broadcasti32x2_128_mask ((__v4si) __A,
- (__v4si)_mm_undefined_si128(),
- (__mmask8) -1);
+ return (__m128i)__builtin_shufflevector((__v4si)__A, (__v4si)__A,
+ 0, 1, 0, 1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_broadcast_i32x2 (__m128i __O, __mmask8 __M, __m128i __A)
{
- return (__m128i) __builtin_ia32_broadcasti32x2_128_mask ((__v4si) __A,
- (__v4si) __O,
- __M);
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__M,
+ (__v4si)_mm_broadcast_i32x2(__A),
+ (__v4si)__O);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_broadcast_i32x2 (__mmask8 __M, __m128i __A)
{
- return (__m128i) __builtin_ia32_broadcasti32x2_128_mask ((__v4si) __A,
- (__v4si) _mm_setzero_si128 (),
- __M);
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__M,
+ (__v4si)_mm_broadcast_i32x2(__A),
+ (__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_broadcast_i32x2 (__m128i __A)
{
- return (__m256i) __builtin_ia32_broadcasti32x2_256_mask ((__v4si) __A,
- (__v8si)_mm256_undefined_si256(),
- (__mmask8) -1);
+ return (__m256i)__builtin_shufflevector((__v4si)__A, (__v4si)__A,
+ 0, 1, 0, 1, 0, 1, 0, 1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_broadcast_i32x2 (__m256i __O, __mmask8 __M, __m128i __A)
{
- return (__m256i) __builtin_ia32_broadcasti32x2_256_mask ((__v4si) __A,
- (__v8si) __O,
- __M);
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
+ (__v8si)_mm256_broadcast_i32x2(__A),
+ (__v8si)__O);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_broadcast_i32x2 (__mmask8 __M, __m128i __A)
{
- return (__m256i) __builtin_ia32_broadcasti32x2_256_mask ((__v4si) __A,
- (__v8si) _mm256_setzero_si256 (),
- __M);
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
+ (__v8si)_mm256_broadcast_i32x2(__A),
+ (__v8si)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_broadcast_i64x2(__m128i __A)
{
return (__m256i)__builtin_shufflevector((__v2di)__A, (__v2di)__A,
0, 1, 0, 1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_broadcast_i64x2(__m256i __O, __mmask8 __M, __m128i __A)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
(__v4di)__O);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_broadcast_i64x2 (__mmask8 __M, __m128i __A)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
(__v4di)_mm256_setzero_si256());
}
-#define _mm256_extractf64x2_pd(A, imm) __extension__ ({ \
- (__m128d)__builtin_shufflevector((__v4df)(__m256d)(A), \
- (__v4df)_mm256_undefined_pd(), \
- ((imm) & 1) ? 2 : 0, \
- ((imm) & 1) ? 3 : 1); })
-
-#define _mm256_mask_extractf64x2_pd(W, U, A, imm) __extension__ ({ \
- (__m128d)__builtin_ia32_selectpd_128((__mmask8)(U), \
- (__v2df)_mm256_extractf64x2_pd((A), (imm)), \
- (__v2df)(W)); })
-
-#define _mm256_maskz_extractf64x2_pd(U, A, imm) __extension__ ({ \
- (__m128d)__builtin_ia32_selectpd_128((__mmask8)(U), \
- (__v2df)_mm256_extractf64x2_pd((A), (imm)), \
- (__v2df)_mm_setzero_pd()); })
-
-#define _mm256_extracti64x2_epi64(A, imm) __extension__ ({ \
- (__m128i)__builtin_shufflevector((__v4di)(__m256i)(A), \
- (__v4di)_mm256_undefined_si256(), \
- ((imm) & 1) ? 2 : 0, \
- ((imm) & 1) ? 3 : 1); })
-
-#define _mm256_mask_extracti64x2_epi64(W, U, A, imm) __extension__ ({ \
- (__m128i)__builtin_ia32_selectq_128((__mmask8)(U), \
- (__v2di)_mm256_extracti64x2_epi64((A), (imm)), \
- (__v2di)(W)); })
-
-#define _mm256_maskz_extracti64x2_epi64(U, A, imm) __extension__ ({ \
- (__m128i)__builtin_ia32_selectq_128((__mmask8)(U), \
- (__v2di)_mm256_extracti64x2_epi64((A), (imm)), \
- (__v2di)_mm_setzero_di()); })
-
-#define _mm256_insertf64x2(A, B, imm) __extension__ ({ \
- (__m256d)__builtin_shufflevector((__v4df)(A), \
- (__v4df)_mm256_castpd128_pd256((__m128d)(B)), \
- ((imm) & 0x1) ? 0 : 4, \
- ((imm) & 0x1) ? 1 : 5, \
- ((imm) & 0x1) ? 4 : 2, \
- ((imm) & 0x1) ? 5 : 3); })
-
-#define _mm256_mask_insertf64x2(W, U, A, B, imm) __extension__ ({ \
+#define _mm256_extractf64x2_pd(A, imm) \
+ (__m128d)__builtin_ia32_extractf64x2_256_mask((__v4df)(__m256d)(A), \
+ (int)(imm), \
+ (__v2df)_mm_undefined_pd(), \
+ (__mmask8)-1)
+
+#define _mm256_mask_extractf64x2_pd(W, U, A, imm) \
+ (__m128d)__builtin_ia32_extractf64x2_256_mask((__v4df)(__m256d)(A), \
+ (int)(imm), \
+ (__v2df)(__m128d)(W), \
+ (__mmask8)(U))
+
+#define _mm256_maskz_extractf64x2_pd(U, A, imm) \
+ (__m128d)__builtin_ia32_extractf64x2_256_mask((__v4df)(__m256d)(A), \
+ (int)(imm), \
+ (__v2df)_mm_setzero_pd(), \
+ (__mmask8)(U))
+
+#define _mm256_extracti64x2_epi64(A, imm) \
+ (__m128i)__builtin_ia32_extracti64x2_256_mask((__v4di)(__m256i)(A), \
+ (int)(imm), \
+ (__v2di)_mm_undefined_si128(), \
+ (__mmask8)-1)
+
+#define _mm256_mask_extracti64x2_epi64(W, U, A, imm) \
+ (__m128i)__builtin_ia32_extracti64x2_256_mask((__v4di)(__m256i)(A), \
+ (int)(imm), \
+ (__v2di)(__m128i)(W), \
+ (__mmask8)(U))
+
+#define _mm256_maskz_extracti64x2_epi64(U, A, imm) \
+ (__m128i)__builtin_ia32_extracti64x2_256_mask((__v4di)(__m256i)(A), \
+ (int)(imm), \
+ (__v2di)_mm_setzero_si128(), \
+ (__mmask8)(U))
+
+#define _mm256_insertf64x2(A, B, imm) \
+ (__m256d)__builtin_ia32_insertf64x2_256((__v4df)(__m256d)(A), \
+ (__v2df)(__m128d)(B), (int)(imm))
+
+#define _mm256_mask_insertf64x2(W, U, A, B, imm) \
(__m256d)__builtin_ia32_selectpd_256((__mmask8)(U), \
(__v4df)_mm256_insertf64x2((A), (B), (imm)), \
- (__v4df)(W)); })
+ (__v4df)(__m256d)(W))
-#define _mm256_maskz_insertf64x2(U, A, B, imm) __extension__ ({ \
+#define _mm256_maskz_insertf64x2(U, A, B, imm) \
(__m256d)__builtin_ia32_selectpd_256((__mmask8)(U), \
(__v4df)_mm256_insertf64x2((A), (B), (imm)), \
- (__v4df)_mm256_setzero_pd()); })
+ (__v4df)_mm256_setzero_pd())
-#define _mm256_inserti64x2(A, B, imm) __extension__ ({ \
- (__m256i)__builtin_shufflevector((__v4di)(A), \
- (__v4di)_mm256_castsi128_si256((__m128i)(B)), \
- ((imm) & 0x1) ? 0 : 4, \
- ((imm) & 0x1) ? 1 : 5, \
- ((imm) & 0x1) ? 4 : 2, \
- ((imm) & 0x1) ? 5 : 3); })
+#define _mm256_inserti64x2(A, B, imm) \
+ (__m256i)__builtin_ia32_inserti64x2_256((__v4di)(__m256i)(A), \
+ (__v2di)(__m128i)(B), (int)(imm))
-#define _mm256_mask_inserti64x2(W, U, A, B, imm) __extension__ ({ \
+#define _mm256_mask_inserti64x2(W, U, A, B, imm) \
(__m256i)__builtin_ia32_selectq_256((__mmask8)(U), \
(__v4di)_mm256_inserti64x2((A), (B), (imm)), \
- (__v4di)(W)); })
+ (__v4di)(__m256i)(W))
-#define _mm256_maskz_inserti64x2(U, A, B, imm) __extension__ ({ \
+#define _mm256_maskz_inserti64x2(U, A, B, imm) \
(__m256i)__builtin_ia32_selectq_256((__mmask8)(U), \
(__v4di)_mm256_inserti64x2((A), (B), (imm)), \
- (__v4di)_mm256_setzero_si256()); })
+ (__v4di)_mm256_setzero_si256())
-#define _mm_mask_fpclass_pd_mask(U, A, imm) __extension__ ({ \
+#define _mm_mask_fpclass_pd_mask(U, A, imm) \
(__mmask8)__builtin_ia32_fpclasspd128_mask((__v2df)(__m128d)(A), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_fpclass_pd_mask(A, imm) __extension__ ({ \
+#define _mm_fpclass_pd_mask(A, imm) \
(__mmask8)__builtin_ia32_fpclasspd128_mask((__v2df)(__m128d)(A), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_fpclass_pd_mask(U, A, imm) __extension__ ({ \
+#define _mm256_mask_fpclass_pd_mask(U, A, imm) \
(__mmask8)__builtin_ia32_fpclasspd256_mask((__v4df)(__m256d)(A), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_fpclass_pd_mask(A, imm) __extension__ ({ \
+#define _mm256_fpclass_pd_mask(A, imm) \
(__mmask8)__builtin_ia32_fpclasspd256_mask((__v4df)(__m256d)(A), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_fpclass_ps_mask(U, A, imm) __extension__ ({ \
+#define _mm_mask_fpclass_ps_mask(U, A, imm) \
(__mmask8)__builtin_ia32_fpclassps128_mask((__v4sf)(__m128)(A), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_fpclass_ps_mask(A, imm) __extension__ ({ \
+#define _mm_fpclass_ps_mask(A, imm) \
(__mmask8)__builtin_ia32_fpclassps128_mask((__v4sf)(__m128)(A), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_fpclass_ps_mask(U, A, imm) __extension__ ({ \
+#define _mm256_mask_fpclass_ps_mask(U, A, imm) \
(__mmask8)__builtin_ia32_fpclassps256_mask((__v8sf)(__m256)(A), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_fpclass_ps_mask(A, imm) __extension__ ({ \
+#define _mm256_fpclass_ps_mask(A, imm) \
(__mmask8)__builtin_ia32_fpclassps256_mask((__v8sf)(__m256)(A), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#undef __DEFAULT_FN_ATTRS
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
#endif
/*===---- avx512vlintrin.h - AVX512VL intrinsics ---------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#ifndef __AVX512VLINTRIN_H
#define __AVX512VLINTRIN_H
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512vl")))
+#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx512vl"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("avx512vl"), __min_vector_width__(256)))
-/* Doesn't require avx512vl, used in avx512dqintrin.h */
-static __inline __m128i __attribute__((__always_inline__, __nodebug__, __target__("avx512f")))
-_mm_setzero_di(void) {
- return (__m128i)(__v2di){ 0LL, 0LL};
-}
+typedef short __v2hi __attribute__((__vector_size__(4)));
+typedef char __v4qi __attribute__((__vector_size__(4)));
+typedef char __v2qi __attribute__((__vector_size__(2)));
/* Integer compare */
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpeq_epi32_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_pcmpeqd128_mask((__v4si)__a, (__v4si)__b,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpeq_epi32_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_pcmpeqd128_mask((__v4si)__a, (__v4si)__b,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpeq_epu32_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpd128_mask((__v4si)__a, (__v4si)__b, 0,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpeq_epu32_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpd128_mask((__v4si)__a, (__v4si)__b, 0,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpeq_epi32_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_pcmpeqd256_mask((__v8si)__a, (__v8si)__b,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpeq_epi32_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_pcmpeqd256_mask((__v8si)__a, (__v8si)__b,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpeq_epu32_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpd256_mask((__v8si)__a, (__v8si)__b, 0,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpeq_epu32_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpd256_mask((__v8si)__a, (__v8si)__b, 0,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpeq_epi64_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_pcmpeqq128_mask((__v2di)__a, (__v2di)__b,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpeq_epi64_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_pcmpeqq128_mask((__v2di)__a, (__v2di)__b,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpeq_epu64_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpq128_mask((__v2di)__a, (__v2di)__b, 0,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpeq_epu64_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpq128_mask((__v2di)__a, (__v2di)__b, 0,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpeq_epi64_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_pcmpeqq256_mask((__v4di)__a, (__v4di)__b,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpeq_epi64_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_pcmpeqq256_mask((__v4di)__a, (__v4di)__b,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpeq_epu64_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpq256_mask((__v4di)__a, (__v4di)__b, 0,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpeq_epu64_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpq256_mask((__v4di)__a, (__v4di)__b, 0,
- __u);
-}
-
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpge_epi32_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpd128_mask((__v4si)__a, (__v4si)__b, 5,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpge_epi32_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpd128_mask((__v4si)__a, (__v4si)__b, 5,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpge_epu32_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpd128_mask((__v4si)__a, (__v4si)__b, 5,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpge_epu32_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpd128_mask((__v4si)__a, (__v4si)__b, 5,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpge_epi32_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpd256_mask((__v8si)__a, (__v8si)__b, 5,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpge_epi32_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpd256_mask((__v8si)__a, (__v8si)__b, 5,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpge_epu32_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpd256_mask((__v8si)__a, (__v8si)__b, 5,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpge_epu32_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpd256_mask((__v8si)__a, (__v8si)__b, 5,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpge_epi64_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpq128_mask((__v2di)__a, (__v2di)__b, 5,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpge_epi64_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpq128_mask((__v2di)__a, (__v2di)__b, 5,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpge_epu64_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpq128_mask((__v2di)__a, (__v2di)__b, 5,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpge_epu64_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpq128_mask((__v2di)__a, (__v2di)__b, 5,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpge_epi64_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpq256_mask((__v4di)__a, (__v4di)__b, 5,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpge_epi64_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpq256_mask((__v4di)__a, (__v4di)__b, 5,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpge_epu64_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpq256_mask((__v4di)__a, (__v4di)__b, 5,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpge_epu64_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpq256_mask((__v4di)__a, (__v4di)__b, 5,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpgt_epi32_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_pcmpgtd128_mask((__v4si)__a, (__v4si)__b,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpgt_epi32_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_pcmpgtd128_mask((__v4si)__a, (__v4si)__b,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpgt_epu32_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpd128_mask((__v4si)__a, (__v4si)__b, 6,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpgt_epu32_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpd128_mask((__v4si)__a, (__v4si)__b, 6,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpgt_epi32_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_pcmpgtd256_mask((__v8si)__a, (__v8si)__b,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpgt_epi32_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_pcmpgtd256_mask((__v8si)__a, (__v8si)__b,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpgt_epu32_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpd256_mask((__v8si)__a, (__v8si)__b, 6,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpgt_epu32_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpd256_mask((__v8si)__a, (__v8si)__b, 6,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpgt_epi64_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_pcmpgtq128_mask((__v2di)__a, (__v2di)__b,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpgt_epi64_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_pcmpgtq128_mask((__v2di)__a, (__v2di)__b,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpgt_epu64_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpq128_mask((__v2di)__a, (__v2di)__b, 6,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpgt_epu64_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpq128_mask((__v2di)__a, (__v2di)__b, 6,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpgt_epi64_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_pcmpgtq256_mask((__v4di)__a, (__v4di)__b,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpgt_epi64_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_pcmpgtq256_mask((__v4di)__a, (__v4di)__b,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpgt_epu64_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpq256_mask((__v4di)__a, (__v4di)__b, 6,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpgt_epu64_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpq256_mask((__v4di)__a, (__v4di)__b, 6,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmple_epi32_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpd128_mask((__v4si)__a, (__v4si)__b, 2,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmple_epi32_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpd128_mask((__v4si)__a, (__v4si)__b, 2,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmple_epu32_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpd128_mask((__v4si)__a, (__v4si)__b, 2,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmple_epu32_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpd128_mask((__v4si)__a, (__v4si)__b, 2,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmple_epi32_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpd256_mask((__v8si)__a, (__v8si)__b, 2,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmple_epi32_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpd256_mask((__v8si)__a, (__v8si)__b, 2,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmple_epu32_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpd256_mask((__v8si)__a, (__v8si)__b, 2,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmple_epu32_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpd256_mask((__v8si)__a, (__v8si)__b, 2,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmple_epi64_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpq128_mask((__v2di)__a, (__v2di)__b, 2,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmple_epi64_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpq128_mask((__v2di)__a, (__v2di)__b, 2,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmple_epu64_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpq128_mask((__v2di)__a, (__v2di)__b, 2,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmple_epu64_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpq128_mask((__v2di)__a, (__v2di)__b, 2,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmple_epi64_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpq256_mask((__v4di)__a, (__v4di)__b, 2,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmple_epi64_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpq256_mask((__v4di)__a, (__v4di)__b, 2,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmple_epu64_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpq256_mask((__v4di)__a, (__v4di)__b, 2,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmple_epu64_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpq256_mask((__v4di)__a, (__v4di)__b, 2,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmplt_epi32_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpd128_mask((__v4si)__a, (__v4si)__b, 1,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmplt_epi32_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpd128_mask((__v4si)__a, (__v4si)__b, 1,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmplt_epu32_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpd128_mask((__v4si)__a, (__v4si)__b, 1,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmplt_epu32_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpd128_mask((__v4si)__a, (__v4si)__b, 1,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmplt_epi32_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpd256_mask((__v8si)__a, (__v8si)__b, 1,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmplt_epi32_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpd256_mask((__v8si)__a, (__v8si)__b, 1,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmplt_epu32_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpd256_mask((__v8si)__a, (__v8si)__b, 1,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmplt_epu32_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpd256_mask((__v8si)__a, (__v8si)__b, 1,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmplt_epi64_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpq128_mask((__v2di)__a, (__v2di)__b, 1,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmplt_epi64_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpq128_mask((__v2di)__a, (__v2di)__b, 1,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmplt_epu64_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpq128_mask((__v2di)__a, (__v2di)__b, 1,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmplt_epu64_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpq128_mask((__v2di)__a, (__v2di)__b, 1,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmplt_epi64_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpq256_mask((__v4di)__a, (__v4di)__b, 1,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmplt_epi64_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpq256_mask((__v4di)__a, (__v4di)__b, 1,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmplt_epu64_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpq256_mask((__v4di)__a, (__v4di)__b, 1,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmplt_epu64_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpq256_mask((__v4di)__a, (__v4di)__b, 1,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpneq_epi32_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpd128_mask((__v4si)__a, (__v4si)__b, 4,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpneq_epi32_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpd128_mask((__v4si)__a, (__v4si)__b, 4,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpneq_epu32_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpd128_mask((__v4si)__a, (__v4si)__b, 4,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpneq_epu32_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpd128_mask((__v4si)__a, (__v4si)__b, 4,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpneq_epi32_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpd256_mask((__v8si)__a, (__v8si)__b, 4,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpneq_epi32_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpd256_mask((__v8si)__a, (__v8si)__b, 4,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpneq_epu32_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpd256_mask((__v8si)__a, (__v8si)__b, 4,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpneq_epu32_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpd256_mask((__v8si)__a, (__v8si)__b, 4,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpneq_epi64_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpq128_mask((__v2di)__a, (__v2di)__b, 4,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpneq_epi64_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_cmpq128_mask((__v2di)__a, (__v2di)__b, 4,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_cmpneq_epu64_mask(__m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpq128_mask((__v2di)__a, (__v2di)__b, 4,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm_mask_cmpneq_epu64_mask(__mmask8 __u, __m128i __a, __m128i __b) {
- return (__mmask8)__builtin_ia32_ucmpq128_mask((__v2di)__a, (__v2di)__b, 4,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpneq_epi64_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpq256_mask((__v4di)__a, (__v4di)__b, 4,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpneq_epi64_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_cmpq256_mask((__v4di)__a, (__v4di)__b, 4,
- __u);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_cmpneq_epu64_mask(__m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpq256_mask((__v4di)__a, (__v4di)__b, 4,
- (__mmask8)-1);
-}
-
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
-_mm256_mask_cmpneq_epu64_mask(__mmask8 __u, __m256i __a, __m256i __b) {
- return (__mmask8)__builtin_ia32_ucmpq256_mask((__v4di)__a, (__v4di)__b, 4,
- __u);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+#define _mm_cmpeq_epi32_mask(A, B) \
+ _mm_cmp_epi32_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm_mask_cmpeq_epi32_mask(k, A, B) \
+ _mm_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm_cmpge_epi32_mask(A, B) \
+ _mm_cmp_epi32_mask((A), (B), _MM_CMPINT_GE)
+#define _mm_mask_cmpge_epi32_mask(k, A, B) \
+ _mm_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm_cmpgt_epi32_mask(A, B) \
+ _mm_cmp_epi32_mask((A), (B), _MM_CMPINT_GT)
+#define _mm_mask_cmpgt_epi32_mask(k, A, B) \
+ _mm_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm_cmple_epi32_mask(A, B) \
+ _mm_cmp_epi32_mask((A), (B), _MM_CMPINT_LE)
+#define _mm_mask_cmple_epi32_mask(k, A, B) \
+ _mm_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm_cmplt_epi32_mask(A, B) \
+ _mm_cmp_epi32_mask((A), (B), _MM_CMPINT_LT)
+#define _mm_mask_cmplt_epi32_mask(k, A, B) \
+ _mm_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm_cmpneq_epi32_mask(A, B) \
+ _mm_cmp_epi32_mask((A), (B), _MM_CMPINT_NE)
+#define _mm_mask_cmpneq_epi32_mask(k, A, B) \
+ _mm_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm256_cmpeq_epi32_mask(A, B) \
+ _mm256_cmp_epi32_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm256_mask_cmpeq_epi32_mask(k, A, B) \
+ _mm256_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm256_cmpge_epi32_mask(A, B) \
+ _mm256_cmp_epi32_mask((A), (B), _MM_CMPINT_GE)
+#define _mm256_mask_cmpge_epi32_mask(k, A, B) \
+ _mm256_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm256_cmpgt_epi32_mask(A, B) \
+ _mm256_cmp_epi32_mask((A), (B), _MM_CMPINT_GT)
+#define _mm256_mask_cmpgt_epi32_mask(k, A, B) \
+ _mm256_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm256_cmple_epi32_mask(A, B) \
+ _mm256_cmp_epi32_mask((A), (B), _MM_CMPINT_LE)
+#define _mm256_mask_cmple_epi32_mask(k, A, B) \
+ _mm256_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm256_cmplt_epi32_mask(A, B) \
+ _mm256_cmp_epi32_mask((A), (B), _MM_CMPINT_LT)
+#define _mm256_mask_cmplt_epi32_mask(k, A, B) \
+ _mm256_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm256_cmpneq_epi32_mask(A, B) \
+ _mm256_cmp_epi32_mask((A), (B), _MM_CMPINT_NE)
+#define _mm256_mask_cmpneq_epi32_mask(k, A, B) \
+ _mm256_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm_cmpeq_epu32_mask(A, B) \
+ _mm_cmp_epu32_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm_mask_cmpeq_epu32_mask(k, A, B) \
+ _mm_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm_cmpge_epu32_mask(A, B) \
+ _mm_cmp_epu32_mask((A), (B), _MM_CMPINT_GE)
+#define _mm_mask_cmpge_epu32_mask(k, A, B) \
+ _mm_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm_cmpgt_epu32_mask(A, B) \
+ _mm_cmp_epu32_mask((A), (B), _MM_CMPINT_GT)
+#define _mm_mask_cmpgt_epu32_mask(k, A, B) \
+ _mm_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm_cmple_epu32_mask(A, B) \
+ _mm_cmp_epu32_mask((A), (B), _MM_CMPINT_LE)
+#define _mm_mask_cmple_epu32_mask(k, A, B) \
+ _mm_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm_cmplt_epu32_mask(A, B) \
+ _mm_cmp_epu32_mask((A), (B), _MM_CMPINT_LT)
+#define _mm_mask_cmplt_epu32_mask(k, A, B) \
+ _mm_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm_cmpneq_epu32_mask(A, B) \
+ _mm_cmp_epu32_mask((A), (B), _MM_CMPINT_NE)
+#define _mm_mask_cmpneq_epu32_mask(k, A, B) \
+ _mm_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm256_cmpeq_epu32_mask(A, B) \
+ _mm256_cmp_epu32_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm256_mask_cmpeq_epu32_mask(k, A, B) \
+ _mm256_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm256_cmpge_epu32_mask(A, B) \
+ _mm256_cmp_epu32_mask((A), (B), _MM_CMPINT_GE)
+#define _mm256_mask_cmpge_epu32_mask(k, A, B) \
+ _mm256_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm256_cmpgt_epu32_mask(A, B) \
+ _mm256_cmp_epu32_mask((A), (B), _MM_CMPINT_GT)
+#define _mm256_mask_cmpgt_epu32_mask(k, A, B) \
+ _mm256_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm256_cmple_epu32_mask(A, B) \
+ _mm256_cmp_epu32_mask((A), (B), _MM_CMPINT_LE)
+#define _mm256_mask_cmple_epu32_mask(k, A, B) \
+ _mm256_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm256_cmplt_epu32_mask(A, B) \
+ _mm256_cmp_epu32_mask((A), (B), _MM_CMPINT_LT)
+#define _mm256_mask_cmplt_epu32_mask(k, A, B) \
+ _mm256_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm256_cmpneq_epu32_mask(A, B) \
+ _mm256_cmp_epu32_mask((A), (B), _MM_CMPINT_NE)
+#define _mm256_mask_cmpneq_epu32_mask(k, A, B) \
+ _mm256_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm_cmpeq_epi64_mask(A, B) \
+ _mm_cmp_epi64_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm_mask_cmpeq_epi64_mask(k, A, B) \
+ _mm_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm_cmpge_epi64_mask(A, B) \
+ _mm_cmp_epi64_mask((A), (B), _MM_CMPINT_GE)
+#define _mm_mask_cmpge_epi64_mask(k, A, B) \
+ _mm_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm_cmpgt_epi64_mask(A, B) \
+ _mm_cmp_epi64_mask((A), (B), _MM_CMPINT_GT)
+#define _mm_mask_cmpgt_epi64_mask(k, A, B) \
+ _mm_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm_cmple_epi64_mask(A, B) \
+ _mm_cmp_epi64_mask((A), (B), _MM_CMPINT_LE)
+#define _mm_mask_cmple_epi64_mask(k, A, B) \
+ _mm_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm_cmplt_epi64_mask(A, B) \
+ _mm_cmp_epi64_mask((A), (B), _MM_CMPINT_LT)
+#define _mm_mask_cmplt_epi64_mask(k, A, B) \
+ _mm_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm_cmpneq_epi64_mask(A, B) \
+ _mm_cmp_epi64_mask((A), (B), _MM_CMPINT_NE)
+#define _mm_mask_cmpneq_epi64_mask(k, A, B) \
+ _mm_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm256_cmpeq_epi64_mask(A, B) \
+ _mm256_cmp_epi64_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm256_mask_cmpeq_epi64_mask(k, A, B) \
+ _mm256_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm256_cmpge_epi64_mask(A, B) \
+ _mm256_cmp_epi64_mask((A), (B), _MM_CMPINT_GE)
+#define _mm256_mask_cmpge_epi64_mask(k, A, B) \
+ _mm256_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm256_cmpgt_epi64_mask(A, B) \
+ _mm256_cmp_epi64_mask((A), (B), _MM_CMPINT_GT)
+#define _mm256_mask_cmpgt_epi64_mask(k, A, B) \
+ _mm256_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm256_cmple_epi64_mask(A, B) \
+ _mm256_cmp_epi64_mask((A), (B), _MM_CMPINT_LE)
+#define _mm256_mask_cmple_epi64_mask(k, A, B) \
+ _mm256_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm256_cmplt_epi64_mask(A, B) \
+ _mm256_cmp_epi64_mask((A), (B), _MM_CMPINT_LT)
+#define _mm256_mask_cmplt_epi64_mask(k, A, B) \
+ _mm256_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm256_cmpneq_epi64_mask(A, B) \
+ _mm256_cmp_epi64_mask((A), (B), _MM_CMPINT_NE)
+#define _mm256_mask_cmpneq_epi64_mask(k, A, B) \
+ _mm256_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm_cmpeq_epu64_mask(A, B) \
+ _mm_cmp_epu64_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm_mask_cmpeq_epu64_mask(k, A, B) \
+ _mm_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm_cmpge_epu64_mask(A, B) \
+ _mm_cmp_epu64_mask((A), (B), _MM_CMPINT_GE)
+#define _mm_mask_cmpge_epu64_mask(k, A, B) \
+ _mm_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm_cmpgt_epu64_mask(A, B) \
+ _mm_cmp_epu64_mask((A), (B), _MM_CMPINT_GT)
+#define _mm_mask_cmpgt_epu64_mask(k, A, B) \
+ _mm_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm_cmple_epu64_mask(A, B) \
+ _mm_cmp_epu64_mask((A), (B), _MM_CMPINT_LE)
+#define _mm_mask_cmple_epu64_mask(k, A, B) \
+ _mm_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm_cmplt_epu64_mask(A, B) \
+ _mm_cmp_epu64_mask((A), (B), _MM_CMPINT_LT)
+#define _mm_mask_cmplt_epu64_mask(k, A, B) \
+ _mm_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm_cmpneq_epu64_mask(A, B) \
+ _mm_cmp_epu64_mask((A), (B), _MM_CMPINT_NE)
+#define _mm_mask_cmpneq_epu64_mask(k, A, B) \
+ _mm_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_NE)
+
+#define _mm256_cmpeq_epu64_mask(A, B) \
+ _mm256_cmp_epu64_mask((A), (B), _MM_CMPINT_EQ)
+#define _mm256_mask_cmpeq_epu64_mask(k, A, B) \
+ _mm256_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_EQ)
+#define _mm256_cmpge_epu64_mask(A, B) \
+ _mm256_cmp_epu64_mask((A), (B), _MM_CMPINT_GE)
+#define _mm256_mask_cmpge_epu64_mask(k, A, B) \
+ _mm256_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_GE)
+#define _mm256_cmpgt_epu64_mask(A, B) \
+ _mm256_cmp_epu64_mask((A), (B), _MM_CMPINT_GT)
+#define _mm256_mask_cmpgt_epu64_mask(k, A, B) \
+ _mm256_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_GT)
+#define _mm256_cmple_epu64_mask(A, B) \
+ _mm256_cmp_epu64_mask((A), (B), _MM_CMPINT_LE)
+#define _mm256_mask_cmple_epu64_mask(k, A, B) \
+ _mm256_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_LE)
+#define _mm256_cmplt_epu64_mask(A, B) \
+ _mm256_cmp_epu64_mask((A), (B), _MM_CMPINT_LT)
+#define _mm256_mask_cmplt_epu64_mask(k, A, B) \
+ _mm256_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_LT)
+#define _mm256_cmpneq_epu64_mask(A, B) \
+ _mm256_cmp_epu64_mask((A), (B), _MM_CMPINT_NE)
+#define _mm256_mask_cmpneq_epu64_mask(k, A, B) \
+ _mm256_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_NE)
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_add_epi32(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_add_epi32(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_add_epi64(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_add_epi64(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_sub_epi32(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_sub_epi32(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_sub_epi64(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_sub_epi64(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_add_epi32(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_add_epi32(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_add_epi64(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_add_epi64(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_sub_epi32(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_sub_epi32(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_sub_epi64(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_sub_epi64(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_mul_epi32(__m256i __W, __mmask8 __M, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_mul_epi32(__mmask8 __M, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
(__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_mul_epi32(__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__M,
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_mul_epi32(__mmask8 __M, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__M,
(__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_mul_epu32(__m256i __W, __mmask8 __M, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_mul_epu32(__mmask8 __M, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
(__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_mul_epu32(__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__M,
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_mul_epu32(__mmask8 __M, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__M,
(__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_mullo_epi32(__mmask8 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_mullo_epi32(__m256i __W, __mmask8 __M, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
(__v8si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_mullo_epi32(__mmask8 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__M,
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_mullo_epi32(__m128i __W, __mmask8 __M, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__M,
(__v4si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_and_epi32(__m256i __a, __m256i __b)
+{
+ return (__m256i)((__v8su)__a & (__v8su)__b);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_and_epi32(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
- (__v8si)_mm256_and_si256(__A, __B),
+ (__v8si)_mm256_and_epi32(__A, __B),
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_and_epi32(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)_mm256_mask_and_epi32(_mm256_setzero_si256(), __U, __A, __B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_and_epi32(__m128i __a, __m128i __b)
+{
+ return (__m128i)((__v4su)__a & (__v4su)__b);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_and_epi32(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
- (__v4si)_mm_and_si128(__A, __B),
+ (__v4si)_mm_and_epi32(__A, __B),
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_and_epi32(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)_mm_mask_and_epi32(_mm_setzero_si128(), __U, __A, __B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_andnot_epi32(__m256i __A, __m256i __B)
+{
+ return (__m256i)(~(__v8su)__A & (__v8su)__B);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_andnot_epi32(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
- (__v8si)_mm256_andnot_si256(__A, __B),
+ (__v8si)_mm256_andnot_epi32(__A, __B),
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_andnot_epi32(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)_mm256_mask_andnot_epi32(_mm256_setzero_si256(),
__U, __A, __B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_andnot_epi32(__m128i __A, __m128i __B)
+{
+ return (__m128i)(~(__v4su)__A & (__v4su)__B);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_andnot_epi32(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
- (__v4si)_mm_andnot_si128(__A, __B),
+ (__v4si)_mm_andnot_epi32(__A, __B),
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_andnot_epi32 (__mmask8 __U, __m128i __A, __m128i __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_andnot_epi32(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)_mm_mask_andnot_epi32(_mm_setzero_si128(), __U, __A, __B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_or_epi32(__m256i __a, __m256i __b)
+{
+ return (__m256i)((__v8su)__a | (__v8su)__b);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_or_epi32 (__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
- (__v8si)_mm256_or_si256(__A, __B),
+ (__v8si)_mm256_or_epi32(__A, __B),
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_or_epi32(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)_mm256_mask_or_epi32(_mm256_setzero_si256(), __U, __A, __B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_or_epi32(__m128i __a, __m128i __b)
+{
+ return (__m128i)((__v4su)__a | (__v4su)__b);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_or_epi32(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
- (__v4si)_mm_or_si128(__A, __B),
+ (__v4si)_mm_or_epi32(__A, __B),
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_or_epi32(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)_mm_mask_or_epi32(_mm_setzero_si128(), __U, __A, __B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_xor_epi32(__m256i __a, __m256i __b)
+{
+ return (__m256i)((__v8su)__a ^ (__v8su)__b);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_xor_epi32(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
- (__v8si)_mm256_xor_si256(__A, __B),
+ (__v8si)_mm256_xor_epi32(__A, __B),
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_xor_epi32(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)_mm256_mask_xor_epi32(_mm256_setzero_si256(), __U, __A, __B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_xor_epi32(__m128i __W, __mmask8 __U, __m128i __A,
- __m128i __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_xor_epi32(__m128i __a, __m128i __b)
+{
+ return (__m128i)((__v4su)__a ^ (__v4su)__b);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_xor_epi32(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
- (__v4si)_mm_xor_si128(__A, __B),
+ (__v4si)_mm_xor_epi32(__A, __B),
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_xor_epi32(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)_mm_mask_xor_epi32(_mm_setzero_si128(), __U, __A, __B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_and_epi64(__m256i __a, __m256i __b)
+{
+ return (__m256i)((__v4du)__a & (__v4du)__b);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_and_epi64(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_and_si256(__A, __B),
+ (__v4di)_mm256_and_epi64(__A, __B),
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_and_epi64(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)_mm256_mask_and_epi64(_mm256_setzero_si256(), __U, __A, __B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_and_epi64(__m128i __a, __m128i __b)
+{
+ return (__m128i)((__v2du)__a & (__v2du)__b);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_and_epi64(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_and_si128(__A, __B),
+ (__v2di)_mm_and_epi64(__A, __B),
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_and_epi64(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)_mm_mask_and_epi64(_mm_setzero_si128(), __U, __A, __B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_andnot_epi64(__m256i __A, __m256i __B)
+{
+ return (__m256i)(~(__v4du)__A & (__v4du)__B);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_andnot_epi64(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_andnot_si256(__A, __B),
+ (__v4di)_mm256_andnot_epi64(__A, __B),
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_andnot_epi64(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)_mm256_mask_andnot_epi64(_mm256_setzero_si256(),
__U, __A, __B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_andnot_epi64(__m128i __A, __m128i __B)
+{
+ return (__m128i)(~(__v2du)__A & (__v2du)__B);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_andnot_epi64(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_andnot_si128(__A, __B),
+ (__v2di)_mm_andnot_epi64(__A, __B),
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_andnot_epi64(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)_mm_mask_andnot_epi64(_mm_setzero_si128(), __U, __A, __B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_or_epi64(__m256i __a, __m256i __b)
+{
+ return (__m256i)((__v4du)__a | (__v4du)__b);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_or_epi64(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_or_si256(__A, __B),
+ (__v4di)_mm256_or_epi64(__A, __B),
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_or_epi64(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)_mm256_mask_or_epi64(_mm256_setzero_si256(), __U, __A, __B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_or_epi64(__m128i __a, __m128i __b)
+{
+ return (__m128i)((__v2du)__a | (__v2du)__b);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_or_epi64(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_or_si128(__A, __B),
+ (__v2di)_mm_or_epi64(__A, __B),
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_or_epi64(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)_mm_mask_or_epi64(_mm_setzero_si128(), __U, __A, __B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_xor_epi64(__m256i __a, __m256i __b)
+{
+ return (__m256i)((__v4du)__a ^ (__v4du)__b);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_xor_epi64(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_xor_si256(__A, __B),
+ (__v4di)_mm256_xor_epi64(__A, __B),
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_xor_epi64(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)_mm256_mask_xor_epi64(_mm256_setzero_si256(), __U, __A, __B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_xor_epi64(__m128i __a, __m128i __b)
+{
+ return (__m128i)((__v2du)__a ^ (__v2du)__b);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_xor_epi64(__m128i __W, __mmask8 __U, __m128i __A,
__m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_xor_si128(__A, __B),
+ (__v2di)_mm_xor_epi64(__A, __B),
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_xor_epi64(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)_mm_mask_xor_epi64(_mm_setzero_si128(), __U, __A, __B);
}
-#define _mm_cmp_epi32_mask(a, b, p) __extension__ ({ \
+#define _mm_cmp_epi32_mask(a, b, p) \
(__mmask8)__builtin_ia32_cmpd128_mask((__v4si)(__m128i)(a), \
(__v4si)(__m128i)(b), (int)(p), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_cmp_epi32_mask(m, a, b, p) __extension__ ({ \
+#define _mm_mask_cmp_epi32_mask(m, a, b, p) \
(__mmask8)__builtin_ia32_cmpd128_mask((__v4si)(__m128i)(a), \
(__v4si)(__m128i)(b), (int)(p), \
- (__mmask8)(m)); })
+ (__mmask8)(m))
-#define _mm_cmp_epu32_mask(a, b, p) __extension__ ({ \
+#define _mm_cmp_epu32_mask(a, b, p) \
(__mmask8)__builtin_ia32_ucmpd128_mask((__v4si)(__m128i)(a), \
(__v4si)(__m128i)(b), (int)(p), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_cmp_epu32_mask(m, a, b, p) __extension__ ({ \
+#define _mm_mask_cmp_epu32_mask(m, a, b, p) \
(__mmask8)__builtin_ia32_ucmpd128_mask((__v4si)(__m128i)(a), \
(__v4si)(__m128i)(b), (int)(p), \
- (__mmask8)(m)); })
+ (__mmask8)(m))
-#define _mm256_cmp_epi32_mask(a, b, p) __extension__ ({ \
+#define _mm256_cmp_epi32_mask(a, b, p) \
(__mmask8)__builtin_ia32_cmpd256_mask((__v8si)(__m256i)(a), \
(__v8si)(__m256i)(b), (int)(p), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_cmp_epi32_mask(m, a, b, p) __extension__ ({ \
+#define _mm256_mask_cmp_epi32_mask(m, a, b, p) \
(__mmask8)__builtin_ia32_cmpd256_mask((__v8si)(__m256i)(a), \
(__v8si)(__m256i)(b), (int)(p), \
- (__mmask8)(m)); })
+ (__mmask8)(m))
-#define _mm256_cmp_epu32_mask(a, b, p) __extension__ ({ \
+#define _mm256_cmp_epu32_mask(a, b, p) \
(__mmask8)__builtin_ia32_ucmpd256_mask((__v8si)(__m256i)(a), \
(__v8si)(__m256i)(b), (int)(p), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_cmp_epu32_mask(m, a, b, p) __extension__ ({ \
+#define _mm256_mask_cmp_epu32_mask(m, a, b, p) \
(__mmask8)__builtin_ia32_ucmpd256_mask((__v8si)(__m256i)(a), \
(__v8si)(__m256i)(b), (int)(p), \
- (__mmask8)(m)); })
+ (__mmask8)(m))
-#define _mm_cmp_epi64_mask(a, b, p) __extension__ ({ \
+#define _mm_cmp_epi64_mask(a, b, p) \
(__mmask8)__builtin_ia32_cmpq128_mask((__v2di)(__m128i)(a), \
(__v2di)(__m128i)(b), (int)(p), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_cmp_epi64_mask(m, a, b, p) __extension__ ({ \
+#define _mm_mask_cmp_epi64_mask(m, a, b, p) \
(__mmask8)__builtin_ia32_cmpq128_mask((__v2di)(__m128i)(a), \
(__v2di)(__m128i)(b), (int)(p), \
- (__mmask8)(m)); })
+ (__mmask8)(m))
-#define _mm_cmp_epu64_mask(a, b, p) __extension__ ({ \
+#define _mm_cmp_epu64_mask(a, b, p) \
(__mmask8)__builtin_ia32_ucmpq128_mask((__v2di)(__m128i)(a), \
(__v2di)(__m128i)(b), (int)(p), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_cmp_epu64_mask(m, a, b, p) __extension__ ({ \
+#define _mm_mask_cmp_epu64_mask(m, a, b, p) \
(__mmask8)__builtin_ia32_ucmpq128_mask((__v2di)(__m128i)(a), \
(__v2di)(__m128i)(b), (int)(p), \
- (__mmask8)(m)); })
+ (__mmask8)(m))
-#define _mm256_cmp_epi64_mask(a, b, p) __extension__ ({ \
+#define _mm256_cmp_epi64_mask(a, b, p) \
(__mmask8)__builtin_ia32_cmpq256_mask((__v4di)(__m256i)(a), \
(__v4di)(__m256i)(b), (int)(p), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_cmp_epi64_mask(m, a, b, p) __extension__ ({ \
+#define _mm256_mask_cmp_epi64_mask(m, a, b, p) \
(__mmask8)__builtin_ia32_cmpq256_mask((__v4di)(__m256i)(a), \
(__v4di)(__m256i)(b), (int)(p), \
- (__mmask8)(m)); })
+ (__mmask8)(m))
-#define _mm256_cmp_epu64_mask(a, b, p) __extension__ ({ \
+#define _mm256_cmp_epu64_mask(a, b, p) \
(__mmask8)__builtin_ia32_ucmpq256_mask((__v4di)(__m256i)(a), \
(__v4di)(__m256i)(b), (int)(p), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_cmp_epu64_mask(m, a, b, p) __extension__ ({ \
+#define _mm256_mask_cmp_epu64_mask(m, a, b, p) \
(__mmask8)__builtin_ia32_ucmpq256_mask((__v4di)(__m256i)(a), \
(__v4di)(__m256i)(b), (int)(p), \
- (__mmask8)(m)); })
+ (__mmask8)(m))
-#define _mm256_cmp_ps_mask(a, b, p) __extension__ ({ \
+#define _mm256_cmp_ps_mask(a, b, p) \
(__mmask8)__builtin_ia32_cmpps256_mask((__v8sf)(__m256)(a), \
(__v8sf)(__m256)(b), (int)(p), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_cmp_ps_mask(m, a, b, p) __extension__ ({ \
+#define _mm256_mask_cmp_ps_mask(m, a, b, p) \
(__mmask8)__builtin_ia32_cmpps256_mask((__v8sf)(__m256)(a), \
(__v8sf)(__m256)(b), (int)(p), \
- (__mmask8)(m)); })
+ (__mmask8)(m))
-#define _mm256_cmp_pd_mask(a, b, p) __extension__ ({ \
+#define _mm256_cmp_pd_mask(a, b, p) \
(__mmask8)__builtin_ia32_cmppd256_mask((__v4df)(__m256d)(a), \
(__v4df)(__m256d)(b), (int)(p), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_cmp_pd_mask(m, a, b, p) __extension__ ({ \
+#define _mm256_mask_cmp_pd_mask(m, a, b, p) \
(__mmask8)__builtin_ia32_cmppd256_mask((__v4df)(__m256d)(a), \
(__v4df)(__m256d)(b), (int)(p), \
- (__mmask8)(m)); })
+ (__mmask8)(m))
-#define _mm_cmp_ps_mask(a, b, p) __extension__ ({ \
+#define _mm_cmp_ps_mask(a, b, p) \
(__mmask8)__builtin_ia32_cmpps128_mask((__v4sf)(__m128)(a), \
(__v4sf)(__m128)(b), (int)(p), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_cmp_ps_mask(m, a, b, p) __extension__ ({ \
+#define _mm_mask_cmp_ps_mask(m, a, b, p) \
(__mmask8)__builtin_ia32_cmpps128_mask((__v4sf)(__m128)(a), \
(__v4sf)(__m128)(b), (int)(p), \
- (__mmask8)(m)); })
+ (__mmask8)(m))
-#define _mm_cmp_pd_mask(a, b, p) __extension__ ({ \
+#define _mm_cmp_pd_mask(a, b, p) \
(__mmask8)__builtin_ia32_cmppd128_mask((__v2df)(__m128d)(a), \
(__v2df)(__m128d)(b), (int)(p), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_cmp_pd_mask(m, a, b, p) __extension__ ({ \
+#define _mm_mask_cmp_pd_mask(m, a, b, p) \
(__mmask8)__builtin_ia32_cmppd128_mask((__v2df)(__m128d)(a), \
(__v2df)(__m128d)(b), (int)(p), \
- (__mmask8)(m)); })
+ (__mmask8)(m))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_fmadd_pd(__m128d __A, __mmask8 __U, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfmaddpd128_mask ((__v2df) __A,
- (__v2df) __B,
- (__v2df) __C,
- (__mmask8) __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddpd ((__v2df) __A,
+ (__v2df) __B,
+ (__v2df) __C),
+ (__v2df) __A);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask3_fmadd_pd(__m128d __A, __m128d __B, __m128d __C, __mmask8 __U)
{
- return (__m128d) __builtin_ia32_vfmaddpd128_mask3 ((__v2df) __A,
- (__v2df) __B,
- (__v2df) __C,
- (__mmask8) __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddpd ((__v2df) __A,
+ (__v2df) __B,
+ (__v2df) __C),
+ (__v2df) __C);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_fmadd_pd(__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfmaddpd128_maskz ((__v2df) __A,
- (__v2df) __B,
- (__v2df) __C,
- (__mmask8) __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddpd ((__v2df) __A,
+ (__v2df) __B,
+ (__v2df) __C),
+ (__v2df)_mm_setzero_pd());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_fmsub_pd(__m128d __A, __mmask8 __U, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfmaddpd128_mask ((__v2df) __A,
- (__v2df) __B,
- -(__v2df) __C,
- (__mmask8) __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddpd ((__v2df) __A,
+ (__v2df) __B,
+ -(__v2df) __C),
+ (__v2df) __A);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_fmsub_pd(__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfmaddpd128_maskz ((__v2df) __A,
- (__v2df) __B,
- -(__v2df) __C,
- (__mmask8) __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddpd ((__v2df) __A,
+ (__v2df) __B,
+ -(__v2df) __C),
+ (__v2df)_mm_setzero_pd());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask3_fnmadd_pd(__m128d __A, __m128d __B, __m128d __C, __mmask8 __U)
{
- return (__m128d) __builtin_ia32_vfmaddpd128_mask3 (-(__v2df) __A,
- (__v2df) __B,
- (__v2df) __C,
- (__mmask8) __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddpd (-(__v2df) __A,
+ (__v2df) __B,
+ (__v2df) __C),
+ (__v2df) __C);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_fnmadd_pd(__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfmaddpd128_maskz (-(__v2df) __A,
- (__v2df) __B,
- (__v2df) __C,
- (__mmask8) __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddpd (-(__v2df) __A,
+ (__v2df) __B,
+ (__v2df) __C),
+ (__v2df)_mm_setzero_pd());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_fnmsub_pd(__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfmaddpd128_maskz (-(__v2df) __A,
- (__v2df) __B,
- -(__v2df) __C,
- (__mmask8) __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddpd (-(__v2df) __A,
+ (__v2df) __B,
+ -(__v2df) __C),
+ (__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_fmadd_pd(__m256d __A, __mmask8 __U, __m256d __B, __m256d __C)
{
- return (__m256d) __builtin_ia32_vfmaddpd256_mask ((__v4df) __A,
- (__v4df) __B,
- (__v4df) __C,
- (__mmask8) __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddpd256 ((__v4df) __A,
+ (__v4df) __B,
+ (__v4df) __C),
+ (__v4df) __A);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask3_fmadd_pd(__m256d __A, __m256d __B, __m256d __C, __mmask8 __U)
{
- return (__m256d) __builtin_ia32_vfmaddpd256_mask3 ((__v4df) __A,
- (__v4df) __B,
- (__v4df) __C,
- (__mmask8) __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddpd256 ((__v4df) __A,
+ (__v4df) __B,
+ (__v4df) __C),
+ (__v4df) __C);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_fmadd_pd(__mmask8 __U, __m256d __A, __m256d __B, __m256d __C)
{
- return (__m256d) __builtin_ia32_vfmaddpd256_maskz ((__v4df) __A,
- (__v4df) __B,
- (__v4df) __C,
- (__mmask8) __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddpd256 ((__v4df) __A,
+ (__v4df) __B,
+ (__v4df) __C),
+ (__v4df)_mm256_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_fmsub_pd(__m256d __A, __mmask8 __U, __m256d __B, __m256d __C)
{
- return (__m256d) __builtin_ia32_vfmaddpd256_mask ((__v4df) __A,
- (__v4df) __B,
- -(__v4df) __C,
- (__mmask8) __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddpd256 ((__v4df) __A,
+ (__v4df) __B,
+ -(__v4df) __C),
+ (__v4df) __A);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_fmsub_pd(__mmask8 __U, __m256d __A, __m256d __B, __m256d __C)
{
- return (__m256d) __builtin_ia32_vfmaddpd256_maskz ((__v4df) __A,
- (__v4df) __B,
- -(__v4df) __C,
- (__mmask8) __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddpd256 ((__v4df) __A,
+ (__v4df) __B,
+ -(__v4df) __C),
+ (__v4df)_mm256_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask3_fnmadd_pd(__m256d __A, __m256d __B, __m256d __C, __mmask8 __U)
{
- return (__m256d) __builtin_ia32_vfmaddpd256_mask3 (-(__v4df) __A,
- (__v4df) __B,
- (__v4df) __C,
- (__mmask8) __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddpd256 (-(__v4df) __A,
+ (__v4df) __B,
+ (__v4df) __C),
+ (__v4df) __C);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_fnmadd_pd(__mmask8 __U, __m256d __A, __m256d __B, __m256d __C)
{
- return (__m256d) __builtin_ia32_vfmaddpd256_maskz (-(__v4df) __A,
- (__v4df) __B,
- (__v4df) __C,
- (__mmask8) __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddpd256 (-(__v4df) __A,
+ (__v4df) __B,
+ (__v4df) __C),
+ (__v4df)_mm256_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_fnmsub_pd(__mmask8 __U, __m256d __A, __m256d __B, __m256d __C)
{
- return (__m256d) __builtin_ia32_vfmaddpd256_maskz (-(__v4df) __A,
- (__v4df) __B,
- -(__v4df) __C,
- (__mmask8) __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddpd256 (-(__v4df) __A,
+ (__v4df) __B,
+ -(__v4df) __C),
+ (__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_fmadd_ps(__m128 __A, __mmask8 __U, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfmaddps128_mask ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddps ((__v4sf) __A,
+ (__v4sf) __B,
+ (__v4sf) __C),
+ (__v4sf) __A);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask3_fmadd_ps(__m128 __A, __m128 __B, __m128 __C, __mmask8 __U)
{
- return (__m128) __builtin_ia32_vfmaddps128_mask3 ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddps ((__v4sf) __A,
+ (__v4sf) __B,
+ (__v4sf) __C),
+ (__v4sf) __C);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_fmadd_ps(__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfmaddps128_maskz ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddps ((__v4sf) __A,
+ (__v4sf) __B,
+ (__v4sf) __C),
+ (__v4sf)_mm_setzero_ps());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_fmsub_ps(__m128 __A, __mmask8 __U, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfmaddps128_mask ((__v4sf) __A,
- (__v4sf) __B,
- -(__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddps ((__v4sf) __A,
+ (__v4sf) __B,
+ -(__v4sf) __C),
+ (__v4sf) __A);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_fmsub_ps(__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfmaddps128_maskz ((__v4sf) __A,
- (__v4sf) __B,
- -(__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddps ((__v4sf) __A,
+ (__v4sf) __B,
+ -(__v4sf) __C),
+ (__v4sf)_mm_setzero_ps());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask3_fnmadd_ps(__m128 __A, __m128 __B, __m128 __C, __mmask8 __U)
{
- return (__m128) __builtin_ia32_vfmaddps128_mask3 (-(__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddps (-(__v4sf) __A,
+ (__v4sf) __B,
+ (__v4sf) __C),
+ (__v4sf) __C);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_fnmadd_ps(__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfmaddps128_maskz (-(__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddps (-(__v4sf) __A,
+ (__v4sf) __B,
+ (__v4sf) __C),
+ (__v4sf)_mm_setzero_ps());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_fnmsub_ps(__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfmaddps128_maskz (-(__v4sf) __A,
- (__v4sf) __B,
- -(__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddps (-(__v4sf) __A,
+ (__v4sf) __B,
+ -(__v4sf) __C),
+ (__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_fmadd_ps(__m256 __A, __mmask8 __U, __m256 __B, __m256 __C)
{
- return (__m256) __builtin_ia32_vfmaddps256_mask ((__v8sf) __A,
- (__v8sf) __B,
- (__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddps256 ((__v8sf) __A,
+ (__v8sf) __B,
+ (__v8sf) __C),
+ (__v8sf) __A);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask3_fmadd_ps(__m256 __A, __m256 __B, __m256 __C, __mmask8 __U)
{
- return (__m256) __builtin_ia32_vfmaddps256_mask3 ((__v8sf) __A,
- (__v8sf) __B,
- (__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddps256 ((__v8sf) __A,
+ (__v8sf) __B,
+ (__v8sf) __C),
+ (__v8sf) __C);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_fmadd_ps(__mmask8 __U, __m256 __A, __m256 __B, __m256 __C)
{
- return (__m256) __builtin_ia32_vfmaddps256_maskz ((__v8sf) __A,
- (__v8sf) __B,
- (__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddps256 ((__v8sf) __A,
+ (__v8sf) __B,
+ (__v8sf) __C),
+ (__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_fmsub_ps(__m256 __A, __mmask8 __U, __m256 __B, __m256 __C)
{
- return (__m256) __builtin_ia32_vfmaddps256_mask ((__v8sf) __A,
- (__v8sf) __B,
- -(__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddps256 ((__v8sf) __A,
+ (__v8sf) __B,
+ -(__v8sf) __C),
+ (__v8sf) __A);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_fmsub_ps(__mmask8 __U, __m256 __A, __m256 __B, __m256 __C)
{
- return (__m256) __builtin_ia32_vfmaddps256_maskz ((__v8sf) __A,
- (__v8sf) __B,
- -(__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddps256 ((__v8sf) __A,
+ (__v8sf) __B,
+ -(__v8sf) __C),
+ (__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask3_fnmadd_ps(__m256 __A, __m256 __B, __m256 __C, __mmask8 __U)
{
- return (__m256) __builtin_ia32_vfmaddps256_mask3 (-(__v8sf) __A,
- (__v8sf) __B,
- (__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddps256 (-(__v8sf) __A,
+ (__v8sf) __B,
+ (__v8sf) __C),
+ (__v8sf) __C);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_fnmadd_ps(__mmask8 __U, __m256 __A, __m256 __B, __m256 __C)
{
- return (__m256) __builtin_ia32_vfmaddps256_maskz (-(__v8sf) __A,
- (__v8sf) __B,
- (__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddps256 (-(__v8sf) __A,
+ (__v8sf) __B,
+ (__v8sf) __C),
+ (__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_fnmsub_ps(__mmask8 __U, __m256 __A, __m256 __B, __m256 __C)
{
- return (__m256) __builtin_ia32_vfmaddps256_maskz (-(__v8sf) __A,
- (__v8sf) __B,
- -(__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddps256 (-(__v8sf) __A,
+ (__v8sf) __B,
+ -(__v8sf) __C),
+ (__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_fmaddsub_pd(__m128d __A, __mmask8 __U, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfmaddsubpd128_mask ((__v2df) __A,
- (__v2df) __B,
- (__v2df) __C,
- (__mmask8) __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddsubpd ((__v2df) __A,
+ (__v2df) __B,
+ (__v2df) __C),
+ (__v2df) __A);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask3_fmaddsub_pd(__m128d __A, __m128d __B, __m128d __C, __mmask8 __U)
{
- return (__m128d) __builtin_ia32_vfmaddsubpd128_mask3 ((__v2df) __A,
- (__v2df) __B,
- (__v2df) __C,
- (__mmask8)
- __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddsubpd ((__v2df) __A,
+ (__v2df) __B,
+ (__v2df) __C),
+ (__v2df) __C);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_fmaddsub_pd(__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfmaddsubpd128_maskz ((__v2df) __A,
- (__v2df) __B,
- (__v2df) __C,
- (__mmask8)
- __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddsubpd ((__v2df) __A,
+ (__v2df) __B,
+ (__v2df) __C),
+ (__v2df)_mm_setzero_pd());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_fmsubadd_pd(__m128d __A, __mmask8 __U, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfmaddsubpd128_mask ((__v2df) __A,
- (__v2df) __B,
- -(__v2df) __C,
- (__mmask8) __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddsubpd ((__v2df) __A,
+ (__v2df) __B,
+ -(__v2df) __C),
+ (__v2df) __A);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_fmsubadd_pd(__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfmaddsubpd128_maskz ((__v2df) __A,
- (__v2df) __B,
- -(__v2df) __C,
- (__mmask8)
- __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddsubpd ((__v2df) __A,
+ (__v2df) __B,
+ -(__v2df) __C),
+ (__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_fmaddsub_pd(__m256d __A, __mmask8 __U, __m256d __B, __m256d __C)
{
- return (__m256d) __builtin_ia32_vfmaddsubpd256_mask ((__v4df) __A,
- (__v4df) __B,
- (__v4df) __C,
- (__mmask8) __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddsubpd256 ((__v4df) __A,
+ (__v4df) __B,
+ (__v4df) __C),
+ (__v4df) __A);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask3_fmaddsub_pd(__m256d __A, __m256d __B, __m256d __C, __mmask8 __U)
{
- return (__m256d) __builtin_ia32_vfmaddsubpd256_mask3 ((__v4df) __A,
- (__v4df) __B,
- (__v4df) __C,
- (__mmask8)
- __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddsubpd256 ((__v4df) __A,
+ (__v4df) __B,
+ (__v4df) __C),
+ (__v4df) __C);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_fmaddsub_pd(__mmask8 __U, __m256d __A, __m256d __B, __m256d __C)
{
- return (__m256d) __builtin_ia32_vfmaddsubpd256_maskz ((__v4df) __A,
- (__v4df) __B,
- (__v4df) __C,
- (__mmask8)
- __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddsubpd256 ((__v4df) __A,
+ (__v4df) __B,
+ (__v4df) __C),
+ (__v4df)_mm256_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_fmsubadd_pd(__m256d __A, __mmask8 __U, __m256d __B, __m256d __C)
{
- return (__m256d) __builtin_ia32_vfmaddsubpd256_mask ((__v4df) __A,
- (__v4df) __B,
- -(__v4df) __C,
- (__mmask8) __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddsubpd256 ((__v4df) __A,
+ (__v4df) __B,
+ -(__v4df) __C),
+ (__v4df) __A);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_fmsubadd_pd(__mmask8 __U, __m256d __A, __m256d __B, __m256d __C)
{
- return (__m256d) __builtin_ia32_vfmaddsubpd256_maskz ((__v4df) __A,
- (__v4df) __B,
- -(__v4df) __C,
- (__mmask8)
- __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddsubpd256 ((__v4df) __A,
+ (__v4df) __B,
+ -(__v4df) __C),
+ (__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_fmaddsub_ps(__m128 __A, __mmask8 __U, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfmaddsubps128_mask ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddsubps ((__v4sf) __A,
+ (__v4sf) __B,
+ (__v4sf) __C),
+ (__v4sf) __A);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask3_fmaddsub_ps(__m128 __A, __m128 __B, __m128 __C, __mmask8 __U)
{
- return (__m128) __builtin_ia32_vfmaddsubps128_mask3 ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddsubps ((__v4sf) __A,
+ (__v4sf) __B,
+ (__v4sf) __C),
+ (__v4sf) __C);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_fmaddsub_ps(__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfmaddsubps128_maskz ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddsubps ((__v4sf) __A,
+ (__v4sf) __B,
+ (__v4sf) __C),
+ (__v4sf)_mm_setzero_ps());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_fmsubadd_ps(__m128 __A, __mmask8 __U, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfmaddsubps128_mask ((__v4sf) __A,
- (__v4sf) __B,
- -(__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddsubps ((__v4sf) __A,
+ (__v4sf) __B,
+ -(__v4sf) __C),
+ (__v4sf) __A);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_fmsubadd_ps(__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfmaddsubps128_maskz ((__v4sf) __A,
- (__v4sf) __B,
- -(__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddsubps ((__v4sf) __A,
+ (__v4sf) __B,
+ -(__v4sf) __C),
+ (__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_fmaddsub_ps(__m256 __A, __mmask8 __U, __m256 __B,
__m256 __C)
{
- return (__m256) __builtin_ia32_vfmaddsubps256_mask ((__v8sf) __A,
- (__v8sf) __B,
- (__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddsubps256 ((__v8sf) __A,
+ (__v8sf) __B,
+ (__v8sf) __C),
+ (__v8sf) __A);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask3_fmaddsub_ps(__m256 __A, __m256 __B, __m256 __C, __mmask8 __U)
{
- return (__m256) __builtin_ia32_vfmaddsubps256_mask3 ((__v8sf) __A,
- (__v8sf) __B,
- (__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddsubps256 ((__v8sf) __A,
+ (__v8sf) __B,
+ (__v8sf) __C),
+ (__v8sf) __C);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_fmaddsub_ps(__mmask8 __U, __m256 __A, __m256 __B, __m256 __C)
{
- return (__m256) __builtin_ia32_vfmaddsubps256_maskz ((__v8sf) __A,
- (__v8sf) __B,
- (__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddsubps256 ((__v8sf) __A,
+ (__v8sf) __B,
+ (__v8sf) __C),
+ (__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_fmsubadd_ps(__m256 __A, __mmask8 __U, __m256 __B, __m256 __C)
{
- return (__m256) __builtin_ia32_vfmaddsubps256_mask ((__v8sf) __A,
- (__v8sf) __B,
- -(__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddsubps256 ((__v8sf) __A,
+ (__v8sf) __B,
+ -(__v8sf) __C),
+ (__v8sf) __A);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_fmsubadd_ps(__mmask8 __U, __m256 __A, __m256 __B, __m256 __C)
{
- return (__m256) __builtin_ia32_vfmaddsubps256_maskz ((__v8sf) __A,
- (__v8sf) __B,
- -(__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddsubps256 ((__v8sf) __A,
+ (__v8sf) __B,
+ -(__v8sf) __C),
+ (__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask3_fmsub_pd(__m128d __A, __m128d __B, __m128d __C, __mmask8 __U)
{
- return (__m128d) __builtin_ia32_vfmsubpd128_mask3 ((__v2df) __A,
- (__v2df) __B,
- (__v2df) __C,
- (__mmask8) __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddpd ((__v2df) __A,
+ (__v2df) __B,
+ -(__v2df) __C),
+ (__v2df) __C);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask3_fmsub_pd(__m256d __A, __m256d __B, __m256d __C, __mmask8 __U)
{
- return (__m256d) __builtin_ia32_vfmsubpd256_mask3 ((__v4df) __A,
- (__v4df) __B,
- (__v4df) __C,
- (__mmask8) __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddpd256 ((__v4df) __A,
+ (__v4df) __B,
+ -(__v4df) __C),
+ (__v4df) __C);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask3_fmsub_ps(__m128 __A, __m128 __B, __m128 __C, __mmask8 __U)
{
- return (__m128) __builtin_ia32_vfmsubps128_mask3 ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddps ((__v4sf) __A,
+ (__v4sf) __B,
+ -(__v4sf) __C),
+ (__v4sf) __C);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask3_fmsub_ps(__m256 __A, __m256 __B, __m256 __C, __mmask8 __U)
{
- return (__m256) __builtin_ia32_vfmsubps256_mask3 ((__v8sf) __A,
- (__v8sf) __B,
- (__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddps256 ((__v8sf) __A,
+ (__v8sf) __B,
+ -(__v8sf) __C),
+ (__v8sf) __C);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask3_fmsubadd_pd(__m128d __A, __m128d __B, __m128d __C, __mmask8 __U)
{
- return (__m128d) __builtin_ia32_vfmsubaddpd128_mask3 ((__v2df) __A,
- (__v2df) __B,
- (__v2df) __C,
- (__mmask8)
- __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddsubpd ((__v2df) __A,
+ (__v2df) __B,
+ -(__v2df) __C),
+ (__v2df) __C);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask3_fmsubadd_pd(__m256d __A, __m256d __B, __m256d __C, __mmask8 __U)
{
- return (__m256d) __builtin_ia32_vfmsubaddpd256_mask3 ((__v4df) __A,
- (__v4df) __B,
- (__v4df) __C,
- (__mmask8)
- __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddsubpd256 ((__v4df) __A,
+ (__v4df) __B,
+ -(__v4df) __C),
+ (__v4df) __C);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask3_fmsubadd_ps(__m128 __A, __m128 __B, __m128 __C, __mmask8 __U)
{
- return (__m128) __builtin_ia32_vfmsubaddps128_mask3 ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddsubps ((__v4sf) __A,
+ (__v4sf) __B,
+ -(__v4sf) __C),
+ (__v4sf) __C);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask3_fmsubadd_ps(__m256 __A, __m256 __B, __m256 __C, __mmask8 __U)
{
- return (__m256) __builtin_ia32_vfmsubaddps256_mask3 ((__v8sf) __A,
- (__v8sf) __B,
- (__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddsubps256 ((__v8sf) __A,
+ (__v8sf) __B,
+ -(__v8sf) __C),
+ (__v8sf) __C);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_fnmadd_pd(__m128d __A, __mmask8 __U, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfnmaddpd128_mask ((__v2df) __A,
- (__v2df) __B,
- (__v2df) __C,
- (__mmask8) __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddpd ((__v2df) __A,
+ -(__v2df) __B,
+ (__v2df) __C),
+ (__v2df) __A);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_fnmadd_pd(__m256d __A, __mmask8 __U, __m256d __B, __m256d __C)
{
- return (__m256d) __builtin_ia32_vfnmaddpd256_mask ((__v4df) __A,
- (__v4df) __B,
- (__v4df) __C,
- (__mmask8) __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddpd256 ((__v4df) __A,
+ -(__v4df) __B,
+ (__v4df) __C),
+ (__v4df) __A);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_fnmadd_ps(__m128 __A, __mmask8 __U, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfnmaddps128_mask ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddps ((__v4sf) __A,
+ -(__v4sf) __B,
+ (__v4sf) __C),
+ (__v4sf) __A);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_fnmadd_ps(__m256 __A, __mmask8 __U, __m256 __B, __m256 __C)
{
- return (__m256) __builtin_ia32_vfnmaddps256_mask ((__v8sf) __A,
- (__v8sf) __B,
- (__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddps256 ((__v8sf) __A,
+ -(__v8sf) __B,
+ (__v8sf) __C),
+ (__v8sf) __A);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_fnmsub_pd(__m128d __A, __mmask8 __U, __m128d __B, __m128d __C)
{
- return (__m128d) __builtin_ia32_vfnmsubpd128_mask ((__v2df) __A,
- (__v2df) __B,
- (__v2df) __C,
- (__mmask8) __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddpd ((__v2df) __A,
+ -(__v2df) __B,
+ -(__v2df) __C),
+ (__v2df) __A);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask3_fnmsub_pd(__m128d __A, __m128d __B, __m128d __C, __mmask8 __U)
{
- return (__m128d) __builtin_ia32_vfnmsubpd128_mask3 ((__v2df) __A,
- (__v2df) __B,
- (__v2df) __C,
- (__mmask8) __U);
+ return (__m128d) __builtin_ia32_selectpd_128((__mmask8) __U,
+ __builtin_ia32_vfmaddpd ((__v2df) __A,
+ -(__v2df) __B,
+ -(__v2df) __C),
+ (__v2df) __C);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_fnmsub_pd(__m256d __A, __mmask8 __U, __m256d __B, __m256d __C)
{
- return (__m256d) __builtin_ia32_vfnmsubpd256_mask ((__v4df) __A,
- (__v4df) __B,
- (__v4df) __C,
- (__mmask8) __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddpd256 ((__v4df) __A,
+ -(__v4df) __B,
+ -(__v4df) __C),
+ (__v4df) __A);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask3_fnmsub_pd(__m256d __A, __m256d __B, __m256d __C, __mmask8 __U)
{
- return (__m256d) __builtin_ia32_vfnmsubpd256_mask3 ((__v4df) __A,
- (__v4df) __B,
- (__v4df) __C,
- (__mmask8) __U);
+ return (__m256d) __builtin_ia32_selectpd_256((__mmask8) __U,
+ __builtin_ia32_vfmaddpd256 ((__v4df) __A,
+ -(__v4df) __B,
+ -(__v4df) __C),
+ (__v4df) __C);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_fnmsub_ps(__m128 __A, __mmask8 __U, __m128 __B, __m128 __C)
{
- return (__m128) __builtin_ia32_vfnmsubps128_mask ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddps ((__v4sf) __A,
+ -(__v4sf) __B,
+ -(__v4sf) __C),
+ (__v4sf) __A);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask3_fnmsub_ps(__m128 __A, __m128 __B, __m128 __C, __mmask8 __U)
{
- return (__m128) __builtin_ia32_vfnmsubps128_mask3 ((__v4sf) __A,
- (__v4sf) __B,
- (__v4sf) __C,
- (__mmask8) __U);
+ return (__m128) __builtin_ia32_selectps_128((__mmask8) __U,
+ __builtin_ia32_vfmaddps ((__v4sf) __A,
+ -(__v4sf) __B,
+ -(__v4sf) __C),
+ (__v4sf) __C);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_fnmsub_ps(__m256 __A, __mmask8 __U, __m256 __B, __m256 __C)
{
- return (__m256) __builtin_ia32_vfnmsubps256_mask ((__v8sf) __A,
- (__v8sf) __B,
- (__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddps256 ((__v8sf) __A,
+ -(__v8sf) __B,
+ -(__v8sf) __C),
+ (__v8sf) __A);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask3_fnmsub_ps(__m256 __A, __m256 __B, __m256 __C, __mmask8 __U)
{
- return (__m256) __builtin_ia32_vfnmsubps256_mask3 ((__v8sf) __A,
- (__v8sf) __B,
- (__v8sf) __C,
- (__mmask8) __U);
+ return (__m256) __builtin_ia32_selectps_256((__mmask8) __U,
+ __builtin_ia32_vfmaddps256 ((__v8sf) __A,
+ -(__v8sf) __B,
+ -(__v8sf) __C),
+ (__v8sf) __C);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_add_pd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_add_pd(__A, __B),
(__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_add_pd(__mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_add_pd(__A, __B),
(__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_add_pd(__m256d __W, __mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_add_pd(__A, __B),
(__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_add_pd(__mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_add_pd(__A, __B),
(__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_add_ps(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_add_ps(__A, __B),
(__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_add_ps(__mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_add_ps(__A, __B),
(__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_add_ps(__m256 __W, __mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_add_ps(__A, __B),
(__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_add_ps(__mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_add_ps(__A, __B),
(__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_blend_epi32 (__mmask8 __U, __m128i __A, __m128i __W) {
return (__m128i) __builtin_ia32_selectd_128 ((__mmask8) __U,
(__v4si) __W,
(__v4si) __A);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_blend_epi32 (__mmask8 __U, __m256i __A, __m256i __W) {
return (__m256i) __builtin_ia32_selectd_256 ((__mmask8) __U,
(__v8si) __W,
(__v8si) __A);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_blend_pd (__mmask8 __U, __m128d __A, __m128d __W) {
return (__m128d) __builtin_ia32_selectpd_128 ((__mmask8) __U,
(__v2df) __W,
(__v2df) __A);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_blend_pd (__mmask8 __U, __m256d __A, __m256d __W) {
return (__m256d) __builtin_ia32_selectpd_256 ((__mmask8) __U,
(__v4df) __W,
(__v4df) __A);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_blend_ps (__mmask8 __U, __m128 __A, __m128 __W) {
return (__m128) __builtin_ia32_selectps_128 ((__mmask8) __U,
(__v4sf) __W,
(__v4sf) __A);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_blend_ps (__mmask8 __U, __m256 __A, __m256 __W) {
return (__m256) __builtin_ia32_selectps_256 ((__mmask8) __U,
(__v8sf) __W,
(__v8sf) __A);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_blend_epi64 (__mmask8 __U, __m128i __A, __m128i __W) {
return (__m128i) __builtin_ia32_selectq_128 ((__mmask8) __U,
(__v2di) __W,
(__v2di) __A);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_blend_epi64 (__mmask8 __U, __m256i __A, __m256i __W) {
return (__m256i) __builtin_ia32_selectq_256 ((__mmask8) __U,
(__v4di) __W,
(__v4di) __A);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_compress_pd (__m128d __W, __mmask8 __U, __m128d __A) {
return (__m128d) __builtin_ia32_compressdf128_mask ((__v2df) __A,
(__v2df) __W,
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_compress_pd (__mmask8 __U, __m128d __A) {
return (__m128d) __builtin_ia32_compressdf128_mask ((__v2df) __A,
(__v2df)
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_compress_pd (__m256d __W, __mmask8 __U, __m256d __A) {
return (__m256d) __builtin_ia32_compressdf256_mask ((__v4df) __A,
(__v4df) __W,
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_compress_pd (__mmask8 __U, __m256d __A) {
return (__m256d) __builtin_ia32_compressdf256_mask ((__v4df) __A,
(__v4df)
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_compress_epi64 (__m128i __W, __mmask8 __U, __m128i __A) {
return (__m128i) __builtin_ia32_compressdi128_mask ((__v2di) __A,
(__v2di) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_compress_epi64 (__mmask8 __U, __m128i __A) {
return (__m128i) __builtin_ia32_compressdi128_mask ((__v2di) __A,
(__v2di)
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_compress_epi64 (__m256i __W, __mmask8 __U, __m256i __A) {
return (__m256i) __builtin_ia32_compressdi256_mask ((__v4di) __A,
(__v4di) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_compress_epi64 (__mmask8 __U, __m256i __A) {
return (__m256i) __builtin_ia32_compressdi256_mask ((__v4di) __A,
(__v4di)
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_compress_ps (__m128 __W, __mmask8 __U, __m128 __A) {
return (__m128) __builtin_ia32_compresssf128_mask ((__v4sf) __A,
(__v4sf) __W,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_compress_ps (__mmask8 __U, __m128 __A) {
return (__m128) __builtin_ia32_compresssf128_mask ((__v4sf) __A,
(__v4sf)
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_compress_ps (__m256 __W, __mmask8 __U, __m256 __A) {
return (__m256) __builtin_ia32_compresssf256_mask ((__v8sf) __A,
(__v8sf) __W,
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_compress_ps (__mmask8 __U, __m256 __A) {
return (__m256) __builtin_ia32_compresssf256_mask ((__v8sf) __A,
(__v8sf)
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_compress_epi32 (__m128i __W, __mmask8 __U, __m128i __A) {
return (__m128i) __builtin_ia32_compresssi128_mask ((__v4si) __A,
(__v4si) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_compress_epi32 (__mmask8 __U, __m128i __A) {
return (__m128i) __builtin_ia32_compresssi128_mask ((__v4si) __A,
(__v4si)
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_compress_epi32 (__m256i __W, __mmask8 __U, __m256i __A) {
return (__m256i) __builtin_ia32_compresssi256_mask ((__v8si) __A,
(__v8si) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_compress_epi32 (__mmask8 __U, __m256i __A) {
return (__m256i) __builtin_ia32_compresssi256_mask ((__v8si) __A,
(__v8si)
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_compressstoreu_pd (void *__P, __mmask8 __U, __m128d __A) {
__builtin_ia32_compressstoredf128_mask ((__v2df *) __P,
(__v2df) __A,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_compressstoreu_pd (void *__P, __mmask8 __U, __m256d __A) {
__builtin_ia32_compressstoredf256_mask ((__v4df *) __P,
(__v4df) __A,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_compressstoreu_epi64 (void *__P, __mmask8 __U, __m128i __A) {
__builtin_ia32_compressstoredi128_mask ((__v2di *) __P,
(__v2di) __A,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_compressstoreu_epi64 (void *__P, __mmask8 __U, __m256i __A) {
__builtin_ia32_compressstoredi256_mask ((__v4di *) __P,
(__v4di) __A,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_compressstoreu_ps (void *__P, __mmask8 __U, __m128 __A) {
__builtin_ia32_compressstoresf128_mask ((__v4sf *) __P,
(__v4sf) __A,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_compressstoreu_ps (void *__P, __mmask8 __U, __m256 __A) {
__builtin_ia32_compressstoresf256_mask ((__v8sf *) __P,
(__v8sf) __A,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_compressstoreu_epi32 (void *__P, __mmask8 __U, __m128i __A) {
__builtin_ia32_compressstoresi128_mask ((__v4si *) __P,
(__v4si) __A,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_compressstoreu_epi32 (void *__P, __mmask8 __U, __m256i __A) {
__builtin_ia32_compressstoresi256_mask ((__v8si *) __P,
(__v8si) __A,
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi32_pd (__m128d __W, __mmask8 __U, __m128i __A) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8) __U,
(__v2df)_mm_cvtepi32_pd(__A),
(__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_cvtepi32_pd (__mmask8 __U, __m128i __A) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8) __U,
(__v2df)_mm_cvtepi32_pd(__A),
(__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi32_pd (__m256d __W, __mmask8 __U, __m128i __A) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8) __U,
(__v4df)_mm256_cvtepi32_pd(__A),
(__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtepi32_pd (__mmask8 __U, __m128i __A) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8) __U,
(__v4df)_mm256_cvtepi32_pd(__A),
(__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi32_ps (__m128 __W, __mmask8 __U, __m128i __A) {
- return (__m128) __builtin_ia32_cvtdq2ps128_mask ((__v4si) __A,
- (__v4sf) __W,
- (__mmask8) __U);
-}
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm_cvtepi32_ps(__A),
+ (__v4sf)__W);
+}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_maskz_cvtepi32_ps (__mmask16 __U, __m128i __A) {
- return (__m128) __builtin_ia32_cvtdq2ps128_mask ((__v4si) __A,
- (__v4sf)
- _mm_setzero_ps (),
- (__mmask8) __U);
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_maskz_cvtepi32_ps (__mmask8 __U, __m128i __A) {
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm_cvtepi32_ps(__A),
+ (__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi32_ps (__m256 __W, __mmask8 __U, __m256i __A) {
- return (__m256) __builtin_ia32_cvtdq2ps256_mask ((__v8si) __A,
- (__v8sf) __W,
- (__mmask8) __U);
+ return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
+ (__v8sf)_mm256_cvtepi32_ps(__A),
+ (__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_maskz_cvtepi32_ps (__mmask16 __U, __m256i __A) {
- return (__m256) __builtin_ia32_cvtdq2ps256_mask ((__v8si) __A,
- (__v8sf)
- _mm256_setzero_ps (),
- (__mmask8) __U);
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_maskz_cvtepi32_ps (__mmask8 __U, __m256i __A) {
+ return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
+ (__v8sf)_mm256_cvtepi32_ps(__A),
+ (__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtpd_epi32 (__m128i __W, __mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvtpd2dq128_mask ((__v2df) __A,
(__v4si) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtpd_epi32 (__mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvtpd2dq128_mask ((__v2df) __A,
(__v4si)
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtpd_epi32 (__m128i __W, __mmask8 __U, __m256d __A) {
- return (__m128i) __builtin_ia32_cvtpd2dq256_mask ((__v4df) __A,
- (__v4si) __W,
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm256_cvtpd_epi32(__A),
+ (__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtpd_epi32 (__mmask8 __U, __m256d __A) {
- return (__m128i) __builtin_ia32_cvtpd2dq256_mask ((__v4df) __A,
- (__v4si)
- _mm_setzero_si128 (),
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm256_cvtpd_epi32(__A),
+ (__v4si)_mm_setzero_si128());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_cvtpd_ps (__m128 __W, __mmask8 __U, __m128d __A) {
return (__m128) __builtin_ia32_cvtpd2ps_mask ((__v2df) __A,
(__v4sf) __W,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_cvtpd_ps (__mmask8 __U, __m128d __A) {
return (__m128) __builtin_ia32_cvtpd2ps_mask ((__v2df) __A,
(__v4sf)
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS256
_mm256_mask_cvtpd_ps (__m128 __W, __mmask8 __U, __m256d __A) {
- return (__m128) __builtin_ia32_cvtpd2ps256_mask ((__v4df) __A,
- (__v4sf) __W,
- (__mmask8) __U);
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm256_cvtpd_ps(__A),
+ (__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtpd_ps (__mmask8 __U, __m256d __A) {
- return (__m128) __builtin_ia32_cvtpd2ps256_mask ((__v4df) __A,
- (__v4sf)
- _mm_setzero_ps (),
- (__mmask8) __U);
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm256_cvtpd_ps(__A),
+ (__v4sf)_mm_setzero_ps());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtpd_epu32 (__m128d __A) {
return (__m128i) __builtin_ia32_cvtpd2udq128_mask ((__v2df) __A,
(__v4si)
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtpd_epu32 (__m128i __W, __mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvtpd2udq128_mask ((__v2df) __A,
(__v4si) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtpd_epu32 (__mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvtpd2udq128_mask ((__v2df) __A,
(__v4si)
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtpd_epu32 (__m256d __A) {
return (__m128i) __builtin_ia32_cvtpd2udq256_mask ((__v4df) __A,
(__v4si)
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtpd_epu32 (__m128i __W, __mmask8 __U, __m256d __A) {
return (__m128i) __builtin_ia32_cvtpd2udq256_mask ((__v4df) __A,
(__v4si) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtpd_epu32 (__mmask8 __U, __m256d __A) {
return (__m128i) __builtin_ia32_cvtpd2udq256_mask ((__v4df) __A,
(__v4si)
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtps_epi32 (__m128i __W, __mmask8 __U, __m128 __A) {
- return (__m128i) __builtin_ia32_cvtps2dq128_mask ((__v4sf) __A,
- (__v4si) __W,
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_cvtps_epi32(__A),
+ (__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtps_epi32 (__mmask8 __U, __m128 __A) {
- return (__m128i) __builtin_ia32_cvtps2dq128_mask ((__v4sf) __A,
- (__v4si)
- _mm_setzero_si128 (),
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_cvtps_epi32(__A),
+ (__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtps_epi32 (__m256i __W, __mmask8 __U, __m256 __A) {
- return (__m256i) __builtin_ia32_cvtps2dq256_mask ((__v8sf) __A,
- (__v8si) __W,
- (__mmask8) __U);
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_cvtps_epi32(__A),
+ (__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtps_epi32 (__mmask8 __U, __m256 __A) {
- return (__m256i) __builtin_ia32_cvtps2dq256_mask ((__v8sf) __A,
- (__v8si)
- _mm256_setzero_si256 (),
- (__mmask8) __U);
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_cvtps_epi32(__A),
+ (__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_cvtps_pd (__m128d __W, __mmask8 __U, __m128 __A) {
- return (__m128d) __builtin_ia32_cvtps2pd128_mask ((__v4sf) __A,
- (__v2df) __W,
- (__mmask8) __U);
+ return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
+ (__v2df)_mm_cvtps_pd(__A),
+ (__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_cvtps_pd (__mmask8 __U, __m128 __A) {
- return (__m128d) __builtin_ia32_cvtps2pd128_mask ((__v4sf) __A,
- (__v2df)
- _mm_setzero_pd (),
- (__mmask8) __U);
+ return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
+ (__v2df)_mm_cvtps_pd(__A),
+ (__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_cvtps_pd (__m256d __W, __mmask8 __U, __m128 __A) {
- return (__m256d) __builtin_ia32_cvtps2pd256_mask ((__v4sf) __A,
- (__v4df) __W,
- (__mmask8) __U);
+ return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
+ (__v4df)_mm256_cvtps_pd(__A),
+ (__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtps_pd (__mmask8 __U, __m128 __A) {
- return (__m256d) __builtin_ia32_cvtps2pd256_mask ((__v4sf) __A,
- (__v4df)
- _mm256_setzero_pd (),
- (__mmask8) __U);
+ return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
+ (__v4df)_mm256_cvtps_pd(__A),
+ (__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtps_epu32 (__m128 __A) {
return (__m128i) __builtin_ia32_cvtps2udq128_mask ((__v4sf) __A,
(__v4si)
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtps_epu32 (__m128i __W, __mmask8 __U, __m128 __A) {
return (__m128i) __builtin_ia32_cvtps2udq128_mask ((__v4sf) __A,
(__v4si) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtps_epu32 (__mmask8 __U, __m128 __A) {
return (__m128i) __builtin_ia32_cvtps2udq128_mask ((__v4sf) __A,
(__v4si)
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvtps_epu32 (__m256 __A) {
return (__m256i) __builtin_ia32_cvtps2udq256_mask ((__v8sf) __A,
(__v8si)
(__mmask8) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtps_epu32 (__m256i __W, __mmask8 __U, __m256 __A) {
return (__m256i) __builtin_ia32_cvtps2udq256_mask ((__v8sf) __A,
(__v8si) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtps_epu32 (__mmask8 __U, __m256 __A) {
return (__m256i) __builtin_ia32_cvtps2udq256_mask ((__v8sf) __A,
(__v8si)
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvttpd_epi32 (__m128i __W, __mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvttpd2dq128_mask ((__v2df) __A,
(__v4si) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvttpd_epi32 (__mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvttpd2dq128_mask ((__v2df) __A,
(__v4si)
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvttpd_epi32 (__m128i __W, __mmask8 __U, __m256d __A) {
- return (__m128i) __builtin_ia32_cvttpd2dq256_mask ((__v4df) __A,
- (__v4si) __W,
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm256_cvttpd_epi32(__A),
+ (__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvttpd_epi32 (__mmask8 __U, __m256d __A) {
- return (__m128i) __builtin_ia32_cvttpd2dq256_mask ((__v4df) __A,
- (__v4si)
- _mm_setzero_si128 (),
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm256_cvttpd_epi32(__A),
+ (__v4si)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvttpd_epu32 (__m128d __A) {
return (__m128i) __builtin_ia32_cvttpd2udq128_mask ((__v2df) __A,
(__v4si)
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvttpd_epu32 (__m128i __W, __mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvttpd2udq128_mask ((__v2df) __A,
(__v4si) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvttpd_epu32 (__mmask8 __U, __m128d __A) {
return (__m128i) __builtin_ia32_cvttpd2udq128_mask ((__v2df) __A,
(__v4si)
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvttpd_epu32 (__m256d __A) {
return (__m128i) __builtin_ia32_cvttpd2udq256_mask ((__v4df) __A,
(__v4si)
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvttpd_epu32 (__m128i __W, __mmask8 __U, __m256d __A) {
return (__m128i) __builtin_ia32_cvttpd2udq256_mask ((__v4df) __A,
(__v4si) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvttpd_epu32 (__mmask8 __U, __m256d __A) {
return (__m128i) __builtin_ia32_cvttpd2udq256_mask ((__v4df) __A,
(__v4si)
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvttps_epi32 (__m128i __W, __mmask8 __U, __m128 __A) {
- return (__m128i) __builtin_ia32_cvttps2dq128_mask ((__v4sf) __A,
- (__v4si) __W,
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_cvttps_epi32(__A),
+ (__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvttps_epi32 (__mmask8 __U, __m128 __A) {
- return (__m128i) __builtin_ia32_cvttps2dq128_mask ((__v4sf) __A,
- (__v4si)
- _mm_setzero_si128 (),
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_cvttps_epi32(__A),
+ (__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_cvttps_epi32 (__m256i __W, __mmask8 __U, __m256 __A) {
- return (__m256i) __builtin_ia32_cvttps2dq256_mask ((__v8sf) __A,
- (__v8si) __W,
- (__mmask8) __U);
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_cvttps_epi32(__A),
+ (__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvttps_epi32 (__mmask8 __U, __m256 __A) {
- return (__m256i) __builtin_ia32_cvttps2dq256_mask ((__v8sf) __A,
- (__v8si)
- _mm256_setzero_si256 (),
- (__mmask8) __U);
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_cvttps_epi32(__A),
+ (__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvttps_epu32 (__m128 __A) {
return (__m128i) __builtin_ia32_cvttps2udq128_mask ((__v4sf) __A,
(__v4si)
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvttps_epu32 (__m128i __W, __mmask8 __U, __m128 __A) {
return (__m128i) __builtin_ia32_cvttps2udq128_mask ((__v4sf) __A,
(__v4si) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvttps_epu32 (__mmask8 __U, __m128 __A) {
return (__m128i) __builtin_ia32_cvttps2udq128_mask ((__v4sf) __A,
(__v4si)
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cvttps_epu32 (__m256 __A) {
return (__m256i) __builtin_ia32_cvttps2udq256_mask ((__v8sf) __A,
(__v8si)
(__mmask8) -1);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_cvttps_epu32 (__m256i __W, __mmask8 __U, __m256 __A) {
return (__m256i) __builtin_ia32_cvttps2udq256_mask ((__v8sf) __A,
(__v8si) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvttps_epu32 (__mmask8 __U, __m256 __A) {
return (__m256i) __builtin_ia32_cvttps2udq256_mask ((__v8sf) __A,
(__v8si)
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_cvtepu32_pd (__m128i __A) {
return (__m128d) __builtin_convertvector(
__builtin_shufflevector((__v4su)__A, (__v4su)__A, 0, 1), __v2df);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_cvtepu32_pd (__m128d __W, __mmask8 __U, __m128i __A) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8) __U,
(__v2df)_mm_cvtepu32_pd(__A),
(__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_cvtepu32_pd (__mmask8 __U, __m128i __A) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8) __U,
(__v2df)_mm_cvtepu32_pd(__A),
(__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_cvtepu32_pd (__m128i __A) {
return (__m256d)__builtin_convertvector((__v4su)__A, __v4df);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepu32_pd (__m256d __W, __mmask8 __U, __m128i __A) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8) __U,
(__v4df)_mm256_cvtepu32_pd(__A),
(__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtepu32_pd (__mmask8 __U, __m128i __A) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8) __U,
(__v4df)_mm256_cvtepu32_pd(__A),
(__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_cvtepu32_ps (__m128i __A) {
- return (__m128) __builtin_ia32_cvtudq2ps128_mask ((__v4si) __A,
- (__v4sf)
- _mm_setzero_ps (),
- (__mmask8) -1);
+ return (__m128)__builtin_convertvector((__v4su)__A, __v4sf);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_cvtepu32_ps (__m128 __W, __mmask8 __U, __m128i __A) {
- return (__m128) __builtin_ia32_cvtudq2ps128_mask ((__v4si) __A,
- (__v4sf) __W,
- (__mmask8) __U);
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm_cvtepu32_ps(__A),
+ (__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_cvtepu32_ps (__mmask8 __U, __m128i __A) {
- return (__m128) __builtin_ia32_cvtudq2ps128_mask ((__v4si) __A,
- (__v4sf)
- _mm_setzero_ps (),
- (__mmask8) __U);
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm_cvtepu32_ps(__A),
+ (__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_cvtepu32_ps (__m256i __A) {
- return (__m256) __builtin_ia32_cvtudq2ps256_mask ((__v8si) __A,
- (__v8sf)
- _mm256_setzero_ps (),
- (__mmask8) -1);
+ return (__m256)__builtin_convertvector((__v8su)__A, __v8sf);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepu32_ps (__m256 __W, __mmask8 __U, __m256i __A) {
- return (__m256) __builtin_ia32_cvtudq2ps256_mask ((__v8si) __A,
- (__v8sf) __W,
- (__mmask8) __U);
+ return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
+ (__v8sf)_mm256_cvtepu32_ps(__A),
+ (__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtepu32_ps (__mmask8 __U, __m256i __A) {
- return (__m256) __builtin_ia32_cvtudq2ps256_mask ((__v8si) __A,
- (__v8sf)
- _mm256_setzero_ps (),
- (__mmask8) __U);
+ return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
+ (__v8sf)_mm256_cvtepu32_ps(__A),
+ (__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_div_pd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_div_pd(__A, __B),
(__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_div_pd(__mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_div_pd(__A, __B),
(__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_div_pd(__m256d __W, __mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_div_pd(__A, __B),
(__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_div_pd(__mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_div_pd(__A, __B),
(__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_div_ps(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_div_ps(__A, __B),
(__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_div_ps(__mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_div_ps(__A, __B),
(__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_div_ps(__m256 __W, __mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_div_ps(__A, __B),
(__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_div_ps(__mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_div_ps(__A, __B),
(__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_expand_pd (__m128d __W, __mmask8 __U, __m128d __A) {
return (__m128d) __builtin_ia32_expanddf128_mask ((__v2df) __A,
(__v2df) __W,
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_expand_pd (__mmask8 __U, __m128d __A) {
return (__m128d) __builtin_ia32_expanddf128_mask ((__v2df) __A,
(__v2df)
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_expand_pd (__m256d __W, __mmask8 __U, __m256d __A) {
return (__m256d) __builtin_ia32_expanddf256_mask ((__v4df) __A,
(__v4df) __W,
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_expand_pd (__mmask8 __U, __m256d __A) {
return (__m256d) __builtin_ia32_expanddf256_mask ((__v4df) __A,
(__v4df)
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_expand_epi64 (__m128i __W, __mmask8 __U, __m128i __A) {
return (__m128i) __builtin_ia32_expanddi128_mask ((__v2di) __A,
(__v2di) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_expand_epi64 (__mmask8 __U, __m128i __A) {
return (__m128i) __builtin_ia32_expanddi128_mask ((__v2di) __A,
(__v2di)
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_expand_epi64 (__m256i __W, __mmask8 __U, __m256i __A) {
return (__m256i) __builtin_ia32_expanddi256_mask ((__v4di) __A,
(__v4di) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_expand_epi64 (__mmask8 __U, __m256i __A) {
return (__m256i) __builtin_ia32_expanddi256_mask ((__v4di) __A,
(__v4di)
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_expandloadu_pd (__m128d __W, __mmask8 __U, void const *__P) {
- return (__m128d) __builtin_ia32_expandloaddf128_mask ((__v2df *) __P,
+ return (__m128d) __builtin_ia32_expandloaddf128_mask ((const __v2df *) __P,
(__v2df) __W,
(__mmask8)
__U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_expandloadu_pd (__mmask8 __U, void const *__P) {
- return (__m128d) __builtin_ia32_expandloaddf128_mask ((__v2df *) __P,
+ return (__m128d) __builtin_ia32_expandloaddf128_mask ((const __v2df *) __P,
(__v2df)
_mm_setzero_pd (),
(__mmask8)
__U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_expandloadu_pd (__m256d __W, __mmask8 __U, void const *__P) {
- return (__m256d) __builtin_ia32_expandloaddf256_mask ((__v4df *) __P,
+ return (__m256d) __builtin_ia32_expandloaddf256_mask ((const __v4df *) __P,
(__v4df) __W,
(__mmask8)
__U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_expandloadu_pd (__mmask8 __U, void const *__P) {
- return (__m256d) __builtin_ia32_expandloaddf256_mask ((__v4df *) __P,
+ return (__m256d) __builtin_ia32_expandloaddf256_mask ((const __v4df *) __P,
(__v4df)
_mm256_setzero_pd (),
(__mmask8)
__U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_expandloadu_epi64 (__m128i __W, __mmask8 __U, void const *__P) {
- return (__m128i) __builtin_ia32_expandloaddi128_mask ((__v2di *) __P,
+ return (__m128i) __builtin_ia32_expandloaddi128_mask ((const __v2di *) __P,
(__v2di) __W,
(__mmask8)
__U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_expandloadu_epi64 (__mmask8 __U, void const *__P) {
- return (__m128i) __builtin_ia32_expandloaddi128_mask ((__v2di *) __P,
+ return (__m128i) __builtin_ia32_expandloaddi128_mask ((const __v2di *) __P,
(__v2di)
_mm_setzero_si128 (),
(__mmask8)
__U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_expandloadu_epi64 (__m256i __W, __mmask8 __U,
void const *__P) {
- return (__m256i) __builtin_ia32_expandloaddi256_mask ((__v4di *) __P,
+ return (__m256i) __builtin_ia32_expandloaddi256_mask ((const __v4di *) __P,
(__v4di) __W,
(__mmask8)
__U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_expandloadu_epi64 (__mmask8 __U, void const *__P) {
- return (__m256i) __builtin_ia32_expandloaddi256_mask ((__v4di *) __P,
+ return (__m256i) __builtin_ia32_expandloaddi256_mask ((const __v4di *) __P,
(__v4di)
_mm256_setzero_si256 (),
(__mmask8)
__U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_expandloadu_ps (__m128 __W, __mmask8 __U, void const *__P) {
- return (__m128) __builtin_ia32_expandloadsf128_mask ((__v4sf *) __P,
+ return (__m128) __builtin_ia32_expandloadsf128_mask ((const __v4sf *) __P,
(__v4sf) __W,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_expandloadu_ps (__mmask8 __U, void const *__P) {
- return (__m128) __builtin_ia32_expandloadsf128_mask ((__v4sf *) __P,
+ return (__m128) __builtin_ia32_expandloadsf128_mask ((const __v4sf *) __P,
(__v4sf)
_mm_setzero_ps (),
(__mmask8)
__U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_expandloadu_ps (__m256 __W, __mmask8 __U, void const *__P) {
- return (__m256) __builtin_ia32_expandloadsf256_mask ((__v8sf *) __P,
+ return (__m256) __builtin_ia32_expandloadsf256_mask ((const __v8sf *) __P,
(__v8sf) __W,
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_expandloadu_ps (__mmask8 __U, void const *__P) {
- return (__m256) __builtin_ia32_expandloadsf256_mask ((__v8sf *) __P,
+ return (__m256) __builtin_ia32_expandloadsf256_mask ((const __v8sf *) __P,
(__v8sf)
_mm256_setzero_ps (),
(__mmask8)
__U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_expandloadu_epi32 (__m128i __W, __mmask8 __U, void const *__P) {
- return (__m128i) __builtin_ia32_expandloadsi128_mask ((__v4si *) __P,
+ return (__m128i) __builtin_ia32_expandloadsi128_mask ((const __v4si *) __P,
(__v4si) __W,
(__mmask8)
__U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_expandloadu_epi32 (__mmask8 __U, void const *__P) {
- return (__m128i) __builtin_ia32_expandloadsi128_mask ((__v4si *) __P,
+ return (__m128i) __builtin_ia32_expandloadsi128_mask ((const __v4si *) __P,
(__v4si)
_mm_setzero_si128 (),
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_expandloadu_epi32 (__m256i __W, __mmask8 __U,
void const *__P) {
- return (__m256i) __builtin_ia32_expandloadsi256_mask ((__v8si *) __P,
+ return (__m256i) __builtin_ia32_expandloadsi256_mask ((const __v8si *) __P,
(__v8si) __W,
(__mmask8)
__U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_expandloadu_epi32 (__mmask8 __U, void const *__P) {
- return (__m256i) __builtin_ia32_expandloadsi256_mask ((__v8si *) __P,
+ return (__m256i) __builtin_ia32_expandloadsi256_mask ((const __v8si *) __P,
(__v8si)
_mm256_setzero_si256 (),
(__mmask8)
__U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_expand_ps (__m128 __W, __mmask8 __U, __m128 __A) {
return (__m128) __builtin_ia32_expandsf128_mask ((__v4sf) __A,
(__v4sf) __W,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_expand_ps (__mmask8 __U, __m128 __A) {
return (__m128) __builtin_ia32_expandsf128_mask ((__v4sf) __A,
(__v4sf)
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_expand_ps (__m256 __W, __mmask8 __U, __m256 __A) {
return (__m256) __builtin_ia32_expandsf256_mask ((__v8sf) __A,
(__v8sf) __W,
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_expand_ps (__mmask8 __U, __m256 __A) {
return (__m256) __builtin_ia32_expandsf256_mask ((__v8sf) __A,
(__v8sf)
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_expand_epi32 (__m128i __W, __mmask8 __U, __m128i __A) {
return (__m128i) __builtin_ia32_expandsi128_mask ((__v4si) __A,
(__v4si) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_expand_epi32 (__mmask8 __U, __m128i __A) {
return (__m128i) __builtin_ia32_expandsi128_mask ((__v4si) __A,
(__v4si)
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_expand_epi32 (__m256i __W, __mmask8 __U, __m256i __A) {
return (__m256i) __builtin_ia32_expandsi256_mask ((__v8si) __A,
(__v8si) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_expand_epi32 (__mmask8 __U, __m256i __A) {
return (__m256i) __builtin_ia32_expandsi256_mask ((__v8si) __A,
(__v8si)
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_getexp_pd (__m128d __A) {
return (__m128d) __builtin_ia32_getexppd128_mask ((__v2df) __A,
(__v2df)
(__mmask8) -1);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_getexp_pd (__m128d __W, __mmask8 __U, __m128d __A) {
return (__m128d) __builtin_ia32_getexppd128_mask ((__v2df) __A,
(__v2df) __W,
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_getexp_pd (__mmask8 __U, __m128d __A) {
return (__m128d) __builtin_ia32_getexppd128_mask ((__v2df) __A,
(__v2df)
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_getexp_pd (__m256d __A) {
return (__m256d) __builtin_ia32_getexppd256_mask ((__v4df) __A,
(__v4df)
(__mmask8) -1);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_getexp_pd (__m256d __W, __mmask8 __U, __m256d __A) {
return (__m256d) __builtin_ia32_getexppd256_mask ((__v4df) __A,
(__v4df) __W,
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_getexp_pd (__mmask8 __U, __m256d __A) {
return (__m256d) __builtin_ia32_getexppd256_mask ((__v4df) __A,
(__v4df)
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_getexp_ps (__m128 __A) {
return (__m128) __builtin_ia32_getexpps128_mask ((__v4sf) __A,
(__v4sf)
(__mmask8) -1);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_getexp_ps (__m128 __W, __mmask8 __U, __m128 __A) {
return (__m128) __builtin_ia32_getexpps128_mask ((__v4sf) __A,
(__v4sf) __W,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_getexp_ps (__mmask8 __U, __m128 __A) {
return (__m128) __builtin_ia32_getexpps128_mask ((__v4sf) __A,
(__v4sf)
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_getexp_ps (__m256 __A) {
return (__m256) __builtin_ia32_getexpps256_mask ((__v8sf) __A,
(__v8sf)
(__mmask8) -1);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_getexp_ps (__m256 __W, __mmask8 __U, __m256 __A) {
return (__m256) __builtin_ia32_getexpps256_mask ((__v8sf) __A,
(__v8sf) __W,
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_getexp_ps (__mmask8 __U, __m256 __A) {
return (__m256) __builtin_ia32_getexpps256_mask ((__v8sf) __A,
(__v8sf)
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_max_pd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_max_pd(__A, __B),
(__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_max_pd(__mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_max_pd(__A, __B),
(__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_max_pd(__m256d __W, __mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_max_pd(__A, __B),
(__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_max_pd(__mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_max_pd(__A, __B),
(__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_max_ps(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_max_ps(__A, __B),
(__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_max_ps(__mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_max_ps(__A, __B),
(__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_max_ps(__m256 __W, __mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_max_ps(__A, __B),
(__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_max_ps(__mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_max_ps(__A, __B),
(__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_min_pd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_min_pd(__A, __B),
(__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_min_pd(__mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_min_pd(__A, __B),
(__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_min_pd(__m256d __W, __mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_min_pd(__A, __B),
(__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_min_pd(__mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_min_pd(__A, __B),
(__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_min_ps(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_min_ps(__A, __B),
(__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_min_ps(__mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_min_ps(__A, __B),
(__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_min_ps(__m256 __W, __mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_min_ps(__A, __B),
(__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_min_ps(__mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_min_ps(__A, __B),
(__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_mul_pd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_mul_pd(__A, __B),
(__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_mul_pd(__mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_mul_pd(__A, __B),
(__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_mul_pd(__m256d __W, __mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_mul_pd(__A, __B),
(__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_mul_pd(__mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_mul_pd(__A, __B),
(__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_mul_ps(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_mul_ps(__A, __B),
(__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_mul_ps(__mmask8 __U, __m128 __A, __m128 __B) {
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_mul_ps(__A, __B),
(__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_mul_ps(__m256 __W, __mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_mul_ps(__A, __B),
(__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_mul_ps(__mmask8 __U, __m256 __A, __m256 __B) {
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_mul_ps(__A, __B),
(__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_abs_epi32(__m128i __W, __mmask8 __U, __m128i __A) {
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_abs_epi32(__A),
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_abs_epi32(__mmask8 __U, __m128i __A) {
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_abs_epi32(__A),
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_abs_epi32(__m256i __W, __mmask8 __U, __m256i __A) {
- return (__m256i)__builtin_ia32_selectd_256((__mmask16)__U,
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_abs_epi32(__A),
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_abs_epi32(__mmask8 __U, __m256i __A) {
- return (__m256i)__builtin_ia32_selectd_256((__mmask16)__U,
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_abs_epi32(__A),
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_abs_epi64 (__m128i __A) {
- return (__m128i) __builtin_ia32_pabsq128_mask ((__v2di) __A,
- (__v2di)
- _mm_setzero_si128 (),
- (__mmask8) -1);
+ return (__m128i)__builtin_ia32_pabsq128((__v2di)__A);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_abs_epi64 (__m128i __W, __mmask8 __U, __m128i __A) {
- return (__m128i) __builtin_ia32_pabsq128_mask ((__v2di) __A,
- (__v2di) __W,
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_abs_epi64(__A),
+ (__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_abs_epi64 (__mmask8 __U, __m128i __A) {
- return (__m128i) __builtin_ia32_pabsq128_mask ((__v2di) __A,
- (__v2di)
- _mm_setzero_si128 (),
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_abs_epi64(__A),
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_abs_epi64 (__m256i __A) {
- return (__m256i) __builtin_ia32_pabsq256_mask ((__v4di) __A,
- (__v4di)
- _mm256_setzero_si256 (),
- (__mmask8) -1);
+ return (__m256i)__builtin_ia32_pabsq256 ((__v4di)__A);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_abs_epi64 (__m256i __W, __mmask8 __U, __m256i __A) {
- return (__m256i) __builtin_ia32_pabsq256_mask ((__v4di) __A,
- (__v4di) __W,
- (__mmask8) __U);
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_abs_epi64(__A),
+ (__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_abs_epi64 (__mmask8 __U, __m256i __A) {
- return (__m256i) __builtin_ia32_pabsq256_mask ((__v4di) __A,
- (__v4di)
- _mm256_setzero_si256 (),
- (__mmask8) __U);
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_abs_epi64(__A),
+ (__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_max_epi32(__mmask8 __M, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__M,
(__v4si)_mm_max_epi32(__A, __B),
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_max_epi32(__m128i __W, __mmask8 __M, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__M,
(__v4si)_mm_max_epi32(__A, __B),
(__v4si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_max_epi32(__mmask8 __M, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
(__v8si)_mm256_max_epi32(__A, __B),
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_max_epi32(__m256i __W, __mmask8 __M, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
(__v8si)_mm256_max_epi32(__A, __B),
(__v8si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_max_epi64 (__mmask8 __M, __m128i __A, __m128i __B) {
- return (__m128i) __builtin_ia32_pmaxsq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di)
- _mm_setzero_si128 (),
- __M);
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_max_epi64 (__m128i __A, __m128i __B) {
+ return (__m128i)__builtin_ia32_pmaxsq128((__v2di)__A, (__v2di)__B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_max_epi64 (__m128i __W, __mmask8 __M, __m128i __A,
- __m128i __B) {
- return (__m128i) __builtin_ia32_pmaxsq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di) __W, __M);
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_max_epi64 (__mmask8 __M, __m128i __A, __m128i __B) {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__M,
+ (__v2di)_mm_max_epi64(__A, __B),
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_max_epi64 (__m128i __A, __m128i __B) {
- return (__m128i) __builtin_ia32_pmaxsq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di)
- _mm_setzero_si128 (),
- (__mmask8) -1);
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_max_epi64 (__m128i __W, __mmask8 __M, __m128i __A, __m128i __B) {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__M,
+ (__v2di)_mm_max_epi64(__A, __B),
+ (__v2di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_max_epi64 (__mmask8 __M, __m256i __A, __m256i __B) {
- return (__m256i) __builtin_ia32_pmaxsq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di)
- _mm256_setzero_si256 (),
- __M);
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_max_epi64 (__m256i __A, __m256i __B) {
+ return (__m256i)__builtin_ia32_pmaxsq256((__v4di)__A, (__v4di)__B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_max_epi64 (__m256i __W, __mmask8 __M, __m256i __A,
- __m256i __B) {
- return (__m256i) __builtin_ia32_pmaxsq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di) __W, __M);
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_max_epi64 (__mmask8 __M, __m256i __A, __m256i __B) {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
+ (__v4di)_mm256_max_epi64(__A, __B),
+ (__v4di)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_max_epi64 (__m256i __A, __m256i __B) {
- return (__m256i) __builtin_ia32_pmaxsq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di)
- _mm256_setzero_si256 (),
- (__mmask8) -1);
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_max_epi64 (__m256i __W, __mmask8 __M, __m256i __A, __m256i __B) {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
+ (__v4di)_mm256_max_epi64(__A, __B),
+ (__v4di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_max_epu32(__mmask8 __M, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__M,
(__v4si)_mm_max_epu32(__A, __B),
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_max_epu32(__m128i __W, __mmask8 __M, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__M,
(__v4si)_mm_max_epu32(__A, __B),
(__v4si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_max_epu32(__mmask8 __M, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
(__v8si)_mm256_max_epu32(__A, __B),
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_max_epu32(__m256i __W, __mmask8 __M, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
(__v8si)_mm256_max_epu32(__A, __B),
(__v8si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_max_epu64 (__mmask8 __M, __m128i __A, __m128i __B) {
- return (__m128i) __builtin_ia32_pmaxuq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di)
- _mm_setzero_si128 (),
- __M);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_max_epu64 (__m128i __A, __m128i __B) {
- return (__m128i) __builtin_ia32_pmaxuq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di)
- _mm_setzero_si128 (),
- (__mmask8) -1);
+ return (__m128i)__builtin_ia32_pmaxuq128((__v2di)__A, (__v2di)__B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_max_epu64 (__m128i __W, __mmask8 __M, __m128i __A,
- __m128i __B) {
- return (__m128i) __builtin_ia32_pmaxuq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di) __W, __M);
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_max_epu64 (__mmask8 __M, __m128i __A, __m128i __B) {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__M,
+ (__v2di)_mm_max_epu64(__A, __B),
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_max_epu64 (__mmask8 __M, __m256i __A, __m256i __B) {
- return (__m256i) __builtin_ia32_pmaxuq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di)
- _mm256_setzero_si256 (),
- __M);
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_max_epu64 (__m128i __W, __mmask8 __M, __m128i __A, __m128i __B) {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__M,
+ (__v2di)_mm_max_epu64(__A, __B),
+ (__v2di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_max_epu64 (__m256i __A, __m256i __B) {
- return (__m256i) __builtin_ia32_pmaxuq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di)
- _mm256_setzero_si256 (),
- (__mmask8) -1);
+ return (__m256i)__builtin_ia32_pmaxuq256((__v4di)__A, (__v4di)__B);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_max_epu64 (__mmask8 __M, __m256i __A, __m256i __B) {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
+ (__v4di)_mm256_max_epu64(__A, __B),
+ (__v4di)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_max_epu64 (__m256i __W, __mmask8 __M, __m256i __A,
- __m256i __B) {
- return (__m256i) __builtin_ia32_pmaxuq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di) __W, __M);
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_max_epu64 (__m256i __W, __mmask8 __M, __m256i __A, __m256i __B) {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
+ (__v4di)_mm256_max_epu64(__A, __B),
+ (__v4di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_min_epi32(__mmask8 __M, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__M,
(__v4si)_mm_min_epi32(__A, __B),
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_min_epi32(__m128i __W, __mmask8 __M, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__M,
(__v4si)_mm_min_epi32(__A, __B),
(__v4si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_min_epi32(__mmask8 __M, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
(__v8si)_mm256_min_epi32(__A, __B),
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_min_epi32(__m256i __W, __mmask8 __M, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
(__v8si)_mm256_min_epi32(__A, __B),
(__v8si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_min_epi64 (__m128i __A, __m128i __B) {
- return (__m128i) __builtin_ia32_pminsq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di)
- _mm_setzero_si128 (),
- (__mmask8) -1);
+ return (__m128i)__builtin_ia32_pminsq128((__v2di)__A, (__v2di)__B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_min_epi64 (__m128i __W, __mmask8 __M, __m128i __A,
- __m128i __B) {
- return (__m128i) __builtin_ia32_pminsq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di) __W, __M);
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_min_epi64 (__m128i __W, __mmask8 __M, __m128i __A, __m128i __B) {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__M,
+ (__v2di)_mm_min_epi64(__A, __B),
+ (__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_min_epi64 (__mmask8 __M, __m128i __A, __m128i __B) {
- return (__m128i) __builtin_ia32_pminsq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di)
- _mm_setzero_si128 (),
- __M);
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__M,
+ (__v2di)_mm_min_epi64(__A, __B),
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_min_epi64 (__m256i __A, __m256i __B) {
- return (__m256i) __builtin_ia32_pminsq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di)
- _mm256_setzero_si256 (),
- (__mmask8) -1);
+ return (__m256i)__builtin_ia32_pminsq256((__v4di)__A, (__v4di)__B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_min_epi64 (__m256i __W, __mmask8 __M, __m256i __A,
- __m256i __B) {
- return (__m256i) __builtin_ia32_pminsq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di) __W, __M);
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_min_epi64 (__m256i __W, __mmask8 __M, __m256i __A, __m256i __B) {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
+ (__v4di)_mm256_min_epi64(__A, __B),
+ (__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_min_epi64 (__mmask8 __M, __m256i __A, __m256i __B) {
- return (__m256i) __builtin_ia32_pminsq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di)
- _mm256_setzero_si256 (),
- __M);
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
+ (__v4di)_mm256_min_epi64(__A, __B),
+ (__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_min_epu32(__mmask8 __M, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__M,
(__v4si)_mm_min_epu32(__A, __B),
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_min_epu32(__m128i __W, __mmask8 __M, __m128i __A, __m128i __B) {
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__M,
(__v4si)_mm_min_epu32(__A, __B),
(__v4si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_min_epu32(__mmask8 __M, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
(__v8si)_mm256_min_epu32(__A, __B),
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_min_epu32(__m256i __W, __mmask8 __M, __m256i __A, __m256i __B) {
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
(__v8si)_mm256_min_epu32(__A, __B),
(__v8si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_min_epu64 (__m128i __A, __m128i __B) {
- return (__m128i) __builtin_ia32_pminuq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di)
- _mm_setzero_si128 (),
- (__mmask8) -1);
+ return (__m128i)__builtin_ia32_pminuq128((__v2di)__A, (__v2di)__B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_min_epu64 (__m128i __W, __mmask8 __M, __m128i __A,
- __m128i __B) {
- return (__m128i) __builtin_ia32_pminuq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di) __W, __M);
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_min_epu64 (__m128i __W, __mmask8 __M, __m128i __A, __m128i __B) {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__M,
+ (__v2di)_mm_min_epu64(__A, __B),
+ (__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_min_epu64 (__mmask8 __M, __m128i __A, __m128i __B) {
- return (__m128i) __builtin_ia32_pminuq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di)
- _mm_setzero_si128 (),
- __M);
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__M,
+ (__v2di)_mm_min_epu64(__A, __B),
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_min_epu64 (__m256i __A, __m256i __B) {
- return (__m256i) __builtin_ia32_pminuq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di)
- _mm256_setzero_si256 (),
- (__mmask8) -1);
+ return (__m256i)__builtin_ia32_pminuq256((__v4di)__A, (__v4di)__B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_min_epu64 (__m256i __W, __mmask8 __M, __m256i __A,
- __m256i __B) {
- return (__m256i) __builtin_ia32_pminuq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di) __W, __M);
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_min_epu64 (__m256i __W, __mmask8 __M, __m256i __A, __m256i __B) {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
+ (__v4di)_mm256_min_epu64(__A, __B),
+ (__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_min_epu64 (__mmask8 __M, __m256i __A, __m256i __B) {
- return (__m256i) __builtin_ia32_pminuq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di)
- _mm256_setzero_si256 (),
- __M);
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
+ (__v4di)_mm256_min_epu64(__A, __B),
+ (__v4di)_mm256_setzero_si256());
}
-#define _mm_roundscale_pd(A, imm) __extension__ ({ \
+#define _mm_roundscale_pd(A, imm) \
(__m128d)__builtin_ia32_rndscalepd_128_mask((__v2df)(__m128d)(A), \
(int)(imm), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_roundscale_pd(W, U, A, imm) __extension__ ({ \
+#define _mm_mask_roundscale_pd(W, U, A, imm) \
(__m128d)__builtin_ia32_rndscalepd_128_mask((__v2df)(__m128d)(A), \
(int)(imm), \
(__v2df)(__m128d)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_maskz_roundscale_pd(U, A, imm) __extension__ ({ \
+#define _mm_maskz_roundscale_pd(U, A, imm) \
(__m128d)__builtin_ia32_rndscalepd_128_mask((__v2df)(__m128d)(A), \
(int)(imm), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_roundscale_pd(A, imm) __extension__ ({ \
+#define _mm256_roundscale_pd(A, imm) \
(__m256d)__builtin_ia32_rndscalepd_256_mask((__v4df)(__m256d)(A), \
(int)(imm), \
(__v4df)_mm256_setzero_pd(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_roundscale_pd(W, U, A, imm) __extension__ ({ \
+#define _mm256_mask_roundscale_pd(W, U, A, imm) \
(__m256d)__builtin_ia32_rndscalepd_256_mask((__v4df)(__m256d)(A), \
(int)(imm), \
(__v4df)(__m256d)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_maskz_roundscale_pd(U, A, imm) __extension__ ({ \
+#define _mm256_maskz_roundscale_pd(U, A, imm) \
(__m256d)__builtin_ia32_rndscalepd_256_mask((__v4df)(__m256d)(A), \
(int)(imm), \
(__v4df)_mm256_setzero_pd(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_roundscale_ps(A, imm) __extension__ ({ \
+#define _mm_roundscale_ps(A, imm) \
(__m128)__builtin_ia32_rndscaleps_128_mask((__v4sf)(__m128)(A), (int)(imm), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_roundscale_ps(W, U, A, imm) __extension__ ({ \
+#define _mm_mask_roundscale_ps(W, U, A, imm) \
(__m128)__builtin_ia32_rndscaleps_128_mask((__v4sf)(__m128)(A), (int)(imm), \
(__v4sf)(__m128)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_maskz_roundscale_ps(U, A, imm) __extension__ ({ \
+#define _mm_maskz_roundscale_ps(U, A, imm) \
(__m128)__builtin_ia32_rndscaleps_128_mask((__v4sf)(__m128)(A), (int)(imm), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_roundscale_ps(A, imm) __extension__ ({ \
+#define _mm256_roundscale_ps(A, imm) \
(__m256)__builtin_ia32_rndscaleps_256_mask((__v8sf)(__m256)(A), (int)(imm), \
(__v8sf)_mm256_setzero_ps(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_roundscale_ps(W, U, A, imm) __extension__ ({ \
+#define _mm256_mask_roundscale_ps(W, U, A, imm) \
(__m256)__builtin_ia32_rndscaleps_256_mask((__v8sf)(__m256)(A), (int)(imm), \
(__v8sf)(__m256)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_maskz_roundscale_ps(U, A, imm) __extension__ ({ \
+#define _mm256_maskz_roundscale_ps(U, A, imm) \
(__m256)__builtin_ia32_rndscaleps_256_mask((__v8sf)(__m256)(A), (int)(imm), \
(__v8sf)_mm256_setzero_ps(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_scalef_pd (__m128d __A, __m128d __B) {
return (__m128d) __builtin_ia32_scalefpd128_mask ((__v2df) __A,
(__v2df) __B,
(__mmask8) -1);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_scalef_pd (__m128d __W, __mmask8 __U, __m128d __A,
__m128d __B) {
return (__m128d) __builtin_ia32_scalefpd128_mask ((__v2df) __A,
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_scalef_pd (__mmask8 __U, __m128d __A, __m128d __B) {
return (__m128d) __builtin_ia32_scalefpd128_mask ((__v2df) __A,
(__v2df) __B,
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_scalef_pd (__m256d __A, __m256d __B) {
return (__m256d) __builtin_ia32_scalefpd256_mask ((__v4df) __A,
(__v4df) __B,
(__mmask8) -1);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_scalef_pd (__m256d __W, __mmask8 __U, __m256d __A,
__m256d __B) {
return (__m256d) __builtin_ia32_scalefpd256_mask ((__v4df) __A,
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_scalef_pd (__mmask8 __U, __m256d __A, __m256d __B) {
return (__m256d) __builtin_ia32_scalefpd256_mask ((__v4df) __A,
(__v4df) __B,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_scalef_ps (__m128 __A, __m128 __B) {
return (__m128) __builtin_ia32_scalefps128_mask ((__v4sf) __A,
(__v4sf) __B,
(__mmask8) -1);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_scalef_ps (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) {
return (__m128) __builtin_ia32_scalefps128_mask ((__v4sf) __A,
(__v4sf) __B,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_scalef_ps (__mmask8 __U, __m128 __A, __m128 __B) {
return (__m128) __builtin_ia32_scalefps128_mask ((__v4sf) __A,
(__v4sf) __B,
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_scalef_ps (__m256 __A, __m256 __B) {
return (__m256) __builtin_ia32_scalefps256_mask ((__v8sf) __A,
(__v8sf) __B,
(__mmask8) -1);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_scalef_ps (__m256 __W, __mmask8 __U, __m256 __A,
__m256 __B) {
return (__m256) __builtin_ia32_scalefps256_mask ((__v8sf) __A,
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_scalef_ps (__mmask8 __U, __m256 __A, __m256 __B) {
return (__m256) __builtin_ia32_scalefps256_mask ((__v8sf) __A,
(__v8sf) __B,
(__mmask8) __U);
}
-#define _mm_i64scatter_pd(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv2df((double *)(addr), (__mmask8)-1, \
- (__v2di)(__m128i)(index), \
- (__v2df)(__m128d)(v1), (int)(scale)); })
-
-#define _mm_mask_i64scatter_pd(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv2df((double *)(addr), (__mmask8)(mask), \
+#define _mm_i64scatter_pd(addr, index, v1, scale) \
+ __builtin_ia32_scatterdiv2df((void *)(addr), (__mmask8)-1, \
(__v2di)(__m128i)(index), \
- (__v2df)(__m128d)(v1), (int)(scale)); })
-
-#define _mm_i64scatter_epi64(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv2di((long long *)(addr), (__mmask8)-1, \
- (__v2di)(__m128i)(index), \
- (__v2di)(__m128i)(v1), (int)(scale)); })
-
-#define _mm_mask_i64scatter_epi64(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv2di((long long *)(addr), (__mmask8)(mask), \
- (__v2di)(__m128i)(index), \
- (__v2di)(__m128i)(v1), (int)(scale)); })
-
-#define _mm256_i64scatter_pd(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv4df((double *)(addr), (__mmask8)-1, \
- (__v4di)(__m256i)(index), \
- (__v4df)(__m256d)(v1), (int)(scale)); })
-
-#define _mm256_mask_i64scatter_pd(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv4df((double *)(addr), (__mmask8)(mask), \
- (__v4di)(__m256i)(index), \
- (__v4df)(__m256d)(v1), (int)(scale)); })
-
-#define _mm256_i64scatter_epi64(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv4di((long long *)(addr), (__mmask8)-1, \
- (__v4di)(__m256i)(index), \
- (__v4di)(__m256i)(v1), (int)(scale)); })
-
-#define _mm256_mask_i64scatter_epi64(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv4di((long long *)(addr), (__mmask8)(mask), \
- (__v4di)(__m256i)(index), \
- (__v4di)(__m256i)(v1), (int)(scale)); })
-
-#define _mm_i64scatter_ps(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv4sf((float *)(addr), (__mmask8)-1, \
- (__v2di)(__m128i)(index), (__v4sf)(__m128)(v1), \
- (int)(scale)); })
-
-#define _mm_mask_i64scatter_ps(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv4sf((float *)(addr), (__mmask8)(mask), \
- (__v2di)(__m128i)(index), (__v4sf)(__m128)(v1), \
- (int)(scale)); })
-
-#define _mm_i64scatter_epi32(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv4si((int *)(addr), (__mmask8)-1, \
- (__v2di)(__m128i)(index), \
- (__v4si)(__m128i)(v1), (int)(scale)); })
-
-#define _mm_mask_i64scatter_epi32(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv4si((int *)(addr), (__mmask8)(mask), \
- (__v2di)(__m128i)(index), \
- (__v4si)(__m128i)(v1), (int)(scale)); })
-
-#define _mm256_i64scatter_ps(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv8sf((float *)(addr), (__mmask8)-1, \
- (__v4di)(__m256i)(index), (__v4sf)(__m128)(v1), \
- (int)(scale)); })
-
-#define _mm256_mask_i64scatter_ps(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv8sf((float *)(addr), (__mmask8)(mask), \
- (__v4di)(__m256i)(index), (__v4sf)(__m128)(v1), \
- (int)(scale)); })
-
-#define _mm256_i64scatter_epi32(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv8si((int *)(addr), (__mmask8)-1, \
- (__v4di)(__m256i)(index), \
- (__v4si)(__m128i)(v1), (int)(scale)); })
-
-#define _mm256_mask_i64scatter_epi32(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scatterdiv8si((int *)(addr), (__mmask8)(mask), \
- (__v4di)(__m256i)(index), \
- (__v4si)(__m128i)(v1), (int)(scale)); })
-
-#define _mm_i32scatter_pd(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv2df((double *)(addr), (__mmask8)-1, \
- (__v4si)(__m128i)(index), \
- (__v2df)(__m128d)(v1), (int)(scale)); })
-
-#define _mm_mask_i32scatter_pd(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv2df((double *)(addr), (__mmask8)(mask), \
- (__v4si)(__m128i)(index), \
- (__v2df)(__m128d)(v1), (int)(scale)); })
-
-#define _mm_i32scatter_epi64(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv2di((long long *)(addr), (__mmask8)-1, \
- (__v4si)(__m128i)(index), \
- (__v2di)(__m128i)(v1), (int)(scale)); })
-
-#define _mm_mask_i32scatter_epi64(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv2di((long long *)(addr), (__mmask8)(mask), \
- (__v4si)(__m128i)(index), \
- (__v2di)(__m128i)(v1), (int)(scale)); })
-
-#define _mm256_i32scatter_pd(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv4df((double *)(addr), (__mmask8)-1, \
- (__v4si)(__m128i)(index), \
- (__v4df)(__m256d)(v1), (int)(scale)); })
-
-#define _mm256_mask_i32scatter_pd(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv4df((double *)(addr), (__mmask8)(mask), \
- (__v4si)(__m128i)(index), \
- (__v4df)(__m256d)(v1), (int)(scale)); })
-
-#define _mm256_i32scatter_epi64(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv4di((long long *)(addr), (__mmask8)-1, \
- (__v4si)(__m128i)(index), \
- (__v4di)(__m256i)(v1), (int)(scale)); })
-
-#define _mm256_mask_i32scatter_epi64(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv4di((long long *)(addr), (__mmask8)(mask), \
- (__v4si)(__m128i)(index), \
- (__v4di)(__m256i)(v1), (int)(scale)); })
-
-#define _mm_i32scatter_ps(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv4sf((float *)(addr), (__mmask8)-1, \
- (__v4si)(__m128i)(index), (__v4sf)(__m128)(v1), \
- (int)(scale)); })
-
-#define _mm_mask_i32scatter_ps(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv4sf((float *)(addr), (__mmask8)(mask), \
- (__v4si)(__m128i)(index), (__v4sf)(__m128)(v1), \
- (int)(scale)); })
-
-#define _mm_i32scatter_epi32(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv4si((int *)(addr), (__mmask8)-1, \
- (__v4si)(__m128i)(index), \
- (__v4si)(__m128i)(v1), (int)(scale)); })
-
-#define _mm_mask_i32scatter_epi32(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv4si((int *)(addr), (__mmask8)(mask), \
- (__v4si)(__m128i)(index), \
- (__v4si)(__m128i)(v1), (int)(scale)); })
-
-#define _mm256_i32scatter_ps(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv8sf((float *)(addr), (__mmask8)-1, \
- (__v8si)(__m256i)(index), (__v8sf)(__m256)(v1), \
- (int)(scale)); })
-
-#define _mm256_mask_i32scatter_ps(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv8sf((float *)(addr), (__mmask8)(mask), \
- (__v8si)(__m256i)(index), (__v8sf)(__m256)(v1), \
- (int)(scale)); })
-
-#define _mm256_i32scatter_epi32(addr, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv8si((int *)(addr), (__mmask8)-1, \
- (__v8si)(__m256i)(index), \
- (__v8si)(__m256i)(v1), (int)(scale)); })
-
-#define _mm256_mask_i32scatter_epi32(addr, mask, index, v1, scale) __extension__ ({ \
- __builtin_ia32_scattersiv8si((int *)(addr), (__mmask8)(mask), \
- (__v8si)(__m256i)(index), \
- (__v8si)(__m256i)(v1), (int)(scale)); })
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_mask_sqrt_pd(__m128d __W, __mmask8 __U, __m128d __A) {
- return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
- (__v2df)_mm_sqrt_pd(__A),
- (__v2df)__W);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_maskz_sqrt_pd(__mmask8 __U, __m128d __A) {
- return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
- (__v2df)_mm_sqrt_pd(__A),
- (__v2df)_mm_setzero_pd());
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_mask_sqrt_pd(__m256d __W, __mmask8 __U, __m256d __A) {
- return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
- (__v4df)_mm256_sqrt_pd(__A),
- (__v4df)__W);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_maskz_sqrt_pd(__mmask8 __U, __m256d __A) {
- return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
- (__v4df)_mm256_sqrt_pd(__A),
- (__v4df)_mm256_setzero_pd());
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_mask_sqrt_ps(__m128 __W, __mmask8 __U, __m128 __A) {
- return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
- (__v4sf)_mm_sqrt_ps(__A),
- (__v4sf)__W);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_maskz_sqrt_ps(__mmask8 __U, __m128 __A) {
- return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
- (__v4sf)_mm_sqrt_ps(__A),
- (__v4sf)_mm_setzero_pd());
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_mask_sqrt_ps(__m256 __W, __mmask8 __U, __m256 __A) {
- return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
- (__v8sf)_mm256_sqrt_ps(__A),
- (__v8sf)__W);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_maskz_sqrt_ps(__mmask8 __U, __m256 __A) {
- return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
- (__v8sf)_mm256_sqrt_ps(__A),
- (__v8sf)_mm256_setzero_ps());
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_mask_sub_pd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B) {
- return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
- (__v2df)_mm_sub_pd(__A, __B),
- (__v2df)__W);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_maskz_sub_pd(__mmask8 __U, __m128d __A, __m128d __B) {
- return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
- (__v2df)_mm_sub_pd(__A, __B),
- (__v2df)_mm_setzero_pd());
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_mask_sub_pd(__m256d __W, __mmask8 __U, __m256d __A, __m256d __B) {
- return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
- (__v4df)_mm256_sub_pd(__A, __B),
- (__v4df)__W);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_maskz_sub_pd(__mmask8 __U, __m256d __A, __m256d __B) {
- return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
- (__v4df)_mm256_sub_pd(__A, __B),
- (__v4df)_mm256_setzero_pd());
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_mask_sub_ps(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) {
- return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
- (__v4sf)_mm_sub_ps(__A, __B),
- (__v4sf)__W);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_maskz_sub_ps(__mmask8 __U, __m128 __A, __m128 __B) {
- return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
- (__v4sf)_mm_sub_ps(__A, __B),
- (__v4sf)_mm_setzero_ps());
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_mask_sub_ps(__m256 __W, __mmask8 __U, __m256 __A, __m256 __B) {
- return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
- (__v8sf)_mm256_sub_ps(__A, __B),
- (__v8sf)__W);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_maskz_sub_ps(__mmask8 __U, __m256 __A, __m256 __B) {
- return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
- (__v8sf)_mm256_sub_ps(__A, __B),
- (__v8sf)_mm256_setzero_ps());
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask2_permutex2var_epi32 (__m128i __A, __m128i __I, __mmask8 __U,
- __m128i __B) {
- return (__m128i) __builtin_ia32_vpermi2vard128_mask ((__v4si) __A,
- (__v4si) __I
- /* idx */ ,
- (__v4si) __B,
- (__mmask8) __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask2_permutex2var_epi32 (__m256i __A, __m256i __I,
- __mmask8 __U, __m256i __B) {
- return (__m256i) __builtin_ia32_vpermi2vard256_mask ((__v8si) __A,
- (__v8si) __I
- /* idx */ ,
- (__v8si) __B,
- (__mmask8) __U);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_mask2_permutex2var_pd (__m128d __A, __m128i __I, __mmask8 __U,
- __m128d __B) {
- return (__m128d) __builtin_ia32_vpermi2varpd128_mask ((__v2df) __A,
- (__v2di) __I
- /* idx */ ,
- (__v2df) __B,
- (__mmask8)
- __U);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_mask2_permutex2var_pd (__m256d __A, __m256i __I, __mmask8 __U,
- __m256d __B) {
- return (__m256d) __builtin_ia32_vpermi2varpd256_mask ((__v4df) __A,
- (__v4di) __I
- /* idx */ ,
- (__v4df) __B,
- (__mmask8)
- __U);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_mask2_permutex2var_ps (__m128 __A, __m128i __I, __mmask8 __U,
- __m128 __B) {
- return (__m128) __builtin_ia32_vpermi2varps128_mask ((__v4sf) __A,
- (__v4si) __I
- /* idx */ ,
- (__v4sf) __B,
- (__mmask8) __U);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_mask2_permutex2var_ps (__m256 __A, __m256i __I, __mmask8 __U,
- __m256 __B) {
- return (__m256) __builtin_ia32_vpermi2varps256_mask ((__v8sf) __A,
- (__v8si) __I
- /* idx */ ,
- (__v8sf) __B,
- (__mmask8) __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask2_permutex2var_epi64 (__m128i __A, __m128i __I, __mmask8 __U,
- __m128i __B) {
- return (__m128i) __builtin_ia32_vpermi2varq128_mask ((__v2di) __A,
- (__v2di) __I
- /* idx */ ,
- (__v2di) __B,
- (__mmask8) __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask2_permutex2var_epi64 (__m256i __A, __m256i __I,
- __mmask8 __U, __m256i __B) {
- return (__m256i) __builtin_ia32_vpermi2varq256_mask ((__v4di) __A,
- (__v4di) __I
- /* idx */ ,
- (__v4di) __B,
- (__mmask8) __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_permutex2var_epi32 (__m128i __A, __m128i __I, __m128i __B) {
- return (__m128i) __builtin_ia32_vpermt2vard128_mask ((__v4si) __I
- /* idx */ ,
- (__v4si) __A,
- (__v4si) __B,
- (__mmask8) -1);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_permutex2var_epi32 (__m128i __A, __mmask8 __U, __m128i __I,
- __m128i __B) {
- return (__m128i) __builtin_ia32_vpermt2vard128_mask ((__v4si) __I
- /* idx */ ,
- (__v4si) __A,
- (__v4si) __B,
- (__mmask8) __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_permutex2var_epi32 (__mmask8 __U, __m128i __A, __m128i __I,
- __m128i __B) {
- return (__m128i) __builtin_ia32_vpermt2vard128_maskz ((__v4si) __I
- /* idx */ ,
- (__v4si) __A,
- (__v4si) __B,
- (__mmask8)
- __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_permutex2var_epi32 (__m256i __A, __m256i __I, __m256i __B) {
- return (__m256i) __builtin_ia32_vpermt2vard256_mask ((__v8si) __I
- /* idx */ ,
- (__v8si) __A,
- (__v8si) __B,
- (__mmask8) -1);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_permutex2var_epi32 (__m256i __A, __mmask8 __U, __m256i __I,
- __m256i __B) {
- return (__m256i) __builtin_ia32_vpermt2vard256_mask ((__v8si) __I
- /* idx */ ,
- (__v8si) __A,
- (__v8si) __B,
- (__mmask8) __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_permutex2var_epi32 (__mmask8 __U, __m256i __A,
- __m256i __I, __m256i __B) {
- return (__m256i) __builtin_ia32_vpermt2vard256_maskz ((__v8si) __I
- /* idx */ ,
- (__v8si) __A,
- (__v8si) __B,
- (__mmask8)
- __U);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_permutex2var_pd (__m128d __A, __m128i __I, __m128d __B) {
- return (__m128d) __builtin_ia32_vpermt2varpd128_mask ((__v2di) __I
- /* idx */ ,
- (__v2df) __A,
- (__v2df) __B,
- (__mmask8) -
- 1);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_mask_permutex2var_pd (__m128d __A, __mmask8 __U, __m128i __I,
- __m128d __B) {
- return (__m128d) __builtin_ia32_vpermt2varpd128_mask ((__v2di) __I
- /* idx */ ,
- (__v2df) __A,
- (__v2df) __B,
- (__mmask8)
- __U);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_maskz_permutex2var_pd (__mmask8 __U, __m128d __A, __m128i __I,
- __m128d __B) {
- return (__m128d) __builtin_ia32_vpermt2varpd128_maskz ((__v2di) __I
- /* idx */ ,
- (__v2df) __A,
- (__v2df) __B,
- (__mmask8)
- __U);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_permutex2var_pd (__m256d __A, __m256i __I, __m256d __B) {
- return (__m256d) __builtin_ia32_vpermt2varpd256_mask ((__v4di) __I
- /* idx */ ,
- (__v4df) __A,
- (__v4df) __B,
- (__mmask8) -
- 1);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_mask_permutex2var_pd (__m256d __A, __mmask8 __U, __m256i __I,
- __m256d __B) {
- return (__m256d) __builtin_ia32_vpermt2varpd256_mask ((__v4di) __I
- /* idx */ ,
- (__v4df) __A,
- (__v4df) __B,
- (__mmask8)
- __U);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_maskz_permutex2var_pd (__mmask8 __U, __m256d __A, __m256i __I,
- __m256d __B) {
- return (__m256d) __builtin_ia32_vpermt2varpd256_maskz ((__v4di) __I
- /* idx */ ,
- (__v4df) __A,
- (__v4df) __B,
- (__mmask8)
- __U);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_permutex2var_ps (__m128 __A, __m128i __I, __m128 __B) {
- return (__m128) __builtin_ia32_vpermt2varps128_mask ((__v4si) __I
- /* idx */ ,
- (__v4sf) __A,
- (__v4sf) __B,
- (__mmask8) -1);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_mask_permutex2var_ps (__m128 __A, __mmask8 __U, __m128i __I,
- __m128 __B) {
- return (__m128) __builtin_ia32_vpermt2varps128_mask ((__v4si) __I
- /* idx */ ,
- (__v4sf) __A,
- (__v4sf) __B,
- (__mmask8) __U);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_maskz_permutex2var_ps (__mmask8 __U, __m128 __A, __m128i __I,
- __m128 __B) {
- return (__m128) __builtin_ia32_vpermt2varps128_maskz ((__v4si) __I
- /* idx */ ,
- (__v4sf) __A,
- (__v4sf) __B,
- (__mmask8)
- __U);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_permutex2var_ps (__m256 __A, __m256i __I, __m256 __B) {
- return (__m256) __builtin_ia32_vpermt2varps256_mask ((__v8si) __I
- /* idx */ ,
- (__v8sf) __A,
- (__v8sf) __B,
- (__mmask8) -1);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_mask_permutex2var_ps (__m256 __A, __mmask8 __U, __m256i __I,
- __m256 __B) {
- return (__m256) __builtin_ia32_vpermt2varps256_mask ((__v8si) __I
- /* idx */ ,
- (__v8sf) __A,
- (__v8sf) __B,
- (__mmask8) __U);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_maskz_permutex2var_ps (__mmask8 __U, __m256 __A, __m256i __I,
- __m256 __B) {
- return (__m256) __builtin_ia32_vpermt2varps256_maskz ((__v8si) __I
- /* idx */ ,
- (__v8sf) __A,
- (__v8sf) __B,
- (__mmask8)
- __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_permutex2var_epi64 (__m128i __A, __m128i __I, __m128i __B) {
- return (__m128i) __builtin_ia32_vpermt2varq128_mask ((__v2di) __I
- /* idx */ ,
- (__v2di) __A,
- (__v2di) __B,
- (__mmask8) -1);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_permutex2var_epi64 (__m128i __A, __mmask8 __U, __m128i __I,
- __m128i __B) {
- return (__m128i) __builtin_ia32_vpermt2varq128_mask ((__v2di) __I
- /* idx */ ,
- (__v2di) __A,
- (__v2di) __B,
- (__mmask8) __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_permutex2var_epi64 (__mmask8 __U, __m128i __A, __m128i __I,
- __m128i __B) {
- return (__m128i) __builtin_ia32_vpermt2varq128_maskz ((__v2di) __I
- /* idx */ ,
- (__v2di) __A,
- (__v2di) __B,
- (__mmask8)
- __U);
-}
-
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_permutex2var_epi64 (__m256i __A, __m256i __I, __m256i __B) {
- return (__m256i) __builtin_ia32_vpermt2varq256_mask ((__v4di) __I
- /* idx */ ,
- (__v4di) __A,
- (__v4di) __B,
- (__mmask8) -1);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_permutex2var_epi64 (__m256i __A, __mmask8 __U, __m256i __I,
- __m256i __B) {
- return (__m256i) __builtin_ia32_vpermt2varq256_mask ((__v4di) __I
- /* idx */ ,
- (__v4di) __A,
- (__v4di) __B,
- (__mmask8) __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_permutex2var_epi64 (__mmask8 __U, __m256i __A,
- __m256i __I, __m256i __B) {
- return (__m256i) __builtin_ia32_vpermt2varq256_maskz ((__v4di) __I
- /* idx */ ,
- (__v4di) __A,
- (__v4di) __B,
- (__mmask8)
- __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_cvtepi8_epi32(__m128i __W, __mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
- (__v4si)_mm_cvtepi8_epi32(__A),
- (__v4si)__W);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_cvtepi8_epi32(__mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
- (__v4si)_mm_cvtepi8_epi32(__A),
- (__v4si)_mm_setzero_si128());
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_cvtepi8_epi32 (__m256i __W, __mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
- (__v8si)_mm256_cvtepi8_epi32(__A),
- (__v8si)__W);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_cvtepi8_epi32 (__mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
- (__v8si)_mm256_cvtepi8_epi32(__A),
- (__v8si)_mm256_setzero_si256());
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_cvtepi8_epi64(__m128i __W, __mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_cvtepi8_epi64(__A),
- (__v2di)__W);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_cvtepi8_epi64(__mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_cvtepi8_epi64(__A),
- (__v2di)_mm_setzero_si128());
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_cvtepi8_epi64(__m256i __W, __mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_cvtepi8_epi64(__A),
- (__v4di)__W);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_cvtepi8_epi64(__mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_cvtepi8_epi64(__A),
- (__v4di)_mm256_setzero_si256());
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_cvtepi32_epi64(__m128i __W, __mmask8 __U, __m128i __X)
-{
- return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_cvtepi32_epi64(__X),
- (__v2di)__W);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_cvtepi32_epi64(__mmask8 __U, __m128i __X)
-{
- return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_cvtepi32_epi64(__X),
- (__v2di)_mm_setzero_si128());
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_cvtepi32_epi64(__m256i __W, __mmask8 __U, __m128i __X)
-{
- return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_cvtepi32_epi64(__X),
- (__v4di)__W);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_cvtepi32_epi64(__mmask8 __U, __m128i __X)
-{
- return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_cvtepi32_epi64(__X),
- (__v4di)_mm256_setzero_si256());
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_cvtepi16_epi32(__m128i __W, __mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
- (__v4si)_mm_cvtepi16_epi32(__A),
- (__v4si)__W);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_cvtepi16_epi32(__mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
- (__v4si)_mm_cvtepi16_epi32(__A),
- (__v4si)_mm_setzero_si128());
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_cvtepi16_epi32(__m256i __W, __mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
- (__v8si)_mm256_cvtepi16_epi32(__A),
- (__v8si)__W);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_cvtepi16_epi32 (__mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
- (__v8si)_mm256_cvtepi16_epi32(__A),
- (__v8si)_mm256_setzero_si256());
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_cvtepi16_epi64(__m128i __W, __mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_cvtepi16_epi64(__A),
- (__v2di)__W);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_cvtepi16_epi64(__mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_cvtepi16_epi64(__A),
- (__v2di)_mm_setzero_si128());
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_cvtepi16_epi64(__m256i __W, __mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_cvtepi16_epi64(__A),
- (__v4di)__W);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_cvtepi16_epi64(__mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_cvtepi16_epi64(__A),
- (__v4di)_mm256_setzero_si256());
-}
-
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_cvtepu8_epi32(__m128i __W, __mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
- (__v4si)_mm_cvtepu8_epi32(__A),
- (__v4si)__W);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_cvtepu8_epi32(__mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
- (__v4si)_mm_cvtepu8_epi32(__A),
- (__v4si)_mm_setzero_si128());
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_cvtepu8_epi32(__m256i __W, __mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
- (__v8si)_mm256_cvtepu8_epi32(__A),
- (__v8si)__W);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_cvtepu8_epi32(__mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
- (__v8si)_mm256_cvtepu8_epi32(__A),
- (__v8si)_mm256_setzero_si256());
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_cvtepu8_epi64(__m128i __W, __mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_cvtepu8_epi64(__A),
- (__v2di)__W);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_cvtepu8_epi64(__mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_cvtepu8_epi64(__A),
- (__v2di)_mm_setzero_si128());
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_cvtepu8_epi64(__m256i __W, __mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_cvtepu8_epi64(__A),
- (__v4di)__W);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_cvtepu8_epi64 (__mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_cvtepu8_epi64(__A),
- (__v4di)_mm256_setzero_si256());
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_cvtepu32_epi64(__m128i __W, __mmask8 __U, __m128i __X)
-{
- return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_cvtepu32_epi64(__X),
- (__v2di)__W);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_cvtepu32_epi64(__mmask8 __U, __m128i __X)
-{
- return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_cvtepu32_epi64(__X),
- (__v2di)_mm_setzero_si128());
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_cvtepu32_epi64(__m256i __W, __mmask8 __U, __m128i __X)
-{
- return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_cvtepu32_epi64(__X),
- (__v4di)__W);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_cvtepu32_epi64(__mmask8 __U, __m128i __X)
-{
- return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_cvtepu32_epi64(__X),
- (__v4di)_mm256_setzero_si256());
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_cvtepu16_epi32(__m128i __W, __mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
- (__v4si)_mm_cvtepu16_epi32(__A),
- (__v4si)__W);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_cvtepu16_epi32(__mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
- (__v4si)_mm_cvtepu16_epi32(__A),
- (__v4si)_mm_setzero_si128());
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_cvtepu16_epi32(__m256i __W, __mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
- (__v8si)_mm256_cvtepu16_epi32(__A),
- (__v8si)__W);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_cvtepu16_epi32(__mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
- (__v8si)_mm256_cvtepu16_epi32(__A),
- (__v8si)_mm256_setzero_si256());
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_cvtepu16_epi64(__m128i __W, __mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_cvtepu16_epi64(__A),
- (__v2di)__W);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_cvtepu16_epi64(__mmask8 __U, __m128i __A)
-{
- return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
- (__v2di)_mm_cvtepu16_epi64(__A),
- (__v2di)_mm_setzero_si128());
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_cvtepu16_epi64(__m256i __W, __mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_cvtepu16_epi64(__A),
- (__v4di)__W);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_cvtepu16_epi64(__mmask8 __U, __m128i __A)
-{
- return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
- (__v4di)_mm256_cvtepu16_epi64(__A),
- (__v4di)_mm256_setzero_si256());
-}
+ (__v2df)(__m128d)(v1), (int)(scale))
+#define _mm_mask_i64scatter_pd(addr, mask, index, v1, scale) \
+ __builtin_ia32_scatterdiv2df((void *)(addr), (__mmask8)(mask), \
+ (__v2di)(__m128i)(index), \
+ (__v2df)(__m128d)(v1), (int)(scale))
+
+#define _mm_i64scatter_epi64(addr, index, v1, scale) \
+ __builtin_ia32_scatterdiv2di((void *)(addr), (__mmask8)-1, \
+ (__v2di)(__m128i)(index), \
+ (__v2di)(__m128i)(v1), (int)(scale))
+
+#define _mm_mask_i64scatter_epi64(addr, mask, index, v1, scale) \
+ __builtin_ia32_scatterdiv2di((void *)(addr), (__mmask8)(mask), \
+ (__v2di)(__m128i)(index), \
+ (__v2di)(__m128i)(v1), (int)(scale))
-#define _mm_rol_epi32(a, b) __extension__ ({\
- (__m128i)__builtin_ia32_prold128_mask((__v4si)(__m128i)(a), (int)(b), \
- (__v4si)_mm_setzero_si128(), \
- (__mmask8)-1); })
+#define _mm256_i64scatter_pd(addr, index, v1, scale) \
+ __builtin_ia32_scatterdiv4df((void *)(addr), (__mmask8)-1, \
+ (__v4di)(__m256i)(index), \
+ (__v4df)(__m256d)(v1), (int)(scale))
-#define _mm_mask_rol_epi32(w, u, a, b) __extension__ ({\
- (__m128i)__builtin_ia32_prold128_mask((__v4si)(__m128i)(a), (int)(b), \
- (__v4si)(__m128i)(w), (__mmask8)(u)); })
+#define _mm256_mask_i64scatter_pd(addr, mask, index, v1, scale) \
+ __builtin_ia32_scatterdiv4df((void *)(addr), (__mmask8)(mask), \
+ (__v4di)(__m256i)(index), \
+ (__v4df)(__m256d)(v1), (int)(scale))
-#define _mm_maskz_rol_epi32(u, a, b) __extension__ ({\
- (__m128i)__builtin_ia32_prold128_mask((__v4si)(__m128i)(a), (int)(b), \
- (__v4si)_mm_setzero_si128(), \
- (__mmask8)(u)); })
+#define _mm256_i64scatter_epi64(addr, index, v1, scale) \
+ __builtin_ia32_scatterdiv4di((void *)(addr), (__mmask8)-1, \
+ (__v4di)(__m256i)(index), \
+ (__v4di)(__m256i)(v1), (int)(scale))
-#define _mm256_rol_epi32(a, b) __extension__ ({\
- (__m256i)__builtin_ia32_prold256_mask((__v8si)(__m256i)(a), (int)(b), \
- (__v8si)_mm256_setzero_si256(), \
- (__mmask8)-1); })
+#define _mm256_mask_i64scatter_epi64(addr, mask, index, v1, scale) \
+ __builtin_ia32_scatterdiv4di((void *)(addr), (__mmask8)(mask), \
+ (__v4di)(__m256i)(index), \
+ (__v4di)(__m256i)(v1), (int)(scale))
-#define _mm256_mask_rol_epi32(w, u, a, b) __extension__ ({\
- (__m256i)__builtin_ia32_prold256_mask((__v8si)(__m256i)(a), (int)(b), \
- (__v8si)(__m256i)(w), (__mmask8)(u)); })
+#define _mm_i64scatter_ps(addr, index, v1, scale) \
+ __builtin_ia32_scatterdiv4sf((void *)(addr), (__mmask8)-1, \
+ (__v2di)(__m128i)(index), (__v4sf)(__m128)(v1), \
+ (int)(scale))
-#define _mm256_maskz_rol_epi32(u, a, b) __extension__ ({\
- (__m256i)__builtin_ia32_prold256_mask((__v8si)(__m256i)(a), (int)(b), \
- (__v8si)_mm256_setzero_si256(), \
- (__mmask8)(u)); })
+#define _mm_mask_i64scatter_ps(addr, mask, index, v1, scale) \
+ __builtin_ia32_scatterdiv4sf((void *)(addr), (__mmask8)(mask), \
+ (__v2di)(__m128i)(index), (__v4sf)(__m128)(v1), \
+ (int)(scale))
-#define _mm_rol_epi64(a, b) __extension__ ({\
- (__m128i)__builtin_ia32_prolq128_mask((__v2di)(__m128i)(a), (int)(b), \
- (__v2di)_mm_setzero_di(), \
- (__mmask8)-1); })
+#define _mm_i64scatter_epi32(addr, index, v1, scale) \
+ __builtin_ia32_scatterdiv4si((void *)(addr), (__mmask8)-1, \
+ (__v2di)(__m128i)(index), \
+ (__v4si)(__m128i)(v1), (int)(scale))
-#define _mm_mask_rol_epi64(w, u, a, b) __extension__ ({\
- (__m128i)__builtin_ia32_prolq128_mask((__v2di)(__m128i)(a), (int)(b), \
- (__v2di)(__m128i)(w), (__mmask8)(u)); })
+#define _mm_mask_i64scatter_epi32(addr, mask, index, v1, scale) \
+ __builtin_ia32_scatterdiv4si((void *)(addr), (__mmask8)(mask), \
+ (__v2di)(__m128i)(index), \
+ (__v4si)(__m128i)(v1), (int)(scale))
-#define _mm_maskz_rol_epi64(u, a, b) __extension__ ({\
- (__m128i)__builtin_ia32_prolq128_mask((__v2di)(__m128i)(a), (int)(b), \
- (__v2di)_mm_setzero_di(), \
- (__mmask8)(u)); })
+#define _mm256_i64scatter_ps(addr, index, v1, scale) \
+ __builtin_ia32_scatterdiv8sf((void *)(addr), (__mmask8)-1, \
+ (__v4di)(__m256i)(index), (__v4sf)(__m128)(v1), \
+ (int)(scale))
-#define _mm256_rol_epi64(a, b) __extension__ ({\
- (__m256i)__builtin_ia32_prolq256_mask((__v4di)(__m256i)(a), (int)(b), \
- (__v4di)_mm256_setzero_si256(), \
- (__mmask8)-1); })
+#define _mm256_mask_i64scatter_ps(addr, mask, index, v1, scale) \
+ __builtin_ia32_scatterdiv8sf((void *)(addr), (__mmask8)(mask), \
+ (__v4di)(__m256i)(index), (__v4sf)(__m128)(v1), \
+ (int)(scale))
-#define _mm256_mask_rol_epi64(w, u, a, b) __extension__ ({\
- (__m256i)__builtin_ia32_prolq256_mask((__v4di)(__m256i)(a), (int)(b), \
- (__v4di)(__m256i)(w), (__mmask8)(u)); })
+#define _mm256_i64scatter_epi32(addr, index, v1, scale) \
+ __builtin_ia32_scatterdiv8si((void *)(addr), (__mmask8)-1, \
+ (__v4di)(__m256i)(index), \
+ (__v4si)(__m128i)(v1), (int)(scale))
-#define _mm256_maskz_rol_epi64(u, a, b) __extension__ ({\
- (__m256i)__builtin_ia32_prolq256_mask((__v4di)(__m256i)(a), (int)(b), \
- (__v4di)_mm256_setzero_si256(), \
- (__mmask8)(u)); })
+#define _mm256_mask_i64scatter_epi32(addr, mask, index, v1, scale) \
+ __builtin_ia32_scatterdiv8si((void *)(addr), (__mmask8)(mask), \
+ (__v4di)(__m256i)(index), \
+ (__v4si)(__m128i)(v1), (int)(scale))
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+#define _mm_i32scatter_pd(addr, index, v1, scale) \
+ __builtin_ia32_scattersiv2df((void *)(addr), (__mmask8)-1, \
+ (__v4si)(__m128i)(index), \
+ (__v2df)(__m128d)(v1), (int)(scale))
+
+#define _mm_mask_i32scatter_pd(addr, mask, index, v1, scale) \
+ __builtin_ia32_scattersiv2df((void *)(addr), (__mmask8)(mask), \
+ (__v4si)(__m128i)(index), \
+ (__v2df)(__m128d)(v1), (int)(scale))
+
+#define _mm_i32scatter_epi64(addr, index, v1, scale) \
+ __builtin_ia32_scattersiv2di((void *)(addr), (__mmask8)-1, \
+ (__v4si)(__m128i)(index), \
+ (__v2di)(__m128i)(v1), (int)(scale))
+
+#define _mm_mask_i32scatter_epi64(addr, mask, index, v1, scale) \
+ __builtin_ia32_scattersiv2di((void *)(addr), (__mmask8)(mask), \
+ (__v4si)(__m128i)(index), \
+ (__v2di)(__m128i)(v1), (int)(scale))
+
+#define _mm256_i32scatter_pd(addr, index, v1, scale) \
+ __builtin_ia32_scattersiv4df((void *)(addr), (__mmask8)-1, \
+ (__v4si)(__m128i)(index), \
+ (__v4df)(__m256d)(v1), (int)(scale))
+
+#define _mm256_mask_i32scatter_pd(addr, mask, index, v1, scale) \
+ __builtin_ia32_scattersiv4df((void *)(addr), (__mmask8)(mask), \
+ (__v4si)(__m128i)(index), \
+ (__v4df)(__m256d)(v1), (int)(scale))
+
+#define _mm256_i32scatter_epi64(addr, index, v1, scale) \
+ __builtin_ia32_scattersiv4di((void *)(addr), (__mmask8)-1, \
+ (__v4si)(__m128i)(index), \
+ (__v4di)(__m256i)(v1), (int)(scale))
+
+#define _mm256_mask_i32scatter_epi64(addr, mask, index, v1, scale) \
+ __builtin_ia32_scattersiv4di((void *)(addr), (__mmask8)(mask), \
+ (__v4si)(__m128i)(index), \
+ (__v4di)(__m256i)(v1), (int)(scale))
+
+#define _mm_i32scatter_ps(addr, index, v1, scale) \
+ __builtin_ia32_scattersiv4sf((void *)(addr), (__mmask8)-1, \
+ (__v4si)(__m128i)(index), (__v4sf)(__m128)(v1), \
+ (int)(scale))
+
+#define _mm_mask_i32scatter_ps(addr, mask, index, v1, scale) \
+ __builtin_ia32_scattersiv4sf((void *)(addr), (__mmask8)(mask), \
+ (__v4si)(__m128i)(index), (__v4sf)(__m128)(v1), \
+ (int)(scale))
+
+#define _mm_i32scatter_epi32(addr, index, v1, scale) \
+ __builtin_ia32_scattersiv4si((void *)(addr), (__mmask8)-1, \
+ (__v4si)(__m128i)(index), \
+ (__v4si)(__m128i)(v1), (int)(scale))
+
+#define _mm_mask_i32scatter_epi32(addr, mask, index, v1, scale) \
+ __builtin_ia32_scattersiv4si((void *)(addr), (__mmask8)(mask), \
+ (__v4si)(__m128i)(index), \
+ (__v4si)(__m128i)(v1), (int)(scale))
+
+#define _mm256_i32scatter_ps(addr, index, v1, scale) \
+ __builtin_ia32_scattersiv8sf((void *)(addr), (__mmask8)-1, \
+ (__v8si)(__m256i)(index), (__v8sf)(__m256)(v1), \
+ (int)(scale))
+
+#define _mm256_mask_i32scatter_ps(addr, mask, index, v1, scale) \
+ __builtin_ia32_scattersiv8sf((void *)(addr), (__mmask8)(mask), \
+ (__v8si)(__m256i)(index), (__v8sf)(__m256)(v1), \
+ (int)(scale))
+
+#define _mm256_i32scatter_epi32(addr, index, v1, scale) \
+ __builtin_ia32_scattersiv8si((void *)(addr), (__mmask8)-1, \
+ (__v8si)(__m256i)(index), \
+ (__v8si)(__m256i)(v1), (int)(scale))
+
+#define _mm256_mask_i32scatter_epi32(addr, mask, index, v1, scale) \
+ __builtin_ia32_scattersiv8si((void *)(addr), (__mmask8)(mask), \
+ (__v8si)(__m256i)(index), \
+ (__v8si)(__m256i)(v1), (int)(scale))
+
+ static __inline__ __m128d __DEFAULT_FN_ATTRS128
+ _mm_mask_sqrt_pd(__m128d __W, __mmask8 __U, __m128d __A) {
+ return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
+ (__v2df)_mm_sqrt_pd(__A),
+ (__v2df)__W);
+ }
+
+ static __inline__ __m128d __DEFAULT_FN_ATTRS128
+ _mm_maskz_sqrt_pd(__mmask8 __U, __m128d __A) {
+ return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
+ (__v2df)_mm_sqrt_pd(__A),
+ (__v2df)_mm_setzero_pd());
+ }
+
+ static __inline__ __m256d __DEFAULT_FN_ATTRS256
+ _mm256_mask_sqrt_pd(__m256d __W, __mmask8 __U, __m256d __A) {
+ return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
+ (__v4df)_mm256_sqrt_pd(__A),
+ (__v4df)__W);
+ }
+
+ static __inline__ __m256d __DEFAULT_FN_ATTRS256
+ _mm256_maskz_sqrt_pd(__mmask8 __U, __m256d __A) {
+ return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
+ (__v4df)_mm256_sqrt_pd(__A),
+ (__v4df)_mm256_setzero_pd());
+ }
+
+ static __inline__ __m128 __DEFAULT_FN_ATTRS128
+ _mm_mask_sqrt_ps(__m128 __W, __mmask8 __U, __m128 __A) {
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm_sqrt_ps(__A),
+ (__v4sf)__W);
+ }
+
+ static __inline__ __m128 __DEFAULT_FN_ATTRS128
+ _mm_maskz_sqrt_ps(__mmask8 __U, __m128 __A) {
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm_sqrt_ps(__A),
+ (__v4sf)_mm_setzero_ps());
+ }
+
+ static __inline__ __m256 __DEFAULT_FN_ATTRS256
+ _mm256_mask_sqrt_ps(__m256 __W, __mmask8 __U, __m256 __A) {
+ return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
+ (__v8sf)_mm256_sqrt_ps(__A),
+ (__v8sf)__W);
+ }
+
+ static __inline__ __m256 __DEFAULT_FN_ATTRS256
+ _mm256_maskz_sqrt_ps(__mmask8 __U, __m256 __A) {
+ return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
+ (__v8sf)_mm256_sqrt_ps(__A),
+ (__v8sf)_mm256_setzero_ps());
+ }
+
+ static __inline__ __m128d __DEFAULT_FN_ATTRS128
+ _mm_mask_sub_pd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B) {
+ return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
+ (__v2df)_mm_sub_pd(__A, __B),
+ (__v2df)__W);
+ }
+
+ static __inline__ __m128d __DEFAULT_FN_ATTRS128
+ _mm_maskz_sub_pd(__mmask8 __U, __m128d __A, __m128d __B) {
+ return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
+ (__v2df)_mm_sub_pd(__A, __B),
+ (__v2df)_mm_setzero_pd());
+ }
+
+ static __inline__ __m256d __DEFAULT_FN_ATTRS256
+ _mm256_mask_sub_pd(__m256d __W, __mmask8 __U, __m256d __A, __m256d __B) {
+ return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
+ (__v4df)_mm256_sub_pd(__A, __B),
+ (__v4df)__W);
+ }
+
+ static __inline__ __m256d __DEFAULT_FN_ATTRS256
+ _mm256_maskz_sub_pd(__mmask8 __U, __m256d __A, __m256d __B) {
+ return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
+ (__v4df)_mm256_sub_pd(__A, __B),
+ (__v4df)_mm256_setzero_pd());
+ }
+
+ static __inline__ __m128 __DEFAULT_FN_ATTRS128
+ _mm_mask_sub_ps(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) {
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm_sub_ps(__A, __B),
+ (__v4sf)__W);
+ }
+
+ static __inline__ __m128 __DEFAULT_FN_ATTRS128
+ _mm_maskz_sub_ps(__mmask8 __U, __m128 __A, __m128 __B) {
+ return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
+ (__v4sf)_mm_sub_ps(__A, __B),
+ (__v4sf)_mm_setzero_ps());
+ }
+
+ static __inline__ __m256 __DEFAULT_FN_ATTRS256
+ _mm256_mask_sub_ps(__m256 __W, __mmask8 __U, __m256 __A, __m256 __B) {
+ return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
+ (__v8sf)_mm256_sub_ps(__A, __B),
+ (__v8sf)__W);
+ }
+
+ static __inline__ __m256 __DEFAULT_FN_ATTRS256
+ _mm256_maskz_sub_ps(__mmask8 __U, __m256 __A, __m256 __B) {
+ return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
+ (__v8sf)_mm256_sub_ps(__A, __B),
+ (__v8sf)_mm256_setzero_ps());
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_permutex2var_epi32(__m128i __A, __m128i __I, __m128i __B) {
+ return (__m128i)__builtin_ia32_vpermi2vard128((__v4si) __A, (__v4si)__I,
+ (__v4si)__B);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_mask_permutex2var_epi32(__m128i __A, __mmask8 __U, __m128i __I,
+ __m128i __B) {
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_permutex2var_epi32(__A, __I, __B),
+ (__v4si)__A);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_mask2_permutex2var_epi32(__m128i __A, __m128i __I, __mmask8 __U,
+ __m128i __B) {
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_permutex2var_epi32(__A, __I, __B),
+ (__v4si)__I);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_maskz_permutex2var_epi32(__mmask8 __U, __m128i __A, __m128i __I,
+ __m128i __B) {
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_permutex2var_epi32(__A, __I, __B),
+ (__v4si)_mm_setzero_si128());
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_permutex2var_epi32(__m256i __A, __m256i __I, __m256i __B) {
+ return (__m256i)__builtin_ia32_vpermi2vard256((__v8si)__A, (__v8si) __I,
+ (__v8si) __B);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_mask_permutex2var_epi32(__m256i __A, __mmask8 __U, __m256i __I,
+ __m256i __B) {
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_permutex2var_epi32(__A, __I, __B),
+ (__v8si)__A);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_mask2_permutex2var_epi32(__m256i __A, __m256i __I, __mmask8 __U,
+ __m256i __B) {
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_permutex2var_epi32(__A, __I, __B),
+ (__v8si)__I);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_maskz_permutex2var_epi32(__mmask8 __U, __m256i __A, __m256i __I,
+ __m256i __B) {
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_permutex2var_epi32(__A, __I, __B),
+ (__v8si)_mm256_setzero_si256());
+ }
+
+ static __inline__ __m128d __DEFAULT_FN_ATTRS128
+ _mm_permutex2var_pd(__m128d __A, __m128i __I, __m128d __B) {
+ return (__m128d)__builtin_ia32_vpermi2varpd128((__v2df)__A, (__v2di)__I,
+ (__v2df)__B);
+ }
+
+ static __inline__ __m128d __DEFAULT_FN_ATTRS128
+ _mm_mask_permutex2var_pd(__m128d __A, __mmask8 __U, __m128i __I, __m128d __B) {
+ return (__m128d)__builtin_ia32_selectpd_128(__U,
+ (__v2df)_mm_permutex2var_pd(__A, __I, __B),
+ (__v2df)__A);
+ }
+
+ static __inline__ __m128d __DEFAULT_FN_ATTRS128
+ _mm_mask2_permutex2var_pd(__m128d __A, __m128i __I, __mmask8 __U, __m128d __B) {
+ return (__m128d)__builtin_ia32_selectpd_128(__U,
+ (__v2df)_mm_permutex2var_pd(__A, __I, __B),
+ (__v2df)(__m128d)__I);
+ }
+
+ static __inline__ __m128d __DEFAULT_FN_ATTRS128
+ _mm_maskz_permutex2var_pd(__mmask8 __U, __m128d __A, __m128i __I, __m128d __B) {
+ return (__m128d)__builtin_ia32_selectpd_128(__U,
+ (__v2df)_mm_permutex2var_pd(__A, __I, __B),
+ (__v2df)_mm_setzero_pd());
+ }
+
+ static __inline__ __m256d __DEFAULT_FN_ATTRS256
+ _mm256_permutex2var_pd(__m256d __A, __m256i __I, __m256d __B) {
+ return (__m256d)__builtin_ia32_vpermi2varpd256((__v4df)__A, (__v4di)__I,
+ (__v4df)__B);
+ }
+
+ static __inline__ __m256d __DEFAULT_FN_ATTRS256
+ _mm256_mask_permutex2var_pd(__m256d __A, __mmask8 __U, __m256i __I,
+ __m256d __B) {
+ return (__m256d)__builtin_ia32_selectpd_256(__U,
+ (__v4df)_mm256_permutex2var_pd(__A, __I, __B),
+ (__v4df)__A);
+ }
+
+ static __inline__ __m256d __DEFAULT_FN_ATTRS256
+ _mm256_mask2_permutex2var_pd(__m256d __A, __m256i __I, __mmask8 __U,
+ __m256d __B) {
+ return (__m256d)__builtin_ia32_selectpd_256(__U,
+ (__v4df)_mm256_permutex2var_pd(__A, __I, __B),
+ (__v4df)(__m256d)__I);
+ }
+
+ static __inline__ __m256d __DEFAULT_FN_ATTRS256
+ _mm256_maskz_permutex2var_pd(__mmask8 __U, __m256d __A, __m256i __I,
+ __m256d __B) {
+ return (__m256d)__builtin_ia32_selectpd_256(__U,
+ (__v4df)_mm256_permutex2var_pd(__A, __I, __B),
+ (__v4df)_mm256_setzero_pd());
+ }
+
+ static __inline__ __m128 __DEFAULT_FN_ATTRS128
+ _mm_permutex2var_ps(__m128 __A, __m128i __I, __m128 __B) {
+ return (__m128)__builtin_ia32_vpermi2varps128((__v4sf)__A, (__v4si)__I,
+ (__v4sf)__B);
+ }
+
+ static __inline__ __m128 __DEFAULT_FN_ATTRS128
+ _mm_mask_permutex2var_ps(__m128 __A, __mmask8 __U, __m128i __I, __m128 __B) {
+ return (__m128)__builtin_ia32_selectps_128(__U,
+ (__v4sf)_mm_permutex2var_ps(__A, __I, __B),
+ (__v4sf)__A);
+ }
+
+ static __inline__ __m128 __DEFAULT_FN_ATTRS128
+ _mm_mask2_permutex2var_ps(__m128 __A, __m128i __I, __mmask8 __U, __m128 __B) {
+ return (__m128)__builtin_ia32_selectps_128(__U,
+ (__v4sf)_mm_permutex2var_ps(__A, __I, __B),
+ (__v4sf)(__m128)__I);
+ }
+
+ static __inline__ __m128 __DEFAULT_FN_ATTRS128
+ _mm_maskz_permutex2var_ps(__mmask8 __U, __m128 __A, __m128i __I, __m128 __B) {
+ return (__m128)__builtin_ia32_selectps_128(__U,
+ (__v4sf)_mm_permutex2var_ps(__A, __I, __B),
+ (__v4sf)_mm_setzero_ps());
+ }
+
+ static __inline__ __m256 __DEFAULT_FN_ATTRS256
+ _mm256_permutex2var_ps(__m256 __A, __m256i __I, __m256 __B) {
+ return (__m256)__builtin_ia32_vpermi2varps256((__v8sf)__A, (__v8si)__I,
+ (__v8sf) __B);
+ }
+
+ static __inline__ __m256 __DEFAULT_FN_ATTRS256
+ _mm256_mask_permutex2var_ps(__m256 __A, __mmask8 __U, __m256i __I, __m256 __B) {
+ return (__m256)__builtin_ia32_selectps_256(__U,
+ (__v8sf)_mm256_permutex2var_ps(__A, __I, __B),
+ (__v8sf)__A);
+ }
+
+ static __inline__ __m256 __DEFAULT_FN_ATTRS256
+ _mm256_mask2_permutex2var_ps(__m256 __A, __m256i __I, __mmask8 __U,
+ __m256 __B) {
+ return (__m256)__builtin_ia32_selectps_256(__U,
+ (__v8sf)_mm256_permutex2var_ps(__A, __I, __B),
+ (__v8sf)(__m256)__I);
+ }
+
+ static __inline__ __m256 __DEFAULT_FN_ATTRS256
+ _mm256_maskz_permutex2var_ps(__mmask8 __U, __m256 __A, __m256i __I,
+ __m256 __B) {
+ return (__m256)__builtin_ia32_selectps_256(__U,
+ (__v8sf)_mm256_permutex2var_ps(__A, __I, __B),
+ (__v8sf)_mm256_setzero_ps());
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_permutex2var_epi64(__m128i __A, __m128i __I, __m128i __B) {
+ return (__m128i)__builtin_ia32_vpermi2varq128((__v2di)__A, (__v2di)__I,
+ (__v2di)__B);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_mask_permutex2var_epi64(__m128i __A, __mmask8 __U, __m128i __I,
+ __m128i __B) {
+ return (__m128i)__builtin_ia32_selectq_128(__U,
+ (__v2di)_mm_permutex2var_epi64(__A, __I, __B),
+ (__v2di)__A);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_mask2_permutex2var_epi64(__m128i __A, __m128i __I, __mmask8 __U,
+ __m128i __B) {
+ return (__m128i)__builtin_ia32_selectq_128(__U,
+ (__v2di)_mm_permutex2var_epi64(__A, __I, __B),
+ (__v2di)__I);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_maskz_permutex2var_epi64(__mmask8 __U, __m128i __A, __m128i __I,
+ __m128i __B) {
+ return (__m128i)__builtin_ia32_selectq_128(__U,
+ (__v2di)_mm_permutex2var_epi64(__A, __I, __B),
+ (__v2di)_mm_setzero_si128());
+ }
+
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_permutex2var_epi64(__m256i __A, __m256i __I, __m256i __B) {
+ return (__m256i)__builtin_ia32_vpermi2varq256((__v4di)__A, (__v4di) __I,
+ (__v4di) __B);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_mask_permutex2var_epi64(__m256i __A, __mmask8 __U, __m256i __I,
+ __m256i __B) {
+ return (__m256i)__builtin_ia32_selectq_256(__U,
+ (__v4di)_mm256_permutex2var_epi64(__A, __I, __B),
+ (__v4di)__A);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_mask2_permutex2var_epi64(__m256i __A, __m256i __I, __mmask8 __U,
+ __m256i __B) {
+ return (__m256i)__builtin_ia32_selectq_256(__U,
+ (__v4di)_mm256_permutex2var_epi64(__A, __I, __B),
+ (__v4di)__I);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_maskz_permutex2var_epi64(__mmask8 __U, __m256i __A, __m256i __I,
+ __m256i __B) {
+ return (__m256i)__builtin_ia32_selectq_256(__U,
+ (__v4di)_mm256_permutex2var_epi64(__A, __I, __B),
+ (__v4di)_mm256_setzero_si256());
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_mask_cvtepi8_epi32(__m128i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_cvtepi8_epi32(__A),
+ (__v4si)__W);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_maskz_cvtepi8_epi32(__mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_cvtepi8_epi32(__A),
+ (__v4si)_mm_setzero_si128());
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_mask_cvtepi8_epi32 (__m256i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_cvtepi8_epi32(__A),
+ (__v8si)__W);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_maskz_cvtepi8_epi32 (__mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_cvtepi8_epi32(__A),
+ (__v8si)_mm256_setzero_si256());
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_mask_cvtepi8_epi64(__m128i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_cvtepi8_epi64(__A),
+ (__v2di)__W);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_maskz_cvtepi8_epi64(__mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_cvtepi8_epi64(__A),
+ (__v2di)_mm_setzero_si128());
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_mask_cvtepi8_epi64(__m256i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_cvtepi8_epi64(__A),
+ (__v4di)__W);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_maskz_cvtepi8_epi64(__mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_cvtepi8_epi64(__A),
+ (__v4di)_mm256_setzero_si256());
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_mask_cvtepi32_epi64(__m128i __W, __mmask8 __U, __m128i __X)
+ {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_cvtepi32_epi64(__X),
+ (__v2di)__W);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_maskz_cvtepi32_epi64(__mmask8 __U, __m128i __X)
+ {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_cvtepi32_epi64(__X),
+ (__v2di)_mm_setzero_si128());
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_mask_cvtepi32_epi64(__m256i __W, __mmask8 __U, __m128i __X)
+ {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_cvtepi32_epi64(__X),
+ (__v4di)__W);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_maskz_cvtepi32_epi64(__mmask8 __U, __m128i __X)
+ {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_cvtepi32_epi64(__X),
+ (__v4di)_mm256_setzero_si256());
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_mask_cvtepi16_epi32(__m128i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_cvtepi16_epi32(__A),
+ (__v4si)__W);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_maskz_cvtepi16_epi32(__mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_cvtepi16_epi32(__A),
+ (__v4si)_mm_setzero_si128());
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_mask_cvtepi16_epi32(__m256i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_cvtepi16_epi32(__A),
+ (__v8si)__W);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_maskz_cvtepi16_epi32 (__mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_cvtepi16_epi32(__A),
+ (__v8si)_mm256_setzero_si256());
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_mask_cvtepi16_epi64(__m128i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_cvtepi16_epi64(__A),
+ (__v2di)__W);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_maskz_cvtepi16_epi64(__mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_cvtepi16_epi64(__A),
+ (__v2di)_mm_setzero_si128());
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_mask_cvtepi16_epi64(__m256i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_cvtepi16_epi64(__A),
+ (__v4di)__W);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_maskz_cvtepi16_epi64(__mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_cvtepi16_epi64(__A),
+ (__v4di)_mm256_setzero_si256());
+ }
+
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_mask_cvtepu8_epi32(__m128i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_cvtepu8_epi32(__A),
+ (__v4si)__W);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_maskz_cvtepu8_epi32(__mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_cvtepu8_epi32(__A),
+ (__v4si)_mm_setzero_si128());
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_mask_cvtepu8_epi32(__m256i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_cvtepu8_epi32(__A),
+ (__v8si)__W);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_maskz_cvtepu8_epi32(__mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_cvtepu8_epi32(__A),
+ (__v8si)_mm256_setzero_si256());
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_mask_cvtepu8_epi64(__m128i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_cvtepu8_epi64(__A),
+ (__v2di)__W);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_maskz_cvtepu8_epi64(__mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_cvtepu8_epi64(__A),
+ (__v2di)_mm_setzero_si128());
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_mask_cvtepu8_epi64(__m256i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_cvtepu8_epi64(__A),
+ (__v4di)__W);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_maskz_cvtepu8_epi64 (__mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_cvtepu8_epi64(__A),
+ (__v4di)_mm256_setzero_si256());
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_mask_cvtepu32_epi64(__m128i __W, __mmask8 __U, __m128i __X)
+ {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_cvtepu32_epi64(__X),
+ (__v2di)__W);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_maskz_cvtepu32_epi64(__mmask8 __U, __m128i __X)
+ {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_cvtepu32_epi64(__X),
+ (__v2di)_mm_setzero_si128());
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_mask_cvtepu32_epi64(__m256i __W, __mmask8 __U, __m128i __X)
+ {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_cvtepu32_epi64(__X),
+ (__v4di)__W);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_maskz_cvtepu32_epi64(__mmask8 __U, __m128i __X)
+ {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_cvtepu32_epi64(__X),
+ (__v4di)_mm256_setzero_si256());
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_mask_cvtepu16_epi32(__m128i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_cvtepu16_epi32(__A),
+ (__v4si)__W);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_maskz_cvtepu16_epi32(__mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
+ (__v4si)_mm_cvtepu16_epi32(__A),
+ (__v4si)_mm_setzero_si128());
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_mask_cvtepu16_epi32(__m256i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_cvtepu16_epi32(__A),
+ (__v8si)__W);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_maskz_cvtepu16_epi32(__mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
+ (__v8si)_mm256_cvtepu16_epi32(__A),
+ (__v8si)_mm256_setzero_si256());
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_mask_cvtepu16_epi64(__m128i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_cvtepu16_epi64(__A),
+ (__v2di)__W);
+ }
+
+ static __inline__ __m128i __DEFAULT_FN_ATTRS128
+ _mm_maskz_cvtepu16_epi64(__mmask8 __U, __m128i __A)
+ {
+ return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
+ (__v2di)_mm_cvtepu16_epi64(__A),
+ (__v2di)_mm_setzero_si128());
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_mask_cvtepu16_epi64(__m256i __W, __mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_cvtepu16_epi64(__A),
+ (__v4di)__W);
+ }
+
+ static __inline__ __m256i __DEFAULT_FN_ATTRS256
+ _mm256_maskz_cvtepu16_epi64(__mmask8 __U, __m128i __A)
+ {
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
+ (__v4di)_mm256_cvtepu16_epi64(__A),
+ (__v4di)_mm256_setzero_si256());
+ }
+
+
+#define _mm_rol_epi32(a, b) \
+ (__m128i)__builtin_ia32_prold128((__v4si)(__m128i)(a), (int)(b))
+
+#define _mm_mask_rol_epi32(w, u, a, b) \
+ (__m128i)__builtin_ia32_selectd_128((__mmask8)(u), \
+ (__v4si)_mm_rol_epi32((a), (b)), \
+ (__v4si)(__m128i)(w))
+
+#define _mm_maskz_rol_epi32(u, a, b) \
+ (__m128i)__builtin_ia32_selectd_128((__mmask8)(u), \
+ (__v4si)_mm_rol_epi32((a), (b)), \
+ (__v4si)_mm_setzero_si128())
+
+#define _mm256_rol_epi32(a, b) \
+ (__m256i)__builtin_ia32_prold256((__v8si)(__m256i)(a), (int)(b))
+
+#define _mm256_mask_rol_epi32(w, u, a, b) \
+ (__m256i)__builtin_ia32_selectd_256((__mmask8)(u), \
+ (__v8si)_mm256_rol_epi32((a), (b)), \
+ (__v8si)(__m256i)(w))
+
+#define _mm256_maskz_rol_epi32(u, a, b) \
+ (__m256i)__builtin_ia32_selectd_256((__mmask8)(u), \
+ (__v8si)_mm256_rol_epi32((a), (b)), \
+ (__v8si)_mm256_setzero_si256())
+
+#define _mm_rol_epi64(a, b) \
+ (__m128i)__builtin_ia32_prolq128((__v2di)(__m128i)(a), (int)(b))
+
+#define _mm_mask_rol_epi64(w, u, a, b) \
+ (__m128i)__builtin_ia32_selectq_128((__mmask8)(u), \
+ (__v2di)_mm_rol_epi64((a), (b)), \
+ (__v2di)(__m128i)(w))
+
+#define _mm_maskz_rol_epi64(u, a, b) \
+ (__m128i)__builtin_ia32_selectq_128((__mmask8)(u), \
+ (__v2di)_mm_rol_epi64((a), (b)), \
+ (__v2di)_mm_setzero_si128())
+
+#define _mm256_rol_epi64(a, b) \
+ (__m256i)__builtin_ia32_prolq256((__v4di)(__m256i)(a), (int)(b))
+
+#define _mm256_mask_rol_epi64(w, u, a, b) \
+ (__m256i)__builtin_ia32_selectq_256((__mmask8)(u), \
+ (__v4di)_mm256_rol_epi64((a), (b)), \
+ (__v4di)(__m256i)(w))
+
+#define _mm256_maskz_rol_epi64(u, a, b) \
+ (__m256i)__builtin_ia32_selectq_256((__mmask8)(u), \
+ (__v4di)_mm256_rol_epi64((a), (b)), \
+ (__v4di)_mm256_setzero_si256())
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_rolv_epi32 (__m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_prolvd128_mask ((__v4si) __A,
- (__v4si) __B,
- (__v4si)
- _mm_setzero_si128 (),
- (__mmask8) -1);
+ return (__m128i)__builtin_ia32_prolvd128((__v4si)__A, (__v4si)__B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_rolv_epi32 (__m128i __W, __mmask8 __U, __m128i __A,
- __m128i __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_rolv_epi32 (__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_prolvd128_mask ((__v4si) __A,
- (__v4si) __B,
- (__v4si) __W,
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_rolv_epi32(__A, __B),
+ (__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_rolv_epi32 (__mmask8 __U, __m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_prolvd128_mask ((__v4si) __A,
- (__v4si) __B,
- (__v4si)
- _mm_setzero_si128 (),
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_rolv_epi32(__A, __B),
+ (__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_rolv_epi32 (__m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_prolvd256_mask ((__v8si) __A,
- (__v8si) __B,
- (__v8si)
- _mm256_setzero_si256 (),
- (__mmask8) -1);
+ return (__m256i)__builtin_ia32_prolvd256((__v8si)__A, (__v8si)__B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_rolv_epi32 (__m256i __W, __mmask8 __U, __m256i __A,
- __m256i __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_rolv_epi32 (__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_prolvd256_mask ((__v8si) __A,
- (__v8si) __B,
- (__v8si) __W,
- (__mmask8) __U);
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_rolv_epi32(__A, __B),
+ (__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_rolv_epi32 (__mmask8 __U, __m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_prolvd256_mask ((__v8si) __A,
- (__v8si) __B,
- (__v8si)
- _mm256_setzero_si256 (),
- (__mmask8) __U);
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_rolv_epi32(__A, __B),
+ (__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_rolv_epi64 (__m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_prolvq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di)
- _mm_setzero_di (),
- (__mmask8) -1);
+ return (__m128i)__builtin_ia32_prolvq128((__v2di)__A, (__v2di)__B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_rolv_epi64 (__m128i __W, __mmask8 __U, __m128i __A,
- __m128i __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_rolv_epi64 (__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_prolvq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di) __W,
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectq_128(__U,
+ (__v2di)_mm_rolv_epi64(__A, __B),
+ (__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_rolv_epi64 (__mmask8 __U, __m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_prolvq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di)
- _mm_setzero_di (),
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectq_128(__U,
+ (__v2di)_mm_rolv_epi64(__A, __B),
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_rolv_epi64 (__m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_prolvq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di)
- _mm256_setzero_si256 (),
- (__mmask8) -1);
+ return (__m256i)__builtin_ia32_prolvq256((__v4di)__A, (__v4di)__B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_rolv_epi64 (__m256i __W, __mmask8 __U, __m256i __A,
- __m256i __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_rolv_epi64 (__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_prolvq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di) __W,
- (__mmask8) __U);
+ return (__m256i)__builtin_ia32_selectq_256(__U,
+ (__v4di)_mm256_rolv_epi64(__A, __B),
+ (__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_rolv_epi64 (__mmask8 __U, __m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_prolvq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di)
- _mm256_setzero_si256 (),
- (__mmask8) __U);
+ return (__m256i)__builtin_ia32_selectq_256(__U,
+ (__v4di)_mm256_rolv_epi64(__A, __B),
+ (__v4di)_mm256_setzero_si256());
}
-#define _mm_ror_epi32(A, B) __extension__ ({ \
- (__m128i)__builtin_ia32_prord128_mask((__v4si)(__m128i)(A), (int)(B), \
- (__v4si)_mm_setzero_si128(), \
- (__mmask8)-1); })
+#define _mm_ror_epi32(a, b) \
+ (__m128i)__builtin_ia32_prord128((__v4si)(__m128i)(a), (int)(b))
-#define _mm_mask_ror_epi32(W, U, A, B) __extension__ ({ \
- (__m128i)__builtin_ia32_prord128_mask((__v4si)(__m128i)(A), (int)(B), \
- (__v4si)(__m128i)(W), (__mmask8)(U)); })
+#define _mm_mask_ror_epi32(w, u, a, b) \
+ (__m128i)__builtin_ia32_selectd_128((__mmask8)(u), \
+ (__v4si)_mm_ror_epi32((a), (b)), \
+ (__v4si)(__m128i)(w))
-#define _mm_maskz_ror_epi32(U, A, B) __extension__ ({ \
- (__m128i)__builtin_ia32_prord128_mask((__v4si)(__m128i)(A), (int)(B), \
- (__v4si)_mm_setzero_si128(), \
- (__mmask8)(U)); })
+#define _mm_maskz_ror_epi32(u, a, b) \
+ (__m128i)__builtin_ia32_selectd_128((__mmask8)(u), \
+ (__v4si)_mm_ror_epi32((a), (b)), \
+ (__v4si)_mm_setzero_si128())
-#define _mm256_ror_epi32(A, B) __extension__ ({ \
- (__m256i)__builtin_ia32_prord256_mask((__v8si)(__m256i)(A), (int)(B), \
- (__v8si)_mm256_setzero_si256(), \
- (__mmask8)-1); })
+#define _mm256_ror_epi32(a, b) \
+ (__m256i)__builtin_ia32_prord256((__v8si)(__m256i)(a), (int)(b))
-#define _mm256_mask_ror_epi32(W, U, A, B) __extension__ ({ \
- (__m256i)__builtin_ia32_prord256_mask((__v8si)(__m256i)(A), (int)(B), \
- (__v8si)(__m256i)(W), (__mmask8)(U)); })
+#define _mm256_mask_ror_epi32(w, u, a, b) \
+ (__m256i)__builtin_ia32_selectd_256((__mmask8)(u), \
+ (__v8si)_mm256_ror_epi32((a), (b)), \
+ (__v8si)(__m256i)(w))
-#define _mm256_maskz_ror_epi32(U, A, B) __extension__ ({ \
- (__m256i)__builtin_ia32_prord256_mask((__v8si)(__m256i)(A), (int)(B), \
- (__v8si)_mm256_setzero_si256(), \
- (__mmask8)(U)); })
+#define _mm256_maskz_ror_epi32(u, a, b) \
+ (__m256i)__builtin_ia32_selectd_256((__mmask8)(u), \
+ (__v8si)_mm256_ror_epi32((a), (b)), \
+ (__v8si)_mm256_setzero_si256())
-#define _mm_ror_epi64(A, B) __extension__ ({ \
- (__m128i)__builtin_ia32_prorq128_mask((__v2di)(__m128i)(A), (int)(B), \
- (__v2di)_mm_setzero_di(), \
- (__mmask8)-1); })
+#define _mm_ror_epi64(a, b) \
+ (__m128i)__builtin_ia32_prorq128((__v2di)(__m128i)(a), (int)(b))
-#define _mm_mask_ror_epi64(W, U, A, B) __extension__ ({ \
- (__m128i)__builtin_ia32_prorq128_mask((__v2di)(__m128i)(A), (int)(B), \
- (__v2di)(__m128i)(W), (__mmask8)(U)); })
+#define _mm_mask_ror_epi64(w, u, a, b) \
+ (__m128i)__builtin_ia32_selectq_128((__mmask8)(u), \
+ (__v2di)_mm_ror_epi64((a), (b)), \
+ (__v2di)(__m128i)(w))
-#define _mm_maskz_ror_epi64(U, A, B) __extension__ ({ \
- (__m128i)__builtin_ia32_prorq128_mask((__v2di)(__m128i)(A), (int)(B), \
- (__v2di)_mm_setzero_di(), \
- (__mmask8)(U)); })
+#define _mm_maskz_ror_epi64(u, a, b) \
+ (__m128i)__builtin_ia32_selectq_128((__mmask8)(u), \
+ (__v2di)_mm_ror_epi64((a), (b)), \
+ (__v2di)_mm_setzero_si128())
-#define _mm256_ror_epi64(A, B) __extension__ ({ \
- (__m256i)__builtin_ia32_prorq256_mask((__v4di)(__m256i)(A), (int)(B), \
- (__v4di)_mm256_setzero_si256(), \
- (__mmask8)-1); })
+#define _mm256_ror_epi64(a, b) \
+ (__m256i)__builtin_ia32_prorq256((__v4di)(__m256i)(a), (int)(b))
-#define _mm256_mask_ror_epi64(W, U, A, B) __extension__ ({ \
- (__m256i)__builtin_ia32_prorq256_mask((__v4di)(__m256i)(A), (int)(B), \
- (__v4di)(__m256i)(W), (__mmask8)(U)); })
+#define _mm256_mask_ror_epi64(w, u, a, b) \
+ (__m256i)__builtin_ia32_selectq_256((__mmask8)(u), \
+ (__v4di)_mm256_ror_epi64((a), (b)), \
+ (__v4di)(__m256i)(w))
-#define _mm256_maskz_ror_epi64(U, A, B) __extension__ ({ \
- (__m256i)__builtin_ia32_prorq256_mask((__v4di)(__m256i)(A), (int)(B), \
- (__v4di)_mm256_setzero_si256(), \
- (__mmask8)(U)); })
+#define _mm256_maskz_ror_epi64(u, a, b) \
+ (__m256i)__builtin_ia32_selectq_256((__mmask8)(u), \
+ (__v4di)_mm256_ror_epi64((a), (b)), \
+ (__v4di)_mm256_setzero_si256())
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_sll_epi32(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_sll_epi32(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_sll_epi32(__m256i __W, __mmask8 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_sll_epi32(__mmask8 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_slli_epi32(__m128i __W, __mmask8 __U, __m128i __A, int __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_slli_epi32(__m128i __W, __mmask8 __U, __m128i __A, unsigned int __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_slli_epi32(__A, __B),
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_slli_epi32(__mmask8 __U, __m128i __A, int __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_slli_epi32(__mmask8 __U, __m128i __A, unsigned int __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_slli_epi32(__A, __B),
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_slli_epi32(__m256i __W, __mmask8 __U, __m256i __A, int __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_slli_epi32(__m256i __W, __mmask8 __U, __m256i __A, unsigned int __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_slli_epi32(__A, __B),
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_slli_epi32(__mmask8 __U, __m256i __A, int __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_slli_epi32(__mmask8 __U, __m256i __A, unsigned int __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_slli_epi32(__A, __B),
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_sll_epi64(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_sll_epi64(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)_mm_sll_epi64(__A, __B),
- (__v2di)_mm_setzero_di());
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_sll_epi64(__m256i __W, __mmask8 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_sll_epi64(__mmask8 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_slli_epi64(__m128i __W, __mmask8 __U, __m128i __A, int __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_slli_epi64(__m128i __W, __mmask8 __U, __m128i __A, unsigned int __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)_mm_slli_epi64(__A, __B),
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_slli_epi64(__mmask8 __U, __m128i __A, int __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_slli_epi64(__mmask8 __U, __m128i __A, unsigned int __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)_mm_slli_epi64(__A, __B),
- (__v2di)_mm_setzero_di());
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_slli_epi64(__m256i __W, __mmask8 __U, __m256i __A, int __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_slli_epi64(__m256i __W, __mmask8 __U, __m256i __A, unsigned int __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)_mm256_slli_epi64(__A, __B),
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_slli_epi64(__mmask8 __U, __m256i __A, int __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_slli_epi64(__mmask8 __U, __m256i __A, unsigned int __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)_mm256_slli_epi64(__A, __B),
(__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_rorv_epi32 (__m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_prorvd128_mask ((__v4si) __A,
- (__v4si) __B,
- (__v4si)
- _mm_setzero_si128 (),
- (__mmask8) -1);
+ return (__m128i)__builtin_ia32_prorvd128((__v4si)__A, (__v4si)__B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_rorv_epi32 (__m128i __W, __mmask8 __U, __m128i __A,
- __m128i __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_rorv_epi32 (__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_prorvd128_mask ((__v4si) __A,
- (__v4si) __B,
- (__v4si) __W,
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_rorv_epi32(__A, __B),
+ (__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_rorv_epi32 (__mmask8 __U, __m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_prorvd128_mask ((__v4si) __A,
- (__v4si) __B,
- (__v4si)
- _mm_setzero_si128 (),
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_rorv_epi32(__A, __B),
+ (__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_rorv_epi32 (__m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_prorvd256_mask ((__v8si) __A,
- (__v8si) __B,
- (__v8si)
- _mm256_setzero_si256 (),
- (__mmask8) -1);
+ return (__m256i)__builtin_ia32_prorvd256((__v8si)__A, (__v8si)__B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_rorv_epi32 (__m256i __W, __mmask8 __U, __m256i __A,
- __m256i __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_rorv_epi32 (__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_prorvd256_mask ((__v8si) __A,
- (__v8si) __B,
- (__v8si) __W,
- (__mmask8) __U);
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_rorv_epi32(__A, __B),
+ (__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_rorv_epi32 (__mmask8 __U, __m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_prorvd256_mask ((__v8si) __A,
- (__v8si) __B,
- (__v8si)
- _mm256_setzero_si256 (),
- (__mmask8) __U);
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_rorv_epi32(__A, __B),
+ (__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_rorv_epi64 (__m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_prorvq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di)
- _mm_setzero_di (),
- (__mmask8) -1);
+ return (__m128i)__builtin_ia32_prorvq128((__v2di)__A, (__v2di)__B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_rorv_epi64 (__m128i __W, __mmask8 __U, __m128i __A,
- __m128i __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_rorv_epi64 (__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_prorvq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di) __W,
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectq_128(__U,
+ (__v2di)_mm_rorv_epi64(__A, __B),
+ (__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_rorv_epi64 (__mmask8 __U, __m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_prorvq128_mask ((__v2di) __A,
- (__v2di) __B,
- (__v2di)
- _mm_setzero_di (),
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectq_128(__U,
+ (__v2di)_mm_rorv_epi64(__A, __B),
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_rorv_epi64 (__m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_prorvq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di)
- _mm256_setzero_si256 (),
- (__mmask8) -1);
+ return (__m256i)__builtin_ia32_prorvq256((__v4di)__A, (__v4di)__B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_rorv_epi64 (__m256i __W, __mmask8 __U, __m256i __A,
- __m256i __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_rorv_epi64 (__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_prorvq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di) __W,
- (__mmask8) __U);
+ return (__m256i)__builtin_ia32_selectq_256(__U,
+ (__v4di)_mm256_rorv_epi64(__A, __B),
+ (__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_rorv_epi64 (__mmask8 __U, __m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_prorvq256_mask ((__v4di) __A,
- (__v4di) __B,
- (__v4di)
- _mm256_setzero_si256 (),
- (__mmask8) __U);
+ return (__m256i)__builtin_ia32_selectq_256(__U,
+ (__v4di)_mm256_rorv_epi64(__A, __B),
+ (__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_sllv_epi64(__m128i __W, __mmask8 __U, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_sllv_epi64(__mmask8 __U, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)_mm_sllv_epi64(__X, __Y),
- (__v2di)_mm_setzero_di());
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_sllv_epi64(__m256i __W, __mmask8 __U, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_sllv_epi64(__mmask8 __U, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_sllv_epi32(__m128i __W, __mmask8 __U, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_sllv_epi32(__mmask8 __U, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_sllv_epi32(__m256i __W, __mmask8 __U, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_sllv_epi32(__mmask8 __U, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_srlv_epi64(__m128i __W, __mmask8 __U, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_srlv_epi64(__mmask8 __U, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)_mm_srlv_epi64(__X, __Y),
- (__v2di)_mm_setzero_di());
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_srlv_epi64(__m256i __W, __mmask8 __U, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_srlv_epi64(__mmask8 __U, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_srlv_epi32(__m128i __W, __mmask8 __U, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_srlv_epi32(__mmask8 __U, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_srlv_epi32(__m256i __W, __mmask8 __U, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_srlv_epi32(__mmask8 __U, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_srl_epi32(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_srl_epi32(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_srl_epi32(__m256i __W, __mmask8 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_srl_epi32(__mmask8 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_srli_epi32(__m128i __W, __mmask8 __U, __m128i __A, int __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_srli_epi32(__m128i __W, __mmask8 __U, __m128i __A, unsigned int __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_srli_epi32(__A, __B),
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_srli_epi32(__mmask8 __U, __m128i __A, int __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_srli_epi32(__mmask8 __U, __m128i __A, unsigned int __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_srli_epi32(__A, __B),
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_srli_epi32(__m256i __W, __mmask8 __U, __m256i __A, int __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_srli_epi32(__m256i __W, __mmask8 __U, __m256i __A, unsigned int __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_srli_epi32(__A, __B),
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_srli_epi32(__mmask8 __U, __m256i __A, int __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_srli_epi32(__mmask8 __U, __m256i __A, unsigned int __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_srli_epi32(__A, __B),
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_srl_epi64(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_srl_epi64(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)_mm_srl_epi64(__A, __B),
- (__v2di)_mm_setzero_di());
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_srl_epi64(__m256i __W, __mmask8 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_srl_epi64(__mmask8 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_srli_epi64(__m128i __W, __mmask8 __U, __m128i __A, int __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_srli_epi64(__m128i __W, __mmask8 __U, __m128i __A, unsigned int __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)_mm_srli_epi64(__A, __B),
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_srli_epi64(__mmask8 __U, __m128i __A, int __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_srli_epi64(__mmask8 __U, __m128i __A, unsigned int __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)_mm_srli_epi64(__A, __B),
- (__v2di)_mm_setzero_di());
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_srli_epi64(__m256i __W, __mmask8 __U, __m256i __A, int __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_srli_epi64(__m256i __W, __mmask8 __U, __m256i __A, unsigned int __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)_mm256_srli_epi64(__A, __B),
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_srli_epi64(__mmask8 __U, __m256i __A, int __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_srli_epi64(__mmask8 __U, __m256i __A, unsigned int __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)_mm256_srli_epi64(__A, __B),
(__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_srav_epi32(__m128i __W, __mmask8 __U, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_srav_epi32(__mmask8 __U, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_srav_epi32(__m256i __W, __mmask8 __U, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_srav_epi32(__mmask8 __U, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_srav_epi64(__m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_psravq128((__v2di)__X, (__v2di)__Y);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_srav_epi64(__m128i __W, __mmask8 __U, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_srav_epi64(__mmask8 __U, __m128i __X, __m128i __Y)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)_mm_srav_epi64(__X, __Y),
- (__v2di)_mm_setzero_di());
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_srav_epi64(__m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_psravq256((__v4di)__X, (__v4di) __Y);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_srav_epi64(__m256i __W, __mmask8 __U, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_srav_epi64 (__mmask8 __U, __m256i __X, __m256i __Y)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_mov_epi32 (__m128i __W, __mmask8 __U, __m128i __A)
{
return (__m128i) __builtin_ia32_selectd_128 ((__mmask8) __U,
(__v4si) __W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_mov_epi32 (__mmask8 __U, __m128i __A)
{
return (__m128i) __builtin_ia32_selectd_128 ((__mmask8) __U,
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_mov_epi32 (__m256i __W, __mmask8 __U, __m256i __A)
{
return (__m256i) __builtin_ia32_selectd_256 ((__mmask8) __U,
(__v8si) __W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_mov_epi32 (__mmask8 __U, __m256i __A)
{
return (__m256i) __builtin_ia32_selectd_256 ((__mmask8) __U,
(__v8si) _mm256_setzero_si256 ());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline __m128i __DEFAULT_FN_ATTRS128
+_mm_load_epi32 (void const *__P)
+{
+ return *(const __m128i *) __P;
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_load_epi32 (__m128i __W, __mmask8 __U, void const *__P)
{
- return (__m128i) __builtin_ia32_movdqa32load128_mask ((__v4si *) __P,
+ return (__m128i) __builtin_ia32_movdqa32load128_mask ((const __v4si *) __P,
(__v4si) __W,
(__mmask8)
__U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_load_epi32 (__mmask8 __U, void const *__P)
{
- return (__m128i) __builtin_ia32_movdqa32load128_mask ((__v4si *) __P,
+ return (__m128i) __builtin_ia32_movdqa32load128_mask ((const __v4si *) __P,
(__v4si)
_mm_setzero_si128 (),
(__mmask8)
__U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline __m256i __DEFAULT_FN_ATTRS256
+_mm256_load_epi32 (void const *__P)
+{
+ return *(const __m256i *) __P;
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_load_epi32 (__m256i __W, __mmask8 __U, void const *__P)
{
- return (__m256i) __builtin_ia32_movdqa32load256_mask ((__v8si *) __P,
+ return (__m256i) __builtin_ia32_movdqa32load256_mask ((const __v8si *) __P,
(__v8si) __W,
(__mmask8)
__U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_load_epi32 (__mmask8 __U, void const *__P)
{
- return (__m256i) __builtin_ia32_movdqa32load256_mask ((__v8si *) __P,
+ return (__m256i) __builtin_ia32_movdqa32load256_mask ((const __v8si *) __P,
(__v8si)
_mm256_setzero_si256 (),
(__mmask8)
__U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS128
+_mm_store_epi32 (void *__P, __m128i __A)
+{
+ *(__m128i *) __P = __A;
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_store_epi32 (void *__P, __mmask8 __U, __m128i __A)
{
__builtin_ia32_movdqa32store128_mask ((__v4si *) __P,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS256
+_mm256_store_epi32 (void *__P, __m256i __A)
+{
+ *(__m256i *) __P = __A;
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_store_epi32 (void *__P, __mmask8 __U, __m256i __A)
{
__builtin_ia32_movdqa32store256_mask ((__v8si *) __P,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_mov_epi64 (__m128i __W, __mmask8 __U, __m128i __A)
{
return (__m128i) __builtin_ia32_selectq_128 ((__mmask8) __U,
(__v2di) __W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_mov_epi64 (__mmask8 __U, __m128i __A)
{
return (__m128i) __builtin_ia32_selectq_128 ((__mmask8) __U,
(__v2di) __A,
- (__v2di) _mm_setzero_di ());
+ (__v2di) _mm_setzero_si128 ());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_mov_epi64 (__m256i __W, __mmask8 __U, __m256i __A)
{
return (__m256i) __builtin_ia32_selectq_256 ((__mmask8) __U,
(__v4di) __W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_mov_epi64 (__mmask8 __U, __m256i __A)
{
return (__m256i) __builtin_ia32_selectq_256 ((__mmask8) __U,
(__v4di) _mm256_setzero_si256 ());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline __m128i __DEFAULT_FN_ATTRS128
+_mm_load_epi64 (void const *__P)
+{
+ return *(const __m128i *) __P;
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_load_epi64 (__m128i __W, __mmask8 __U, void const *__P)
{
- return (__m128i) __builtin_ia32_movdqa64load128_mask ((__v2di *) __P,
+ return (__m128i) __builtin_ia32_movdqa64load128_mask ((const __v2di *) __P,
(__v2di) __W,
(__mmask8)
__U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_load_epi64 (__mmask8 __U, void const *__P)
{
- return (__m128i) __builtin_ia32_movdqa64load128_mask ((__v2di *) __P,
+ return (__m128i) __builtin_ia32_movdqa64load128_mask ((const __v2di *) __P,
(__v2di)
- _mm_setzero_di (),
+ _mm_setzero_si128 (),
(__mmask8)
__U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline __m256i __DEFAULT_FN_ATTRS256
+_mm256_load_epi64 (void const *__P)
+{
+ return *(const __m256i *) __P;
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_load_epi64 (__m256i __W, __mmask8 __U, void const *__P)
{
- return (__m256i) __builtin_ia32_movdqa64load256_mask ((__v4di *) __P,
+ return (__m256i) __builtin_ia32_movdqa64load256_mask ((const __v4di *) __P,
(__v4di) __W,
(__mmask8)
__U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_load_epi64 (__mmask8 __U, void const *__P)
{
- return (__m256i) __builtin_ia32_movdqa64load256_mask ((__v4di *) __P,
+ return (__m256i) __builtin_ia32_movdqa64load256_mask ((const __v4di *) __P,
(__v4di)
_mm256_setzero_si256 (),
(__mmask8)
__U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS128
+_mm_store_epi64 (void *__P, __m128i __A)
+{
+ *(__m128i *) __P = __A;
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_store_epi64 (void *__P, __mmask8 __U, __m128i __A)
{
__builtin_ia32_movdqa64store128_mask ((__v2di *) __P,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS256
+_mm256_store_epi64 (void *__P, __m256i __A)
+{
+ *(__m256i *) __P = __A;
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_store_epi64 (void *__P, __mmask8 __U, __m256i __A)
{
__builtin_ia32_movdqa64store256_mask ((__v4di *) __P,
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_movedup_pd (__m128d __W, __mmask8 __U, __m128d __A)
{
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_movedup_pd (__mmask8 __U, __m128d __A)
{
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_movedup_pd (__m256d __W, __mmask8 __U, __m256d __A)
{
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_movedup_pd (__mmask8 __U, __m256d __A)
{
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_setzero_pd());
}
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_set1_epi32(__m128i __O, __mmask8 __M, int __A)
+{
+ return (__m128i)__builtin_ia32_selectd_128(__M,
+ (__v4si) _mm_set1_epi32(__A),
+ (__v4si)__O);
+}
-#define _mm_mask_set1_epi32(O, M, A) __extension__ ({ \
- (__m128i)__builtin_ia32_pbroadcastd128_gpr_mask((int)(A), \
- (__v4si)(__m128i)(O), \
- (__mmask8)(M)); })
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_set1_epi32( __mmask8 __M, int __A)
+{
+ return (__m128i)__builtin_ia32_selectd_128(__M,
+ (__v4si) _mm_set1_epi32(__A),
+ (__v4si)_mm_setzero_si128());
+}
-#define _mm_maskz_set1_epi32(M, A) __extension__ ({ \
- (__m128i)__builtin_ia32_pbroadcastd128_gpr_mask((int)(A), \
- (__v4si)_mm_setzero_si128(), \
- (__mmask8)(M)); })
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_set1_epi32(__m256i __O, __mmask8 __M, int __A)
+{
+ return (__m256i)__builtin_ia32_selectd_256(__M,
+ (__v8si) _mm256_set1_epi32(__A),
+ (__v8si)__O);
+}
-#define _mm256_mask_set1_epi32(O, M, A) __extension__ ({ \
- (__m256i)__builtin_ia32_pbroadcastd256_gpr_mask((int)(A), \
- (__v8si)(__m256i)(O), \
- (__mmask8)(M)); })
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_set1_epi32( __mmask8 __M, int __A)
+{
+ return (__m256i)__builtin_ia32_selectd_256(__M,
+ (__v8si) _mm256_set1_epi32(__A),
+ (__v8si)_mm256_setzero_si256());
+}
-#define _mm256_maskz_set1_epi32(M, A) __extension__ ({ \
- (__m256i)__builtin_ia32_pbroadcastd256_gpr_mask((int)(A), \
- (__v8si)_mm256_setzero_si256(), \
- (__mmask8)(M)); })
-#ifdef __x86_64__
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_set1_epi64 (__m128i __O, __mmask8 __M, long long __A)
{
- return (__m128i) __builtin_ia32_pbroadcastq128_gpr_mask (__A, (__v2di) __O,
- __M);
+ return (__m128i) __builtin_ia32_selectq_128(__M,
+ (__v2di) _mm_set1_epi64x(__A),
+ (__v2di) __O);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_set1_epi64 (__mmask8 __M, long long __A)
{
- return (__m128i) __builtin_ia32_pbroadcastq128_gpr_mask (__A,
- (__v2di)
- _mm_setzero_si128 (),
- __M);
+ return (__m128i) __builtin_ia32_selectq_128(__M,
+ (__v2di) _mm_set1_epi64x(__A),
+ (__v2di) _mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_set1_epi64 (__m256i __O, __mmask8 __M, long long __A)
{
- return (__m256i) __builtin_ia32_pbroadcastq256_gpr_mask (__A, (__v4di) __O,
- __M);
+ return (__m256i) __builtin_ia32_selectq_256(__M,
+ (__v4di) _mm256_set1_epi64x(__A),
+ (__v4di) __O) ;
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_set1_epi64 (__mmask8 __M, long long __A)
{
- return (__m256i) __builtin_ia32_pbroadcastq256_gpr_mask (__A,
- (__v4di)
- _mm256_setzero_si256 (),
- __M);
+ return (__m256i) __builtin_ia32_selectq_256(__M,
+ (__v4di) _mm256_set1_epi64x(__A),
+ (__v4di) _mm256_setzero_si256());
}
-#endif
-#define _mm_fixupimm_pd(A, B, C, imm) __extension__ ({ \
+#define _mm_fixupimm_pd(A, B, C, imm) \
(__m128d)__builtin_ia32_fixupimmpd128_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2di)(__m128i)(C), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_fixupimm_pd(A, U, B, C, imm) __extension__ ({ \
+#define _mm_mask_fixupimm_pd(A, U, B, C, imm) \
(__m128d)__builtin_ia32_fixupimmpd128_mask((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2di)(__m128i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_maskz_fixupimm_pd(U, A, B, C, imm) __extension__ ({ \
+#define _mm_maskz_fixupimm_pd(U, A, B, C, imm) \
(__m128d)__builtin_ia32_fixupimmpd128_maskz((__v2df)(__m128d)(A), \
(__v2df)(__m128d)(B), \
(__v2di)(__m128i)(C), \
- (int)(imm), (__mmask8)(U)); })
+ (int)(imm), (__mmask8)(U))
-#define _mm256_fixupimm_pd(A, B, C, imm) __extension__ ({ \
+#define _mm256_fixupimm_pd(A, B, C, imm) \
(__m256d)__builtin_ia32_fixupimmpd256_mask((__v4df)(__m256d)(A), \
(__v4df)(__m256d)(B), \
(__v4di)(__m256i)(C), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_fixupimm_pd(A, U, B, C, imm) __extension__ ({ \
+#define _mm256_mask_fixupimm_pd(A, U, B, C, imm) \
(__m256d)__builtin_ia32_fixupimmpd256_mask((__v4df)(__m256d)(A), \
(__v4df)(__m256d)(B), \
(__v4di)(__m256i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_maskz_fixupimm_pd(U, A, B, C, imm) __extension__ ({ \
+#define _mm256_maskz_fixupimm_pd(U, A, B, C, imm) \
(__m256d)__builtin_ia32_fixupimmpd256_maskz((__v4df)(__m256d)(A), \
(__v4df)(__m256d)(B), \
(__v4di)(__m256i)(C), \
- (int)(imm), (__mmask8)(U)); })
+ (int)(imm), (__mmask8)(U))
-#define _mm_fixupimm_ps(A, B, C, imm) __extension__ ({ \
+#define _mm_fixupimm_ps(A, B, C, imm) \
(__m128)__builtin_ia32_fixupimmps128_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4si)(__m128i)(C), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_fixupimm_ps(A, U, B, C, imm) __extension__ ({ \
+#define _mm_mask_fixupimm_ps(A, U, B, C, imm) \
(__m128)__builtin_ia32_fixupimmps128_mask((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4si)(__m128i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_maskz_fixupimm_ps(U, A, B, C, imm) __extension__ ({ \
+#define _mm_maskz_fixupimm_ps(U, A, B, C, imm) \
(__m128)__builtin_ia32_fixupimmps128_maskz((__v4sf)(__m128)(A), \
(__v4sf)(__m128)(B), \
(__v4si)(__m128i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_fixupimm_ps(A, B, C, imm) __extension__ ({ \
+#define _mm256_fixupimm_ps(A, B, C, imm) \
(__m256)__builtin_ia32_fixupimmps256_mask((__v8sf)(__m256)(A), \
(__v8sf)(__m256)(B), \
(__v8si)(__m256i)(C), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_fixupimm_ps(A, U, B, C, imm) __extension__ ({ \
+#define _mm256_mask_fixupimm_ps(A, U, B, C, imm) \
(__m256)__builtin_ia32_fixupimmps256_mask((__v8sf)(__m256)(A), \
(__v8sf)(__m256)(B), \
(__v8si)(__m256i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_maskz_fixupimm_ps(U, A, B, C, imm) __extension__ ({ \
+#define _mm256_maskz_fixupimm_ps(U, A, B, C, imm) \
(__m256)__builtin_ia32_fixupimmps256_maskz((__v8sf)(__m256)(A), \
(__v8sf)(__m256)(B), \
(__v8si)(__m256i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_load_pd (__m128d __W, __mmask8 __U, void const *__P)
{
- return (__m128d) __builtin_ia32_loadapd128_mask ((__v2df *) __P,
+ return (__m128d) __builtin_ia32_loadapd128_mask ((const __v2df *) __P,
(__v2df) __W,
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_load_pd (__mmask8 __U, void const *__P)
{
- return (__m128d) __builtin_ia32_loadapd128_mask ((__v2df *) __P,
+ return (__m128d) __builtin_ia32_loadapd128_mask ((const __v2df *) __P,
(__v2df)
_mm_setzero_pd (),
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_load_pd (__m256d __W, __mmask8 __U, void const *__P)
{
- return (__m256d) __builtin_ia32_loadapd256_mask ((__v4df *) __P,
+ return (__m256d) __builtin_ia32_loadapd256_mask ((const __v4df *) __P,
(__v4df) __W,
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_load_pd (__mmask8 __U, void const *__P)
{
- return (__m256d) __builtin_ia32_loadapd256_mask ((__v4df *) __P,
+ return (__m256d) __builtin_ia32_loadapd256_mask ((const __v4df *) __P,
(__v4df)
_mm256_setzero_pd (),
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_load_ps (__m128 __W, __mmask8 __U, void const *__P)
{
- return (__m128) __builtin_ia32_loadaps128_mask ((__v4sf *) __P,
+ return (__m128) __builtin_ia32_loadaps128_mask ((const __v4sf *) __P,
(__v4sf) __W,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_load_ps (__mmask8 __U, void const *__P)
{
- return (__m128) __builtin_ia32_loadaps128_mask ((__v4sf *) __P,
+ return (__m128) __builtin_ia32_loadaps128_mask ((const __v4sf *) __P,
(__v4sf)
_mm_setzero_ps (),
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_load_ps (__m256 __W, __mmask8 __U, void const *__P)
{
- return (__m256) __builtin_ia32_loadaps256_mask ((__v8sf *) __P,
+ return (__m256) __builtin_ia32_loadaps256_mask ((const __v8sf *) __P,
(__v8sf) __W,
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_load_ps (__mmask8 __U, void const *__P)
{
- return (__m256) __builtin_ia32_loadaps256_mask ((__v8sf *) __P,
+ return (__m256) __builtin_ia32_loadaps256_mask ((const __v8sf *) __P,
(__v8sf)
_mm256_setzero_ps (),
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline __m128i __DEFAULT_FN_ATTRS128
+_mm_loadu_epi64 (void const *__P)
+{
+ struct __loadu_epi64 {
+ __m128i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ return ((const struct __loadu_epi64*)__P)->__v;
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_loadu_epi64 (__m128i __W, __mmask8 __U, void const *__P)
{
- return (__m128i) __builtin_ia32_loaddqudi128_mask ((__v2di *) __P,
+ return (__m128i) __builtin_ia32_loaddqudi128_mask ((const __v2di *) __P,
(__v2di) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_loadu_epi64 (__mmask8 __U, void const *__P)
{
- return (__m128i) __builtin_ia32_loaddqudi128_mask ((__v2di *) __P,
+ return (__m128i) __builtin_ia32_loaddqudi128_mask ((const __v2di *) __P,
(__v2di)
_mm_setzero_si128 (),
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline __m256i __DEFAULT_FN_ATTRS256
+_mm256_loadu_epi64 (void const *__P)
+{
+ struct __loadu_epi64 {
+ __m256i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ return ((const struct __loadu_epi64*)__P)->__v;
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_loadu_epi64 (__m256i __W, __mmask8 __U, void const *__P)
{
- return (__m256i) __builtin_ia32_loaddqudi256_mask ((__v4di *) __P,
+ return (__m256i) __builtin_ia32_loaddqudi256_mask ((const __v4di *) __P,
(__v4di) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_loadu_epi64 (__mmask8 __U, void const *__P)
{
- return (__m256i) __builtin_ia32_loaddqudi256_mask ((__v4di *) __P,
+ return (__m256i) __builtin_ia32_loaddqudi256_mask ((const __v4di *) __P,
(__v4di)
_mm256_setzero_si256 (),
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline __m128i __DEFAULT_FN_ATTRS128
+_mm_loadu_epi32 (void const *__P)
+{
+ struct __loadu_epi32 {
+ __m128i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ return ((const struct __loadu_epi32*)__P)->__v;
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_loadu_epi32 (__m128i __W, __mmask8 __U, void const *__P)
{
- return (__m128i) __builtin_ia32_loaddqusi128_mask ((__v4si *) __P,
+ return (__m128i) __builtin_ia32_loaddqusi128_mask ((const __v4si *) __P,
(__v4si) __W,
(__mmask8) __U);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_loadu_epi32 (__mmask8 __U, void const *__P)
{
- return (__m128i) __builtin_ia32_loaddqusi128_mask ((__v4si *) __P,
+ return (__m128i) __builtin_ia32_loaddqusi128_mask ((const __v4si *) __P,
(__v4si)
_mm_setzero_si128 (),
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline __m256i __DEFAULT_FN_ATTRS256
+_mm256_loadu_epi32 (void const *__P)
+{
+ struct __loadu_epi32 {
+ __m256i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ return ((const struct __loadu_epi32*)__P)->__v;
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_loadu_epi32 (__m256i __W, __mmask8 __U, void const *__P)
{
- return (__m256i) __builtin_ia32_loaddqusi256_mask ((__v8si *) __P,
+ return (__m256i) __builtin_ia32_loaddqusi256_mask ((const __v8si *) __P,
(__v8si) __W,
(__mmask8) __U);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_loadu_epi32 (__mmask8 __U, void const *__P)
{
- return (__m256i) __builtin_ia32_loaddqusi256_mask ((__v8si *) __P,
+ return (__m256i) __builtin_ia32_loaddqusi256_mask ((const __v8si *) __P,
(__v8si)
_mm256_setzero_si256 (),
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_loadu_pd (__m128d __W, __mmask8 __U, void const *__P)
{
- return (__m128d) __builtin_ia32_loadupd128_mask ((__v2df *) __P,
+ return (__m128d) __builtin_ia32_loadupd128_mask ((const __v2df *) __P,
(__v2df) __W,
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_loadu_pd (__mmask8 __U, void const *__P)
{
- return (__m128d) __builtin_ia32_loadupd128_mask ((__v2df *) __P,
+ return (__m128d) __builtin_ia32_loadupd128_mask ((const __v2df *) __P,
(__v2df)
_mm_setzero_pd (),
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_loadu_pd (__m256d __W, __mmask8 __U, void const *__P)
{
- return (__m256d) __builtin_ia32_loadupd256_mask ((__v4df *) __P,
+ return (__m256d) __builtin_ia32_loadupd256_mask ((const __v4df *) __P,
(__v4df) __W,
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_loadu_pd (__mmask8 __U, void const *__P)
{
- return (__m256d) __builtin_ia32_loadupd256_mask ((__v4df *) __P,
+ return (__m256d) __builtin_ia32_loadupd256_mask ((const __v4df *) __P,
(__v4df)
_mm256_setzero_pd (),
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_loadu_ps (__m128 __W, __mmask8 __U, void const *__P)
{
- return (__m128) __builtin_ia32_loadups128_mask ((__v4sf *) __P,
+ return (__m128) __builtin_ia32_loadups128_mask ((const __v4sf *) __P,
(__v4sf) __W,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_loadu_ps (__mmask8 __U, void const *__P)
{
- return (__m128) __builtin_ia32_loadups128_mask ((__v4sf *) __P,
+ return (__m128) __builtin_ia32_loadups128_mask ((const __v4sf *) __P,
(__v4sf)
_mm_setzero_ps (),
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_loadu_ps (__m256 __W, __mmask8 __U, void const *__P)
{
- return (__m256) __builtin_ia32_loadups256_mask ((__v8sf *) __P,
+ return (__m256) __builtin_ia32_loadups256_mask ((const __v8sf *) __P,
(__v8sf) __W,
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_loadu_ps (__mmask8 __U, void const *__P)
{
- return (__m256) __builtin_ia32_loadups256_mask ((__v8sf *) __P,
+ return (__m256) __builtin_ia32_loadups256_mask ((const __v8sf *) __P,
(__v8sf)
_mm256_setzero_ps (),
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_store_pd (void *__P, __mmask8 __U, __m128d __A)
{
__builtin_ia32_storeapd128_mask ((__v2df *) __P,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_store_pd (void *__P, __mmask8 __U, __m256d __A)
{
__builtin_ia32_storeapd256_mask ((__v4df *) __P,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_store_ps (void *__P, __mmask8 __U, __m128 __A)
{
__builtin_ia32_storeaps128_mask ((__v4sf *) __P,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_store_ps (void *__P, __mmask8 __U, __m256 __A)
{
__builtin_ia32_storeaps256_mask ((__v8sf *) __P,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS128
+_mm_storeu_epi64 (void *__P, __m128i __A)
+{
+ struct __storeu_epi64 {
+ __m128i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_epi64*)__P)->__v = __A;
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_storeu_epi64 (void *__P, __mmask8 __U, __m128i __A)
{
__builtin_ia32_storedqudi128_mask ((__v2di *) __P,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS256
+_mm256_storeu_epi64 (void *__P, __m256i __A)
+{
+ struct __storeu_epi64 {
+ __m256i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_epi64*)__P)->__v = __A;
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_storeu_epi64 (void *__P, __mmask8 __U, __m256i __A)
{
__builtin_ia32_storedqudi256_mask ((__v4di *) __P,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS128
+_mm_storeu_epi32 (void *__P, __m128i __A)
+{
+ struct __storeu_epi32 {
+ __m128i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_epi32*)__P)->__v = __A;
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_storeu_epi32 (void *__P, __mmask8 __U, __m128i __A)
{
__builtin_ia32_storedqusi128_mask ((__v4si *) __P,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS256
+_mm256_storeu_epi32 (void *__P, __m256i __A)
+{
+ struct __storeu_epi32 {
+ __m256i_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_epi32*)__P)->__v = __A;
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_storeu_epi32 (void *__P, __mmask8 __U, __m256i __A)
{
__builtin_ia32_storedqusi256_mask ((__v8si *) __P,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_storeu_pd (void *__P, __mmask8 __U, __m128d __A)
{
__builtin_ia32_storeupd128_mask ((__v2df *) __P,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_storeu_pd (void *__P, __mmask8 __U, __m256d __A)
{
__builtin_ia32_storeupd256_mask ((__v4df *) __P,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_storeu_ps (void *__P, __mmask8 __U, __m128 __A)
{
__builtin_ia32_storeups128_mask ((__v4sf *) __P,
(__mmask8) __U);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_storeu_ps (void *__P, __mmask8 __U, __m256 __A)
{
__builtin_ia32_storeups256_mask ((__v8sf *) __P,
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_unpackhi_pd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_unpackhi_pd(__mmask8 __U, __m128d __A, __m128d __B)
{
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_unpackhi_pd(__m256d __W, __mmask8 __U, __m256d __A, __m256d __B)
{
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_unpackhi_pd(__mmask8 __U, __m256d __A, __m256d __B)
{
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_unpackhi_ps(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_unpackhi_ps(__mmask8 __U, __m128 __A, __m128 __B)
{
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_unpackhi_ps(__m256 __W, __mmask8 __U, __m256 __A, __m256 __B)
{
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_unpackhi_ps(__mmask8 __U, __m256 __A, __m256 __B)
{
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_unpacklo_pd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_unpacklo_pd(__mmask8 __U, __m128d __A, __m128d __B)
{
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_unpacklo_pd(__m256d __W, __mmask8 __U, __m256d __A, __m256d __B)
{
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_unpacklo_pd(__mmask8 __U, __m256d __A, __m256d __B)
{
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_unpacklo_ps(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_unpacklo_ps(__mmask8 __U, __m128 __A, __m128 __B)
{
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_unpacklo_ps(__m256 __W, __mmask8 __U, __m256 __A, __m256 __B)
{
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_unpacklo_ps(__mmask8 __U, __m256 __A, __m256 __B)
{
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_rcp14_pd (__m128d __A)
{
return (__m128d) __builtin_ia32_rcp14pd128_mask ((__v2df) __A,
(__mmask8) -1);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_rcp14_pd (__m128d __W, __mmask8 __U, __m128d __A)
{
return (__m128d) __builtin_ia32_rcp14pd128_mask ((__v2df) __A,
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_rcp14_pd (__mmask8 __U, __m128d __A)
{
return (__m128d) __builtin_ia32_rcp14pd128_mask ((__v2df) __A,
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_rcp14_pd (__m256d __A)
{
return (__m256d) __builtin_ia32_rcp14pd256_mask ((__v4df) __A,
(__mmask8) -1);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_rcp14_pd (__m256d __W, __mmask8 __U, __m256d __A)
{
return (__m256d) __builtin_ia32_rcp14pd256_mask ((__v4df) __A,
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_rcp14_pd (__mmask8 __U, __m256d __A)
{
return (__m256d) __builtin_ia32_rcp14pd256_mask ((__v4df) __A,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_rcp14_ps (__m128 __A)
{
return (__m128) __builtin_ia32_rcp14ps128_mask ((__v4sf) __A,
(__mmask8) -1);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_rcp14_ps (__m128 __W, __mmask8 __U, __m128 __A)
{
return (__m128) __builtin_ia32_rcp14ps128_mask ((__v4sf) __A,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_rcp14_ps (__mmask8 __U, __m128 __A)
{
return (__m128) __builtin_ia32_rcp14ps128_mask ((__v4sf) __A,
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_rcp14_ps (__m256 __A)
{
return (__m256) __builtin_ia32_rcp14ps256_mask ((__v8sf) __A,
(__mmask8) -1);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_rcp14_ps (__m256 __W, __mmask8 __U, __m256 __A)
{
return (__m256) __builtin_ia32_rcp14ps256_mask ((__v8sf) __A,
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_rcp14_ps (__mmask8 __U, __m256 __A)
{
return (__m256) __builtin_ia32_rcp14ps256_mask ((__v8sf) __A,
(__mmask8) __U);
}
-#define _mm_mask_permute_pd(W, U, X, C) __extension__ ({ \
+#define _mm_mask_permute_pd(W, U, X, C) \
(__m128d)__builtin_ia32_selectpd_128((__mmask8)(U), \
(__v2df)_mm_permute_pd((X), (C)), \
- (__v2df)(__m128d)(W)); })
+ (__v2df)(__m128d)(W))
-#define _mm_maskz_permute_pd(U, X, C) __extension__ ({ \
+#define _mm_maskz_permute_pd(U, X, C) \
(__m128d)__builtin_ia32_selectpd_128((__mmask8)(U), \
(__v2df)_mm_permute_pd((X), (C)), \
- (__v2df)_mm_setzero_pd()); })
+ (__v2df)_mm_setzero_pd())
-#define _mm256_mask_permute_pd(W, U, X, C) __extension__ ({ \
+#define _mm256_mask_permute_pd(W, U, X, C) \
(__m256d)__builtin_ia32_selectpd_256((__mmask8)(U), \
(__v4df)_mm256_permute_pd((X), (C)), \
- (__v4df)(__m256d)(W)); })
+ (__v4df)(__m256d)(W))
-#define _mm256_maskz_permute_pd(U, X, C) __extension__ ({ \
+#define _mm256_maskz_permute_pd(U, X, C) \
(__m256d)__builtin_ia32_selectpd_256((__mmask8)(U), \
(__v4df)_mm256_permute_pd((X), (C)), \
- (__v4df)_mm256_setzero_pd()); })
+ (__v4df)_mm256_setzero_pd())
-#define _mm_mask_permute_ps(W, U, X, C) __extension__ ({ \
+#define _mm_mask_permute_ps(W, U, X, C) \
(__m128)__builtin_ia32_selectps_128((__mmask8)(U), \
(__v4sf)_mm_permute_ps((X), (C)), \
- (__v4sf)(__m128)(W)); })
+ (__v4sf)(__m128)(W))
-#define _mm_maskz_permute_ps(U, X, C) __extension__ ({ \
+#define _mm_maskz_permute_ps(U, X, C) \
(__m128)__builtin_ia32_selectps_128((__mmask8)(U), \
(__v4sf)_mm_permute_ps((X), (C)), \
- (__v4sf)_mm_setzero_ps()); })
+ (__v4sf)_mm_setzero_ps())
-#define _mm256_mask_permute_ps(W, U, X, C) __extension__ ({ \
+#define _mm256_mask_permute_ps(W, U, X, C) \
(__m256)__builtin_ia32_selectps_256((__mmask8)(U), \
(__v8sf)_mm256_permute_ps((X), (C)), \
- (__v8sf)(__m256)(W)); })
+ (__v8sf)(__m256)(W))
-#define _mm256_maskz_permute_ps(U, X, C) __extension__ ({ \
+#define _mm256_maskz_permute_ps(U, X, C) \
(__m256)__builtin_ia32_selectps_256((__mmask8)(U), \
(__v8sf)_mm256_permute_ps((X), (C)), \
- (__v8sf)_mm256_setzero_ps()); })
+ (__v8sf)_mm256_setzero_ps())
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_permutevar_pd(__m128d __W, __mmask8 __U, __m128d __A, __m128i __C)
{
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)__W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_permutevar_pd(__mmask8 __U, __m128d __A, __m128i __C)
{
return (__m128d)__builtin_ia32_selectpd_128((__mmask8)__U,
(__v2df)_mm_setzero_pd());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_permutevar_pd(__m256d __W, __mmask8 __U, __m256d __A, __m256i __C)
{
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_permutevar_pd(__mmask8 __U, __m256d __A, __m256i __C)
{
return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
(__v4df)_mm256_setzero_pd());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_permutevar_ps(__m128 __W, __mmask8 __U, __m128 __A, __m128i __C)
{
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_permutevar_ps(__mmask8 __U, __m128 __A, __m128i __C)
{
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_permutevar_ps(__m256 __W, __mmask8 __U, __m256 __A, __m256i __C)
{
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_permutevar_ps(__mmask8 __U, __m256 __A, __m256i __C)
{
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_setzero_ps());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS128
_mm_test_epi32_mask (__m128i __A, __m128i __B)
{
- return (__mmask8) __builtin_ia32_ptestmd128 ((__v4si) __A,
- (__v4si) __B,
- (__mmask8) -1);
+ return _mm_cmpneq_epi32_mask (_mm_and_si128 (__A, __B), _mm_setzero_si128());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS128
_mm_mask_test_epi32_mask (__mmask8 __U, __m128i __A, __m128i __B)
{
- return (__mmask8) __builtin_ia32_ptestmd128 ((__v4si) __A,
- (__v4si) __B, __U);
+ return _mm_mask_cmpneq_epi32_mask (__U, _mm_and_si128 (__A, __B),
+ _mm_setzero_si128());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS256
_mm256_test_epi32_mask (__m256i __A, __m256i __B)
{
- return (__mmask8) __builtin_ia32_ptestmd256 ((__v8si) __A,
- (__v8si) __B,
- (__mmask8) -1);
+ return _mm256_cmpneq_epi32_mask (_mm256_and_si256 (__A, __B),
+ _mm256_setzero_si256());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS256
_mm256_mask_test_epi32_mask (__mmask8 __U, __m256i __A, __m256i __B)
{
- return (__mmask8) __builtin_ia32_ptestmd256 ((__v8si) __A,
- (__v8si) __B, __U);
+ return _mm256_mask_cmpneq_epi32_mask (__U, _mm256_and_si256 (__A, __B),
+ _mm256_setzero_si256());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS128
_mm_test_epi64_mask (__m128i __A, __m128i __B)
{
- return (__mmask8) __builtin_ia32_ptestmq128 ((__v2di) __A,
- (__v2di) __B,
- (__mmask8) -1);
+ return _mm_cmpneq_epi64_mask (_mm_and_si128 (__A, __B), _mm_setzero_si128());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS128
_mm_mask_test_epi64_mask (__mmask8 __U, __m128i __A, __m128i __B)
{
- return (__mmask8) __builtin_ia32_ptestmq128 ((__v2di) __A,
- (__v2di) __B, __U);
+ return _mm_mask_cmpneq_epi64_mask (__U, _mm_and_si128 (__A, __B),
+ _mm_setzero_si128());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS256
_mm256_test_epi64_mask (__m256i __A, __m256i __B)
{
- return (__mmask8) __builtin_ia32_ptestmq256 ((__v4di) __A,
- (__v4di) __B,
- (__mmask8) -1);
+ return _mm256_cmpneq_epi64_mask (_mm256_and_si256 (__A, __B),
+ _mm256_setzero_si256());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS256
_mm256_mask_test_epi64_mask (__mmask8 __U, __m256i __A, __m256i __B)
{
- return (__mmask8) __builtin_ia32_ptestmq256 ((__v4di) __A,
- (__v4di) __B, __U);
+ return _mm256_mask_cmpneq_epi64_mask (__U, _mm256_and_si256 (__A, __B),
+ _mm256_setzero_si256());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS128
_mm_testn_epi32_mask (__m128i __A, __m128i __B)
{
- return (__mmask8) __builtin_ia32_ptestnmd128 ((__v4si) __A,
- (__v4si) __B,
- (__mmask8) -1);
+ return _mm_cmpeq_epi32_mask (_mm_and_si128 (__A, __B), _mm_setzero_si128());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS128
_mm_mask_testn_epi32_mask (__mmask8 __U, __m128i __A, __m128i __B)
{
- return (__mmask8) __builtin_ia32_ptestnmd128 ((__v4si) __A,
- (__v4si) __B, __U);
+ return _mm_mask_cmpeq_epi32_mask (__U, _mm_and_si128 (__A, __B),
+ _mm_setzero_si128());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS256
_mm256_testn_epi32_mask (__m256i __A, __m256i __B)
{
- return (__mmask8) __builtin_ia32_ptestnmd256 ((__v8si) __A,
- (__v8si) __B,
- (__mmask8) -1);
+ return _mm256_cmpeq_epi32_mask (_mm256_and_si256 (__A, __B),
+ _mm256_setzero_si256());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS256
_mm256_mask_testn_epi32_mask (__mmask8 __U, __m256i __A, __m256i __B)
{
- return (__mmask8) __builtin_ia32_ptestnmd256 ((__v8si) __A,
- (__v8si) __B, __U);
+ return _mm256_mask_cmpeq_epi32_mask (__U, _mm256_and_si256 (__A, __B),
+ _mm256_setzero_si256());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS128
_mm_testn_epi64_mask (__m128i __A, __m128i __B)
{
- return (__mmask8) __builtin_ia32_ptestnmq128 ((__v2di) __A,
- (__v2di) __B,
- (__mmask8) -1);
+ return _mm_cmpeq_epi64_mask (_mm_and_si128 (__A, __B), _mm_setzero_si128());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS128
_mm_mask_testn_epi64_mask (__mmask8 __U, __m128i __A, __m128i __B)
{
- return (__mmask8) __builtin_ia32_ptestnmq128 ((__v2di) __A,
- (__v2di) __B, __U);
+ return _mm_mask_cmpeq_epi64_mask (__U, _mm_and_si128 (__A, __B),
+ _mm_setzero_si128());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS256
_mm256_testn_epi64_mask (__m256i __A, __m256i __B)
{
- return (__mmask8) __builtin_ia32_ptestnmq256 ((__v4di) __A,
- (__v4di) __B,
- (__mmask8) -1);
+ return _mm256_cmpeq_epi64_mask (_mm256_and_si256 (__A, __B),
+ _mm256_setzero_si256());
}
-static __inline__ __mmask8 __DEFAULT_FN_ATTRS
+static __inline__ __mmask8 __DEFAULT_FN_ATTRS256
_mm256_mask_testn_epi64_mask (__mmask8 __U, __m256i __A, __m256i __B)
{
- return (__mmask8) __builtin_ia32_ptestnmq256 ((__v4di) __A,
- (__v4di) __B, __U);
+ return _mm256_mask_cmpeq_epi64_mask (__U, _mm256_and_si256 (__A, __B),
+ _mm256_setzero_si256());
}
-
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_unpackhi_epi32(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_unpackhi_epi32(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_unpackhi_epi32(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_unpackhi_epi32(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_unpackhi_epi64(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_unpackhi_epi64(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)_mm_unpackhi_epi64(__A, __B),
- (__v2di)_mm_setzero_di());
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_unpackhi_epi64(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_unpackhi_epi64(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_unpacklo_epi32(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_unpacklo_epi32(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_unpacklo_epi32(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_unpacklo_epi32(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_unpacklo_epi64(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_unpacklo_epi64(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U,
(__v2di)_mm_unpacklo_epi64(__A, __B),
- (__v2di)_mm_setzero_di());
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_unpacklo_epi64(__m256i __W, __mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_unpacklo_epi64(__mmask8 __U, __m256i __A, __m256i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U,
(__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_sra_epi32(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_sra_epi32(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_sra_epi32(__m256i __W, __mmask8 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_sra_epi32(__mmask8 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_srai_epi32(__m128i __W, __mmask8 __U, __m128i __A, int __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_srai_epi32(__m128i __W, __mmask8 __U, __m128i __A, unsigned int __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_srai_epi32(__A, __B),
(__v4si)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_srai_epi32(__mmask8 __U, __m128i __A, int __B)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_srai_epi32(__mmask8 __U, __m128i __A, unsigned int __B)
{
return (__m128i)__builtin_ia32_selectd_128((__mmask8)__U,
(__v4si)_mm_srai_epi32(__A, __B),
(__v4si)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_srai_epi32(__m256i __W, __mmask8 __U, __m256i __A, int __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_srai_epi32(__m256i __W, __mmask8 __U, __m256i __A, unsigned int __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_srai_epi32(__A, __B),
(__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_srai_epi32(__mmask8 __U, __m256i __A, int __B)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_srai_epi32(__mmask8 __U, __m256i __A, unsigned int __B)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__U,
(__v8si)_mm256_srai_epi32(__A, __B),
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_sra_epi64(__m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_psraq128((__v2di)__A, (__v2di)__B);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_sra_epi64(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U, \
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_sra_epi64(__mmask8 __U, __m128i __A, __m128i __B)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U, \
(__v2di)_mm_sra_epi64(__A, __B), \
- (__v2di)_mm_setzero_di());
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_sra_epi64(__m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_psraq256((__v4di) __A, (__v2di) __B);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_sra_epi64(__m256i __W, __mmask8 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U, \
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_sra_epi64(__mmask8 __U, __m256i __A, __m128i __B)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U, \
(__v4di)_mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_srai_epi64(__m128i __A, int __imm)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_srai_epi64(__m128i __A, unsigned int __imm)
{
return (__m128i)__builtin_ia32_psraqi128((__v2di)__A, __imm);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_srai_epi64(__m128i __W, __mmask8 __U, __m128i __A, int __imm)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_srai_epi64(__m128i __W, __mmask8 __U, __m128i __A, unsigned int __imm)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U, \
(__v2di)_mm_srai_epi64(__A, __imm), \
(__v2di)__W);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_srai_epi64(__mmask8 __U, __m128i __A, int __imm)
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_srai_epi64(__mmask8 __U, __m128i __A, unsigned int __imm)
{
return (__m128i)__builtin_ia32_selectq_128((__mmask8)__U, \
(__v2di)_mm_srai_epi64(__A, __imm), \
- (__v2di)_mm_setzero_di());
+ (__v2di)_mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_srai_epi64(__m256i __A, int __imm)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_srai_epi64(__m256i __A, unsigned int __imm)
{
return (__m256i)__builtin_ia32_psraqi256((__v4di)__A, __imm);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_srai_epi64(__m256i __W, __mmask8 __U, __m256i __A, int __imm)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_srai_epi64(__m256i __W, __mmask8 __U, __m256i __A,
+ unsigned int __imm)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U, \
(__v4di)_mm256_srai_epi64(__A, __imm), \
(__v4di)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_srai_epi64(__mmask8 __U, __m256i __A, int __imm)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_srai_epi64(__mmask8 __U, __m256i __A, unsigned int __imm)
{
return (__m256i)__builtin_ia32_selectq_256((__mmask8)__U, \
(__v4di)_mm256_srai_epi64(__A, __imm), \
(__v4di)_mm256_setzero_si256());
}
-#define _mm_ternarylogic_epi32(A, B, C, imm) __extension__ ({ \
+#define _mm_ternarylogic_epi32(A, B, C, imm) \
(__m128i)__builtin_ia32_pternlogd128_mask((__v4si)(__m128i)(A), \
(__v4si)(__m128i)(B), \
(__v4si)(__m128i)(C), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_ternarylogic_epi32(A, U, B, C, imm) __extension__ ({ \
+#define _mm_mask_ternarylogic_epi32(A, U, B, C, imm) \
(__m128i)__builtin_ia32_pternlogd128_mask((__v4si)(__m128i)(A), \
(__v4si)(__m128i)(B), \
(__v4si)(__m128i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_maskz_ternarylogic_epi32(U, A, B, C, imm) __extension__ ({ \
+#define _mm_maskz_ternarylogic_epi32(U, A, B, C, imm) \
(__m128i)__builtin_ia32_pternlogd128_maskz((__v4si)(__m128i)(A), \
(__v4si)(__m128i)(B), \
(__v4si)(__m128i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_ternarylogic_epi32(A, B, C, imm) __extension__ ({ \
+#define _mm256_ternarylogic_epi32(A, B, C, imm) \
(__m256i)__builtin_ia32_pternlogd256_mask((__v8si)(__m256i)(A), \
(__v8si)(__m256i)(B), \
(__v8si)(__m256i)(C), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_ternarylogic_epi32(A, U, B, C, imm) __extension__ ({ \
+#define _mm256_mask_ternarylogic_epi32(A, U, B, C, imm) \
(__m256i)__builtin_ia32_pternlogd256_mask((__v8si)(__m256i)(A), \
(__v8si)(__m256i)(B), \
(__v8si)(__m256i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_maskz_ternarylogic_epi32(U, A, B, C, imm) __extension__ ({ \
+#define _mm256_maskz_ternarylogic_epi32(U, A, B, C, imm) \
(__m256i)__builtin_ia32_pternlogd256_maskz((__v8si)(__m256i)(A), \
(__v8si)(__m256i)(B), \
(__v8si)(__m256i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_ternarylogic_epi64(A, B, C, imm) __extension__ ({ \
+#define _mm_ternarylogic_epi64(A, B, C, imm) \
(__m128i)__builtin_ia32_pternlogq128_mask((__v2di)(__m128i)(A), \
(__v2di)(__m128i)(B), \
(__v2di)(__m128i)(C), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_ternarylogic_epi64(A, U, B, C, imm) __extension__ ({ \
+#define _mm_mask_ternarylogic_epi64(A, U, B, C, imm) \
(__m128i)__builtin_ia32_pternlogq128_mask((__v2di)(__m128i)(A), \
(__v2di)(__m128i)(B), \
(__v2di)(__m128i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_maskz_ternarylogic_epi64(U, A, B, C, imm) __extension__ ({ \
+#define _mm_maskz_ternarylogic_epi64(U, A, B, C, imm) \
(__m128i)__builtin_ia32_pternlogq128_maskz((__v2di)(__m128i)(A), \
(__v2di)(__m128i)(B), \
(__v2di)(__m128i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_ternarylogic_epi64(A, B, C, imm) __extension__ ({ \
+#define _mm256_ternarylogic_epi64(A, B, C, imm) \
(__m256i)__builtin_ia32_pternlogq256_mask((__v4di)(__m256i)(A), \
(__v4di)(__m256i)(B), \
(__v4di)(__m256i)(C), (int)(imm), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_ternarylogic_epi64(A, U, B, C, imm) __extension__ ({ \
+#define _mm256_mask_ternarylogic_epi64(A, U, B, C, imm) \
(__m256i)__builtin_ia32_pternlogq256_mask((__v4di)(__m256i)(A), \
(__v4di)(__m256i)(B), \
(__v4di)(__m256i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_maskz_ternarylogic_epi64(U, A, B, C, imm) __extension__ ({ \
+#define _mm256_maskz_ternarylogic_epi64(U, A, B, C, imm) \
(__m256i)__builtin_ia32_pternlogq256_maskz((__v4di)(__m256i)(A), \
(__v4di)(__m256i)(B), \
(__v4di)(__m256i)(C), (int)(imm), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_shuffle_f32x4(A, B, imm) __extension__ ({ \
- (__m256)__builtin_ia32_shuf_f32x4_256_mask((__v8sf)(__m256)(A), \
- (__v8sf)(__m256)(B), (int)(imm), \
- (__v8sf)_mm256_setzero_ps(), \
- (__mmask8)-1); })
+#define _mm256_shuffle_f32x4(A, B, imm) \
+ (__m256)__builtin_ia32_shuf_f32x4_256((__v8sf)(__m256)(A), \
+ (__v8sf)(__m256)(B), (int)(imm))
-#define _mm256_mask_shuffle_f32x4(W, U, A, B, imm) __extension__ ({ \
- (__m256)__builtin_ia32_shuf_f32x4_256_mask((__v8sf)(__m256)(A), \
- (__v8sf)(__m256)(B), (int)(imm), \
- (__v8sf)(__m256)(W), \
- (__mmask8)(U)); })
+#define _mm256_mask_shuffle_f32x4(W, U, A, B, imm) \
+ (__m256)__builtin_ia32_selectps_256((__mmask8)(U), \
+ (__v8sf)_mm256_shuffle_f32x4((A), (B), (imm)), \
+ (__v8sf)(__m256)(W))
-#define _mm256_maskz_shuffle_f32x4(U, A, B, imm) __extension__ ({ \
- (__m256)__builtin_ia32_shuf_f32x4_256_mask((__v8sf)(__m256)(A), \
- (__v8sf)(__m256)(B), (int)(imm), \
- (__v8sf)_mm256_setzero_ps(), \
- (__mmask8)(U)); })
+#define _mm256_maskz_shuffle_f32x4(U, A, B, imm) \
+ (__m256)__builtin_ia32_selectps_256((__mmask8)(U), \
+ (__v8sf)_mm256_shuffle_f32x4((A), (B), (imm)), \
+ (__v8sf)_mm256_setzero_ps())
-#define _mm256_shuffle_f64x2(A, B, imm) __extension__ ({ \
- (__m256d)__builtin_ia32_shuf_f64x2_256_mask((__v4df)(__m256d)(A), \
- (__v4df)(__m256d)(B), \
- (int)(imm), \
- (__v4df)_mm256_setzero_pd(), \
- (__mmask8)-1); })
+#define _mm256_shuffle_f64x2(A, B, imm) \
+ (__m256d)__builtin_ia32_shuf_f64x2_256((__v4df)(__m256d)(A), \
+ (__v4df)(__m256d)(B), (int)(imm))
-#define _mm256_mask_shuffle_f64x2(W, U, A, B, imm) __extension__ ({ \
- (__m256d)__builtin_ia32_shuf_f64x2_256_mask((__v4df)(__m256d)(A), \
- (__v4df)(__m256d)(B), \
- (int)(imm), \
- (__v4df)(__m256d)(W), \
- (__mmask8)(U)); })
+#define _mm256_mask_shuffle_f64x2(W, U, A, B, imm) \
+ (__m256d)__builtin_ia32_selectpd_256((__mmask8)(U), \
+ (__v4df)_mm256_shuffle_f64x2((A), (B), (imm)), \
+ (__v4df)(__m256d)(W))
-#define _mm256_maskz_shuffle_f64x2(U, A, B, imm) __extension__ ({ \
- (__m256d)__builtin_ia32_shuf_f64x2_256_mask((__v4df)(__m256d)(A), \
- (__v4df)(__m256d)(B), \
- (int)(imm), \
- (__v4df)_mm256_setzero_pd(), \
- (__mmask8)(U)); })
+#define _mm256_maskz_shuffle_f64x2(U, A, B, imm) \
+ (__m256d)__builtin_ia32_selectpd_256((__mmask8)(U), \
+ (__v4df)_mm256_shuffle_f64x2((A), (B), (imm)), \
+ (__v4df)_mm256_setzero_pd())
-#define _mm256_shuffle_i32x4(A, B, imm) __extension__ ({ \
- (__m256i)__builtin_ia32_shuf_i32x4_256_mask((__v8si)(__m256i)(A), \
- (__v8si)(__m256i)(B), \
- (int)(imm), \
- (__v8si)_mm256_setzero_si256(), \
- (__mmask8)-1); })
+#define _mm256_shuffle_i32x4(A, B, imm) \
+ (__m256i)__builtin_ia32_shuf_i32x4_256((__v8si)(__m256i)(A), \
+ (__v8si)(__m256i)(B), (int)(imm))
-#define _mm256_mask_shuffle_i32x4(W, U, A, B, imm) __extension__ ({ \
- (__m256i)__builtin_ia32_shuf_i32x4_256_mask((__v8si)(__m256i)(A), \
- (__v8si)(__m256i)(B), \
- (int)(imm), \
- (__v8si)(__m256i)(W), \
- (__mmask8)(U)); })
+#define _mm256_mask_shuffle_i32x4(W, U, A, B, imm) \
+ (__m256i)__builtin_ia32_selectd_256((__mmask8)(U), \
+ (__v8si)_mm256_shuffle_i32x4((A), (B), (imm)), \
+ (__v8si)(__m256i)(W))
-#define _mm256_maskz_shuffle_i32x4(U, A, B, imm) __extension__ ({ \
- (__m256i)__builtin_ia32_shuf_i32x4_256_mask((__v8si)(__m256i)(A), \
- (__v8si)(__m256i)(B), \
- (int)(imm), \
- (__v8si)_mm256_setzero_si256(), \
- (__mmask8)(U)); })
+#define _mm256_maskz_shuffle_i32x4(U, A, B, imm) \
+ (__m256i)__builtin_ia32_selectd_256((__mmask8)(U), \
+ (__v8si)_mm256_shuffle_i32x4((A), (B), (imm)), \
+ (__v8si)_mm256_setzero_si256())
-#define _mm256_shuffle_i64x2(A, B, imm) __extension__ ({ \
- (__m256i)__builtin_ia32_shuf_i64x2_256_mask((__v4di)(__m256i)(A), \
- (__v4di)(__m256i)(B), \
- (int)(imm), \
- (__v4di)_mm256_setzero_si256(), \
- (__mmask8)-1); })
+#define _mm256_shuffle_i64x2(A, B, imm) \
+ (__m256i)__builtin_ia32_shuf_i64x2_256((__v4di)(__m256i)(A), \
+ (__v4di)(__m256i)(B), (int)(imm))
-#define _mm256_mask_shuffle_i64x2(W, U, A, B, imm) __extension__ ({ \
- (__m256i)__builtin_ia32_shuf_i64x2_256_mask((__v4di)(__m256i)(A), \
- (__v4di)(__m256i)(B), \
- (int)(imm), \
- (__v4di)(__m256i)(W), \
- (__mmask8)(U)); })
+#define _mm256_mask_shuffle_i64x2(W, U, A, B, imm) \
+ (__m256i)__builtin_ia32_selectq_256((__mmask8)(U), \
+ (__v4di)_mm256_shuffle_i64x2((A), (B), (imm)), \
+ (__v4di)(__m256i)(W))
-#define _mm256_maskz_shuffle_i64x2(U, A, B, imm) __extension__ ({ \
- (__m256i)__builtin_ia32_shuf_i64x2_256_mask((__v4di)(__m256i)(A), \
- (__v4di)(__m256i)(B), \
- (int)(imm), \
- (__v4di)_mm256_setzero_si256(), \
- (__mmask8)(U)); })
-#define _mm_mask_shuffle_pd(W, U, A, B, M) __extension__ ({ \
+#define _mm256_maskz_shuffle_i64x2(U, A, B, imm) \
+ (__m256i)__builtin_ia32_selectq_256((__mmask8)(U), \
+ (__v4di)_mm256_shuffle_i64x2((A), (B), (imm)), \
+ (__v4di)_mm256_setzero_si256())
+
+#define _mm_mask_shuffle_pd(W, U, A, B, M) \
(__m128d)__builtin_ia32_selectpd_128((__mmask8)(U), \
(__v2df)_mm_shuffle_pd((A), (B), (M)), \
- (__v2df)(__m128d)(W)); })
+ (__v2df)(__m128d)(W))
-#define _mm_maskz_shuffle_pd(U, A, B, M) __extension__ ({ \
+#define _mm_maskz_shuffle_pd(U, A, B, M) \
(__m128d)__builtin_ia32_selectpd_128((__mmask8)(U), \
(__v2df)_mm_shuffle_pd((A), (B), (M)), \
- (__v2df)_mm_setzero_pd()); })
+ (__v2df)_mm_setzero_pd())
-#define _mm256_mask_shuffle_pd(W, U, A, B, M) __extension__ ({ \
+#define _mm256_mask_shuffle_pd(W, U, A, B, M) \
(__m256d)__builtin_ia32_selectpd_256((__mmask8)(U), \
(__v4df)_mm256_shuffle_pd((A), (B), (M)), \
- (__v4df)(__m256d)(W)); })
+ (__v4df)(__m256d)(W))
-#define _mm256_maskz_shuffle_pd(U, A, B, M) __extension__ ({ \
+#define _mm256_maskz_shuffle_pd(U, A, B, M) \
(__m256d)__builtin_ia32_selectpd_256((__mmask8)(U), \
(__v4df)_mm256_shuffle_pd((A), (B), (M)), \
- (__v4df)_mm256_setzero_pd()); })
+ (__v4df)_mm256_setzero_pd())
-#define _mm_mask_shuffle_ps(W, U, A, B, M) __extension__ ({ \
+#define _mm_mask_shuffle_ps(W, U, A, B, M) \
(__m128)__builtin_ia32_selectps_128((__mmask8)(U), \
(__v4sf)_mm_shuffle_ps((A), (B), (M)), \
- (__v4sf)(__m128)(W)); })
+ (__v4sf)(__m128)(W))
-#define _mm_maskz_shuffle_ps(U, A, B, M) __extension__ ({ \
+#define _mm_maskz_shuffle_ps(U, A, B, M) \
(__m128)__builtin_ia32_selectps_128((__mmask8)(U), \
(__v4sf)_mm_shuffle_ps((A), (B), (M)), \
- (__v4sf)_mm_setzero_ps()); })
+ (__v4sf)_mm_setzero_ps())
-#define _mm256_mask_shuffle_ps(W, U, A, B, M) __extension__ ({ \
+#define _mm256_mask_shuffle_ps(W, U, A, B, M) \
(__m256)__builtin_ia32_selectps_256((__mmask8)(U), \
(__v8sf)_mm256_shuffle_ps((A), (B), (M)), \
- (__v8sf)(__m256)(W)); })
+ (__v8sf)(__m256)(W))
-#define _mm256_maskz_shuffle_ps(U, A, B, M) __extension__ ({ \
+#define _mm256_maskz_shuffle_ps(U, A, B, M) \
(__m256)__builtin_ia32_selectps_256((__mmask8)(U), \
(__v8sf)_mm256_shuffle_ps((A), (B), (M)), \
- (__v8sf)_mm256_setzero_ps()); })
+ (__v8sf)_mm256_setzero_ps())
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_rsqrt14_pd (__m128d __A)
{
return (__m128d) __builtin_ia32_rsqrt14pd128_mask ((__v2df) __A,
(__mmask8) -1);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_rsqrt14_pd (__m128d __W, __mmask8 __U, __m128d __A)
{
return (__m128d) __builtin_ia32_rsqrt14pd128_mask ((__v2df) __A,
(__mmask8) __U);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_rsqrt14_pd (__mmask8 __U, __m128d __A)
{
return (__m128d) __builtin_ia32_rsqrt14pd128_mask ((__v2df) __A,
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_rsqrt14_pd (__m256d __A)
{
return (__m256d) __builtin_ia32_rsqrt14pd256_mask ((__v4df) __A,
(__mmask8) -1);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_rsqrt14_pd (__m256d __W, __mmask8 __U, __m256d __A)
{
return (__m256d) __builtin_ia32_rsqrt14pd256_mask ((__v4df) __A,
(__mmask8) __U);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_rsqrt14_pd (__mmask8 __U, __m256d __A)
{
return (__m256d) __builtin_ia32_rsqrt14pd256_mask ((__v4df) __A,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_rsqrt14_ps (__m128 __A)
{
return (__m128) __builtin_ia32_rsqrt14ps128_mask ((__v4sf) __A,
(__mmask8) -1);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_rsqrt14_ps (__m128 __W, __mmask8 __U, __m128 __A)
{
return (__m128) __builtin_ia32_rsqrt14ps128_mask ((__v4sf) __A,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_rsqrt14_ps (__mmask8 __U, __m128 __A)
{
return (__m128) __builtin_ia32_rsqrt14ps128_mask ((__v4sf) __A,
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_rsqrt14_ps (__m256 __A)
{
return (__m256) __builtin_ia32_rsqrt14ps256_mask ((__v8sf) __A,
(__mmask8) -1);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_rsqrt14_ps (__m256 __W, __mmask8 __U, __m256 __A)
{
return (__m256) __builtin_ia32_rsqrt14ps256_mask ((__v8sf) __A,
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_rsqrt14_ps (__mmask8 __U, __m256 __A)
{
return (__m256) __builtin_ia32_rsqrt14ps256_mask ((__v8sf) __A,
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_broadcast_f32x4(__m128 __A)
{
return (__m256)__builtin_shufflevector((__v4sf)__A, (__v4sf)__A,
0, 1, 2, 3, 0, 1, 2, 3);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_broadcast_f32x4(__m256 __O, __mmask8 __M, __m128 __A)
{
return (__m256)__builtin_ia32_selectps_256((__mmask8)__M,
(__v8sf)__O);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_broadcast_f32x4 (__mmask8 __M, __m128 __A)
{
return (__m256)__builtin_ia32_selectps_256((__mmask8)__M,
(__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_broadcast_i32x4(__m128i __A)
{
return (__m256i)__builtin_shufflevector((__v4si)__A, (__v4si)__A,
0, 1, 2, 3, 0, 1, 2, 3);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_broadcast_i32x4(__m256i __O, __mmask8 __M, __m128i __A)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
(__v8si)__O);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_broadcast_i32x4(__mmask8 __M, __m128i __A)
{
return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
(__v8si)_mm256_setzero_si256());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_broadcastsd_pd (__m256d __O, __mmask8 __M, __m128d __A)
{
return (__m256d)__builtin_ia32_selectpd_256(__M,
(__v4df) __O);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_broadcastsd_pd (__mmask8 __M, __m128d __A)
{
return (__m256d)__builtin_ia32_selectpd_256(__M,
(__v4df) _mm256_setzero_pd());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_broadcastss_ps (__m128 __O, __mmask8 __M, __m128 __A)
{
return (__m128)__builtin_ia32_selectps_128(__M,
(__v4sf) __O);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_broadcastss_ps (__mmask8 __M, __m128 __A)
{
return (__m128)__builtin_ia32_selectps_128(__M,
(__v4sf) _mm_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_broadcastss_ps (__m256 __O, __mmask8 __M, __m128 __A)
{
return (__m256)__builtin_ia32_selectps_256(__M,
(__v8sf) __O);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_broadcastss_ps (__mmask8 __M, __m128 __A)
{
return (__m256)__builtin_ia32_selectps_256(__M,
(__v8sf) _mm256_setzero_ps());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_broadcastd_epi32 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i)__builtin_ia32_selectd_128(__M,
(__v4si) __O);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_broadcastd_epi32 (__mmask8 __M, __m128i __A)
{
return (__m128i)__builtin_ia32_selectd_128(__M,
(__v4si) _mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_broadcastd_epi32 (__m256i __O, __mmask8 __M, __m128i __A)
{
return (__m256i)__builtin_ia32_selectd_256(__M,
(__v8si) __O);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_broadcastd_epi32 (__mmask8 __M, __m128i __A)
{
return (__m256i)__builtin_ia32_selectd_256(__M,
(__v8si) _mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_broadcastq_epi64 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i)__builtin_ia32_selectq_128(__M,
(__v2di) __O);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_broadcastq_epi64 (__mmask8 __M, __m128i __A)
{
return (__m128i)__builtin_ia32_selectq_128(__M,
(__v2di) _mm_setzero_si128());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_broadcastq_epi64 (__m256i __O, __mmask8 __M, __m128i __A)
{
return (__m256i)__builtin_ia32_selectq_256(__M,
(__v4di) __O);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_maskz_broadcastq_epi64 (__mmask8 __M, __m128i __A)
{
return (__m256i)__builtin_ia32_selectq_256(__M,
(__v4di) _mm256_setzero_si256());
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtsepi32_epi8 (__m128i __A)
{
return (__m128i) __builtin_ia32_pmovsdb128_mask ((__v4si) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtsepi32_epi8 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovsdb128_mask ((__v4si) __A,
(__v16qi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtsepi32_epi8 (__mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovsdb128_mask ((__v4si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtsepi32_storeu_epi8 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovsdb128mem_mask ((__v16qi *) __P, (__v4si) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtsepi32_epi8 (__m256i __A)
{
return (__m128i) __builtin_ia32_pmovsdb256_mask ((__v8si) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtsepi32_epi8 (__m128i __O, __mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovsdb256_mask ((__v8si) __A,
(__v16qi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtsepi32_epi8 (__mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovsdb256_mask ((__v8si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtsepi32_storeu_epi8 (void * __P, __mmask8 __M, __m256i __A)
{
__builtin_ia32_pmovsdb256mem_mask ((__v16qi *) __P, (__v8si) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtsepi32_epi16 (__m128i __A)
{
return (__m128i) __builtin_ia32_pmovsdw128_mask ((__v4si) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtsepi32_epi16 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovsdw128_mask ((__v4si) __A,
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtsepi32_epi16 (__mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovsdw128_mask ((__v4si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtsepi32_storeu_epi16 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovsdw128mem_mask ((__v8hi *) __P, (__v4si) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtsepi32_epi16 (__m256i __A)
{
return (__m128i) __builtin_ia32_pmovsdw256_mask ((__v8si) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtsepi32_epi16 (__m128i __O, __mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovsdw256_mask ((__v8si) __A,
(__v8hi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtsepi32_epi16 (__mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovsdw256_mask ((__v8si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtsepi32_storeu_epi16 (void * __P, __mmask8 __M, __m256i __A)
{
__builtin_ia32_pmovsdw256mem_mask ((__v8hi *) __P, (__v8si) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtsepi64_epi8 (__m128i __A)
{
return (__m128i) __builtin_ia32_pmovsqb128_mask ((__v2di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtsepi64_epi8 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovsqb128_mask ((__v2di) __A,
(__v16qi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtsepi64_epi8 (__mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovsqb128_mask ((__v2di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtsepi64_storeu_epi8 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovsqb128mem_mask ((__v16qi *) __P, (__v2di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtsepi64_epi8 (__m256i __A)
{
return (__m128i) __builtin_ia32_pmovsqb256_mask ((__v4di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtsepi64_epi8 (__m128i __O, __mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovsqb256_mask ((__v4di) __A,
(__v16qi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtsepi64_epi8 (__mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovsqb256_mask ((__v4di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtsepi64_storeu_epi8 (void * __P, __mmask8 __M, __m256i __A)
{
__builtin_ia32_pmovsqb256mem_mask ((__v16qi *) __P, (__v4di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtsepi64_epi32 (__m128i __A)
{
return (__m128i) __builtin_ia32_pmovsqd128_mask ((__v2di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtsepi64_epi32 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovsqd128_mask ((__v2di) __A,
(__v4si) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtsepi64_epi32 (__mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovsqd128_mask ((__v2di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtsepi64_storeu_epi32 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovsqd128mem_mask ((__v4si *) __P, (__v2di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtsepi64_epi32 (__m256i __A)
{
return (__m128i) __builtin_ia32_pmovsqd256_mask ((__v4di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtsepi64_epi32 (__m128i __O, __mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovsqd256_mask ((__v4di) __A,
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtsepi64_epi32 (__mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovsqd256_mask ((__v4di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtsepi64_storeu_epi32 (void * __P, __mmask8 __M, __m256i __A)
{
__builtin_ia32_pmovsqd256mem_mask ((__v4si *) __P, (__v4di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtsepi64_epi16 (__m128i __A)
{
return (__m128i) __builtin_ia32_pmovsqw128_mask ((__v2di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtsepi64_epi16 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovsqw128_mask ((__v2di) __A,
(__v8hi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtsepi64_epi16 (__mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovsqw128_mask ((__v2di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtsepi64_storeu_epi16 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovsqw128mem_mask ((__v8hi *) __P, (__v2di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtsepi64_epi16 (__m256i __A)
{
return (__m128i) __builtin_ia32_pmovsqw256_mask ((__v4di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtsepi64_epi16 (__m128i __O, __mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovsqw256_mask ((__v4di) __A,
(__v8hi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtsepi64_epi16 (__mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovsqw256_mask ((__v4di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtsepi64_storeu_epi16 (void * __P, __mmask8 __M, __m256i __A)
{
__builtin_ia32_pmovsqw256mem_mask ((__v8hi *) __P, (__v4di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtusepi32_epi8 (__m128i __A)
{
return (__m128i) __builtin_ia32_pmovusdb128_mask ((__v4si) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtusepi32_epi8 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovusdb128_mask ((__v4si) __A,
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtusepi32_epi8 (__mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovusdb128_mask ((__v4si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtusepi32_storeu_epi8 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovusdb128mem_mask ((__v16qi *) __P, (__v4si) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtusepi32_epi8 (__m256i __A)
{
return (__m128i) __builtin_ia32_pmovusdb256_mask ((__v8si) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtusepi32_epi8 (__m128i __O, __mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovusdb256_mask ((__v8si) __A,
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtusepi32_epi8 (__mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovusdb256_mask ((__v8si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtusepi32_storeu_epi8 (void * __P, __mmask8 __M, __m256i __A)
{
__builtin_ia32_pmovusdb256mem_mask ((__v16qi*) __P, (__v8si) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtusepi32_epi16 (__m128i __A)
{
return (__m128i) __builtin_ia32_pmovusdw128_mask ((__v4si) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtusepi32_epi16 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovusdw128_mask ((__v4si) __A,
(__v8hi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtusepi32_epi16 (__mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovusdw128_mask ((__v4si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtusepi32_storeu_epi16 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovusdw128mem_mask ((__v8hi *) __P, (__v4si) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtusepi32_epi16 (__m256i __A)
{
return (__m128i) __builtin_ia32_pmovusdw256_mask ((__v8si) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtusepi32_epi16 (__m128i __O, __mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovusdw256_mask ((__v8si) __A,
(__v8hi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtusepi32_epi16 (__mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovusdw256_mask ((__v8si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtusepi32_storeu_epi16 (void * __P, __mmask8 __M, __m256i __A)
{
__builtin_ia32_pmovusdw256mem_mask ((__v8hi *) __P, (__v8si) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtusepi64_epi8 (__m128i __A)
{
return (__m128i) __builtin_ia32_pmovusqb128_mask ((__v2di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtusepi64_epi8 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovusqb128_mask ((__v2di) __A,
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtusepi64_epi8 (__mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovusqb128_mask ((__v2di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtusepi64_storeu_epi8 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovusqb128mem_mask ((__v16qi *) __P, (__v2di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtusepi64_epi8 (__m256i __A)
{
return (__m128i) __builtin_ia32_pmovusqb256_mask ((__v4di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtusepi64_epi8 (__m128i __O, __mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovusqb256_mask ((__v4di) __A,
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtusepi64_epi8 (__mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovusqb256_mask ((__v4di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtusepi64_storeu_epi8 (void * __P, __mmask8 __M, __m256i __A)
{
__builtin_ia32_pmovusqb256mem_mask ((__v16qi *) __P, (__v4di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtusepi64_epi32 (__m128i __A)
{
return (__m128i) __builtin_ia32_pmovusqd128_mask ((__v2di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtusepi64_epi32 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovusqd128_mask ((__v2di) __A,
(__v4si) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtusepi64_epi32 (__mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovusqd128_mask ((__v2di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtusepi64_storeu_epi32 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovusqd128mem_mask ((__v4si *) __P, (__v2di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtusepi64_epi32 (__m256i __A)
{
return (__m128i) __builtin_ia32_pmovusqd256_mask ((__v4di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtusepi64_epi32 (__m128i __O, __mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovusqd256_mask ((__v4di) __A,
(__v4si) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtusepi64_epi32 (__mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovusqd256_mask ((__v4di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtusepi64_storeu_epi32 (void * __P, __mmask8 __M, __m256i __A)
{
__builtin_ia32_pmovusqd256mem_mask ((__v4si *) __P, (__v4di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtusepi64_epi16 (__m128i __A)
{
return (__m128i) __builtin_ia32_pmovusqw128_mask ((__v2di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtusepi64_epi16 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovusqw128_mask ((__v2di) __A,
(__v8hi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtusepi64_epi16 (__mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovusqw128_mask ((__v2di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtusepi64_storeu_epi16 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovusqw128mem_mask ((__v8hi *) __P, (__v2di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtusepi64_epi16 (__m256i __A)
{
return (__m128i) __builtin_ia32_pmovusqw256_mask ((__v4di) __A,
(__mmask8) -1);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtusepi64_epi16 (__m128i __O, __mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovusqw256_mask ((__v4di) __A,
(__v8hi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtusepi64_epi16 (__mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovusqw256_mask ((__v4di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtusepi64_storeu_epi16 (void * __P, __mmask8 __M, __m256i __A)
{
- return __builtin_ia32_pmovusqw256mem_mask ((__v8hi *) __P, (__v4di) __A, __M);
+ __builtin_ia32_pmovusqw256mem_mask ((__v8hi *) __P, (__v4di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtepi32_epi8 (__m128i __A)
{
- return (__m128i) __builtin_ia32_pmovdb128_mask ((__v4si) __A,
- (__v16qi)_mm_undefined_si128(),
- (__mmask8) -1);
+ return (__m128i)__builtin_shufflevector(
+ __builtin_convertvector((__v4si)__A, __v4qi), (__v4qi){0, 0, 0, 0}, 0, 1,
+ 2, 3, 4, 5, 6, 7, 7, 7, 7, 7, 7, 7, 7, 7);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi32_epi8 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovdb128_mask ((__v4si) __A,
(__v16qi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtepi32_epi8 (__mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovdb128_mask ((__v4si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi32_storeu_epi8 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovdb128mem_mask ((__v16qi *) __P, (__v4si) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtepi32_epi8 (__m256i __A)
{
- return (__m128i) __builtin_ia32_pmovdb256_mask ((__v8si) __A,
- (__v16qi)_mm_undefined_si128(),
- (__mmask8) -1);
+ return (__m128i)__builtin_shufflevector(
+ __builtin_convertvector((__v8si)__A, __v8qi),
+ (__v8qi){0, 0, 0, 0, 0, 0, 0, 0}, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11,
+ 12, 13, 14, 15);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi32_epi8 (__m128i __O, __mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovdb256_mask ((__v8si) __A,
(__v16qi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtepi32_epi8 (__mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovdb256_mask ((__v8si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi32_storeu_epi8 (void * __P, __mmask8 __M, __m256i __A)
{
__builtin_ia32_pmovdb256mem_mask ((__v16qi *) __P, (__v8si) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtepi32_epi16 (__m128i __A)
{
- return (__m128i) __builtin_ia32_pmovdw128_mask ((__v4si) __A,
- (__v8hi) _mm_setzero_si128 (),
- (__mmask8) -1);
+ return (__m128i)__builtin_shufflevector(
+ __builtin_convertvector((__v4si)__A, __v4hi), (__v4hi){0, 0, 0, 0}, 0, 1,
+ 2, 3, 4, 5, 6, 7);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi32_epi16 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovdw128_mask ((__v4si) __A,
(__v8hi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtepi32_epi16 (__mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovdw128_mask ((__v4si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi32_storeu_epi16 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovdw128mem_mask ((__v8hi *) __P, (__v4si) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtepi32_epi16 (__m256i __A)
{
- return (__m128i) __builtin_ia32_pmovdw256_mask ((__v8si) __A,
- (__v8hi)_mm_setzero_si128 (),
- (__mmask8) -1);
+ return (__m128i)__builtin_convertvector((__v8si)__A, __v8hi);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi32_epi16 (__m128i __O, __mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovdw256_mask ((__v8si) __A,
(__v8hi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtepi32_epi16 (__mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovdw256_mask ((__v8si) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi32_storeu_epi16 (void * __P, __mmask8 __M, __m256i __A)
{
__builtin_ia32_pmovdw256mem_mask ((__v8hi *) __P, (__v8si) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtepi64_epi8 (__m128i __A)
{
- return (__m128i) __builtin_ia32_pmovqb128_mask ((__v2di) __A,
- (__v16qi) _mm_undefined_si128(),
- (__mmask8) -1);
+ return (__m128i)__builtin_shufflevector(
+ __builtin_convertvector((__v2di)__A, __v2qi), (__v2qi){0, 0}, 0, 1, 2, 3,
+ 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi64_epi8 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovqb128_mask ((__v2di) __A,
(__v16qi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtepi64_epi8 (__mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovqb128_mask ((__v2di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi64_storeu_epi8 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovqb128mem_mask ((__v16qi *) __P, (__v2di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtepi64_epi8 (__m256i __A)
{
- return (__m128i) __builtin_ia32_pmovqb256_mask ((__v4di) __A,
- (__v16qi) _mm_undefined_si128(),
- (__mmask8) -1);
+ return (__m128i)__builtin_shufflevector(
+ __builtin_convertvector((__v4di)__A, __v4qi), (__v4qi){0, 0, 0, 0}, 0, 1,
+ 2, 3, 4, 5, 6, 7, 7, 7, 7, 7, 7, 7, 7, 7);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi64_epi8 (__m128i __O, __mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovqb256_mask ((__v4di) __A,
(__v16qi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtepi64_epi8 (__mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovqb256_mask ((__v4di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi64_storeu_epi8 (void * __P, __mmask8 __M, __m256i __A)
{
__builtin_ia32_pmovqb256mem_mask ((__v16qi *) __P, (__v4di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtepi64_epi32 (__m128i __A)
{
- return (__m128i) __builtin_ia32_pmovqd128_mask ((__v2di) __A,
- (__v4si)_mm_undefined_si128(),
- (__mmask8) -1);
+ return (__m128i)__builtin_shufflevector(
+ __builtin_convertvector((__v2di)__A, __v2si), (__v2si){0, 0}, 0, 1, 2, 3);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi64_epi32 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovqd128_mask ((__v2di) __A,
(__v4si) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtepi64_epi32 (__mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovqd128_mask ((__v2di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi64_storeu_epi32 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovqd128mem_mask ((__v4si *) __P, (__v2di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtepi64_epi32 (__m256i __A)
{
- return (__m128i) __builtin_ia32_pmovqd256_mask ((__v4di) __A,
- (__v4si) _mm_undefined_si128(),
- (__mmask8) -1);
+ return (__m128i)__builtin_convertvector((__v4di)__A, __v4si);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi64_epi32 (__m128i __O, __mmask8 __M, __m256i __A)
{
- return (__m128i) __builtin_ia32_pmovqd256_mask ((__v4di) __A,
- (__v4si) __O, __M);
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__M,
+ (__v4si)_mm256_cvtepi64_epi32(__A),
+ (__v4si)__O);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtepi64_epi32 (__mmask8 __M, __m256i __A)
{
- return (__m128i) __builtin_ia32_pmovqd256_mask ((__v4di) __A,
- (__v4si) _mm_setzero_si128 (),
- __M);
+ return (__m128i)__builtin_ia32_selectd_128((__mmask8)__M,
+ (__v4si)_mm256_cvtepi64_epi32(__A),
+ (__v4si)_mm_setzero_si128());
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi64_storeu_epi32 (void * __P, __mmask8 __M, __m256i __A)
{
__builtin_ia32_pmovqd256mem_mask ((__v4si *) __P, (__v4di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_cvtepi64_epi16 (__m128i __A)
{
- return (__m128i) __builtin_ia32_pmovqw128_mask ((__v2di) __A,
- (__v8hi) _mm_undefined_si128(),
- (__mmask8) -1);
+ return (__m128i)__builtin_shufflevector(
+ __builtin_convertvector((__v2di)__A, __v2hi), (__v2hi){0, 0}, 0, 1, 2, 3,
+ 3, 3, 3, 3);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi64_epi16 (__m128i __O, __mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovqw128_mask ((__v2di) __A,
__M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
_mm_maskz_cvtepi64_epi16 (__mmask8 __M, __m128i __A)
{
return (__m128i) __builtin_ia32_pmovqw128_mask ((__v2di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_cvtepi64_storeu_epi16 (void * __P, __mmask8 __M, __m128i __A)
{
__builtin_ia32_pmovqw128mem_mask ((__v8hi *) __P, (__v2di) __A, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_cvtepi64_epi16 (__m256i __A)
{
- return (__m128i) __builtin_ia32_pmovqw256_mask ((__v4di) __A,
- (__v8hi)_mm_undefined_si128(),
- (__mmask8) -1);
+ return (__m128i)__builtin_shufflevector(
+ __builtin_convertvector((__v4di)__A, __v4hi), (__v4hi){0, 0, 0, 0}, 0, 1,
+ 2, 3, 4, 5, 6, 7);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi64_epi16 (__m128i __O, __mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovqw256_mask ((__v4di) __A,
(__v8hi) __O, __M);
}
-static __inline__ __m128i __DEFAULT_FN_ATTRS
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtepi64_epi16 (__mmask8 __M, __m256i __A)
{
return (__m128i) __builtin_ia32_pmovqw256_mask ((__v4di) __A,
__M);
}
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS256
_mm256_mask_cvtepi64_storeu_epi16 (void * __P, __mmask8 __M, __m256i __A)
{
__builtin_ia32_pmovqw256mem_mask ((__v8hi *) __P, (__v4di) __A, __M);
}
-#define _mm256_extractf32x4_ps(A, imm) __extension__ ({ \
- (__m128)__builtin_shufflevector((__v8sf)(__m256)(A), \
- (__v8sf)_mm256_undefined_ps(), \
- ((imm) & 1) ? 4 : 0, \
- ((imm) & 1) ? 5 : 1, \
- ((imm) & 1) ? 6 : 2, \
- ((imm) & 1) ? 7 : 3); })
-
-#define _mm256_mask_extractf32x4_ps(W, U, A, imm) __extension__ ({ \
- (__m128)__builtin_ia32_selectps_128((__mmask8)(U), \
- (__v4sf)_mm256_extractf32x4_ps((A), (imm)), \
- (__v4sf)(W)); })
-
-#define _mm256_maskz_extractf32x4_ps(U, A, imm) __extension__ ({ \
- (__m128)__builtin_ia32_selectps_128((__mmask8)(U), \
- (__v4sf)_mm256_extractf32x4_ps((A), (imm)), \
- (__v4sf)_mm_setzero_ps()); })
-
-#define _mm256_extracti32x4_epi32(A, imm) __extension__ ({ \
- (__m128i)__builtin_shufflevector((__v8si)(__m256)(A), \
- (__v8si)_mm256_undefined_si256(), \
- ((imm) & 1) ? 4 : 0, \
- ((imm) & 1) ? 5 : 1, \
- ((imm) & 1) ? 6 : 2, \
- ((imm) & 1) ? 7 : 3); })
-
-#define _mm256_mask_extracti32x4_epi32(W, U, A, imm) __extension__ ({ \
- (__m128i)__builtin_ia32_selectd_128((__mmask8)(U), \
- (__v4si)_mm256_extracti32x4_epi32((A), (imm)), \
- (__v4si)(W)); })
-
-#define _mm256_maskz_extracti32x4_epi32(U, A, imm) __extension__ ({ \
- (__m128i)__builtin_ia32_selectd_128((__mmask8)(U), \
- (__v4si)_mm256_extracti32x4_epi32((A), (imm)), \
- (__v4si)_mm_setzero_si128()); })
-
-#define _mm256_insertf32x4(A, B, imm) __extension__ ({ \
- (__m256)__builtin_shufflevector((__v8sf)(A), \
- (__v8sf)_mm256_castps128_ps256((__m128)(B)), \
- ((imm) & 0x1) ? 0 : 8, \
- ((imm) & 0x1) ? 1 : 9, \
- ((imm) & 0x1) ? 2 : 10, \
- ((imm) & 0x1) ? 3 : 11, \
- ((imm) & 0x1) ? 8 : 4, \
- ((imm) & 0x1) ? 9 : 5, \
- ((imm) & 0x1) ? 10 : 6, \
- ((imm) & 0x1) ? 11 : 7); })
-
-#define _mm256_mask_insertf32x4(W, U, A, B, imm) __extension__ ({ \
+#define _mm256_extractf32x4_ps(A, imm) \
+ (__m128)__builtin_ia32_extractf32x4_256_mask((__v8sf)(__m256)(A), \
+ (int)(imm), \
+ (__v4sf)_mm_undefined_ps(), \
+ (__mmask8)-1)
+
+#define _mm256_mask_extractf32x4_ps(W, U, A, imm) \
+ (__m128)__builtin_ia32_extractf32x4_256_mask((__v8sf)(__m256)(A), \
+ (int)(imm), \
+ (__v4sf)(__m128)(W), \
+ (__mmask8)(U))
+
+#define _mm256_maskz_extractf32x4_ps(U, A, imm) \
+ (__m128)__builtin_ia32_extractf32x4_256_mask((__v8sf)(__m256)(A), \
+ (int)(imm), \
+ (__v4sf)_mm_setzero_ps(), \
+ (__mmask8)(U))
+
+#define _mm256_extracti32x4_epi32(A, imm) \
+ (__m128i)__builtin_ia32_extracti32x4_256_mask((__v8si)(__m256i)(A), \
+ (int)(imm), \
+ (__v4si)_mm_undefined_si128(), \
+ (__mmask8)-1)
+
+#define _mm256_mask_extracti32x4_epi32(W, U, A, imm) \
+ (__m128i)__builtin_ia32_extracti32x4_256_mask((__v8si)(__m256i)(A), \
+ (int)(imm), \
+ (__v4si)(__m128i)(W), \
+ (__mmask8)(U))
+
+#define _mm256_maskz_extracti32x4_epi32(U, A, imm) \
+ (__m128i)__builtin_ia32_extracti32x4_256_mask((__v8si)(__m256i)(A), \
+ (int)(imm), \
+ (__v4si)_mm_setzero_si128(), \
+ (__mmask8)(U))
+
+#define _mm256_insertf32x4(A, B, imm) \
+ (__m256)__builtin_ia32_insertf32x4_256((__v8sf)(__m256)(A), \
+ (__v4sf)(__m128)(B), (int)(imm))
+
+#define _mm256_mask_insertf32x4(W, U, A, B, imm) \
(__m256)__builtin_ia32_selectps_256((__mmask8)(U), \
(__v8sf)_mm256_insertf32x4((A), (B), (imm)), \
- (__v8sf)(W)); })
+ (__v8sf)(__m256)(W))
-#define _mm256_maskz_insertf32x4(U, A, B, imm) __extension__ ({ \
+#define _mm256_maskz_insertf32x4(U, A, B, imm) \
(__m256)__builtin_ia32_selectps_256((__mmask8)(U), \
(__v8sf)_mm256_insertf32x4((A), (B), (imm)), \
- (__v8sf)_mm256_setzero_ps()); })
-
-#define _mm256_inserti32x4(A, B, imm) __extension__ ({ \
- (__m256i)__builtin_shufflevector((__v8si)(A), \
- (__v8si)_mm256_castsi128_si256((__m128i)(B)), \
- ((imm) & 0x1) ? 0 : 8, \
- ((imm) & 0x1) ? 1 : 9, \
- ((imm) & 0x1) ? 2 : 10, \
- ((imm) & 0x1) ? 3 : 11, \
- ((imm) & 0x1) ? 8 : 4, \
- ((imm) & 0x1) ? 9 : 5, \
- ((imm) & 0x1) ? 10 : 6, \
- ((imm) & 0x1) ? 11 : 7); })
-
-#define _mm256_mask_inserti32x4(W, U, A, B, imm) __extension__ ({ \
+ (__v8sf)_mm256_setzero_ps())
+
+#define _mm256_inserti32x4(A, B, imm) \
+ (__m256i)__builtin_ia32_inserti32x4_256((__v8si)(__m256i)(A), \
+ (__v4si)(__m128i)(B), (int)(imm))
+
+#define _mm256_mask_inserti32x4(W, U, A, B, imm) \
(__m256i)__builtin_ia32_selectd_256((__mmask8)(U), \
(__v8si)_mm256_inserti32x4((A), (B), (imm)), \
- (__v8si)(W)); })
+ (__v8si)(__m256i)(W))
-#define _mm256_maskz_inserti32x4(U, A, B, imm) __extension__ ({ \
+#define _mm256_maskz_inserti32x4(U, A, B, imm) \
(__m256i)__builtin_ia32_selectd_256((__mmask8)(U), \
(__v8si)_mm256_inserti32x4((A), (B), (imm)), \
- (__v8si)_mm256_setzero_si256()); })
+ (__v8si)_mm256_setzero_si256())
-#define _mm_getmant_pd(A, B, C) __extension__({\
+#define _mm_getmant_pd(A, B, C) \
(__m128d)__builtin_ia32_getmantpd128_mask((__v2df)(__m128d)(A), \
(int)(((C)<<2) | (B)), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_getmant_pd(W, U, A, B, C) __extension__({\
+#define _mm_mask_getmant_pd(W, U, A, B, C) \
(__m128d)__builtin_ia32_getmantpd128_mask((__v2df)(__m128d)(A), \
(int)(((C)<<2) | (B)), \
(__v2df)(__m128d)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_maskz_getmant_pd(U, A, B, C) __extension__({\
+#define _mm_maskz_getmant_pd(U, A, B, C) \
(__m128d)__builtin_ia32_getmantpd128_mask((__v2df)(__m128d)(A), \
(int)(((C)<<2) | (B)), \
(__v2df)_mm_setzero_pd(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_getmant_pd(A, B, C) __extension__ ({ \
+#define _mm256_getmant_pd(A, B, C) \
(__m256d)__builtin_ia32_getmantpd256_mask((__v4df)(__m256d)(A), \
(int)(((C)<<2) | (B)), \
(__v4df)_mm256_setzero_pd(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_getmant_pd(W, U, A, B, C) __extension__ ({ \
+#define _mm256_mask_getmant_pd(W, U, A, B, C) \
(__m256d)__builtin_ia32_getmantpd256_mask((__v4df)(__m256d)(A), \
(int)(((C)<<2) | (B)), \
(__v4df)(__m256d)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_maskz_getmant_pd(U, A, B, C) __extension__ ({ \
+#define _mm256_maskz_getmant_pd(U, A, B, C) \
(__m256d)__builtin_ia32_getmantpd256_mask((__v4df)(__m256d)(A), \
(int)(((C)<<2) | (B)), \
(__v4df)_mm256_setzero_pd(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_getmant_ps(A, B, C) __extension__ ({ \
+#define _mm_getmant_ps(A, B, C) \
(__m128)__builtin_ia32_getmantps128_mask((__v4sf)(__m128)(A), \
(int)(((C)<<2) | (B)), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm_mask_getmant_ps(W, U, A, B, C) __extension__ ({ \
+#define _mm_mask_getmant_ps(W, U, A, B, C) \
(__m128)__builtin_ia32_getmantps128_mask((__v4sf)(__m128)(A), \
(int)(((C)<<2) | (B)), \
(__v4sf)(__m128)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_maskz_getmant_ps(U, A, B, C) __extension__ ({ \
+#define _mm_maskz_getmant_ps(U, A, B, C) \
(__m128)__builtin_ia32_getmantps128_mask((__v4sf)(__m128)(A), \
(int)(((C)<<2) | (B)), \
(__v4sf)_mm_setzero_ps(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_getmant_ps(A, B, C) __extension__ ({ \
+#define _mm256_getmant_ps(A, B, C) \
(__m256)__builtin_ia32_getmantps256_mask((__v8sf)(__m256)(A), \
(int)(((C)<<2) | (B)), \
(__v8sf)_mm256_setzero_ps(), \
- (__mmask8)-1); })
+ (__mmask8)-1)
-#define _mm256_mask_getmant_ps(W, U, A, B, C) __extension__ ({ \
+#define _mm256_mask_getmant_ps(W, U, A, B, C) \
(__m256)__builtin_ia32_getmantps256_mask((__v8sf)(__m256)(A), \
(int)(((C)<<2) | (B)), \
(__v8sf)(__m256)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_maskz_getmant_ps(U, A, B, C) __extension__ ({ \
+#define _mm256_maskz_getmant_ps(U, A, B, C) \
(__m256)__builtin_ia32_getmantps256_mask((__v8sf)(__m256)(A), \
(int)(((C)<<2) | (B)), \
(__v8sf)_mm256_setzero_ps(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_mmask_i64gather_pd(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm_mmask_i64gather_pd(v1_old, mask, index, addr, scale) \
(__m128d)__builtin_ia32_gather3div2df((__v2df)(__m128d)(v1_old), \
- (double const *)(addr), \
+ (void const *)(addr), \
(__v2di)(__m128i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm_mmask_i64gather_epi64(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm_mmask_i64gather_epi64(v1_old, mask, index, addr, scale) \
(__m128i)__builtin_ia32_gather3div2di((__v2di)(__m128i)(v1_old), \
- (long long const *)(addr), \
+ (void const *)(addr), \
(__v2di)(__m128i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm256_mmask_i64gather_pd(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm256_mmask_i64gather_pd(v1_old, mask, index, addr, scale) \
(__m256d)__builtin_ia32_gather3div4df((__v4df)(__m256d)(v1_old), \
- (double const *)(addr), \
+ (void const *)(addr), \
(__v4di)(__m256i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm256_mmask_i64gather_epi64(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm256_mmask_i64gather_epi64(v1_old, mask, index, addr, scale) \
(__m256i)__builtin_ia32_gather3div4di((__v4di)(__m256i)(v1_old), \
- (long long const *)(addr), \
+ (void const *)(addr), \
(__v4di)(__m256i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm_mmask_i64gather_ps(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm_mmask_i64gather_ps(v1_old, mask, index, addr, scale) \
(__m128)__builtin_ia32_gather3div4sf((__v4sf)(__m128)(v1_old), \
- (float const *)(addr), \
+ (void const *)(addr), \
(__v2di)(__m128i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm_mmask_i64gather_epi32(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm_mmask_i64gather_epi32(v1_old, mask, index, addr, scale) \
(__m128i)__builtin_ia32_gather3div4si((__v4si)(__m128i)(v1_old), \
- (int const *)(addr), \
+ (void const *)(addr), \
(__v2di)(__m128i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm256_mmask_i64gather_ps(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm256_mmask_i64gather_ps(v1_old, mask, index, addr, scale) \
(__m128)__builtin_ia32_gather3div8sf((__v4sf)(__m128)(v1_old), \
- (float const *)(addr), \
+ (void const *)(addr), \
(__v4di)(__m256i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm256_mmask_i64gather_epi32(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm256_mmask_i64gather_epi32(v1_old, mask, index, addr, scale) \
(__m128i)__builtin_ia32_gather3div8si((__v4si)(__m128i)(v1_old), \
- (int const *)(addr), \
+ (void const *)(addr), \
(__v4di)(__m256i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm_mmask_i32gather_pd(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm_mmask_i32gather_pd(v1_old, mask, index, addr, scale) \
(__m128d)__builtin_ia32_gather3siv2df((__v2df)(__m128d)(v1_old), \
- (double const *)(addr), \
+ (void const *)(addr), \
(__v4si)(__m128i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm_mmask_i32gather_epi64(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm_mmask_i32gather_epi64(v1_old, mask, index, addr, scale) \
(__m128i)__builtin_ia32_gather3siv2di((__v2di)(__m128i)(v1_old), \
- (long long const *)(addr), \
+ (void const *)(addr), \
(__v4si)(__m128i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm256_mmask_i32gather_pd(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm256_mmask_i32gather_pd(v1_old, mask, index, addr, scale) \
(__m256d)__builtin_ia32_gather3siv4df((__v4df)(__m256d)(v1_old), \
- (double const *)(addr), \
+ (void const *)(addr), \
(__v4si)(__m128i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm256_mmask_i32gather_epi64(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm256_mmask_i32gather_epi64(v1_old, mask, index, addr, scale) \
(__m256i)__builtin_ia32_gather3siv4di((__v4di)(__m256i)(v1_old), \
- (long long const *)(addr), \
+ (void const *)(addr), \
(__v4si)(__m128i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm_mmask_i32gather_ps(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm_mmask_i32gather_ps(v1_old, mask, index, addr, scale) \
(__m128)__builtin_ia32_gather3siv4sf((__v4sf)(__m128)(v1_old), \
- (float const *)(addr), \
+ (void const *)(addr), \
(__v4si)(__m128i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm_mmask_i32gather_epi32(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm_mmask_i32gather_epi32(v1_old, mask, index, addr, scale) \
(__m128i)__builtin_ia32_gather3siv4si((__v4si)(__m128i)(v1_old), \
- (int const *)(addr), \
+ (void const *)(addr), \
(__v4si)(__m128i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm256_mmask_i32gather_ps(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm256_mmask_i32gather_ps(v1_old, mask, index, addr, scale) \
(__m256)__builtin_ia32_gather3siv8sf((__v8sf)(__m256)(v1_old), \
- (float const *)(addr), \
+ (void const *)(addr), \
(__v8si)(__m256i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm256_mmask_i32gather_epi32(v1_old, mask, index, addr, scale) __extension__ ({\
+#define _mm256_mmask_i32gather_epi32(v1_old, mask, index, addr, scale) \
(__m256i)__builtin_ia32_gather3siv8si((__v8si)(__m256i)(v1_old), \
- (int const *)(addr), \
+ (void const *)(addr), \
(__v8si)(__m256i)(index), \
- (__mmask8)(mask), (int)(scale)); })
+ (__mmask8)(mask), (int)(scale))
-#define _mm256_permutex_pd(X, C) __extension__ ({ \
- (__m256d)__builtin_shufflevector((__v4df)(__m256d)(X), \
- (__v4df)_mm256_undefined_pd(), \
- ((C) >> 0) & 0x3, ((C) >> 2) & 0x3, \
- ((C) >> 4) & 0x3, ((C) >> 6) & 0x3); })
+#define _mm256_permutex_pd(X, C) \
+ (__m256d)__builtin_ia32_permdf256((__v4df)(__m256d)(X), (int)(C))
-#define _mm256_mask_permutex_pd(W, U, X, C) __extension__ ({ \
+#define _mm256_mask_permutex_pd(W, U, X, C) \
(__m256d)__builtin_ia32_selectpd_256((__mmask8)(U), \
(__v4df)_mm256_permutex_pd((X), (C)), \
- (__v4df)(__m256d)(W)); })
+ (__v4df)(__m256d)(W))
-#define _mm256_maskz_permutex_pd(U, X, C) __extension__ ({ \
+#define _mm256_maskz_permutex_pd(U, X, C) \
(__m256d)__builtin_ia32_selectpd_256((__mmask8)(U), \
(__v4df)_mm256_permutex_pd((X), (C)), \
- (__v4df)_mm256_setzero_pd()); })
+ (__v4df)_mm256_setzero_pd())
-#define _mm256_permutex_epi64(X, C) __extension__ ({ \
- (__m256i)__builtin_shufflevector((__v4di)(__m256i)(X), \
- (__v4di)_mm256_undefined_si256(), \
- ((C) >> 0) & 0x3, ((C) >> 2) & 0x3, \
- ((C) >> 4) & 0x3, ((C) >> 6) & 0x3); })
+#define _mm256_permutex_epi64(X, C) \
+ (__m256i)__builtin_ia32_permdi256((__v4di)(__m256i)(X), (int)(C))
-#define _mm256_mask_permutex_epi64(W, U, X, C) __extension__ ({ \
+#define _mm256_mask_permutex_epi64(W, U, X, C) \
(__m256i)__builtin_ia32_selectq_256((__mmask8)(U), \
(__v4di)_mm256_permutex_epi64((X), (C)), \
- (__v4di)(__m256i)(W)); })
+ (__v4di)(__m256i)(W))
-#define _mm256_maskz_permutex_epi64(U, X, C) __extension__ ({ \
+#define _mm256_maskz_permutex_epi64(U, X, C) \
(__m256i)__builtin_ia32_selectq_256((__mmask8)(U), \
(__v4di)_mm256_permutex_epi64((X), (C)), \
- (__v4di)_mm256_setzero_si256()); })
+ (__v4di)_mm256_setzero_si256())
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_permutexvar_pd (__m256i __X, __m256d __Y)
{
- return (__m256d) __builtin_ia32_permvardf256_mask ((__v4df) __Y,
- (__v4di) __X,
- (__v4df) _mm256_undefined_si256 (),
- (__mmask8) -1);
+ return (__m256d)__builtin_ia32_permvardf256((__v4df)__Y, (__v4di)__X);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_permutexvar_pd (__m256d __W, __mmask8 __U, __m256i __X,
__m256d __Y)
{
- return (__m256d) __builtin_ia32_permvardf256_mask ((__v4df) __Y,
- (__v4di) __X,
- (__v4df) __W,
- (__mmask8) __U);
+ return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
+ (__v4df)_mm256_permutexvar_pd(__X, __Y),
+ (__v4df)__W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_permutexvar_pd (__mmask8 __U, __m256i __X, __m256d __Y)
{
- return (__m256d) __builtin_ia32_permvardf256_mask ((__v4df) __Y,
- (__v4di) __X,
- (__v4df) _mm256_setzero_pd (),
- (__mmask8) __U);
+ return (__m256d)__builtin_ia32_selectpd_256((__mmask8)__U,
+ (__v4df)_mm256_permutexvar_pd(__X, __Y),
+ (__v4df)_mm256_setzero_pd());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_permutexvar_epi64 (__mmask8 __M, __m256i __X, __m256i __Y)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_permutexvar_epi64 ( __m256i __X, __m256i __Y)
{
- return (__m256i) __builtin_ia32_permvardi256_mask ((__v4di) __Y,
- (__v4di) __X,
- (__v4di) _mm256_setzero_si256 (),
- (__mmask8) __M);
+ return (__m256i)__builtin_ia32_permvardi256((__v4di) __Y, (__v4di) __X);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_permutexvar_epi64 ( __m256i __X, __m256i __Y)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_permutexvar_epi64 (__mmask8 __M, __m256i __X, __m256i __Y)
{
- return (__m256i) __builtin_ia32_permvardi256_mask ((__v4di) __Y,
- (__v4di) __X,
- (__v4di) _mm256_undefined_si256 (),
- (__mmask8) -1);
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
+ (__v4di)_mm256_permutexvar_epi64(__X, __Y),
+ (__v4di)_mm256_setzero_si256());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_mask_permutexvar_epi64 (__m256i __W, __mmask8 __M, __m256i __X,
__m256i __Y)
{
- return (__m256i) __builtin_ia32_permvardi256_mask ((__v4di) __Y,
- (__v4di) __X,
- (__v4di) __W,
- __M);
+ return (__m256i)__builtin_ia32_selectq_256((__mmask8)__M,
+ (__v4di)_mm256_permutexvar_epi64(__X, __Y),
+ (__v4di)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_mask_permutexvar_ps (__m256 __W, __mmask8 __U, __m256i __X,
- __m256 __Y)
-{
- return (__m256) __builtin_ia32_permvarsf256_mask ((__v8sf) __Y,
- (__v8si) __X,
- (__v8sf) __W,
- (__mmask8) __U);
-}
+#define _mm256_permutexvar_ps(A, B) _mm256_permutevar8x32_ps((B), (A))
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_maskz_permutexvar_ps (__mmask8 __U, __m256i __X, __m256 __Y)
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_mask_permutexvar_ps(__m256 __W, __mmask8 __U, __m256i __X, __m256 __Y)
{
- return (__m256) __builtin_ia32_permvarsf256_mask ((__v8sf) __Y,
- (__v8si) __X,
- (__v8sf) _mm256_setzero_ps (),
- (__mmask8) __U);
+ return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
+ (__v8sf)_mm256_permutexvar_ps(__X, __Y),
+ (__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_permutexvar_ps (__m256i __X, __m256 __Y)
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_maskz_permutexvar_ps(__mmask8 __U, __m256i __X, __m256 __Y)
{
- return (__m256) __builtin_ia32_permvarsf256_mask ((__v8sf) __Y,
- (__v8si) __X,
- (__v8sf) _mm256_undefined_si256 (),
- (__mmask8) -1);
+ return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
+ (__v8sf)_mm256_permutexvar_ps(__X, __Y),
+ (__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_permutexvar_epi32 (__mmask8 __M, __m256i __X, __m256i __Y)
-{
- return (__m256i) __builtin_ia32_permvarsi256_mask ((__v8si) __Y,
- (__v8si) __X,
- (__v8si) _mm256_setzero_si256 (),
- __M);
-}
+#define _mm256_permutexvar_epi32(A, B) _mm256_permutevar8x32_epi32((B), (A))
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_permutexvar_epi32 (__m256i __W, __mmask8 __M, __m256i __X,
- __m256i __Y)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_permutexvar_epi32(__m256i __W, __mmask8 __M, __m256i __X,
+ __m256i __Y)
{
- return (__m256i) __builtin_ia32_permvarsi256_mask ((__v8si) __Y,
- (__v8si) __X,
- (__v8si) __W,
- (__mmask8) __M);
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
+ (__v8si)_mm256_permutexvar_epi32(__X, __Y),
+ (__v8si)__W);
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_permutexvar_epi32 (__m256i __X, __m256i __Y)
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_permutexvar_epi32(__mmask8 __M, __m256i __X, __m256i __Y)
{
- return (__m256i) __builtin_ia32_permvarsi256_mask ((__v8si) __Y,
- (__v8si) __X,
- (__v8si) _mm256_undefined_si256(),
- (__mmask8) -1);
+ return (__m256i)__builtin_ia32_selectd_256((__mmask8)__M,
+ (__v8si)_mm256_permutexvar_epi32(__X, __Y),
+ (__v8si)_mm256_setzero_si256());
}
-#define _mm_alignr_epi32(A, B, imm) __extension__ ({ \
- (__m128i)__builtin_shufflevector((__v4si)(__m128i)(B), \
- (__v4si)(__m128i)(A), \
- ((int)(imm) & 0x3) + 0, \
- ((int)(imm) & 0x3) + 1, \
- ((int)(imm) & 0x3) + 2, \
- ((int)(imm) & 0x3) + 3); })
+#define _mm_alignr_epi32(A, B, imm) \
+ (__m128i)__builtin_ia32_alignd128((__v4si)(__m128i)(A), \
+ (__v4si)(__m128i)(B), (int)(imm))
-#define _mm_mask_alignr_epi32(W, U, A, B, imm) __extension__ ({ \
+#define _mm_mask_alignr_epi32(W, U, A, B, imm) \
(__m128i)__builtin_ia32_selectd_128((__mmask8)(U), \
(__v4si)_mm_alignr_epi32((A), (B), (imm)), \
- (__v4si)(__m128i)(W)); })
+ (__v4si)(__m128i)(W))
-#define _mm_maskz_alignr_epi32(U, A, B, imm) __extension__ ({ \
+#define _mm_maskz_alignr_epi32(U, A, B, imm) \
(__m128i)__builtin_ia32_selectd_128((__mmask8)(U), \
(__v4si)_mm_alignr_epi32((A), (B), (imm)), \
- (__v4si)_mm_setzero_si128()); })
-
-#define _mm256_alignr_epi32(A, B, imm) __extension__ ({ \
- (__m256i)__builtin_shufflevector((__v8si)(__m256i)(B), \
- (__v8si)(__m256i)(A), \
- ((int)(imm) & 0x7) + 0, \
- ((int)(imm) & 0x7) + 1, \
- ((int)(imm) & 0x7) + 2, \
- ((int)(imm) & 0x7) + 3, \
- ((int)(imm) & 0x7) + 4, \
- ((int)(imm) & 0x7) + 5, \
- ((int)(imm) & 0x7) + 6, \
- ((int)(imm) & 0x7) + 7); })
-
-#define _mm256_mask_alignr_epi32(W, U, A, B, imm) __extension__ ({ \
+ (__v4si)_mm_setzero_si128())
+
+#define _mm256_alignr_epi32(A, B, imm) \
+ (__m256i)__builtin_ia32_alignd256((__v8si)(__m256i)(A), \
+ (__v8si)(__m256i)(B), (int)(imm))
+
+#define _mm256_mask_alignr_epi32(W, U, A, B, imm) \
(__m256i)__builtin_ia32_selectd_256((__mmask8)(U), \
(__v8si)_mm256_alignr_epi32((A), (B), (imm)), \
- (__v8si)(__m256i)(W)); })
+ (__v8si)(__m256i)(W))
-#define _mm256_maskz_alignr_epi32(U, A, B, imm) __extension__ ({ \
+#define _mm256_maskz_alignr_epi32(U, A, B, imm) \
(__m256i)__builtin_ia32_selectd_256((__mmask8)(U), \
(__v8si)_mm256_alignr_epi32((A), (B), (imm)), \
- (__v8si)_mm256_setzero_si256()); })
+ (__v8si)_mm256_setzero_si256())
-#define _mm_alignr_epi64(A, B, imm) __extension__ ({ \
- (__m128i)__builtin_shufflevector((__v2di)(__m128i)(B), \
- (__v2di)(__m128i)(A), \
- ((int)(imm) & 0x1) + 0, \
- ((int)(imm) & 0x1) + 1); })
+#define _mm_alignr_epi64(A, B, imm) \
+ (__m128i)__builtin_ia32_alignq128((__v2di)(__m128i)(A), \
+ (__v2di)(__m128i)(B), (int)(imm))
-#define _mm_mask_alignr_epi64(W, U, A, B, imm) __extension__ ({ \
+#define _mm_mask_alignr_epi64(W, U, A, B, imm) \
(__m128i)__builtin_ia32_selectq_128((__mmask8)(U), \
(__v2di)_mm_alignr_epi64((A), (B), (imm)), \
- (__v2di)(__m128i)(W)); })
+ (__v2di)(__m128i)(W))
-#define _mm_maskz_alignr_epi64(U, A, B, imm) __extension__ ({ \
+#define _mm_maskz_alignr_epi64(U, A, B, imm) \
(__m128i)__builtin_ia32_selectq_128((__mmask8)(U), \
(__v2di)_mm_alignr_epi64((A), (B), (imm)), \
- (__v2di)_mm_setzero_di()); })
+ (__v2di)_mm_setzero_si128())
-#define _mm256_alignr_epi64(A, B, imm) __extension__ ({ \
- (__m256i)__builtin_shufflevector((__v4di)(__m256i)(B), \
- (__v4di)(__m256i)(A), \
- ((int)(imm) & 0x3) + 0, \
- ((int)(imm) & 0x3) + 1, \
- ((int)(imm) & 0x3) + 2, \
- ((int)(imm) & 0x3) + 3); })
+#define _mm256_alignr_epi64(A, B, imm) \
+ (__m256i)__builtin_ia32_alignq256((__v4di)(__m256i)(A), \
+ (__v4di)(__m256i)(B), (int)(imm))
-#define _mm256_mask_alignr_epi64(W, U, A, B, imm) __extension__ ({ \
+#define _mm256_mask_alignr_epi64(W, U, A, B, imm) \
(__m256i)__builtin_ia32_selectq_256((__mmask8)(U), \
(__v4di)_mm256_alignr_epi64((A), (B), (imm)), \
- (__v4di)(__m256i)(W)); })
+ (__v4di)(__m256i)(W))
-#define _mm256_maskz_alignr_epi64(U, A, B, imm) __extension__ ({ \
+#define _mm256_maskz_alignr_epi64(U, A, B, imm) \
(__m256i)__builtin_ia32_selectq_256((__mmask8)(U), \
(__v4di)_mm256_alignr_epi64((A), (B), (imm)), \
- (__v4di)_mm256_setzero_si256()); })
+ (__v4di)_mm256_setzero_si256())
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_movehdup_ps (__m128 __W, __mmask8 __U, __m128 __A)
{
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_movehdup_ps (__mmask8 __U, __m128 __A)
{
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_movehdup_ps (__m256 __W, __mmask8 __U, __m256 __A)
{
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_movehdup_ps (__mmask8 __U, __m256 __A)
{
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_setzero_ps());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_moveldup_ps (__m128 __W, __mmask8 __U, __m128 __A)
{
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)__W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_moveldup_ps (__mmask8 __U, __m128 __A)
{
return (__m128)__builtin_ia32_selectps_128((__mmask8)__U,
(__v4sf)_mm_setzero_ps());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_moveldup_ps (__m256 __W, __mmask8 __U, __m256 __A)
{
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)__W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_moveldup_ps (__mmask8 __U, __m256 __A)
{
return (__m256)__builtin_ia32_selectps_256((__mmask8)__U,
(__v8sf)_mm256_setzero_ps());
}
-#define _mm256_mask_shuffle_epi32(W, U, A, I) __extension__({\
+#define _mm256_mask_shuffle_epi32(W, U, A, I) \
(__m256i)__builtin_ia32_selectd_256((__mmask8)(U), \
(__v8si)_mm256_shuffle_epi32((A), (I)), \
- (__v8si)(__m256i)(W)); })
+ (__v8si)(__m256i)(W))
-#define _mm256_maskz_shuffle_epi32(U, A, I) __extension__({\
+#define _mm256_maskz_shuffle_epi32(U, A, I) \
(__m256i)__builtin_ia32_selectd_256((__mmask8)(U), \
(__v8si)_mm256_shuffle_epi32((A), (I)), \
- (__v8si)_mm256_setzero_si256()); })
+ (__v8si)_mm256_setzero_si256())
-#define _mm_mask_shuffle_epi32(W, U, A, I) __extension__({\
+#define _mm_mask_shuffle_epi32(W, U, A, I) \
(__m128i)__builtin_ia32_selectd_128((__mmask8)(U), \
(__v4si)_mm_shuffle_epi32((A), (I)), \
- (__v4si)(__m128i)(W)); })
+ (__v4si)(__m128i)(W))
-#define _mm_maskz_shuffle_epi32(U, A, I) __extension__({\
+#define _mm_maskz_shuffle_epi32(U, A, I) \
(__m128i)__builtin_ia32_selectd_128((__mmask8)(U), \
(__v4si)_mm_shuffle_epi32((A), (I)), \
- (__v4si)_mm_setzero_si128()); })
+ (__v4si)_mm_setzero_si128())
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_mov_pd (__m128d __W, __mmask8 __U, __m128d __A)
{
return (__m128d) __builtin_ia32_selectpd_128 ((__mmask8) __U,
(__v2df) __W);
}
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_mov_pd (__mmask8 __U, __m128d __A)
{
return (__m128d) __builtin_ia32_selectpd_128 ((__mmask8) __U,
(__v2df) _mm_setzero_pd ());
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_mask_mov_pd (__m256d __W, __mmask8 __U, __m256d __A)
{
return (__m256d) __builtin_ia32_selectpd_256 ((__mmask8) __U,
(__v4df) __W);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_maskz_mov_pd (__mmask8 __U, __m256d __A)
{
return (__m256d) __builtin_ia32_selectpd_256 ((__mmask8) __U,
(__v4df) _mm256_setzero_pd ());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_mov_ps (__m128 __W, __mmask8 __U, __m128 __A)
{
return (__m128) __builtin_ia32_selectps_128 ((__mmask8) __U,
(__v4sf) __W);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_mov_ps (__mmask8 __U, __m128 __A)
{
return (__m128) __builtin_ia32_selectps_128 ((__mmask8) __U,
(__v4sf) _mm_setzero_ps ());
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_mov_ps (__m256 __W, __mmask8 __U, __m256 __A)
{
return (__m256) __builtin_ia32_selectps_256 ((__mmask8) __U,
(__v8sf) __W);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_mov_ps (__mmask8 __U, __m256 __A)
{
return (__m256) __builtin_ia32_selectps_256 ((__mmask8) __U,
(__v8sf) _mm256_setzero_ps ());
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_cvtph_ps (__m128 __W, __mmask8 __U, __m128i __A)
{
return (__m128) __builtin_ia32_vcvtph2ps_mask ((__v8hi) __A,
(__mmask8) __U);
}
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_cvtph_ps (__mmask8 __U, __m128i __A)
{
return (__m128) __builtin_ia32_vcvtph2ps_mask ((__v8hi) __A,
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_mask_cvtph_ps (__m256 __W, __mmask8 __U, __m128i __A)
{
return (__m256) __builtin_ia32_vcvtph2ps256_mask ((__v8hi) __A,
(__mmask8) __U);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_maskz_cvtph_ps (__mmask8 __U, __m128i __A)
{
return (__m256) __builtin_ia32_vcvtph2ps256_mask ((__v8hi) __A,
(__mmask8) __U);
}
-static __inline __m128i __DEFAULT_FN_ATTRS
-_mm_mask_cvtps_ph (__m128i __W, __mmask8 __U, __m128 __A)
-{
- return (__m128i) __builtin_ia32_vcvtps2ph_mask ((__v4sf) __A, _MM_FROUND_CUR_DIRECTION,
- (__v8hi) __W,
- (__mmask8) __U);
-}
-
-static __inline __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_cvtps_ph (__mmask8 __U, __m128 __A)
-{
- return (__m128i) __builtin_ia32_vcvtps2ph_mask ((__v4sf) __A, _MM_FROUND_CUR_DIRECTION,
- (__v8hi) _mm_setzero_si128 (),
- (__mmask8) __U);
-}
-
-#define _mm_mask_cvt_roundps_ph(W, U, A, I) __extension__ ({ \
+#define _mm_mask_cvt_roundps_ph(W, U, A, I) \
(__m128i)__builtin_ia32_vcvtps2ph_mask((__v4sf)(__m128)(A), (int)(I), \
(__v8hi)(__m128i)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm_maskz_cvt_roundps_ph(U, A, I) __extension__ ({ \
+#define _mm_maskz_cvt_roundps_ph(U, A, I) \
(__m128i)__builtin_ia32_vcvtps2ph_mask((__v4sf)(__m128)(A), (int)(I), \
(__v8hi)_mm_setzero_si128(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-static __inline __m128i __DEFAULT_FN_ATTRS
-_mm256_mask_cvtps_ph (__m128i __W, __mmask8 __U, __m256 __A)
-{
- return (__m128i) __builtin_ia32_vcvtps2ph256_mask ((__v8sf) __A, _MM_FROUND_CUR_DIRECTION,
- (__v8hi) __W,
- (__mmask8) __U);
-}
+#define _mm_mask_cvtps_ph _mm_mask_cvt_roundps_ph
+#define _mm_maskz_cvtps_ph _mm_maskz_cvt_roundps_ph
-static __inline __m128i __DEFAULT_FN_ATTRS
-_mm256_maskz_cvtps_ph ( __mmask8 __U, __m256 __A)
-{
- return (__m128i) __builtin_ia32_vcvtps2ph256_mask ((__v8sf) __A, _MM_FROUND_CUR_DIRECTION,
- (__v8hi) _mm_setzero_si128(),
- (__mmask8) __U);
-}
-#define _mm256_mask_cvt_roundps_ph(W, U, A, I) __extension__ ({ \
+#define _mm256_mask_cvt_roundps_ph(W, U, A, I) \
(__m128i)__builtin_ia32_vcvtps2ph256_mask((__v8sf)(__m256)(A), (int)(I), \
(__v8hi)(__m128i)(W), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
-#define _mm256_maskz_cvt_roundps_ph(U, A, I) __extension__ ({ \
+#define _mm256_maskz_cvt_roundps_ph(U, A, I) \
(__m128i)__builtin_ia32_vcvtps2ph256_mask((__v8sf)(__m256)(A), (int)(I), \
(__v8hi)_mm_setzero_si128(), \
- (__mmask8)(U)); })
+ (__mmask8)(U))
+
+#define _mm256_mask_cvtps_ph _mm256_mask_cvt_roundps_ph
+#define _mm256_maskz_cvtps_ph _mm256_maskz_cvt_roundps_ph
-#undef __DEFAULT_FN_ATTRS
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
#endif /* __AVX512VLINTRIN_H */
--- /dev/null
+/*===------------- avx512vlvbmi2intrin.h - VBMI2 intrinsics -----------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512vlvbmi2intrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __AVX512VLVBMI2INTRIN_H
+#define __AVX512VLVBMI2INTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512vbmi2"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512vbmi2"), __min_vector_width__(256)))
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_compress_epi16(__m128i __S, __mmask8 __U, __m128i __D)
+{
+ return (__m128i) __builtin_ia32_compresshi128_mask ((__v8hi) __D,
+ (__v8hi) __S,
+ __U);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_compress_epi16(__mmask8 __U, __m128i __D)
+{
+ return (__m128i) __builtin_ia32_compresshi128_mask ((__v8hi) __D,
+ (__v8hi) _mm_setzero_si128(),
+ __U);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_compress_epi8(__m128i __S, __mmask16 __U, __m128i __D)
+{
+ return (__m128i) __builtin_ia32_compressqi128_mask ((__v16qi) __D,
+ (__v16qi) __S,
+ __U);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_compress_epi8(__mmask16 __U, __m128i __D)
+{
+ return (__m128i) __builtin_ia32_compressqi128_mask ((__v16qi) __D,
+ (__v16qi) _mm_setzero_si128(),
+ __U);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS128
+_mm_mask_compressstoreu_epi16(void *__P, __mmask8 __U, __m128i __D)
+{
+ __builtin_ia32_compressstorehi128_mask ((__v8hi *) __P, (__v8hi) __D,
+ __U);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS128
+_mm_mask_compressstoreu_epi8(void *__P, __mmask16 __U, __m128i __D)
+{
+ __builtin_ia32_compressstoreqi128_mask ((__v16qi *) __P, (__v16qi) __D,
+ __U);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_expand_epi16(__m128i __S, __mmask8 __U, __m128i __D)
+{
+ return (__m128i) __builtin_ia32_expandhi128_mask ((__v8hi) __D,
+ (__v8hi) __S,
+ __U);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_expand_epi16(__mmask8 __U, __m128i __D)
+{
+ return (__m128i) __builtin_ia32_expandhi128_mask ((__v8hi) __D,
+ (__v8hi) _mm_setzero_si128(),
+ __U);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_expand_epi8(__m128i __S, __mmask16 __U, __m128i __D)
+{
+ return (__m128i) __builtin_ia32_expandqi128_mask ((__v16qi) __D,
+ (__v16qi) __S,
+ __U);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_expand_epi8(__mmask16 __U, __m128i __D)
+{
+ return (__m128i) __builtin_ia32_expandqi128_mask ((__v16qi) __D,
+ (__v16qi) _mm_setzero_si128(),
+ __U);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_expandloadu_epi16(__m128i __S, __mmask8 __U, void const *__P)
+{
+ return (__m128i) __builtin_ia32_expandloadhi128_mask ((const __v8hi *)__P,
+ (__v8hi) __S,
+ __U);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_expandloadu_epi16(__mmask8 __U, void const *__P)
+{
+ return (__m128i) __builtin_ia32_expandloadhi128_mask ((const __v8hi *)__P,
+ (__v8hi) _mm_setzero_si128(),
+ __U);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_expandloadu_epi8(__m128i __S, __mmask16 __U, void const *__P)
+{
+ return (__m128i) __builtin_ia32_expandloadqi128_mask ((const __v16qi *)__P,
+ (__v16qi) __S,
+ __U);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_expandloadu_epi8(__mmask16 __U, void const *__P)
+{
+ return (__m128i) __builtin_ia32_expandloadqi128_mask ((const __v16qi *)__P,
+ (__v16qi) _mm_setzero_si128(),
+ __U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_compress_epi16(__m256i __S, __mmask16 __U, __m256i __D)
+{
+ return (__m256i) __builtin_ia32_compresshi256_mask ((__v16hi) __D,
+ (__v16hi) __S,
+ __U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_compress_epi16(__mmask16 __U, __m256i __D)
+{
+ return (__m256i) __builtin_ia32_compresshi256_mask ((__v16hi) __D,
+ (__v16hi) _mm256_setzero_si256(),
+ __U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_compress_epi8(__m256i __S, __mmask32 __U, __m256i __D)
+{
+ return (__m256i) __builtin_ia32_compressqi256_mask ((__v32qi) __D,
+ (__v32qi) __S,
+ __U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_compress_epi8(__mmask32 __U, __m256i __D)
+{
+ return (__m256i) __builtin_ia32_compressqi256_mask ((__v32qi) __D,
+ (__v32qi) _mm256_setzero_si256(),
+ __U);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS256
+_mm256_mask_compressstoreu_epi16(void *__P, __mmask16 __U, __m256i __D)
+{
+ __builtin_ia32_compressstorehi256_mask ((__v16hi *) __P, (__v16hi) __D,
+ __U);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS256
+_mm256_mask_compressstoreu_epi8(void *__P, __mmask32 __U, __m256i __D)
+{
+ __builtin_ia32_compressstoreqi256_mask ((__v32qi *) __P, (__v32qi) __D,
+ __U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_expand_epi16(__m256i __S, __mmask16 __U, __m256i __D)
+{
+ return (__m256i) __builtin_ia32_expandhi256_mask ((__v16hi) __D,
+ (__v16hi) __S,
+ __U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_expand_epi16(__mmask16 __U, __m256i __D)
+{
+ return (__m256i) __builtin_ia32_expandhi256_mask ((__v16hi) __D,
+ (__v16hi) _mm256_setzero_si256(),
+ __U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_expand_epi8(__m256i __S, __mmask32 __U, __m256i __D)
+{
+ return (__m256i) __builtin_ia32_expandqi256_mask ((__v32qi) __D,
+ (__v32qi) __S,
+ __U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_expand_epi8(__mmask32 __U, __m256i __D)
+{
+ return (__m256i) __builtin_ia32_expandqi256_mask ((__v32qi) __D,
+ (__v32qi) _mm256_setzero_si256(),
+ __U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_expandloadu_epi16(__m256i __S, __mmask16 __U, void const *__P)
+{
+ return (__m256i) __builtin_ia32_expandloadhi256_mask ((const __v16hi *)__P,
+ (__v16hi) __S,
+ __U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_expandloadu_epi16(__mmask16 __U, void const *__P)
+{
+ return (__m256i) __builtin_ia32_expandloadhi256_mask ((const __v16hi *)__P,
+ (__v16hi) _mm256_setzero_si256(),
+ __U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_expandloadu_epi8(__m256i __S, __mmask32 __U, void const *__P)
+{
+ return (__m256i) __builtin_ia32_expandloadqi256_mask ((const __v32qi *)__P,
+ (__v32qi) __S,
+ __U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_expandloadu_epi8(__mmask32 __U, void const *__P)
+{
+ return (__m256i) __builtin_ia32_expandloadqi256_mask ((const __v32qi *)__P,
+ (__v32qi) _mm256_setzero_si256(),
+ __U);
+}
+
+#define _mm256_shldi_epi64(A, B, I) \
+ (__m256i)__builtin_ia32_vpshldq256((__v4di)(__m256i)(A), \
+ (__v4di)(__m256i)(B), (int)(I))
+
+#define _mm256_mask_shldi_epi64(S, U, A, B, I) \
+ (__m256i)__builtin_ia32_selectq_256((__mmask8)(U), \
+ (__v4di)_mm256_shldi_epi64((A), (B), (I)), \
+ (__v4di)(__m256i)(S))
+
+#define _mm256_maskz_shldi_epi64(U, A, B, I) \
+ (__m256i)__builtin_ia32_selectq_256((__mmask8)(U), \
+ (__v4di)_mm256_shldi_epi64((A), (B), (I)), \
+ (__v4di)_mm256_setzero_si256())
+
+#define _mm_shldi_epi64(A, B, I) \
+ (__m128i)__builtin_ia32_vpshldq128((__v2di)(__m128i)(A), \
+ (__v2di)(__m128i)(B), (int)(I))
+
+#define _mm_mask_shldi_epi64(S, U, A, B, I) \
+ (__m128i)__builtin_ia32_selectq_128((__mmask8)(U), \
+ (__v2di)_mm_shldi_epi64((A), (B), (I)), \
+ (__v2di)(__m128i)(S))
+
+#define _mm_maskz_shldi_epi64(U, A, B, I) \
+ (__m128i)__builtin_ia32_selectq_128((__mmask8)(U), \
+ (__v2di)_mm_shldi_epi64((A), (B), (I)), \
+ (__v2di)_mm_setzero_si128())
+
+#define _mm256_shldi_epi32(A, B, I) \
+ (__m256i)__builtin_ia32_vpshldd256((__v8si)(__m256i)(A), \
+ (__v8si)(__m256i)(B), (int)(I))
+
+#define _mm256_mask_shldi_epi32(S, U, A, B, I) \
+ (__m256i)__builtin_ia32_selectd_256((__mmask8)(U), \
+ (__v8si)_mm256_shldi_epi32((A), (B), (I)), \
+ (__v8si)(__m256i)(S))
+
+#define _mm256_maskz_shldi_epi32(U, A, B, I) \
+ (__m256i)__builtin_ia32_selectd_256((__mmask8)(U), \
+ (__v8si)_mm256_shldi_epi32((A), (B), (I)), \
+ (__v8si)_mm256_setzero_si256())
+
+#define _mm_shldi_epi32(A, B, I) \
+ (__m128i)__builtin_ia32_vpshldd128((__v4si)(__m128i)(A), \
+ (__v4si)(__m128i)(B), (int)(I))
+
+#define _mm_mask_shldi_epi32(S, U, A, B, I) \
+ (__m128i)__builtin_ia32_selectd_128((__mmask8)(U), \
+ (__v4si)_mm_shldi_epi32((A), (B), (I)), \
+ (__v4si)(__m128i)(S))
+
+#define _mm_maskz_shldi_epi32(U, A, B, I) \
+ (__m128i)__builtin_ia32_selectd_128((__mmask8)(U), \
+ (__v4si)_mm_shldi_epi32((A), (B), (I)), \
+ (__v4si)_mm_setzero_si128())
+
+#define _mm256_shldi_epi16(A, B, I) \
+ (__m256i)__builtin_ia32_vpshldw256((__v16hi)(__m256i)(A), \
+ (__v16hi)(__m256i)(B), (int)(I))
+
+#define _mm256_mask_shldi_epi16(S, U, A, B, I) \
+ (__m256i)__builtin_ia32_selectw_256((__mmask16)(U), \
+ (__v16hi)_mm256_shldi_epi16((A), (B), (I)), \
+ (__v16hi)(__m256i)(S))
+
+#define _mm256_maskz_shldi_epi16(U, A, B, I) \
+ (__m256i)__builtin_ia32_selectw_256((__mmask16)(U), \
+ (__v16hi)_mm256_shldi_epi16((A), (B), (I)), \
+ (__v16hi)_mm256_setzero_si256())
+
+#define _mm_shldi_epi16(A, B, I) \
+ (__m128i)__builtin_ia32_vpshldw128((__v8hi)(__m128i)(A), \
+ (__v8hi)(__m128i)(B), (int)(I))
+
+#define _mm_mask_shldi_epi16(S, U, A, B, I) \
+ (__m128i)__builtin_ia32_selectw_128((__mmask8)(U), \
+ (__v8hi)_mm_shldi_epi16((A), (B), (I)), \
+ (__v8hi)(__m128i)(S))
+
+#define _mm_maskz_shldi_epi16(U, A, B, I) \
+ (__m128i)__builtin_ia32_selectw_128((__mmask8)(U), \
+ (__v8hi)_mm_shldi_epi16((A), (B), (I)), \
+ (__v8hi)_mm_setzero_si128())
+
+#define _mm256_shrdi_epi64(A, B, I) \
+ (__m256i)__builtin_ia32_vpshrdq256((__v4di)(__m256i)(A), \
+ (__v4di)(__m256i)(B), (int)(I))
+
+#define _mm256_mask_shrdi_epi64(S, U, A, B, I) \
+ (__m256i)__builtin_ia32_selectq_256((__mmask8)(U), \
+ (__v4di)_mm256_shrdi_epi64((A), (B), (I)), \
+ (__v4di)(__m256i)(S))
+
+#define _mm256_maskz_shrdi_epi64(U, A, B, I) \
+ (__m256i)__builtin_ia32_selectq_256((__mmask8)(U), \
+ (__v4di)_mm256_shrdi_epi64((A), (B), (I)), \
+ (__v4di)_mm256_setzero_si256())
+
+#define _mm_shrdi_epi64(A, B, I) \
+ (__m128i)__builtin_ia32_vpshrdq128((__v2di)(__m128i)(A), \
+ (__v2di)(__m128i)(B), (int)(I))
+
+#define _mm_mask_shrdi_epi64(S, U, A, B, I) \
+ (__m128i)__builtin_ia32_selectq_128((__mmask8)(U), \
+ (__v2di)_mm_shrdi_epi64((A), (B), (I)), \
+ (__v2di)(__m128i)(S))
+
+#define _mm_maskz_shrdi_epi64(U, A, B, I) \
+ (__m128i)__builtin_ia32_selectq_128((__mmask8)(U), \
+ (__v2di)_mm_shrdi_epi64((A), (B), (I)), \
+ (__v2di)_mm_setzero_si128())
+
+#define _mm256_shrdi_epi32(A, B, I) \
+ (__m256i)__builtin_ia32_vpshrdd256((__v8si)(__m256i)(A), \
+ (__v8si)(__m256i)(B), (int)(I))
+
+#define _mm256_mask_shrdi_epi32(S, U, A, B, I) \
+ (__m256i)__builtin_ia32_selectd_256((__mmask8)(U), \
+ (__v8si)_mm256_shrdi_epi32((A), (B), (I)), \
+ (__v8si)(__m256i)(S))
+
+#define _mm256_maskz_shrdi_epi32(U, A, B, I) \
+ (__m256i)__builtin_ia32_selectd_256((__mmask8)(U), \
+ (__v8si)_mm256_shrdi_epi32((A), (B), (I)), \
+ (__v8si)_mm256_setzero_si256())
+
+#define _mm_shrdi_epi32(A, B, I) \
+ (__m128i)__builtin_ia32_vpshrdd128((__v4si)(__m128i)(A), \
+ (__v4si)(__m128i)(B), (int)(I))
+
+#define _mm_mask_shrdi_epi32(S, U, A, B, I) \
+ (__m128i)__builtin_ia32_selectd_128((__mmask8)(U), \
+ (__v4si)_mm_shrdi_epi32((A), (B), (I)), \
+ (__v4si)(__m128i)(S))
+
+#define _mm_maskz_shrdi_epi32(U, A, B, I) \
+ (__m128i)__builtin_ia32_selectd_128((__mmask8)(U), \
+ (__v4si)_mm_shrdi_epi32((A), (B), (I)), \
+ (__v4si)_mm_setzero_si128())
+
+#define _mm256_shrdi_epi16(A, B, I) \
+ (__m256i)__builtin_ia32_vpshrdw256((__v16hi)(__m256i)(A), \
+ (__v16hi)(__m256i)(B), (int)(I))
+
+#define _mm256_mask_shrdi_epi16(S, U, A, B, I) \
+ (__m256i)__builtin_ia32_selectw_256((__mmask16)(U), \
+ (__v16hi)_mm256_shrdi_epi16((A), (B), (I)), \
+ (__v16hi)(__m256i)(S))
+
+#define _mm256_maskz_shrdi_epi16(U, A, B, I) \
+ (__m256i)__builtin_ia32_selectw_256((__mmask16)(U), \
+ (__v16hi)_mm256_shrdi_epi16((A), (B), (I)), \
+ (__v16hi)_mm256_setzero_si256())
+
+#define _mm_shrdi_epi16(A, B, I) \
+ (__m128i)__builtin_ia32_vpshrdw128((__v8hi)(__m128i)(A), \
+ (__v8hi)(__m128i)(B), (int)(I))
+
+#define _mm_mask_shrdi_epi16(S, U, A, B, I) \
+ (__m128i)__builtin_ia32_selectw_128((__mmask8)(U), \
+ (__v8hi)_mm_shrdi_epi16((A), (B), (I)), \
+ (__v8hi)(__m128i)(S))
+
+#define _mm_maskz_shrdi_epi16(U, A, B, I) \
+ (__m128i)__builtin_ia32_selectw_128((__mmask8)(U), \
+ (__v8hi)_mm_shrdi_epi16((A), (B), (I)), \
+ (__v8hi)_mm_setzero_si128())
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_shldv_epi64(__m256i __A, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_vpshldvq256((__v4di)__A, (__v4di)__B,
+ (__v4di)__C);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_shldv_epi64(__m256i __A, __mmask8 __U, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_selectq_256(__U,
+ (__v4di)_mm256_shldv_epi64(__A, __B, __C),
+ (__v4di)__A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_shldv_epi64(__mmask8 __U, __m256i __A, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_selectq_256(__U,
+ (__v4di)_mm256_shldv_epi64(__A, __B, __C),
+ (__v4di)_mm256_setzero_si256());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_shldv_epi64(__m128i __A, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_vpshldvq128((__v2di)__A, (__v2di)__B,
+ (__v2di)__C);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_shldv_epi64(__m128i __A, __mmask8 __U, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_selectq_128(__U,
+ (__v2di)_mm_shldv_epi64(__A, __B, __C),
+ (__v2di)__A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_shldv_epi64(__mmask8 __U, __m128i __A, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_selectq_128(__U,
+ (__v2di)_mm_shldv_epi64(__A, __B, __C),
+ (__v2di)_mm_setzero_si128());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_shldv_epi32(__m256i __A, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_vpshldvd256((__v8si)__A, (__v8si)__B,
+ (__v8si)__C);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_shldv_epi32(__m256i __A, __mmask8 __U, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_shldv_epi32(__A, __B, __C),
+ (__v8si)__A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_shldv_epi32(__mmask8 __U, __m256i __A, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_shldv_epi32(__A, __B, __C),
+ (__v8si)_mm256_setzero_si256());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_shldv_epi32(__m128i __A, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_vpshldvd128((__v4si)__A, (__v4si)__B,
+ (__v4si)__C);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_shldv_epi32(__m128i __A, __mmask8 __U, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_shldv_epi32(__A, __B, __C),
+ (__v4si)__A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_shldv_epi32(__mmask8 __U, __m128i __A, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_shldv_epi32(__A, __B, __C),
+ (__v4si)_mm_setzero_si128());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_shldv_epi16(__m256i __A, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_vpshldvw256((__v16hi)__A, (__v16hi)__B,
+ (__v16hi)__C);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_shldv_epi16(__m256i __A, __mmask16 __U, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_selectw_256(__U,
+ (__v16hi)_mm256_shldv_epi16(__A, __B, __C),
+ (__v16hi)__A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_shldv_epi16(__mmask16 __U, __m256i __A, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_selectw_256(__U,
+ (__v16hi)_mm256_shldv_epi16(__A, __B, __C),
+ (__v16hi)_mm256_setzero_si256());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_shldv_epi16(__m128i __A, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_vpshldvw128((__v8hi)__A, (__v8hi)__B,
+ (__v8hi)__C);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_shldv_epi16(__m128i __A, __mmask8 __U, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_selectw_128(__U,
+ (__v8hi)_mm_shldv_epi16(__A, __B, __C),
+ (__v8hi)__A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_shldv_epi16(__mmask8 __U, __m128i __A, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_selectw_128(__U,
+ (__v8hi)_mm_shldv_epi16(__A, __B, __C),
+ (__v8hi)_mm_setzero_si128());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_shrdv_epi64(__m256i __A, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_vpshrdvq256((__v4di)__A, (__v4di)__B,
+ (__v4di)__C);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_shrdv_epi64(__m256i __A, __mmask8 __U, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_selectq_256(__U,
+ (__v4di)_mm256_shrdv_epi64(__A, __B, __C),
+ (__v4di)__A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_shrdv_epi64(__mmask8 __U, __m256i __A, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_selectq_256(__U,
+ (__v4di)_mm256_shrdv_epi64(__A, __B, __C),
+ (__v4di)_mm256_setzero_si256());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_shrdv_epi64(__m128i __A, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_vpshrdvq128((__v2di)__A, (__v2di)__B,
+ (__v2di)__C);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_shrdv_epi64(__m128i __A, __mmask8 __U, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_selectq_128(__U,
+ (__v2di)_mm_shrdv_epi64(__A, __B, __C),
+ (__v2di)__A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_shrdv_epi64(__mmask8 __U, __m128i __A, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_selectq_128(__U,
+ (__v2di)_mm_shrdv_epi64(__A, __B, __C),
+ (__v2di)_mm_setzero_si128());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_shrdv_epi32(__m256i __A, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_vpshrdvd256((__v8si)__A, (__v8si)__B,
+ (__v8si)__C);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_shrdv_epi32(__m256i __A, __mmask8 __U, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_shrdv_epi32(__A, __B, __C),
+ (__v8si)__A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_shrdv_epi32(__mmask8 __U, __m256i __A, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_shrdv_epi32(__A, __B, __C),
+ (__v8si)_mm256_setzero_si256());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_shrdv_epi32(__m128i __A, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_vpshrdvd128((__v4si)__A, (__v4si)__B,
+ (__v4si)__C);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_shrdv_epi32(__m128i __A, __mmask8 __U, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_shrdv_epi32(__A, __B, __C),
+ (__v4si)__A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_shrdv_epi32(__mmask8 __U, __m128i __A, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_shrdv_epi32(__A, __B, __C),
+ (__v4si)_mm_setzero_si128());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_shrdv_epi16(__m256i __A, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_vpshrdvw256((__v16hi)__A, (__v16hi)__B,
+ (__v16hi)__C);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_shrdv_epi16(__m256i __A, __mmask16 __U, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_selectw_256(__U,
+ (__v16hi)_mm256_shrdv_epi16(__A, __B, __C),
+ (__v16hi)__A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_shrdv_epi16(__mmask16 __U, __m256i __A, __m256i __B, __m256i __C)
+{
+ return (__m256i)__builtin_ia32_selectw_256(__U,
+ (__v16hi)_mm256_shrdv_epi16(__A, __B, __C),
+ (__v16hi)_mm256_setzero_si256());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_shrdv_epi16(__m128i __A, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_vpshrdvw128((__v8hi)__A, (__v8hi)__B,
+ (__v8hi)__C);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_shrdv_epi16(__m128i __A, __mmask8 __U, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_selectw_128(__U,
+ (__v8hi)_mm_shrdv_epi16(__A, __B, __C),
+ (__v8hi)__A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_shrdv_epi16(__mmask8 __U, __m128i __A, __m128i __B, __m128i __C)
+{
+ return (__m128i)__builtin_ia32_selectw_128(__U,
+ (__v8hi)_mm_shrdv_epi16(__A, __B, __C),
+ (__v8hi)_mm_setzero_si128());
+}
+
+
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
+
+#endif
--- /dev/null
+/*===------------- avx512vlvnniintrin.h - VNNI intrinsics ------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512vlvnniintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __AVX512VLVNNIINTRIN_H
+#define __AVX512VLVNNIINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512vnni"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512vnni"), __min_vector_width__(256)))
+
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_dpbusd_epi32(__m256i __S, __m256i __A, __m256i __B)
+{
+ return (__m256i)__builtin_ia32_vpdpbusd256((__v8si)__S, (__v8si)__A,
+ (__v8si)__B);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_dpbusd_epi32(__m256i __S, __mmask8 __U, __m256i __A, __m256i __B)
+{
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_dpbusd_epi32(__S, __A, __B),
+ (__v8si)__S);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_dpbusd_epi32(__mmask8 __U, __m256i __S, __m256i __A, __m256i __B)
+{
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_dpbusd_epi32(__S, __A, __B),
+ (__v8si)_mm256_setzero_si256());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_dpbusds_epi32(__m256i __S, __m256i __A, __m256i __B)
+{
+ return (__m256i)__builtin_ia32_vpdpbusds256((__v8si)__S, (__v8si)__A,
+ (__v8si)__B);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_dpbusds_epi32(__m256i __S, __mmask8 __U, __m256i __A, __m256i __B)
+{
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_dpbusds_epi32(__S, __A, __B),
+ (__v8si)__S);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_dpbusds_epi32(__mmask8 __U, __m256i __S, __m256i __A, __m256i __B)
+{
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_dpbusds_epi32(__S, __A, __B),
+ (__v8si)_mm256_setzero_si256());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_dpwssd_epi32(__m256i __S, __m256i __A, __m256i __B)
+{
+ return (__m256i)__builtin_ia32_vpdpwssd256((__v8si)__S, (__v8si)__A,
+ (__v8si)__B);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_dpwssd_epi32(__m256i __S, __mmask8 __U, __m256i __A, __m256i __B)
+{
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_dpwssd_epi32(__S, __A, __B),
+ (__v8si)__S);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_dpwssd_epi32(__mmask8 __U, __m256i __S, __m256i __A, __m256i __B)
+{
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_dpwssd_epi32(__S, __A, __B),
+ (__v8si)_mm256_setzero_si256());
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_dpwssds_epi32(__m256i __S, __m256i __A, __m256i __B)
+{
+ return (__m256i)__builtin_ia32_vpdpwssds256((__v8si)__S, (__v8si)__A,
+ (__v8si)__B);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_dpwssds_epi32(__m256i __S, __mmask8 __U, __m256i __A, __m256i __B)
+{
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_dpwssds_epi32(__S, __A, __B),
+ (__v8si)__S);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_dpwssds_epi32(__mmask8 __U, __m256i __S, __m256i __A, __m256i __B)
+{
+ return (__m256i)__builtin_ia32_selectd_256(__U,
+ (__v8si)_mm256_dpwssds_epi32(__S, __A, __B),
+ (__v8si)_mm256_setzero_si256());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_dpbusd_epi32(__m128i __S, __m128i __A, __m128i __B)
+{
+ return (__m128i)__builtin_ia32_vpdpbusd128((__v4si)__S, (__v4si)__A,
+ (__v4si)__B);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_dpbusd_epi32(__m128i __S, __mmask8 __U, __m128i __A, __m128i __B)
+{
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_dpbusd_epi32(__S, __A, __B),
+ (__v4si)__S);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_dpbusd_epi32(__mmask8 __U, __m128i __S, __m128i __A, __m128i __B)
+{
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_dpbusd_epi32(__S, __A, __B),
+ (__v4si)_mm_setzero_si128());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_dpbusds_epi32(__m128i __S, __m128i __A, __m128i __B)
+{
+ return (__m128i)__builtin_ia32_vpdpbusds128((__v4si)__S, (__v4si)__A,
+ (__v4si)__B);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_dpbusds_epi32(__m128i __S, __mmask8 __U, __m128i __A, __m128i __B)
+{
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_dpbusds_epi32(__S, __A, __B),
+ (__v4si)__S);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_dpbusds_epi32(__mmask8 __U, __m128i __S, __m128i __A, __m128i __B)
+{
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_dpbusds_epi32(__S, __A, __B),
+ (__v4si)_mm_setzero_si128());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_dpwssd_epi32(__m128i __S, __m128i __A, __m128i __B)
+{
+ return (__m128i)__builtin_ia32_vpdpwssd128((__v4si)__S, (__v4si)__A,
+ (__v4si)__B);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_dpwssd_epi32(__m128i __S, __mmask8 __U, __m128i __A, __m128i __B)
+{
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_dpwssd_epi32(__S, __A, __B),
+ (__v4si)__S);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_dpwssd_epi32(__mmask8 __U, __m128i __S, __m128i __A, __m128i __B)
+{
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_dpwssd_epi32(__S, __A, __B),
+ (__v4si)_mm_setzero_si128());
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_dpwssds_epi32(__m128i __S, __m128i __A, __m128i __B)
+{
+ return (__m128i)__builtin_ia32_vpdpwssds128((__v4si)__S, (__v4si)__A,
+ (__v4si)__B);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_dpwssds_epi32(__m128i __S, __mmask8 __U, __m128i __A, __m128i __B)
+{
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_dpwssds_epi32(__S, __A, __B),
+ (__v4si)__S);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_dpwssds_epi32(__mmask8 __U, __m128i __S, __m128i __A, __m128i __B)
+{
+ return (__m128i)__builtin_ia32_selectd_128(__U,
+ (__v4si)_mm_dpwssds_epi32(__S, __A, __B),
+ (__v4si)_mm_setzero_si128());
+}
+
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
+
+#endif
--- /dev/null
+/*===------ avx512vlvp2intersectintrin.h - VL VP2INTERSECT intrinsics ------===
+ *
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512vlvp2intersectintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef _AVX512VLVP2INTERSECT_H
+#define _AVX512VLVP2INTERSECT_H
+
+#define __DEFAULT_FN_ATTRS128 \
+ __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512vp2intersect"), \
+ __min_vector_width__(128)))
+
+#define __DEFAULT_FN_ATTRS256 \
+ __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512vp2intersect"), \
+ __min_vector_width__(256)))
+/// Store, in an even/odd pair of mask registers, the indicators of the
+/// locations of value matches between dwords in operands __a and __b.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VP2INTERSECTD </c> instruction.
+///
+/// \param __a
+/// A 256-bit vector of [8 x i32].
+/// \param __b
+/// A 256-bit vector of [8 x i32]
+/// \param __m0
+/// A pointer point to 8-bit mask
+/// \param __m1
+/// A pointer point to 8-bit mask
+static __inline__ void __DEFAULT_FN_ATTRS256
+_mm256_2intersect_epi32(__m256i __a, __m256i __b, __mmask8 *__m0, __mmask8 *__m1) {
+ __builtin_ia32_vp2intersect_d_256((__v8si)__a, (__v8si)__b, __m0, __m1);
+}
+
+/// Store, in an even/odd pair of mask registers, the indicators of the
+/// locations of value matches between quadwords in operands __a and __b.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VP2INTERSECTQ </c> instruction.
+///
+/// \param __a
+/// A 256-bit vector of [4 x i64].
+/// \param __b
+/// A 256-bit vector of [4 x i64]
+/// \param __m0
+/// A pointer point to 8-bit mask
+/// \param __m1
+/// A pointer point to 8-bit mask
+static __inline__ void __DEFAULT_FN_ATTRS256
+_mm256_2intersect_epi64(__m256i __a, __m256i __b, __mmask8 *__m0, __mmask8 *__m1) {
+ __builtin_ia32_vp2intersect_q_256((__v4di)__a, (__v4di)__b, __m0, __m1);
+}
+
+/// Store, in an even/odd pair of mask registers, the indicators of the
+/// locations of value matches between dwords in operands __a and __b.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VP2INTERSECTD </c> instruction.
+///
+/// \param __a
+/// A 128-bit vector of [4 x i32].
+/// \param __b
+/// A 128-bit vector of [4 x i32]
+/// \param __m0
+/// A pointer point to 8-bit mask
+/// \param __m1
+/// A pointer point to 8-bit mask
+static __inline__ void __DEFAULT_FN_ATTRS128
+_mm_2intersect_epi32(__m128i __a, __m128i __b, __mmask8 *__m0, __mmask8 *__m1) {
+ __builtin_ia32_vp2intersect_d_128((__v4si)__a, (__v4si)__b, __m0, __m1);
+}
+
+/// Store, in an even/odd pair of mask registers, the indicators of the
+/// locations of value matches between quadwords in operands __a and __b.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VP2INTERSECTQ </c> instruction.
+///
+/// \param __a
+/// A 128-bit vector of [2 x i64].
+/// \param __b
+/// A 128-bit vector of [2 x i64]
+/// \param __m0
+/// A pointer point to 8-bit mask
+/// \param __m1
+/// A pointer point to 8-bit mask
+static __inline__ void __DEFAULT_FN_ATTRS128
+_mm_2intersect_epi64(__m128i __a, __m128i __b, __mmask8 *__m0, __mmask8 *__m1) {
+ __builtin_ia32_vp2intersect_q_128((__v2di)__a, (__v2di)__b, __m0, __m1);
+}
+
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
+
+#endif
--- /dev/null
+/*===------------- avx512vnniintrin.h - VNNI intrinsics ------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512vnniintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __AVX512VNNIINTRIN_H
+#define __AVX512VNNIINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512vnni"), __min_vector_width__(512)))
+
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_dpbusd_epi32(__m512i __S, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_vpdpbusd512((__v16si)__S, (__v16si)__A,
+ (__v16si)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_dpbusd_epi32(__m512i __S, __mmask16 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_dpbusd_epi32(__S, __A, __B),
+ (__v16si)__S);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_dpbusd_epi32(__mmask16 __U, __m512i __S, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_dpbusd_epi32(__S, __A, __B),
+ (__v16si)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_dpbusds_epi32(__m512i __S, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_vpdpbusds512((__v16si)__S, (__v16si)__A,
+ (__v16si)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_dpbusds_epi32(__m512i __S, __mmask16 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_dpbusds_epi32(__S, __A, __B),
+ (__v16si)__S);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_dpbusds_epi32(__mmask16 __U, __m512i __S, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_dpbusds_epi32(__S, __A, __B),
+ (__v16si)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_dpwssd_epi32(__m512i __S, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_vpdpwssd512((__v16si)__S, (__v16si)__A,
+ (__v16si)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_dpwssd_epi32(__m512i __S, __mmask16 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_dpwssd_epi32(__S, __A, __B),
+ (__v16si)__S);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_dpwssd_epi32(__mmask16 __U, __m512i __S, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_dpwssd_epi32(__S, __A, __B),
+ (__v16si)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_dpwssds_epi32(__m512i __S, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_vpdpwssds512((__v16si)__S, (__v16si)__A,
+ (__v16si)__B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_mask_dpwssds_epi32(__m512i __S, __mmask16 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_dpwssds_epi32(__S, __A, __B),
+ (__v16si)__S);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_maskz_dpwssds_epi32(__mmask16 __U, __m512i __S, __m512i __A, __m512i __B)
+{
+ return (__m512i)__builtin_ia32_selectd_512(__U,
+ (__v16si)_mm512_dpwssds_epi32(__S, __A, __B),
+ (__v16si)_mm512_setzero_si512());
+}
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif
--- /dev/null
+/*===------- avx512vpintersectintrin.h - VP2INTERSECT intrinsics ------------===
+ *
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <avx512vp2intersect.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef _AVX512VP2INTERSECT_H
+#define _AVX512VP2INTERSECT_H
+
+#define __DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("avx512vp2intersect"), \
+ __min_vector_width__(512)))
+
+/// Store, in an even/odd pair of mask registers, the indicators of the
+/// locations of value matches between dwords in operands __a and __b.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VP2INTERSECTD </c> instruction.
+///
+/// \param __a
+/// A 512-bit vector of [16 x i32].
+/// \param __b
+/// A 512-bit vector of [16 x i32]
+/// \param __m0
+/// A pointer point to 16-bit mask
+/// \param __m1
+/// A pointer point to 16-bit mask
+static __inline__ void __DEFAULT_FN_ATTRS
+_mm512_2intersect_epi32(__m512i __a, __m512i __b, __mmask16 *__m0, __mmask16 *__m1) {
+ __builtin_ia32_vp2intersect_d_512((__v16si)__a, (__v16si)__b, __m0, __m1);
+}
+
+/// Store, in an even/odd pair of mask registers, the indicators of the
+/// locations of value matches between quadwords in operands __a and __b.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VP2INTERSECTQ </c> instruction.
+///
+/// \param __a
+/// A 512-bit vector of [8 x i64].
+/// \param __b
+/// A 512-bit vector of [8 x i64]
+/// \param __m0
+/// A pointer point to 8-bit mask
+/// \param __m1
+/// A pointer point to 8-bit mask
+static __inline__ void __DEFAULT_FN_ATTRS
+_mm512_2intersect_epi64(__m512i __a, __m512i __b, __mmask8 *__m0, __mmask8 *__m1) {
+ __builtin_ia32_vp2intersect_q_512((__v8di)__a, (__v8di)__b, __m0, __m1);
+}
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif
-/*===------------- avx512vpopcntdqintrin.h - AVX512VPOPCNTDQ intrinsics
- *------------------===
+/*===----- avx512vpopcntdqintrin.h - AVX512VPOPCNTDQ intrinsics-------------===
*
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
/* Define the default attributes for the functions in this file. */
#define __DEFAULT_FN_ATTRS \
- __attribute__((__always_inline__, __nodebug__, __target__("avx512vpopcntd" \
- "q")))
+ __attribute__((__always_inline__, __nodebug__, __target__("avx512vpopcntdq"), __min_vector_width__(512)))
static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_popcnt_epi64(__m512i __A) {
return (__m512i)__builtin_ia32_vpopcntq_512((__v8di)__A);
--- /dev/null
+/*===---- avx512vpopcntdqintrin.h - AVX512VPOPCNTDQ intrinsics -------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error \
+ "Never use <avx512vpopcntdqvlintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __AVX512VPOPCNTDQVLINTRIN_H
+#define __AVX512VPOPCNTDQVLINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS128 \
+ __attribute__((__always_inline__, __nodebug__, __target__("avx512vpopcntdq,avx512vl"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 \
+ __attribute__((__always_inline__, __nodebug__, __target__("avx512vpopcntdq,avx512vl"), __min_vector_width__(256)))
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_popcnt_epi64(__m128i __A) {
+ return (__m128i)__builtin_ia32_vpopcntq_128((__v2di)__A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_popcnt_epi64(__m128i __W, __mmask8 __U, __m128i __A) {
+ return (__m128i)__builtin_ia32_selectq_128(
+ (__mmask8)__U, (__v2di)_mm_popcnt_epi64(__A), (__v2di)__W);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_popcnt_epi64(__mmask8 __U, __m128i __A) {
+ return _mm_mask_popcnt_epi64((__m128i)_mm_setzero_si128(), __U, __A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_popcnt_epi32(__m128i __A) {
+ return (__m128i)__builtin_ia32_vpopcntd_128((__v4si)__A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_mask_popcnt_epi32(__m128i __W, __mmask8 __U, __m128i __A) {
+ return (__m128i)__builtin_ia32_selectd_128(
+ (__mmask8)__U, (__v4si)_mm_popcnt_epi32(__A), (__v4si)__W);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_maskz_popcnt_epi32(__mmask8 __U, __m128i __A) {
+ return _mm_mask_popcnt_epi32((__m128i)_mm_setzero_si128(), __U, __A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_popcnt_epi64(__m256i __A) {
+ return (__m256i)__builtin_ia32_vpopcntq_256((__v4di)__A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_popcnt_epi64(__m256i __W, __mmask8 __U, __m256i __A) {
+ return (__m256i)__builtin_ia32_selectq_256(
+ (__mmask8)__U, (__v4di)_mm256_popcnt_epi64(__A), (__v4di)__W);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_popcnt_epi64(__mmask8 __U, __m256i __A) {
+ return _mm256_mask_popcnt_epi64((__m256i)_mm256_setzero_si256(), __U, __A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_popcnt_epi32(__m256i __A) {
+ return (__m256i)__builtin_ia32_vpopcntd_256((__v8si)__A);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_mask_popcnt_epi32(__m256i __W, __mmask8 __U, __m256i __A) {
+ return (__m256i)__builtin_ia32_selectd_256(
+ (__mmask8)__U, (__v8si)_mm256_popcnt_epi32(__A), (__v8si)__W);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_maskz_popcnt_epi32(__mmask8 __U, __m256i __A) {
+ return _mm256_mask_popcnt_epi32((__m256i)_mm256_setzero_si256(), __U, __A);
+}
+
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
+
+#endif
/*===---- avxintrin.h - AVX intrinsics -------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
* appear in the interface though. */
typedef signed char __v32qs __attribute__((__vector_size__(32)));
-typedef float __m256 __attribute__ ((__vector_size__ (32)));
-typedef double __m256d __attribute__((__vector_size__(32)));
-typedef long long __m256i __attribute__((__vector_size__(32)));
+typedef float __m256 __attribute__ ((__vector_size__ (32), __aligned__(32)));
+typedef double __m256d __attribute__((__vector_size__(32), __aligned__(32)));
+typedef long long __m256i __attribute__((__vector_size__(32), __aligned__(32)));
+
+typedef float __m256_u __attribute__ ((__vector_size__ (32), __aligned__(1)));
+typedef double __m256d_u __attribute__((__vector_size__(32), __aligned__(1)));
+typedef long long __m256i_u __attribute__((__vector_size__(32), __aligned__(1)));
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx")))
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx"), __min_vector_width__(256)))
+#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx"), __min_vector_width__(128)))
/* Arithmetic */
-/// \brief Adds two 256-bit vectors of [4 x double].
+/// Adds two 256-bit vectors of [4 x double].
///
/// \headerfile <x86intrin.h>
///
return (__m256d)((__v4df)__a+(__v4df)__b);
}
-/// \brief Adds two 256-bit vectors of [8 x float].
+/// Adds two 256-bit vectors of [8 x float].
///
/// \headerfile <x86intrin.h>
///
return (__m256)((__v8sf)__a+(__v8sf)__b);
}
-/// \brief Subtracts two 256-bit vectors of [4 x double].
+/// Subtracts two 256-bit vectors of [4 x double].
///
/// \headerfile <x86intrin.h>
///
return (__m256d)((__v4df)__a-(__v4df)__b);
}
-/// \brief Subtracts two 256-bit vectors of [8 x float].
+/// Subtracts two 256-bit vectors of [8 x float].
///
/// \headerfile <x86intrin.h>
///
return (__m256)((__v8sf)__a-(__v8sf)__b);
}
-/// \brief Adds the even-indexed values and subtracts the odd-indexed values of
+/// Adds the even-indexed values and subtracts the odd-indexed values of
/// two 256-bit vectors of [4 x double].
///
/// \headerfile <x86intrin.h>
return (__m256d)__builtin_ia32_addsubpd256((__v4df)__a, (__v4df)__b);
}
-/// \brief Adds the even-indexed values and subtracts the odd-indexed values of
+/// Adds the even-indexed values and subtracts the odd-indexed values of
/// two 256-bit vectors of [8 x float].
///
/// \headerfile <x86intrin.h>
return (__m256)__builtin_ia32_addsubps256((__v8sf)__a, (__v8sf)__b);
}
-/// \brief Divides two 256-bit vectors of [4 x double].
+/// Divides two 256-bit vectors of [4 x double].
///
/// \headerfile <x86intrin.h>
///
return (__m256d)((__v4df)__a/(__v4df)__b);
}
-/// \brief Divides two 256-bit vectors of [8 x float].
+/// Divides two 256-bit vectors of [8 x float].
///
/// \headerfile <x86intrin.h>
///
return (__m256)((__v8sf)__a/(__v8sf)__b);
}
-/// \brief Compares two 256-bit vectors of [4 x double] and returns the greater
+/// Compares two 256-bit vectors of [4 x double] and returns the greater
/// of each pair of values.
///
/// \headerfile <x86intrin.h>
return (__m256d)__builtin_ia32_maxpd256((__v4df)__a, (__v4df)__b);
}
-/// \brief Compares two 256-bit vectors of [8 x float] and returns the greater
+/// Compares two 256-bit vectors of [8 x float] and returns the greater
/// of each pair of values.
///
/// \headerfile <x86intrin.h>
return (__m256)__builtin_ia32_maxps256((__v8sf)__a, (__v8sf)__b);
}
-/// \brief Compares two 256-bit vectors of [4 x double] and returns the lesser
+/// Compares two 256-bit vectors of [4 x double] and returns the lesser
/// of each pair of values.
///
/// \headerfile <x86intrin.h>
return (__m256d)__builtin_ia32_minpd256((__v4df)__a, (__v4df)__b);
}
-/// \brief Compares two 256-bit vectors of [8 x float] and returns the lesser
+/// Compares two 256-bit vectors of [8 x float] and returns the lesser
/// of each pair of values.
///
/// \headerfile <x86intrin.h>
return (__m256)__builtin_ia32_minps256((__v8sf)__a, (__v8sf)__b);
}
-/// \brief Multiplies two 256-bit vectors of [4 x double].
+/// Multiplies two 256-bit vectors of [4 x double].
///
/// \headerfile <x86intrin.h>
///
return (__m256d)((__v4df)__a * (__v4df)__b);
}
-/// \brief Multiplies two 256-bit vectors of [8 x float].
+/// Multiplies two 256-bit vectors of [8 x float].
///
/// \headerfile <x86intrin.h>
///
return (__m256)((__v8sf)__a * (__v8sf)__b);
}
-/// \brief Calculates the square roots of the values in a 256-bit vector of
+/// Calculates the square roots of the values in a 256-bit vector of
/// [4 x double].
///
/// \headerfile <x86intrin.h>
return (__m256d)__builtin_ia32_sqrtpd256((__v4df)__a);
}
-/// \brief Calculates the square roots of the values in a 256-bit vector of
+/// Calculates the square roots of the values in a 256-bit vector of
/// [8 x float].
///
/// \headerfile <x86intrin.h>
return (__m256)__builtin_ia32_sqrtps256((__v8sf)__a);
}
-/// \brief Calculates the reciprocal square roots of the values in a 256-bit
+/// Calculates the reciprocal square roots of the values in a 256-bit
/// vector of [8 x float].
///
/// \headerfile <x86intrin.h>
return (__m256)__builtin_ia32_rsqrtps256((__v8sf)__a);
}
-/// \brief Calculates the reciprocals of the values in a 256-bit vector of
+/// Calculates the reciprocals of the values in a 256-bit vector of
/// [8 x float].
///
/// \headerfile <x86intrin.h>
return (__m256)__builtin_ia32_rcpps256((__v8sf)__a);
}
-/// \brief Rounds the values in a 256-bit vector of [4 x double] as specified
+/// Rounds the values in a 256-bit vector of [4 x double] as specified
/// by the byte operand. The source values are rounded to integer values and
/// returned as 64-bit double-precision floating-point values.
///
/// 10: Upward (toward positive infinity). \n
/// 11: Truncated.
/// \returns A 256-bit vector of [4 x double] containing the rounded values.
-#define _mm256_round_pd(V, M) __extension__ ({ \
- (__m256d)__builtin_ia32_roundpd256((__v4df)(__m256d)(V), (M)); })
+#define _mm256_round_pd(V, M) \
+ (__m256d)__builtin_ia32_roundpd256((__v4df)(__m256d)(V), (M))
-/// \brief Rounds the values stored in a 256-bit vector of [8 x float] as
+/// Rounds the values stored in a 256-bit vector of [8 x float] as
/// specified by the byte operand. The source values are rounded to integer
/// values and returned as floating-point values.
///
/// 10: Upward (toward positive infinity). \n
/// 11: Truncated.
/// \returns A 256-bit vector of [8 x float] containing the rounded values.
-#define _mm256_round_ps(V, M) __extension__ ({ \
- (__m256)__builtin_ia32_roundps256((__v8sf)(__m256)(V), (M)); })
+#define _mm256_round_ps(V, M) \
+ (__m256)__builtin_ia32_roundps256((__v8sf)(__m256)(V), (M))
-/// \brief Rounds up the values stored in a 256-bit vector of [4 x double]. The
+/// Rounds up the values stored in a 256-bit vector of [4 x double]. The
/// source values are rounded up to integer values and returned as 64-bit
/// double-precision floating-point values.
///
/// \returns A 256-bit vector of [4 x double] containing the rounded up values.
#define _mm256_ceil_pd(V) _mm256_round_pd((V), _MM_FROUND_CEIL)
-/// \brief Rounds down the values stored in a 256-bit vector of [4 x double].
+/// Rounds down the values stored in a 256-bit vector of [4 x double].
/// The source values are rounded down to integer values and returned as
/// 64-bit double-precision floating-point values.
///
/// values.
#define _mm256_floor_pd(V) _mm256_round_pd((V), _MM_FROUND_FLOOR)
-/// \brief Rounds up the values stored in a 256-bit vector of [8 x float]. The
+/// Rounds up the values stored in a 256-bit vector of [8 x float]. The
/// source values are rounded up to integer values and returned as
/// floating-point values.
///
/// \returns A 256-bit vector of [8 x float] containing the rounded up values.
#define _mm256_ceil_ps(V) _mm256_round_ps((V), _MM_FROUND_CEIL)
-/// \brief Rounds down the values stored in a 256-bit vector of [8 x float]. The
+/// Rounds down the values stored in a 256-bit vector of [8 x float]. The
/// source values are rounded down to integer values and returned as
/// floating-point values.
///
#define _mm256_floor_ps(V) _mm256_round_ps((V), _MM_FROUND_FLOOR)
/* Logical */
-/// \brief Performs a bitwise AND of two 256-bit vectors of [4 x double].
+/// Performs a bitwise AND of two 256-bit vectors of [4 x double].
///
/// \headerfile <x86intrin.h>
///
return (__m256d)((__v4du)__a & (__v4du)__b);
}
-/// \brief Performs a bitwise AND of two 256-bit vectors of [8 x float].
+/// Performs a bitwise AND of two 256-bit vectors of [8 x float].
///
/// \headerfile <x86intrin.h>
///
return (__m256)((__v8su)__a & (__v8su)__b);
}
-/// \brief Performs a bitwise AND of two 256-bit vectors of [4 x double], using
+/// Performs a bitwise AND of two 256-bit vectors of [4 x double], using
/// the one's complement of the values contained in the first source operand.
///
/// \headerfile <x86intrin.h>
return (__m256d)(~(__v4du)__a & (__v4du)__b);
}
-/// \brief Performs a bitwise AND of two 256-bit vectors of [8 x float], using
+/// Performs a bitwise AND of two 256-bit vectors of [8 x float], using
/// the one's complement of the values contained in the first source operand.
///
/// \headerfile <x86intrin.h>
return (__m256)(~(__v8su)__a & (__v8su)__b);
}
-/// \brief Performs a bitwise OR of two 256-bit vectors of [4 x double].
+/// Performs a bitwise OR of two 256-bit vectors of [4 x double].
///
/// \headerfile <x86intrin.h>
///
return (__m256d)((__v4du)__a | (__v4du)__b);
}
-/// \brief Performs a bitwise OR of two 256-bit vectors of [8 x float].
+/// Performs a bitwise OR of two 256-bit vectors of [8 x float].
///
/// \headerfile <x86intrin.h>
///
return (__m256)((__v8su)__a | (__v8su)__b);
}
-/// \brief Performs a bitwise XOR of two 256-bit vectors of [4 x double].
+/// Performs a bitwise XOR of two 256-bit vectors of [4 x double].
///
/// \headerfile <x86intrin.h>
///
return (__m256d)((__v4du)__a ^ (__v4du)__b);
}
-/// \brief Performs a bitwise XOR of two 256-bit vectors of [8 x float].
+/// Performs a bitwise XOR of two 256-bit vectors of [8 x float].
///
/// \headerfile <x86intrin.h>
///
}
/* Horizontal arithmetic */
-/// \brief Horizontally adds the adjacent pairs of values contained in two
+/// Horizontally adds the adjacent pairs of values contained in two
/// 256-bit vectors of [4 x double].
///
/// \headerfile <x86intrin.h>
return (__m256d)__builtin_ia32_haddpd256((__v4df)__a, (__v4df)__b);
}
-/// \brief Horizontally adds the adjacent pairs of values contained in two
+/// Horizontally adds the adjacent pairs of values contained in two
/// 256-bit vectors of [8 x float].
///
/// \headerfile <x86intrin.h>
return (__m256)__builtin_ia32_haddps256((__v8sf)__a, (__v8sf)__b);
}
-/// \brief Horizontally subtracts the adjacent pairs of values contained in two
+/// Horizontally subtracts the adjacent pairs of values contained in two
/// 256-bit vectors of [4 x double].
///
/// \headerfile <x86intrin.h>
return (__m256d)__builtin_ia32_hsubpd256((__v4df)__a, (__v4df)__b);
}
-/// \brief Horizontally subtracts the adjacent pairs of values contained in two
+/// Horizontally subtracts the adjacent pairs of values contained in two
/// 256-bit vectors of [8 x float].
///
/// \headerfile <x86intrin.h>
}
/* Vector permutations */
-/// \brief Copies the values in a 128-bit vector of [2 x double] as specified
+/// Copies the values in a 128-bit vector of [2 x double] as specified
/// by the 128-bit integer vector operand.
///
/// \headerfile <x86intrin.h>
/// 1: Bits [127:64] of the source are copied to bits [127:64] of the
/// returned vector.
/// \returns A 128-bit vector of [2 x double] containing the copied values.
-static __inline __m128d __DEFAULT_FN_ATTRS
+static __inline __m128d __DEFAULT_FN_ATTRS128
_mm_permutevar_pd(__m128d __a, __m128i __c)
{
return (__m128d)__builtin_ia32_vpermilvarpd((__v2df)__a, (__v2di)__c);
}
-/// \brief Copies the values in a 256-bit vector of [4 x double] as specified
+/// Copies the values in a 256-bit vector of [4 x double] as specified
/// by the 256-bit integer vector operand.
///
/// \headerfile <x86intrin.h>
return (__m256d)__builtin_ia32_vpermilvarpd256((__v4df)__a, (__v4di)__c);
}
-/// \brief Copies the values stored in a 128-bit vector of [4 x float] as
+/// Copies the values stored in a 128-bit vector of [4 x float] as
/// specified by the 128-bit integer vector operand.
/// \headerfile <x86intrin.h>
///
/// 11: Bits [127:96] of the source are copied to bits [127:96] of the
/// returned vector.
/// \returns A 128-bit vector of [4 x float] containing the copied values.
-static __inline __m128 __DEFAULT_FN_ATTRS
+static __inline __m128 __DEFAULT_FN_ATTRS128
_mm_permutevar_ps(__m128 __a, __m128i __c)
{
return (__m128)__builtin_ia32_vpermilvarps((__v4sf)__a, (__v4si)__c);
}
-/// \brief Copies the values stored in a 256-bit vector of [8 x float] as
+/// Copies the values stored in a 256-bit vector of [8 x float] as
/// specified by the 256-bit integer vector operand.
///
/// \headerfile <x86intrin.h>
return (__m256)__builtin_ia32_vpermilvarps256((__v8sf)__a, (__v8si)__c);
}
-/// \brief Copies the values in a 128-bit vector of [2 x double] as specified
+/// Copies the values in a 128-bit vector of [2 x double] as specified
/// by the immediate integer operand.
///
/// \headerfile <x86intrin.h>
/// 1: Bits [127:64] of the source are copied to bits [127:64] of the
/// returned vector.
/// \returns A 128-bit vector of [2 x double] containing the copied values.
-#define _mm_permute_pd(A, C) __extension__ ({ \
- (__m128d)__builtin_shufflevector((__v2df)(__m128d)(A), \
- (__v2df)_mm_undefined_pd(), \
- ((C) >> 0) & 0x1, ((C) >> 1) & 0x1); })
+#define _mm_permute_pd(A, C) \
+ (__m128d)__builtin_ia32_vpermilpd((__v2df)(__m128d)(A), (int)(C))
-/// \brief Copies the values in a 256-bit vector of [4 x double] as specified by
+/// Copies the values in a 256-bit vector of [4 x double] as specified by
/// the immediate integer operand.
///
/// \headerfile <x86intrin.h>
/// 1: Bits [255:192] of the source are copied to bits [255:192] of the
/// returned vector.
/// \returns A 256-bit vector of [4 x double] containing the copied values.
-#define _mm256_permute_pd(A, C) __extension__ ({ \
- (__m256d)__builtin_shufflevector((__v4df)(__m256d)(A), \
- (__v4df)_mm256_undefined_pd(), \
- 0 + (((C) >> 0) & 0x1), \
- 0 + (((C) >> 1) & 0x1), \
- 2 + (((C) >> 2) & 0x1), \
- 2 + (((C) >> 3) & 0x1)); })
-
-/// \brief Copies the values in a 128-bit vector of [4 x float] as specified by
+#define _mm256_permute_pd(A, C) \
+ (__m256d)__builtin_ia32_vpermilpd256((__v4df)(__m256d)(A), (int)(C))
+
+/// Copies the values in a 128-bit vector of [4 x float] as specified by
/// the immediate integer operand.
///
/// \headerfile <x86intrin.h>
/// 11: Bits [127:96] of the source are copied to bits [127:96] of the
/// returned vector.
/// \returns A 128-bit vector of [4 x float] containing the copied values.
-#define _mm_permute_ps(A, C) __extension__ ({ \
- (__m128)__builtin_shufflevector((__v4sf)(__m128)(A), \
- (__v4sf)_mm_undefined_ps(), \
- ((C) >> 0) & 0x3, ((C) >> 2) & 0x3, \
- ((C) >> 4) & 0x3, ((C) >> 6) & 0x3); })
+#define _mm_permute_ps(A, C) \
+ (__m128)__builtin_ia32_vpermilps((__v4sf)(__m128)(A), (int)(C))
-/// \brief Copies the values in a 256-bit vector of [8 x float] as specified by
+/// Copies the values in a 256-bit vector of [8 x float] as specified by
/// the immediate integer operand.
///
/// \headerfile <x86intrin.h>
/// \param A
/// A 256-bit vector of [8 x float].
/// \param C
-/// An immediate integer operand specifying how the values are to be \n
+/// An immediate integer operand specifying how the values are to be
/// copied. \n
/// Bits [1:0]: \n
/// 00: Bits [31:0] of the source are copied to bits [31:0] of the
/// 11: Bits [127:96] of the source are copied to bits [95:64] of the
/// returned vector. \n
/// Bits [7:6]: \n
-/// 00: Bits [31:qq0] of the source are copied to bits [127:96] of the
+/// 00: Bits [31:0] of the source are copied to bits [127:96] of the
/// returned vector. \n
/// 01: Bits [63:32] of the source are copied to bits [127:96] of the
/// returned vector. \n
/// 11: Bits [255:224] of the source are copied to bits [255:224] of the
/// returned vector.
/// \returns A 256-bit vector of [8 x float] containing the copied values.
-#define _mm256_permute_ps(A, C) __extension__ ({ \
- (__m256)__builtin_shufflevector((__v8sf)(__m256)(A), \
- (__v8sf)_mm256_undefined_ps(), \
- 0 + (((C) >> 0) & 0x3), \
- 0 + (((C) >> 2) & 0x3), \
- 0 + (((C) >> 4) & 0x3), \
- 0 + (((C) >> 6) & 0x3), \
- 4 + (((C) >> 0) & 0x3), \
- 4 + (((C) >> 2) & 0x3), \
- 4 + (((C) >> 4) & 0x3), \
- 4 + (((C) >> 6) & 0x3)); })
-
-/// \brief Permutes 128-bit data values stored in two 256-bit vectors of
+#define _mm256_permute_ps(A, C) \
+ (__m256)__builtin_ia32_vpermilps256((__v8sf)(__m256)(A), (int)(C))
+
+/// Permutes 128-bit data values stored in two 256-bit vectors of
/// [4 x double], as specified by the immediate integer operand.
///
/// \headerfile <x86intrin.h>
/// 11: Bits [255:128] of operand \a V2 are copied to bits [255:128] of the
/// destination.
/// \returns A 256-bit vector of [4 x double] containing the copied values.
-#define _mm256_permute2f128_pd(V1, V2, M) __extension__ ({ \
+#define _mm256_permute2f128_pd(V1, V2, M) \
(__m256d)__builtin_ia32_vperm2f128_pd256((__v4df)(__m256d)(V1), \
- (__v4df)(__m256d)(V2), (M)); })
+ (__v4df)(__m256d)(V2), (int)(M))
-/// \brief Permutes 128-bit data values stored in two 256-bit vectors of
+/// Permutes 128-bit data values stored in two 256-bit vectors of
/// [8 x float], as specified by the immediate integer operand.
///
/// \headerfile <x86intrin.h>
/// 11: Bits [255:128] of operand \a V2 are copied to bits [255:128] of the
/// destination.
/// \returns A 256-bit vector of [8 x float] containing the copied values.
-#define _mm256_permute2f128_ps(V1, V2, M) __extension__ ({ \
+#define _mm256_permute2f128_ps(V1, V2, M) \
(__m256)__builtin_ia32_vperm2f128_ps256((__v8sf)(__m256)(V1), \
- (__v8sf)(__m256)(V2), (M)); })
+ (__v8sf)(__m256)(V2), (int)(M))
-/// \brief Permutes 128-bit data values stored in two 256-bit integer vectors,
+/// Permutes 128-bit data values stored in two 256-bit integer vectors,
/// as specified by the immediate integer operand.
///
/// \headerfile <x86intrin.h>
/// 11: Bits [255:128] of operand \a V2 are copied to bits [255:128] of the
/// destination.
/// \returns A 256-bit integer vector containing the copied values.
-#define _mm256_permute2f128_si256(V1, V2, M) __extension__ ({ \
+#define _mm256_permute2f128_si256(V1, V2, M) \
(__m256i)__builtin_ia32_vperm2f128_si256((__v8si)(__m256i)(V1), \
- (__v8si)(__m256i)(V2), (M)); })
+ (__v8si)(__m256i)(V2), (int)(M))
/* Vector Blend */
-/// \brief Merges 64-bit double-precision data values stored in either of the
+/// Merges 64-bit double-precision data values stored in either of the
/// two 256-bit vectors of [4 x double], as specified by the immediate
/// integer operand.
///
/// destination. When a mask bit is 1, the corresponding 64-bit element in
/// operand \a V2 is copied to the same position in the destination.
/// \returns A 256-bit vector of [4 x double] containing the copied values.
-#define _mm256_blend_pd(V1, V2, M) __extension__ ({ \
- (__m256d)__builtin_shufflevector((__v4df)(__m256d)(V1), \
- (__v4df)(__m256d)(V2), \
- (((M) & 0x01) ? 4 : 0), \
- (((M) & 0x02) ? 5 : 1), \
- (((M) & 0x04) ? 6 : 2), \
- (((M) & 0x08) ? 7 : 3)); })
-
-/// \brief Merges 32-bit single-precision data values stored in either of the
+#define _mm256_blend_pd(V1, V2, M) \
+ (__m256d)__builtin_ia32_blendpd256((__v4df)(__m256d)(V1), \
+ (__v4df)(__m256d)(V2), (int)(M))
+
+/// Merges 32-bit single-precision data values stored in either of the
/// two 256-bit vectors of [8 x float], as specified by the immediate
/// integer operand.
///
/// destination. When a mask bit is 1, the corresponding 32-bit element in
/// operand \a V2 is copied to the same position in the destination.
/// \returns A 256-bit vector of [8 x float] containing the copied values.
-#define _mm256_blend_ps(V1, V2, M) __extension__ ({ \
- (__m256)__builtin_shufflevector((__v8sf)(__m256)(V1), \
- (__v8sf)(__m256)(V2), \
- (((M) & 0x01) ? 8 : 0), \
- (((M) & 0x02) ? 9 : 1), \
- (((M) & 0x04) ? 10 : 2), \
- (((M) & 0x08) ? 11 : 3), \
- (((M) & 0x10) ? 12 : 4), \
- (((M) & 0x20) ? 13 : 5), \
- (((M) & 0x40) ? 14 : 6), \
- (((M) & 0x80) ? 15 : 7)); })
-
-/// \brief Merges 64-bit double-precision data values stored in either of the
+#define _mm256_blend_ps(V1, V2, M) \
+ (__m256)__builtin_ia32_blendps256((__v8sf)(__m256)(V1), \
+ (__v8sf)(__m256)(V2), (int)(M))
+
+/// Merges 64-bit double-precision data values stored in either of the
/// two 256-bit vectors of [4 x double], as specified by the 256-bit vector
/// operand.
///
(__v4df)__a, (__v4df)__b, (__v4df)__c);
}
-/// \brief Merges 32-bit single-precision data values stored in either of the
+/// Merges 32-bit single-precision data values stored in either of the
/// two 256-bit vectors of [8 x float], as specified by the 256-bit vector
/// operand.
///
}
/* Vector Dot Product */
-/// \brief Computes two dot products in parallel, using the lower and upper
+/// Computes two dot products in parallel, using the lower and upper
/// halves of two [8 x float] vectors as input to the two computations, and
/// returning the two dot products in the lower and upper halves of the
/// [8 x float] result.
/// is set to zero. The bitmask is applied in the same way to each of the
/// two parallel dot product computations.
/// \returns A 256-bit vector of [8 x float] containing the two dot products.
-#define _mm256_dp_ps(V1, V2, M) __extension__ ({ \
+#define _mm256_dp_ps(V1, V2, M) \
(__m256)__builtin_ia32_dpps256((__v8sf)(__m256)(V1), \
- (__v8sf)(__m256)(V2), (M)); })
+ (__v8sf)(__m256)(V2), (M))
/* Vector shuffle */
-/// \brief Selects 8 float values from the 256-bit operands of [8 x float], as
+/// Selects 8 float values from the 256-bit operands of [8 x float], as
/// specified by the immediate value operand.
///
/// The four selected elements in each operand are copied to the destination
/// 10: Bits [95:64] and [223:192] are copied from the selected operand. \n
/// 11: Bits [127:96] and [255:224] are copied from the selected operand.
/// \returns A 256-bit vector of [8 x float] containing the shuffled values.
-#define _mm256_shuffle_ps(a, b, mask) __extension__ ({ \
- (__m256)__builtin_shufflevector((__v8sf)(__m256)(a), \
- (__v8sf)(__m256)(b), \
- 0 + (((mask) >> 0) & 0x3), \
- 0 + (((mask) >> 2) & 0x3), \
- 8 + (((mask) >> 4) & 0x3), \
- 8 + (((mask) >> 6) & 0x3), \
- 4 + (((mask) >> 0) & 0x3), \
- 4 + (((mask) >> 2) & 0x3), \
- 12 + (((mask) >> 4) & 0x3), \
- 12 + (((mask) >> 6) & 0x3)); })
-
-/// \brief Selects four double-precision values from the 256-bit operands of
+#define _mm256_shuffle_ps(a, b, mask) \
+ (__m256)__builtin_ia32_shufps256((__v8sf)(__m256)(a), \
+ (__v8sf)(__m256)(b), (int)(mask))
+
+/// Selects four double-precision values from the 256-bit operands of
/// [4 x double], as specified by the immediate value operand.
///
/// The selected elements from the first 256-bit operand are copied to bits
/// Bit [3]=1: Bits [255:192] are copied from \a b to bits [255:192] of the
/// destination.
/// \returns A 256-bit vector of [4 x double] containing the shuffled values.
-#define _mm256_shuffle_pd(a, b, mask) __extension__ ({ \
- (__m256d)__builtin_shufflevector((__v4df)(__m256d)(a), \
- (__v4df)(__m256d)(b), \
- 0 + (((mask) >> 0) & 0x1), \
- 4 + (((mask) >> 1) & 0x1), \
- 2 + (((mask) >> 2) & 0x1), \
- 6 + (((mask) >> 3) & 0x1)); })
+#define _mm256_shuffle_pd(a, b, mask) \
+ (__m256d)__builtin_ia32_shufpd256((__v4df)(__m256d)(a), \
+ (__v4df)(__m256d)(b), (int)(mask))
/* Compare */
#define _CMP_EQ_OQ 0x00 /* Equal (ordered, non-signaling) */
#define _CMP_GT_OQ 0x1e /* Greater-than (ordered, non-signaling) */
#define _CMP_TRUE_US 0x1f /* True (unordered, signaling) */
-/// \brief Compares each of the corresponding double-precision values of two
+/// Compares each of the corresponding double-precision values of two
/// 128-bit vectors of [2 x double], using the operation specified by the
/// immediate integer operand.
///
/// \param c
/// An immediate integer operand, with bits [4:0] specifying which comparison
/// operation to use: \n
-/// 0x00 : Equal (ordered, non-signaling)
-/// 0x01 : Less-than (ordered, signaling)
-/// 0x02 : Less-than-or-equal (ordered, signaling)
-/// 0x03 : Unordered (non-signaling)
-/// 0x04 : Not-equal (unordered, non-signaling)
-/// 0x05 : Not-less-than (unordered, signaling)
-/// 0x06 : Not-less-than-or-equal (unordered, signaling)
-/// 0x07 : Ordered (non-signaling)
-/// 0x08 : Equal (unordered, non-signaling)
-/// 0x09 : Not-greater-than-or-equal (unordered, signaling)
-/// 0x0a : Not-greater-than (unordered, signaling)
-/// 0x0b : False (ordered, non-signaling)
-/// 0x0c : Not-equal (ordered, non-signaling)
-/// 0x0d : Greater-than-or-equal (ordered, signaling)
-/// 0x0e : Greater-than (ordered, signaling)
-/// 0x0f : True (unordered, non-signaling)
-/// 0x10 : Equal (ordered, signaling)
-/// 0x11 : Less-than (ordered, non-signaling)
-/// 0x12 : Less-than-or-equal (ordered, non-signaling)
-/// 0x13 : Unordered (signaling)
-/// 0x14 : Not-equal (unordered, signaling)
-/// 0x15 : Not-less-than (unordered, non-signaling)
-/// 0x16 : Not-less-than-or-equal (unordered, non-signaling)
-/// 0x17 : Ordered (signaling)
-/// 0x18 : Equal (unordered, signaling)
-/// 0x19 : Not-greater-than-or-equal (unordered, non-signaling)
-/// 0x1a : Not-greater-than (unordered, non-signaling)
-/// 0x1b : False (ordered, signaling)
-/// 0x1c : Not-equal (ordered, signaling)
-/// 0x1d : Greater-than-or-equal (ordered, non-signaling)
-/// 0x1e : Greater-than (ordered, non-signaling)
-/// 0x1f : True (unordered, signaling)
+/// 0x00: Equal (ordered, non-signaling) \n
+/// 0x01: Less-than (ordered, signaling) \n
+/// 0x02: Less-than-or-equal (ordered, signaling) \n
+/// 0x03: Unordered (non-signaling) \n
+/// 0x04: Not-equal (unordered, non-signaling) \n
+/// 0x05: Not-less-than (unordered, signaling) \n
+/// 0x06: Not-less-than-or-equal (unordered, signaling) \n
+/// 0x07: Ordered (non-signaling) \n
+/// 0x08: Equal (unordered, non-signaling) \n
+/// 0x09: Not-greater-than-or-equal (unordered, signaling) \n
+/// 0x0A: Not-greater-than (unordered, signaling) \n
+/// 0x0B: False (ordered, non-signaling) \n
+/// 0x0C: Not-equal (ordered, non-signaling) \n
+/// 0x0D: Greater-than-or-equal (ordered, signaling) \n
+/// 0x0E: Greater-than (ordered, signaling) \n
+/// 0x0F: True (unordered, non-signaling) \n
+/// 0x10: Equal (ordered, signaling) \n
+/// 0x11: Less-than (ordered, non-signaling) \n
+/// 0x12: Less-than-or-equal (ordered, non-signaling) \n
+/// 0x13: Unordered (signaling) \n
+/// 0x14: Not-equal (unordered, signaling) \n
+/// 0x15: Not-less-than (unordered, non-signaling) \n
+/// 0x16: Not-less-than-or-equal (unordered, non-signaling) \n
+/// 0x17: Ordered (signaling) \n
+/// 0x18: Equal (unordered, signaling) \n
+/// 0x19: Not-greater-than-or-equal (unordered, non-signaling) \n
+/// 0x1A: Not-greater-than (unordered, non-signaling) \n
+/// 0x1B: False (ordered, signaling) \n
+/// 0x1C: Not-equal (ordered, signaling) \n
+/// 0x1D: Greater-than-or-equal (ordered, non-signaling) \n
+/// 0x1E: Greater-than (ordered, non-signaling) \n
+/// 0x1F: True (unordered, signaling)
/// \returns A 128-bit vector of [2 x double] containing the comparison results.
-#define _mm_cmp_pd(a, b, c) __extension__ ({ \
+#define _mm_cmp_pd(a, b, c) \
(__m128d)__builtin_ia32_cmppd((__v2df)(__m128d)(a), \
- (__v2df)(__m128d)(b), (c)); })
+ (__v2df)(__m128d)(b), (c))
-/// \brief Compares each of the corresponding values of two 128-bit vectors of
+/// Compares each of the corresponding values of two 128-bit vectors of
/// [4 x float], using the operation specified by the immediate integer
/// operand.
///
/// \param c
/// An immediate integer operand, with bits [4:0] specifying which comparison
/// operation to use: \n
-/// 0x00 : Equal (ordered, non-signaling)
-/// 0x01 : Less-than (ordered, signaling)
-/// 0x02 : Less-than-or-equal (ordered, signaling)
-/// 0x03 : Unordered (non-signaling)
-/// 0x04 : Not-equal (unordered, non-signaling)
-/// 0x05 : Not-less-than (unordered, signaling)
-/// 0x06 : Not-less-than-or-equal (unordered, signaling)
-/// 0x07 : Ordered (non-signaling)
-/// 0x08 : Equal (unordered, non-signaling)
-/// 0x09 : Not-greater-than-or-equal (unordered, signaling)
-/// 0x0a : Not-greater-than (unordered, signaling)
-/// 0x0b : False (ordered, non-signaling)
-/// 0x0c : Not-equal (ordered, non-signaling)
-/// 0x0d : Greater-than-or-equal (ordered, signaling)
-/// 0x0e : Greater-than (ordered, signaling)
-/// 0x0f : True (unordered, non-signaling)
-/// 0x10 : Equal (ordered, signaling)
-/// 0x11 : Less-than (ordered, non-signaling)
-/// 0x12 : Less-than-or-equal (ordered, non-signaling)
-/// 0x13 : Unordered (signaling)
-/// 0x14 : Not-equal (unordered, signaling)
-/// 0x15 : Not-less-than (unordered, non-signaling)
-/// 0x16 : Not-less-than-or-equal (unordered, non-signaling)
-/// 0x17 : Ordered (signaling)
-/// 0x18 : Equal (unordered, signaling)
-/// 0x19 : Not-greater-than-or-equal (unordered, non-signaling)
-/// 0x1a : Not-greater-than (unordered, non-signaling)
-/// 0x1b : False (ordered, signaling)
-/// 0x1c : Not-equal (ordered, signaling)
-/// 0x1d : Greater-than-or-equal (ordered, non-signaling)
-/// 0x1e : Greater-than (ordered, non-signaling)
-/// 0x1f : True (unordered, signaling)
+/// 0x00: Equal (ordered, non-signaling) \n
+/// 0x01: Less-than (ordered, signaling) \n
+/// 0x02: Less-than-or-equal (ordered, signaling) \n
+/// 0x03: Unordered (non-signaling) \n
+/// 0x04: Not-equal (unordered, non-signaling) \n
+/// 0x05: Not-less-than (unordered, signaling) \n
+/// 0x06: Not-less-than-or-equal (unordered, signaling) \n
+/// 0x07: Ordered (non-signaling) \n
+/// 0x08: Equal (unordered, non-signaling) \n
+/// 0x09: Not-greater-than-or-equal (unordered, signaling) \n
+/// 0x0A: Not-greater-than (unordered, signaling) \n
+/// 0x0B: False (ordered, non-signaling) \n
+/// 0x0C: Not-equal (ordered, non-signaling) \n
+/// 0x0D: Greater-than-or-equal (ordered, signaling) \n
+/// 0x0E: Greater-than (ordered, signaling) \n
+/// 0x0F: True (unordered, non-signaling) \n
+/// 0x10: Equal (ordered, signaling) \n
+/// 0x11: Less-than (ordered, non-signaling) \n
+/// 0x12: Less-than-or-equal (ordered, non-signaling) \n
+/// 0x13: Unordered (signaling) \n
+/// 0x14: Not-equal (unordered, signaling) \n
+/// 0x15: Not-less-than (unordered, non-signaling) \n
+/// 0x16: Not-less-than-or-equal (unordered, non-signaling) \n
+/// 0x17: Ordered (signaling) \n
+/// 0x18: Equal (unordered, signaling) \n
+/// 0x19: Not-greater-than-or-equal (unordered, non-signaling) \n
+/// 0x1A: Not-greater-than (unordered, non-signaling) \n
+/// 0x1B: False (ordered, signaling) \n
+/// 0x1C: Not-equal (ordered, signaling) \n
+/// 0x1D: Greater-than-or-equal (ordered, non-signaling) \n
+/// 0x1E: Greater-than (ordered, non-signaling) \n
+/// 0x1F: True (unordered, signaling)
/// \returns A 128-bit vector of [4 x float] containing the comparison results.
-#define _mm_cmp_ps(a, b, c) __extension__ ({ \
+#define _mm_cmp_ps(a, b, c) \
(__m128)__builtin_ia32_cmpps((__v4sf)(__m128)(a), \
- (__v4sf)(__m128)(b), (c)); })
+ (__v4sf)(__m128)(b), (c))
-/// \brief Compares each of the corresponding double-precision values of two
+/// Compares each of the corresponding double-precision values of two
/// 256-bit vectors of [4 x double], using the operation specified by the
/// immediate integer operand.
///
/// \param c
/// An immediate integer operand, with bits [4:0] specifying which comparison
/// operation to use: \n
-/// 0x00 : Equal (ordered, non-signaling)
-/// 0x01 : Less-than (ordered, signaling)
-/// 0x02 : Less-than-or-equal (ordered, signaling)
-/// 0x03 : Unordered (non-signaling)
-/// 0x04 : Not-equal (unordered, non-signaling)
-/// 0x05 : Not-less-than (unordered, signaling)
-/// 0x06 : Not-less-than-or-equal (unordered, signaling)
-/// 0x07 : Ordered (non-signaling)
-/// 0x08 : Equal (unordered, non-signaling)
-/// 0x09 : Not-greater-than-or-equal (unordered, signaling)
-/// 0x0a : Not-greater-than (unordered, signaling)
-/// 0x0b : False (ordered, non-signaling)
-/// 0x0c : Not-equal (ordered, non-signaling)
-/// 0x0d : Greater-than-or-equal (ordered, signaling)
-/// 0x0e : Greater-than (ordered, signaling)
-/// 0x0f : True (unordered, non-signaling)
-/// 0x10 : Equal (ordered, signaling)
-/// 0x11 : Less-than (ordered, non-signaling)
-/// 0x12 : Less-than-or-equal (ordered, non-signaling)
-/// 0x13 : Unordered (signaling)
-/// 0x14 : Not-equal (unordered, signaling)
-/// 0x15 : Not-less-than (unordered, non-signaling)
-/// 0x16 : Not-less-than-or-equal (unordered, non-signaling)
-/// 0x17 : Ordered (signaling)
-/// 0x18 : Equal (unordered, signaling)
-/// 0x19 : Not-greater-than-or-equal (unordered, non-signaling)
-/// 0x1a : Not-greater-than (unordered, non-signaling)
-/// 0x1b : False (ordered, signaling)
-/// 0x1c : Not-equal (ordered, signaling)
-/// 0x1d : Greater-than-or-equal (ordered, non-signaling)
-/// 0x1e : Greater-than (ordered, non-signaling)
-/// 0x1f : True (unordered, signaling)
+/// 0x00: Equal (ordered, non-signaling) \n
+/// 0x01: Less-than (ordered, signaling) \n
+/// 0x02: Less-than-or-equal (ordered, signaling) \n
+/// 0x03: Unordered (non-signaling) \n
+/// 0x04: Not-equal (unordered, non-signaling) \n
+/// 0x05: Not-less-than (unordered, signaling) \n
+/// 0x06: Not-less-than-or-equal (unordered, signaling) \n
+/// 0x07: Ordered (non-signaling) \n
+/// 0x08: Equal (unordered, non-signaling) \n
+/// 0x09: Not-greater-than-or-equal (unordered, signaling) \n
+/// 0x0A: Not-greater-than (unordered, signaling) \n
+/// 0x0B: False (ordered, non-signaling) \n
+/// 0x0C: Not-equal (ordered, non-signaling) \n
+/// 0x0D: Greater-than-or-equal (ordered, signaling) \n
+/// 0x0E: Greater-than (ordered, signaling) \n
+/// 0x0F: True (unordered, non-signaling) \n
+/// 0x10: Equal (ordered, signaling) \n
+/// 0x11: Less-than (ordered, non-signaling) \n
+/// 0x12: Less-than-or-equal (ordered, non-signaling) \n
+/// 0x13: Unordered (signaling) \n
+/// 0x14: Not-equal (unordered, signaling) \n
+/// 0x15: Not-less-than (unordered, non-signaling) \n
+/// 0x16: Not-less-than-or-equal (unordered, non-signaling) \n
+/// 0x17: Ordered (signaling) \n
+/// 0x18: Equal (unordered, signaling) \n
+/// 0x19: Not-greater-than-or-equal (unordered, non-signaling) \n
+/// 0x1A: Not-greater-than (unordered, non-signaling) \n
+/// 0x1B: False (ordered, signaling) \n
+/// 0x1C: Not-equal (ordered, signaling) \n
+/// 0x1D: Greater-than-or-equal (ordered, non-signaling) \n
+/// 0x1E: Greater-than (ordered, non-signaling) \n
+/// 0x1F: True (unordered, signaling)
/// \returns A 256-bit vector of [4 x double] containing the comparison results.
-#define _mm256_cmp_pd(a, b, c) __extension__ ({ \
+#define _mm256_cmp_pd(a, b, c) \
(__m256d)__builtin_ia32_cmppd256((__v4df)(__m256d)(a), \
- (__v4df)(__m256d)(b), (c)); })
+ (__v4df)(__m256d)(b), (c))
-/// \brief Compares each of the corresponding values of two 256-bit vectors of
+/// Compares each of the corresponding values of two 256-bit vectors of
/// [8 x float], using the operation specified by the immediate integer
/// operand.
///
/// \param c
/// An immediate integer operand, with bits [4:0] specifying which comparison
/// operation to use: \n
-/// 0x00 : Equal (ordered, non-signaling)
-/// 0x01 : Less-than (ordered, signaling)
-/// 0x02 : Less-than-or-equal (ordered, signaling)
-/// 0x03 : Unordered (non-signaling)
-/// 0x04 : Not-equal (unordered, non-signaling)
-/// 0x05 : Not-less-than (unordered, signaling)
-/// 0x06 : Not-less-than-or-equal (unordered, signaling)
-/// 0x07 : Ordered (non-signaling)
-/// 0x08 : Equal (unordered, non-signaling)
-/// 0x09 : Not-greater-than-or-equal (unordered, signaling)
-/// 0x0a : Not-greater-than (unordered, signaling)
-/// 0x0b : False (ordered, non-signaling)
-/// 0x0c : Not-equal (ordered, non-signaling)
-/// 0x0d : Greater-than-or-equal (ordered, signaling)
-/// 0x0e : Greater-than (ordered, signaling)
-/// 0x0f : True (unordered, non-signaling)
-/// 0x10 : Equal (ordered, signaling)
-/// 0x11 : Less-than (ordered, non-signaling)
-/// 0x12 : Less-than-or-equal (ordered, non-signaling)
-/// 0x13 : Unordered (signaling)
-/// 0x14 : Not-equal (unordered, signaling)
-/// 0x15 : Not-less-than (unordered, non-signaling)
-/// 0x16 : Not-less-than-or-equal (unordered, non-signaling)
-/// 0x17 : Ordered (signaling)
-/// 0x18 : Equal (unordered, signaling)
-/// 0x19 : Not-greater-than-or-equal (unordered, non-signaling)
-/// 0x1a : Not-greater-than (unordered, non-signaling)
-/// 0x1b : False (ordered, signaling)
-/// 0x1c : Not-equal (ordered, signaling)
-/// 0x1d : Greater-than-or-equal (ordered, non-signaling)
-/// 0x1e : Greater-than (ordered, non-signaling)
-/// 0x1f : True (unordered, signaling)
+/// 0x00: Equal (ordered, non-signaling) \n
+/// 0x01: Less-than (ordered, signaling) \n
+/// 0x02: Less-than-or-equal (ordered, signaling) \n
+/// 0x03: Unordered (non-signaling) \n
+/// 0x04: Not-equal (unordered, non-signaling) \n
+/// 0x05: Not-less-than (unordered, signaling) \n
+/// 0x06: Not-less-than-or-equal (unordered, signaling) \n
+/// 0x07: Ordered (non-signaling) \n
+/// 0x08: Equal (unordered, non-signaling) \n
+/// 0x09: Not-greater-than-or-equal (unordered, signaling) \n
+/// 0x0A: Not-greater-than (unordered, signaling) \n
+/// 0x0B: False (ordered, non-signaling) \n
+/// 0x0C: Not-equal (ordered, non-signaling) \n
+/// 0x0D: Greater-than-or-equal (ordered, signaling) \n
+/// 0x0E: Greater-than (ordered, signaling) \n
+/// 0x0F: True (unordered, non-signaling) \n
+/// 0x10: Equal (ordered, signaling) \n
+/// 0x11: Less-than (ordered, non-signaling) \n
+/// 0x12: Less-than-or-equal (ordered, non-signaling) \n
+/// 0x13: Unordered (signaling) \n
+/// 0x14: Not-equal (unordered, signaling) \n
+/// 0x15: Not-less-than (unordered, non-signaling) \n
+/// 0x16: Not-less-than-or-equal (unordered, non-signaling) \n
+/// 0x17: Ordered (signaling) \n
+/// 0x18: Equal (unordered, signaling) \n
+/// 0x19: Not-greater-than-or-equal (unordered, non-signaling) \n
+/// 0x1A: Not-greater-than (unordered, non-signaling) \n
+/// 0x1B: False (ordered, signaling) \n
+/// 0x1C: Not-equal (ordered, signaling) \n
+/// 0x1D: Greater-than-or-equal (ordered, non-signaling) \n
+/// 0x1E: Greater-than (ordered, non-signaling) \n
+/// 0x1F: True (unordered, signaling)
/// \returns A 256-bit vector of [8 x float] containing the comparison results.
-#define _mm256_cmp_ps(a, b, c) __extension__ ({ \
+#define _mm256_cmp_ps(a, b, c) \
(__m256)__builtin_ia32_cmpps256((__v8sf)(__m256)(a), \
- (__v8sf)(__m256)(b), (c)); })
+ (__v8sf)(__m256)(b), (c))
-/// \brief Compares each of the corresponding scalar double-precision values of
+/// Compares each of the corresponding scalar double-precision values of
/// two 128-bit vectors of [2 x double], using the operation specified by the
/// immediate integer operand.
///
/// \param c
/// An immediate integer operand, with bits [4:0] specifying which comparison
/// operation to use: \n
-/// 0x00 : Equal (ordered, non-signaling)
-/// 0x01 : Less-than (ordered, signaling)
-/// 0x02 : Less-than-or-equal (ordered, signaling)
-/// 0x03 : Unordered (non-signaling)
-/// 0x04 : Not-equal (unordered, non-signaling)
-/// 0x05 : Not-less-than (unordered, signaling)
-/// 0x06 : Not-less-than-or-equal (unordered, signaling)
-/// 0x07 : Ordered (non-signaling)
-/// 0x08 : Equal (unordered, non-signaling)
-/// 0x09 : Not-greater-than-or-equal (unordered, signaling)
-/// 0x0a : Not-greater-than (unordered, signaling)
-/// 0x0b : False (ordered, non-signaling)
-/// 0x0c : Not-equal (ordered, non-signaling)
-/// 0x0d : Greater-than-or-equal (ordered, signaling)
-/// 0x0e : Greater-than (ordered, signaling)
-/// 0x0f : True (unordered, non-signaling)
-/// 0x10 : Equal (ordered, signaling)
-/// 0x11 : Less-than (ordered, non-signaling)
-/// 0x12 : Less-than-or-equal (ordered, non-signaling)
-/// 0x13 : Unordered (signaling)
-/// 0x14 : Not-equal (unordered, signaling)
-/// 0x15 : Not-less-than (unordered, non-signaling)
-/// 0x16 : Not-less-than-or-equal (unordered, non-signaling)
-/// 0x17 : Ordered (signaling)
-/// 0x18 : Equal (unordered, signaling)
-/// 0x19 : Not-greater-than-or-equal (unordered, non-signaling)
-/// 0x1a : Not-greater-than (unordered, non-signaling)
-/// 0x1b : False (ordered, signaling)
-/// 0x1c : Not-equal (ordered, signaling)
-/// 0x1d : Greater-than-or-equal (ordered, non-signaling)
-/// 0x1e : Greater-than (ordered, non-signaling)
-/// 0x1f : True (unordered, signaling)
+/// 0x00: Equal (ordered, non-signaling) \n
+/// 0x01: Less-than (ordered, signaling) \n
+/// 0x02: Less-than-or-equal (ordered, signaling) \n
+/// 0x03: Unordered (non-signaling) \n
+/// 0x04: Not-equal (unordered, non-signaling) \n
+/// 0x05: Not-less-than (unordered, signaling) \n
+/// 0x06: Not-less-than-or-equal (unordered, signaling) \n
+/// 0x07: Ordered (non-signaling) \n
+/// 0x08: Equal (unordered, non-signaling) \n
+/// 0x09: Not-greater-than-or-equal (unordered, signaling) \n
+/// 0x0A: Not-greater-than (unordered, signaling) \n
+/// 0x0B: False (ordered, non-signaling) \n
+/// 0x0C: Not-equal (ordered, non-signaling) \n
+/// 0x0D: Greater-than-or-equal (ordered, signaling) \n
+/// 0x0E: Greater-than (ordered, signaling) \n
+/// 0x0F: True (unordered, non-signaling) \n
+/// 0x10: Equal (ordered, signaling) \n
+/// 0x11: Less-than (ordered, non-signaling) \n
+/// 0x12: Less-than-or-equal (ordered, non-signaling) \n
+/// 0x13: Unordered (signaling) \n
+/// 0x14: Not-equal (unordered, signaling) \n
+/// 0x15: Not-less-than (unordered, non-signaling) \n
+/// 0x16: Not-less-than-or-equal (unordered, non-signaling) \n
+/// 0x17: Ordered (signaling) \n
+/// 0x18: Equal (unordered, signaling) \n
+/// 0x19: Not-greater-than-or-equal (unordered, non-signaling) \n
+/// 0x1A: Not-greater-than (unordered, non-signaling) \n
+/// 0x1B: False (ordered, signaling) \n
+/// 0x1C: Not-equal (ordered, signaling) \n
+/// 0x1D: Greater-than-or-equal (ordered, non-signaling) \n
+/// 0x1E: Greater-than (ordered, non-signaling) \n
+/// 0x1F: True (unordered, signaling)
/// \returns A 128-bit vector of [2 x double] containing the comparison results.
-#define _mm_cmp_sd(a, b, c) __extension__ ({ \
+#define _mm_cmp_sd(a, b, c) \
(__m128d)__builtin_ia32_cmpsd((__v2df)(__m128d)(a), \
- (__v2df)(__m128d)(b), (c)); })
+ (__v2df)(__m128d)(b), (c))
-/// \brief Compares each of the corresponding scalar values of two 128-bit
+/// Compares each of the corresponding scalar values of two 128-bit
/// vectors of [4 x float], using the operation specified by the immediate
/// integer operand.
///
/// \param c
/// An immediate integer operand, with bits [4:0] specifying which comparison
/// operation to use: \n
-/// 0x00 : Equal (ordered, non-signaling)
-/// 0x01 : Less-than (ordered, signaling)
-/// 0x02 : Less-than-or-equal (ordered, signaling)
-/// 0x03 : Unordered (non-signaling)
-/// 0x04 : Not-equal (unordered, non-signaling)
-/// 0x05 : Not-less-than (unordered, signaling)
-/// 0x06 : Not-less-than-or-equal (unordered, signaling)
-/// 0x07 : Ordered (non-signaling)
-/// 0x08 : Equal (unordered, non-signaling)
-/// 0x09 : Not-greater-than-or-equal (unordered, signaling)
-/// 0x0a : Not-greater-than (unordered, signaling)
-/// 0x0b : False (ordered, non-signaling)
-/// 0x0c : Not-equal (ordered, non-signaling)
-/// 0x0d : Greater-than-or-equal (ordered, signaling)
-/// 0x0e : Greater-than (ordered, signaling)
-/// 0x0f : True (unordered, non-signaling)
-/// 0x10 : Equal (ordered, signaling)
-/// 0x11 : Less-than (ordered, non-signaling)
-/// 0x12 : Less-than-or-equal (ordered, non-signaling)
-/// 0x13 : Unordered (signaling)
-/// 0x14 : Not-equal (unordered, signaling)
-/// 0x15 : Not-less-than (unordered, non-signaling)
-/// 0x16 : Not-less-than-or-equal (unordered, non-signaling)
-/// 0x17 : Ordered (signaling)
-/// 0x18 : Equal (unordered, signaling)
-/// 0x19 : Not-greater-than-or-equal (unordered, non-signaling)
-/// 0x1a : Not-greater-than (unordered, non-signaling)
-/// 0x1b : False (ordered, signaling)
-/// 0x1c : Not-equal (ordered, signaling)
-/// 0x1d : Greater-than-or-equal (ordered, non-signaling)
-/// 0x1e : Greater-than (ordered, non-signaling)
-/// 0x1f : True (unordered, signaling)
+/// 0x00: Equal (ordered, non-signaling) \n
+/// 0x01: Less-than (ordered, signaling) \n
+/// 0x02: Less-than-or-equal (ordered, signaling) \n
+/// 0x03: Unordered (non-signaling) \n
+/// 0x04: Not-equal (unordered, non-signaling) \n
+/// 0x05: Not-less-than (unordered, signaling) \n
+/// 0x06: Not-less-than-or-equal (unordered, signaling) \n
+/// 0x07: Ordered (non-signaling) \n
+/// 0x08: Equal (unordered, non-signaling) \n
+/// 0x09: Not-greater-than-or-equal (unordered, signaling) \n
+/// 0x0A: Not-greater-than (unordered, signaling) \n
+/// 0x0B: False (ordered, non-signaling) \n
+/// 0x0C: Not-equal (ordered, non-signaling) \n
+/// 0x0D: Greater-than-or-equal (ordered, signaling) \n
+/// 0x0E: Greater-than (ordered, signaling) \n
+/// 0x0F: True (unordered, non-signaling) \n
+/// 0x10: Equal (ordered, signaling) \n
+/// 0x11: Less-than (ordered, non-signaling) \n
+/// 0x12: Less-than-or-equal (ordered, non-signaling) \n
+/// 0x13: Unordered (signaling) \n
+/// 0x14: Not-equal (unordered, signaling) \n
+/// 0x15: Not-less-than (unordered, non-signaling) \n
+/// 0x16: Not-less-than-or-equal (unordered, non-signaling) \n
+/// 0x17: Ordered (signaling) \n
+/// 0x18: Equal (unordered, signaling) \n
+/// 0x19: Not-greater-than-or-equal (unordered, non-signaling) \n
+/// 0x1A: Not-greater-than (unordered, non-signaling) \n
+/// 0x1B: False (ordered, signaling) \n
+/// 0x1C: Not-equal (ordered, signaling) \n
+/// 0x1D: Greater-than-or-equal (ordered, non-signaling) \n
+/// 0x1E: Greater-than (ordered, non-signaling) \n
+/// 0x1F: True (unordered, signaling)
/// \returns A 128-bit vector of [4 x float] containing the comparison results.
-#define _mm_cmp_ss(a, b, c) __extension__ ({ \
+#define _mm_cmp_ss(a, b, c) \
(__m128)__builtin_ia32_cmpss((__v4sf)(__m128)(a), \
- (__v4sf)(__m128)(b), (c)); })
+ (__v4sf)(__m128)(b), (c))
-/// \brief Takes a [8 x i32] vector and returns the vector element value
+/// Takes a [8 x i32] vector and returns the vector element value
/// indexed by the immediate constant operand.
///
/// \headerfile <x86intrin.h>
/// element is extracted and returned.
/// \returns A 32-bit integer containing the extracted 32 bits of extended
/// packed data.
-static __inline int __DEFAULT_FN_ATTRS
-_mm256_extract_epi32(__m256i __a, const int __imm)
-{
- __v8si __b = (__v8si)__a;
- return __b[__imm & 7];
-}
+#define _mm256_extract_epi32(X, N) \
+ (int)__builtin_ia32_vec_ext_v8si((__v8si)(__m256i)(X), (int)(N))
-/// \brief Takes a [16 x i16] vector and returns the vector element value
+/// Takes a [16 x i16] vector and returns the vector element value
/// indexed by the immediate constant operand.
///
/// \headerfile <x86intrin.h>
/// element is extracted and returned.
/// \returns A 32-bit integer containing the extracted 16 bits of zero extended
/// packed data.
-static __inline int __DEFAULT_FN_ATTRS
-_mm256_extract_epi16(__m256i __a, const int __imm)
-{
- __v16hi __b = (__v16hi)__a;
- return (unsigned short)__b[__imm & 15];
-}
+#define _mm256_extract_epi16(X, N) \
+ (int)(unsigned short)__builtin_ia32_vec_ext_v16hi((__v16hi)(__m256i)(X), \
+ (int)(N))
-/// \brief Takes a [32 x i8] vector and returns the vector element value
+/// Takes a [32 x i8] vector and returns the vector element value
/// indexed by the immediate constant operand.
///
/// \headerfile <x86intrin.h>
/// element is extracted and returned.
/// \returns A 32-bit integer containing the extracted 8 bits of zero extended
/// packed data.
-static __inline int __DEFAULT_FN_ATTRS
-_mm256_extract_epi8(__m256i __a, const int __imm)
-{
- __v32qi __b = (__v32qi)__a;
- return (unsigned char)__b[__imm & 31];
-}
+#define _mm256_extract_epi8(X, N) \
+ (int)(unsigned char)__builtin_ia32_vec_ext_v32qi((__v32qi)(__m256i)(X), \
+ (int)(N))
#ifdef __x86_64__
-/// \brief Takes a [4 x i64] vector and returns the vector element value
+/// Takes a [4 x i64] vector and returns the vector element value
/// indexed by the immediate constant operand.
///
/// \headerfile <x86intrin.h>
/// element is extracted and returned.
/// \returns A 64-bit integer containing the extracted 64 bits of extended
/// packed data.
-static __inline long long __DEFAULT_FN_ATTRS
-_mm256_extract_epi64(__m256i __a, const int __imm)
-{
- __v4di __b = (__v4di)__a;
- return __b[__imm & 3];
-}
+#define _mm256_extract_epi64(X, N) \
+ (long long)__builtin_ia32_vec_ext_v4di((__v4di)(__m256i)(X), (int)(N))
#endif
-/// \brief Takes a [8 x i32] vector and replaces the vector element value
+/// Takes a [8 x i32] vector and replaces the vector element value
/// indexed by the immediate constant operand by a new value. Returns the
/// modified vector.
///
/// replaced.
/// \returns A copy of vector \a __a, after replacing its element indexed by
/// \a __imm with \a __b.
-static __inline __m256i __DEFAULT_FN_ATTRS
-_mm256_insert_epi32(__m256i __a, int __b, int const __imm)
-{
- __v8si __c = (__v8si)__a;
- __c[__imm & 7] = __b;
- return (__m256i)__c;
-}
+#define _mm256_insert_epi32(X, I, N) \
+ (__m256i)__builtin_ia32_vec_set_v8si((__v8si)(__m256i)(X), \
+ (int)(I), (int)(N))
-/// \brief Takes a [16 x i16] vector and replaces the vector element value
+/// Takes a [16 x i16] vector and replaces the vector element value
/// indexed by the immediate constant operand with a new value. Returns the
/// modified vector.
///
/// replaced.
/// \returns A copy of vector \a __a, after replacing its element indexed by
/// \a __imm with \a __b.
-static __inline __m256i __DEFAULT_FN_ATTRS
-_mm256_insert_epi16(__m256i __a, int __b, int const __imm)
-{
- __v16hi __c = (__v16hi)__a;
- __c[__imm & 15] = __b;
- return (__m256i)__c;
-}
+#define _mm256_insert_epi16(X, I, N) \
+ (__m256i)__builtin_ia32_vec_set_v16hi((__v16hi)(__m256i)(X), \
+ (int)(I), (int)(N))
-/// \brief Takes a [32 x i8] vector and replaces the vector element value
+/// Takes a [32 x i8] vector and replaces the vector element value
/// indexed by the immediate constant operand with a new value. Returns the
/// modified vector.
///
/// replaced.
/// \returns A copy of vector \a __a, after replacing its element indexed by
/// \a __imm with \a __b.
-static __inline __m256i __DEFAULT_FN_ATTRS
-_mm256_insert_epi8(__m256i __a, int __b, int const __imm)
-{
- __v32qi __c = (__v32qi)__a;
- __c[__imm & 31] = __b;
- return (__m256i)__c;
-}
+#define _mm256_insert_epi8(X, I, N) \
+ (__m256i)__builtin_ia32_vec_set_v32qi((__v32qi)(__m256i)(X), \
+ (int)(I), (int)(N))
#ifdef __x86_64__
-/// \brief Takes a [4 x i64] vector and replaces the vector element value
+/// Takes a [4 x i64] vector and replaces the vector element value
/// indexed by the immediate constant operand with a new value. Returns the
/// modified vector.
///
/// replaced.
/// \returns A copy of vector \a __a, after replacing its element indexed by
/// \a __imm with \a __b.
-static __inline __m256i __DEFAULT_FN_ATTRS
-_mm256_insert_epi64(__m256i __a, long long __b, int const __imm)
-{
- __v4di __c = (__v4di)__a;
- __c[__imm & 3] = __b;
- return (__m256i)__c;
-}
+#define _mm256_insert_epi64(X, I, N) \
+ (__m256i)__builtin_ia32_vec_set_v4di((__v4di)(__m256i)(X), \
+ (long long)(I), (int)(N))
#endif
/* Conversion */
-/// \brief Converts a vector of [4 x i32] into a vector of [4 x double].
+/// Converts a vector of [4 x i32] into a vector of [4 x double].
///
/// \headerfile <x86intrin.h>
///
return (__m256d)__builtin_convertvector((__v4si)__a, __v4df);
}
-/// \brief Converts a vector of [8 x i32] into a vector of [8 x float].
+/// Converts a vector of [8 x i32] into a vector of [8 x float].
///
/// \headerfile <x86intrin.h>
///
static __inline __m256 __DEFAULT_FN_ATTRS
_mm256_cvtepi32_ps(__m256i __a)
{
- return (__m256)__builtin_ia32_cvtdq2ps256((__v8si) __a);
+ return (__m256)__builtin_convertvector((__v8si)__a, __v8sf);
}
-/// \brief Converts a 256-bit vector of [4 x double] into a 128-bit vector of
+/// Converts a 256-bit vector of [4 x double] into a 128-bit vector of
/// [4 x float].
///
/// \headerfile <x86intrin.h>
return (__m128)__builtin_ia32_cvtpd2ps256((__v4df) __a);
}
-/// \brief Converts a vector of [8 x float] into a vector of [8 x i32].
+/// Converts a vector of [8 x float] into a vector of [8 x i32].
///
/// \headerfile <x86intrin.h>
///
return (__m256i)__builtin_ia32_cvtps2dq256((__v8sf) __a);
}
-/// \brief Converts a 128-bit vector of [4 x float] into a 256-bit vector of [4
+/// Converts a 128-bit vector of [4 x float] into a 256-bit vector of [4
/// x double].
///
/// \headerfile <x86intrin.h>
return (__m256d)__builtin_convertvector((__v4sf)__a, __v4df);
}
-/// \brief Converts a 256-bit vector of [4 x double] into a 128-bit vector of [4
+/// Converts a 256-bit vector of [4 x double] into a 128-bit vector of [4
/// x i32], truncating the result by rounding towards zero when it is
/// inexact.
///
return (__m128i)__builtin_ia32_cvttpd2dq256((__v4df) __a);
}
-/// \brief Converts a 256-bit vector of [4 x double] into a 128-bit vector of [4
+/// Converts a 256-bit vector of [4 x double] into a 128-bit vector of [4
/// x i32]. When a conversion is inexact, the value returned is rounded
/// according to the rounding control bits in the MXCSR register.
///
return (__m128i)__builtin_ia32_cvtpd2dq256((__v4df) __a);
}
-/// \brief Converts a vector of [8 x float] into a vector of [8 x i32],
+/// Converts a vector of [8 x float] into a vector of [8 x i32],
/// truncating the result by rounding towards zero when it is inexact.
///
/// \headerfile <x86intrin.h>
return (__m256i)__builtin_ia32_cvttps2dq256((__v8sf) __a);
}
-/// \brief Returns the first element of the input vector of [4 x double].
+/// Returns the first element of the input vector of [4 x double].
///
/// \headerfile <avxintrin.h>
///
return __a[0];
}
-/// \brief Returns the first element of the input vector of [8 x i32].
+/// Returns the first element of the input vector of [8 x i32].
///
/// \headerfile <avxintrin.h>
///
return __b[0];
}
-/// \brief Returns the first element of the input vector of [8 x float].
+/// Returns the first element of the input vector of [8 x float].
///
/// \headerfile <avxintrin.h>
///
}
/* Vector replicate */
-/// \brief Moves and duplicates high-order (odd-indexed) values from a 256-bit
-/// vector of [8 x float] to float values in a 256-bit vector of
-/// [8 x float].
+/// Moves and duplicates odd-indexed values from a 256-bit vector of
+/// [8 x float] to float values in a 256-bit vector of [8 x float].
///
/// \headerfile <x86intrin.h>
///
return __builtin_shufflevector((__v8sf)__a, (__v8sf)__a, 1, 1, 3, 3, 5, 5, 7, 7);
}
-/// \brief Moves and duplicates low-order (even-indexed) values from a 256-bit
-/// vector of [8 x float] to float values in a 256-bit vector of [8 x float].
+/// Moves and duplicates even-indexed values from a 256-bit vector of
+/// [8 x float] to float values in a 256-bit vector of [8 x float].
///
/// \headerfile <x86intrin.h>
///
return __builtin_shufflevector((__v8sf)__a, (__v8sf)__a, 0, 0, 2, 2, 4, 4, 6, 6);
}
-/// \brief Moves and duplicates double-precision floating point values from a
+/// Moves and duplicates double-precision floating point values from a
/// 256-bit vector of [4 x double] to double-precision values in a 256-bit
/// vector of [4 x double].
///
}
/* Unpack and Interleave */
-/// \brief Unpacks the odd-indexed vector elements from two 256-bit vectors of
+/// Unpacks the odd-indexed vector elements from two 256-bit vectors of
/// [4 x double] and interleaves them into a 256-bit vector of [4 x double].
///
/// \headerfile <x86intrin.h>
return __builtin_shufflevector((__v4df)__a, (__v4df)__b, 1, 5, 1+2, 5+2);
}
-/// \brief Unpacks the even-indexed vector elements from two 256-bit vectors of
+/// Unpacks the even-indexed vector elements from two 256-bit vectors of
/// [4 x double] and interleaves them into a 256-bit vector of [4 x double].
///
/// \headerfile <x86intrin.h>
return __builtin_shufflevector((__v4df)__a, (__v4df)__b, 0, 4, 0+2, 4+2);
}
-/// \brief Unpacks the 32-bit vector elements 2, 3, 6 and 7 from each of the
+/// Unpacks the 32-bit vector elements 2, 3, 6 and 7 from each of the
/// two 256-bit vectors of [8 x float] and interleaves them into a 256-bit
/// vector of [8 x float].
///
return __builtin_shufflevector((__v8sf)__a, (__v8sf)__b, 2, 10, 2+1, 10+1, 6, 14, 6+1, 14+1);
}
-/// \brief Unpacks the 32-bit vector elements 0, 1, 4 and 5 from each of the
+/// Unpacks the 32-bit vector elements 0, 1, 4 and 5 from each of the
/// two 256-bit vectors of [8 x float] and interleaves them into a 256-bit
/// vector of [8 x float].
///
}
/* Bit Test */
-/// \brief Given two 128-bit floating-point vectors of [2 x double], perform an
+/// Given two 128-bit floating-point vectors of [2 x double], perform an
/// element-by-element comparison of the double-precision element in the
/// first source vector and the corresponding element in the second source
/// vector.
/// \param __b
/// A 128-bit vector of [2 x double].
/// \returns the ZF flag in the EFLAGS register.
-static __inline int __DEFAULT_FN_ATTRS
+static __inline int __DEFAULT_FN_ATTRS128
_mm_testz_pd(__m128d __a, __m128d __b)
{
return __builtin_ia32_vtestzpd((__v2df)__a, (__v2df)__b);
}
-/// \brief Given two 128-bit floating-point vectors of [2 x double], perform an
+/// Given two 128-bit floating-point vectors of [2 x double], perform an
/// element-by-element comparison of the double-precision element in the
/// first source vector and the corresponding element in the second source
/// vector.
/// \param __b
/// A 128-bit vector of [2 x double].
/// \returns the CF flag in the EFLAGS register.
-static __inline int __DEFAULT_FN_ATTRS
+static __inline int __DEFAULT_FN_ATTRS128
_mm_testc_pd(__m128d __a, __m128d __b)
{
return __builtin_ia32_vtestcpd((__v2df)__a, (__v2df)__b);
}
-/// \brief Given two 128-bit floating-point vectors of [2 x double], perform an
+/// Given two 128-bit floating-point vectors of [2 x double], perform an
/// element-by-element comparison of the double-precision element in the
/// first source vector and the corresponding element in the second source
/// vector.
/// \param __b
/// A 128-bit vector of [2 x double].
/// \returns 1 if both the ZF and CF flags are set to 0, otherwise returns 0.
-static __inline int __DEFAULT_FN_ATTRS
+static __inline int __DEFAULT_FN_ATTRS128
_mm_testnzc_pd(__m128d __a, __m128d __b)
{
return __builtin_ia32_vtestnzcpd((__v2df)__a, (__v2df)__b);
}
-/// \brief Given two 128-bit floating-point vectors of [4 x float], perform an
+/// Given two 128-bit floating-point vectors of [4 x float], perform an
/// element-by-element comparison of the single-precision element in the
/// first source vector and the corresponding element in the second source
/// vector.
/// \param __b
/// A 128-bit vector of [4 x float].
/// \returns the ZF flag.
-static __inline int __DEFAULT_FN_ATTRS
+static __inline int __DEFAULT_FN_ATTRS128
_mm_testz_ps(__m128 __a, __m128 __b)
{
return __builtin_ia32_vtestzps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Given two 128-bit floating-point vectors of [4 x float], perform an
+/// Given two 128-bit floating-point vectors of [4 x float], perform an
/// element-by-element comparison of the single-precision element in the
/// first source vector and the corresponding element in the second source
/// vector.
/// \param __b
/// A 128-bit vector of [4 x float].
/// \returns the CF flag.
-static __inline int __DEFAULT_FN_ATTRS
+static __inline int __DEFAULT_FN_ATTRS128
_mm_testc_ps(__m128 __a, __m128 __b)
{
return __builtin_ia32_vtestcps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Given two 128-bit floating-point vectors of [4 x float], perform an
+/// Given two 128-bit floating-point vectors of [4 x float], perform an
/// element-by-element comparison of the single-precision element in the
/// first source vector and the corresponding element in the second source
/// vector.
/// \param __b
/// A 128-bit vector of [4 x float].
/// \returns 1 if both the ZF and CF flags are set to 0, otherwise returns 0.
-static __inline int __DEFAULT_FN_ATTRS
+static __inline int __DEFAULT_FN_ATTRS128
_mm_testnzc_ps(__m128 __a, __m128 __b)
{
return __builtin_ia32_vtestnzcps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Given two 256-bit floating-point vectors of [4 x double], perform an
+/// Given two 256-bit floating-point vectors of [4 x double], perform an
/// element-by-element comparison of the double-precision elements in the
/// first source vector and the corresponding elements in the second source
/// vector.
return __builtin_ia32_vtestzpd256((__v4df)__a, (__v4df)__b);
}
-/// \brief Given two 256-bit floating-point vectors of [4 x double], perform an
+/// Given two 256-bit floating-point vectors of [4 x double], perform an
/// element-by-element comparison of the double-precision elements in the
/// first source vector and the corresponding elements in the second source
/// vector.
return __builtin_ia32_vtestcpd256((__v4df)__a, (__v4df)__b);
}
-/// \brief Given two 256-bit floating-point vectors of [4 x double], perform an
+/// Given two 256-bit floating-point vectors of [4 x double], perform an
/// element-by-element comparison of the double-precision elements in the
/// first source vector and the corresponding elements in the second source
/// vector.
return __builtin_ia32_vtestnzcpd256((__v4df)__a, (__v4df)__b);
}
-/// \brief Given two 256-bit floating-point vectors of [8 x float], perform an
+/// Given two 256-bit floating-point vectors of [8 x float], perform an
/// element-by-element comparison of the single-precision element in the
/// first source vector and the corresponding element in the second source
/// vector.
return __builtin_ia32_vtestzps256((__v8sf)__a, (__v8sf)__b);
}
-/// \brief Given two 256-bit floating-point vectors of [8 x float], perform an
+/// Given two 256-bit floating-point vectors of [8 x float], perform an
/// element-by-element comparison of the single-precision element in the
/// first source vector and the corresponding element in the second source
/// vector.
return __builtin_ia32_vtestcps256((__v8sf)__a, (__v8sf)__b);
}
-/// \brief Given two 256-bit floating-point vectors of [8 x float], perform an
+/// Given two 256-bit floating-point vectors of [8 x float], perform an
/// element-by-element comparison of the single-precision elements in the
/// first source vector and the corresponding elements in the second source
/// vector.
return __builtin_ia32_vtestnzcps256((__v8sf)__a, (__v8sf)__b);
}
-/// \brief Given two 256-bit integer vectors, perform a bit-by-bit comparison
+/// Given two 256-bit integer vectors, perform a bit-by-bit comparison
/// of the two source vectors.
///
/// The EFLAGS register is updated as follows: \n
return __builtin_ia32_ptestz256((__v4di)__a, (__v4di)__b);
}
-/// \brief Given two 256-bit integer vectors, perform a bit-by-bit comparison
+/// Given two 256-bit integer vectors, perform a bit-by-bit comparison
/// of the two source vectors.
///
/// The EFLAGS register is updated as follows: \n
return __builtin_ia32_ptestc256((__v4di)__a, (__v4di)__b);
}
-/// \brief Given two 256-bit integer vectors, perform a bit-by-bit comparison
+/// Given two 256-bit integer vectors, perform a bit-by-bit comparison
/// of the two source vectors.
///
/// The EFLAGS register is updated as follows: \n
}
/* Vector extract sign mask */
-/// \brief Extracts the sign bits of double-precision floating point elements
+/// Extracts the sign bits of double-precision floating point elements
/// in a 256-bit vector of [4 x double] and writes them to the lower order
/// bits of the return value.
///
return __builtin_ia32_movmskpd256((__v4df)__a);
}
-/// \brief Extracts the sign bits of double-precision floating point elements
+/// Extracts the sign bits of single-precision floating point elements
/// in a 256-bit vector of [8 x float] and writes them to the lower order
/// bits of the return value.
///
/// This intrinsic corresponds to the <c> VMOVMSKPS </c> instruction.
///
/// \param __a
-/// A 256-bit vector of [8 x float] containing the double-precision floating
+/// A 256-bit vector of [8 x float] containing the single-precision floating
/// point values with sign bits to be extracted.
/// \returns The sign bits from the operand, written to bits [7:0].
static __inline int __DEFAULT_FN_ATTRS
}
/* Vector __zero */
-/// \brief Zeroes the contents of all XMM or YMM registers.
+/// Zeroes the contents of all XMM or YMM registers.
///
/// \headerfile <x86intrin.h>
///
/// This intrinsic corresponds to the <c> VZEROALL </c> instruction.
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __attribute__((__always_inline__, __nodebug__, __target__("avx")))
_mm256_zeroall(void)
{
__builtin_ia32_vzeroall();
}
-/// \brief Zeroes the upper 128 bits (bits 255:128) of all YMM registers.
+/// Zeroes the upper 128 bits (bits 255:128) of all YMM registers.
///
/// \headerfile <x86intrin.h>
///
/// This intrinsic corresponds to the <c> VZEROUPPER </c> instruction.
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __attribute__((__always_inline__, __nodebug__, __target__("avx")))
_mm256_zeroupper(void)
{
__builtin_ia32_vzeroupper();
}
/* Vector load with broadcast */
-/// \brief Loads a scalar single-precision floating point value from the
+/// Loads a scalar single-precision floating point value from the
/// specified address pointed to by \a __a and broadcasts it to the elements
/// of a [4 x float] vector.
///
/// The single-precision floating point value to be broadcast.
/// \returns A 128-bit vector of [4 x float] whose 32-bit elements are set
/// equal to the broadcast value.
-static __inline __m128 __DEFAULT_FN_ATTRS
+static __inline __m128 __DEFAULT_FN_ATTRS128
_mm_broadcast_ss(float const *__a)
{
float __f = *__a;
- return (__m128)(__v4sf){ __f, __f, __f, __f };
+ return __extension__ (__m128)(__v4sf){ __f, __f, __f, __f };
}
-/// \brief Loads a scalar double-precision floating point value from the
+/// Loads a scalar double-precision floating point value from the
/// specified address pointed to by \a __a and broadcasts it to the elements
/// of a [4 x double] vector.
///
_mm256_broadcast_sd(double const *__a)
{
double __d = *__a;
- return (__m256d)(__v4df){ __d, __d, __d, __d };
+ return __extension__ (__m256d)(__v4df){ __d, __d, __d, __d };
}
-/// \brief Loads a scalar single-precision floating point value from the
+/// Loads a scalar single-precision floating point value from the
/// specified address pointed to by \a __a and broadcasts it to the elements
/// of a [8 x float] vector.
///
_mm256_broadcast_ss(float const *__a)
{
float __f = *__a;
- return (__m256)(__v8sf){ __f, __f, __f, __f, __f, __f, __f, __f };
+ return __extension__ (__m256)(__v8sf){ __f, __f, __f, __f, __f, __f, __f, __f };
}
-/// \brief Loads the data from a 128-bit vector of [2 x double] from the
+/// Loads the data from a 128-bit vector of [2 x double] from the
/// specified address pointed to by \a __a and broadcasts it to 128-bit
/// elements in a 256-bit vector of [4 x double].
///
static __inline __m256d __DEFAULT_FN_ATTRS
_mm256_broadcast_pd(__m128d const *__a)
{
- return (__m256d)__builtin_ia32_vbroadcastf128_pd256((__v2df const *)__a);
+ __m128d __b = _mm_loadu_pd((const double *)__a);
+ return (__m256d)__builtin_shufflevector((__v2df)__b, (__v2df)__b,
+ 0, 1, 0, 1);
}
-/// \brief Loads the data from a 128-bit vector of [4 x float] from the
+/// Loads the data from a 128-bit vector of [4 x float] from the
/// specified address pointed to by \a __a and broadcasts it to 128-bit
/// elements in a 256-bit vector of [8 x float].
///
static __inline __m256 __DEFAULT_FN_ATTRS
_mm256_broadcast_ps(__m128 const *__a)
{
- return (__m256)__builtin_ia32_vbroadcastf128_ps256((__v4sf const *)__a);
+ __m128 __b = _mm_loadu_ps((const float *)__a);
+ return (__m256)__builtin_shufflevector((__v4sf)__b, (__v4sf)__b,
+ 0, 1, 2, 3, 0, 1, 2, 3);
}
/* SIMD load ops */
-/// \brief Loads 4 double-precision floating point values from a 32-byte aligned
+/// Loads 4 double-precision floating point values from a 32-byte aligned
/// memory location pointed to by \a __p into a vector of [4 x double].
///
/// \headerfile <x86intrin.h>
static __inline __m256d __DEFAULT_FN_ATTRS
_mm256_load_pd(double const *__p)
{
- return *(__m256d *)__p;
+ return *(const __m256d *)__p;
}
-/// \brief Loads 8 single-precision floating point values from a 32-byte aligned
+/// Loads 8 single-precision floating point values from a 32-byte aligned
/// memory location pointed to by \a __p into a vector of [8 x float].
///
/// \headerfile <x86intrin.h>
static __inline __m256 __DEFAULT_FN_ATTRS
_mm256_load_ps(float const *__p)
{
- return *(__m256 *)__p;
+ return *(const __m256 *)__p;
}
-/// \brief Loads 4 double-precision floating point values from an unaligned
+/// Loads 4 double-precision floating point values from an unaligned
/// memory location pointed to by \a __p into a vector of [4 x double].
///
/// \headerfile <x86intrin.h>
_mm256_loadu_pd(double const *__p)
{
struct __loadu_pd {
- __m256d __v;
+ __m256d_u __v;
} __attribute__((__packed__, __may_alias__));
- return ((struct __loadu_pd*)__p)->__v;
+ return ((const struct __loadu_pd*)__p)->__v;
}
-/// \brief Loads 8 single-precision floating point values from an unaligned
+/// Loads 8 single-precision floating point values from an unaligned
/// memory location pointed to by \a __p into a vector of [8 x float].
///
/// \headerfile <x86intrin.h>
_mm256_loadu_ps(float const *__p)
{
struct __loadu_ps {
- __m256 __v;
+ __m256_u __v;
} __attribute__((__packed__, __may_alias__));
- return ((struct __loadu_ps*)__p)->__v;
+ return ((const struct __loadu_ps*)__p)->__v;
}
-/// \brief Loads 256 bits of integer data from a 32-byte aligned memory
+/// Loads 256 bits of integer data from a 32-byte aligned memory
/// location pointed to by \a __p into elements of a 256-bit integer vector.
///
/// \headerfile <x86intrin.h>
return *__p;
}
-/// \brief Loads 256 bits of integer data from an unaligned memory location
+/// Loads 256 bits of integer data from an unaligned memory location
/// pointed to by \a __p into a 256-bit integer vector.
///
/// \headerfile <x86intrin.h>
/// A pointer to a 256-bit integer vector containing integer values.
/// \returns A 256-bit integer vector containing the moved values.
static __inline __m256i __DEFAULT_FN_ATTRS
-_mm256_loadu_si256(__m256i const *__p)
+_mm256_loadu_si256(__m256i_u const *__p)
{
struct __loadu_si256 {
- __m256i __v;
+ __m256i_u __v;
} __attribute__((__packed__, __may_alias__));
- return ((struct __loadu_si256*)__p)->__v;
+ return ((const struct __loadu_si256*)__p)->__v;
}
-/// \brief Loads 256 bits of integer data from an unaligned memory location
+/// Loads 256 bits of integer data from an unaligned memory location
/// pointed to by \a __p into a 256-bit integer vector. This intrinsic may
/// perform better than \c _mm256_loadu_si256 when the data crosses a cache
/// line boundary.
}
/* SIMD store ops */
-/// \brief Stores double-precision floating point values from a 256-bit vector
+/// Stores double-precision floating point values from a 256-bit vector
/// of [4 x double] to a 32-byte aligned memory location pointed to by
/// \a __p.
///
*(__m256d *)__p = __a;
}
-/// \brief Stores single-precision floating point values from a 256-bit vector
+/// Stores single-precision floating point values from a 256-bit vector
/// of [8 x float] to a 32-byte aligned memory location pointed to by \a __p.
///
/// \headerfile <x86intrin.h>
*(__m256 *)__p = __a;
}
-/// \brief Stores double-precision floating point values from a 256-bit vector
+/// Stores double-precision floating point values from a 256-bit vector
/// of [4 x double] to an unaligned memory location pointed to by \a __p.
///
/// \headerfile <x86intrin.h>
_mm256_storeu_pd(double *__p, __m256d __a)
{
struct __storeu_pd {
- __m256d __v;
+ __m256d_u __v;
} __attribute__((__packed__, __may_alias__));
((struct __storeu_pd*)__p)->__v = __a;
}
-/// \brief Stores single-precision floating point values from a 256-bit vector
+/// Stores single-precision floating point values from a 256-bit vector
/// of [8 x float] to an unaligned memory location pointed to by \a __p.
///
/// \headerfile <x86intrin.h>
_mm256_storeu_ps(float *__p, __m256 __a)
{
struct __storeu_ps {
- __m256 __v;
+ __m256_u __v;
} __attribute__((__packed__, __may_alias__));
((struct __storeu_ps*)__p)->__v = __a;
}
-/// \brief Stores integer values from a 256-bit integer vector to a 32-byte
+/// Stores integer values from a 256-bit integer vector to a 32-byte
/// aligned memory location pointed to by \a __p.
///
/// \headerfile <x86intrin.h>
*__p = __a;
}
-/// \brief Stores integer values from a 256-bit integer vector to an unaligned
+/// Stores integer values from a 256-bit integer vector to an unaligned
/// memory location pointed to by \a __p.
///
/// \headerfile <x86intrin.h>
/// \param __a
/// A 256-bit integer vector containing the values to be moved.
static __inline void __DEFAULT_FN_ATTRS
-_mm256_storeu_si256(__m256i *__p, __m256i __a)
+_mm256_storeu_si256(__m256i_u *__p, __m256i __a)
{
struct __storeu_si256 {
- __m256i __v;
+ __m256i_u __v;
} __attribute__((__packed__, __may_alias__));
((struct __storeu_si256*)__p)->__v = __a;
}
/* Conditional load ops */
-/// \brief Conditionally loads double-precision floating point elements from a
+/// Conditionally loads double-precision floating point elements from a
/// memory location pointed to by \a __p into a 128-bit vector of
/// [2 x double], depending on the mask bits associated with each data
/// element.
/// corresponding value in the memory location is not loaded and the
/// corresponding field in the return value is set to zero.
/// \returns A 128-bit vector of [2 x double] containing the loaded values.
-static __inline __m128d __DEFAULT_FN_ATTRS
+static __inline __m128d __DEFAULT_FN_ATTRS128
_mm_maskload_pd(double const *__p, __m128i __m)
{
return (__m128d)__builtin_ia32_maskloadpd((const __v2df *)__p, (__v2di)__m);
}
-/// \brief Conditionally loads double-precision floating point elements from a
+/// Conditionally loads double-precision floating point elements from a
/// memory location pointed to by \a __p into a 256-bit vector of
/// [4 x double], depending on the mask bits associated with each data
/// element.
(__v4di)__m);
}
-/// \brief Conditionally loads single-precision floating point elements from a
+/// Conditionally loads single-precision floating point elements from a
/// memory location pointed to by \a __p into a 128-bit vector of
/// [4 x float], depending on the mask bits associated with each data
/// element.
/// corresponding value in the memory location is not loaded and the
/// corresponding field in the return value is set to zero.
/// \returns A 128-bit vector of [4 x float] containing the loaded values.
-static __inline __m128 __DEFAULT_FN_ATTRS
+static __inline __m128 __DEFAULT_FN_ATTRS128
_mm_maskload_ps(float const *__p, __m128i __m)
{
return (__m128)__builtin_ia32_maskloadps((const __v4sf *)__p, (__v4si)__m);
}
-/// \brief Conditionally loads single-precision floating point elements from a
+/// Conditionally loads single-precision floating point elements from a
/// memory location pointed to by \a __p into a 256-bit vector of
/// [8 x float], depending on the mask bits associated with each data
/// element.
}
/* Conditional store ops */
-/// \brief Moves single-precision floating point values from a 256-bit vector
+/// Moves single-precision floating point values from a 256-bit vector
/// of [8 x float] to a memory location pointed to by \a __p, according to
/// the specified mask.
///
__builtin_ia32_maskstoreps256((__v8sf *)__p, (__v8si)__m, (__v8sf)__a);
}
-/// \brief Moves double-precision values from a 128-bit vector of [2 x double]
+/// Moves double-precision values from a 128-bit vector of [2 x double]
/// to a memory location pointed to by \a __p, according to the specified
/// mask.
///
/// changed.
/// \param __a
/// A 128-bit vector of [2 x double] containing the values to be stored.
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS128
_mm_maskstore_pd(double *__p, __m128i __m, __m128d __a)
{
__builtin_ia32_maskstorepd((__v2df *)__p, (__v2di)__m, (__v2df)__a);
}
-/// \brief Moves double-precision values from a 256-bit vector of [4 x double]
+/// Moves double-precision values from a 256-bit vector of [4 x double]
/// to a memory location pointed to by \a __p, according to the specified
/// mask.
///
__builtin_ia32_maskstorepd256((__v4df *)__p, (__v4di)__m, (__v4df)__a);
}
-/// \brief Moves single-precision floating point values from a 128-bit vector
+/// Moves single-precision floating point values from a 128-bit vector
/// of [4 x float] to a memory location pointed to by \a __p, according to
/// the specified mask.
///
/// changed.
/// \param __a
/// A 128-bit vector of [4 x float] containing the values to be stored.
-static __inline void __DEFAULT_FN_ATTRS
+static __inline void __DEFAULT_FN_ATTRS128
_mm_maskstore_ps(float *__p, __m128i __m, __m128 __a)
{
__builtin_ia32_maskstoreps((__v4sf *)__p, (__v4si)__m, (__v4sf)__a);
}
/* Cacheability support ops */
-/// \brief Moves integer data from a 256-bit integer vector to a 32-byte
+/// Moves integer data from a 256-bit integer vector to a 32-byte
/// aligned memory location. To minimize caching, the data is flagged as
/// non-temporal (unlikely to be used again soon).
///
__builtin_nontemporal_store((__v4di_aligned)__b, (__v4di_aligned*)__a);
}
-/// \brief Moves double-precision values from a 256-bit vector of [4 x double]
+/// Moves double-precision values from a 256-bit vector of [4 x double]
/// to a 32-byte aligned memory location. To minimize caching, the data is
/// flagged as non-temporal (unlikely to be used again soon).
///
__builtin_nontemporal_store((__v4df_aligned)__b, (__v4df_aligned*)__a);
}
-/// \brief Moves single-precision floating point values from a 256-bit vector
+/// Moves single-precision floating point values from a 256-bit vector
/// of [8 x float] to a 32-byte aligned memory location. To minimize
/// caching, the data is flagged as non-temporal (unlikely to be used again
/// soon).
}
/* Create vectors */
-/// \brief Create a 256-bit vector of [4 x double] with undefined values.
+/// Create a 256-bit vector of [4 x double] with undefined values.
///
/// \headerfile <x86intrin.h>
///
return (__m256d)__builtin_ia32_undef256();
}
-/// \brief Create a 256-bit vector of [8 x float] with undefined values.
+/// Create a 256-bit vector of [8 x float] with undefined values.
///
/// \headerfile <x86intrin.h>
///
return (__m256)__builtin_ia32_undef256();
}
-/// \brief Create a 256-bit integer vector with undefined values.
+/// Create a 256-bit integer vector with undefined values.
///
/// \headerfile <x86intrin.h>
///
return (__m256i)__builtin_ia32_undef256();
}
-/// \brief Constructs a 256-bit floating-point vector of [4 x double]
+/// Constructs a 256-bit floating-point vector of [4 x double]
/// initialized with the specified double-precision floating-point values.
///
/// \headerfile <x86intrin.h>
static __inline __m256d __DEFAULT_FN_ATTRS
_mm256_set_pd(double __a, double __b, double __c, double __d)
{
- return (__m256d){ __d, __c, __b, __a };
+ return __extension__ (__m256d){ __d, __c, __b, __a };
}
-/// \brief Constructs a 256-bit floating-point vector of [8 x float] initialized
+/// Constructs a 256-bit floating-point vector of [8 x float] initialized
/// with the specified single-precision floating-point values.
///
/// \headerfile <x86intrin.h>
_mm256_set_ps(float __a, float __b, float __c, float __d,
float __e, float __f, float __g, float __h)
{
- return (__m256){ __h, __g, __f, __e, __d, __c, __b, __a };
+ return __extension__ (__m256){ __h, __g, __f, __e, __d, __c, __b, __a };
}
-/// \brief Constructs a 256-bit integer vector initialized with the specified
+/// Constructs a 256-bit integer vector initialized with the specified
/// 32-bit integral values.
///
/// \headerfile <x86intrin.h>
_mm256_set_epi32(int __i0, int __i1, int __i2, int __i3,
int __i4, int __i5, int __i6, int __i7)
{
- return (__m256i)(__v8si){ __i7, __i6, __i5, __i4, __i3, __i2, __i1, __i0 };
+ return __extension__ (__m256i)(__v8si){ __i7, __i6, __i5, __i4, __i3, __i2, __i1, __i0 };
}
-/// \brief Constructs a 256-bit integer vector initialized with the specified
+/// Constructs a 256-bit integer vector initialized with the specified
/// 16-bit integral values.
///
/// \headerfile <x86intrin.h>
short __w07, short __w06, short __w05, short __w04,
short __w03, short __w02, short __w01, short __w00)
{
- return (__m256i)(__v16hi){ __w00, __w01, __w02, __w03, __w04, __w05, __w06,
+ return __extension__ (__m256i)(__v16hi){ __w00, __w01, __w02, __w03, __w04, __w05, __w06,
__w07, __w08, __w09, __w10, __w11, __w12, __w13, __w14, __w15 };
}
-/// \brief Constructs a 256-bit integer vector initialized with the specified
+/// Constructs a 256-bit integer vector initialized with the specified
/// 8-bit integral values.
///
/// \headerfile <x86intrin.h>
char __b07, char __b06, char __b05, char __b04,
char __b03, char __b02, char __b01, char __b00)
{
- return (__m256i)(__v32qi){
+ return __extension__ (__m256i)(__v32qi){
__b00, __b01, __b02, __b03, __b04, __b05, __b06, __b07,
__b08, __b09, __b10, __b11, __b12, __b13, __b14, __b15,
__b16, __b17, __b18, __b19, __b20, __b21, __b22, __b23,
};
}
-/// \brief Constructs a 256-bit integer vector initialized with the specified
+/// Constructs a 256-bit integer vector initialized with the specified
/// 64-bit integral values.
///
/// \headerfile <x86intrin.h>
static __inline __m256i __DEFAULT_FN_ATTRS
_mm256_set_epi64x(long long __a, long long __b, long long __c, long long __d)
{
- return (__m256i)(__v4di){ __d, __c, __b, __a };
+ return __extension__ (__m256i)(__v4di){ __d, __c, __b, __a };
}
/* Create vectors with elements in reverse order */
-/// \brief Constructs a 256-bit floating-point vector of [4 x double],
+/// Constructs a 256-bit floating-point vector of [4 x double],
/// initialized in reverse order with the specified double-precision
/// floating-point values.
///
static __inline __m256d __DEFAULT_FN_ATTRS
_mm256_setr_pd(double __a, double __b, double __c, double __d)
{
- return (__m256d){ __a, __b, __c, __d };
+ return _mm256_set_pd(__d, __c, __b, __a);
}
-/// \brief Constructs a 256-bit floating-point vector of [8 x float],
+/// Constructs a 256-bit floating-point vector of [8 x float],
/// initialized in reverse order with the specified single-precision
/// float-point values.
///
_mm256_setr_ps(float __a, float __b, float __c, float __d,
float __e, float __f, float __g, float __h)
{
- return (__m256){ __a, __b, __c, __d, __e, __f, __g, __h };
+ return _mm256_set_ps(__h, __g, __f, __e, __d, __c, __b, __a);
}
-/// \brief Constructs a 256-bit integer vector, initialized in reverse order
+/// Constructs a 256-bit integer vector, initialized in reverse order
/// with the specified 32-bit integral values.
///
/// \headerfile <x86intrin.h>
_mm256_setr_epi32(int __i0, int __i1, int __i2, int __i3,
int __i4, int __i5, int __i6, int __i7)
{
- return (__m256i)(__v8si){ __i0, __i1, __i2, __i3, __i4, __i5, __i6, __i7 };
+ return _mm256_set_epi32(__i7, __i6, __i5, __i4, __i3, __i2, __i1, __i0);
}
-/// \brief Constructs a 256-bit integer vector, initialized in reverse order
+/// Constructs a 256-bit integer vector, initialized in reverse order
/// with the specified 16-bit integral values.
///
/// \headerfile <x86intrin.h>
short __w07, short __w06, short __w05, short __w04,
short __w03, short __w02, short __w01, short __w00)
{
- return (__m256i)(__v16hi){ __w15, __w14, __w13, __w12, __w11, __w10, __w09,
- __w08, __w07, __w06, __w05, __w04, __w03, __w02, __w01, __w00 };
+ return _mm256_set_epi16(__w00, __w01, __w02, __w03,
+ __w04, __w05, __w06, __w07,
+ __w08, __w09, __w10, __w11,
+ __w12, __w13, __w14, __w15);
}
-/// \brief Constructs a 256-bit integer vector, initialized in reverse order
+/// Constructs a 256-bit integer vector, initialized in reverse order
/// with the specified 8-bit integral values.
///
/// \headerfile <x86intrin.h>
char __b07, char __b06, char __b05, char __b04,
char __b03, char __b02, char __b01, char __b00)
{
- return (__m256i)(__v32qi){
- __b31, __b30, __b29, __b28, __b27, __b26, __b25, __b24,
- __b23, __b22, __b21, __b20, __b19, __b18, __b17, __b16,
- __b15, __b14, __b13, __b12, __b11, __b10, __b09, __b08,
- __b07, __b06, __b05, __b04, __b03, __b02, __b01, __b00 };
+ return _mm256_set_epi8(__b00, __b01, __b02, __b03, __b04, __b05, __b06, __b07,
+ __b08, __b09, __b10, __b11, __b12, __b13, __b14, __b15,
+ __b16, __b17, __b18, __b19, __b20, __b21, __b22, __b23,
+ __b24, __b25, __b26, __b27, __b28, __b29, __b30, __b31);
}
-/// \brief Constructs a 256-bit integer vector, initialized in reverse order
+/// Constructs a 256-bit integer vector, initialized in reverse order
/// with the specified 64-bit integral values.
///
/// \headerfile <x86intrin.h>
static __inline __m256i __DEFAULT_FN_ATTRS
_mm256_setr_epi64x(long long __a, long long __b, long long __c, long long __d)
{
- return (__m256i)(__v4di){ __a, __b, __c, __d };
+ return _mm256_set_epi64x(__d, __c, __b, __a);
}
/* Create vectors with repeated elements */
-/// \brief Constructs a 256-bit floating-point vector of [4 x double], with each
+/// Constructs a 256-bit floating-point vector of [4 x double], with each
/// of the four double-precision floating-point vector elements set to the
/// specified double-precision floating-point value.
///
static __inline __m256d __DEFAULT_FN_ATTRS
_mm256_set1_pd(double __w)
{
- return (__m256d){ __w, __w, __w, __w };
+ return _mm256_set_pd(__w, __w, __w, __w);
}
-/// \brief Constructs a 256-bit floating-point vector of [8 x float], with each
+/// Constructs a 256-bit floating-point vector of [8 x float], with each
/// of the eight single-precision floating-point vector elements set to the
/// specified single-precision floating-point value.
///
static __inline __m256 __DEFAULT_FN_ATTRS
_mm256_set1_ps(float __w)
{
- return (__m256){ __w, __w, __w, __w, __w, __w, __w, __w };
+ return _mm256_set_ps(__w, __w, __w, __w, __w, __w, __w, __w);
}
-/// \brief Constructs a 256-bit integer vector of [8 x i32], with each of the
+/// Constructs a 256-bit integer vector of [8 x i32], with each of the
/// 32-bit integral vector elements set to the specified 32-bit integral
/// value.
///
static __inline __m256i __DEFAULT_FN_ATTRS
_mm256_set1_epi32(int __i)
{
- return (__m256i)(__v8si){ __i, __i, __i, __i, __i, __i, __i, __i };
+ return _mm256_set_epi32(__i, __i, __i, __i, __i, __i, __i, __i);
}
-/// \brief Constructs a 256-bit integer vector of [16 x i16], with each of the
+/// Constructs a 256-bit integer vector of [16 x i16], with each of the
/// 16-bit integral vector elements set to the specified 16-bit integral
/// value.
///
static __inline __m256i __DEFAULT_FN_ATTRS
_mm256_set1_epi16(short __w)
{
- return (__m256i)(__v16hi){ __w, __w, __w, __w, __w, __w, __w, __w, __w, __w,
- __w, __w, __w, __w, __w, __w };
+ return _mm256_set_epi16(__w, __w, __w, __w, __w, __w, __w, __w,
+ __w, __w, __w, __w, __w, __w, __w, __w);
}
-/// \brief Constructs a 256-bit integer vector of [32 x i8], with each of the
+/// Constructs a 256-bit integer vector of [32 x i8], with each of the
/// 8-bit integral vector elements set to the specified 8-bit integral value.
///
/// \headerfile <x86intrin.h>
static __inline __m256i __DEFAULT_FN_ATTRS
_mm256_set1_epi8(char __b)
{
- return (__m256i)(__v32qi){ __b, __b, __b, __b, __b, __b, __b, __b, __b, __b,
- __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b,
- __b, __b, __b, __b, __b, __b, __b };
+ return _mm256_set_epi8(__b, __b, __b, __b, __b, __b, __b, __b,
+ __b, __b, __b, __b, __b, __b, __b, __b,
+ __b, __b, __b, __b, __b, __b, __b, __b,
+ __b, __b, __b, __b, __b, __b, __b, __b);
}
-/// \brief Constructs a 256-bit integer vector of [4 x i64], with each of the
+/// Constructs a 256-bit integer vector of [4 x i64], with each of the
/// 64-bit integral vector elements set to the specified 64-bit integral
/// value.
///
static __inline __m256i __DEFAULT_FN_ATTRS
_mm256_set1_epi64x(long long __q)
{
- return (__m256i)(__v4di){ __q, __q, __q, __q };
+ return _mm256_set_epi64x(__q, __q, __q, __q);
}
/* Create __zeroed vectors */
-/// \brief Constructs a 256-bit floating-point vector of [4 x double] with all
+/// Constructs a 256-bit floating-point vector of [4 x double] with all
/// vector elements initialized to zero.
///
/// \headerfile <x86intrin.h>
static __inline __m256d __DEFAULT_FN_ATTRS
_mm256_setzero_pd(void)
{
- return (__m256d){ 0, 0, 0, 0 };
+ return __extension__ (__m256d){ 0, 0, 0, 0 };
}
-/// \brief Constructs a 256-bit floating-point vector of [8 x float] with all
+/// Constructs a 256-bit floating-point vector of [8 x float] with all
/// vector elements initialized to zero.
///
/// \headerfile <x86intrin.h>
static __inline __m256 __DEFAULT_FN_ATTRS
_mm256_setzero_ps(void)
{
- return (__m256){ 0, 0, 0, 0, 0, 0, 0, 0 };
+ return __extension__ (__m256){ 0, 0, 0, 0, 0, 0, 0, 0 };
}
-/// \brief Constructs a 256-bit integer vector initialized to zero.
+/// Constructs a 256-bit integer vector initialized to zero.
///
/// \headerfile <x86intrin.h>
///
static __inline __m256i __DEFAULT_FN_ATTRS
_mm256_setzero_si256(void)
{
- return (__m256i){ 0LL, 0LL, 0LL, 0LL };
+ return __extension__ (__m256i)(__v4di){ 0, 0, 0, 0 };
}
/* Cast between vector types */
-/// \brief Casts a 256-bit floating-point vector of [4 x double] into a 256-bit
+/// Casts a 256-bit floating-point vector of [4 x double] into a 256-bit
/// floating-point vector of [8 x float].
///
/// \headerfile <x86intrin.h>
return (__m256)__a;
}
-/// \brief Casts a 256-bit floating-point vector of [4 x double] into a 256-bit
+/// Casts a 256-bit floating-point vector of [4 x double] into a 256-bit
/// integer vector.
///
/// \headerfile <x86intrin.h>
return (__m256i)__a;
}
-/// \brief Casts a 256-bit floating-point vector of [8 x float] into a 256-bit
+/// Casts a 256-bit floating-point vector of [8 x float] into a 256-bit
/// floating-point vector of [4 x double].
///
/// \headerfile <x86intrin.h>
return (__m256d)__a;
}
-/// \brief Casts a 256-bit floating-point vector of [8 x float] into a 256-bit
+/// Casts a 256-bit floating-point vector of [8 x float] into a 256-bit
/// integer vector.
///
/// \headerfile <x86intrin.h>
return (__m256i)__a;
}
-/// \brief Casts a 256-bit integer vector into a 256-bit floating-point vector
+/// Casts a 256-bit integer vector into a 256-bit floating-point vector
/// of [8 x float].
///
/// \headerfile <x86intrin.h>
return (__m256)__a;
}
-/// \brief Casts a 256-bit integer vector into a 256-bit floating-point vector
+/// Casts a 256-bit integer vector into a 256-bit floating-point vector
/// of [4 x double].
///
/// \headerfile <x86intrin.h>
return (__m256d)__a;
}
-/// \brief Returns the lower 128 bits of a 256-bit floating-point vector of
+/// Returns the lower 128 bits of a 256-bit floating-point vector of
/// [4 x double] as a 128-bit floating-point vector of [2 x double].
///
/// \headerfile <x86intrin.h>
return __builtin_shufflevector((__v4df)__a, (__v4df)__a, 0, 1);
}
-/// \brief Returns the lower 128 bits of a 256-bit floating-point vector of
+/// Returns the lower 128 bits of a 256-bit floating-point vector of
/// [8 x float] as a 128-bit floating-point vector of [4 x float].
///
/// \headerfile <x86intrin.h>
return __builtin_shufflevector((__v8sf)__a, (__v8sf)__a, 0, 1, 2, 3);
}
-/// \brief Truncates a 256-bit integer vector into a 128-bit integer vector.
+/// Truncates a 256-bit integer vector into a 128-bit integer vector.
///
/// \headerfile <x86intrin.h>
///
return __builtin_shufflevector((__v4di)__a, (__v4di)__a, 0, 1);
}
-/// \brief Constructs a 256-bit floating-point vector of [4 x double] from a
+/// Constructs a 256-bit floating-point vector of [4 x double] from a
/// 128-bit floating-point vector of [2 x double].
///
/// The lower 128 bits contain the value of the source vector. The contents
return __builtin_shufflevector((__v2df)__a, (__v2df)__a, 0, 1, -1, -1);
}
-/// \brief Constructs a 256-bit floating-point vector of [8 x float] from a
+/// Constructs a 256-bit floating-point vector of [8 x float] from a
/// 128-bit floating-point vector of [4 x float].
///
/// The lower 128 bits contain the value of the source vector. The contents
return __builtin_shufflevector((__v4sf)__a, (__v4sf)__a, 0, 1, 2, 3, -1, -1, -1, -1);
}
-/// \brief Constructs a 256-bit integer vector from a 128-bit integer vector.
+/// Constructs a 256-bit integer vector from a 128-bit integer vector.
///
/// The lower 128 bits contain the value of the source vector. The contents
/// of the upper 128 bits are undefined.
return __builtin_shufflevector((__v2di)__a, (__v2di)__a, 0, 1, -1, -1);
}
-/// \brief Constructs a 256-bit floating-point vector of [4 x double] from a
+/// Constructs a 256-bit floating-point vector of [4 x double] from a
/// 128-bit floating-point vector of [2 x double]. The lower 128 bits
/// contain the value of the source vector. The upper 128 bits are set
/// to zero.
return __builtin_shufflevector((__v2df)__a, (__v2df)_mm_setzero_pd(), 0, 1, 2, 3);
}
-/// \brief Constructs a 256-bit floating-point vector of [8 x float] from a
+/// Constructs a 256-bit floating-point vector of [8 x float] from a
/// 128-bit floating-point vector of [4 x float]. The lower 128 bits contain
/// the value of the source vector. The upper 128 bits are set to zero.
///
return __builtin_shufflevector((__v4sf)__a, (__v4sf)_mm_setzero_ps(), 0, 1, 2, 3, 4, 5, 6, 7);
}
-/// \brief Constructs a 256-bit integer vector from a 128-bit integer vector.
+/// Constructs a 256-bit integer vector from a 128-bit integer vector.
/// The lower 128 bits contain the value of the source vector. The upper
/// 128 bits are set to zero.
///
We use macros rather than inlines because we only want to accept
invocations where the immediate M is a constant expression.
*/
-/// \brief Constructs a new 256-bit vector of [8 x float] by first duplicating
+/// Constructs a new 256-bit vector of [8 x float] by first duplicating
/// a 256-bit vector of [8 x float] given in the first parameter, and then
/// replacing either the upper or the lower 128 bits with the contents of a
/// 128-bit vector of [4 x float] in the second parameter.
/// result, and bits [127:0] of \a V1 are copied to bits [127:0] of the
/// result.
/// \returns A 256-bit vector of [8 x float] containing the interleaved values.
-#define _mm256_insertf128_ps(V1, V2, M) __extension__ ({ \
- (__m256)__builtin_shufflevector( \
- (__v8sf)(__m256)(V1), \
- (__v8sf)_mm256_castps128_ps256((__m128)(V2)), \
- (((M) & 1) ? 0 : 8), \
- (((M) & 1) ? 1 : 9), \
- (((M) & 1) ? 2 : 10), \
- (((M) & 1) ? 3 : 11), \
- (((M) & 1) ? 8 : 4), \
- (((M) & 1) ? 9 : 5), \
- (((M) & 1) ? 10 : 6), \
- (((M) & 1) ? 11 : 7) );})
-
-/// \brief Constructs a new 256-bit vector of [4 x double] by first duplicating
+#define _mm256_insertf128_ps(V1, V2, M) \
+ (__m256)__builtin_ia32_vinsertf128_ps256((__v8sf)(__m256)(V1), \
+ (__v4sf)(__m128)(V2), (int)(M))
+
+/// Constructs a new 256-bit vector of [4 x double] by first duplicating
/// a 256-bit vector of [4 x double] given in the first parameter, and then
/// replacing either the upper or the lower 128 bits with the contents of a
/// 128-bit vector of [2 x double] in the second parameter.
/// result, and bits [127:0] of \a V1 are copied to bits [127:0] of the
/// result.
/// \returns A 256-bit vector of [4 x double] containing the interleaved values.
-#define _mm256_insertf128_pd(V1, V2, M) __extension__ ({ \
- (__m256d)__builtin_shufflevector( \
- (__v4df)(__m256d)(V1), \
- (__v4df)_mm256_castpd128_pd256((__m128d)(V2)), \
- (((M) & 1) ? 0 : 4), \
- (((M) & 1) ? 1 : 5), \
- (((M) & 1) ? 4 : 2), \
- (((M) & 1) ? 5 : 3) );})
-
-/// \brief Constructs a new 256-bit integer vector by first duplicating a
+#define _mm256_insertf128_pd(V1, V2, M) \
+ (__m256d)__builtin_ia32_vinsertf128_pd256((__v4df)(__m256d)(V1), \
+ (__v2df)(__m128d)(V2), (int)(M))
+
+/// Constructs a new 256-bit integer vector by first duplicating a
/// 256-bit integer vector given in the first parameter, and then replacing
/// either the upper or the lower 128 bits with the contents of a 128-bit
/// integer vector in the second parameter.
/// result, and bits [127:0] of \a V1 are copied to bits [127:0] of the
/// result.
/// \returns A 256-bit integer vector containing the interleaved values.
-#define _mm256_insertf128_si256(V1, V2, M) __extension__ ({ \
- (__m256i)__builtin_shufflevector( \
- (__v4di)(__m256i)(V1), \
- (__v4di)_mm256_castsi128_si256((__m128i)(V2)), \
- (((M) & 1) ? 0 : 4), \
- (((M) & 1) ? 1 : 5), \
- (((M) & 1) ? 4 : 2), \
- (((M) & 1) ? 5 : 3) );})
+#define _mm256_insertf128_si256(V1, V2, M) \
+ (__m256i)__builtin_ia32_vinsertf128_si256((__v8si)(__m256i)(V1), \
+ (__v4si)(__m128i)(V2), (int)(M))
/*
Vector extract.
We use macros rather than inlines because we only want to accept
invocations where the immediate M is a constant expression.
*/
-/// \brief Extracts either the upper or the lower 128 bits from a 256-bit vector
+/// Extracts either the upper or the lower 128 bits from a 256-bit vector
/// of [8 x float], as determined by the immediate integer parameter, and
/// returns the extracted bits as a 128-bit vector of [4 x float].
///
/// result. \n
/// If bit [0] of \a M is 1, bits [255:128] of \a V are copied to the result.
/// \returns A 128-bit vector of [4 x float] containing the extracted bits.
-#define _mm256_extractf128_ps(V, M) __extension__ ({ \
- (__m128)__builtin_shufflevector( \
- (__v8sf)(__m256)(V), \
- (__v8sf)(_mm256_undefined_ps()), \
- (((M) & 1) ? 4 : 0), \
- (((M) & 1) ? 5 : 1), \
- (((M) & 1) ? 6 : 2), \
- (((M) & 1) ? 7 : 3) );})
-
-/// \brief Extracts either the upper or the lower 128 bits from a 256-bit vector
+#define _mm256_extractf128_ps(V, M) \
+ (__m128)__builtin_ia32_vextractf128_ps256((__v8sf)(__m256)(V), (int)(M))
+
+/// Extracts either the upper or the lower 128 bits from a 256-bit vector
/// of [4 x double], as determined by the immediate integer parameter, and
/// returns the extracted bits as a 128-bit vector of [2 x double].
///
/// result. \n
/// If bit [0] of \a M is 1, bits [255:128] of \a V are copied to the result.
/// \returns A 128-bit vector of [2 x double] containing the extracted bits.
-#define _mm256_extractf128_pd(V, M) __extension__ ({ \
- (__m128d)__builtin_shufflevector( \
- (__v4df)(__m256d)(V), \
- (__v4df)(_mm256_undefined_pd()), \
- (((M) & 1) ? 2 : 0), \
- (((M) & 1) ? 3 : 1) );})
-
-/// \brief Extracts either the upper or the lower 128 bits from a 256-bit
+#define _mm256_extractf128_pd(V, M) \
+ (__m128d)__builtin_ia32_vextractf128_pd256((__v4df)(__m256d)(V), (int)(M))
+
+/// Extracts either the upper or the lower 128 bits from a 256-bit
/// integer vector, as determined by the immediate integer parameter, and
/// returns the extracted bits as a 128-bit integer vector.
///
/// result. \n
/// If bit [0] of \a M is 1, bits [255:128] of \a V are copied to the result.
/// \returns A 128-bit integer vector containing the extracted bits.
-#define _mm256_extractf128_si256(V, M) __extension__ ({ \
- (__m128i)__builtin_shufflevector( \
- (__v4di)(__m256i)(V), \
- (__v4di)(_mm256_undefined_si256()), \
- (((M) & 1) ? 2 : 0), \
- (((M) & 1) ? 3 : 1) );})
+#define _mm256_extractf128_si256(V, M) \
+ (__m128i)__builtin_ia32_vextractf128_si256((__v8si)(__m256i)(V), (int)(M))
/* SIMD load ops (unaligned) */
-/// \brief Loads two 128-bit floating-point vectors of [4 x float] from
+/// Loads two 128-bit floating-point vectors of [4 x float] from
/// unaligned memory locations and constructs a 256-bit floating-point vector
/// of [8 x float] by concatenating the two 128-bit vectors.
///
return _mm256_insertf128_ps(__v256, _mm_loadu_ps(__addr_hi), 1);
}
-/// \brief Loads two 128-bit floating-point vectors of [2 x double] from
+/// Loads two 128-bit floating-point vectors of [2 x double] from
/// unaligned memory locations and constructs a 256-bit floating-point vector
/// of [4 x double] by concatenating the two 128-bit vectors.
///
return _mm256_insertf128_pd(__v256, _mm_loadu_pd(__addr_hi), 1);
}
-/// \brief Loads two 128-bit integer vectors from unaligned memory locations and
+/// Loads two 128-bit integer vectors from unaligned memory locations and
/// constructs a 256-bit integer vector by concatenating the two 128-bit
/// vectors.
///
/// address of the memory location does not have to be aligned.
/// \returns A 256-bit integer vector containing the concatenated result.
static __inline __m256i __DEFAULT_FN_ATTRS
-_mm256_loadu2_m128i(__m128i const *__addr_hi, __m128i const *__addr_lo)
+_mm256_loadu2_m128i(__m128i_u const *__addr_hi, __m128i_u const *__addr_lo)
{
__m256i __v256 = _mm256_castsi128_si256(_mm_loadu_si128(__addr_lo));
return _mm256_insertf128_si256(__v256, _mm_loadu_si128(__addr_hi), 1);
}
/* SIMD store ops (unaligned) */
-/// \brief Stores the upper and lower 128 bits of a 256-bit floating-point
+/// Stores the upper and lower 128 bits of a 256-bit floating-point
/// vector of [8 x float] into two different unaligned memory locations.
///
/// \headerfile <x86intrin.h>
_mm_storeu_ps(__addr_hi, __v128);
}
-/// \brief Stores the upper and lower 128 bits of a 256-bit floating-point
+/// Stores the upper and lower 128 bits of a 256-bit floating-point
/// vector of [4 x double] into two different unaligned memory locations.
///
/// \headerfile <x86intrin.h>
_mm_storeu_pd(__addr_hi, __v128);
}
-/// \brief Stores the upper and lower 128 bits of a 256-bit integer vector into
+/// Stores the upper and lower 128 bits of a 256-bit integer vector into
/// two different unaligned memory locations.
///
/// \headerfile <x86intrin.h>
/// \param __a
/// A 256-bit integer vector.
static __inline void __DEFAULT_FN_ATTRS
-_mm256_storeu2_m128i(__m128i *__addr_hi, __m128i *__addr_lo, __m256i __a)
+_mm256_storeu2_m128i(__m128i_u *__addr_hi, __m128i_u *__addr_lo, __m256i __a)
{
__m128i __v128;
_mm_storeu_si128(__addr_hi, __v128);
}
-/// \brief Constructs a 256-bit floating-point vector of [8 x float] by
+/// Constructs a 256-bit floating-point vector of [8 x float] by
/// concatenating two 128-bit floating-point vectors of [4 x float].
///
/// \headerfile <x86intrin.h>
return (__m256) __builtin_shufflevector((__v4sf)__lo, (__v4sf)__hi, 0, 1, 2, 3, 4, 5, 6, 7);
}
-/// \brief Constructs a 256-bit floating-point vector of [4 x double] by
+/// Constructs a 256-bit floating-point vector of [4 x double] by
/// concatenating two 128-bit floating-point vectors of [2 x double].
///
/// \headerfile <x86intrin.h>
static __inline __m256d __DEFAULT_FN_ATTRS
_mm256_set_m128d (__m128d __hi, __m128d __lo)
{
- return (__m256d)_mm256_set_m128((__m128)__hi, (__m128)__lo);
+ return (__m256d) __builtin_shufflevector((__v2df)__lo, (__v2df)__hi, 0, 1, 2, 3);
}
-/// \brief Constructs a 256-bit integer vector by concatenating two 128-bit
+/// Constructs a 256-bit integer vector by concatenating two 128-bit
/// integer vectors.
///
/// \headerfile <x86intrin.h>
static __inline __m256i __DEFAULT_FN_ATTRS
_mm256_set_m128i (__m128i __hi, __m128i __lo)
{
- return (__m256i)_mm256_set_m128((__m128)__hi, (__m128)__lo);
+ return (__m256i) __builtin_shufflevector((__v2di)__lo, (__v2di)__hi, 0, 1, 2, 3);
}
-/// \brief Constructs a 256-bit floating-point vector of [8 x float] by
+/// Constructs a 256-bit floating-point vector of [8 x float] by
/// concatenating two 128-bit floating-point vectors of [4 x float]. This is
/// similar to _mm256_set_m128, but the order of the input parameters is
/// swapped.
return _mm256_set_m128(__hi, __lo);
}
-/// \brief Constructs a 256-bit floating-point vector of [4 x double] by
+/// Constructs a 256-bit floating-point vector of [4 x double] by
/// concatenating two 128-bit floating-point vectors of [2 x double]. This is
/// similar to _mm256_set_m128d, but the order of the input parameters is
/// swapped.
static __inline __m256d __DEFAULT_FN_ATTRS
_mm256_setr_m128d (__m128d __lo, __m128d __hi)
{
- return (__m256d)_mm256_set_m128((__m128)__hi, (__m128)__lo);
+ return (__m256d)_mm256_set_m128d(__hi, __lo);
}
-/// \brief Constructs a 256-bit integer vector by concatenating two 128-bit
+/// Constructs a 256-bit integer vector by concatenating two 128-bit
/// integer vectors. This is similar to _mm256_set_m128i, but the order of
/// the input parameters is swapped.
///
static __inline __m256i __DEFAULT_FN_ATTRS
_mm256_setr_m128i (__m128i __lo, __m128i __hi)
{
- return (__m256i)_mm256_set_m128((__m128)__hi, (__m128)__lo);
+ return (__m256i)_mm256_set_m128i(__hi, __lo);
}
#undef __DEFAULT_FN_ATTRS
+#undef __DEFAULT_FN_ATTRS128
#endif /* __AVXINTRIN_H */
/*===---- bmi2intrin.h - BMI2 intrinsics -----------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
/*===---- bmiintrin.h - BMI intrinsics -------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#ifndef __BMIINTRIN_H
#define __BMIINTRIN_H
-#define _tzcnt_u16(a) (__tzcnt_u16((a)))
-
-#define _andn_u32(a, b) (__andn_u32((a), (b)))
-
-/* _bextr_u32 != __bextr_u32 */
-#define _blsi_u32(a) (__blsi_u32((a)))
-
-#define _blsmsk_u32(a) (__blsmsk_u32((a)))
-
-#define _blsr_u32(a) (__blsr_u32((a)))
-
-#define _tzcnt_u32(a) (__tzcnt_u32((a)))
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("bmi")))
-
/* Allow using the tzcnt intrinsics even for non-BMI targets. Since the TZCNT
instruction behaves as BSF on non-BMI targets, there is code that expects
to use it as a potentially faster version of BSF. */
#define __RELAXED_FN_ATTRS __attribute__((__always_inline__, __nodebug__))
-/// \brief Counts the number of trailing zero bits in the operand.
+#define _tzcnt_u16(a) (__tzcnt_u16((a)))
+
+/// Counts the number of trailing zero bits in the operand.
///
/// \headerfile <x86intrin.h>
///
static __inline__ unsigned short __RELAXED_FN_ATTRS
__tzcnt_u16(unsigned short __X)
{
- return __X ? __builtin_ctzs(__X) : 16;
+ return __builtin_ia32_tzcnt_u16(__X);
+}
+
+/// Counts the number of trailing zero bits in the operand.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> TZCNT </c> instruction.
+///
+/// \param __X
+/// An unsigned 32-bit integer whose trailing zeros are to be counted.
+/// \returns An unsigned 32-bit integer containing the number of trailing zero
+/// bits in the operand.
+static __inline__ unsigned int __RELAXED_FN_ATTRS
+__tzcnt_u32(unsigned int __X)
+{
+ return __builtin_ia32_tzcnt_u32(__X);
+}
+
+/// Counts the number of trailing zero bits in the operand.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> TZCNT </c> instruction.
+///
+/// \param __X
+/// An unsigned 32-bit integer whose trailing zeros are to be counted.
+/// \returns An 32-bit integer containing the number of trailing zero bits in
+/// the operand.
+static __inline__ int __RELAXED_FN_ATTRS
+_mm_tzcnt_32(unsigned int __X)
+{
+ return __builtin_ia32_tzcnt_u32(__X);
+}
+
+#define _tzcnt_u32(a) (__tzcnt_u32((a)))
+
+#ifdef __x86_64__
+
+/// Counts the number of trailing zero bits in the operand.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> TZCNT </c> instruction.
+///
+/// \param __X
+/// An unsigned 64-bit integer whose trailing zeros are to be counted.
+/// \returns An unsigned 64-bit integer containing the number of trailing zero
+/// bits in the operand.
+static __inline__ unsigned long long __RELAXED_FN_ATTRS
+__tzcnt_u64(unsigned long long __X)
+{
+ return __builtin_ia32_tzcnt_u64(__X);
}
-/// \brief Performs a bitwise AND of the second operand with the one's
+/// Counts the number of trailing zero bits in the operand.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> TZCNT </c> instruction.
+///
+/// \param __X
+/// An unsigned 64-bit integer whose trailing zeros are to be counted.
+/// \returns An 64-bit integer containing the number of trailing zero bits in
+/// the operand.
+static __inline__ long long __RELAXED_FN_ATTRS
+_mm_tzcnt_64(unsigned long long __X)
+{
+ return __builtin_ia32_tzcnt_u64(__X);
+}
+
+#define _tzcnt_u64(a) (__tzcnt_u64((a)))
+
+#endif /* __x86_64__ */
+
+#undef __RELAXED_FN_ATTRS
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__BMI__)
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("bmi")))
+
+#define _andn_u32(a, b) (__andn_u32((a), (b)))
+
+/* _bextr_u32 != __bextr_u32 */
+#define _blsi_u32(a) (__blsi_u32((a)))
+
+#define _blsmsk_u32(a) (__blsmsk_u32((a)))
+
+#define _blsr_u32(a) (__blsr_u32((a)))
+
+/// Performs a bitwise AND of the second operand with the one's
/// complement of the first operand.
///
/// \headerfile <x86intrin.h>
}
/* AMD-specified, double-leading-underscore version of BEXTR */
-/// \brief Extracts the specified bits from the first operand and returns them
+/// Extracts the specified bits from the first operand and returns them
/// in the least significant bits of the result.
///
/// \headerfile <x86intrin.h>
/// number of bits to be extracted.
/// \returns An unsigned integer whose least significant bits contain the
/// extracted bits.
+/// \see _bextr_u32
static __inline__ unsigned int __DEFAULT_FN_ATTRS
__bextr_u32(unsigned int __X, unsigned int __Y)
{
}
/* Intel-specified, single-leading-underscore version of BEXTR */
-/// \brief Extracts the specified bits from the first operand and returns them
+/// Extracts the specified bits from the first operand and returns them
/// in the least significant bits of the result.
///
/// \headerfile <x86intrin.h>
/// Bits [7:0] specify the number of bits.
/// \returns An unsigned integer whose least significant bits contain the
/// extracted bits.
+/// \see __bextr_u32
static __inline__ unsigned int __DEFAULT_FN_ATTRS
_bextr_u32(unsigned int __X, unsigned int __Y, unsigned int __Z)
{
return __builtin_ia32_bextr_u32 (__X, ((__Y & 0xff) | ((__Z & 0xff) << 8)));
}
-/// \brief Clears all bits in the source except for the least significant bit
+/* Intel-specified, single-leading-underscore version of BEXTR2 */
+/// Extracts the specified bits from the first operand and returns them
+/// in the least significant bits of the result.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> BEXTR </c> instruction.
+///
+/// \param __X
+/// An unsigned integer whose bits are to be extracted.
+/// \param __Y
+/// An unsigned integer used to specify which bits are extracted. Bits [7:0]
+/// specify the index of the least significant bit. Bits [15:8] specify the
+/// number of bits to be extracted.
+/// \returns An unsigned integer whose least significant bits contain the
+/// extracted bits.
+/// \see __bextr_u32
+static __inline__ unsigned int __DEFAULT_FN_ATTRS
+_bextr2_u32(unsigned int __X, unsigned int __Y) {
+ return __builtin_ia32_bextr_u32(__X, __Y);
+}
+
+/// Clears all bits in the source except for the least significant bit
/// containing a value of 1 and returns the result.
///
/// \headerfile <x86intrin.h>
return __X & -__X;
}
-/// \brief Creates a mask whose bits are set to 1, using bit 0 up to and
+/// Creates a mask whose bits are set to 1, using bit 0 up to and
/// including the least significant bit that is set to 1 in the source
/// operand and returns the result.
///
return __X ^ (__X - 1);
}
-/// \brief Clears the least significant bit that is set to 1 in the source
+/// Clears the least significant bit that is set to 1 in the source
/// operand and returns the result.
///
/// \headerfile <x86intrin.h>
return __X & (__X - 1);
}
-/// \brief Counts the number of trailing zero bits in the operand.
-///
-/// \headerfile <x86intrin.h>
-///
-/// This intrinsic corresponds to the <c> TZCNT </c> instruction.
-///
-/// \param __X
-/// An unsigned 32-bit integer whose trailing zeros are to be counted.
-/// \returns An unsigned 32-bit integer containing the number of trailing zero
-/// bits in the operand.
-static __inline__ unsigned int __RELAXED_FN_ATTRS
-__tzcnt_u32(unsigned int __X)
-{
- return __X ? __builtin_ctz(__X) : 32;
-}
-
-/// \brief Counts the number of trailing zero bits in the operand.
-///
-/// \headerfile <x86intrin.h>
-///
-/// This intrinsic corresponds to the <c> TZCNT </c> instruction.
-///
-/// \param __X
-/// An unsigned 32-bit integer whose trailing zeros are to be counted.
-/// \returns An 32-bit integer containing the number of trailing zero bits in
-/// the operand.
-static __inline__ int __RELAXED_FN_ATTRS
-_mm_tzcnt_32(unsigned int __X)
-{
- return __X ? __builtin_ctz(__X) : 32;
-}
-
#ifdef __x86_64__
#define _andn_u64(a, b) (__andn_u64((a), (b)))
#define _blsr_u64(a) (__blsr_u64((a)))
-#define _tzcnt_u64(a) (__tzcnt_u64((a)))
-
-/// \brief Performs a bitwise AND of the second operand with the one's
+/// Performs a bitwise AND of the second operand with the one's
/// complement of the first operand.
///
/// \headerfile <x86intrin.h>
}
/* AMD-specified, double-leading-underscore version of BEXTR */
-/// \brief Extracts the specified bits from the first operand and returns them
+/// Extracts the specified bits from the first operand and returns them
/// in the least significant bits of the result.
///
/// \headerfile <x86intrin.h>
/// the number of bits to be extracted.
/// \returns An unsigned 64-bit integer whose least significant bits contain the
/// extracted bits.
+/// \see _bextr_u64
static __inline__ unsigned long long __DEFAULT_FN_ATTRS
__bextr_u64(unsigned long long __X, unsigned long long __Y)
{
}
/* Intel-specified, single-leading-underscore version of BEXTR */
-/// \brief Extracts the specified bits from the first operand and returns them
+/// Extracts the specified bits from the first operand and returns them
/// in the least significant bits of the result.
///
/// \headerfile <x86intrin.h>
/// Bits [7:0] specify the number of bits.
/// \returns An unsigned 64-bit integer whose least significant bits contain the
/// extracted bits.
+/// \see __bextr_u64
static __inline__ unsigned long long __DEFAULT_FN_ATTRS
_bextr_u64(unsigned long long __X, unsigned int __Y, unsigned int __Z)
{
return __builtin_ia32_bextr_u64 (__X, ((__Y & 0xff) | ((__Z & 0xff) << 8)));
}
-/// \brief Clears all bits in the source except for the least significant bit
+/* Intel-specified, single-leading-underscore version of BEXTR2 */
+/// Extracts the specified bits from the first operand and returns them
+/// in the least significant bits of the result.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> BEXTR </c> instruction.
+///
+/// \param __X
+/// An unsigned 64-bit integer whose bits are to be extracted.
+/// \param __Y
+/// An unsigned 64-bit integer used to specify which bits are extracted. Bits
+/// [7:0] specify the index of the least significant bit. Bits [15:8] specify
+/// the number of bits to be extracted.
+/// \returns An unsigned 64-bit integer whose least significant bits contain the
+/// extracted bits.
+/// \see __bextr_u64
+static __inline__ unsigned long long __DEFAULT_FN_ATTRS
+_bextr2_u64(unsigned long long __X, unsigned long long __Y) {
+ return __builtin_ia32_bextr_u64(__X, __Y);
+}
+
+/// Clears all bits in the source except for the least significant bit
/// containing a value of 1 and returns the result.
///
/// \headerfile <x86intrin.h>
return __X & -__X;
}
-/// \brief Creates a mask whose bits are set to 1, using bit 0 up to and
+/// Creates a mask whose bits are set to 1, using bit 0 up to and
/// including the least significant bit that is set to 1 in the source
/// operand and returns the result.
///
return __X ^ (__X - 1);
}
-/// \brief Clears the least significant bit that is set to 1 in the source
+/// Clears the least significant bit that is set to 1 in the source
/// operand and returns the result.
///
/// \headerfile <x86intrin.h>
return __X & (__X - 1);
}
-/// \brief Counts the number of trailing zero bits in the operand.
-///
-/// \headerfile <x86intrin.h>
-///
-/// This intrinsic corresponds to the <c> TZCNT </c> instruction.
-///
-/// \param __X
-/// An unsigned 64-bit integer whose trailing zeros are to be counted.
-/// \returns An unsigned 64-bit integer containing the number of trailing zero
-/// bits in the operand.
-static __inline__ unsigned long long __RELAXED_FN_ATTRS
-__tzcnt_u64(unsigned long long __X)
-{
- return __X ? __builtin_ctzll(__X) : 64;
-}
-
-/// \brief Counts the number of trailing zero bits in the operand.
-///
-/// \headerfile <x86intrin.h>
-///
-/// This intrinsic corresponds to the <c> TZCNT </c> instruction.
-///
-/// \param __X
-/// An unsigned 64-bit integer whose trailing zeros are to be counted.
-/// \returns An 64-bit integer containing the number of trailing zero bits in
-/// the operand.
-static __inline__ long long __RELAXED_FN_ATTRS
-_mm_tzcnt_64(unsigned long long __X)
-{
- return __X ? __builtin_ctzll(__X) : 64;
-}
-
#endif /* __x86_64__ */
#undef __DEFAULT_FN_ATTRS
-#undef __RELAXED_FN_ATTRS
+
+#endif /* !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) \
+ || defined(__BMI__) */
#endif /* __BMIINTRIN_H */
--- /dev/null
+/*===------ cet.h -Control-flow Enforcement Technology feature ------------===
+ * Add x86 feature with IBT and/or SHSTK bits to ELF program property if they
+ * are enabled. Otherwise, contents in this header file are unused. This file
+ * is mainly design for assembly source code which want to enable CET.
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __CET_H
+#define __CET_H
+
+#ifdef __ASSEMBLER__
+
+#ifndef __CET__
+# define _CET_ENDBR
+#endif
+
+#ifdef __CET__
+
+# ifdef __LP64__
+# if __CET__ & 0x1
+# define _CET_ENDBR endbr64
+# else
+# define _CET_ENDBR
+# endif
+# else
+# if __CET__ & 0x1
+# define _CET_ENDBR endbr32
+# else
+# define _CET_ENDBR
+# endif
+# endif
+
+
+# ifdef __LP64__
+# define __PROPERTY_ALIGN 3
+# else
+# define __PROPERTY_ALIGN 2
+# endif
+
+ .pushsection ".note.gnu.property", "a"
+ .p2align __PROPERTY_ALIGN
+ .long 1f - 0f /* name length. */
+ .long 4f - 1f /* data length. */
+ /* NT_GNU_PROPERTY_TYPE_0. */
+ .long 5 /* note type. */
+0:
+ .asciz "GNU" /* vendor name. */
+1:
+ .p2align __PROPERTY_ALIGN
+ /* GNU_PROPERTY_X86_FEATURE_1_AND. */
+ .long 0xc0000002 /* pr_type. */
+ .long 3f - 2f /* pr_datasz. */
+2:
+ /* GNU_PROPERTY_X86_FEATURE_1_XXX. */
+ .long __CET__
+3:
+ .p2align __PROPERTY_ALIGN
+4:
+ .popsection
+#endif
+#endif
+#endif
--- /dev/null
+/*===---- cetintrin.h - CET intrinsic --------------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <cetintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __CETINTRIN_H
+#define __CETINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("shstk")))
+
+static __inline__ void __DEFAULT_FN_ATTRS _incsspd(int __a) {
+ __builtin_ia32_incsspd(__a);
+}
+
+#ifdef __x86_64__
+static __inline__ void __DEFAULT_FN_ATTRS _incsspq(unsigned long long __a) {
+ __builtin_ia32_incsspq(__a);
+}
+#endif /* __x86_64__ */
+
+#ifdef __x86_64__
+static __inline__ void __DEFAULT_FN_ATTRS _inc_ssp(unsigned int __a) {
+ __builtin_ia32_incsspq(__a);
+}
+#else /* __x86_64__ */
+static __inline__ void __DEFAULT_FN_ATTRS _inc_ssp(unsigned int __a) {
+ __builtin_ia32_incsspd((int)__a);
+}
+#endif /* __x86_64__ */
+
+static __inline__ unsigned int __DEFAULT_FN_ATTRS _rdsspd(unsigned int __a) {
+ return __builtin_ia32_rdsspd(__a);
+}
+
+#ifdef __x86_64__
+static __inline__ unsigned long long __DEFAULT_FN_ATTRS _rdsspq(unsigned long long __a) {
+ return __builtin_ia32_rdsspq(__a);
+}
+#endif /* __x86_64__ */
+
+#ifdef __x86_64__
+static __inline__ unsigned long long __DEFAULT_FN_ATTRS _get_ssp(void) {
+ return __builtin_ia32_rdsspq(0);
+}
+#else /* __x86_64__ */
+static __inline__ unsigned int __DEFAULT_FN_ATTRS _get_ssp(void) {
+ return __builtin_ia32_rdsspd(0);
+}
+#endif /* __x86_64__ */
+
+static __inline__ void __DEFAULT_FN_ATTRS _saveprevssp() {
+ __builtin_ia32_saveprevssp();
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS _rstorssp(void * __p) {
+ __builtin_ia32_rstorssp(__p);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS _wrssd(unsigned int __a, void * __p) {
+ __builtin_ia32_wrssd(__a, __p);
+}
+
+#ifdef __x86_64__
+static __inline__ void __DEFAULT_FN_ATTRS _wrssq(unsigned long long __a, void * __p) {
+ __builtin_ia32_wrssq(__a, __p);
+}
+#endif /* __x86_64__ */
+
+static __inline__ void __DEFAULT_FN_ATTRS _wrussd(unsigned int __a, void * __p) {
+ __builtin_ia32_wrussd(__a, __p);
+}
+
+#ifdef __x86_64__
+static __inline__ void __DEFAULT_FN_ATTRS _wrussq(unsigned long long __a, void * __p) {
+ __builtin_ia32_wrussq(__a, __p);
+}
+#endif /* __x86_64__ */
+
+static __inline__ void __DEFAULT_FN_ATTRS _setssbsy() {
+ __builtin_ia32_setssbsy();
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS _clrssbsy(void * __p) {
+ __builtin_ia32_clrssbsy(__p);
+}
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif /* __CETINTRIN_H */
--- /dev/null
+/*===---- cldemoteintrin.h - CLDEMOTE intrinsic ----------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#if !defined __X86INTRIN_H && !defined __IMMINTRIN_H
+#error "Never use <cldemoteintrin.h> directly; include <x86intrin.h> instead."
+#endif
+
+#ifndef __CLDEMOTEINTRIN_H
+#define __CLDEMOTEINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("cldemote")))
+
+/// Hint to hardware that the cache line that contains \p __P should be demoted
+/// from the cache closest to the processor core to a level more distant from
+/// the processor core.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> CLDEMOTE </c> instruction.
+static __inline__ void __DEFAULT_FN_ATTRS
+_cldemote(const void * __P) {
+ __builtin_ia32_cldemote(__P);
+}
+
+#define _mm_cldemote(p) _cldemote(p)
+#undef __DEFAULT_FN_ATTRS
+
+#endif
--- /dev/null
+/*===---- clflushoptintrin.h - CLFLUSHOPT intrinsic ------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <clflushoptintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __CLFLUSHOPTINTRIN_H
+#define __CLFLUSHOPTINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("clflushopt")))
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_mm_clflushopt(void const * __m) {
+ __builtin_ia32_clflushopt(__m);
+}
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif
--- /dev/null
+/*===---- clwbintrin.h - CLWB intrinsic ------------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <clwbintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __CLWBINTRIN_H
+#define __CLWBINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("clwb")))
+
+/// Writes back to memory the cache line (if modified) that contains the
+/// linear address specified in \a __p from any level of the cache hierarchy in
+/// the cache coherence domain
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the <c> CLWB </c> instruction.
+///
+/// \param __p
+/// A pointer to the memory location used to identify the cache line to be
+/// written back.
+static __inline__ void __DEFAULT_FN_ATTRS
+_mm_clwb(void const *__p) {
+ __builtin_ia32_clwb(__p);
+}
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif
--- /dev/null
+/*===----------------------- clzerointrin.h - CLZERO ----------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#if !defined __X86INTRIN_H && !defined __IMMINTRIN_H
+#error "Never use <clzerointrin.h> directly; include <x86intrin.h> instead."
+#endif
+
+#ifndef __CLZEROINTRIN_H
+#define __CLZEROINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("clzero")))
+
+/// Loads the cache line address and zero's out the cacheline
+///
+/// \headerfile <clzerointrin.h>
+///
+/// This intrinsic corresponds to the <c> CLZERO </c> instruction.
+///
+/// \param __line
+/// A pointer to a cacheline which needs to be zeroed out.
+static __inline__ void __DEFAULT_FN_ATTRS
+_mm_clzero (void * __line)
+{
+ __builtin_ia32_clzero ((void *)__line);
+}
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif /* __CLZEROINTRIN_H */
/*===---- cpuid.h - X86 cpu model detection --------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#define signature_CYRIX_ebx 0x69727943
#define signature_CYRIX_edx 0x736e4978
#define signature_CYRIX_ecx 0x64616574
+/* HYGON: "HygonGenuine" */
+#define signature_HYGON_ebx 0x6f677948
+#define signature_HYGON_edx 0x6e65476e
+#define signature_HYGON_ecx 0x656e6975
/* INTEL: "GenuineIntel" */
#define signature_INTEL_ebx 0x756e6547
#define signature_INTEL_edx 0x49656e69
#define signature_TM2_ecx 0x3638784d
/* NSC: "Geode by NSC" */
#define signature_NSC_ebx 0x646f6547
-#define signature_NSC_edx 0x43534e20
-#define signature_NSC_ecx 0x79622065
+#define signature_NSC_edx 0x79622065
+#define signature_NSC_ecx 0x43534e20
/* NEXGEN: "NexGenDriven" */
#define signature_NEXGEN_ebx 0x4778654e
#define signature_NEXGEN_edx 0x72446e65
#define bit_SMEP 0x00000080
#define bit_BMI2 0x00000100
#define bit_ENH_MOVSB 0x00000200
+#define bit_INVPCID 0x00000400
#define bit_RTM 0x00000800
#define bit_MPX 0x00004000
#define bit_AVX512F 0x00010000
#define bit_CLFLUSHOPT 0x00800000
#define bit_CLWB 0x01000000
#define bit_AVX512PF 0x04000000
-#define bit_AVX51SER 0x08000000
+#define bit_AVX512ER 0x08000000
#define bit_AVX512CD 0x10000000
#define bit_SHA 0x20000000
#define bit_AVX512BW 0x40000000
#define bit_AVX512VL 0x80000000
/* Features in %ecx for leaf 7 sub-leaf 0 */
-#define bit_PREFTCHWT1 0x00000001
-#define bit_AVX512VBMI 0x00000002
-#define bit_PKU 0x00000004
-#define bit_OSPKE 0x00000010
+#define bit_PREFTCHWT1 0x00000001
+#define bit_AVX512VBMI 0x00000002
+#define bit_PKU 0x00000004
+#define bit_OSPKE 0x00000010
+#define bit_WAITPKG 0x00000020
+#define bit_AVX512VBMI2 0x00000040
+#define bit_SHSTK 0x00000080
+#define bit_GFNI 0x00000100
+#define bit_VAES 0x00000200
+#define bit_VPCLMULQDQ 0x00000400
+#define bit_AVX512VNNI 0x00000800
+#define bit_AVX512BITALG 0x00001000
#define bit_AVX512VPOPCNTDQ 0x00004000
-#define bit_RDPID 0x00400000
+#define bit_RDPID 0x00400000
+#define bit_CLDEMOTE 0x02000000
+#define bit_MOVDIRI 0x08000000
+#define bit_MOVDIR64B 0x10000000
+#define bit_ENQCMD 0x20000000
/* Features in %edx for leaf 7 sub-leaf 0 */
#define bit_AVX5124VNNIW 0x00000004
#define bit_AVX5124FMAPS 0x00000008
+#define bit_SERIALIZE 0x00004000
+#define bit_TSXLDTRK 0x00010000
+#define bit_PCONFIG 0x00040000
+#define bit_IBT 0x00100000
+#define bit_AMXBF16 0x00400000
+#define bit_AMXTILE 0x01000000
+#define bit_AMXINT8 0x02000000
+
+/* Features in %eax for leaf 7 sub-leaf 1 */
+#define bit_AVX512BF16 0x00000020
/* Features in %eax for leaf 13 sub-leaf 1 */
#define bit_XSAVEOPT 0x00000001
#define bit_XSAVEC 0x00000002
#define bit_XSAVES 0x00000008
+/* Features in %eax for leaf 0x14 sub-leaf 0 */
+#define bit_PTWRITE 0x00000010
+
/* Features in %ecx for leaf 0x80000001 */
#define bit_LAHF_LM 0x00000001
#define bit_ABM 0x00000020
+#define bit_LZCNT bit_ABM /* for gcc compat */
#define bit_SSE4a 0x00000040
#define bit_PRFCHW 0x00000100
#define bit_XOP 0x00000800
#define bit_3DNOWP 0x40000000
#define bit_3DNOW 0x80000000
-/* Features in %ebx for leaf 0x80000001 */
+/* Features in %ebx for leaf 0x80000008 */
#define bit_CLZERO 0x00000001
+#define bit_WBNOINVD 0x00000200
#if __i386__
#ifndef __CLANG_CUDA_WRAPPERS_ALGORITHM
#define __CLANG_CUDA_WRAPPERS_ALGORITHM
-// This header defines __device__ overloads of std::min/max, but only if we're
-// <= C++11. In C++14, these functions are constexpr, and so are implicitly
-// __host__ __device__.
+// This header defines __device__ overloads of std::min/max.
//
-// We don't support the initializer_list overloads because
-// initializer_list::begin() and end() are not __host__ __device__ functions.
+// Ideally we'd declare these functions only if we're <= C++11. In C++14,
+// these functions are constexpr, and so are implicitly __host__ __device__.
//
-// When compiling in C++14 mode, we could force std::min/max to have different
-// implementations for host and device, by declaring the device overloads
-// before the constexpr overloads appear. We choose not to do this because
-
-// a) why write our own implementation when we can use one from the standard
-// library? and
-// b) libstdc++ is evil and declares min/max inside a header that is included
-// *before* we include <algorithm>. So we'd have to unconditionally
-// declare our __device__ overloads of min/max, but that would pollute
-// things for people who choose not to include <algorithm>.
+// However, the compiler being in C++14 mode does not imply that the standard
+// library supports C++14. There is no macro we can test to check that the
+// stdlib has constexpr std::min/max. Thus we have to unconditionally define
+// our device overloads.
+//
+// A host+device function cannot be overloaded, and a constexpr function
+// implicitly become host device if there's no explicitly host or device
+// overload preceding it. So the simple thing to do would be to declare our
+// device min/max overloads, and then #include_next <algorithm>. This way our
+// device overloads would come first, and so if we have a C++14 stdlib, its
+// min/max won't become host+device and conflict with our device overloads.
+//
+// But that also doesn't work. libstdc++ is evil and declares std::min/max in
+// an internal header that is included *before* <algorithm>. Thus by the time
+// we're inside of this file, std::min/max may already have been declared, and
+// thus we can't prevent them from becoming host+device if they're constexpr.
+//
+// Therefore we perpetrate the following hack: We mark our __device__ overloads
+// with __attribute__((enable_if(true, ""))). This causes the signature of the
+// function to change without changing anything else about it. (Except that
+// overload resolution will prefer it over the __host__ __device__ version
+// rather than considering them equally good).
#include_next <algorithm>
-#if __cplusplus <= 201103L
-
// We need to define these overloads in exactly the namespace our standard
// library uses (including the right inline namespace), otherwise they won't be
// picked up by other functions in the standard library (e.g. functions in
#endif
#endif
+#pragma push_macro("_CPP14_CONSTEXPR")
+#if __cplusplus >= 201402L
+#define _CPP14_CONSTEXPR constexpr
+#else
+#define _CPP14_CONSTEXPR
+#endif
+
template <class __T, class __Cmp>
-inline __device__ const __T &
+__attribute__((enable_if(true, "")))
+inline _CPP14_CONSTEXPR __host__ __device__ const __T &
max(const __T &__a, const __T &__b, __Cmp __cmp) {
return __cmp(__a, __b) ? __b : __a;
}
template <class __T>
-inline __device__ const __T &
+__attribute__((enable_if(true, "")))
+inline _CPP14_CONSTEXPR __host__ __device__ const __T &
max(const __T &__a, const __T &__b) {
return __a < __b ? __b : __a;
}
template <class __T, class __Cmp>
-inline __device__ const __T &
+__attribute__((enable_if(true, "")))
+inline _CPP14_CONSTEXPR __host__ __device__ const __T &
min(const __T &__a, const __T &__b, __Cmp __cmp) {
return __cmp(__b, __a) ? __b : __a;
}
template <class __T>
-inline __device__ const __T &
+__attribute__((enable_if(true, "")))
+inline _CPP14_CONSTEXPR __host__ __device__ const __T &
min(const __T &__a, const __T &__b) {
- return __a < __b ? __b : __a;
+ return __a < __b ? __a : __b;
}
+#pragma pop_macro("_CPP14_CONSTEXPR")
+
#ifdef _LIBCPP_END_NAMESPACE_STD
_LIBCPP_END_NAMESPACE_STD
#else
} // namespace std
#endif
-#endif // __cplusplus <= 201103L
#endif // __CLANG_CUDA_WRAPPERS_ALGORITHM
#include_next <new>
-// Device overrides for placement new and delete.
#pragma push_macro("CUDA_NOEXCEPT")
#if __cplusplus >= 201103L
#define CUDA_NOEXCEPT noexcept
#define CUDA_NOEXCEPT
#endif
+// Device overrides for non-placement new and delete.
+__device__ inline void *operator new(__SIZE_TYPE__ size) {
+ if (size == 0) {
+ size = 1;
+ }
+ return ::malloc(size);
+}
+__device__ inline void *operator new(__SIZE_TYPE__ size,
+ const std::nothrow_t &) CUDA_NOEXCEPT {
+ return ::operator new(size);
+}
+
+__device__ inline void *operator new[](__SIZE_TYPE__ size) {
+ return ::operator new(size);
+}
+__device__ inline void *operator new[](__SIZE_TYPE__ size,
+ const std::nothrow_t &) {
+ return ::operator new(size);
+}
+
+__device__ inline void operator delete(void* ptr) CUDA_NOEXCEPT {
+ if (ptr) {
+ ::free(ptr);
+ }
+}
+__device__ inline void operator delete(void *ptr,
+ const std::nothrow_t &) CUDA_NOEXCEPT {
+ ::operator delete(ptr);
+}
+
+__device__ inline void operator delete[](void* ptr) CUDA_NOEXCEPT {
+ ::operator delete(ptr);
+}
+__device__ inline void operator delete[](void *ptr,
+ const std::nothrow_t &) CUDA_NOEXCEPT {
+ ::operator delete(ptr);
+}
+
+// Sized delete, C++14 only.
+#if __cplusplus >= 201402L
+__device__ inline void operator delete(void *ptr,
+ __SIZE_TYPE__ size) CUDA_NOEXCEPT {
+ ::operator delete(ptr);
+}
+__device__ inline void operator delete[](void *ptr,
+ __SIZE_TYPE__ size) CUDA_NOEXCEPT {
+ ::operator delete(ptr);
+}
+#endif
+
+// Device overrides for placement new and delete.
__device__ inline void *operator new(__SIZE_TYPE__, void *__ptr) CUDA_NOEXCEPT {
return __ptr;
}
}
__device__ inline void operator delete(void *, void *) CUDA_NOEXCEPT {}
__device__ inline void operator delete[](void *, void *) CUDA_NOEXCEPT {}
+
#pragma pop_macro("CUDA_NOEXCEPT")
#endif // include guard
/*===---- emmintrin.h - SSE2 intrinsics ------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#include <xmmintrin.h>
-typedef double __m128d __attribute__((__vector_size__(16)));
-typedef long long __m128i __attribute__((__vector_size__(16)));
+typedef double __m128d __attribute__((__vector_size__(16), __aligned__(16)));
+typedef long long __m128i __attribute__((__vector_size__(16), __aligned__(16)));
+
+typedef double __m128d_u __attribute__((__vector_size__(16), __aligned__(1)));
+typedef long long __m128i_u __attribute__((__vector_size__(16), __aligned__(1)));
/* Type defines. */
typedef double __v2df __attribute__ ((__vector_size__ (16)));
* appear in the interface though. */
typedef signed char __v16qs __attribute__((__vector_size__(16)));
-#include <f16cintrin.h>
-
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("sse2")))
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("sse2"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS_MMX __attribute__((__always_inline__, __nodebug__, __target__("mmx,sse2"), __min_vector_width__(64)))
-/// \brief Adds lower double-precision values in both operands and returns the
+/// Adds lower double-precision values in both operands and returns the
/// sum in the lower 64 bits of the result. The upper 64 bits of the result
/// are copied from the upper double-precision value of the first operand.
///
return __a;
}
-/// \brief Adds two 128-bit vectors of [2 x double].
+/// Adds two 128-bit vectors of [2 x double].
///
/// \headerfile <x86intrin.h>
///
return (__m128d)((__v2df)__a + (__v2df)__b);
}
-/// \brief Subtracts the lower double-precision value of the second operand
+/// Subtracts the lower double-precision value of the second operand
/// from the lower double-precision value of the first operand and returns
/// the difference in the lower 64 bits of the result. The upper 64 bits of
/// the result are copied from the upper double-precision value of the first
return __a;
}
-/// \brief Subtracts two 128-bit vectors of [2 x double].
+/// Subtracts two 128-bit vectors of [2 x double].
///
/// \headerfile <x86intrin.h>
///
return (__m128d)((__v2df)__a - (__v2df)__b);
}
-/// \brief Multiplies lower double-precision values in both operands and returns
+/// Multiplies lower double-precision values in both operands and returns
/// the product in the lower 64 bits of the result. The upper 64 bits of the
/// result are copied from the upper double-precision value of the first
/// operand.
return __a;
}
-/// \brief Multiplies two 128-bit vectors of [2 x double].
+/// Multiplies two 128-bit vectors of [2 x double].
///
/// \headerfile <x86intrin.h>
///
return (__m128d)((__v2df)__a * (__v2df)__b);
}
-/// \brief Divides the lower double-precision value of the first operand by the
+/// Divides the lower double-precision value of the first operand by the
/// lower double-precision value of the second operand and returns the
/// quotient in the lower 64 bits of the result. The upper 64 bits of the
/// result are copied from the upper double-precision value of the first
return __a;
}
-/// \brief Performs an element-by-element division of two 128-bit vectors of
+/// Performs an element-by-element division of two 128-bit vectors of
/// [2 x double].
///
/// \headerfile <x86intrin.h>
return (__m128d)((__v2df)__a / (__v2df)__b);
}
-/// \brief Calculates the square root of the lower double-precision value of
+/// Calculates the square root of the lower double-precision value of
/// the second operand and returns it in the lower 64 bits of the result.
-/// The upper 64 bits of the result are copied from the upper double-
-/// precision value of the first operand.
+/// The upper 64 bits of the result are copied from the upper
+/// double-precision value of the first operand.
///
/// \headerfile <x86intrin.h>
///
_mm_sqrt_sd(__m128d __a, __m128d __b)
{
__m128d __c = __builtin_ia32_sqrtsd((__v2df)__b);
- return (__m128d) { __c[0], __a[1] };
+ return __extension__ (__m128d) { __c[0], __a[1] };
}
-/// \brief Calculates the square root of the each of two values stored in a
+/// Calculates the square root of the each of two values stored in a
/// 128-bit vector of [2 x double].
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_sqrtpd((__v2df)__a);
}
-/// \brief Compares lower 64-bit double-precision values of both operands, and
+/// Compares lower 64-bit double-precision values of both operands, and
/// returns the lesser of the pair of values in the lower 64-bits of the
-/// result. The upper 64 bits of the result are copied from the upper double-
-/// precision value of the first operand.
+/// result. The upper 64 bits of the result are copied from the upper
+/// double-precision value of the first operand.
///
/// \headerfile <x86intrin.h>
///
return __builtin_ia32_minsd((__v2df)__a, (__v2df)__b);
}
-/// \brief Performs element-by-element comparison of the two 128-bit vectors of
+/// Performs element-by-element comparison of the two 128-bit vectors of
/// [2 x double] and returns the vector containing the lesser of each pair of
/// values.
///
return __builtin_ia32_minpd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares lower 64-bit double-precision values of both operands, and
+/// Compares lower 64-bit double-precision values of both operands, and
/// returns the greater of the pair of values in the lower 64-bits of the
-/// result. The upper 64 bits of the result are copied from the upper double-
-/// precision value of the first operand.
+/// result. The upper 64 bits of the result are copied from the upper
+/// double-precision value of the first operand.
///
/// \headerfile <x86intrin.h>
///
return __builtin_ia32_maxsd((__v2df)__a, (__v2df)__b);
}
-/// \brief Performs element-by-element comparison of the two 128-bit vectors of
+/// Performs element-by-element comparison of the two 128-bit vectors of
/// [2 x double] and returns the vector containing the greater of each pair
/// of values.
///
return __builtin_ia32_maxpd((__v2df)__a, (__v2df)__b);
}
-/// \brief Performs a bitwise AND of two 128-bit vectors of [2 x double].
+/// Performs a bitwise AND of two 128-bit vectors of [2 x double].
///
/// \headerfile <x86intrin.h>
///
return (__m128d)((__v2du)__a & (__v2du)__b);
}
-/// \brief Performs a bitwise AND of two 128-bit vectors of [2 x double], using
+/// Performs a bitwise AND of two 128-bit vectors of [2 x double], using
/// the one's complement of the values contained in the first source operand.
///
/// \headerfile <x86intrin.h>
return (__m128d)(~(__v2du)__a & (__v2du)__b);
}
-/// \brief Performs a bitwise OR of two 128-bit vectors of [2 x double].
+/// Performs a bitwise OR of two 128-bit vectors of [2 x double].
///
/// \headerfile <x86intrin.h>
///
return (__m128d)((__v2du)__a | (__v2du)__b);
}
-/// \brief Performs a bitwise XOR of two 128-bit vectors of [2 x double].
+/// Performs a bitwise XOR of two 128-bit vectors of [2 x double].
///
/// \headerfile <x86intrin.h>
///
return (__m128d)((__v2du)__a ^ (__v2du)__b);
}
-/// \brief Compares each of the corresponding double-precision values of the
-/// 128-bit vectors of [2 x double] for equality. Each comparison yields 0h
-/// for false, FFFFFFFFFFFFFFFFh for true.
+/// Compares each of the corresponding double-precision values of the
+/// 128-bit vectors of [2 x double] for equality. Each comparison yields 0x0
+/// for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpeqpd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares each of the corresponding double-precision values of the
+/// Compares each of the corresponding double-precision values of the
/// 128-bit vectors of [2 x double] to determine if the values in the first
/// operand are less than those in the second operand. Each comparison
-/// yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpltpd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares each of the corresponding double-precision values of the
+/// Compares each of the corresponding double-precision values of the
/// 128-bit vectors of [2 x double] to determine if the values in the first
/// operand are less than or equal to those in the second operand.
///
-/// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// Each comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmplepd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares each of the corresponding double-precision values of the
+/// Compares each of the corresponding double-precision values of the
/// 128-bit vectors of [2 x double] to determine if the values in the first
/// operand are greater than those in the second operand.
///
-/// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// Each comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpltpd((__v2df)__b, (__v2df)__a);
}
-/// \brief Compares each of the corresponding double-precision values of the
+/// Compares each of the corresponding double-precision values of the
/// 128-bit vectors of [2 x double] to determine if the values in the first
/// operand are greater than or equal to those in the second operand.
///
-/// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// Each comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmplepd((__v2df)__b, (__v2df)__a);
}
-/// \brief Compares each of the corresponding double-precision values of the
+/// Compares each of the corresponding double-precision values of the
/// 128-bit vectors of [2 x double] to determine if the values in the first
/// operand are ordered with respect to those in the second operand.
///
/// A pair of double-precision values are "ordered" with respect to each
-/// other if neither value is a NaN. Each comparison yields 0h for false,
-/// FFFFFFFFFFFFFFFFh for true.
+/// other if neither value is a NaN. Each comparison yields 0x0 for false,
+/// 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpordpd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares each of the corresponding double-precision values of the
+/// Compares each of the corresponding double-precision values of the
/// 128-bit vectors of [2 x double] to determine if the values in the first
/// operand are unordered with respect to those in the second operand.
///
/// A pair of double-precision values are "unordered" with respect to each
-/// other if one or both values are NaN. Each comparison yields 0h for false,
-/// FFFFFFFFFFFFFFFFh for true.
+/// other if one or both values are NaN. Each comparison yields 0x0 for
+/// false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpunordpd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares each of the corresponding double-precision values of the
+/// Compares each of the corresponding double-precision values of the
/// 128-bit vectors of [2 x double] to determine if the values in the first
/// operand are unequal to those in the second operand.
///
-/// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// Each comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpneqpd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares each of the corresponding double-precision values of the
+/// Compares each of the corresponding double-precision values of the
/// 128-bit vectors of [2 x double] to determine if the values in the first
/// operand are not less than those in the second operand.
///
-/// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// Each comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpnltpd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares each of the corresponding double-precision values of the
+/// Compares each of the corresponding double-precision values of the
/// 128-bit vectors of [2 x double] to determine if the values in the first
/// operand are not less than or equal to those in the second operand.
///
-/// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// Each comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpnlepd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares each of the corresponding double-precision values of the
+/// Compares each of the corresponding double-precision values of the
/// 128-bit vectors of [2 x double] to determine if the values in the first
/// operand are not greater than those in the second operand.
///
-/// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// Each comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpnltpd((__v2df)__b, (__v2df)__a);
}
-/// \brief Compares each of the corresponding double-precision values of the
+/// Compares each of the corresponding double-precision values of the
/// 128-bit vectors of [2 x double] to determine if the values in the first
/// operand are not greater than or equal to those in the second operand.
///
-/// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// Each comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpnlepd((__v2df)__b, (__v2df)__a);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] for equality.
///
-/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// The comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpeqsd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is less than the corresponding value in
/// the second parameter.
///
-/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// The comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpltsd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is less than or equal to the
/// corresponding value in the second parameter.
///
-/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// The comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmplesd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is greater than the corresponding value
/// in the second parameter.
///
-/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// The comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
_mm_cmpgt_sd(__m128d __a, __m128d __b)
{
__m128d __c = __builtin_ia32_cmpltsd((__v2df)__b, (__v2df)__a);
- return (__m128d) { __c[0], __a[1] };
+ return __extension__ (__m128d) { __c[0], __a[1] };
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is greater than or equal to the
/// corresponding value in the second parameter.
///
-/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// The comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
_mm_cmpge_sd(__m128d __a, __m128d __b)
{
__m128d __c = __builtin_ia32_cmplesd((__v2df)__b, (__v2df)__a);
- return (__m128d) { __c[0], __a[1] };
+ return __extension__ (__m128d) { __c[0], __a[1] };
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is "ordered" with respect to the
/// corresponding value in the second parameter.
///
-/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. A pair of
-/// double-precision values are "ordered" with respect to each other if
+/// The comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true. A pair
+/// of double-precision values are "ordered" with respect to each other if
/// neither value is a NaN.
///
/// \headerfile <x86intrin.h>
return (__m128d)__builtin_ia32_cmpordsd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is "unordered" with respect to the
/// corresponding value in the second parameter.
///
-/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. A pair of
-/// double-precision values are "unordered" with respect to each other if one
-/// or both values are NaN.
+/// The comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true. A pair
+/// of double-precision values are "unordered" with respect to each other if
+/// one or both values are NaN.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpunordsd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is unequal to the corresponding value in
/// the second parameter.
///
-/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// The comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpneqsd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is not less than the corresponding
/// value in the second parameter.
///
-/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// The comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpnltsd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is not less than or equal to the
/// corresponding value in the second parameter.
///
-/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// The comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128d)__builtin_ia32_cmpnlesd((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is not greater than the corresponding
/// value in the second parameter.
///
-/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// The comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
_mm_cmpngt_sd(__m128d __a, __m128d __b)
{
__m128d __c = __builtin_ia32_cmpnltsd((__v2df)__b, (__v2df)__a);
- return (__m128d) { __c[0], __a[1] };
+ return __extension__ (__m128d) { __c[0], __a[1] };
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is not greater than or equal to the
/// corresponding value in the second parameter.
///
-/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+/// The comparison yields 0x0 for false, 0xFFFFFFFFFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
_mm_cmpnge_sd(__m128d __a, __m128d __b)
{
__m128d __c = __builtin_ia32_cmpnlesd((__v2df)__b, (__v2df)__a);
- return (__m128d) { __c[0], __a[1] };
+ return __extension__ (__m128d) { __c[0], __a[1] };
}
-/// \brief Compares the lower double-precision floating-point values in each of
-/// the two 128-bit floating-point vectors of [2 x double] for equality. The
-/// comparison yields 0 for false, 1 for true.
+/// Compares the lower double-precision floating-point values in each of
+/// the two 128-bit floating-point vectors of [2 x double] for equality.
+///
+/// The comparison yields 0 for false, 1 for true. If either of the two
+/// lower double-precision values is NaN, 0 is returned.
///
/// \headerfile <x86intrin.h>
///
/// \param __b
/// A 128-bit vector of [2 x double]. The lower double-precision value is
/// compared to the lower double-precision value of \a __a.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the two
+/// lower double-precision values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_comieq_sd(__m128d __a, __m128d __b)
{
return __builtin_ia32_comisdeq((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is less than the corresponding value in
/// the second parameter.
///
-/// The comparison yields 0 for false, 1 for true.
+/// The comparison yields 0 for false, 1 for true. If either of the two
+/// lower double-precision values is NaN, 0 is returned.
///
/// \headerfile <x86intrin.h>
///
/// \param __b
/// A 128-bit vector of [2 x double]. The lower double-precision value is
/// compared to the lower double-precision value of \a __a.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the two
+/// lower double-precision values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_comilt_sd(__m128d __a, __m128d __b)
{
return __builtin_ia32_comisdlt((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is less than or equal to the
/// corresponding value in the second parameter.
///
-/// The comparison yields 0 for false, 1 for true.
+/// The comparison yields 0 for false, 1 for true. If either of the two
+/// lower double-precision values is NaN, 0 is returned.
///
/// \headerfile <x86intrin.h>
///
/// \param __b
/// A 128-bit vector of [2 x double]. The lower double-precision value is
/// compared to the lower double-precision value of \a __a.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the two
+/// lower double-precision values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_comile_sd(__m128d __a, __m128d __b)
{
return __builtin_ia32_comisdle((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is greater than the corresponding value
/// in the second parameter.
///
-/// The comparison yields 0 for false, 1 for true.
+/// The comparison yields 0 for false, 1 for true. If either of the two
+/// lower double-precision values is NaN, 0 is returned.
///
/// \headerfile <x86intrin.h>
///
/// \param __b
/// A 128-bit vector of [2 x double]. The lower double-precision value is
/// compared to the lower double-precision value of \a __a.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the two
+/// lower double-precision values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_comigt_sd(__m128d __a, __m128d __b)
{
return __builtin_ia32_comisdgt((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is greater than or equal to the
/// corresponding value in the second parameter.
///
-/// The comparison yields 0 for false, 1 for true.
+/// The comparison yields 0 for false, 1 for true. If either of the two
+/// lower double-precision values is NaN, 0 is returned.
///
/// \headerfile <x86intrin.h>
///
/// \param __b
/// A 128-bit vector of [2 x double]. The lower double-precision value is
/// compared to the lower double-precision value of \a __a.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the two
+/// lower double-precision values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_comige_sd(__m128d __a, __m128d __b)
{
return __builtin_ia32_comisdge((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is unequal to the corresponding value in
/// the second parameter.
///
-/// The comparison yields 0 for false, 1 for true.
+/// The comparison yields 0 for false, 1 for true. If either of the two
+/// lower double-precision values is NaN, 1 is returned.
///
/// \headerfile <x86intrin.h>
///
/// \param __b
/// A 128-bit vector of [2 x double]. The lower double-precision value is
/// compared to the lower double-precision value of \a __a.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the two
+/// lower double-precision values is NaN, 1 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_comineq_sd(__m128d __a, __m128d __b)
{
return __builtin_ia32_comisdneq((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] for equality. The
/// comparison yields 0 for false, 1 for true.
///
-/// If either of the two lower double-precision values is NaN, 1 is returned.
+/// If either of the two lower double-precision values is NaN, 0 is returned.
///
/// \headerfile <x86intrin.h>
///
/// A 128-bit vector of [2 x double]. The lower double-precision value is
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison results. If either of the two
-/// lower double-precision values is NaN, 1 is returned.
+/// lower double-precision values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_ucomieq_sd(__m128d __a, __m128d __b)
{
return __builtin_ia32_ucomisdeq((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is less than the corresponding value in
/// the second parameter.
///
/// The comparison yields 0 for false, 1 for true. If either of the two lower
-/// double-precision values is NaN, 1 is returned.
+/// double-precision values is NaN, 0 is returned.
///
/// \headerfile <x86intrin.h>
///
/// A 128-bit vector of [2 x double]. The lower double-precision value is
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison results. If either of the two
-/// lower double-precision values is NaN, 1 is returned.
+/// lower double-precision values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_ucomilt_sd(__m128d __a, __m128d __b)
{
return __builtin_ia32_ucomisdlt((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is less than or equal to the
/// corresponding value in the second parameter.
///
/// The comparison yields 0 for false, 1 for true. If either of the two lower
-/// double-precision values is NaN, 1 is returned.
+/// double-precision values is NaN, 0 is returned.
///
/// \headerfile <x86intrin.h>
///
/// A 128-bit vector of [2 x double]. The lower double-precision value is
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison results. If either of the two
-/// lower double-precision values is NaN, 1 is returned.
+/// lower double-precision values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_ucomile_sd(__m128d __a, __m128d __b)
{
return __builtin_ia32_ucomisdle((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is greater than the corresponding value
/// in the second parameter.
return __builtin_ia32_ucomisdgt((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is greater than or equal to the
/// corresponding value in the second parameter.
return __builtin_ia32_ucomisdge((__v2df)__a, (__v2df)__b);
}
-/// \brief Compares the lower double-precision floating-point values in each of
+/// Compares the lower double-precision floating-point values in each of
/// the two 128-bit floating-point vectors of [2 x double] to determine if
/// the value in the first parameter is unequal to the corresponding value in
/// the second parameter.
///
/// The comparison yields 0 for false, 1 for true. If either of the two lower
-/// double-precision values is NaN, 0 is returned.
+/// double-precision values is NaN, 1 is returned.
///
/// \headerfile <x86intrin.h>
///
/// A 128-bit vector of [2 x double]. The lower double-precision value is
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison result. If either of the two
-/// lower double-precision values is NaN, 0 is returned.
+/// lower double-precision values is NaN, 1 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_ucomineq_sd(__m128d __a, __m128d __b)
{
return __builtin_ia32_ucomisdneq((__v2df)__a, (__v2df)__b);
}
-/// \brief Converts the two double-precision floating-point elements of a
+/// Converts the two double-precision floating-point elements of a
/// 128-bit vector of [2 x double] into two single-precision floating-point
/// values, returned in the lower 64 bits of a 128-bit vector of [4 x float].
/// The upper 64 bits of the result vector are set to zero.
return __builtin_ia32_cvtpd2ps((__v2df)__a);
}
-/// \brief Converts the lower two single-precision floating-point elements of a
+/// Converts the lower two single-precision floating-point elements of a
/// 128-bit vector of [4 x float] into two double-precision floating-point
/// values, returned in a 128-bit vector of [2 x double]. The upper two
/// elements of the input vector are unused.
__builtin_shufflevector((__v4sf)__a, (__v4sf)__a, 0, 1), __v2df);
}
-/// \brief Converts the lower two integer elements of a 128-bit vector of
+/// Converts the lower two integer elements of a 128-bit vector of
/// [4 x i32] into two double-precision floating-point values, returned in a
/// 128-bit vector of [2 x double].
///
__builtin_shufflevector((__v4si)__a, (__v4si)__a, 0, 1), __v2df);
}
-/// \brief Converts the two double-precision floating-point elements of a
+/// Converts the two double-precision floating-point elements of a
/// 128-bit vector of [2 x double] into two signed 32-bit integer values,
/// returned in the lower 64 bits of a 128-bit vector of [4 x i32]. The upper
/// 64 bits of the result vector are set to zero.
return __builtin_ia32_cvtpd2dq((__v2df)__a);
}
-/// \brief Converts the low-order element of a 128-bit vector of [2 x double]
+/// Converts the low-order element of a 128-bit vector of [2 x double]
/// into a 32-bit signed integer value.
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_cvtsd2si((__v2df)__a);
}
-/// \brief Converts the lower double-precision floating-point element of a
+/// Converts the lower double-precision floating-point element of a
/// 128-bit vector of [2 x double], in the second parameter, into a
/// single-precision floating-point value, returned in the lower 32 bits of a
/// 128-bit vector of [4 x float]. The upper 96 bits of the result vector are
return (__m128)__builtin_ia32_cvtsd2ss((__v4sf)__a, (__v2df)__b);
}
-/// \brief Converts a 32-bit signed integer value, in the second parameter, into
+/// Converts a 32-bit signed integer value, in the second parameter, into
/// a double-precision floating-point value, returned in the lower 64 bits of
/// a 128-bit vector of [2 x double]. The upper 64 bits of the result vector
/// are copied from the upper 64 bits of the first parameter.
return __a;
}
-/// \brief Converts the lower single-precision floating-point element of a
+/// Converts the lower single-precision floating-point element of a
/// 128-bit vector of [4 x float], in the second parameter, into a
/// double-precision floating-point value, returned in the lower 64 bits of
/// a 128-bit vector of [2 x double]. The upper 64 bits of the result vector
return __a;
}
-/// \brief Converts the two double-precision floating-point elements of a
+/// Converts the two double-precision floating-point elements of a
/// 128-bit vector of [2 x double] into two signed 32-bit integer values,
/// returned in the lower 64 bits of a 128-bit vector of [4 x i32].
///
return (__m128i)__builtin_ia32_cvttpd2dq((__v2df)__a);
}
-/// \brief Converts the low-order element of a [2 x double] vector into a 32-bit
+/// Converts the low-order element of a [2 x double] vector into a 32-bit
/// signed integer value, truncating the result when it is inexact.
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_cvttsd2si((__v2df)__a);
}
-/// \brief Converts the two double-precision floating-point elements of a
+/// Converts the two double-precision floating-point elements of a
/// 128-bit vector of [2 x double] into two signed 32-bit integer values,
/// returned in a 64-bit vector of [2 x i32].
///
/// \param __a
/// A 128-bit vector of [2 x double].
/// \returns A 64-bit vector of [2 x i32] containing the converted values.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_cvtpd_pi32(__m128d __a)
{
return (__m64)__builtin_ia32_cvtpd2pi((__v2df)__a);
}
-/// \brief Converts the two double-precision floating-point elements of a
+/// Converts the two double-precision floating-point elements of a
/// 128-bit vector of [2 x double] into two signed 32-bit integer values,
/// returned in a 64-bit vector of [2 x i32].
///
/// \param __a
/// A 128-bit vector of [2 x double].
/// \returns A 64-bit vector of [2 x i32] containing the converted values.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_cvttpd_pi32(__m128d __a)
{
return (__m64)__builtin_ia32_cvttpd2pi((__v2df)__a);
}
-/// \brief Converts the two signed 32-bit integer elements of a 64-bit vector of
+/// Converts the two signed 32-bit integer elements of a 64-bit vector of
/// [2 x i32] into two double-precision floating-point values, returned in a
/// 128-bit vector of [2 x double].
///
/// \param __a
/// A 64-bit vector of [2 x i32].
/// \returns A 128-bit vector of [2 x double] containing the converted values.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
+static __inline__ __m128d __DEFAULT_FN_ATTRS_MMX
_mm_cvtpi32_pd(__m64 __a)
{
return __builtin_ia32_cvtpi2pd((__v2si)__a);
}
-/// \brief Returns the low-order element of a 128-bit vector of [2 x double] as
+/// Returns the low-order element of a 128-bit vector of [2 x double] as
/// a double-precision floating-point value.
///
/// \headerfile <x86intrin.h>
return __a[0];
}
-/// \brief Loads a 128-bit floating-point vector of [2 x double] from an aligned
+/// Loads a 128-bit floating-point vector of [2 x double] from an aligned
/// memory location.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128d __DEFAULT_FN_ATTRS
_mm_load_pd(double const *__dp)
{
- return *(__m128d*)__dp;
+ return *(const __m128d*)__dp;
}
-/// \brief Loads a double-precision floating-point value from a specified memory
+/// Loads a double-precision floating-point value from a specified memory
/// location and duplicates it to both vector elements of a 128-bit vector of
/// [2 x double].
///
struct __mm_load1_pd_struct {
double __u;
} __attribute__((__packed__, __may_alias__));
- double __u = ((struct __mm_load1_pd_struct*)__dp)->__u;
- return (__m128d){ __u, __u };
+ double __u = ((const struct __mm_load1_pd_struct*)__dp)->__u;
+ return __extension__ (__m128d){ __u, __u };
}
#define _mm_load_pd1(dp) _mm_load1_pd(dp)
-/// \brief Loads two double-precision values, in reverse order, from an aligned
+/// Loads two double-precision values, in reverse order, from an aligned
/// memory location into a 128-bit vector of [2 x double].
///
/// \headerfile <x86intrin.h>
static __inline__ __m128d __DEFAULT_FN_ATTRS
_mm_loadr_pd(double const *__dp)
{
- __m128d __u = *(__m128d*)__dp;
+ __m128d __u = *(const __m128d*)__dp;
return __builtin_shufflevector((__v2df)__u, (__v2df)__u, 1, 0);
}
-/// \brief Loads a 128-bit floating-point vector of [2 x double] from an
+/// Loads a 128-bit floating-point vector of [2 x double] from an
/// unaligned memory location.
///
/// \headerfile <x86intrin.h>
_mm_loadu_pd(double const *__dp)
{
struct __loadu_pd {
- __m128d __v;
+ __m128d_u __v;
} __attribute__((__packed__, __may_alias__));
- return ((struct __loadu_pd*)__dp)->__v;
+ return ((const struct __loadu_pd*)__dp)->__v;
}
-/// \brief Loads a 64-bit integer value to the low element of a 128-bit integer
+/// Loads a 64-bit integer value to the low element of a 128-bit integer
/// vector and clears the upper element.
///
/// \headerfile <x86intrin.h>
struct __loadu_si64 {
long long __v;
} __attribute__((__packed__, __may_alias__));
- long long __u = ((struct __loadu_si64*)__a)->__v;
- return (__m128i){__u, 0L};
+ long long __u = ((const struct __loadu_si64*)__a)->__v;
+ return __extension__ (__m128i)(__v2di){__u, 0LL};
}
-/// \brief Loads a 64-bit double-precision value to the low element of a
+/// Loads a 32-bit integer value to the low element of a 128-bit integer
+/// vector and clears the upper element.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VMOVD / MOVD </c> instruction.
+///
+/// \param __a
+/// A pointer to a 32-bit memory location. The address of the memory
+/// location does not have to be aligned.
+/// \returns A 128-bit vector of [4 x i32] containing the loaded value.
+static __inline__ __m128i __DEFAULT_FN_ATTRS
+_mm_loadu_si32(void const *__a)
+{
+ struct __loadu_si32 {
+ int __v;
+ } __attribute__((__packed__, __may_alias__));
+ int __u = ((const struct __loadu_si32*)__a)->__v;
+ return __extension__ (__m128i)(__v4si){__u, 0, 0, 0};
+}
+
+/// Loads a 16-bit integer value to the low element of a 128-bit integer
+/// vector and clears the upper element.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic does not correspond to a specific instruction.
+///
+/// \param __a
+/// A pointer to a 16-bit memory location. The address of the memory
+/// location does not have to be aligned.
+/// \returns A 128-bit vector of [8 x i16] containing the loaded value.
+static __inline__ __m128i __DEFAULT_FN_ATTRS
+_mm_loadu_si16(void const *__a)
+{
+ struct __loadu_si16 {
+ short __v;
+ } __attribute__((__packed__, __may_alias__));
+ short __u = ((const struct __loadu_si16*)__a)->__v;
+ return __extension__ (__m128i)(__v8hi){__u, 0, 0, 0, 0, 0, 0, 0};
+}
+
+/// Loads a 64-bit double-precision value to the low element of a
/// 128-bit integer vector and clears the upper element.
///
/// \headerfile <x86intrin.h>
struct __mm_load_sd_struct {
double __u;
} __attribute__((__packed__, __may_alias__));
- double __u = ((struct __mm_load_sd_struct*)__dp)->__u;
- return (__m128d){ __u, 0 };
+ double __u = ((const struct __mm_load_sd_struct*)__dp)->__u;
+ return __extension__ (__m128d){ __u, 0 };
}
-/// \brief Loads a double-precision value into the high-order bits of a 128-bit
+/// Loads a double-precision value into the high-order bits of a 128-bit
/// vector of [2 x double]. The low-order bits are copied from the low-order
/// bits of the first operand.
///
struct __mm_loadh_pd_struct {
double __u;
} __attribute__((__packed__, __may_alias__));
- double __u = ((struct __mm_loadh_pd_struct*)__dp)->__u;
- return (__m128d){ __a[0], __u };
+ double __u = ((const struct __mm_loadh_pd_struct*)__dp)->__u;
+ return __extension__ (__m128d){ __a[0], __u };
}
-/// \brief Loads a double-precision value into the low-order bits of a 128-bit
+/// Loads a double-precision value into the low-order bits of a 128-bit
/// vector of [2 x double]. The high-order bits are copied from the
/// high-order bits of the first operand.
///
struct __mm_loadl_pd_struct {
double __u;
} __attribute__((__packed__, __may_alias__));
- double __u = ((struct __mm_loadl_pd_struct*)__dp)->__u;
- return (__m128d){ __u, __a[1] };
+ double __u = ((const struct __mm_loadl_pd_struct*)__dp)->__u;
+ return __extension__ (__m128d){ __u, __a[1] };
}
-/// \brief Constructs a 128-bit floating-point vector of [2 x double] with
+/// Constructs a 128-bit floating-point vector of [2 x double] with
/// unspecified content. This could be used as an argument to another
/// intrinsic function where the argument is required but the value is not
/// actually used.
return (__m128d)__builtin_ia32_undef128();
}
-/// \brief Constructs a 128-bit floating-point vector of [2 x double]. The lower
+/// Constructs a 128-bit floating-point vector of [2 x double]. The lower
/// 64 bits of the vector are initialized with the specified double-precision
/// floating-point value. The upper 64 bits are set to zero.
///
static __inline__ __m128d __DEFAULT_FN_ATTRS
_mm_set_sd(double __w)
{
- return (__m128d){ __w, 0 };
+ return __extension__ (__m128d){ __w, 0 };
}
-/// \brief Constructs a 128-bit floating-point vector of [2 x double], with each
+/// Constructs a 128-bit floating-point vector of [2 x double], with each
/// of the two double-precision floating-point vector elements set to the
/// specified double-precision floating-point value.
///
static __inline__ __m128d __DEFAULT_FN_ATTRS
_mm_set1_pd(double __w)
{
- return (__m128d){ __w, __w };
+ return __extension__ (__m128d){ __w, __w };
}
-/// \brief Constructs a 128-bit floating-point vector of [2 x double], with each
+/// Constructs a 128-bit floating-point vector of [2 x double], with each
/// of the two double-precision floating-point vector elements set to the
/// specified double-precision floating-point value.
///
return _mm_set1_pd(__w);
}
-/// \brief Constructs a 128-bit floating-point vector of [2 x double]
+/// Constructs a 128-bit floating-point vector of [2 x double]
/// initialized with the specified double-precision floating-point values.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128d __DEFAULT_FN_ATTRS
_mm_set_pd(double __w, double __x)
{
- return (__m128d){ __x, __w };
+ return __extension__ (__m128d){ __x, __w };
}
-/// \brief Constructs a 128-bit floating-point vector of [2 x double],
+/// Constructs a 128-bit floating-point vector of [2 x double],
/// initialized in reverse order with the specified double-precision
/// floating-point values.
///
static __inline__ __m128d __DEFAULT_FN_ATTRS
_mm_setr_pd(double __w, double __x)
{
- return (__m128d){ __w, __x };
+ return __extension__ (__m128d){ __w, __x };
}
-/// \brief Constructs a 128-bit floating-point vector of [2 x double]
+/// Constructs a 128-bit floating-point vector of [2 x double]
/// initialized to zero.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128d __DEFAULT_FN_ATTRS
_mm_setzero_pd(void)
{
- return (__m128d){ 0, 0 };
+ return __extension__ (__m128d){ 0, 0 };
}
-/// \brief Constructs a 128-bit floating-point vector of [2 x double]. The lower
+/// Constructs a 128-bit floating-point vector of [2 x double]. The lower
/// 64 bits are set to the lower 64 bits of the second parameter. The upper
/// 64 bits are set to the upper 64 bits of the first parameter.
///
static __inline__ __m128d __DEFAULT_FN_ATTRS
_mm_move_sd(__m128d __a, __m128d __b)
{
- return (__m128d){ __b[0], __a[1] };
+ __a[0] = __b[0];
+ return __a;
}
-/// \brief Stores the lower 64 bits of a 128-bit vector of [2 x double] to a
+/// Stores the lower 64 bits of a 128-bit vector of [2 x double] to a
/// memory location.
///
/// \headerfile <x86intrin.h>
((struct __mm_store_sd_struct*)__dp)->__u = __a[0];
}
-/// \brief Moves packed double-precision values from a 128-bit vector of
+/// Moves packed double-precision values from a 128-bit vector of
/// [2 x double] to a memory location.
///
/// \headerfile <x86intrin.h>
*(__m128d*)__dp = __a;
}
-/// \brief Moves the lower 64 bits of a 128-bit vector of [2 x double] twice to
+/// Moves the lower 64 bits of a 128-bit vector of [2 x double] twice to
/// the upper and lower 64 bits of a memory location.
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c>VMOVDDUP + VMOVAPD / MOVLHPS + MOVAPS </c> instruction.
+/// This intrinsic corresponds to the
+/// <c> VMOVDDUP + VMOVAPD / MOVLHPS + MOVAPS </c> instruction.
///
/// \param __dp
/// A pointer to a memory location that can store two double-precision
/// values.
/// \param __a
/// A 128-bit vector of [2 x double] whose lower 64 bits are copied to each
-/// of the values in \a dp.
+/// of the values in \a __dp.
static __inline__ void __DEFAULT_FN_ATTRS
_mm_store1_pd(double *__dp, __m128d __a)
{
_mm_store_pd(__dp, __a);
}
-/// \brief Stores a 128-bit vector of [2 x double] into an aligned memory
-/// location.
+/// Moves the lower 64 bits of a 128-bit vector of [2 x double] twice to
+/// the upper and lower 64 bits of a memory location.
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c> VMOVAPD / MOVAPD </c> instruction.
+/// This intrinsic corresponds to the
+/// <c> VMOVDDUP + VMOVAPD / MOVLHPS + MOVAPS </c> instruction.
///
/// \param __dp
-/// A pointer to a 128-bit memory location. The address of the memory
-/// location has to be 16-byte aligned.
+/// A pointer to a memory location that can store two double-precision
+/// values.
/// \param __a
-/// A 128-bit vector of [2 x double] containing the values to be stored.
+/// A 128-bit vector of [2 x double] whose lower 64 bits are copied to each
+/// of the values in \a __dp.
static __inline__ void __DEFAULT_FN_ATTRS
_mm_store_pd1(double *__dp, __m128d __a)
{
- return _mm_store1_pd(__dp, __a);
+ _mm_store1_pd(__dp, __a);
}
-/// \brief Stores a 128-bit vector of [2 x double] into an unaligned memory
+/// Stores a 128-bit vector of [2 x double] into an unaligned memory
/// location.
///
/// \headerfile <x86intrin.h>
_mm_storeu_pd(double *__dp, __m128d __a)
{
struct __storeu_pd {
- __m128d __v;
+ __m128d_u __v;
} __attribute__((__packed__, __may_alias__));
((struct __storeu_pd*)__dp)->__v = __a;
}
-/// \brief Stores two double-precision values, in reverse order, from a 128-bit
+/// Stores two double-precision values, in reverse order, from a 128-bit
/// vector of [2 x double] to a 16-byte aligned memory location.
///
/// \headerfile <x86intrin.h>
*(__m128d *)__dp = __a;
}
-/// \brief Stores the upper 64 bits of a 128-bit vector of [2 x double] to a
+/// Stores the upper 64 bits of a 128-bit vector of [2 x double] to a
/// memory location.
///
/// \headerfile <x86intrin.h>
((struct __mm_storeh_pd_struct*)__dp)->__u = __a[1];
}
-/// \brief Stores the lower 64 bits of a 128-bit vector of [2 x double] to a
+/// Stores the lower 64 bits of a 128-bit vector of [2 x double] to a
/// memory location.
///
/// \headerfile <x86intrin.h>
((struct __mm_storeh_pd_struct*)__dp)->__u = __a[0];
}
-/// \brief Adds the corresponding elements of two 128-bit vectors of [16 x i8],
+/// Adds the corresponding elements of two 128-bit vectors of [16 x i8],
/// saving the lower 8 bits of each sum in the corresponding element of a
/// 128-bit result vector of [16 x i8].
///
return (__m128i)((__v16qu)__a + (__v16qu)__b);
}
-/// \brief Adds the corresponding elements of two 128-bit vectors of [8 x i16],
+/// Adds the corresponding elements of two 128-bit vectors of [8 x i16],
/// saving the lower 16 bits of each sum in the corresponding element of a
/// 128-bit result vector of [8 x i16].
///
return (__m128i)((__v8hu)__a + (__v8hu)__b);
}
-/// \brief Adds the corresponding elements of two 128-bit vectors of [4 x i32],
+/// Adds the corresponding elements of two 128-bit vectors of [4 x i32],
/// saving the lower 32 bits of each sum in the corresponding element of a
/// 128-bit result vector of [4 x i32].
///
return (__m128i)((__v4su)__a + (__v4su)__b);
}
-/// \brief Adds two signed or unsigned 64-bit integer values, returning the
+/// Adds two signed or unsigned 64-bit integer values, returning the
/// lower 64 bits of the sum.
///
/// \headerfile <x86intrin.h>
/// \param __b
/// A 64-bit integer.
/// \returns A 64-bit integer containing the sum of both parameters.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_add_si64(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_paddq((__v1di)__a, (__v1di)__b);
}
-/// \brief Adds the corresponding elements of two 128-bit vectors of [2 x i64],
+/// Adds the corresponding elements of two 128-bit vectors of [2 x i64],
/// saving the lower 64 bits of each sum in the corresponding element of a
/// 128-bit result vector of [2 x i64].
///
return (__m128i)((__v2du)__a + (__v2du)__b);
}
-/// \brief Adds, with saturation, the corresponding elements of two 128-bit
+/// Adds, with saturation, the corresponding elements of two 128-bit
/// signed [16 x i8] vectors, saving each sum in the corresponding element of
-/// a 128-bit result vector of [16 x i8]. Positive sums greater than 7Fh are
-/// saturated to 7Fh. Negative sums less than 80h are saturated to 80h.
+/// a 128-bit result vector of [16 x i8]. Positive sums greater than 0x7F are
+/// saturated to 0x7F. Negative sums less than 0x80 are saturated to 0x80.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)__builtin_ia32_paddsb128((__v16qi)__a, (__v16qi)__b);
}
-/// \brief Adds, with saturation, the corresponding elements of two 128-bit
+/// Adds, with saturation, the corresponding elements of two 128-bit
/// signed [8 x i16] vectors, saving each sum in the corresponding element of
-/// a 128-bit result vector of [8 x i16]. Positive sums greater than 7FFFh
-/// are saturated to 7FFFh. Negative sums less than 8000h are saturated to
-/// 8000h.
+/// a 128-bit result vector of [8 x i16]. Positive sums greater than 0x7FFF
+/// are saturated to 0x7FFF. Negative sums less than 0x8000 are saturated to
+/// 0x8000.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)__builtin_ia32_paddsw128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Adds, with saturation, the corresponding elements of two 128-bit
+/// Adds, with saturation, the corresponding elements of two 128-bit
/// unsigned [16 x i8] vectors, saving each sum in the corresponding element
-/// of a 128-bit result vector of [16 x i8]. Positive sums greater than FFh
-/// are saturated to FFh. Negative sums are saturated to 00h.
+/// of a 128-bit result vector of [16 x i8]. Positive sums greater than 0xFF
+/// are saturated to 0xFF. Negative sums are saturated to 0x00.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)__builtin_ia32_paddusb128((__v16qi)__a, (__v16qi)__b);
}
-/// \brief Adds, with saturation, the corresponding elements of two 128-bit
+/// Adds, with saturation, the corresponding elements of two 128-bit
/// unsigned [8 x i16] vectors, saving each sum in the corresponding element
-/// of a 128-bit result vector of [8 x i16]. Positive sums greater than FFFFh
-/// are saturated to FFFFh. Negative sums are saturated to 0000h.
+/// of a 128-bit result vector of [8 x i16]. Positive sums greater than
+/// 0xFFFF are saturated to 0xFFFF. Negative sums are saturated to 0x0000.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)__builtin_ia32_paddusw128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Computes the rounded avarages of corresponding elements of two
+/// Computes the rounded averages of corresponding elements of two
/// 128-bit unsigned [16 x i8] vectors, saving each result in the
/// corresponding element of a 128-bit result vector of [16 x i8].
///
return (__m128i)__builtin_ia32_pavgb128((__v16qi)__a, (__v16qi)__b);
}
-/// \brief Computes the rounded avarages of corresponding elements of two
+/// Computes the rounded averages of corresponding elements of two
/// 128-bit unsigned [8 x i16] vectors, saving each result in the
/// corresponding element of a 128-bit result vector of [8 x i16].
///
return (__m128i)__builtin_ia32_pavgw128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Multiplies the corresponding elements of two 128-bit signed [8 x i16]
+/// Multiplies the corresponding elements of two 128-bit signed [8 x i16]
/// vectors, producing eight intermediate 32-bit signed integer products, and
/// adds the consecutive pairs of 32-bit products to form a 128-bit signed
/// [4 x i32] vector.
return (__m128i)__builtin_ia32_pmaddwd128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Compares corresponding elements of two 128-bit signed [8 x i16]
+/// Compares corresponding elements of two 128-bit signed [8 x i16]
/// vectors, saving the greater value from each comparison in the
/// corresponding element of a 128-bit result vector of [8 x i16].
///
return (__m128i)__builtin_ia32_pmaxsw128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Compares corresponding elements of two 128-bit unsigned [16 x i8]
+/// Compares corresponding elements of two 128-bit unsigned [16 x i8]
/// vectors, saving the greater value from each comparison in the
/// corresponding element of a 128-bit result vector of [16 x i8].
///
return (__m128i)__builtin_ia32_pmaxub128((__v16qi)__a, (__v16qi)__b);
}
-/// \brief Compares corresponding elements of two 128-bit signed [8 x i16]
+/// Compares corresponding elements of two 128-bit signed [8 x i16]
/// vectors, saving the smaller value from each comparison in the
/// corresponding element of a 128-bit result vector of [8 x i16].
///
return (__m128i)__builtin_ia32_pminsw128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Compares corresponding elements of two 128-bit unsigned [16 x i8]
+/// Compares corresponding elements of two 128-bit unsigned [16 x i8]
/// vectors, saving the smaller value from each comparison in the
/// corresponding element of a 128-bit result vector of [16 x i8].
///
return (__m128i)__builtin_ia32_pminub128((__v16qi)__a, (__v16qi)__b);
}
-/// \brief Multiplies the corresponding elements of two signed [8 x i16]
+/// Multiplies the corresponding elements of two signed [8 x i16]
/// vectors, saving the upper 16 bits of each 32-bit product in the
/// corresponding element of a 128-bit signed [8 x i16] result vector.
///
return (__m128i)__builtin_ia32_pmulhw128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Multiplies the corresponding elements of two unsigned [8 x i16]
+/// Multiplies the corresponding elements of two unsigned [8 x i16]
/// vectors, saving the upper 16 bits of each 32-bit product in the
/// corresponding element of a 128-bit unsigned [8 x i16] result vector.
///
return (__m128i)__builtin_ia32_pmulhuw128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Multiplies the corresponding elements of two signed [8 x i16]
+/// Multiplies the corresponding elements of two signed [8 x i16]
/// vectors, saving the lower 16 bits of each 32-bit product in the
/// corresponding element of a 128-bit signed [8 x i16] result vector.
///
return (__m128i)((__v8hu)__a * (__v8hu)__b);
}
-/// \brief Multiplies 32-bit unsigned integer values contained in the lower bits
+/// Multiplies 32-bit unsigned integer values contained in the lower bits
/// of the two 64-bit integer vectors and returns the 64-bit unsigned
/// product.
///
/// \param __b
/// A 64-bit integer containing one of the source operands.
/// \returns A 64-bit integer vector containing the product of both operands.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_mul_su32(__m64 __a, __m64 __b)
{
return __builtin_ia32_pmuludq((__v2si)__a, (__v2si)__b);
}
-/// \brief Multiplies 32-bit unsigned integer values contained in the lower
+/// Multiplies 32-bit unsigned integer values contained in the lower
/// bits of the corresponding elements of two [2 x i64] vectors, and returns
/// the 64-bit products in the corresponding elements of a [2 x i64] vector.
///
return __builtin_ia32_pmuludq128((__v4si)__a, (__v4si)__b);
}
-/// \brief Computes the absolute differences of corresponding 8-bit integer
+/// Computes the absolute differences of corresponding 8-bit integer
/// values in two 128-bit vectors. Sums the first 8 absolute differences, and
/// separately sums the second 8 absolute differences. Packs these two
/// unsigned 16-bit integer sums into the upper and lower elements of a
return __builtin_ia32_psadbw128((__v16qi)__a, (__v16qi)__b);
}
-/// \brief Subtracts the corresponding 8-bit integer values in the operands.
+/// Subtracts the corresponding 8-bit integer values in the operands.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)((__v16qu)__a - (__v16qu)__b);
}
-/// \brief Subtracts the corresponding 16-bit integer values in the operands.
+/// Subtracts the corresponding 16-bit integer values in the operands.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)((__v8hu)__a - (__v8hu)__b);
}
-/// \brief Subtracts the corresponding 32-bit integer values in the operands.
+/// Subtracts the corresponding 32-bit integer values in the operands.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)((__v4su)__a - (__v4su)__b);
}
-/// \brief Subtracts signed or unsigned 64-bit integer values and writes the
+/// Subtracts signed or unsigned 64-bit integer values and writes the
/// difference to the corresponding bits in the destination.
///
/// \headerfile <x86intrin.h>
/// A 64-bit integer vector containing the subtrahend.
/// \returns A 64-bit integer vector containing the difference of the values in
/// the operands.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_sub_si64(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_psubq((__v1di)__a, (__v1di)__b);
}
-/// \brief Subtracts the corresponding elements of two [2 x i64] vectors.
+/// Subtracts the corresponding elements of two [2 x i64] vectors.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)((__v2du)__a - (__v2du)__b);
}
-/// \brief Subtracts corresponding 8-bit signed integer values in the input and
+/// Subtracts corresponding 8-bit signed integer values in the input and
/// returns the differences in the corresponding bytes in the destination.
-/// Differences greater than 7Fh are saturated to 7Fh, and differences less
-/// than 80h are saturated to 80h.
+/// Differences greater than 0x7F are saturated to 0x7F, and differences less
+/// than 0x80 are saturated to 0x80.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)__builtin_ia32_psubsb128((__v16qi)__a, (__v16qi)__b);
}
-/// \brief Subtracts corresponding 16-bit signed integer values in the input and
+/// Subtracts corresponding 16-bit signed integer values in the input and
/// returns the differences in the corresponding bytes in the destination.
-/// Differences greater than 7FFFh are saturated to 7FFFh, and values less
-/// than 8000h are saturated to 8000h.
+/// Differences greater than 0x7FFF are saturated to 0x7FFF, and values less
+/// than 0x8000 are saturated to 0x8000.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)__builtin_ia32_psubsw128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Subtracts corresponding 8-bit unsigned integer values in the input
+/// Subtracts corresponding 8-bit unsigned integer values in the input
/// and returns the differences in the corresponding bytes in the
-/// destination. Differences less than 00h are saturated to 00h.
+/// destination. Differences less than 0x00 are saturated to 0x00.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)__builtin_ia32_psubusb128((__v16qi)__a, (__v16qi)__b);
}
-/// \brief Subtracts corresponding 16-bit unsigned integer values in the input
+/// Subtracts corresponding 16-bit unsigned integer values in the input
/// and returns the differences in the corresponding bytes in the
-/// destination. Differences less than 0000h are saturated to 0000h.
+/// destination. Differences less than 0x0000 are saturated to 0x0000.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)__builtin_ia32_psubusw128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Performs a bitwise AND of two 128-bit integer vectors.
+/// Performs a bitwise AND of two 128-bit integer vectors.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)((__v2du)__a & (__v2du)__b);
}
-/// \brief Performs a bitwise AND of two 128-bit integer vectors, using the
+/// Performs a bitwise AND of two 128-bit integer vectors, using the
/// one's complement of the values contained in the first source operand.
///
/// \headerfile <x86intrin.h>
{
return (__m128i)(~(__v2du)__a & (__v2du)__b);
}
-/// \brief Performs a bitwise OR of two 128-bit integer vectors.
+/// Performs a bitwise OR of two 128-bit integer vectors.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)((__v2du)__a | (__v2du)__b);
}
-/// \brief Performs a bitwise exclusive OR of two 128-bit integer vectors.
+/// Performs a bitwise exclusive OR of two 128-bit integer vectors.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)((__v2du)__a ^ (__v2du)__b);
}
-/// \brief Left-shifts the 128-bit integer vector operand by the specified
+/// Left-shifts the 128-bit integer vector operand by the specified
/// number of bytes. Low-order bits are cleared.
///
/// \headerfile <x86intrin.h>
/// An immediate value specifying the number of bytes to left-shift operand
/// \a a.
/// \returns A 128-bit integer vector containing the left-shifted value.
-#define _mm_slli_si128(a, imm) __extension__ ({ \
- (__m128i)__builtin_shufflevector( \
- (__v16qi)_mm_setzero_si128(), \
- (__v16qi)(__m128i)(a), \
- ((char)(imm)&0xF0) ? 0 : 16 - (char)(imm), \
- ((char)(imm)&0xF0) ? 1 : 17 - (char)(imm), \
- ((char)(imm)&0xF0) ? 2 : 18 - (char)(imm), \
- ((char)(imm)&0xF0) ? 3 : 19 - (char)(imm), \
- ((char)(imm)&0xF0) ? 4 : 20 - (char)(imm), \
- ((char)(imm)&0xF0) ? 5 : 21 - (char)(imm), \
- ((char)(imm)&0xF0) ? 6 : 22 - (char)(imm), \
- ((char)(imm)&0xF0) ? 7 : 23 - (char)(imm), \
- ((char)(imm)&0xF0) ? 8 : 24 - (char)(imm), \
- ((char)(imm)&0xF0) ? 9 : 25 - (char)(imm), \
- ((char)(imm)&0xF0) ? 10 : 26 - (char)(imm), \
- ((char)(imm)&0xF0) ? 11 : 27 - (char)(imm), \
- ((char)(imm)&0xF0) ? 12 : 28 - (char)(imm), \
- ((char)(imm)&0xF0) ? 13 : 29 - (char)(imm), \
- ((char)(imm)&0xF0) ? 14 : 30 - (char)(imm), \
- ((char)(imm)&0xF0) ? 15 : 31 - (char)(imm)); })
+#define _mm_slli_si128(a, imm) \
+ (__m128i)__builtin_ia32_pslldqi128_byteshift((__v2di)(__m128i)(a), (int)(imm))
#define _mm_bslli_si128(a, imm) \
- _mm_slli_si128((a), (imm))
+ (__m128i)__builtin_ia32_pslldqi128_byteshift((__v2di)(__m128i)(a), (int)(imm))
-/// \brief Left-shifts each 16-bit value in the 128-bit integer vector operand
+/// Left-shifts each 16-bit value in the 128-bit integer vector operand
/// by the specified number of bits. Low-order bits are cleared.
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_ia32_psllwi128((__v8hi)__a, __count);
}
-/// \brief Left-shifts each 16-bit value in the 128-bit integer vector operand
+/// Left-shifts each 16-bit value in the 128-bit integer vector operand
/// by the specified number of bits. Low-order bits are cleared.
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_ia32_psllw128((__v8hi)__a, (__v8hi)__count);
}
-/// \brief Left-shifts each 32-bit value in the 128-bit integer vector operand
+/// Left-shifts each 32-bit value in the 128-bit integer vector operand
/// by the specified number of bits. Low-order bits are cleared.
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_ia32_pslldi128((__v4si)__a, __count);
}
-/// \brief Left-shifts each 32-bit value in the 128-bit integer vector operand
+/// Left-shifts each 32-bit value in the 128-bit integer vector operand
/// by the specified number of bits. Low-order bits are cleared.
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_ia32_pslld128((__v4si)__a, (__v4si)__count);
}
-/// \brief Left-shifts each 64-bit value in the 128-bit integer vector operand
+/// Left-shifts each 64-bit value in the 128-bit integer vector operand
/// by the specified number of bits. Low-order bits are cleared.
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_psllqi128((__v2di)__a, __count);
}
-/// \brief Left-shifts each 64-bit value in the 128-bit integer vector operand
+/// Left-shifts each 64-bit value in the 128-bit integer vector operand
/// by the specified number of bits. Low-order bits are cleared.
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_psllq128((__v2di)__a, (__v2di)__count);
}
-/// \brief Right-shifts each 16-bit value in the 128-bit integer vector operand
+/// Right-shifts each 16-bit value in the 128-bit integer vector operand
/// by the specified number of bits. High-order bits are filled with the sign
/// bit of the initial value.
///
return (__m128i)__builtin_ia32_psrawi128((__v8hi)__a, __count);
}
-/// \brief Right-shifts each 16-bit value in the 128-bit integer vector operand
+/// Right-shifts each 16-bit value in the 128-bit integer vector operand
/// by the specified number of bits. High-order bits are filled with the sign
/// bit of the initial value.
///
return (__m128i)__builtin_ia32_psraw128((__v8hi)__a, (__v8hi)__count);
}
-/// \brief Right-shifts each 32-bit value in the 128-bit integer vector operand
+/// Right-shifts each 32-bit value in the 128-bit integer vector operand
/// by the specified number of bits. High-order bits are filled with the sign
/// bit of the initial value.
///
return (__m128i)__builtin_ia32_psradi128((__v4si)__a, __count);
}
-/// \brief Right-shifts each 32-bit value in the 128-bit integer vector operand
+/// Right-shifts each 32-bit value in the 128-bit integer vector operand
/// by the specified number of bits. High-order bits are filled with the sign
/// bit of the initial value.
///
return (__m128i)__builtin_ia32_psrad128((__v4si)__a, (__v4si)__count);
}
-/// \brief Right-shifts the 128-bit integer vector operand by the specified
+/// Right-shifts the 128-bit integer vector operand by the specified
/// number of bytes. High-order bits are cleared.
///
/// \headerfile <x86intrin.h>
/// An immediate value specifying the number of bytes to right-shift operand
/// \a a.
/// \returns A 128-bit integer vector containing the right-shifted value.
-#define _mm_srli_si128(a, imm) __extension__ ({ \
- (__m128i)__builtin_shufflevector( \
- (__v16qi)(__m128i)(a), \
- (__v16qi)_mm_setzero_si128(), \
- ((char)(imm)&0xF0) ? 16 : (char)(imm) + 0, \
- ((char)(imm)&0xF0) ? 17 : (char)(imm) + 1, \
- ((char)(imm)&0xF0) ? 18 : (char)(imm) + 2, \
- ((char)(imm)&0xF0) ? 19 : (char)(imm) + 3, \
- ((char)(imm)&0xF0) ? 20 : (char)(imm) + 4, \
- ((char)(imm)&0xF0) ? 21 : (char)(imm) + 5, \
- ((char)(imm)&0xF0) ? 22 : (char)(imm) + 6, \
- ((char)(imm)&0xF0) ? 23 : (char)(imm) + 7, \
- ((char)(imm)&0xF0) ? 24 : (char)(imm) + 8, \
- ((char)(imm)&0xF0) ? 25 : (char)(imm) + 9, \
- ((char)(imm)&0xF0) ? 26 : (char)(imm) + 10, \
- ((char)(imm)&0xF0) ? 27 : (char)(imm) + 11, \
- ((char)(imm)&0xF0) ? 28 : (char)(imm) + 12, \
- ((char)(imm)&0xF0) ? 29 : (char)(imm) + 13, \
- ((char)(imm)&0xF0) ? 30 : (char)(imm) + 14, \
- ((char)(imm)&0xF0) ? 31 : (char)(imm) + 15); })
+#define _mm_srli_si128(a, imm) \
+ (__m128i)__builtin_ia32_psrldqi128_byteshift((__v2di)(__m128i)(a), (int)(imm))
#define _mm_bsrli_si128(a, imm) \
- _mm_srli_si128((a), (imm))
+ (__m128i)__builtin_ia32_psrldqi128_byteshift((__v2di)(__m128i)(a), (int)(imm))
-/// \brief Right-shifts each of 16-bit values in the 128-bit integer vector
+/// Right-shifts each of 16-bit values in the 128-bit integer vector
/// operand by the specified number of bits. High-order bits are cleared.
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_ia32_psrlwi128((__v8hi)__a, __count);
}
-/// \brief Right-shifts each of 16-bit values in the 128-bit integer vector
+/// Right-shifts each of 16-bit values in the 128-bit integer vector
/// operand by the specified number of bits. High-order bits are cleared.
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_ia32_psrlw128((__v8hi)__a, (__v8hi)__count);
}
-/// \brief Right-shifts each of 32-bit values in the 128-bit integer vector
+/// Right-shifts each of 32-bit values in the 128-bit integer vector
/// operand by the specified number of bits. High-order bits are cleared.
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_ia32_psrldi128((__v4si)__a, __count);
}
-/// \brief Right-shifts each of 32-bit values in the 128-bit integer vector
+/// Right-shifts each of 32-bit values in the 128-bit integer vector
/// operand by the specified number of bits. High-order bits are cleared.
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_ia32_psrld128((__v4si)__a, (__v4si)__count);
}
-/// \brief Right-shifts each of 64-bit values in the 128-bit integer vector
+/// Right-shifts each of 64-bit values in the 128-bit integer vector
/// operand by the specified number of bits. High-order bits are cleared.
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_psrlqi128((__v2di)__a, __count);
}
-/// \brief Right-shifts each of 64-bit values in the 128-bit integer vector
+/// Right-shifts each of 64-bit values in the 128-bit integer vector
/// operand by the specified number of bits. High-order bits are cleared.
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_psrlq128((__v2di)__a, (__v2di)__count);
}
-/// \brief Compares each of the corresponding 8-bit values of the 128-bit
-/// integer vectors for equality. Each comparison yields 0h for false, FFh
+/// Compares each of the corresponding 8-bit values of the 128-bit
+/// integer vectors for equality. Each comparison yields 0x0 for false, 0xFF
/// for true.
///
/// \headerfile <x86intrin.h>
return (__m128i)((__v16qi)__a == (__v16qi)__b);
}
-/// \brief Compares each of the corresponding 16-bit values of the 128-bit
-/// integer vectors for equality. Each comparison yields 0h for false, FFFFh
-/// for true.
+/// Compares each of the corresponding 16-bit values of the 128-bit
+/// integer vectors for equality. Each comparison yields 0x0 for false,
+/// 0xFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)((__v8hi)__a == (__v8hi)__b);
}
-/// \brief Compares each of the corresponding 32-bit values of the 128-bit
-/// integer vectors for equality. Each comparison yields 0h for false,
-/// FFFFFFFFh for true.
+/// Compares each of the corresponding 32-bit values of the 128-bit
+/// integer vectors for equality. Each comparison yields 0x0 for false,
+/// 0xFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)((__v4si)__a == (__v4si)__b);
}
-/// \brief Compares each of the corresponding signed 8-bit values of the 128-bit
+/// Compares each of the corresponding signed 8-bit values of the 128-bit
/// integer vectors to determine if the values in the first operand are
-/// greater than those in the second operand. Each comparison yields 0h for
-/// false, FFh for true.
+/// greater than those in the second operand. Each comparison yields 0x0 for
+/// false, 0xFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)((__v16qs)__a > (__v16qs)__b);
}
-/// \brief Compares each of the corresponding signed 16-bit values of the
+/// Compares each of the corresponding signed 16-bit values of the
/// 128-bit integer vectors to determine if the values in the first operand
/// are greater than those in the second operand.
///
-/// Each comparison yields 0h for false, FFFFh for true.
+/// Each comparison yields 0x0 for false, 0xFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)((__v8hi)__a > (__v8hi)__b);
}
-/// \brief Compares each of the corresponding signed 32-bit values of the
+/// Compares each of the corresponding signed 32-bit values of the
/// 128-bit integer vectors to determine if the values in the first operand
/// are greater than those in the second operand.
///
-/// Each comparison yields 0h for false, FFFFFFFFh for true.
+/// Each comparison yields 0x0 for false, 0xFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)((__v4si)__a > (__v4si)__b);
}
-/// \brief Compares each of the corresponding signed 8-bit values of the 128-bit
+/// Compares each of the corresponding signed 8-bit values of the 128-bit
/// integer vectors to determine if the values in the first operand are less
/// than those in the second operand.
///
-/// Each comparison yields 0h for false, FFh for true.
+/// Each comparison yields 0x0 for false, 0xFF for true.
///
/// \headerfile <x86intrin.h>
///
return _mm_cmpgt_epi8(__b, __a);
}
-/// \brief Compares each of the corresponding signed 16-bit values of the
+/// Compares each of the corresponding signed 16-bit values of the
/// 128-bit integer vectors to determine if the values in the first operand
/// are less than those in the second operand.
///
-/// Each comparison yields 0h for false, FFFFh for true.
+/// Each comparison yields 0x0 for false, 0xFFFF for true.
///
/// \headerfile <x86intrin.h>
///
return _mm_cmpgt_epi16(__b, __a);
}
-/// \brief Compares each of the corresponding signed 32-bit values of the
+/// Compares each of the corresponding signed 32-bit values of the
/// 128-bit integer vectors to determine if the values in the first operand
/// are less than those in the second operand.
///
-/// Each comparison yields 0h for false, FFFFFFFFh for true.
+/// Each comparison yields 0x0 for false, 0xFFFFFFFF for true.
///
/// \headerfile <x86intrin.h>
///
}
#ifdef __x86_64__
-/// \brief Converts a 64-bit signed integer value from the second operand into a
+/// Converts a 64-bit signed integer value from the second operand into a
/// double-precision value and returns it in the lower element of a [2 x
/// double] vector; the upper element of the returned vector is copied from
/// the upper element of the first operand.
return __a;
}
-/// \brief Converts the first (lower) element of a vector of [2 x double] into a
+/// Converts the first (lower) element of a vector of [2 x double] into a
/// 64-bit signed integer value, according to the current rounding mode.
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_cvtsd2si64((__v2df)__a);
}
-/// \brief Converts the first (lower) element of a vector of [2 x double] into a
+/// Converts the first (lower) element of a vector of [2 x double] into a
/// 64-bit signed integer value, truncating the result when it is inexact.
///
/// \headerfile <x86intrin.h>
}
#endif
-/// \brief Converts a vector of [4 x i32] into a vector of [4 x float].
+/// Converts a vector of [4 x i32] into a vector of [4 x float].
///
/// \headerfile <x86intrin.h>
///
static __inline__ __m128 __DEFAULT_FN_ATTRS
_mm_cvtepi32_ps(__m128i __a)
{
- return __builtin_ia32_cvtdq2ps((__v4si)__a);
+ return (__m128)__builtin_convertvector((__v4si)__a, __v4sf);
}
-/// \brief Converts a vector of [4 x float] into a vector of [4 x i32].
+/// Converts a vector of [4 x float] into a vector of [4 x i32].
///
/// \headerfile <x86intrin.h>
///
return (__m128i)__builtin_ia32_cvtps2dq((__v4sf)__a);
}
-/// \brief Converts a vector of [4 x float] into a vector of [4 x i32],
+/// Converts a vector of [4 x float] into a vector of [4 x i32],
/// truncating the result when it is inexact.
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_ia32_cvttps2dq((__v4sf)__a);
}
-/// \brief Returns a vector of [4 x i32] where the lowest element is the input
+/// Returns a vector of [4 x i32] where the lowest element is the input
/// operand and the remaining elements are zero.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_cvtsi32_si128(int __a)
{
- return (__m128i)(__v4si){ __a, 0, 0, 0 };
+ return __extension__ (__m128i)(__v4si){ __a, 0, 0, 0 };
}
#ifdef __x86_64__
-/// \brief Returns a vector of [2 x i64] where the lower element is the input
+/// Returns a vector of [2 x i64] where the lower element is the input
/// operand and the upper element is zero.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_cvtsi64_si128(long long __a)
{
- return (__m128i){ __a, 0 };
+ return __extension__ (__m128i)(__v2di){ __a, 0 };
}
#endif
-/// \brief Moves the least significant 32 bits of a vector of [4 x i32] to a
+/// Moves the least significant 32 bits of a vector of [4 x i32] to a
/// 32-bit signed integer value.
///
/// \headerfile <x86intrin.h>
}
#ifdef __x86_64__
-/// \brief Moves the least significant 64 bits of a vector of [2 x i64] to a
+/// Moves the least significant 64 bits of a vector of [2 x i64] to a
/// 64-bit signed integer value.
///
/// \headerfile <x86intrin.h>
}
#endif
-/// \brief Moves packed integer values from an aligned 128-bit memory location
+/// Moves packed integer values from an aligned 128-bit memory location
/// to elements in a 128-bit integer vector.
///
/// \headerfile <x86intrin.h>
return *__p;
}
-/// \brief Moves packed integer values from an unaligned 128-bit memory location
+/// Moves packed integer values from an unaligned 128-bit memory location
/// to elements in a 128-bit integer vector.
///
/// \headerfile <x86intrin.h>
/// A pointer to a memory location containing integer values.
/// \returns A 128-bit integer vector containing the moved values.
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_loadu_si128(__m128i const *__p)
+_mm_loadu_si128(__m128i_u const *__p)
{
struct __loadu_si128 {
- __m128i __v;
+ __m128i_u __v;
} __attribute__((__packed__, __may_alias__));
- return ((struct __loadu_si128*)__p)->__v;
+ return ((const struct __loadu_si128*)__p)->__v;
}
-/// \brief Returns a vector of [2 x i64] where the lower element is taken from
+/// Returns a vector of [2 x i64] where the lower element is taken from
/// the lower element of the operand, and the upper element is zero.
///
/// \headerfile <x86intrin.h>
/// \returns A 128-bit vector of [2 x i64]. The lower order bits contain the
/// moved value. The higher order bits are cleared.
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_loadl_epi64(__m128i const *__p)
+_mm_loadl_epi64(__m128i_u const *__p)
{
struct __mm_loadl_epi64_struct {
long long __u;
} __attribute__((__packed__, __may_alias__));
- return (__m128i) { ((struct __mm_loadl_epi64_struct*)__p)->__u, 0};
+ return __extension__ (__m128i) { ((const struct __mm_loadl_epi64_struct*)__p)->__u, 0};
}
-/// \brief Generates a 128-bit vector of [4 x i32] with unspecified content.
+/// Generates a 128-bit vector of [4 x i32] with unspecified content.
/// This could be used as an argument to another intrinsic function where the
/// argument is required but the value is not actually used.
///
return (__m128i)__builtin_ia32_undef128();
}
-/// \brief Initializes both 64-bit values in a 128-bit vector of [2 x i64] with
+/// Initializes both 64-bit values in a 128-bit vector of [2 x i64] with
/// the specified 64-bit integer values.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_set_epi64x(long long __q1, long long __q0)
{
- return (__m128i){ __q0, __q1 };
+ return __extension__ (__m128i)(__v2di){ __q0, __q1 };
}
-/// \brief Initializes both 64-bit values in a 128-bit vector of [2 x i64] with
+/// Initializes both 64-bit values in a 128-bit vector of [2 x i64] with
/// the specified 64-bit integer values.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_set_epi64(__m64 __q1, __m64 __q0)
{
- return (__m128i){ (long long)__q0, (long long)__q1 };
+ return _mm_set_epi64x((long long)__q1, (long long)__q0);
}
-/// \brief Initializes the 32-bit values in a 128-bit vector of [4 x i32] with
+/// Initializes the 32-bit values in a 128-bit vector of [4 x i32] with
/// the specified 32-bit integer values.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_set_epi32(int __i3, int __i2, int __i1, int __i0)
{
- return (__m128i)(__v4si){ __i0, __i1, __i2, __i3};
+ return __extension__ (__m128i)(__v4si){ __i0, __i1, __i2, __i3};
}
-/// \brief Initializes the 16-bit values in a 128-bit vector of [8 x i16] with
+/// Initializes the 16-bit values in a 128-bit vector of [8 x i16] with
/// the specified 16-bit integer values.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_set_epi16(short __w7, short __w6, short __w5, short __w4, short __w3, short __w2, short __w1, short __w0)
{
- return (__m128i)(__v8hi){ __w0, __w1, __w2, __w3, __w4, __w5, __w6, __w7 };
+ return __extension__ (__m128i)(__v8hi){ __w0, __w1, __w2, __w3, __w4, __w5, __w6, __w7 };
}
-/// \brief Initializes the 8-bit values in a 128-bit vector of [16 x i8] with
+/// Initializes the 8-bit values in a 128-bit vector of [16 x i8] with
/// the specified 8-bit integer values.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_set_epi8(char __b15, char __b14, char __b13, char __b12, char __b11, char __b10, char __b9, char __b8, char __b7, char __b6, char __b5, char __b4, char __b3, char __b2, char __b1, char __b0)
{
- return (__m128i)(__v16qi){ __b0, __b1, __b2, __b3, __b4, __b5, __b6, __b7, __b8, __b9, __b10, __b11, __b12, __b13, __b14, __b15 };
+ return __extension__ (__m128i)(__v16qi){ __b0, __b1, __b2, __b3, __b4, __b5, __b6, __b7, __b8, __b9, __b10, __b11, __b12, __b13, __b14, __b15 };
}
-/// \brief Initializes both values in a 128-bit integer vector with the
+/// Initializes both values in a 128-bit integer vector with the
/// specified 64-bit integer value.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_set1_epi64x(long long __q)
{
- return (__m128i){ __q, __q };
+ return _mm_set_epi64x(__q, __q);
}
-/// \brief Initializes both values in a 128-bit vector of [2 x i64] with the
+/// Initializes both values in a 128-bit vector of [2 x i64] with the
/// specified 64-bit value.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_set1_epi64(__m64 __q)
{
- return (__m128i){ (long long)__q, (long long)__q };
+ return _mm_set_epi64(__q, __q);
}
-/// \brief Initializes all values in a 128-bit vector of [4 x i32] with the
+/// Initializes all values in a 128-bit vector of [4 x i32] with the
/// specified 32-bit value.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_set1_epi32(int __i)
{
- return (__m128i)(__v4si){ __i, __i, __i, __i };
+ return _mm_set_epi32(__i, __i, __i, __i);
}
-/// \brief Initializes all values in a 128-bit vector of [8 x i16] with the
+/// Initializes all values in a 128-bit vector of [8 x i16] with the
/// specified 16-bit value.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_set1_epi16(short __w)
{
- return (__m128i)(__v8hi){ __w, __w, __w, __w, __w, __w, __w, __w };
+ return _mm_set_epi16(__w, __w, __w, __w, __w, __w, __w, __w);
}
-/// \brief Initializes all values in a 128-bit vector of [16 x i8] with the
+/// Initializes all values in a 128-bit vector of [16 x i8] with the
/// specified 8-bit value.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_set1_epi8(char __b)
{
- return (__m128i)(__v16qi){ __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b };
+ return _mm_set_epi8(__b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b);
}
-/// \brief Constructs a 128-bit integer vector, initialized in reverse order
+/// Constructs a 128-bit integer vector, initialized in reverse order
/// with the specified 64-bit integral values.
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c> VPUNPCKLQDQ / PUNPCKLQDQ </c>
-/// instruction.
+/// This intrinsic does not correspond to a specific instruction.
///
/// \param __q0
/// A 64-bit integral value used to initialize the lower 64 bits of the
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_setr_epi64(__m64 __q0, __m64 __q1)
{
- return (__m128i){ (long long)__q0, (long long)__q1 };
+ return _mm_set_epi64(__q1, __q0);
}
-/// \brief Constructs a 128-bit integer vector, initialized in reverse order
+/// Constructs a 128-bit integer vector, initialized in reverse order
/// with the specified 32-bit integral values.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_setr_epi32(int __i0, int __i1, int __i2, int __i3)
{
- return (__m128i)(__v4si){ __i0, __i1, __i2, __i3};
+ return _mm_set_epi32(__i3, __i2, __i1, __i0);
}
-/// \brief Constructs a 128-bit integer vector, initialized in reverse order
+/// Constructs a 128-bit integer vector, initialized in reverse order
/// with the specified 16-bit integral values.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_setr_epi16(short __w0, short __w1, short __w2, short __w3, short __w4, short __w5, short __w6, short __w7)
{
- return (__m128i)(__v8hi){ __w0, __w1, __w2, __w3, __w4, __w5, __w6, __w7 };
+ return _mm_set_epi16(__w7, __w6, __w5, __w4, __w3, __w2, __w1, __w0);
}
-/// \brief Constructs a 128-bit integer vector, initialized in reverse order
+/// Constructs a 128-bit integer vector, initialized in reverse order
/// with the specified 8-bit integral values.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_setr_epi8(char __b0, char __b1, char __b2, char __b3, char __b4, char __b5, char __b6, char __b7, char __b8, char __b9, char __b10, char __b11, char __b12, char __b13, char __b14, char __b15)
{
- return (__m128i)(__v16qi){ __b0, __b1, __b2, __b3, __b4, __b5, __b6, __b7, __b8, __b9, __b10, __b11, __b12, __b13, __b14, __b15 };
+ return _mm_set_epi8(__b15, __b14, __b13, __b12, __b11, __b10, __b9, __b8, __b7, __b6, __b5, __b4, __b3, __b2, __b1, __b0);
}
-/// \brief Creates a 128-bit integer vector initialized to zero.
+/// Creates a 128-bit integer vector initialized to zero.
///
/// \headerfile <x86intrin.h>
///
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_setzero_si128(void)
{
- return (__m128i){ 0LL, 0LL };
+ return __extension__ (__m128i)(__v2di){ 0LL, 0LL };
}
-/// \brief Stores a 128-bit integer vector to a memory location aligned on a
+/// Stores a 128-bit integer vector to a memory location aligned on a
/// 128-bit boundary.
///
/// \headerfile <x86intrin.h>
*__p = __b;
}
-/// \brief Stores a 128-bit integer vector to an unaligned memory location.
+/// Stores a 128-bit integer vector to an unaligned memory location.
///
/// \headerfile <x86intrin.h>
///
/// \param __b
/// A 128-bit integer vector containing the values to be moved.
static __inline__ void __DEFAULT_FN_ATTRS
-_mm_storeu_si128(__m128i *__p, __m128i __b)
+_mm_storeu_si128(__m128i_u *__p, __m128i __b)
{
struct __storeu_si128 {
- __m128i __v;
+ __m128i_u __v;
} __attribute__((__packed__, __may_alias__));
((struct __storeu_si128*)__p)->__v = __b;
}
-/// \brief Moves bytes selected by the mask from the first operand to the
+/// Stores a 64-bit integer value from the low element of a 128-bit integer
+/// vector.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VMOVQ / MOVQ </c> instruction.
+///
+/// \param __p
+/// A pointer to a 64-bit memory location. The address of the memory
+/// location does not have to be algned.
+/// \param __b
+/// A 128-bit integer vector containing the value to be stored.
+static __inline__ void __DEFAULT_FN_ATTRS
+_mm_storeu_si64(void *__p, __m128i __b)
+{
+ struct __storeu_si64 {
+ long long __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_si64*)__p)->__v = ((__v2di)__b)[0];
+}
+
+/// Stores a 32-bit integer value from the low element of a 128-bit integer
+/// vector.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VMOVD / MOVD </c> instruction.
+///
+/// \param __p
+/// A pointer to a 32-bit memory location. The address of the memory
+/// location does not have to be aligned.
+/// \param __b
+/// A 128-bit integer vector containing the value to be stored.
+static __inline__ void __DEFAULT_FN_ATTRS
+_mm_storeu_si32(void *__p, __m128i __b)
+{
+ struct __storeu_si32 {
+ int __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_si32*)__p)->__v = ((__v4si)__b)[0];
+}
+
+/// Stores a 16-bit integer value from the low element of a 128-bit integer
+/// vector.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic does not correspond to a specific instruction.
+///
+/// \param __p
+/// A pointer to a 16-bit memory location. The address of the memory
+/// location does not have to be aligned.
+/// \param __b
+/// A 128-bit integer vector containing the value to be stored.
+static __inline__ void __DEFAULT_FN_ATTRS
+_mm_storeu_si16(void *__p, __m128i __b)
+{
+ struct __storeu_si16 {
+ short __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_si16*)__p)->__v = ((__v8hi)__b)[0];
+}
+
+/// Moves bytes selected by the mask from the first operand to the
/// specified unaligned memory location. When a mask bit is 1, the
/// corresponding byte is written, otherwise it is not written.
///
-/// To minimize caching, the date is flagged as non-temporal (unlikely to be
+/// To minimize caching, the data is flagged as non-temporal (unlikely to be
/// used again soon). Exception and trap behavior for elements not selected
/// for storage to memory are implementation dependent.
///
__builtin_ia32_maskmovdqu((__v16qi)__d, (__v16qi)__n, __p);
}
-/// \brief Stores the lower 64 bits of a 128-bit integer vector of [2 x i64] to
+/// Stores the lower 64 bits of a 128-bit integer vector of [2 x i64] to
/// a memory location.
///
/// \headerfile <x86intrin.h>
/// A 128-bit integer vector of [2 x i64]. The lower 64 bits contain the
/// value to be stored.
static __inline__ void __DEFAULT_FN_ATTRS
-_mm_storel_epi64(__m128i *__p, __m128i __a)
+_mm_storel_epi64(__m128i_u *__p, __m128i __a)
{
struct __mm_storel_epi64_struct {
long long __u;
((struct __mm_storel_epi64_struct*)__p)->__u = __a[0];
}
-/// \brief Stores a 128-bit floating point vector of [2 x double] to a 128-bit
+/// Stores a 128-bit floating point vector of [2 x double] to a 128-bit
/// aligned memory location.
///
/// To minimize caching, the data is flagged as non-temporal (unlikely to be
__builtin_nontemporal_store((__v2df)__a, (__v2df*)__p);
}
-/// \brief Stores a 128-bit integer vector to a 128-bit aligned memory location.
+/// Stores a 128-bit integer vector to a 128-bit aligned memory location.
///
/// To minimize caching, the data is flagged as non-temporal (unlikely to be
/// used again soon).
__builtin_nontemporal_store((__v2di)__a, (__v2di*)__p);
}
-/// \brief Stores a 32-bit integer value in the specified memory location.
+/// Stores a 32-bit integer value in the specified memory location.
///
/// To minimize caching, the data is flagged as non-temporal (unlikely to be
/// used again soon).
/// A pointer to the 32-bit memory location used to store the value.
/// \param __a
/// A 32-bit integer containing the value to be stored.
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("sse2")))
_mm_stream_si32(int *__p, int __a)
{
__builtin_ia32_movnti(__p, __a);
}
#ifdef __x86_64__
-/// \brief Stores a 64-bit integer value in the specified memory location.
+/// Stores a 64-bit integer value in the specified memory location.
///
/// To minimize caching, the data is flagged as non-temporal (unlikely to be
/// used again soon).
/// A pointer to the 64-bit memory location used to store the value.
/// \param __a
/// A 64-bit integer containing the value to be stored.
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("sse2")))
_mm_stream_si64(long long *__p, long long __a)
{
__builtin_ia32_movnti64(__p, __a);
extern "C" {
#endif
-/// \brief The cache line containing \a __p is flushed and invalidated from all
+/// The cache line containing \a __p is flushed and invalidated from all
/// caches in the coherency domain.
///
/// \headerfile <x86intrin.h>
/// flushed.
void _mm_clflush(void const * __p);
-/// \brief Forces strong memory ordering (serialization) between load
+/// Forces strong memory ordering (serialization) between load
/// instructions preceding this instruction and load instructions following
/// this instruction, ensuring the system completes all previous loads before
/// executing subsequent loads.
///
void _mm_lfence(void);
-/// \brief Forces strong memory ordering (serialization) between load and store
+/// Forces strong memory ordering (serialization) between load and store
/// instructions preceding this instruction and load and store instructions
/// following this instruction, ensuring that the system completes all
/// previous memory accesses before executing subsequent memory accesses.
} // extern "C"
#endif
-/// \brief Converts 16-bit signed integers from both 128-bit integer vector
+/// Converts 16-bit signed integers from both 128-bit integer vector
/// operands into 8-bit signed integers, and packs the results into the
/// destination. Positive values greater than 0x7F are saturated to 0x7F.
/// Negative values less than 0x80 are saturated to 0x80.
return (__m128i)__builtin_ia32_packsswb128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Converts 32-bit signed integers from both 128-bit integer vector
+/// Converts 32-bit signed integers from both 128-bit integer vector
/// operands into 16-bit signed integers, and packs the results into the
/// destination. Positive values greater than 0x7FFF are saturated to 0x7FFF.
/// Negative values less than 0x8000 are saturated to 0x8000.
return (__m128i)__builtin_ia32_packssdw128((__v4si)__a, (__v4si)__b);
}
-/// \brief Converts 16-bit signed integers from both 128-bit integer vector
+/// Converts 16-bit signed integers from both 128-bit integer vector
/// operands into 8-bit unsigned integers, and packs the results into the
/// destination. Values greater than 0xFF are saturated to 0xFF. Values less
/// than 0x00 are saturated to 0x00.
return (__m128i)__builtin_ia32_packuswb128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Extracts 16 bits from a 128-bit integer vector of [8 x i16], using
+/// Extracts 16 bits from a 128-bit integer vector of [8 x i16], using
/// the immediate-value parameter as a selector.
///
/// \headerfile <x86intrin.h>
/// 111: assign values from bits [127:112] of \a __a.
/// \returns An integer, whose lower 16 bits are selected from the 128-bit
/// integer vector parameter and the remaining bits are assigned zeros.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_extract_epi16(__m128i __a, int __imm)
-{
- __v8hi __b = (__v8hi)__a;
- return (unsigned short)__b[__imm & 7];
-}
+#define _mm_extract_epi16(a, imm) \
+ (int)(unsigned short)__builtin_ia32_vec_ext_v8hi((__v8hi)(__m128i)(a), \
+ (int)(imm))
-/// \brief Constructs a 128-bit integer vector by first making a copy of the
+/// Constructs a 128-bit integer vector by first making a copy of the
/// 128-bit integer vector parameter, and then inserting the lower 16 bits
/// of an integer parameter into an offset specified by the immediate-value
/// parameter.
/// An immediate value specifying the bit offset in the result at which the
/// lower 16 bits of \a __b are written.
/// \returns A 128-bit integer vector containing the constructed values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_insert_epi16(__m128i __a, int __b, int __imm)
-{
- __v8hi __c = (__v8hi)__a;
- __c[__imm & 7] = __b;
- return (__m128i)__c;
-}
+#define _mm_insert_epi16(a, b, imm) \
+ (__m128i)__builtin_ia32_vec_set_v8hi((__v8hi)(__m128i)(a), (int)(b), \
+ (int)(imm))
-/// \brief Copies the values of the most significant bits from each 8-bit
+/// Copies the values of the most significant bits from each 8-bit
/// element in a 128-bit integer vector of [16 x i8] to create a 16-bit mask
/// value, zero-extends the value, and writes it to the destination.
///
return __builtin_ia32_pmovmskb128((__v16qi)__a);
}
-/// \brief Constructs a 128-bit integer vector by shuffling four 32-bit
+/// Constructs a 128-bit integer vector by shuffling four 32-bit
/// elements of a 128-bit integer vector parameter, using the immediate-value
/// parameter as a specifier.
///
/// 10: assign values from bits [95:64] of \a a. \n
/// 11: assign values from bits [127:96] of \a a.
/// \returns A 128-bit integer vector containing the shuffled values.
-#define _mm_shuffle_epi32(a, imm) __extension__ ({ \
- (__m128i)__builtin_shufflevector((__v4si)(__m128i)(a), \
- (__v4si)_mm_undefined_si128(), \
- ((imm) >> 0) & 0x3, ((imm) >> 2) & 0x3, \
- ((imm) >> 4) & 0x3, ((imm) >> 6) & 0x3); })
+#define _mm_shuffle_epi32(a, imm) \
+ (__m128i)__builtin_ia32_pshufd((__v4si)(__m128i)(a), (int)(imm))
-/// \brief Constructs a 128-bit integer vector by shuffling four lower 16-bit
+/// Constructs a 128-bit integer vector by shuffling four lower 16-bit
/// elements of a 128-bit integer vector of [8 x i16], using the immediate
/// value parameter as a specifier.
///
/// 10: assign values from bits [47:32] of \a a. \n
/// 11: assign values from bits [63:48] of \a a. \n
/// \returns A 128-bit integer vector containing the shuffled values.
-#define _mm_shufflelo_epi16(a, imm) __extension__ ({ \
- (__m128i)__builtin_shufflevector((__v8hi)(__m128i)(a), \
- (__v8hi)_mm_undefined_si128(), \
- ((imm) >> 0) & 0x3, ((imm) >> 2) & 0x3, \
- ((imm) >> 4) & 0x3, ((imm) >> 6) & 0x3, \
- 4, 5, 6, 7); })
-
-/// \brief Constructs a 128-bit integer vector by shuffling four upper 16-bit
+#define _mm_shufflelo_epi16(a, imm) \
+ (__m128i)__builtin_ia32_pshuflw((__v8hi)(__m128i)(a), (int)(imm))
+
+/// Constructs a 128-bit integer vector by shuffling four upper 16-bit
/// elements of a 128-bit integer vector of [8 x i16], using the immediate
/// value parameter as a specifier.
///
/// 10: assign values from bits [111:96] of \a a. \n
/// 11: assign values from bits [127:112] of \a a. \n
/// \returns A 128-bit integer vector containing the shuffled values.
-#define _mm_shufflehi_epi16(a, imm) __extension__ ({ \
- (__m128i)__builtin_shufflevector((__v8hi)(__m128i)(a), \
- (__v8hi)_mm_undefined_si128(), \
- 0, 1, 2, 3, \
- 4 + (((imm) >> 0) & 0x3), \
- 4 + (((imm) >> 2) & 0x3), \
- 4 + (((imm) >> 4) & 0x3), \
- 4 + (((imm) >> 6) & 0x3)); })
-
-/// \brief Unpacks the high-order (index 8-15) values from two 128-bit vectors
+#define _mm_shufflehi_epi16(a, imm) \
+ (__m128i)__builtin_ia32_pshufhw((__v8hi)(__m128i)(a), (int)(imm))
+
+/// Unpacks the high-order (index 8-15) values from two 128-bit vectors
/// of [16 x i8] and interleaves them into a 128-bit vector of [16 x i8].
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_shufflevector((__v16qi)__a, (__v16qi)__b, 8, 16+8, 9, 16+9, 10, 16+10, 11, 16+11, 12, 16+12, 13, 16+13, 14, 16+14, 15, 16+15);
}
-/// \brief Unpacks the high-order (index 4-7) values from two 128-bit vectors of
+/// Unpacks the high-order (index 4-7) values from two 128-bit vectors of
/// [8 x i16] and interleaves them into a 128-bit vector of [8 x i16].
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_shufflevector((__v8hi)__a, (__v8hi)__b, 4, 8+4, 5, 8+5, 6, 8+6, 7, 8+7);
}
-/// \brief Unpacks the high-order (index 2,3) values from two 128-bit vectors of
+/// Unpacks the high-order (index 2,3) values from two 128-bit vectors of
/// [4 x i32] and interleaves them into a 128-bit vector of [4 x i32].
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_shufflevector((__v4si)__a, (__v4si)__b, 2, 4+2, 3, 4+3);
}
-/// \brief Unpacks the high-order (odd-indexed) values from two 128-bit vectors
-/// of [2 x i64] and interleaves them into a 128-bit vector of [2 x i64].
+/// Unpacks the high-order 64-bit elements from two 128-bit vectors of
+/// [2 x i64] and interleaves them into a 128-bit vector of [2 x i64].
///
/// \headerfile <x86intrin.h>
///
return (__m128i)__builtin_shufflevector((__v2di)__a, (__v2di)__b, 1, 2+1);
}
-/// \brief Unpacks the low-order (index 0-7) values from two 128-bit vectors of
+/// Unpacks the low-order (index 0-7) values from two 128-bit vectors of
/// [16 x i8] and interleaves them into a 128-bit vector of [16 x i8].
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_shufflevector((__v16qi)__a, (__v16qi)__b, 0, 16+0, 1, 16+1, 2, 16+2, 3, 16+3, 4, 16+4, 5, 16+5, 6, 16+6, 7, 16+7);
}
-/// \brief Unpacks the low-order (index 0-3) values from each of the two 128-bit
+/// Unpacks the low-order (index 0-3) values from each of the two 128-bit
/// vectors of [8 x i16] and interleaves them into a 128-bit vector of
/// [8 x i16].
///
return (__m128i)__builtin_shufflevector((__v8hi)__a, (__v8hi)__b, 0, 8+0, 1, 8+1, 2, 8+2, 3, 8+3);
}
-/// \brief Unpacks the low-order (index 0,1) values from two 128-bit vectors of
+/// Unpacks the low-order (index 0,1) values from two 128-bit vectors of
/// [4 x i32] and interleaves them into a 128-bit vector of [4 x i32].
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_shufflevector((__v4si)__a, (__v4si)__b, 0, 4+0, 1, 4+1);
}
-/// \brief Unpacks the low-order 64-bit elements from two 128-bit vectors of
+/// Unpacks the low-order 64-bit elements from two 128-bit vectors of
/// [2 x i64] and interleaves them into a 128-bit vector of [2 x i64].
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_shufflevector((__v2di)__a, (__v2di)__b, 0, 2+0);
}
-/// \brief Returns the lower 64 bits of a 128-bit integer vector as a 64-bit
+/// Returns the lower 64 bits of a 128-bit integer vector as a 64-bit
/// integer.
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic has no corresponding instruction.
+/// This intrinsic corresponds to the <c> MOVDQ2Q </c> instruction.
///
/// \param __a
/// A 128-bit integer vector operand. The lower 64 bits are moved to the
return (__m64)__a[0];
}
-/// \brief Moves the 64-bit operand to a 128-bit integer vector, zeroing the
+/// Moves the 64-bit operand to a 128-bit integer vector, zeroing the
/// upper bits.
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c> VMOVQ / MOVQ / MOVD </c> instruction.
+/// This intrinsic corresponds to the <c> MOVD+VMOVQ </c> instruction.
///
/// \param __a
/// A 64-bit value.
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_movpi64_epi64(__m64 __a)
{
- return (__m128i){ (long long)__a, 0 };
+ return __extension__ (__m128i)(__v2di){ (long long)__a, 0 };
}
-/// \brief Moves the lower 64 bits of a 128-bit integer vector to a 128-bit
+/// Moves the lower 64 bits of a 128-bit integer vector to a 128-bit
/// integer vector, zeroing the upper bits.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_move_epi64(__m128i __a)
{
- return __builtin_shufflevector((__v2di)__a, (__m128i){ 0 }, 0, 2);
+ return __builtin_shufflevector((__v2di)__a, _mm_setzero_si128(), 0, 2);
}
-/// \brief Unpacks the high-order (odd-indexed) values from two 128-bit vectors
-/// of [2 x double] and interleaves them into a 128-bit vector of [2 x
+/// Unpacks the high-order 64-bit elements from two 128-bit vectors of
+/// [2 x double] and interleaves them into a 128-bit vector of [2 x
/// double].
///
/// \headerfile <x86intrin.h>
return __builtin_shufflevector((__v2df)__a, (__v2df)__b, 1, 2+1);
}
-/// \brief Unpacks the low-order (even-indexed) values from two 128-bit vectors
+/// Unpacks the low-order 64-bit elements from two 128-bit vectors
/// of [2 x double] and interleaves them into a 128-bit vector of [2 x
/// double].
///
return __builtin_shufflevector((__v2df)__a, (__v2df)__b, 0, 2+0);
}
-/// \brief Extracts the sign bits of the double-precision values in the 128-bit
+/// Extracts the sign bits of the double-precision values in the 128-bit
/// vector of [2 x double], zero-extends the value, and writes it to the
/// low-order bits of the destination.
///
}
-/// \brief Constructs a 128-bit floating-point vector of [2 x double] from two
+/// Constructs a 128-bit floating-point vector of [2 x double] from two
/// 128-bit vector parameters of [2 x double], using the immediate-value
/// parameter as a specifier.
///
/// A 128-bit vector of [2 x double].
/// \param i
/// An 8-bit immediate value. The least significant two bits specify which
-/// elements to copy from a and b: \n
-/// Bit[0] = 0: lower element of a copied to lower element of result. \n
-/// Bit[0] = 1: upper element of a copied to lower element of result. \n
+/// elements to copy from \a a and \a b: \n
+/// Bit[0] = 0: lower element of \a a copied to lower element of result. \n
+/// Bit[0] = 1: upper element of \a a copied to lower element of result. \n
/// Bit[1] = 0: lower element of \a b copied to upper element of result. \n
/// Bit[1] = 1: upper element of \a b copied to upper element of result. \n
/// \returns A 128-bit vector of [2 x double] containing the shuffled values.
-#define _mm_shuffle_pd(a, b, i) __extension__ ({ \
- (__m128d)__builtin_shufflevector((__v2df)(__m128d)(a), (__v2df)(__m128d)(b), \
- 0 + (((i) >> 0) & 0x1), \
- 2 + (((i) >> 1) & 0x1)); })
+#define _mm_shuffle_pd(a, b, i) \
+ (__m128d)__builtin_ia32_shufpd((__v2df)(__m128d)(a), (__v2df)(__m128d)(b), \
+ (int)(i))
-/// \brief Casts a 128-bit floating-point vector of [2 x double] into a 128-bit
+/// Casts a 128-bit floating-point vector of [2 x double] into a 128-bit
/// floating-point vector of [4 x float].
///
/// \headerfile <x86intrin.h>
return (__m128)__a;
}
-/// \brief Casts a 128-bit floating-point vector of [2 x double] into a 128-bit
+/// Casts a 128-bit floating-point vector of [2 x double] into a 128-bit
/// integer vector.
///
/// \headerfile <x86intrin.h>
return (__m128i)__a;
}
-/// \brief Casts a 128-bit floating-point vector of [4 x float] into a 128-bit
+/// Casts a 128-bit floating-point vector of [4 x float] into a 128-bit
/// floating-point vector of [2 x double].
///
/// \headerfile <x86intrin.h>
return (__m128d)__a;
}
-/// \brief Casts a 128-bit floating-point vector of [4 x float] into a 128-bit
+/// Casts a 128-bit floating-point vector of [4 x float] into a 128-bit
/// integer vector.
///
/// \headerfile <x86intrin.h>
return (__m128i)__a;
}
-/// \brief Casts a 128-bit integer vector into a 128-bit floating-point vector
+/// Casts a 128-bit integer vector into a 128-bit floating-point vector
/// of [4 x float].
///
/// \headerfile <x86intrin.h>
return (__m128)__a;
}
-/// \brief Casts a 128-bit integer vector into a 128-bit floating-point vector
+/// Casts a 128-bit integer vector into a 128-bit floating-point vector
/// of [2 x double].
///
/// \headerfile <x86intrin.h>
extern "C" {
#endif
-/// \brief Indicates that a spin loop is being executed for the purposes of
+/// Indicates that a spin loop is being executed for the purposes of
/// optimizing power consumption during the loop.
///
/// \headerfile <x86intrin.h>
} // extern "C"
#endif
#undef __DEFAULT_FN_ATTRS
+#undef __DEFAULT_FN_ATTRS_MMX
#define _MM_SHUFFLE2(x, y) (((x) << 1) | (y))
-#define _MM_DENORMALS_ZERO_ON (0x0040)
-#define _MM_DENORMALS_ZERO_OFF (0x0000)
+#define _MM_DENORMALS_ZERO_ON (0x0040U)
+#define _MM_DENORMALS_ZERO_OFF (0x0000U)
-#define _MM_DENORMALS_ZERO_MASK (0x0040)
+#define _MM_DENORMALS_ZERO_MASK (0x0040U)
#define _MM_GET_DENORMALS_ZERO_MODE() (_mm_getcsr() & _MM_DENORMALS_ZERO_MASK)
#define _MM_SET_DENORMALS_ZERO_MODE(x) (_mm_setcsr((_mm_getcsr() & ~_MM_DENORMALS_ZERO_MASK) | (x)))
--- /dev/null
+/*===------------------ enqcmdintrin.h - enqcmd intrinsics -----------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <enqcmdintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __ENQCMDINTRIN_H
+#define __ENQCMDINTRIN_H
+
+/* Define the default attributes for the functions in this file */
+#define _DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("enqcmd")))
+
+/// Reads 64-byte command pointed by \a __src, formats 64-byte enqueue store
+/// data, and performs 64-byte enqueue store to memory pointed by \a __dst.
+/// This intrinsics may only be used in User mode.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsics corresponds to the <c> ENQCMD </c> instruction.
+///
+/// \param __dst
+/// Pointer to the destination of the enqueue store.
+/// \param __src
+/// Pointer to 64-byte command data.
+/// \returns If the command data is successfully written to \a __dst then 0 is
+/// returned. Otherwise 1 is returned.
+static __inline__ int _DEFAULT_FN_ATTRS
+_enqcmd (void *__dst, const void *__src)
+{
+ return __builtin_ia32_enqcmd(__dst, __src);
+}
+
+/// Reads 64-byte command pointed by \a __src, formats 64-byte enqueue store
+/// data, and performs 64-byte enqueue store to memory pointed by \a __dst
+/// This intrinsic may only be used in Privileged mode.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsics corresponds to the <c> ENQCMDS </c> instruction.
+///
+/// \param __dst
+/// Pointer to the destination of the enqueue store.
+/// \param __src
+/// Pointer to 64-byte command data.
+/// \returns If the command data is successfully written to \a __dst then 0 is
+/// returned. Otherwise 1 is returned.
+static __inline__ int _DEFAULT_FN_ATTRS
+_enqcmds (void *__dst, const void *__src)
+{
+ return __builtin_ia32_enqcmds(__dst, __src);
+}
+
+#undef _DEFAULT_FN_ATTRS
+
+#endif /* __ENQCMDINTRIN_H */
--- /dev/null
+/*===---- f16cintrin.h - F16C intrinsics -----------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#if !defined __IMMINTRIN_H
+#error "Never use <f16cintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __F16CINTRIN_H
+#define __F16CINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS128 \
+ __attribute__((__always_inline__, __nodebug__, __target__("f16c"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 \
+ __attribute__((__always_inline__, __nodebug__, __target__("f16c"), __min_vector_width__(256)))
+
+/* NOTE: Intel documents the 128-bit versions of these as being in emmintrin.h,
+ * but that's because icc can emulate these without f16c using a library call.
+ * Since we don't do that let's leave these in f16cintrin.h.
+ */
+
+/// Converts a 16-bit half-precision float value into a 32-bit float
+/// value.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTPH2PS </c> instruction.
+///
+/// \param __a
+/// A 16-bit half-precision float value.
+/// \returns The converted 32-bit float value.
+static __inline float __DEFAULT_FN_ATTRS128
+_cvtsh_ss(unsigned short __a)
+{
+ __v8hi __v = {(short)__a, 0, 0, 0, 0, 0, 0, 0};
+ __v4sf __r = __builtin_ia32_vcvtph2ps(__v);
+ return __r[0];
+}
+
+/// Converts a 32-bit single-precision float value to a 16-bit
+/// half-precision float value.
+///
+/// \headerfile <x86intrin.h>
+///
+/// \code
+/// unsigned short _cvtss_sh(float a, const int imm);
+/// \endcode
+///
+/// This intrinsic corresponds to the <c> VCVTPS2PH </c> instruction.
+///
+/// \param a
+/// A 32-bit single-precision float value to be converted to a 16-bit
+/// half-precision float value.
+/// \param imm
+/// An immediate value controlling rounding using bits [2:0]: \n
+/// 000: Nearest \n
+/// 001: Down \n
+/// 010: Up \n
+/// 011: Truncate \n
+/// 1XX: Use MXCSR.RC for rounding
+/// \returns The converted 16-bit half-precision float value.
+#define _cvtss_sh(a, imm) \
+ (unsigned short)(((__v8hi)__builtin_ia32_vcvtps2ph((__v4sf){a, 0, 0, 0}, \
+ (imm)))[0])
+
+/// Converts a 128-bit vector containing 32-bit float values into a
+/// 128-bit vector containing 16-bit half-precision float values.
+///
+/// \headerfile <x86intrin.h>
+///
+/// \code
+/// __m128i _mm_cvtps_ph(__m128 a, const int imm);
+/// \endcode
+///
+/// This intrinsic corresponds to the <c> VCVTPS2PH </c> instruction.
+///
+/// \param a
+/// A 128-bit vector containing 32-bit float values.
+/// \param imm
+/// An immediate value controlling rounding using bits [2:0]: \n
+/// 000: Nearest \n
+/// 001: Down \n
+/// 010: Up \n
+/// 011: Truncate \n
+/// 1XX: Use MXCSR.RC for rounding
+/// \returns A 128-bit vector containing converted 16-bit half-precision float
+/// values. The lower 64 bits are used to store the converted 16-bit
+/// half-precision floating-point values.
+#define _mm_cvtps_ph(a, imm) \
+ (__m128i)__builtin_ia32_vcvtps2ph((__v4sf)(__m128)(a), (imm))
+
+/// Converts a 128-bit vector containing 16-bit half-precision float
+/// values into a 128-bit vector containing 32-bit float values.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTPH2PS </c> instruction.
+///
+/// \param __a
+/// A 128-bit vector containing 16-bit half-precision float values. The lower
+/// 64 bits are used in the conversion.
+/// \returns A 128-bit vector of [4 x float] containing converted float values.
+static __inline __m128 __DEFAULT_FN_ATTRS128
+_mm_cvtph_ps(__m128i __a)
+{
+ return (__m128)__builtin_ia32_vcvtph2ps((__v8hi)__a);
+}
+
+/// Converts a 256-bit vector of [8 x float] into a 128-bit vector
+/// containing 16-bit half-precision float values.
+///
+/// \headerfile <x86intrin.h>
+///
+/// \code
+/// __m128i _mm256_cvtps_ph(__m256 a, const int imm);
+/// \endcode
+///
+/// This intrinsic corresponds to the <c> VCVTPS2PH </c> instruction.
+///
+/// \param a
+/// A 256-bit vector containing 32-bit single-precision float values to be
+/// converted to 16-bit half-precision float values.
+/// \param imm
+/// An immediate value controlling rounding using bits [2:0]: \n
+/// 000: Nearest \n
+/// 001: Down \n
+/// 010: Up \n
+/// 011: Truncate \n
+/// 1XX: Use MXCSR.RC for rounding
+/// \returns A 128-bit vector containing the converted 16-bit half-precision
+/// float values.
+#define _mm256_cvtps_ph(a, imm) \
+ (__m128i)__builtin_ia32_vcvtps2ph256((__v8sf)(__m256)(a), (imm))
+
+/// Converts a 128-bit vector containing 16-bit half-precision float
+/// values into a 256-bit vector of [8 x float].
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> VCVTPH2PS </c> instruction.
+///
+/// \param __a
+/// A 128-bit vector containing 16-bit half-precision float values to be
+/// converted to 32-bit single-precision float values.
+/// \returns A vector of [8 x float] containing the converted 32-bit
+/// single-precision float values.
+static __inline __m256 __DEFAULT_FN_ATTRS256
+_mm256_cvtph_ps(__m128i __a)
+{
+ return (__m256)__builtin_ia32_vcvtph2ps256((__v8hi)__a);
+}
+
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
+
+#endif /* __F16CINTRIN_H */
/*===---- float.h - Characteristics of floating point types ----------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
-#ifndef __FLOAT_H
-#define __FLOAT_H
+#ifndef __CLANG_FLOAT_H
+#define __CLANG_FLOAT_H
/* If we're on MinGW, fall back to the system's float.h, which might have
* additional definitions provided for Windows.
# undef FLT_MANT_DIG
# undef DBL_MANT_DIG
# undef LDBL_MANT_DIG
-# if __STDC_VERSION__ >= 199901L || !defined(__STRICT_ANSI__)
+# if __STDC_VERSION__ >= 199901L || !defined(__STRICT_ANSI__) || __cplusplus >= 201103L
# undef DECIMAL_DIG
# endif
# undef FLT_DIG
# undef FLT_MIN
# undef DBL_MIN
# undef LDBL_MIN
-# if __STDC_VERSION__ >= 201112L || !defined(__STRICT_ANSI__)
+# if __STDC_VERSION__ >= 201112L || !defined(__STRICT_ANSI__) || __cplusplus >= 201703L
# undef FLT_TRUE_MIN
# undef DBL_TRUE_MIN
# undef LDBL_TRUE_MIN
# undef FLT_DECIMAL_DIG
# undef DBL_DECIMAL_DIG
# undef LDBL_DECIMAL_DIG
+# undef FLT_HAS_SUBNORM
+# undef DBL_HAS_SUBNORM
+# undef LDBL_HAS_SUBNORM
# endif
#endif
#define DBL_MANT_DIG __DBL_MANT_DIG__
#define LDBL_MANT_DIG __LDBL_MANT_DIG__
-#if __STDC_VERSION__ >= 199901L || !defined(__STRICT_ANSI__)
+#if __STDC_VERSION__ >= 199901L || !defined(__STRICT_ANSI__) || __cplusplus >= 201103L
# define DECIMAL_DIG __DECIMAL_DIG__
#endif
#define DBL_MIN __DBL_MIN__
#define LDBL_MIN __LDBL_MIN__
-#if __STDC_VERSION__ >= 201112L || !defined(__STRICT_ANSI__)
+#if __STDC_VERSION__ >= 201112L || !defined(__STRICT_ANSI__) || __cplusplus >= 201703L
# define FLT_TRUE_MIN __FLT_DENORM_MIN__
# define DBL_TRUE_MIN __DBL_DENORM_MIN__
# define LDBL_TRUE_MIN __LDBL_DENORM_MIN__
# define FLT_DECIMAL_DIG __FLT_DECIMAL_DIG__
# define DBL_DECIMAL_DIG __DBL_DECIMAL_DIG__
# define LDBL_DECIMAL_DIG __LDBL_DECIMAL_DIG__
+# define FLT_HAS_SUBNORM __FLT_HAS_DENORM__
+# define DBL_HAS_SUBNORM __DBL_HAS_DENORM__
+# define LDBL_HAS_SUBNORM __LDBL_HAS_DENORM__
#endif
-#endif /* __FLOAT_H */
+#ifdef __STDC_WANT_IEC_60559_TYPES_EXT__
+# define FLT16_MANT_DIG __FLT16_MANT_DIG__
+# define FLT16_DECIMAL_DIG __FLT16_DECIMAL_DIG__
+# define FLT16_DIG __FLT16_DIG__
+# define FLT16_MIN_EXP __FLT16_MIN_EXP__
+# define FLT16_MIN_10_EXP __FLT16_MIN_10_EXP__
+# define FLT16_MAX_EXP __FLT16_MAX_EXP__
+# define FLT16_MAX_10_EXP __FLT16_MAX_10_EXP__
+# define FLT16_MAX __FLT16_MAX__
+# define FLT16_EPSILON __FLT16_EPSILON__
+# define FLT16_MIN __FLT16_MIN__
+# define FLT16_TRUE_MIN __FLT16_TRUE_MIN__
+#endif /* __STDC_WANT_IEC_60559_TYPES_EXT__ */
+
+#endif /* __CLANG_FLOAT_H */
--- /dev/null
+/*===---- fma4intrin.h - FMA4 intrinsics -----------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __X86INTRIN_H
+#error "Never use <fma4intrin.h> directly; include <x86intrin.h> instead."
+#endif
+
+#ifndef __FMA4INTRIN_H
+#define __FMA4INTRIN_H
+
+#include <pmmintrin.h>
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("fma4"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("fma4"), __min_vector_width__(256)))
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_macc_ps(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_macc_pd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_macc_ss(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddss((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_macc_sd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddsd((__v2df)__A, (__v2df)__B, (__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_msub_ps(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddps((__v4sf)__A, (__v4sf)__B, -(__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_msub_pd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddpd((__v2df)__A, (__v2df)__B, -(__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_msub_ss(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddss((__v4sf)__A, (__v4sf)__B, -(__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_msub_sd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddsd((__v2df)__A, (__v2df)__B, -(__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_nmacc_ps(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddps(-(__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_nmacc_pd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddpd(-(__v2df)__A, (__v2df)__B, (__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_nmacc_ss(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddss(-(__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_nmacc_sd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddsd(-(__v2df)__A, (__v2df)__B, (__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_nmsub_ps(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddps(-(__v4sf)__A, (__v4sf)__B, -(__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_nmsub_pd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddpd(-(__v2df)__A, (__v2df)__B, -(__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_nmsub_ss(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddss(-(__v4sf)__A, (__v4sf)__B, -(__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_nmsub_sd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddsd(-(__v2df)__A, (__v2df)__B, -(__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_maddsub_ps(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddsubps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_maddsub_pd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddsubpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_msubadd_ps(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddsubps((__v4sf)__A, (__v4sf)__B, -(__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_msubadd_pd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddsubpd((__v2df)__A, (__v2df)__B, -(__v2df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_macc_ps(__m256 __A, __m256 __B, __m256 __C)
+{
+ return (__m256)__builtin_ia32_vfmaddps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
+_mm256_macc_pd(__m256d __A, __m256d __B, __m256d __C)
+{
+ return (__m256d)__builtin_ia32_vfmaddpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_msub_ps(__m256 __A, __m256 __B, __m256 __C)
+{
+ return (__m256)__builtin_ia32_vfmaddps256((__v8sf)__A, (__v8sf)__B, -(__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
+_mm256_msub_pd(__m256d __A, __m256d __B, __m256d __C)
+{
+ return (__m256d)__builtin_ia32_vfmaddpd256((__v4df)__A, (__v4df)__B, -(__v4df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_nmacc_ps(__m256 __A, __m256 __B, __m256 __C)
+{
+ return (__m256)__builtin_ia32_vfmaddps256(-(__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
+_mm256_nmacc_pd(__m256d __A, __m256d __B, __m256d __C)
+{
+ return (__m256d)__builtin_ia32_vfmaddpd256(-(__v4df)__A, (__v4df)__B, (__v4df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_nmsub_ps(__m256 __A, __m256 __B, __m256 __C)
+{
+ return (__m256)__builtin_ia32_vfmaddps256(-(__v8sf)__A, (__v8sf)__B, -(__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
+_mm256_nmsub_pd(__m256d __A, __m256d __B, __m256d __C)
+{
+ return (__m256d)__builtin_ia32_vfmaddpd256(-(__v4df)__A, (__v4df)__B, -(__v4df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_maddsub_ps(__m256 __A, __m256 __B, __m256 __C)
+{
+ return (__m256)__builtin_ia32_vfmaddsubps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
+_mm256_maddsub_pd(__m256d __A, __m256d __B, __m256d __C)
+{
+ return (__m256d)__builtin_ia32_vfmaddsubpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_msubadd_ps(__m256 __A, __m256 __B, __m256 __C)
+{
+ return (__m256)__builtin_ia32_vfmaddsubps256((__v8sf)__A, (__v8sf)__B, -(__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
+_mm256_msubadd_pd(__m256d __A, __m256d __B, __m256d __C)
+{
+ return (__m256d)__builtin_ia32_vfmaddsubpd256((__v4df)__A, (__v4df)__B, -(__v4df)__C);
+}
+
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
+
+#endif /* __FMA4INTRIN_H */
--- /dev/null
+/*===---- fmaintrin.h - FMA intrinsics -------------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <fmaintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __FMAINTRIN_H
+#define __FMAINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("fma"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("fma"), __min_vector_width__(256)))
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_fmadd_ps(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_fmadd_pd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_fmadd_ss(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddss3((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_fmadd_sd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddsd3((__v2df)__A, (__v2df)__B, (__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_fmsub_ps(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddps((__v4sf)__A, (__v4sf)__B, -(__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_fmsub_pd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddpd((__v2df)__A, (__v2df)__B, -(__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_fmsub_ss(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddss3((__v4sf)__A, (__v4sf)__B, -(__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_fmsub_sd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddsd3((__v2df)__A, (__v2df)__B, -(__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_fnmadd_ps(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddps(-(__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_fnmadd_pd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddpd(-(__v2df)__A, (__v2df)__B, (__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_fnmadd_ss(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddss3((__v4sf)__A, -(__v4sf)__B, (__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_fnmadd_sd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddsd3((__v2df)__A, -(__v2df)__B, (__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_fnmsub_ps(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddps(-(__v4sf)__A, (__v4sf)__B, -(__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_fnmsub_pd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddpd(-(__v2df)__A, (__v2df)__B, -(__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_fnmsub_ss(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddss3((__v4sf)__A, -(__v4sf)__B, -(__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_fnmsub_sd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddsd3((__v2df)__A, -(__v2df)__B, -(__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_fmaddsub_ps(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddsubps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_fmaddsub_pd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddsubpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_fmsubadd_ps(__m128 __A, __m128 __B, __m128 __C)
+{
+ return (__m128)__builtin_ia32_vfmaddsubps((__v4sf)__A, (__v4sf)__B, -(__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_fmsubadd_pd(__m128d __A, __m128d __B, __m128d __C)
+{
+ return (__m128d)__builtin_ia32_vfmaddsubpd((__v2df)__A, (__v2df)__B, -(__v2df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_fmadd_ps(__m256 __A, __m256 __B, __m256 __C)
+{
+ return (__m256)__builtin_ia32_vfmaddps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
+_mm256_fmadd_pd(__m256d __A, __m256d __B, __m256d __C)
+{
+ return (__m256d)__builtin_ia32_vfmaddpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_fmsub_ps(__m256 __A, __m256 __B, __m256 __C)
+{
+ return (__m256)__builtin_ia32_vfmaddps256((__v8sf)__A, (__v8sf)__B, -(__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
+_mm256_fmsub_pd(__m256d __A, __m256d __B, __m256d __C)
+{
+ return (__m256d)__builtin_ia32_vfmaddpd256((__v4df)__A, (__v4df)__B, -(__v4df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_fnmadd_ps(__m256 __A, __m256 __B, __m256 __C)
+{
+ return (__m256)__builtin_ia32_vfmaddps256(-(__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
+_mm256_fnmadd_pd(__m256d __A, __m256d __B, __m256d __C)
+{
+ return (__m256d)__builtin_ia32_vfmaddpd256(-(__v4df)__A, (__v4df)__B, (__v4df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_fnmsub_ps(__m256 __A, __m256 __B, __m256 __C)
+{
+ return (__m256)__builtin_ia32_vfmaddps256(-(__v8sf)__A, (__v8sf)__B, -(__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
+_mm256_fnmsub_pd(__m256d __A, __m256d __B, __m256d __C)
+{
+ return (__m256d)__builtin_ia32_vfmaddpd256(-(__v4df)__A, (__v4df)__B, -(__v4df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_fmaddsub_ps(__m256 __A, __m256 __B, __m256 __C)
+{
+ return (__m256)__builtin_ia32_vfmaddsubps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
+_mm256_fmaddsub_pd(__m256d __A, __m256d __B, __m256d __C)
+{
+ return (__m256d)__builtin_ia32_vfmaddsubpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_fmsubadd_ps(__m256 __A, __m256 __B, __m256 __C)
+{
+ return (__m256)__builtin_ia32_vfmaddsubps256((__v8sf)__A, (__v8sf)__B, -(__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
+_mm256_fmsubadd_pd(__m256d __A, __m256d __B, __m256d __C)
+{
+ return (__m256d)__builtin_ia32_vfmaddsubpd256((__v4df)__A, (__v4df)__B, -(__v4df)__C);
+}
+
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
+
+#endif /* __FMAINTRIN_H */
--- /dev/null
+//===- FuzzedDataProvider.h - Utility header for fuzz targets ---*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// A single header library providing an utility class to break up an array of
+// bytes. Whenever run on the same input, provides the same output, as long as
+// its methods are called in the same order, with the same arguments.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_FUZZED_DATA_PROVIDER_H_
+#define LLVM_FUZZER_FUZZED_DATA_PROVIDER_H_
+
+#include <algorithm>
+#include <climits>
+#include <cstddef>
+#include <cstdint>
+#include <cstring>
+#include <initializer_list>
+#include <string>
+#include <type_traits>
+#include <utility>
+#include <vector>
+
+// In addition to the comments below, the API is also briefly documented at
+// https://github.com/google/fuzzing/blob/master/docs/split-inputs.md#fuzzed-data-provider
+class FuzzedDataProvider {
+ public:
+ // |data| is an array of length |size| that the FuzzedDataProvider wraps to
+ // provide more granular access. |data| must outlive the FuzzedDataProvider.
+ FuzzedDataProvider(const uint8_t *data, size_t size)
+ : data_ptr_(data), remaining_bytes_(size) {}
+ ~FuzzedDataProvider() = default;
+
+ // See the implementation below (after the class definition) for more verbose
+ // comments for each of the methods.
+
+ // Methods returning std::vector of bytes. These are the most popular choice
+ // when splitting fuzzing input into pieces, as every piece is put into a
+ // separate buffer (i.e. ASan would catch any under-/overflow) and the memory
+ // will be released automatically.
+ template <typename T> std::vector<T> ConsumeBytes(size_t num_bytes);
+ template <typename T>
+ std::vector<T> ConsumeBytesWithTerminator(size_t num_bytes, T terminator = 0);
+ template <typename T> std::vector<T> ConsumeRemainingBytes();
+
+ // Methods returning strings. Use only when you need a std::string or a null
+ // terminated C-string. Otherwise, prefer the methods returning std::vector.
+ std::string ConsumeBytesAsString(size_t num_bytes);
+ std::string ConsumeRandomLengthString(size_t max_length);
+ std::string ConsumeRandomLengthString();
+ std::string ConsumeRemainingBytesAsString();
+
+ // Methods returning integer values.
+ template <typename T> T ConsumeIntegral();
+ template <typename T> T ConsumeIntegralInRange(T min, T max);
+
+ // Methods returning floating point values.
+ template <typename T> T ConsumeFloatingPoint();
+ template <typename T> T ConsumeFloatingPointInRange(T min, T max);
+
+ // 0 <= return value <= 1.
+ template <typename T> T ConsumeProbability();
+
+ bool ConsumeBool();
+
+ // Returns a value chosen from the given enum.
+ template <typename T> T ConsumeEnum();
+
+ // Returns a value from the given array.
+ template <typename T, size_t size> T PickValueInArray(const T (&array)[size]);
+ template <typename T> T PickValueInArray(std::initializer_list<const T> list);
+
+ // Writes data to the given destination and returns number of bytes written.
+ size_t ConsumeData(void *destination, size_t num_bytes);
+
+ // Reports the remaining bytes available for fuzzed input.
+ size_t remaining_bytes() { return remaining_bytes_; }
+
+ private:
+ FuzzedDataProvider(const FuzzedDataProvider &) = delete;
+ FuzzedDataProvider &operator=(const FuzzedDataProvider &) = delete;
+
+ void CopyAndAdvance(void *destination, size_t num_bytes);
+
+ void Advance(size_t num_bytes);
+
+ template <typename T>
+ std::vector<T> ConsumeBytes(size_t size, size_t num_bytes);
+
+ template <typename TS, typename TU> TS ConvertUnsignedToSigned(TU value);
+
+ const uint8_t *data_ptr_;
+ size_t remaining_bytes_;
+};
+
+// Returns a std::vector containing |num_bytes| of input data. If fewer than
+// |num_bytes| of data remain, returns a shorter std::vector containing all
+// of the data that's left. Can be used with any byte sized type, such as
+// char, unsigned char, uint8_t, etc.
+template <typename T>
+std::vector<T> FuzzedDataProvider::ConsumeBytes(size_t num_bytes) {
+ num_bytes = std::min(num_bytes, remaining_bytes_);
+ return ConsumeBytes<T>(num_bytes, num_bytes);
+}
+
+// Similar to |ConsumeBytes|, but also appends the terminator value at the end
+// of the resulting vector. Useful, when a mutable null-terminated C-string is
+// needed, for example. But that is a rare case. Better avoid it, if possible,
+// and prefer using |ConsumeBytes| or |ConsumeBytesAsString| methods.
+template <typename T>
+std::vector<T> FuzzedDataProvider::ConsumeBytesWithTerminator(size_t num_bytes,
+ T terminator) {
+ num_bytes = std::min(num_bytes, remaining_bytes_);
+ std::vector<T> result = ConsumeBytes<T>(num_bytes + 1, num_bytes);
+ result.back() = terminator;
+ return result;
+}
+
+// Returns a std::vector containing all remaining bytes of the input data.
+template <typename T>
+std::vector<T> FuzzedDataProvider::ConsumeRemainingBytes() {
+ return ConsumeBytes<T>(remaining_bytes_);
+}
+
+// Returns a std::string containing |num_bytes| of input data. Using this and
+// |.c_str()| on the resulting string is the best way to get an immutable
+// null-terminated C string. If fewer than |num_bytes| of data remain, returns
+// a shorter std::string containing all of the data that's left.
+inline std::string FuzzedDataProvider::ConsumeBytesAsString(size_t num_bytes) {
+ static_assert(sizeof(std::string::value_type) == sizeof(uint8_t),
+ "ConsumeBytesAsString cannot convert the data to a string.");
+
+ num_bytes = std::min(num_bytes, remaining_bytes_);
+ std::string result(
+ reinterpret_cast<const std::string::value_type *>(data_ptr_), num_bytes);
+ Advance(num_bytes);
+ return result;
+}
+
+// Returns a std::string of length from 0 to |max_length|. When it runs out of
+// input data, returns what remains of the input. Designed to be more stable
+// with respect to a fuzzer inserting characters than just picking a random
+// length and then consuming that many bytes with |ConsumeBytes|.
+inline std::string
+FuzzedDataProvider::ConsumeRandomLengthString(size_t max_length) {
+ // Reads bytes from the start of |data_ptr_|. Maps "\\" to "\", and maps "\"
+ // followed by anything else to the end of the string. As a result of this
+ // logic, a fuzzer can insert characters into the string, and the string
+ // will be lengthened to include those new characters, resulting in a more
+ // stable fuzzer than picking the length of a string independently from
+ // picking its contents.
+ std::string result;
+
+ // Reserve the anticipated capaticity to prevent several reallocations.
+ result.reserve(std::min(max_length, remaining_bytes_));
+ for (size_t i = 0; i < max_length && remaining_bytes_ != 0; ++i) {
+ char next = ConvertUnsignedToSigned<char>(data_ptr_[0]);
+ Advance(1);
+ if (next == '\\' && remaining_bytes_ != 0) {
+ next = ConvertUnsignedToSigned<char>(data_ptr_[0]);
+ Advance(1);
+ if (next != '\\')
+ break;
+ }
+ result += next;
+ }
+
+ result.shrink_to_fit();
+ return result;
+}
+
+// Returns a std::string of length from 0 to |remaining_bytes_|.
+inline std::string FuzzedDataProvider::ConsumeRandomLengthString() {
+ return ConsumeRandomLengthString(remaining_bytes_);
+}
+
+// Returns a std::string containing all remaining bytes of the input data.
+// Prefer using |ConsumeRemainingBytes| unless you actually need a std::string
+// object.
+inline std::string FuzzedDataProvider::ConsumeRemainingBytesAsString() {
+ return ConsumeBytesAsString(remaining_bytes_);
+}
+
+// Returns a number in the range [Type's min, Type's max]. The value might
+// not be uniformly distributed in the given range. If there's no input data
+// left, always returns |min|.
+template <typename T> T FuzzedDataProvider::ConsumeIntegral() {
+ return ConsumeIntegralInRange(std::numeric_limits<T>::min(),
+ std::numeric_limits<T>::max());
+}
+
+// Returns a number in the range [min, max] by consuming bytes from the
+// input data. The value might not be uniformly distributed in the given
+// range. If there's no input data left, always returns |min|. |min| must
+// be less than or equal to |max|.
+template <typename T>
+T FuzzedDataProvider::ConsumeIntegralInRange(T min, T max) {
+ static_assert(std::is_integral<T>::value, "An integral type is required.");
+ static_assert(sizeof(T) <= sizeof(uint64_t), "Unsupported integral type.");
+
+ if (min > max)
+ abort();
+
+ // Use the biggest type possible to hold the range and the result.
+ uint64_t range = static_cast<uint64_t>(max) - min;
+ uint64_t result = 0;
+ size_t offset = 0;
+
+ while (offset < sizeof(T) * CHAR_BIT && (range >> offset) > 0 &&
+ remaining_bytes_ != 0) {
+ // Pull bytes off the end of the seed data. Experimentally, this seems to
+ // allow the fuzzer to more easily explore the input space. This makes
+ // sense, since it works by modifying inputs that caused new code to run,
+ // and this data is often used to encode length of data read by
+ // |ConsumeBytes|. Separating out read lengths makes it easier modify the
+ // contents of the data that is actually read.
+ --remaining_bytes_;
+ result = (result << CHAR_BIT) | data_ptr_[remaining_bytes_];
+ offset += CHAR_BIT;
+ }
+
+ // Avoid division by 0, in case |range + 1| results in overflow.
+ if (range != std::numeric_limits<decltype(range)>::max())
+ result = result % (range + 1);
+
+ return static_cast<T>(min + result);
+}
+
+// Returns a floating point value in the range [Type's lowest, Type's max] by
+// consuming bytes from the input data. If there's no input data left, always
+// returns approximately 0.
+template <typename T> T FuzzedDataProvider::ConsumeFloatingPoint() {
+ return ConsumeFloatingPointInRange<T>(std::numeric_limits<T>::lowest(),
+ std::numeric_limits<T>::max());
+}
+
+// Returns a floating point value in the given range by consuming bytes from
+// the input data. If there's no input data left, returns |min|. Note that
+// |min| must be less than or equal to |max|.
+template <typename T>
+T FuzzedDataProvider::ConsumeFloatingPointInRange(T min, T max) {
+ if (min > max)
+ abort();
+
+ T range = .0;
+ T result = min;
+ constexpr T zero(.0);
+ if (max > zero && min < zero && max > min + std::numeric_limits<T>::max()) {
+ // The diff |max - min| would overflow the given floating point type. Use
+ // the half of the diff as the range and consume a bool to decide whether
+ // the result is in the first of the second part of the diff.
+ range = (max / 2.0) - (min / 2.0);
+ if (ConsumeBool()) {
+ result += range;
+ }
+ } else {
+ range = max - min;
+ }
+
+ return result + range * ConsumeProbability<T>();
+}
+
+// Returns a floating point number in the range [0.0, 1.0]. If there's no
+// input data left, always returns 0.
+template <typename T> T FuzzedDataProvider::ConsumeProbability() {
+ static_assert(std::is_floating_point<T>::value,
+ "A floating point type is required.");
+
+ // Use different integral types for different floating point types in order
+ // to provide better density of the resulting values.
+ using IntegralType =
+ typename std::conditional<(sizeof(T) <= sizeof(uint32_t)), uint32_t,
+ uint64_t>::type;
+
+ T result = static_cast<T>(ConsumeIntegral<IntegralType>());
+ result /= static_cast<T>(std::numeric_limits<IntegralType>::max());
+ return result;
+}
+
+// Reads one byte and returns a bool, or false when no data remains.
+inline bool FuzzedDataProvider::ConsumeBool() {
+ return 1 & ConsumeIntegral<uint8_t>();
+}
+
+// Returns an enum value. The enum must start at 0 and be contiguous. It must
+// also contain |kMaxValue| aliased to its largest (inclusive) value. Such as:
+// enum class Foo { SomeValue, OtherValue, kMaxValue = OtherValue };
+template <typename T> T FuzzedDataProvider::ConsumeEnum() {
+ static_assert(std::is_enum<T>::value, "|T| must be an enum type.");
+ return static_cast<T>(
+ ConsumeIntegralInRange<uint32_t>(0, static_cast<uint32_t>(T::kMaxValue)));
+}
+
+// Returns a copy of the value selected from the given fixed-size |array|.
+template <typename T, size_t size>
+T FuzzedDataProvider::PickValueInArray(const T (&array)[size]) {
+ static_assert(size > 0, "The array must be non empty.");
+ return array[ConsumeIntegralInRange<size_t>(0, size - 1)];
+}
+
+template <typename T>
+T FuzzedDataProvider::PickValueInArray(std::initializer_list<const T> list) {
+ // TODO(Dor1s): switch to static_assert once C++14 is allowed.
+ if (!list.size())
+ abort();
+
+ return *(list.begin() + ConsumeIntegralInRange<size_t>(0, list.size() - 1));
+}
+
+// Writes |num_bytes| of input data to the given destination pointer. If there
+// is not enough data left, writes all remaining bytes. Return value is the
+// number of bytes written.
+// In general, it's better to avoid using this function, but it may be useful
+// in cases when it's necessary to fill a certain buffer or object with
+// fuzzing data.
+inline size_t FuzzedDataProvider::ConsumeData(void *destination,
+ size_t num_bytes) {
+ num_bytes = std::min(num_bytes, remaining_bytes_);
+ CopyAndAdvance(destination, num_bytes);
+ return num_bytes;
+}
+
+// Private methods.
+inline void FuzzedDataProvider::CopyAndAdvance(void *destination,
+ size_t num_bytes) {
+ std::memcpy(destination, data_ptr_, num_bytes);
+ Advance(num_bytes);
+}
+
+inline void FuzzedDataProvider::Advance(size_t num_bytes) {
+ if (num_bytes > remaining_bytes_)
+ abort();
+
+ data_ptr_ += num_bytes;
+ remaining_bytes_ -= num_bytes;
+}
+
+template <typename T>
+std::vector<T> FuzzedDataProvider::ConsumeBytes(size_t size, size_t num_bytes) {
+ static_assert(sizeof(T) == sizeof(uint8_t), "Incompatible data type.");
+
+ // The point of using the size-based constructor below is to increase the
+ // odds of having a vector object with capacity being equal to the length.
+ // That part is always implementation specific, but at least both libc++ and
+ // libstdc++ allocate the requested number of bytes in that constructor,
+ // which seems to be a natural choice for other implementations as well.
+ // To increase the odds even more, we also call |shrink_to_fit| below.
+ std::vector<T> result(size);
+ if (size == 0) {
+ if (num_bytes != 0)
+ abort();
+ return result;
+ }
+
+ CopyAndAdvance(result.data(), num_bytes);
+
+ // Even though |shrink_to_fit| is also implementation specific, we expect it
+ // to provide an additional assurance in case vector's constructor allocated
+ // a buffer which is larger than the actual amount of data we put inside it.
+ result.shrink_to_fit();
+ return result;
+}
+
+template <typename TS, typename TU>
+TS FuzzedDataProvider::ConvertUnsignedToSigned(TU value) {
+ static_assert(sizeof(TS) == sizeof(TU), "Incompatible data types.");
+ static_assert(!std::numeric_limits<TU>::is_signed,
+ "Source type must be unsigned.");
+
+ // TODO(Dor1s): change to `if constexpr` once C++17 becomes mainstream.
+ if (std::numeric_limits<TS>::is_modulo)
+ return static_cast<TS>(value);
+
+ // Avoid using implementation-defined unsigned to signed conversions.
+ // To learn more, see https://stackoverflow.com/questions/13150449.
+ if (value <= std::numeric_limits<TS>::max()) {
+ return static_cast<TS>(value);
+ } else {
+ constexpr auto TS_min = std::numeric_limits<TS>::min();
+ return TS_min + static_cast<char>(value - TS_min);
+ }
+}
+
+#endif // LLVM_FUZZER_FUZZED_DATA_PROVIDER_H_
/*===---- fxsrintrin.h - FXSR intrinsic ------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("fxsr")))
-/// \brief Saves the XMM, MMX, MXCSR and x87 FPU registers into a 512-byte
+/// Saves the XMM, MMX, MXCSR and x87 FPU registers into a 512-byte
/// memory region pointed to by the input parameter \a __p.
///
/// \headerfile <x86intrin.h>
static __inline__ void __DEFAULT_FN_ATTRS
_fxsave(void *__p)
{
- return __builtin_ia32_fxsave(__p);
+ __builtin_ia32_fxsave(__p);
}
-/// \brief Restores the XMM, MMX, MXCSR and x87 FPU registers from the 512-byte
+/// Restores the XMM, MMX, MXCSR and x87 FPU registers from the 512-byte
/// memory region pointed to by the input parameter \a __p. The contents of
/// this memory region should have been written to by a previous \c _fxsave
/// or \c _fxsave64 intrinsic.
static __inline__ void __DEFAULT_FN_ATTRS
_fxrstor(void *__p)
{
- return __builtin_ia32_fxrstor(__p);
+ __builtin_ia32_fxrstor(__p);
}
#ifdef __x86_64__
-/// \brief Saves the XMM, MMX, MXCSR and x87 FPU registers into a 512-byte
+/// Saves the XMM, MMX, MXCSR and x87 FPU registers into a 512-byte
/// memory region pointed to by the input parameter \a __p.
///
/// \headerfile <x86intrin.h>
static __inline__ void __DEFAULT_FN_ATTRS
_fxsave64(void *__p)
{
- return __builtin_ia32_fxsave64(__p);
+ __builtin_ia32_fxsave64(__p);
}
-/// \brief Restores the XMM, MMX, MXCSR and x87 FPU registers from the 512-byte
+/// Restores the XMM, MMX, MXCSR and x87 FPU registers from the 512-byte
/// memory region pointed to by the input parameter \a __p. The contents of
/// this memory region should have been written to by a previous \c _fxsave
/// or \c _fxsave64 intrinsic.
static __inline__ void __DEFAULT_FN_ATTRS
_fxrstor64(void *__p)
{
- return __builtin_ia32_fxrstor64(__p);
+ __builtin_ia32_fxrstor64(__p);
}
#endif
--- /dev/null
+/*===----------------- gfniintrin.h - GFNI intrinsics ----------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <gfniintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __GFNIINTRIN_H
+#define __GFNIINTRIN_H
+
+
+#define _mm_gf2p8affineinv_epi64_epi8(A, B, I) \
+ (__m128i)__builtin_ia32_vgf2p8affineinvqb_v16qi((__v16qi)(__m128i)(A), \
+ (__v16qi)(__m128i)(B), \
+ (char)(I))
+
+#define _mm_mask_gf2p8affineinv_epi64_epi8(S, U, A, B, I) \
+ (__m128i)__builtin_ia32_selectb_128((__mmask16)(U), \
+ (__v16qi)_mm_gf2p8affineinv_epi64_epi8(A, B, I), \
+ (__v16qi)(__m128i)(S))
+
+
+#define _mm_maskz_gf2p8affineinv_epi64_epi8(U, A, B, I) \
+ (__m128i)_mm_mask_gf2p8affineinv_epi64_epi8((__m128i)_mm_setzero_si128(), \
+ U, A, B, I)
+
+
+#define _mm256_gf2p8affineinv_epi64_epi8(A, B, I) \
+ (__m256i)__builtin_ia32_vgf2p8affineinvqb_v32qi((__v32qi)(__m256i)(A), \
+ (__v32qi)(__m256i)(B), \
+ (char)(I))
+
+#define _mm256_mask_gf2p8affineinv_epi64_epi8(S, U, A, B, I) \
+ (__m256i)__builtin_ia32_selectb_256((__mmask32)(U), \
+ (__v32qi)_mm256_gf2p8affineinv_epi64_epi8(A, B, I), \
+ (__v32qi)(__m256i)(S))
+
+#define _mm256_maskz_gf2p8affineinv_epi64_epi8(U, A, B, I) \
+ (__m256i)_mm256_mask_gf2p8affineinv_epi64_epi8((__m256i)_mm256_setzero_si256(), \
+ U, A, B, I)
+
+
+#define _mm512_gf2p8affineinv_epi64_epi8(A, B, I) \
+ (__m512i)__builtin_ia32_vgf2p8affineinvqb_v64qi((__v64qi)(__m512i)(A), \
+ (__v64qi)(__m512i)(B), \
+ (char)(I))
+
+#define _mm512_mask_gf2p8affineinv_epi64_epi8(S, U, A, B, I) \
+ (__m512i)__builtin_ia32_selectb_512((__mmask64)(U), \
+ (__v64qi)_mm512_gf2p8affineinv_epi64_epi8(A, B, I), \
+ (__v64qi)(__m512i)(S))
+
+#define _mm512_maskz_gf2p8affineinv_epi64_epi8(U, A, B, I) \
+ (__m512i)_mm512_mask_gf2p8affineinv_epi64_epi8((__m512i)_mm512_setzero_si512(), \
+ U, A, B, I)
+
+#define _mm_gf2p8affine_epi64_epi8(A, B, I) \
+ (__m128i)__builtin_ia32_vgf2p8affineqb_v16qi((__v16qi)(__m128i)(A), \
+ (__v16qi)(__m128i)(B), \
+ (char)(I))
+
+#define _mm_mask_gf2p8affine_epi64_epi8(S, U, A, B, I) \
+ (__m128i)__builtin_ia32_selectb_128((__mmask16)(U), \
+ (__v16qi)_mm_gf2p8affine_epi64_epi8(A, B, I), \
+ (__v16qi)(__m128i)(S))
+
+
+#define _mm_maskz_gf2p8affine_epi64_epi8(U, A, B, I) \
+ (__m128i)_mm_mask_gf2p8affine_epi64_epi8((__m128i)_mm_setzero_si128(), \
+ U, A, B, I)
+
+
+#define _mm256_gf2p8affine_epi64_epi8(A, B, I) \
+ (__m256i)__builtin_ia32_vgf2p8affineqb_v32qi((__v32qi)(__m256i)(A), \
+ (__v32qi)(__m256i)(B), \
+ (char)(I))
+
+#define _mm256_mask_gf2p8affine_epi64_epi8(S, U, A, B, I) \
+ (__m256i)__builtin_ia32_selectb_256((__mmask32)(U), \
+ (__v32qi)_mm256_gf2p8affine_epi64_epi8(A, B, I), \
+ (__v32qi)(__m256i)(S))
+
+#define _mm256_maskz_gf2p8affine_epi64_epi8(U, A, B, I) \
+ (__m256i)_mm256_mask_gf2p8affine_epi64_epi8((__m256i)_mm256_setzero_si256(), \
+ U, A, B, I)
+
+
+#define _mm512_gf2p8affine_epi64_epi8(A, B, I) \
+ (__m512i)__builtin_ia32_vgf2p8affineqb_v64qi((__v64qi)(__m512i)(A), \
+ (__v64qi)(__m512i)(B), \
+ (char)(I))
+
+#define _mm512_mask_gf2p8affine_epi64_epi8(S, U, A, B, I) \
+ (__m512i)__builtin_ia32_selectb_512((__mmask64)(U), \
+ (__v64qi)_mm512_gf2p8affine_epi64_epi8(A, B, I), \
+ (__v64qi)(__m512i)(S))
+
+#define _mm512_maskz_gf2p8affine_epi64_epi8(U, A, B, I) \
+ (__m512i)_mm512_mask_gf2p8affine_epi64_epi8((__m512i)_mm512_setzero_si512(), \
+ U, A, B, I)
+
+/* Default attributes for simple form (no masking). */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("gfni"), __min_vector_width__(128)))
+
+/* Default attributes for YMM unmasked form. */
+#define __DEFAULT_FN_ATTRS_Y __attribute__((__always_inline__, __nodebug__, __target__("avx,gfni"), __min_vector_width__(256)))
+
+/* Default attributes for ZMM forms. */
+#define __DEFAULT_FN_ATTRS_Z __attribute__((__always_inline__, __nodebug__, __target__("avx512bw,gfni"), __min_vector_width__(512)))
+
+/* Default attributes for VLX forms. */
+#define __DEFAULT_FN_ATTRS_VL128 __attribute__((__always_inline__, __nodebug__, __target__("avx512bw,avx512vl,gfni"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS_VL256 __attribute__((__always_inline__, __nodebug__, __target__("avx512bw,avx512vl,gfni"), __min_vector_width__(256)))
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS
+_mm_gf2p8mul_epi8(__m128i __A, __m128i __B)
+{
+ return (__m128i) __builtin_ia32_vgf2p8mulb_v16qi((__v16qi) __A,
+ (__v16qi) __B);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS_VL128
+_mm_mask_gf2p8mul_epi8(__m128i __S, __mmask16 __U, __m128i __A, __m128i __B)
+{
+ return (__m128i) __builtin_ia32_selectb_128(__U,
+ (__v16qi) _mm_gf2p8mul_epi8(__A, __B),
+ (__v16qi) __S);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS_VL128
+_mm_maskz_gf2p8mul_epi8(__mmask16 __U, __m128i __A, __m128i __B)
+{
+ return _mm_mask_gf2p8mul_epi8((__m128i)_mm_setzero_si128(),
+ __U, __A, __B);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS_Y
+_mm256_gf2p8mul_epi8(__m256i __A, __m256i __B)
+{
+ return (__m256i) __builtin_ia32_vgf2p8mulb_v32qi((__v32qi) __A,
+ (__v32qi) __B);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS_VL256
+_mm256_mask_gf2p8mul_epi8(__m256i __S, __mmask32 __U, __m256i __A, __m256i __B)
+{
+ return (__m256i) __builtin_ia32_selectb_256(__U,
+ (__v32qi) _mm256_gf2p8mul_epi8(__A, __B),
+ (__v32qi) __S);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS_VL256
+_mm256_maskz_gf2p8mul_epi8(__mmask32 __U, __m256i __A, __m256i __B)
+{
+ return _mm256_mask_gf2p8mul_epi8((__m256i)_mm256_setzero_si256(),
+ __U, __A, __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS_Z
+_mm512_gf2p8mul_epi8(__m512i __A, __m512i __B)
+{
+ return (__m512i) __builtin_ia32_vgf2p8mulb_v64qi((__v64qi) __A,
+ (__v64qi) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS_Z
+_mm512_mask_gf2p8mul_epi8(__m512i __S, __mmask64 __U, __m512i __A, __m512i __B)
+{
+ return (__m512i) __builtin_ia32_selectb_512(__U,
+ (__v64qi) _mm512_gf2p8mul_epi8(__A, __B),
+ (__v64qi) __S);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS_Z
+_mm512_maskz_gf2p8mul_epi8(__mmask64 __U, __m512i __A, __m512i __B)
+{
+ return _mm512_mask_gf2p8mul_epi8((__m512i)_mm512_setzero_si512(),
+ __U, __A, __B);
+}
+
+#undef __DEFAULT_FN_ATTRS
+#undef __DEFAULT_FN_ATTRS_Y
+#undef __DEFAULT_FN_ATTRS_Z
+#undef __DEFAULT_FN_ATTRS_VL128
+#undef __DEFAULT_FN_ATTRS_VL256
+
+#endif /* __GFNIINTRIN_H */
+
/*===---- htmintrin.h - Standard header for PowerPC HTM ---------------===*\
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
\*===----------------------------------------------------------------------===*/
/*===---- htmxlintrin.h - XL compiler HTM execution intrinsics-------------===*\
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
\*===----------------------------------------------------------------------===*/
/* These intrinsics are being made available for compatibility with
the IBM XL compiler. For documentation please see the "z/OS XL
- C/C++ Programming Guide" publically available on the web. */
+ C/C++ Programming Guide" publicly available on the web. */
static __inline long __attribute__((__always_inline__, __nodebug__))
__TM_simple_begin ()
--- /dev/null
+/* ===-------- ia32intrin.h ---------------------------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __X86INTRIN_H
+#error "Never use <ia32intrin.h> directly; include <x86intrin.h> instead."
+#endif
+
+#ifndef __IA32INTRIN_H
+#define __IA32INTRIN_H
+
+/** Find the first set bit starting from the lsb. Result is undefined if
+ * input is 0.
+ *
+ * \headerfile <x86intrin.h>
+ *
+ * This intrinsic corresponds to the <c> BSF </c> instruction or the
+ * <c> TZCNT </c> instruction.
+ *
+ * \param __A
+ * A 32-bit integer operand.
+ * \returns A 32-bit integer containing the bit number.
+ */
+static __inline__ int __attribute__((__always_inline__, __nodebug__))
+__bsfd(int __A) {
+ return __builtin_ctz(__A);
+}
+
+/** Find the first set bit starting from the msb. Result is undefined if
+ * input is 0.
+ *
+ * \headerfile <x86intrin.h>
+ *
+ * This intrinsic corresponds to the <c> BSR </c> instruction or the
+ * <c> LZCNT </c> instruction and an <c> XOR </c>.
+ *
+ * \param __A
+ * A 32-bit integer operand.
+ * \returns A 32-bit integer containing the bit number.
+ */
+static __inline__ int __attribute__((__always_inline__, __nodebug__))
+__bsrd(int __A) {
+ return 31 - __builtin_clz(__A);
+}
+
+/** Swaps the bytes in the input. Converting little endian to big endian or
+ * vice versa.
+ *
+ * \headerfile <x86intrin.h>
+ *
+ * This intrinsic corresponds to the <c> BSWAP </c> instruction.
+ *
+ * \param __A
+ * A 32-bit integer operand.
+ * \returns A 32-bit integer containing the swapped bytes.
+ */
+static __inline__ int __attribute__((__always_inline__, __nodebug__))
+__bswapd(int __A) {
+ return __builtin_bswap32(__A);
+}
+
+static __inline__ int __attribute__((__always_inline__, __nodebug__))
+_bswap(int __A) {
+ return __builtin_bswap32(__A);
+}
+
+#define _bit_scan_forward(A) __bsfd((A))
+#define _bit_scan_reverse(A) __bsrd((A))
+
+#ifdef __x86_64__
+/** Find the first set bit starting from the lsb. Result is undefined if
+ * input is 0.
+ *
+ * \headerfile <x86intrin.h>
+ *
+ * This intrinsic corresponds to the <c> BSF </c> instruction or the
+ * <c> TZCNT </c> instruction.
+ *
+ * \param __A
+ * A 64-bit integer operand.
+ * \returns A 32-bit integer containing the bit number.
+ */
+static __inline__ int __attribute__((__always_inline__, __nodebug__))
+__bsfq(long long __A) {
+ return __builtin_ctzll(__A);
+}
+
+/** Find the first set bit starting from the msb. Result is undefined if
+ * input is 0.
+ *
+ * \headerfile <x86intrin.h>
+ *
+ * This intrinsic corresponds to the <c> BSR </c> instruction or the
+ * <c> LZCNT </c> instruction and an <c> XOR </c>.
+ *
+ * \param __A
+ * A 64-bit integer operand.
+ * \returns A 32-bit integer containing the bit number.
+ */
+static __inline__ int __attribute__((__always_inline__, __nodebug__))
+__bsrq(long long __A) {
+ return 63 - __builtin_clzll(__A);
+}
+
+/** Swaps the bytes in the input. Converting little endian to big endian or
+ * vice versa.
+ *
+ * \headerfile <x86intrin.h>
+ *
+ * This intrinsic corresponds to the <c> BSWAP </c> instruction.
+ *
+ * \param __A
+ * A 64-bit integer operand.
+ * \returns A 64-bit integer containing the swapped bytes.
+ */
+static __inline__ long long __attribute__((__always_inline__, __nodebug__))
+__bswapq(long long __A) {
+ return __builtin_bswap64(__A);
+}
+
+#define _bswap64(A) __bswapq((A))
+#endif
+
+/** Counts the number of bits in the source operand having a value of 1.
+ *
+ * \headerfile <x86intrin.h>
+ *
+ * This intrinsic corresponds to the <c> POPCNT </c> instruction or a
+ * a sequence of arithmetic and logic ops to calculate it.
+ *
+ * \param __A
+ * An unsigned 32-bit integer operand.
+ * \returns A 32-bit integer containing the number of bits with value 1 in the
+ * source operand.
+ */
+static __inline__ int __attribute__((__always_inline__, __nodebug__))
+__popcntd(unsigned int __A)
+{
+ return __builtin_popcount(__A);
+}
+
+#define _popcnt32(A) __popcntd((A))
+
+#ifdef __x86_64__
+/** Counts the number of bits in the source operand having a value of 1.
+ *
+ * \headerfile <x86intrin.h>
+ *
+ * This intrinsic corresponds to the <c> POPCNT </c> instruction or a
+ * a sequence of arithmetic and logic ops to calculate it.
+ *
+ * \param __A
+ * An unsigned 64-bit integer operand.
+ * \returns A 64-bit integer containing the number of bits with value 1 in the
+ * source operand.
+ */
+static __inline__ long long __attribute__((__always_inline__, __nodebug__))
+__popcntq(unsigned long long __A)
+{
+ return __builtin_popcountll(__A);
+}
+
+#define _popcnt64(A) __popcntq((A))
+#endif /* __x86_64__ */
+
+#ifdef __x86_64__
+static __inline__ unsigned long long __attribute__((__always_inline__, __nodebug__))
+__readeflags(void)
+{
+ return __builtin_ia32_readeflags_u64();
+}
+
+static __inline__ void __attribute__((__always_inline__, __nodebug__))
+__writeeflags(unsigned long long __f)
+{
+ __builtin_ia32_writeeflags_u64(__f);
+}
+
+#else /* !__x86_64__ */
+static __inline__ unsigned int __attribute__((__always_inline__, __nodebug__))
+__readeflags(void)
+{
+ return __builtin_ia32_readeflags_u32();
+}
+
+static __inline__ void __attribute__((__always_inline__, __nodebug__))
+__writeeflags(unsigned int __f)
+{
+ __builtin_ia32_writeeflags_u32(__f);
+}
+#endif /* !__x86_64__ */
+
+/** Cast a 32-bit float value to a 32-bit unsigned integer value
+ *
+ * \headerfile <x86intrin.h>
+ * This intrinsic corresponds to the <c> VMOVD / MOVD </c> instruction in x86_64,
+ * and corresponds to the <c> VMOVL / MOVL </c> instruction in ia32.
+ *
+ * \param __A
+ * A 32-bit float value.
+ * \returns a 32-bit unsigned integer containing the converted value.
+ */
+static __inline__ unsigned int __attribute__((__always_inline__))
+_castf32_u32(float __A) {
+ unsigned int D;
+ __builtin_memcpy(&D, &__A, sizeof(__A));
+ return D;
+}
+
+/** Cast a 64-bit float value to a 64-bit unsigned integer value
+ *
+ * \headerfile <x86intrin.h>
+ * This intrinsic corresponds to the <c> VMOVQ / MOVQ </c> instruction in x86_64,
+ * and corresponds to the <c> VMOVL / MOVL </c> instruction in ia32.
+ *
+ * \param __A
+ * A 64-bit float value.
+ * \returns a 64-bit unsigned integer containing the converted value.
+ */
+static __inline__ unsigned long long __attribute__((__always_inline__))
+_castf64_u64(double __A) {
+ unsigned long long D;
+ __builtin_memcpy(&D, &__A, sizeof(__A));
+ return D;
+}
+
+/** Cast a 32-bit unsigned integer value to a 32-bit float value
+ *
+ * \headerfile <x86intrin.h>
+ * This intrinsic corresponds to the <c> VMOVQ / MOVQ </c> instruction in x86_64,
+ * and corresponds to the <c> FLDS </c> instruction in ia32.
+ *
+ * \param __A
+ * A 32-bit unsigned integer value.
+ * \returns a 32-bit float value containing the converted value.
+ */
+static __inline__ float __attribute__((__always_inline__))
+_castu32_f32(unsigned int __A) {
+ float D;
+ __builtin_memcpy(&D, &__A, sizeof(__A));
+ return D;
+}
+
+/** Cast a 64-bit unsigned integer value to a 64-bit float value
+ *
+ * \headerfile <x86intrin.h>
+ * This intrinsic corresponds to the <c> VMOVQ / MOVQ </c> instruction in x86_64,
+ * and corresponds to the <c> FLDL </c> instruction in ia32.
+ *
+ * \param __A
+ * A 64-bit unsigned integer value.
+ * \returns a 64-bit float value containing the converted value.
+ */
+static __inline__ double __attribute__((__always_inline__))
+_castu64_f64(unsigned long long __A) {
+ double D;
+ __builtin_memcpy(&D, &__A, sizeof(__A));
+ return D;
+}
+
+/** Adds the unsigned integer operand to the CRC-32C checksum of the
+ * unsigned char operand.
+ *
+ * \headerfile <x86intrin.h>
+ *
+ * This intrinsic corresponds to the <c> CRC32B </c> instruction.
+ *
+ * \param __C
+ * An unsigned integer operand to add to the CRC-32C checksum of operand
+ * \a __D.
+ * \param __D
+ * An unsigned 8-bit integer operand used to compute the CRC-32C checksum.
+ * \returns The result of adding operand \a __C to the CRC-32C checksum of
+ * operand \a __D.
+ */
+static __inline__ unsigned int __attribute__((__always_inline__, __nodebug__, __target__("sse4.2")))
+__crc32b(unsigned int __C, unsigned char __D)
+{
+ return __builtin_ia32_crc32qi(__C, __D);
+}
+
+/** Adds the unsigned integer operand to the CRC-32C checksum of the
+ * unsigned short operand.
+ *
+ * \headerfile <x86intrin.h>
+ *
+ * This intrinsic corresponds to the <c> CRC32W </c> instruction.
+ *
+ * \param __C
+ * An unsigned integer operand to add to the CRC-32C checksum of operand
+ * \a __D.
+ * \param __D
+ * An unsigned 16-bit integer operand used to compute the CRC-32C checksum.
+ * \returns The result of adding operand \a __C to the CRC-32C checksum of
+ * operand \a __D.
+ */
+static __inline__ unsigned int __attribute__((__always_inline__, __nodebug__, __target__("sse4.2")))
+__crc32w(unsigned int __C, unsigned short __D)
+{
+ return __builtin_ia32_crc32hi(__C, __D);
+}
+
+/** Adds the unsigned integer operand to the CRC-32C checksum of the
+ * second unsigned integer operand.
+ *
+ * \headerfile <x86intrin.h>
+ *
+ * This intrinsic corresponds to the <c> CRC32D </c> instruction.
+ *
+ * \param __C
+ * An unsigned integer operand to add to the CRC-32C checksum of operand
+ * \a __D.
+ * \param __D
+ * An unsigned 32-bit integer operand used to compute the CRC-32C checksum.
+ * \returns The result of adding operand \a __C to the CRC-32C checksum of
+ * operand \a __D.
+ */
+static __inline__ unsigned int __attribute__((__always_inline__, __nodebug__, __target__("sse4.2")))
+__crc32d(unsigned int __C, unsigned int __D)
+{
+ return __builtin_ia32_crc32si(__C, __D);
+}
+
+#ifdef __x86_64__
+/** Adds the unsigned integer operand to the CRC-32C checksum of the
+ * unsigned 64-bit integer operand.
+ *
+ * \headerfile <x86intrin.h>
+ *
+ * This intrinsic corresponds to the <c> CRC32Q </c> instruction.
+ *
+ * \param __C
+ * An unsigned integer operand to add to the CRC-32C checksum of operand
+ * \a __D.
+ * \param __D
+ * An unsigned 64-bit integer operand used to compute the CRC-32C checksum.
+ * \returns The result of adding operand \a __C to the CRC-32C checksum of
+ * operand \a __D.
+ */
+static __inline__ unsigned long long __attribute__((__always_inline__, __nodebug__, __target__("sse4.2")))
+__crc32q(unsigned long long __C, unsigned long long __D)
+{
+ return __builtin_ia32_crc32di(__C, __D);
+}
+#endif /* __x86_64__ */
+
+static __inline__ unsigned long long __attribute__((__always_inline__, __nodebug__))
+__rdpmc(int __A) {
+ return __builtin_ia32_rdpmc(__A);
+}
+
+/* __rdtscp */
+static __inline__ unsigned long long __attribute__((__always_inline__, __nodebug__))
+__rdtscp(unsigned int *__A) {
+ return __builtin_ia32_rdtscp(__A);
+}
+
+#define _rdtsc() __rdtsc()
+
+#define _rdpmc(A) __rdpmc(A)
+
+static __inline__ void __attribute__((__always_inline__, __nodebug__))
+_wbinvd(void) {
+ __builtin_ia32_wbinvd();
+}
+
+static __inline__ unsigned char __attribute__((__always_inline__, __nodebug__))
+__rolb(unsigned char __X, int __C) {
+ return __builtin_rotateleft8(__X, __C);
+}
+
+static __inline__ unsigned char __attribute__((__always_inline__, __nodebug__))
+__rorb(unsigned char __X, int __C) {
+ return __builtin_rotateright8(__X, __C);
+}
+
+static __inline__ unsigned short __attribute__((__always_inline__, __nodebug__))
+__rolw(unsigned short __X, int __C) {
+ return __builtin_rotateleft16(__X, __C);
+}
+
+static __inline__ unsigned short __attribute__((__always_inline__, __nodebug__))
+__rorw(unsigned short __X, int __C) {
+ return __builtin_rotateright16(__X, __C);
+}
+
+static __inline__ unsigned int __attribute__((__always_inline__, __nodebug__))
+__rold(unsigned int __X, int __C) {
+ return __builtin_rotateleft32(__X, __C);
+}
+
+static __inline__ unsigned int __attribute__((__always_inline__, __nodebug__))
+__rord(unsigned int __X, int __C) {
+ return __builtin_rotateright32(__X, __C);
+}
+
+#ifdef __x86_64__
+static __inline__ unsigned long long __attribute__((__always_inline__, __nodebug__))
+__rolq(unsigned long long __X, int __C) {
+ return __builtin_rotateleft64(__X, __C);
+}
+
+static __inline__ unsigned long long __attribute__((__always_inline__, __nodebug__))
+__rorq(unsigned long long __X, int __C) {
+ return __builtin_rotateright64(__X, __C);
+}
+#endif /* __x86_64__ */
+
+#ifndef _MSC_VER
+/* These are already provided as builtins for MSVC. */
+/* Select the correct function based on the size of long. */
+#ifdef __LP64__
+#define _lrotl(a,b) __rolq((a), (b))
+#define _lrotr(a,b) __rorq((a), (b))
+#else
+#define _lrotl(a,b) __rold((a), (b))
+#define _lrotr(a,b) __rord((a), (b))
+#endif
+#define _rotl(a,b) __rold((a), (b))
+#define _rotr(a,b) __rord((a), (b))
+#endif // _MSC_VER
+
+/* These are not builtins so need to be provided in all modes. */
+#define _rotwl(a,b) __rolw((a), (b))
+#define _rotwr(a,b) __rorw((a), (b))
+
+#endif /* __IA32INTRIN_H */
--- /dev/null
+/*===---- immintrin.h - Intel intrinsics -----------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#define __IMMINTRIN_H
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__MMX__)
+#include <mmintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__SSE__)
+#include <xmmintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__SSE2__)
+#include <emmintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__SSE3__)
+#include <pmmintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__SSSE3__)
+#include <tmmintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ (defined(__SSE4_2__) || defined(__SSE4_1__))
+#include <smmintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ (defined(__AES__) || defined(__PCLMUL__))
+#include <wmmintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__CLFLUSHOPT__)
+#include <clflushoptintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__CLWB__)
+#include <clwbintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX__)
+#include <avxintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX2__)
+#include <avx2intrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__F16C__)
+#include <f16cintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__VPCLMULQDQ__)
+#include <vpclmulqdqintrin.h>
+#endif
+
+/* No feature check desired due to internal checks */
+#include <bmiintrin.h>
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__BMI2__)
+#include <bmi2intrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__LZCNT__)
+#include <lzcntintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__POPCNT__)
+#include <popcntintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__FMA__)
+#include <fmaintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX512F__)
+#include <avx512fintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX512VL__)
+#include <avx512vlintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX512BW__)
+#include <avx512bwintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX512BITALG__)
+#include <avx512bitalgintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX512CD__)
+#include <avx512cdintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX512VPOPCNTDQ__)
+#include <avx512vpopcntdqintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ (defined(__AVX512VL__) && defined(__AVX512VPOPCNTDQ__))
+#include <avx512vpopcntdqvlintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX512VNNI__)
+#include <avx512vnniintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ (defined(__AVX512VL__) && defined(__AVX512VNNI__))
+#include <avx512vlvnniintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX512DQ__)
+#include <avx512dqintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ (defined(__AVX512VL__) && defined(__AVX512BITALG__))
+#include <avx512vlbitalgintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ (defined(__AVX512VL__) && defined(__AVX512BW__))
+#include <avx512vlbwintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ (defined(__AVX512VL__) && defined(__AVX512CD__))
+#include <avx512vlcdintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ (defined(__AVX512VL__) && defined(__AVX512DQ__))
+#include <avx512vldqintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX512ER__)
+#include <avx512erintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX512IFMA__)
+#include <avx512ifmaintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ (defined(__AVX512IFMA__) && defined(__AVX512VL__))
+#include <avx512ifmavlintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX512VBMI__)
+#include <avx512vbmiintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ (defined(__AVX512VBMI__) && defined(__AVX512VL__))
+#include <avx512vbmivlintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX512VBMI2__)
+#include <avx512vbmi2intrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ (defined(__AVX512VBMI2__) && defined(__AVX512VL__))
+#include <avx512vlvbmi2intrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX512PF__)
+#include <avx512pfintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX512BF16__)
+#include <avx512bf16intrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ (defined(__AVX512VL__) && defined(__AVX512BF16__))
+#include <avx512vlbf16intrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__PKU__)
+#include <pkuintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__VAES__)
+#include <vaesintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__GFNI__)
+#include <gfniintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__RDPID__)
+/// Returns the value of the IA32_TSC_AUX MSR (0xc0000103).
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the <c> RDPID </c> instruction.
+static __inline__ unsigned int __attribute__((__always_inline__, __nodebug__, __target__("rdpid")))
+_rdpid_u32(void) {
+ return __builtin_ia32_rdpid();
+}
+#endif // __RDPID__
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__RDRND__)
+static __inline__ int __attribute__((__always_inline__, __nodebug__, __target__("rdrnd")))
+_rdrand16_step(unsigned short *__p)
+{
+ return __builtin_ia32_rdrand16_step(__p);
+}
+
+static __inline__ int __attribute__((__always_inline__, __nodebug__, __target__("rdrnd")))
+_rdrand32_step(unsigned int *__p)
+{
+ return __builtin_ia32_rdrand32_step(__p);
+}
+
+#ifdef __x86_64__
+static __inline__ int __attribute__((__always_inline__, __nodebug__, __target__("rdrnd")))
+_rdrand64_step(unsigned long long *__p)
+{
+ return __builtin_ia32_rdrand64_step(__p);
+}
+#endif
+#endif /* __RDRND__ */
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__FSGSBASE__)
+#ifdef __x86_64__
+static __inline__ unsigned int __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
+_readfsbase_u32(void)
+{
+ return __builtin_ia32_rdfsbase32();
+}
+
+static __inline__ unsigned long long __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
+_readfsbase_u64(void)
+{
+ return __builtin_ia32_rdfsbase64();
+}
+
+static __inline__ unsigned int __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
+_readgsbase_u32(void)
+{
+ return __builtin_ia32_rdgsbase32();
+}
+
+static __inline__ unsigned long long __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
+_readgsbase_u64(void)
+{
+ return __builtin_ia32_rdgsbase64();
+}
+
+static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
+_writefsbase_u32(unsigned int __V)
+{
+ __builtin_ia32_wrfsbase32(__V);
+}
+
+static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
+_writefsbase_u64(unsigned long long __V)
+{
+ __builtin_ia32_wrfsbase64(__V);
+}
+
+static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
+_writegsbase_u32(unsigned int __V)
+{
+ __builtin_ia32_wrgsbase32(__V);
+}
+
+static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
+_writegsbase_u64(unsigned long long __V)
+{
+ __builtin_ia32_wrgsbase64(__V);
+}
+
+#endif
+#endif /* __FSGSBASE__ */
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__MOVBE__)
+
+/* The structs used below are to force the load/store to be unaligned. This
+ * is accomplished with the __packed__ attribute. The __may_alias__ prevents
+ * tbaa metadata from being generated based on the struct and the type of the
+ * field inside of it.
+ */
+
+static __inline__ short __attribute__((__always_inline__, __nodebug__, __target__("movbe")))
+_loadbe_i16(void const * __P) {
+ struct __loadu_i16 {
+ short __v;
+ } __attribute__((__packed__, __may_alias__));
+ return __builtin_bswap16(((const struct __loadu_i16*)__P)->__v);
+}
+
+static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("movbe")))
+_storebe_i16(void * __P, short __D) {
+ struct __storeu_i16 {
+ short __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_i16*)__P)->__v = __builtin_bswap16(__D);
+}
+
+static __inline__ int __attribute__((__always_inline__, __nodebug__, __target__("movbe")))
+_loadbe_i32(void const * __P) {
+ struct __loadu_i32 {
+ int __v;
+ } __attribute__((__packed__, __may_alias__));
+ return __builtin_bswap32(((const struct __loadu_i32*)__P)->__v);
+}
+
+static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("movbe")))
+_storebe_i32(void * __P, int __D) {
+ struct __storeu_i32 {
+ int __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_i32*)__P)->__v = __builtin_bswap32(__D);
+}
+
+#ifdef __x86_64__
+static __inline__ long long __attribute__((__always_inline__, __nodebug__, __target__("movbe")))
+_loadbe_i64(void const * __P) {
+ struct __loadu_i64 {
+ long long __v;
+ } __attribute__((__packed__, __may_alias__));
+ return __builtin_bswap64(((const struct __loadu_i64*)__P)->__v);
+}
+
+static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("movbe")))
+_storebe_i64(void * __P, long long __D) {
+ struct __storeu_i64 {
+ long long __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __storeu_i64*)__P)->__v = __builtin_bswap64(__D);
+}
+#endif
+#endif /* __MOVBE */
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__RTM__)
+#include <rtmintrin.h>
+#include <xtestintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__SHA__)
+#include <shaintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__FXSR__)
+#include <fxsrintrin.h>
+#endif
+
+/* No feature check desired due to internal MSC_VER checks */
+#include <xsaveintrin.h>
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__XSAVEOPT__)
+#include <xsaveoptintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__XSAVEC__)
+#include <xsavecintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__XSAVES__)
+#include <xsavesintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__SHSTK__)
+#include <cetintrin.h>
+#endif
+
+/* Some intrinsics inside adxintrin.h are available only on processors with ADX,
+ * whereas others are also available at all times. */
+#include <adxintrin.h>
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__RDSEED__)
+#include <rdseedintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__WBNOINVD__)
+#include <wbnoinvdintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__CLDEMOTE__)
+#include <cldemoteintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__WAITPKG__)
+#include <waitpkgintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__MOVDIRI__) || defined(__MOVDIR64B__)
+#include <movdirintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__PCONFIG__)
+#include <pconfigintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__SGX__)
+#include <sgxintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__PTWRITE__)
+#include <ptwriteintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__INVPCID__)
+#include <invpcidintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AMXTILE__) || defined(__AMXINT8__) || defined(__AMXBF16__)
+#include <amxintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__AVX512VP2INTERSECT__)
+#include <avx512vp2intersectintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ (defined(__AVX512VL__) && defined(__AVX512VP2INTERSECT__))
+#include <avx512vlvp2intersectintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__ENQCMD__)
+#include <enqcmdintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__SERIALIZE__)
+#include <serializeintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__TSXLDTRK__)
+#include <tsxldtrkintrin.h>
+#endif
+
+#if defined(_MSC_VER) && __has_extension(gnu_asm)
+/* Define the default attributes for these intrinsics */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__))
+#ifdef __cplusplus
+extern "C" {
+#endif
+/*----------------------------------------------------------------------------*\
+|* Interlocked Exchange HLE
+\*----------------------------------------------------------------------------*/
+#if defined(__i386__) || defined(__x86_64__)
+static __inline__ long __DEFAULT_FN_ATTRS
+_InterlockedExchange_HLEAcquire(long volatile *_Target, long _Value) {
+ __asm__ __volatile__(".byte 0xf2 ; lock ; xchg %0, %1"
+ : "+r" (_Value), "+m" (*_Target) :: "memory");
+ return _Value;
+}
+static __inline__ long __DEFAULT_FN_ATTRS
+_InterlockedExchange_HLERelease(long volatile *_Target, long _Value) {
+ __asm__ __volatile__(".byte 0xf3 ; lock ; xchg %0, %1"
+ : "+r" (_Value), "+m" (*_Target) :: "memory");
+ return _Value;
+}
+#endif
+#if defined(__x86_64__)
+static __inline__ __int64 __DEFAULT_FN_ATTRS
+_InterlockedExchange64_HLEAcquire(__int64 volatile *_Target, __int64 _Value) {
+ __asm__ __volatile__(".byte 0xf2 ; lock ; xchg %0, %1"
+ : "+r" (_Value), "+m" (*_Target) :: "memory");
+ return _Value;
+}
+static __inline__ __int64 __DEFAULT_FN_ATTRS
+_InterlockedExchange64_HLERelease(__int64 volatile *_Target, __int64 _Value) {
+ __asm__ __volatile__(".byte 0xf3 ; lock ; xchg %0, %1"
+ : "+r" (_Value), "+m" (*_Target) :: "memory");
+ return _Value;
+}
+#endif
+/*----------------------------------------------------------------------------*\
+|* Interlocked Compare Exchange HLE
+\*----------------------------------------------------------------------------*/
+#if defined(__i386__) || defined(__x86_64__)
+static __inline__ long __DEFAULT_FN_ATTRS
+_InterlockedCompareExchange_HLEAcquire(long volatile *_Destination,
+ long _Exchange, long _Comparand) {
+ __asm__ __volatile__(".byte 0xf2 ; lock ; cmpxchg %2, %1"
+ : "+a" (_Comparand), "+m" (*_Destination)
+ : "r" (_Exchange) : "memory");
+ return _Comparand;
+}
+static __inline__ long __DEFAULT_FN_ATTRS
+_InterlockedCompareExchange_HLERelease(long volatile *_Destination,
+ long _Exchange, long _Comparand) {
+ __asm__ __volatile__(".byte 0xf3 ; lock ; cmpxchg %2, %1"
+ : "+a" (_Comparand), "+m" (*_Destination)
+ : "r" (_Exchange) : "memory");
+ return _Comparand;
+}
+#endif
+#if defined(__x86_64__)
+static __inline__ __int64 __DEFAULT_FN_ATTRS
+_InterlockedCompareExchange64_HLEAcquire(__int64 volatile *_Destination,
+ __int64 _Exchange, __int64 _Comparand) {
+ __asm__ __volatile__(".byte 0xf2 ; lock ; cmpxchg %2, %1"
+ : "+a" (_Comparand), "+m" (*_Destination)
+ : "r" (_Exchange) : "memory");
+ return _Comparand;
+}
+static __inline__ __int64 __DEFAULT_FN_ATTRS
+_InterlockedCompareExchange64_HLERelease(__int64 volatile *_Destination,
+ __int64 _Exchange, __int64 _Comparand) {
+ __asm__ __volatile__(".byte 0xf3 ; lock ; cmpxchg %2, %1"
+ : "+a" (_Comparand), "+m" (*_Destination)
+ : "r" (_Exchange) : "memory");
+ return _Comparand;
+}
+#endif
+#ifdef __cplusplus
+}
+#endif
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif /* defined(_MSC_VER) && __has_extension(gnu_asm) */
+
+#endif /* __IMMINTRIN_H */
--- /dev/null
+/* ===-------- intrin.h ---------------------------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+/* Only include this if we're compiling for the windows platform. */
+#ifndef _MSC_VER
+#include_next <intrin.h>
+#else
+
+#ifndef __INTRIN_H
+#define __INTRIN_H
+
+/* First include the standard intrinsics. */
+#if defined(__i386__) || defined(__x86_64__)
+#include <x86intrin.h>
+#endif
+
+#if defined(__arm__)
+#include <armintr.h>
+#endif
+
+#if defined(__aarch64__)
+#include <arm64intr.h>
+#endif
+
+/* For the definition of jmp_buf. */
+#if __STDC_HOSTED__
+#include <setjmp.h>
+#endif
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__))
+
+#if __x86_64__
+#define __LPTRINT_TYPE__ __int64
+#else
+#define __LPTRINT_TYPE__ long
+#endif
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#if defined(__MMX__)
+/* And the random ones that aren't in those files. */
+__m64 _m_from_float(float);
+float _m_to_float(__m64);
+#endif
+
+/* Other assorted instruction intrinsics. */
+void __addfsbyte(unsigned long, unsigned char);
+void __addfsdword(unsigned long, unsigned long);
+void __addfsword(unsigned long, unsigned short);
+void __code_seg(const char *);
+static __inline__
+void __cpuid(int[4], int);
+static __inline__
+void __cpuidex(int[4], int, int);
+static __inline__
+__int64 __emul(int, int);
+static __inline__
+unsigned __int64 __emulu(unsigned int, unsigned int);
+unsigned int __getcallerseflags(void);
+static __inline__
+void __halt(void);
+unsigned char __inbyte(unsigned short);
+void __inbytestring(unsigned short, unsigned char *, unsigned long);
+void __incfsbyte(unsigned long);
+void __incfsdword(unsigned long);
+void __incfsword(unsigned long);
+unsigned long __indword(unsigned short);
+void __indwordstring(unsigned short, unsigned long *, unsigned long);
+void __int2c(void);
+void __invlpg(void *);
+unsigned short __inword(unsigned short);
+void __inwordstring(unsigned short, unsigned short *, unsigned long);
+void __lidt(void *);
+unsigned __int64 __ll_lshift(unsigned __int64, int);
+__int64 __ll_rshift(__int64, int);
+static __inline__
+void __movsb(unsigned char *, unsigned char const *, size_t);
+static __inline__
+void __movsd(unsigned long *, unsigned long const *, size_t);
+static __inline__
+void __movsw(unsigned short *, unsigned short const *, size_t);
+static __inline__
+void __nop(void);
+void __nvreg_restore_fence(void);
+void __nvreg_save_fence(void);
+void __outbyte(unsigned short, unsigned char);
+void __outbytestring(unsigned short, unsigned char *, unsigned long);
+void __outdword(unsigned short, unsigned long);
+void __outdwordstring(unsigned short, unsigned long *, unsigned long);
+void __outword(unsigned short, unsigned short);
+void __outwordstring(unsigned short, unsigned short *, unsigned long);
+unsigned long __readcr0(void);
+unsigned long __readcr2(void);
+unsigned __LPTRINT_TYPE__ __readcr3(void);
+unsigned long __readcr4(void);
+unsigned long __readcr8(void);
+unsigned int __readdr(unsigned int);
+#ifdef __i386__
+static __inline__
+unsigned char __readfsbyte(unsigned long);
+static __inline__
+unsigned __int64 __readfsqword(unsigned long);
+static __inline__
+unsigned short __readfsword(unsigned long);
+#endif
+static __inline__
+unsigned __int64 __readmsr(unsigned long);
+unsigned __int64 __readpmc(unsigned long);
+unsigned long __segmentlimit(unsigned long);
+void __sidt(void *);
+static __inline__
+void __stosb(unsigned char *, unsigned char, size_t);
+static __inline__
+void __stosd(unsigned long *, unsigned long, size_t);
+static __inline__
+void __stosw(unsigned short *, unsigned short, size_t);
+void __svm_clgi(void);
+void __svm_invlpga(void *, int);
+void __svm_skinit(int);
+void __svm_stgi(void);
+void __svm_vmload(size_t);
+void __svm_vmrun(size_t);
+void __svm_vmsave(size_t);
+void __ud2(void);
+unsigned __int64 __ull_rshift(unsigned __int64, int);
+void __vmx_off(void);
+void __vmx_vmptrst(unsigned __int64 *);
+void __wbinvd(void);
+void __writecr0(unsigned int);
+static __inline__
+void __writecr3(unsigned __INTPTR_TYPE__);
+void __writecr4(unsigned int);
+void __writecr8(unsigned int);
+void __writedr(unsigned int, unsigned int);
+void __writefsbyte(unsigned long, unsigned char);
+void __writefsdword(unsigned long, unsigned long);
+void __writefsqword(unsigned long, unsigned __int64);
+void __writefsword(unsigned long, unsigned short);
+void __writemsr(unsigned long, unsigned __int64);
+static __inline__
+void *_AddressOfReturnAddress(void);
+static __inline__
+unsigned char _BitScanForward(unsigned long *_Index, unsigned long _Mask);
+static __inline__
+unsigned char _BitScanReverse(unsigned long *_Index, unsigned long _Mask);
+unsigned char _bittest(long const *, long);
+unsigned char _bittestandcomplement(long *, long);
+unsigned char _bittestandreset(long *, long);
+unsigned char _bittestandset(long *, long);
+void __cdecl _disable(void);
+void __cdecl _enable(void);
+long _InterlockedAddLargeStatistic(__int64 volatile *_Addend, long _Value);
+unsigned char _interlockedbittestandreset(long volatile *, long);
+unsigned char _interlockedbittestandset(long volatile *, long);
+void *_InterlockedCompareExchangePointer_HLEAcquire(void *volatile *, void *,
+ void *);
+void *_InterlockedCompareExchangePointer_HLERelease(void *volatile *, void *,
+ void *);
+long _InterlockedExchangeAdd_HLEAcquire(long volatile *, long);
+long _InterlockedExchangeAdd_HLERelease(long volatile *, long);
+__int64 _InterlockedExchangeAdd64_HLEAcquire(__int64 volatile *, __int64);
+__int64 _InterlockedExchangeAdd64_HLERelease(__int64 volatile *, __int64);
+static __inline__ void
+__attribute__((__deprecated__("use other intrinsics or C++11 atomics instead")))
+_ReadBarrier(void);
+static __inline__ void
+__attribute__((__deprecated__("use other intrinsics or C++11 atomics instead")))
+_ReadWriteBarrier(void);
+unsigned int _rorx_u32(unsigned int, const unsigned int);
+int _sarx_i32(int, unsigned int);
+#if __STDC_HOSTED__
+int __cdecl _setjmp(jmp_buf);
+#endif
+unsigned int _shlx_u32(unsigned int, unsigned int);
+unsigned int _shrx_u32(unsigned int, unsigned int);
+void _Store_HLERelease(long volatile *, long);
+void _Store64_HLERelease(__int64 volatile *, __int64);
+void _StorePointer_HLERelease(void *volatile *, void *);
+static __inline__ void
+__attribute__((__deprecated__("use other intrinsics or C++11 atomics instead")))
+_WriteBarrier(void);
+unsigned __int32 xbegin(void);
+void _xend(void);
+
+/* These additional intrinsics are turned on in x64/amd64/x86_64 mode. */
+#ifdef __x86_64__
+void __addgsbyte(unsigned long, unsigned char);
+void __addgsdword(unsigned long, unsigned long);
+void __addgsqword(unsigned long, unsigned __int64);
+void __addgsword(unsigned long, unsigned short);
+static __inline__
+void __faststorefence(void);
+void __incgsbyte(unsigned long);
+void __incgsdword(unsigned long);
+void __incgsqword(unsigned long);
+void __incgsword(unsigned long);
+static __inline__
+void __movsq(unsigned long long *, unsigned long long const *, size_t);
+static __inline__
+unsigned char __readgsbyte(unsigned long);
+static __inline__
+unsigned long __readgsdword(unsigned long);
+static __inline__
+unsigned __int64 __readgsqword(unsigned long);
+unsigned short __readgsword(unsigned long);
+unsigned __int64 __shiftleft128(unsigned __int64 _LowPart,
+ unsigned __int64 _HighPart,
+ unsigned char _Shift);
+unsigned __int64 __shiftright128(unsigned __int64 _LowPart,
+ unsigned __int64 _HighPart,
+ unsigned char _Shift);
+static __inline__
+void __stosq(unsigned __int64 *, unsigned __int64, size_t);
+unsigned char __vmx_on(unsigned __int64 *);
+unsigned char __vmx_vmclear(unsigned __int64 *);
+unsigned char __vmx_vmlaunch(void);
+unsigned char __vmx_vmptrld(unsigned __int64 *);
+unsigned char __vmx_vmread(size_t, size_t *);
+unsigned char __vmx_vmresume(void);
+unsigned char __vmx_vmwrite(size_t, size_t);
+void __writegsbyte(unsigned long, unsigned char);
+void __writegsdword(unsigned long, unsigned long);
+void __writegsqword(unsigned long, unsigned __int64);
+void __writegsword(unsigned long, unsigned short);
+unsigned char _bittest64(__int64 const *, __int64);
+unsigned char _bittestandcomplement64(__int64 *, __int64);
+unsigned char _bittestandreset64(__int64 *, __int64);
+unsigned char _bittestandset64(__int64 *, __int64);
+long _InterlockedAnd_np(long volatile *_Value, long _Mask);
+short _InterlockedAnd16_np(short volatile *_Value, short _Mask);
+__int64 _InterlockedAnd64_np(__int64 volatile *_Value, __int64 _Mask);
+char _InterlockedAnd8_np(char volatile *_Value, char _Mask);
+unsigned char _interlockedbittestandreset64(__int64 volatile *, __int64);
+unsigned char _interlockedbittestandset64(__int64 volatile *, __int64);
+long _InterlockedCompareExchange_np(long volatile *_Destination, long _Exchange,
+ long _Comparand);
+unsigned char _InterlockedCompareExchange128(__int64 volatile *_Destination,
+ __int64 _ExchangeHigh,
+ __int64 _ExchangeLow,
+ __int64 *_CompareandResult);
+unsigned char _InterlockedCompareExchange128_np(__int64 volatile *_Destination,
+ __int64 _ExchangeHigh,
+ __int64 _ExchangeLow,
+ __int64 *_ComparandResult);
+short _InterlockedCompareExchange16_np(short volatile *_Destination,
+ short _Exchange, short _Comparand);
+__int64 _InterlockedCompareExchange64_np(__int64 volatile *_Destination,
+ __int64 _Exchange, __int64 _Comparand);
+void *_InterlockedCompareExchangePointer_np(void *volatile *_Destination,
+ void *_Exchange, void *_Comparand);
+long _InterlockedOr_np(long volatile *_Value, long _Mask);
+short _InterlockedOr16_np(short volatile *_Value, short _Mask);
+__int64 _InterlockedOr64_np(__int64 volatile *_Value, __int64 _Mask);
+char _InterlockedOr8_np(char volatile *_Value, char _Mask);
+long _InterlockedXor_np(long volatile *_Value, long _Mask);
+short _InterlockedXor16_np(short volatile *_Value, short _Mask);
+__int64 _InterlockedXor64_np(__int64 volatile *_Value, __int64 _Mask);
+char _InterlockedXor8_np(char volatile *_Value, char _Mask);
+unsigned __int64 _rorx_u64(unsigned __int64, const unsigned int);
+__int64 _sarx_i64(__int64, unsigned int);
+unsigned __int64 _shlx_u64(unsigned __int64, unsigned int);
+unsigned __int64 _shrx_u64(unsigned __int64, unsigned int);
+static __inline__
+__int64 __mulh(__int64, __int64);
+static __inline__
+unsigned __int64 __umulh(unsigned __int64, unsigned __int64);
+static __inline__
+__int64 _mul128(__int64, __int64, __int64*);
+static __inline__
+unsigned __int64 _umul128(unsigned __int64,
+ unsigned __int64,
+ unsigned __int64*);
+
+#endif /* __x86_64__ */
+
+#if defined(__x86_64__) || defined(__arm__) || defined(__aarch64__)
+
+static __inline__
+unsigned char _BitScanForward64(unsigned long *_Index, unsigned __int64 _Mask);
+static __inline__
+unsigned char _BitScanReverse64(unsigned long *_Index, unsigned __int64 _Mask);
+
+#endif
+
+#if defined(__i386__) || defined(__x86_64__) || defined(__arm__) || defined(__aarch64__)
+static __inline__
+__int64 _InterlockedDecrement64(__int64 volatile *_Addend);
+static __inline__
+__int64 _InterlockedExchange64(__int64 volatile *_Target, __int64 _Value);
+static __inline__
+__int64 _InterlockedExchangeAdd64(__int64 volatile *_Addend, __int64 _Value);
+static __inline__
+__int64 _InterlockedExchangeSub64(__int64 volatile *_Subend, __int64 _Value);
+static __inline__
+__int64 _InterlockedIncrement64(__int64 volatile *_Addend);
+static __inline__
+__int64 _InterlockedOr64(__int64 volatile *_Value, __int64 _Mask);
+static __inline__
+__int64 _InterlockedXor64(__int64 volatile *_Value, __int64 _Mask);
+static __inline__
+__int64 _InterlockedAnd64(__int64 volatile *_Value, __int64 _Mask);
+
+#endif
+
+/*----------------------------------------------------------------------------*\
+|* Interlocked Exchange Add
+\*----------------------------------------------------------------------------*/
+#if defined(__arm__) || defined(__aarch64__)
+char _InterlockedExchangeAdd8_acq(char volatile *_Addend, char _Value);
+char _InterlockedExchangeAdd8_nf(char volatile *_Addend, char _Value);
+char _InterlockedExchangeAdd8_rel(char volatile *_Addend, char _Value);
+short _InterlockedExchangeAdd16_acq(short volatile *_Addend, short _Value);
+short _InterlockedExchangeAdd16_nf(short volatile *_Addend, short _Value);
+short _InterlockedExchangeAdd16_rel(short volatile *_Addend, short _Value);
+long _InterlockedExchangeAdd_acq(long volatile *_Addend, long _Value);
+long _InterlockedExchangeAdd_nf(long volatile *_Addend, long _Value);
+long _InterlockedExchangeAdd_rel(long volatile *_Addend, long _Value);
+__int64 _InterlockedExchangeAdd64_acq(__int64 volatile *_Addend, __int64 _Value);
+__int64 _InterlockedExchangeAdd64_nf(__int64 volatile *_Addend, __int64 _Value);
+__int64 _InterlockedExchangeAdd64_rel(__int64 volatile *_Addend, __int64 _Value);
+#endif
+/*----------------------------------------------------------------------------*\
+|* Interlocked Increment
+\*----------------------------------------------------------------------------*/
+#if defined(__arm__) || defined(__aarch64__)
+short _InterlockedIncrement16_acq(short volatile *_Value);
+short _InterlockedIncrement16_nf(short volatile *_Value);
+short _InterlockedIncrement16_rel(short volatile *_Value);
+long _InterlockedIncrement_acq(long volatile *_Value);
+long _InterlockedIncrement_nf(long volatile *_Value);
+long _InterlockedIncrement_rel(long volatile *_Value);
+__int64 _InterlockedIncrement64_acq(__int64 volatile *_Value);
+__int64 _InterlockedIncrement64_nf(__int64 volatile *_Value);
+__int64 _InterlockedIncrement64_rel(__int64 volatile *_Value);
+#endif
+/*----------------------------------------------------------------------------*\
+|* Interlocked Decrement
+\*----------------------------------------------------------------------------*/
+#if defined(__arm__) || defined(__aarch64__)
+short _InterlockedDecrement16_acq(short volatile *_Value);
+short _InterlockedDecrement16_nf(short volatile *_Value);
+short _InterlockedDecrement16_rel(short volatile *_Value);
+long _InterlockedDecrement_acq(long volatile *_Value);
+long _InterlockedDecrement_nf(long volatile *_Value);
+long _InterlockedDecrement_rel(long volatile *_Value);
+__int64 _InterlockedDecrement64_acq(__int64 volatile *_Value);
+__int64 _InterlockedDecrement64_nf(__int64 volatile *_Value);
+__int64 _InterlockedDecrement64_rel(__int64 volatile *_Value);
+#endif
+/*----------------------------------------------------------------------------*\
+|* Interlocked And
+\*----------------------------------------------------------------------------*/
+#if defined(__arm__) || defined(__aarch64__)
+char _InterlockedAnd8_acq(char volatile *_Value, char _Mask);
+char _InterlockedAnd8_nf(char volatile *_Value, char _Mask);
+char _InterlockedAnd8_rel(char volatile *_Value, char _Mask);
+short _InterlockedAnd16_acq(short volatile *_Value, short _Mask);
+short _InterlockedAnd16_nf(short volatile *_Value, short _Mask);
+short _InterlockedAnd16_rel(short volatile *_Value, short _Mask);
+long _InterlockedAnd_acq(long volatile *_Value, long _Mask);
+long _InterlockedAnd_nf(long volatile *_Value, long _Mask);
+long _InterlockedAnd_rel(long volatile *_Value, long _Mask);
+__int64 _InterlockedAnd64_acq(__int64 volatile *_Value, __int64 _Mask);
+__int64 _InterlockedAnd64_nf(__int64 volatile *_Value, __int64 _Mask);
+__int64 _InterlockedAnd64_rel(__int64 volatile *_Value, __int64 _Mask);
+#endif
+/*----------------------------------------------------------------------------*\
+|* Bit Counting and Testing
+\*----------------------------------------------------------------------------*/
+#if defined(__arm__) || defined(__aarch64__)
+unsigned char _interlockedbittestandset_acq(long volatile *_BitBase,
+ long _BitPos);
+unsigned char _interlockedbittestandset_nf(long volatile *_BitBase,
+ long _BitPos);
+unsigned char _interlockedbittestandset_rel(long volatile *_BitBase,
+ long _BitPos);
+unsigned char _interlockedbittestandreset_acq(long volatile *_BitBase,
+ long _BitPos);
+unsigned char _interlockedbittestandreset_nf(long volatile *_BitBase,
+ long _BitPos);
+unsigned char _interlockedbittestandreset_rel(long volatile *_BitBase,
+ long _BitPos);
+#endif
+/*----------------------------------------------------------------------------*\
+|* Interlocked Or
+\*----------------------------------------------------------------------------*/
+#if defined(__arm__) || defined(__aarch64__)
+char _InterlockedOr8_acq(char volatile *_Value, char _Mask);
+char _InterlockedOr8_nf(char volatile *_Value, char _Mask);
+char _InterlockedOr8_rel(char volatile *_Value, char _Mask);
+short _InterlockedOr16_acq(short volatile *_Value, short _Mask);
+short _InterlockedOr16_nf(short volatile *_Value, short _Mask);
+short _InterlockedOr16_rel(short volatile *_Value, short _Mask);
+long _InterlockedOr_acq(long volatile *_Value, long _Mask);
+long _InterlockedOr_nf(long volatile *_Value, long _Mask);
+long _InterlockedOr_rel(long volatile *_Value, long _Mask);
+__int64 _InterlockedOr64_acq(__int64 volatile *_Value, __int64 _Mask);
+__int64 _InterlockedOr64_nf(__int64 volatile *_Value, __int64 _Mask);
+__int64 _InterlockedOr64_rel(__int64 volatile *_Value, __int64 _Mask);
+#endif
+/*----------------------------------------------------------------------------*\
+|* Interlocked Xor
+\*----------------------------------------------------------------------------*/
+#if defined(__arm__) || defined(__aarch64__)
+char _InterlockedXor8_acq(char volatile *_Value, char _Mask);
+char _InterlockedXor8_nf(char volatile *_Value, char _Mask);
+char _InterlockedXor8_rel(char volatile *_Value, char _Mask);
+short _InterlockedXor16_acq(short volatile *_Value, short _Mask);
+short _InterlockedXor16_nf(short volatile *_Value, short _Mask);
+short _InterlockedXor16_rel(short volatile *_Value, short _Mask);
+long _InterlockedXor_acq(long volatile *_Value, long _Mask);
+long _InterlockedXor_nf(long volatile *_Value, long _Mask);
+long _InterlockedXor_rel(long volatile *_Value, long _Mask);
+__int64 _InterlockedXor64_acq(__int64 volatile *_Value, __int64 _Mask);
+__int64 _InterlockedXor64_nf(__int64 volatile *_Value, __int64 _Mask);
+__int64 _InterlockedXor64_rel(__int64 volatile *_Value, __int64 _Mask);
+#endif
+/*----------------------------------------------------------------------------*\
+|* Interlocked Exchange
+\*----------------------------------------------------------------------------*/
+#if defined(__arm__) || defined(__aarch64__)
+char _InterlockedExchange8_acq(char volatile *_Target, char _Value);
+char _InterlockedExchange8_nf(char volatile *_Target, char _Value);
+char _InterlockedExchange8_rel(char volatile *_Target, char _Value);
+short _InterlockedExchange16_acq(short volatile *_Target, short _Value);
+short _InterlockedExchange16_nf(short volatile *_Target, short _Value);
+short _InterlockedExchange16_rel(short volatile *_Target, short _Value);
+long _InterlockedExchange_acq(long volatile *_Target, long _Value);
+long _InterlockedExchange_nf(long volatile *_Target, long _Value);
+long _InterlockedExchange_rel(long volatile *_Target, long _Value);
+__int64 _InterlockedExchange64_acq(__int64 volatile *_Target, __int64 _Value);
+__int64 _InterlockedExchange64_nf(__int64 volatile *_Target, __int64 _Value);
+__int64 _InterlockedExchange64_rel(__int64 volatile *_Target, __int64 _Value);
+#endif
+/*----------------------------------------------------------------------------*\
+|* Interlocked Compare Exchange
+\*----------------------------------------------------------------------------*/
+#if defined(__arm__) || defined(__aarch64__)
+char _InterlockedCompareExchange8_acq(char volatile *_Destination,
+ char _Exchange, char _Comparand);
+char _InterlockedCompareExchange8_nf(char volatile *_Destination,
+ char _Exchange, char _Comparand);
+char _InterlockedCompareExchange8_rel(char volatile *_Destination,
+ char _Exchange, char _Comparand);
+short _InterlockedCompareExchange16_acq(short volatile *_Destination,
+ short _Exchange, short _Comparand);
+short _InterlockedCompareExchange16_nf(short volatile *_Destination,
+ short _Exchange, short _Comparand);
+short _InterlockedCompareExchange16_rel(short volatile *_Destination,
+ short _Exchange, short _Comparand);
+long _InterlockedCompareExchange_acq(long volatile *_Destination,
+ long _Exchange, long _Comparand);
+long _InterlockedCompareExchange_nf(long volatile *_Destination,
+ long _Exchange, long _Comparand);
+long _InterlockedCompareExchange_rel(long volatile *_Destination,
+ long _Exchange, long _Comparand);
+__int64 _InterlockedCompareExchange64_acq(__int64 volatile *_Destination,
+ __int64 _Exchange, __int64 _Comparand);
+__int64 _InterlockedCompareExchange64_nf(__int64 volatile *_Destination,
+ __int64 _Exchange, __int64 _Comparand);
+__int64 _InterlockedCompareExchange64_rel(__int64 volatile *_Destination,
+ __int64 _Exchange, __int64 _Comparand);
+#endif
+
+/*----------------------------------------------------------------------------*\
+|* movs, stos
+\*----------------------------------------------------------------------------*/
+#if defined(__i386__) || defined(__x86_64__)
+static __inline__ void __DEFAULT_FN_ATTRS
+__movsb(unsigned char *__dst, unsigned char const *__src, size_t __n) {
+ __asm__ __volatile__("rep movsb" : "+D"(__dst), "+S"(__src), "+c"(__n)
+ : : "memory");
+}
+static __inline__ void __DEFAULT_FN_ATTRS
+__movsd(unsigned long *__dst, unsigned long const *__src, size_t __n) {
+ __asm__ __volatile__("rep movsl" : "+D"(__dst), "+S"(__src), "+c"(__n)
+ : : "memory");
+}
+static __inline__ void __DEFAULT_FN_ATTRS
+__movsw(unsigned short *__dst, unsigned short const *__src, size_t __n) {
+ __asm__ __volatile__("rep movsw" : "+D"(__dst), "+S"(__src), "+c"(__n)
+ : : "memory");
+}
+static __inline__ void __DEFAULT_FN_ATTRS
+__stosd(unsigned long *__dst, unsigned long __x, size_t __n) {
+ __asm__ __volatile__("rep stosl" : "+D"(__dst), "+c"(__n) : "a"(__x)
+ : "memory");
+}
+static __inline__ void __DEFAULT_FN_ATTRS
+__stosw(unsigned short *__dst, unsigned short __x, size_t __n) {
+ __asm__ __volatile__("rep stosw" : "+D"(__dst), "+c"(__n) : "a"(__x)
+ : "memory");
+}
+#endif
+#ifdef __x86_64__
+static __inline__ void __DEFAULT_FN_ATTRS
+__movsq(unsigned long long *__dst, unsigned long long const *__src, size_t __n) {
+ __asm__ __volatile__("rep movsq" : "+D"(__dst), "+S"(__src), "+c"(__n)
+ : : "memory");
+}
+static __inline__ void __DEFAULT_FN_ATTRS
+__stosq(unsigned __int64 *__dst, unsigned __int64 __x, size_t __n) {
+ __asm__ __volatile__("rep stosq" : "+D"(__dst), "+c"(__n) : "a"(__x)
+ : "memory");
+}
+#endif
+
+/*----------------------------------------------------------------------------*\
+|* Misc
+\*----------------------------------------------------------------------------*/
+#if defined(__i386__) || defined(__x86_64__)
+static __inline__ void __DEFAULT_FN_ATTRS
+__cpuid(int __info[4], int __level) {
+ __asm__ ("cpuid" : "=a"(__info[0]), "=b" (__info[1]), "=c"(__info[2]), "=d"(__info[3])
+ : "a"(__level), "c"(0));
+}
+static __inline__ void __DEFAULT_FN_ATTRS
+__cpuidex(int __info[4], int __level, int __ecx) {
+ __asm__ ("cpuid" : "=a"(__info[0]), "=b" (__info[1]), "=c"(__info[2]), "=d"(__info[3])
+ : "a"(__level), "c"(__ecx));
+}
+static __inline__ void __DEFAULT_FN_ATTRS
+__halt(void) {
+ __asm__ volatile ("hlt");
+}
+#endif
+
+#if defined(__i386__) || defined(__x86_64__) || defined(__aarch64__)
+static __inline__ void __DEFAULT_FN_ATTRS
+__nop(void) {
+ __asm__ volatile ("nop");
+}
+#endif
+
+/*----------------------------------------------------------------------------*\
+|* MS AArch64 specific
+\*----------------------------------------------------------------------------*/
+#if defined(__aarch64__)
+unsigned __int64 __getReg(int);
+long _InterlockedAdd(long volatile *Addend, long Value);
+__int64 _ReadStatusReg(int);
+void _WriteStatusReg(int, __int64);
+
+unsigned short __cdecl _byteswap_ushort(unsigned short val);
+unsigned long __cdecl _byteswap_ulong (unsigned long val);
+unsigned __int64 __cdecl _byteswap_uint64(unsigned __int64 val);
+#endif
+
+/*----------------------------------------------------------------------------*\
+|* Privileged intrinsics
+\*----------------------------------------------------------------------------*/
+#if defined(__i386__) || defined(__x86_64__)
+static __inline__ unsigned __int64 __DEFAULT_FN_ATTRS
+__readmsr(unsigned long __register) {
+ // Loads the contents of a 64-bit model specific register (MSR) specified in
+ // the ECX register into registers EDX:EAX. The EDX register is loaded with
+ // the high-order 32 bits of the MSR and the EAX register is loaded with the
+ // low-order 32 bits. If less than 64 bits are implemented in the MSR being
+ // read, the values returned to EDX:EAX in unimplemented bit locations are
+ // undefined.
+ unsigned long __edx;
+ unsigned long __eax;
+ __asm__ ("rdmsr" : "=d"(__edx), "=a"(__eax) : "c"(__register));
+ return (((unsigned __int64)__edx) << 32) | (unsigned __int64)__eax;
+}
+#endif
+
+static __inline__ unsigned __LPTRINT_TYPE__ __DEFAULT_FN_ATTRS
+__readcr3(void) {
+ unsigned __LPTRINT_TYPE__ __cr3_val;
+ __asm__ __volatile__ ("mov %%cr3, %0" : "=r"(__cr3_val) : : "memory");
+ return __cr3_val;
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS
+__writecr3(unsigned __INTPTR_TYPE__ __cr3_val) {
+ __asm__ ("mov %0, %%cr3" : : "r"(__cr3_val) : "memory");
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#undef __LPTRINT_TYPE__
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif /* __INTRIN_H */
+#endif /* _MSC_VER */
/*===---- inttypes.h - Standard header for integer printf macros ----------===*\
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
\*===----------------------------------------------------------------------===*/
#ifndef __CLANG_INTTYPES_H
+// AIX system headers need inttypes.h to be re-enterable while _STD_TYPES_T
+// is defined until an inclusion of it without _STD_TYPES_T occurs, in which
+// case the header guard macro is defined.
+#if !defined(_AIX) || !defined(_STD_TYPES_T)
#define __CLANG_INTTYPES_H
+#endif
#if defined(_MSC_VER) && _MSC_VER < 1800
#error MSVC does not have inttypes.h prior to Visual Studio 2013
--- /dev/null
+/*===------------- invpcidintrin.h - INVPCID intrinsic ---------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <invpcidintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __INVPCIDINTRIN_H
+#define __INVPCIDINTRIN_H
+
+static __inline__ void
+ __attribute__((__always_inline__, __nodebug__, __target__("invpcid")))
+_invpcid(unsigned int __type, void *__descriptor) {
+ __builtin_ia32_invpcid(__type, __descriptor);
+}
+
+#endif /* __INVPCIDINTRIN_H */
--- /dev/null
+/*===---- iso646.h - Standard header for alternate spellings of operators---===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __ISO646_H
+#define __ISO646_H
+
+#ifndef __cplusplus
+#define and &&
+#define and_eq &=
+#define bitand &
+#define bitor |
+#define compl ~
+#define not !
+#define not_eq !=
+#define or ||
+#define or_eq |=
+#define xor ^
+#define xor_eq ^=
+#endif
+
+#endif /* __ISO646_H */
/*===---- limits.h - Standard header for integer sizes --------------------===*\
*
- * Copyright (c) 2009 Chris Lattner
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
\*===----------------------------------------------------------------------===*/
/*===---- lwpintrin.h - LWP intrinsics -------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
/* Define the default attributes for the functions in this file. */
#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("lwp")))
-/// \brief Parses the LWPCB at the specified address and enables
+/// Parses the LWPCB at the specified address and enables
/// profiling if valid.
///
/// \headerfile <x86intrin.h>
__builtin_ia32_llwpcb(__addr);
}
-/// \brief Flushes the LWP state to memory and returns the address of the LWPCB.
+/// Flushes the LWP state to memory and returns the address of the LWPCB.
///
/// \headerfile <x86intrin.h>
///
/// Address to the current Lightweight Profiling Control Block (LWPCB).
/// If LWP is not currently enabled, returns NULL.
static __inline__ void* __DEFAULT_FN_ATTRS
-__slwpcb ()
+__slwpcb (void)
{
return __builtin_ia32_slwpcb();
}
-/// \brief Inserts programmed event record into the LWP event ring buffer
+/// Inserts programmed event record into the LWP event ring buffer
/// and advances the ring buffer pointer.
///
/// \headerfile <x86intrin.h>
(__builtin_ia32_lwpins32((unsigned int) (DATA2), (unsigned int) (DATA1), \
(unsigned int) (FLAGS)))
-/// \brief Decrements the LWP programmed value sample event counter. If the result is
+/// Decrements the LWP programmed value sample event counter. If the result is
/// negative, inserts an event record into the LWP event ring buffer in memory
/// and advances the ring buffer pointer.
///
#ifdef __x86_64__
-/// \brief Inserts programmed event record into the LWP event ring buffer
+/// Inserts programmed event record into the LWP event ring buffer
/// and advances the ring buffer pointer.
///
/// \headerfile <x86intrin.h>
(__builtin_ia32_lwpins64((unsigned long long) (DATA2), (unsigned int) (DATA1), \
(unsigned int) (FLAGS)))
-/// \brief Decrements the LWP programmed value sample event counter. If the result is
+/// Decrements the LWP programmed value sample event counter. If the result is
/// negative, inserts an event record into the LWP event ring buffer in memory
/// and advances the ring buffer pointer.
///
/*===---- lzcntintrin.h - LZCNT intrinsics ---------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
/* Define the default attributes for the functions in this file. */
#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("lzcnt")))
-/// \brief Counts the number of leading zero bits in the operand.
+#ifndef _MSC_VER
+/// Counts the number of leading zero bits in the operand.
///
/// \headerfile <x86intrin.h>
///
/// An unsigned 16-bit integer whose leading zeros are to be counted.
/// \returns An unsigned 16-bit integer containing the number of leading zero
/// bits in the operand.
-static __inline__ unsigned short __DEFAULT_FN_ATTRS
-__lzcnt16(unsigned short __X)
-{
- return __X ? __builtin_clzs(__X) : 16;
-}
+#define __lzcnt16(X) __builtin_ia32_lzcnt_u16((unsigned short)(X))
+#endif // _MSC_VER
-/// \brief Counts the number of leading zero bits in the operand.
+/// Counts the number of leading zero bits in the operand.
///
/// \headerfile <x86intrin.h>
///
/// An unsigned 32-bit integer whose leading zeros are to be counted.
/// \returns An unsigned 32-bit integer containing the number of leading zero
/// bits in the operand.
+/// \see _lzcnt_u32
static __inline__ unsigned int __DEFAULT_FN_ATTRS
__lzcnt32(unsigned int __X)
{
- return __X ? __builtin_clz(__X) : 32;
+ return __builtin_ia32_lzcnt_u32(__X);
}
-/// \brief Counts the number of leading zero bits in the operand.
+/// Counts the number of leading zero bits in the operand.
///
/// \headerfile <x86intrin.h>
///
/// An unsigned 32-bit integer whose leading zeros are to be counted.
/// \returns An unsigned 32-bit integer containing the number of leading zero
/// bits in the operand.
+/// \see __lzcnt32
static __inline__ unsigned int __DEFAULT_FN_ATTRS
_lzcnt_u32(unsigned int __X)
{
- return __X ? __builtin_clz(__X) : 32;
+ return __builtin_ia32_lzcnt_u32(__X);
}
#ifdef __x86_64__
-/// \brief Counts the number of leading zero bits in the operand.
+#ifndef _MSC_VER
+/// Counts the number of leading zero bits in the operand.
///
/// \headerfile <x86intrin.h>
///
/// An unsigned 64-bit integer whose leading zeros are to be counted.
/// \returns An unsigned 64-bit integer containing the number of leading zero
/// bits in the operand.
-static __inline__ unsigned long long __DEFAULT_FN_ATTRS
-__lzcnt64(unsigned long long __X)
-{
- return __X ? __builtin_clzll(__X) : 64;
-}
+/// \see _lzcnt_u64
+#define __lzcnt64(X) __builtin_ia32_lzcnt_u64((unsigned long long)(X))
+#endif // _MSC_VER
-/// \brief Counts the number of leading zero bits in the operand.
+/// Counts the number of leading zero bits in the operand.
///
/// \headerfile <x86intrin.h>
///
/// An unsigned 64-bit integer whose leading zeros are to be counted.
/// \returns An unsigned 64-bit integer containing the number of leading zero
/// bits in the operand.
+/// \see __lzcnt64
static __inline__ unsigned long long __DEFAULT_FN_ATTRS
_lzcnt_u64(unsigned long long __X)
{
- return __X ? __builtin_clzll(__X) : 64;
+ return __builtin_ia32_lzcnt_u64(__X);
}
#endif
/*===---- mm3dnow.h - 3DNow! intrinsics ------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
typedef float __v2sf __attribute__((__vector_size__(8)));
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("3dnow")))
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("3dnow"), __min_vector_width__(64)))
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("3dnow")))
_m_femms(void) {
__builtin_ia32_femms();
}
/* Handle the 3dnowa instructions here. */
#undef __DEFAULT_FN_ATTRS
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("3dnowa")))
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("3dnowa"), __min_vector_width__(64)))
static __inline__ __m64 __DEFAULT_FN_ATTRS
_m_pf2iw(__m64 __m) {
/*===---- mm_malloc.h - Allocating and Freeing Aligned Memory Blocks -------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
/*===---- mmintrin.h - MMX intrinsics --------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#ifndef __MMINTRIN_H
#define __MMINTRIN_H
-typedef long long __m64 __attribute__((__vector_size__(8)));
+typedef long long __m64 __attribute__((__vector_size__(8), __aligned__(8)));
typedef long long __v1di __attribute__((__vector_size__(8)));
typedef int __v2si __attribute__((__vector_size__(8)));
typedef char __v8qi __attribute__((__vector_size__(8)));
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("mmx")))
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("mmx"), __min_vector_width__(64)))
-/// \brief Clears the MMX state by setting the state of the x87 stack registers
+/// Clears the MMX state by setting the state of the x87 stack registers
/// to empty.
///
/// \headerfile <x86intrin.h>
///
/// This intrinsic corresponds to the <c> EMMS </c> instruction.
///
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("mmx")))
_mm_empty(void)
{
__builtin_ia32_emms();
}
-/// \brief Constructs a 64-bit integer vector, setting the lower 32 bits to the
+/// Constructs a 64-bit integer vector, setting the lower 32 bits to the
/// value of the 32-bit integer parameter and setting the upper 32 bits to 0.
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c> VMOVD / MOVD </c> instruction.
+/// This intrinsic corresponds to the <c> MOVD </c> instruction.
///
/// \param __i
/// A 32-bit integer value.
return (__m64)__builtin_ia32_vec_init_v2si(__i, 0);
}
-/// \brief Returns the lower 32 bits of a 64-bit integer vector as a 32-bit
+/// Returns the lower 32 bits of a 64-bit integer vector as a 32-bit
/// signed integer.
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c> VMOVD / MOVD </c> instruction.
+/// This intrinsic corresponds to the <c> MOVD </c> instruction.
///
/// \param __m
/// A 64-bit integer vector.
return __builtin_ia32_vec_ext_v2si((__v2si)__m, 0);
}
-/// \brief Casts a 64-bit signed integer value into a 64-bit integer vector.
+/// Casts a 64-bit signed integer value into a 64-bit integer vector.
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c> VMOVQ / MOVD </c> instruction.
+/// This intrinsic corresponds to the <c> MOVQ </c> instruction.
///
/// \param __i
/// A 64-bit signed integer.
return (__m64)__i;
}
-/// \brief Casts a 64-bit integer vector into a 64-bit signed integer value.
+/// Casts a 64-bit integer vector into a 64-bit signed integer value.
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c> VMOVQ / MOVD </c> instruction.
+/// This intrinsic corresponds to the <c> MOVQ </c> instruction.
///
/// \param __m
/// A 64-bit integer vector.
return (long long)__m;
}
-/// \brief Converts 16-bit signed integers from both 64-bit integer vector
+/// Converts 16-bit signed integers from both 64-bit integer vector
/// parameters of [4 x i16] into 8-bit signed integer values, and constructs
/// a 64-bit integer vector of [8 x i8] as the result. Positive values
/// greater than 0x7F are saturated to 0x7F. Negative values less than 0x80
return (__m64)__builtin_ia32_packsswb((__v4hi)__m1, (__v4hi)__m2);
}
-/// \brief Converts 32-bit signed integers from both 64-bit integer vector
+/// Converts 32-bit signed integers from both 64-bit integer vector
/// parameters of [2 x i32] into 16-bit signed integer values, and constructs
/// a 64-bit integer vector of [4 x i16] as the result. Positive values
/// greater than 0x7FFF are saturated to 0x7FFF. Negative values less than
return (__m64)__builtin_ia32_packssdw((__v2si)__m1, (__v2si)__m2);
}
-/// \brief Converts 16-bit signed integers from both 64-bit integer vector
+/// Converts 16-bit signed integers from both 64-bit integer vector
/// parameters of [4 x i16] into 8-bit unsigned integer values, and
/// constructs a 64-bit integer vector of [8 x i8] as the result. Values
/// greater than 0xFF are saturated to 0xFF. Values less than 0 are saturated
return (__m64)__builtin_ia32_packuswb((__v4hi)__m1, (__v4hi)__m2);
}
-/// \brief Unpacks the upper 32 bits from two 64-bit integer vectors of [8 x i8]
+/// Unpacks the upper 32 bits from two 64-bit integer vectors of [8 x i8]
/// and interleaves them into a 64-bit integer vector of [8 x i8].
///
/// \headerfile <x86intrin.h>
return (__m64)__builtin_ia32_punpckhbw((__v8qi)__m1, (__v8qi)__m2);
}
-/// \brief Unpacks the upper 32 bits from two 64-bit integer vectors of
+/// Unpacks the upper 32 bits from two 64-bit integer vectors of
/// [4 x i16] and interleaves them into a 64-bit integer vector of [4 x i16].
///
/// \headerfile <x86intrin.h>
return (__m64)__builtin_ia32_punpckhwd((__v4hi)__m1, (__v4hi)__m2);
}
-/// \brief Unpacks the upper 32 bits from two 64-bit integer vectors of
+/// Unpacks the upper 32 bits from two 64-bit integer vectors of
/// [2 x i32] and interleaves them into a 64-bit integer vector of [2 x i32].
///
/// \headerfile <x86intrin.h>
return (__m64)__builtin_ia32_punpckhdq((__v2si)__m1, (__v2si)__m2);
}
-/// \brief Unpacks the lower 32 bits from two 64-bit integer vectors of [8 x i8]
+/// Unpacks the lower 32 bits from two 64-bit integer vectors of [8 x i8]
/// and interleaves them into a 64-bit integer vector of [8 x i8].
///
/// \headerfile <x86intrin.h>
return (__m64)__builtin_ia32_punpcklbw((__v8qi)__m1, (__v8qi)__m2);
}
-/// \brief Unpacks the lower 32 bits from two 64-bit integer vectors of
+/// Unpacks the lower 32 bits from two 64-bit integer vectors of
/// [4 x i16] and interleaves them into a 64-bit integer vector of [4 x i16].
///
/// \headerfile <x86intrin.h>
return (__m64)__builtin_ia32_punpcklwd((__v4hi)__m1, (__v4hi)__m2);
}
-/// \brief Unpacks the lower 32 bits from two 64-bit integer vectors of
+/// Unpacks the lower 32 bits from two 64-bit integer vectors of
/// [2 x i32] and interleaves them into a 64-bit integer vector of [2 x i32].
///
/// \headerfile <x86intrin.h>
return (__m64)__builtin_ia32_punpckldq((__v2si)__m1, (__v2si)__m2);
}
-/// \brief Adds each 8-bit integer element of the first 64-bit integer vector
+/// Adds each 8-bit integer element of the first 64-bit integer vector
/// of [8 x i8] to the corresponding 8-bit integer element of the second
/// 64-bit integer vector of [8 x i8]. The lower 8 bits of the results are
/// packed into a 64-bit integer vector of [8 x i8].
return (__m64)__builtin_ia32_paddb((__v8qi)__m1, (__v8qi)__m2);
}
-/// \brief Adds each 16-bit integer element of the first 64-bit integer vector
+/// Adds each 16-bit integer element of the first 64-bit integer vector
/// of [4 x i16] to the corresponding 16-bit integer element of the second
/// 64-bit integer vector of [4 x i16]. The lower 16 bits of the results are
/// packed into a 64-bit integer vector of [4 x i16].
return (__m64)__builtin_ia32_paddw((__v4hi)__m1, (__v4hi)__m2);
}
-/// \brief Adds each 32-bit integer element of the first 64-bit integer vector
+/// Adds each 32-bit integer element of the first 64-bit integer vector
/// of [2 x i32] to the corresponding 32-bit integer element of the second
/// 64-bit integer vector of [2 x i32]. The lower 32 bits of the results are
/// packed into a 64-bit integer vector of [2 x i32].
return (__m64)__builtin_ia32_paddd((__v2si)__m1, (__v2si)__m2);
}
-/// \brief Adds each 8-bit signed integer element of the first 64-bit integer
+/// Adds each 8-bit signed integer element of the first 64-bit integer
/// vector of [8 x i8] to the corresponding 8-bit signed integer element of
/// the second 64-bit integer vector of [8 x i8]. Positive sums greater than
/// 0x7F are saturated to 0x7F. Negative sums less than 0x80 are saturated to
return (__m64)__builtin_ia32_paddsb((__v8qi)__m1, (__v8qi)__m2);
}
-/// \brief Adds each 16-bit signed integer element of the first 64-bit integer
+/// Adds each 16-bit signed integer element of the first 64-bit integer
/// vector of [4 x i16] to the corresponding 16-bit signed integer element of
/// the second 64-bit integer vector of [4 x i16]. Positive sums greater than
/// 0x7FFF are saturated to 0x7FFF. Negative sums less than 0x8000 are
return (__m64)__builtin_ia32_paddsw((__v4hi)__m1, (__v4hi)__m2);
}
-/// \brief Adds each 8-bit unsigned integer element of the first 64-bit integer
+/// Adds each 8-bit unsigned integer element of the first 64-bit integer
/// vector of [8 x i8] to the corresponding 8-bit unsigned integer element of
/// the second 64-bit integer vector of [8 x i8]. Sums greater than 0xFF are
/// saturated to 0xFF. The results are packed into a 64-bit integer vector of
return (__m64)__builtin_ia32_paddusb((__v8qi)__m1, (__v8qi)__m2);
}
-/// \brief Adds each 16-bit unsigned integer element of the first 64-bit integer
+/// Adds each 16-bit unsigned integer element of the first 64-bit integer
/// vector of [4 x i16] to the corresponding 16-bit unsigned integer element
/// of the second 64-bit integer vector of [4 x i16]. Sums greater than
/// 0xFFFF are saturated to 0xFFFF. The results are packed into a 64-bit
return (__m64)__builtin_ia32_paddusw((__v4hi)__m1, (__v4hi)__m2);
}
-/// \brief Subtracts each 8-bit integer element of the second 64-bit integer
+/// Subtracts each 8-bit integer element of the second 64-bit integer
/// vector of [8 x i8] from the corresponding 8-bit integer element of the
/// first 64-bit integer vector of [8 x i8]. The lower 8 bits of the results
/// are packed into a 64-bit integer vector of [8 x i8].
return (__m64)__builtin_ia32_psubb((__v8qi)__m1, (__v8qi)__m2);
}
-/// \brief Subtracts each 16-bit integer element of the second 64-bit integer
+/// Subtracts each 16-bit integer element of the second 64-bit integer
/// vector of [4 x i16] from the corresponding 16-bit integer element of the
/// first 64-bit integer vector of [4 x i16]. The lower 16 bits of the
/// results are packed into a 64-bit integer vector of [4 x i16].
return (__m64)__builtin_ia32_psubw((__v4hi)__m1, (__v4hi)__m2);
}
-/// \brief Subtracts each 32-bit integer element of the second 64-bit integer
+/// Subtracts each 32-bit integer element of the second 64-bit integer
/// vector of [2 x i32] from the corresponding 32-bit integer element of the
/// first 64-bit integer vector of [2 x i32]. The lower 32 bits of the
/// results are packed into a 64-bit integer vector of [2 x i32].
return (__m64)__builtin_ia32_psubd((__v2si)__m1, (__v2si)__m2);
}
-/// \brief Subtracts each 8-bit signed integer element of the second 64-bit
+/// Subtracts each 8-bit signed integer element of the second 64-bit
/// integer vector of [8 x i8] from the corresponding 8-bit signed integer
/// element of the first 64-bit integer vector of [8 x i8]. Positive results
/// greater than 0x7F are saturated to 0x7F. Negative results less than 0x80
return (__m64)__builtin_ia32_psubsb((__v8qi)__m1, (__v8qi)__m2);
}
-/// \brief Subtracts each 16-bit signed integer element of the second 64-bit
+/// Subtracts each 16-bit signed integer element of the second 64-bit
/// integer vector of [4 x i16] from the corresponding 16-bit signed integer
/// element of the first 64-bit integer vector of [4 x i16]. Positive results
/// greater than 0x7FFF are saturated to 0x7FFF. Negative results less than
return (__m64)__builtin_ia32_psubsw((__v4hi)__m1, (__v4hi)__m2);
}
-/// \brief Subtracts each 8-bit unsigned integer element of the second 64-bit
+/// Subtracts each 8-bit unsigned integer element of the second 64-bit
/// integer vector of [8 x i8] from the corresponding 8-bit unsigned integer
/// element of the first 64-bit integer vector of [8 x i8].
///
return (__m64)__builtin_ia32_psubusb((__v8qi)__m1, (__v8qi)__m2);
}
-/// \brief Subtracts each 16-bit unsigned integer element of the second 64-bit
+/// Subtracts each 16-bit unsigned integer element of the second 64-bit
/// integer vector of [4 x i16] from the corresponding 16-bit unsigned
/// integer element of the first 64-bit integer vector of [4 x i16].
///
return (__m64)__builtin_ia32_psubusw((__v4hi)__m1, (__v4hi)__m2);
}
-/// \brief Multiplies each 16-bit signed integer element of the first 64-bit
+/// Multiplies each 16-bit signed integer element of the first 64-bit
/// integer vector of [4 x i16] by the corresponding 16-bit signed integer
/// element of the second 64-bit integer vector of [4 x i16] and get four
/// 32-bit products. Adds adjacent pairs of products to get two 32-bit sums.
return (__m64)__builtin_ia32_pmaddwd((__v4hi)__m1, (__v4hi)__m2);
}
-/// \brief Multiplies each 16-bit signed integer element of the first 64-bit
+/// Multiplies each 16-bit signed integer element of the first 64-bit
/// integer vector of [4 x i16] by the corresponding 16-bit signed integer
/// element of the second 64-bit integer vector of [4 x i16]. Packs the upper
/// 16 bits of the 32-bit products into a 64-bit integer vector of [4 x i16].
return (__m64)__builtin_ia32_pmulhw((__v4hi)__m1, (__v4hi)__m2);
}
-/// \brief Multiplies each 16-bit signed integer element of the first 64-bit
+/// Multiplies each 16-bit signed integer element of the first 64-bit
/// integer vector of [4 x i16] by the corresponding 16-bit signed integer
/// element of the second 64-bit integer vector of [4 x i16]. Packs the lower
/// 16 bits of the 32-bit products into a 64-bit integer vector of [4 x i16].
return (__m64)__builtin_ia32_pmullw((__v4hi)__m1, (__v4hi)__m2);
}
-/// \brief Left-shifts each 16-bit signed integer element of the first
+/// Left-shifts each 16-bit signed integer element of the first
/// parameter, which is a 64-bit integer vector of [4 x i16], by the number
/// of bits specified by the second parameter, which is a 64-bit integer. The
/// lower 16 bits of the results are packed into a 64-bit integer vector of
return (__m64)__builtin_ia32_psllw((__v4hi)__m, __count);
}
-/// \brief Left-shifts each 16-bit signed integer element of a 64-bit integer
+/// Left-shifts each 16-bit signed integer element of a 64-bit integer
/// vector of [4 x i16] by the number of bits specified by a 32-bit integer.
/// The lower 16 bits of the results are packed into a 64-bit integer vector
/// of [4 x i16].
return (__m64)__builtin_ia32_psllwi((__v4hi)__m, __count);
}
-/// \brief Left-shifts each 32-bit signed integer element of the first
+/// Left-shifts each 32-bit signed integer element of the first
/// parameter, which is a 64-bit integer vector of [2 x i32], by the number
/// of bits specified by the second parameter, which is a 64-bit integer. The
/// lower 32 bits of the results are packed into a 64-bit integer vector of
return (__m64)__builtin_ia32_pslld((__v2si)__m, __count);
}
-/// \brief Left-shifts each 32-bit signed integer element of a 64-bit integer
+/// Left-shifts each 32-bit signed integer element of a 64-bit integer
/// vector of [2 x i32] by the number of bits specified by a 32-bit integer.
/// The lower 32 bits of the results are packed into a 64-bit integer vector
/// of [2 x i32].
return (__m64)__builtin_ia32_pslldi((__v2si)__m, __count);
}
-/// \brief Left-shifts the first 64-bit integer parameter by the number of bits
+/// Left-shifts the first 64-bit integer parameter by the number of bits
/// specified by the second 64-bit integer parameter. The lower 64 bits of
/// result are returned.
///
return (__m64)__builtin_ia32_psllq((__v1di)__m, __count);
}
-/// \brief Left-shifts the first parameter, which is a 64-bit integer, by the
+/// Left-shifts the first parameter, which is a 64-bit integer, by the
/// number of bits specified by the second parameter, which is a 32-bit
/// integer. The lower 64 bits of result are returned.
///
return (__m64)__builtin_ia32_psllqi((__v1di)__m, __count);
}
-/// \brief Right-shifts each 16-bit integer element of the first parameter,
+/// Right-shifts each 16-bit integer element of the first parameter,
/// which is a 64-bit integer vector of [4 x i16], by the number of bits
/// specified by the second parameter, which is a 64-bit integer.
///
return (__m64)__builtin_ia32_psraw((__v4hi)__m, __count);
}
-/// \brief Right-shifts each 16-bit integer element of a 64-bit integer vector
+/// Right-shifts each 16-bit integer element of a 64-bit integer vector
/// of [4 x i16] by the number of bits specified by a 32-bit integer.
///
/// High-order bits are filled with the sign bit of the initial value of each
return (__m64)__builtin_ia32_psrawi((__v4hi)__m, __count);
}
-/// \brief Right-shifts each 32-bit integer element of the first parameter,
+/// Right-shifts each 32-bit integer element of the first parameter,
/// which is a 64-bit integer vector of [2 x i32], by the number of bits
/// specified by the second parameter, which is a 64-bit integer.
///
return (__m64)__builtin_ia32_psrad((__v2si)__m, __count);
}
-/// \brief Right-shifts each 32-bit integer element of a 64-bit integer vector
+/// Right-shifts each 32-bit integer element of a 64-bit integer vector
/// of [2 x i32] by the number of bits specified by a 32-bit integer.
///
/// High-order bits are filled with the sign bit of the initial value of each
return (__m64)__builtin_ia32_psradi((__v2si)__m, __count);
}
-/// \brief Right-shifts each 16-bit integer element of the first parameter,
+/// Right-shifts each 16-bit integer element of the first parameter,
/// which is a 64-bit integer vector of [4 x i16], by the number of bits
/// specified by the second parameter, which is a 64-bit integer.
///
return (__m64)__builtin_ia32_psrlw((__v4hi)__m, __count);
}
-/// \brief Right-shifts each 16-bit integer element of a 64-bit integer vector
+/// Right-shifts each 16-bit integer element of a 64-bit integer vector
/// of [4 x i16] by the number of bits specified by a 32-bit integer.
///
/// High-order bits are cleared. The 16-bit results are packed into a 64-bit
return (__m64)__builtin_ia32_psrlwi((__v4hi)__m, __count);
}
-/// \brief Right-shifts each 32-bit integer element of the first parameter,
+/// Right-shifts each 32-bit integer element of the first parameter,
/// which is a 64-bit integer vector of [2 x i32], by the number of bits
/// specified by the second parameter, which is a 64-bit integer.
///
return (__m64)__builtin_ia32_psrld((__v2si)__m, __count);
}
-/// \brief Right-shifts each 32-bit integer element of a 64-bit integer vector
+/// Right-shifts each 32-bit integer element of a 64-bit integer vector
/// of [2 x i32] by the number of bits specified by a 32-bit integer.
///
/// High-order bits are cleared. The 32-bit results are packed into a 64-bit
return (__m64)__builtin_ia32_psrldi((__v2si)__m, __count);
}
-/// \brief Right-shifts the first 64-bit integer parameter by the number of bits
+/// Right-shifts the first 64-bit integer parameter by the number of bits
/// specified by the second 64-bit integer parameter.
///
/// High-order bits are cleared.
return (__m64)__builtin_ia32_psrlq((__v1di)__m, __count);
}
-/// \brief Right-shifts the first parameter, which is a 64-bit integer, by the
+/// Right-shifts the first parameter, which is a 64-bit integer, by the
/// number of bits specified by the second parameter, which is a 32-bit
/// integer.
///
return (__m64)__builtin_ia32_psrlqi((__v1di)__m, __count);
}
-/// \brief Performs a bitwise AND of two 64-bit integer vectors.
+/// Performs a bitwise AND of two 64-bit integer vectors.
///
/// \headerfile <x86intrin.h>
///
return __builtin_ia32_pand((__v1di)__m1, (__v1di)__m2);
}
-/// \brief Performs a bitwise NOT of the first 64-bit integer vector, and then
+/// Performs a bitwise NOT of the first 64-bit integer vector, and then
/// performs a bitwise AND of the intermediate result and the second 64-bit
/// integer vector.
///
return __builtin_ia32_pandn((__v1di)__m1, (__v1di)__m2);
}
-/// \brief Performs a bitwise OR of two 64-bit integer vectors.
+/// Performs a bitwise OR of two 64-bit integer vectors.
///
/// \headerfile <x86intrin.h>
///
return __builtin_ia32_por((__v1di)__m1, (__v1di)__m2);
}
-/// \brief Performs a bitwise exclusive OR of two 64-bit integer vectors.
+/// Performs a bitwise exclusive OR of two 64-bit integer vectors.
///
/// \headerfile <x86intrin.h>
///
return __builtin_ia32_pxor((__v1di)__m1, (__v1di)__m2);
}
-/// \brief Compares the 8-bit integer elements of two 64-bit integer vectors of
+/// Compares the 8-bit integer elements of two 64-bit integer vectors of
/// [8 x i8] to determine if the element of the first vector is equal to the
/// corresponding element of the second vector.
///
return (__m64)__builtin_ia32_pcmpeqb((__v8qi)__m1, (__v8qi)__m2);
}
-/// \brief Compares the 16-bit integer elements of two 64-bit integer vectors of
+/// Compares the 16-bit integer elements of two 64-bit integer vectors of
/// [4 x i16] to determine if the element of the first vector is equal to the
/// corresponding element of the second vector.
///
return (__m64)__builtin_ia32_pcmpeqw((__v4hi)__m1, (__v4hi)__m2);
}
-/// \brief Compares the 32-bit integer elements of two 64-bit integer vectors of
+/// Compares the 32-bit integer elements of two 64-bit integer vectors of
/// [2 x i32] to determine if the element of the first vector is equal to the
/// corresponding element of the second vector.
///
return (__m64)__builtin_ia32_pcmpeqd((__v2si)__m1, (__v2si)__m2);
}
-/// \brief Compares the 8-bit integer elements of two 64-bit integer vectors of
+/// Compares the 8-bit integer elements of two 64-bit integer vectors of
/// [8 x i8] to determine if the element of the first vector is greater than
/// the corresponding element of the second vector.
///
return (__m64)__builtin_ia32_pcmpgtb((__v8qi)__m1, (__v8qi)__m2);
}
-/// \brief Compares the 16-bit integer elements of two 64-bit integer vectors of
+/// Compares the 16-bit integer elements of two 64-bit integer vectors of
/// [4 x i16] to determine if the element of the first vector is greater than
/// the corresponding element of the second vector.
///
return (__m64)__builtin_ia32_pcmpgtw((__v4hi)__m1, (__v4hi)__m2);
}
-/// \brief Compares the 32-bit integer elements of two 64-bit integer vectors of
+/// Compares the 32-bit integer elements of two 64-bit integer vectors of
/// [2 x i32] to determine if the element of the first vector is greater than
/// the corresponding element of the second vector.
///
return (__m64)__builtin_ia32_pcmpgtd((__v2si)__m1, (__v2si)__m2);
}
-/// \brief Constructs a 64-bit integer vector initialized to zero.
+/// Constructs a 64-bit integer vector initialized to zero.
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c> VXORPS / XORPS </c> instruction.
+/// This intrinsic corresponds to the <c> PXOR </c> instruction.
///
/// \returns An initialized 64-bit integer vector with all elements set to zero.
static __inline__ __m64 __DEFAULT_FN_ATTRS
_mm_setzero_si64(void)
{
- return (__m64){ 0LL };
+ return __extension__ (__m64){ 0LL };
}
-/// \brief Constructs a 64-bit integer vector initialized with the specified
+/// Constructs a 64-bit integer vector initialized with the specified
/// 32-bit integer values.
///
/// \headerfile <x86intrin.h>
return (__m64)__builtin_ia32_vec_init_v2si(__i0, __i1);
}
-/// \brief Constructs a 64-bit integer vector initialized with the specified
+/// Constructs a 64-bit integer vector initialized with the specified
/// 16-bit integer values.
///
/// \headerfile <x86intrin.h>
return (__m64)__builtin_ia32_vec_init_v4hi(__s0, __s1, __s2, __s3);
}
-/// \brief Constructs a 64-bit integer vector initialized with the specified
+/// Constructs a 64-bit integer vector initialized with the specified
/// 8-bit integer values.
///
/// \headerfile <x86intrin.h>
__b4, __b5, __b6, __b7);
}
-/// \brief Constructs a 64-bit integer vector of [2 x i32], with each of the
+/// Constructs a 64-bit integer vector of [2 x i32], with each of the
/// 32-bit integer vector elements set to the specified 32-bit integer
/// value.
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c> VPSHUFD / PSHUFD </c> instruction.
+/// This intrinsic is a utility function and does not correspond to a specific
+/// instruction.
///
/// \param __i
/// A 32-bit integer value used to initialize each vector element of the
return _mm_set_pi32(__i, __i);
}
-/// \brief Constructs a 64-bit integer vector of [4 x i16], with each of the
+/// Constructs a 64-bit integer vector of [4 x i16], with each of the
/// 16-bit integer vector elements set to the specified 16-bit integer
/// value.
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c> VPSHUFLW / PSHUFLW </c> instruction.
+/// This intrinsic is a utility function and does not correspond to a specific
+/// instruction.
///
/// \param __w
/// A 16-bit integer value used to initialize each vector element of the
return _mm_set_pi16(__w, __w, __w, __w);
}
-/// \brief Constructs a 64-bit integer vector of [8 x i8], with each of the
+/// Constructs a 64-bit integer vector of [8 x i8], with each of the
/// 8-bit integer vector elements set to the specified 8-bit integer value.
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c> VPUNPCKLBW + VPSHUFLW / PUNPCKLBW +
-/// PSHUFLW </c> instruction.
+/// This intrinsic is a utility function and does not correspond to a specific
+/// instruction.
///
/// \param __b
/// An 8-bit integer value used to initialize each vector element of the
return _mm_set_pi8(__b, __b, __b, __b, __b, __b, __b, __b);
}
-/// \brief Constructs a 64-bit integer vector, initialized in reverse order with
+/// Constructs a 64-bit integer vector, initialized in reverse order with
/// the specified 32-bit integer values.
///
/// \headerfile <x86intrin.h>
return _mm_set_pi32(__i1, __i0);
}
-/// \brief Constructs a 64-bit integer vector, initialized in reverse order with
+/// Constructs a 64-bit integer vector, initialized in reverse order with
/// the specified 16-bit integer values.
///
/// \headerfile <x86intrin.h>
return _mm_set_pi16(__w3, __w2, __w1, __w0);
}
-/// \brief Constructs a 64-bit integer vector, initialized in reverse order with
+/// Constructs a 64-bit integer vector, initialized in reverse order with
/// the specified 8-bit integer values.
///
/// \headerfile <x86intrin.h>
/*===---- module.modulemap - intrinsics module map -------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
explicit module neon {
requires neon
header "arm_neon.h"
+ header "arm_fp16.h"
+ export *
+ }
+
+ explicit module sve {
+ requires sve
+ header "arm_sve.h"
export *
}
}
textual header "fma4intrin.h"
textual header "mwaitxintrin.h"
textual header "clzerointrin.h"
+ textual header "wbnoinvdintrin.h"
+ textual header "cldemoteintrin.h"
+ textual header "waitpkgintrin.h"
+ textual header "movdirintrin.h"
+ textual header "pconfigintrin.h"
+ textual header "sgxintrin.h"
+ textual header "ptwriteintrin.h"
+ textual header "invpcidintrin.h"
+
+ textual header "__wmmintrin_aes.h"
+ textual header "__wmmintrin_pclmul.h"
explicit module mm_malloc {
requires !freestanding
export aes
export pclmul
}
-
- explicit module aes {
- header "__wmmintrin_aes.h"
- }
-
- explicit module pclmul {
- header "__wmmintrin_pclmul.h"
- }
}
explicit module systemz {
module opencl_c {
requires opencl
header "opencl-c.h"
+ header "opencl-c-base.h"
}
--- /dev/null
+/*===------------------------- movdirintrin.h ------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#if !defined __X86INTRIN_H && !defined __IMMINTRIN_H
+#error "Never use <movdirintrin.h> directly; include <x86intrin.h> instead."
+#endif
+
+#ifndef _MOVDIRINTRIN_H
+#define _MOVDIRINTRIN_H
+
+/* Move doubleword as direct store */
+static __inline__ void
+__attribute__((__always_inline__, __nodebug__, __target__("movdiri")))
+_directstoreu_u32 (void *__dst, unsigned int __value)
+{
+ __builtin_ia32_directstore_u32((unsigned int *)__dst, (unsigned int)__value);
+}
+
+#ifdef __x86_64__
+
+/* Move quadword as direct store */
+static __inline__ void
+__attribute__((__always_inline__, __nodebug__, __target__("movdiri")))
+_directstoreu_u64 (void *__dst, unsigned long __value)
+{
+ __builtin_ia32_directstore_u64((unsigned long *)__dst, __value);
+}
+
+#endif /* __x86_64__ */
+
+/*
+ * movdir64b - Move 64 bytes as direct store.
+ * The destination must be 64 byte aligned, and the store is atomic.
+ * The source address has no alignment requirement, and the load from
+ * the source address is not atomic.
+ */
+static __inline__ void
+__attribute__((__always_inline__, __nodebug__, __target__("movdir64b")))
+_movdir64b (void *__dst __attribute__((align_value(64))), const void *__src)
+{
+ __builtin_ia32_movdir64b(__dst, __src);
+}
+
+#endif /* _MOVDIRINTRIN_H */
/*===---- msa.h - MIPS MSA intrinsics --------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#define __msa_ld_h __builtin_msa_ld_h
#define __msa_ld_w __builtin_msa_ld_w
#define __msa_ld_d __builtin_msa_ld_d
+#define __msa_ldr_d __builtin_msa_ldr_d
+#define __msa_ldr_w __builtin_msa_ldrq_w
#define __msa_st_b __builtin_msa_st_b
#define __msa_st_h __builtin_msa_st_h
#define __msa_st_w __builtin_msa_st_w
#define __msa_st_d __builtin_msa_st_d
+#define __msa_str_d __builtin_msa_str_d
+#define __msa_str_w __builtin_msa_strq_w
#define __msa_sat_s_b __builtin_msa_sat_s_b
#define __msa_sat_s_h __builtin_msa_sat_s_h
#define __msa_sat_s_w __builtin_msa_sat_s_w
--- /dev/null
+/*===---- mwaitxintrin.h - MONITORX/MWAITX intrinsics ----------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __X86INTRIN_H
+#error "Never use <mwaitxintrin.h> directly; include <x86intrin.h> instead."
+#endif
+
+#ifndef __MWAITXINTRIN_H
+#define __MWAITXINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("mwaitx")))
+static __inline__ void __DEFAULT_FN_ATTRS
+_mm_monitorx(void * __p, unsigned __extensions, unsigned __hints)
+{
+ __builtin_ia32_monitorx(__p, __extensions, __hints);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_mm_mwaitx(unsigned __extensions, unsigned __hints, unsigned __clock)
+{
+ __builtin_ia32_mwaitx(__extensions, __hints, __clock);
+}
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif /* __MWAITXINTRIN_H */
--- /dev/null
+/*===---- nmmintrin.h - SSE4 intrinsics ------------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __NMMINTRIN_H
+#define __NMMINTRIN_H
+
+/* To match expectations of gcc we put the sse4.2 definitions into smmintrin.h,
+ just include it now then. */
+#include <smmintrin.h>
+#endif /* __NMMINTRIN_H */
--- /dev/null
+//===----- opencl-c-base.h - OpenCL C language base definitions -----------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef _OPENCL_BASE_H_
+#define _OPENCL_BASE_H_
+
+// built-in scalar data types:
+
+/**
+ * An unsigned 8-bit integer.
+ */
+typedef unsigned char uchar;
+
+/**
+ * An unsigned 16-bit integer.
+ */
+typedef unsigned short ushort;
+
+/**
+ * An unsigned 32-bit integer.
+ */
+typedef unsigned int uint;
+
+/**
+ * An unsigned 64-bit integer.
+ */
+typedef unsigned long ulong;
+
+/**
+ * The unsigned integer type of the result of the sizeof operator. This
+ * is a 32-bit unsigned integer if CL_DEVICE_ADDRESS_BITS
+ * defined in table 4.3 is 32-bits and is a 64-bit unsigned integer if
+ * CL_DEVICE_ADDRESS_BITS is 64-bits.
+ */
+typedef __SIZE_TYPE__ size_t;
+
+/**
+ * A signed integer type that is the result of subtracting two pointers.
+ * This is a 32-bit signed integer if CL_DEVICE_ADDRESS_BITS
+ * defined in table 4.3 is 32-bits and is a 64-bit signed integer if
+ * CL_DEVICE_ADDRESS_BITS is 64-bits.
+ */
+typedef __PTRDIFF_TYPE__ ptrdiff_t;
+
+/**
+ * A signed integer type with the property that any valid pointer to
+ * void can be converted to this type, then converted back to pointer
+ * to void, and the result will compare equal to the original pointer.
+ */
+typedef __INTPTR_TYPE__ intptr_t;
+
+/**
+ * An unsigned integer type with the property that any valid pointer to
+ * void can be converted to this type, then converted back to pointer
+ * to void, and the result will compare equal to the original pointer.
+ */
+typedef __UINTPTR_TYPE__ uintptr_t;
+
+// built-in vector data types:
+typedef char char2 __attribute__((ext_vector_type(2)));
+typedef char char3 __attribute__((ext_vector_type(3)));
+typedef char char4 __attribute__((ext_vector_type(4)));
+typedef char char8 __attribute__((ext_vector_type(8)));
+typedef char char16 __attribute__((ext_vector_type(16)));
+typedef uchar uchar2 __attribute__((ext_vector_type(2)));
+typedef uchar uchar3 __attribute__((ext_vector_type(3)));
+typedef uchar uchar4 __attribute__((ext_vector_type(4)));
+typedef uchar uchar8 __attribute__((ext_vector_type(8)));
+typedef uchar uchar16 __attribute__((ext_vector_type(16)));
+typedef short short2 __attribute__((ext_vector_type(2)));
+typedef short short3 __attribute__((ext_vector_type(3)));
+typedef short short4 __attribute__((ext_vector_type(4)));
+typedef short short8 __attribute__((ext_vector_type(8)));
+typedef short short16 __attribute__((ext_vector_type(16)));
+typedef ushort ushort2 __attribute__((ext_vector_type(2)));
+typedef ushort ushort3 __attribute__((ext_vector_type(3)));
+typedef ushort ushort4 __attribute__((ext_vector_type(4)));
+typedef ushort ushort8 __attribute__((ext_vector_type(8)));
+typedef ushort ushort16 __attribute__((ext_vector_type(16)));
+typedef int int2 __attribute__((ext_vector_type(2)));
+typedef int int3 __attribute__((ext_vector_type(3)));
+typedef int int4 __attribute__((ext_vector_type(4)));
+typedef int int8 __attribute__((ext_vector_type(8)));
+typedef int int16 __attribute__((ext_vector_type(16)));
+typedef uint uint2 __attribute__((ext_vector_type(2)));
+typedef uint uint3 __attribute__((ext_vector_type(3)));
+typedef uint uint4 __attribute__((ext_vector_type(4)));
+typedef uint uint8 __attribute__((ext_vector_type(8)));
+typedef uint uint16 __attribute__((ext_vector_type(16)));
+typedef long long2 __attribute__((ext_vector_type(2)));
+typedef long long3 __attribute__((ext_vector_type(3)));
+typedef long long4 __attribute__((ext_vector_type(4)));
+typedef long long8 __attribute__((ext_vector_type(8)));
+typedef long long16 __attribute__((ext_vector_type(16)));
+typedef ulong ulong2 __attribute__((ext_vector_type(2)));
+typedef ulong ulong3 __attribute__((ext_vector_type(3)));
+typedef ulong ulong4 __attribute__((ext_vector_type(4)));
+typedef ulong ulong8 __attribute__((ext_vector_type(8)));
+typedef ulong ulong16 __attribute__((ext_vector_type(16)));
+typedef float float2 __attribute__((ext_vector_type(2)));
+typedef float float3 __attribute__((ext_vector_type(3)));
+typedef float float4 __attribute__((ext_vector_type(4)));
+typedef float float8 __attribute__((ext_vector_type(8)));
+typedef float float16 __attribute__((ext_vector_type(16)));
+#ifdef cl_khr_fp16
+#pragma OPENCL EXTENSION cl_khr_fp16 : enable
+typedef half half2 __attribute__((ext_vector_type(2)));
+typedef half half3 __attribute__((ext_vector_type(3)));
+typedef half half4 __attribute__((ext_vector_type(4)));
+typedef half half8 __attribute__((ext_vector_type(8)));
+typedef half half16 __attribute__((ext_vector_type(16)));
+#endif
+#ifdef cl_khr_fp64
+#if __OPENCL_C_VERSION__ < CL_VERSION_1_2
+#pragma OPENCL EXTENSION cl_khr_fp64 : enable
+#endif
+typedef double double2 __attribute__((ext_vector_type(2)));
+typedef double double3 __attribute__((ext_vector_type(3)));
+typedef double double4 __attribute__((ext_vector_type(4)));
+typedef double double8 __attribute__((ext_vector_type(8)));
+typedef double double16 __attribute__((ext_vector_type(16)));
+#endif
+
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+#define NULL ((void*)0)
+#endif
+
+/**
+ * Value of maximum non-infinite single-precision floating-point
+ * number.
+ */
+#define MAXFLOAT 0x1.fffffep127f
+
+/**
+ * A positive float constant expression. HUGE_VALF evaluates
+ * to +infinity. Used as an error value returned by the built-in
+ * math functions.
+ */
+#define HUGE_VALF (__builtin_huge_valf())
+
+/**
+ * A positive double constant expression. HUGE_VAL evaluates
+ * to +infinity. Used as an error value returned by the built-in
+ * math functions.
+ */
+#define HUGE_VAL (__builtin_huge_val())
+
+/**
+ * A constant expression of type float representing positive or
+ * unsigned infinity.
+ */
+#define INFINITY (__builtin_inff())
+
+/**
+ * A constant expression of type float representing a quiet NaN.
+ */
+#define NAN as_float(INT_MAX)
+
+#define FP_ILOGB0 INT_MIN
+#define FP_ILOGBNAN INT_MAX
+
+#define FLT_DIG 6
+#define FLT_MANT_DIG 24
+#define FLT_MAX_10_EXP +38
+#define FLT_MAX_EXP +128
+#define FLT_MIN_10_EXP -37
+#define FLT_MIN_EXP -125
+#define FLT_RADIX 2
+#define FLT_MAX 0x1.fffffep127f
+#define FLT_MIN 0x1.0p-126f
+#define FLT_EPSILON 0x1.0p-23f
+
+#define M_E_F 2.71828182845904523536028747135266250f
+#define M_LOG2E_F 1.44269504088896340735992468100189214f
+#define M_LOG10E_F 0.434294481903251827651128918916605082f
+#define M_LN2_F 0.693147180559945309417232121458176568f
+#define M_LN10_F 2.30258509299404568401799145468436421f
+#define M_PI_F 3.14159265358979323846264338327950288f
+#define M_PI_2_F 1.57079632679489661923132169163975144f
+#define M_PI_4_F 0.785398163397448309615660845819875721f
+#define M_1_PI_F 0.318309886183790671537767526745028724f
+#define M_2_PI_F 0.636619772367581343075535053490057448f
+#define M_2_SQRTPI_F 1.12837916709551257389615890312154517f
+#define M_SQRT2_F 1.41421356237309504880168872420969808f
+#define M_SQRT1_2_F 0.707106781186547524400844362104849039f
+
+#define DBL_DIG 15
+#define DBL_MANT_DIG 53
+#define DBL_MAX_10_EXP +308
+#define DBL_MAX_EXP +1024
+#define DBL_MIN_10_EXP -307
+#define DBL_MIN_EXP -1021
+#define DBL_RADIX 2
+#define DBL_MAX 0x1.fffffffffffffp1023
+#define DBL_MIN 0x1.0p-1022
+#define DBL_EPSILON 0x1.0p-52
+
+#define M_E 0x1.5bf0a8b145769p+1
+#define M_LOG2E 0x1.71547652b82fep+0
+#define M_LOG10E 0x1.bcb7b1526e50ep-2
+#define M_LN2 0x1.62e42fefa39efp-1
+#define M_LN10 0x1.26bb1bbb55516p+1
+#define M_PI 0x1.921fb54442d18p+1
+#define M_PI_2 0x1.921fb54442d18p+0
+#define M_PI_4 0x1.921fb54442d18p-1
+#define M_1_PI 0x1.45f306dc9c883p-2
+#define M_2_PI 0x1.45f306dc9c883p-1
+#define M_2_SQRTPI 0x1.20dd750429b6dp+0
+#define M_SQRT2 0x1.6a09e667f3bcdp+0
+#define M_SQRT1_2 0x1.6a09e667f3bcdp-1
+
+#ifdef cl_khr_fp16
+
+#define HALF_DIG 3
+#define HALF_MANT_DIG 11
+#define HALF_MAX_10_EXP +4
+#define HALF_MAX_EXP +16
+#define HALF_MIN_10_EXP -4
+#define HALF_MIN_EXP -13
+#define HALF_RADIX 2
+#define HALF_MAX ((0x1.ffcp15h))
+#define HALF_MIN ((0x1.0p-14h))
+#define HALF_EPSILON ((0x1.0p-10h))
+
+#define M_E_H 2.71828182845904523536028747135266250h
+#define M_LOG2E_H 1.44269504088896340735992468100189214h
+#define M_LOG10E_H 0.434294481903251827651128918916605082h
+#define M_LN2_H 0.693147180559945309417232121458176568h
+#define M_LN10_H 2.30258509299404568401799145468436421h
+#define M_PI_H 3.14159265358979323846264338327950288h
+#define M_PI_2_H 1.57079632679489661923132169163975144h
+#define M_PI_4_H 0.785398163397448309615660845819875721h
+#define M_1_PI_H 0.318309886183790671537767526745028724h
+#define M_2_PI_H 0.636619772367581343075535053490057448h
+#define M_2_SQRTPI_H 1.12837916709551257389615890312154517h
+#define M_SQRT2_H 1.41421356237309504880168872420969808h
+#define M_SQRT1_2_H 0.707106781186547524400844362104849039h
+
+#endif //cl_khr_fp16
+
+#define CHAR_BIT 8
+#define SCHAR_MAX 127
+#define SCHAR_MIN (-128)
+#define UCHAR_MAX 255
+#define CHAR_MAX SCHAR_MAX
+#define CHAR_MIN SCHAR_MIN
+#define USHRT_MAX 65535
+#define SHRT_MAX 32767
+#define SHRT_MIN (-32768)
+#define UINT_MAX 0xffffffff
+#define INT_MAX 2147483647
+#define INT_MIN (-2147483647-1)
+#define ULONG_MAX 0xffffffffffffffffUL
+#define LONG_MAX 0x7fffffffffffffffL
+#define LONG_MIN (-0x7fffffffffffffffL-1)
+
+// OpenCL v1.1 s6.11.8, v1.2 s6.12.8, v2.0 s6.13.8 - Synchronization Functions
+
+// Flag type and values for barrier, mem_fence, read_mem_fence, write_mem_fence
+typedef uint cl_mem_fence_flags;
+
+/**
+ * Queue a memory fence to ensure correct
+ * ordering of memory operations to local memory
+ */
+#define CLK_LOCAL_MEM_FENCE 0x01
+
+/**
+ * Queue a memory fence to ensure correct
+ * ordering of memory operations to global memory
+ */
+#define CLK_GLOBAL_MEM_FENCE 0x02
+
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+
+typedef enum memory_scope {
+ memory_scope_work_item = __OPENCL_MEMORY_SCOPE_WORK_ITEM,
+ memory_scope_work_group = __OPENCL_MEMORY_SCOPE_WORK_GROUP,
+ memory_scope_device = __OPENCL_MEMORY_SCOPE_DEVICE,
+ memory_scope_all_svm_devices = __OPENCL_MEMORY_SCOPE_ALL_SVM_DEVICES,
+#if defined(cl_intel_subgroups) || defined(cl_khr_subgroups)
+ memory_scope_sub_group = __OPENCL_MEMORY_SCOPE_SUB_GROUP
+#endif
+} memory_scope;
+
+/**
+ * Queue a memory fence to ensure correct ordering of memory
+ * operations between work-items of a work-group to
+ * image memory.
+ */
+#define CLK_IMAGE_MEM_FENCE 0x04
+
+#ifndef ATOMIC_VAR_INIT
+#define ATOMIC_VAR_INIT(x) (x)
+#endif //ATOMIC_VAR_INIT
+#define ATOMIC_FLAG_INIT 0
+
+// enum values aligned with what clang uses in EmitAtomicExpr()
+typedef enum memory_order
+{
+ memory_order_relaxed = __ATOMIC_RELAXED,
+ memory_order_acquire = __ATOMIC_ACQUIRE,
+ memory_order_release = __ATOMIC_RELEASE,
+ memory_order_acq_rel = __ATOMIC_ACQ_REL,
+ memory_order_seq_cst = __ATOMIC_SEQ_CST
+} memory_order;
+
+#endif // defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+
+// OpenCL v1.1 s6.11.3, v1.2 s6.12.14, v2.0 s6.13.14 - Image Read and Write Functions
+
+// These values need to match the runtime equivalent
+//
+// Addressing Mode.
+//
+#define CLK_ADDRESS_NONE 0
+#define CLK_ADDRESS_CLAMP_TO_EDGE 2
+#define CLK_ADDRESS_CLAMP 4
+#define CLK_ADDRESS_REPEAT 6
+#define CLK_ADDRESS_MIRRORED_REPEAT 8
+
+//
+// Coordination Normalization
+//
+#define CLK_NORMALIZED_COORDS_FALSE 0
+#define CLK_NORMALIZED_COORDS_TRUE 1
+
+//
+// Filtering Mode.
+//
+#define CLK_FILTER_NEAREST 0x10
+#define CLK_FILTER_LINEAR 0x20
+
+#ifdef cl_khr_gl_msaa_sharing
+#pragma OPENCL EXTENSION cl_khr_gl_msaa_sharing : enable
+#endif //cl_khr_gl_msaa_sharing
+
+//
+// Channel Datatype.
+//
+#define CLK_SNORM_INT8 0x10D0
+#define CLK_SNORM_INT16 0x10D1
+#define CLK_UNORM_INT8 0x10D2
+#define CLK_UNORM_INT16 0x10D3
+#define CLK_UNORM_SHORT_565 0x10D4
+#define CLK_UNORM_SHORT_555 0x10D5
+#define CLK_UNORM_INT_101010 0x10D6
+#define CLK_SIGNED_INT8 0x10D7
+#define CLK_SIGNED_INT16 0x10D8
+#define CLK_SIGNED_INT32 0x10D9
+#define CLK_UNSIGNED_INT8 0x10DA
+#define CLK_UNSIGNED_INT16 0x10DB
+#define CLK_UNSIGNED_INT32 0x10DC
+#define CLK_HALF_FLOAT 0x10DD
+#define CLK_FLOAT 0x10DE
+#define CLK_UNORM_INT24 0x10DF
+
+// Channel order, numbering must be aligned with cl_channel_order in cl.h
+//
+#define CLK_R 0x10B0
+#define CLK_A 0x10B1
+#define CLK_RG 0x10B2
+#define CLK_RA 0x10B3
+#define CLK_RGB 0x10B4
+#define CLK_RGBA 0x10B5
+#define CLK_BGRA 0x10B6
+#define CLK_ARGB 0x10B7
+#define CLK_INTENSITY 0x10B8
+#define CLK_LUMINANCE 0x10B9
+#define CLK_Rx 0x10BA
+#define CLK_RGx 0x10BB
+#define CLK_RGBx 0x10BC
+#define CLK_DEPTH 0x10BD
+#define CLK_DEPTH_STENCIL 0x10BE
+#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#define CLK_sRGB 0x10BF
+#define CLK_sRGBx 0x10C0
+#define CLK_sRGBA 0x10C1
+#define CLK_sBGRA 0x10C2
+#define CLK_ABGR 0x10C3
+#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+
+// OpenCL v2.0 s6.13.16 - Pipe Functions
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+#define CLK_NULL_RESERVE_ID (__builtin_astype(((void*)(__SIZE_MAX__)), reserve_id_t))
+
+// OpenCL v2.0 s6.13.17 - Enqueue Kernels
+#define CL_COMPLETE 0x0
+#define CL_RUNNING 0x1
+#define CL_SUBMITTED 0x2
+#define CL_QUEUED 0x3
+
+#define CLK_SUCCESS 0
+#define CLK_ENQUEUE_FAILURE -101
+#define CLK_INVALID_QUEUE -102
+#define CLK_INVALID_NDRANGE -160
+#define CLK_INVALID_EVENT_WAIT_LIST -57
+#define CLK_DEVICE_QUEUE_FULL -161
+#define CLK_INVALID_ARG_SIZE -51
+#define CLK_EVENT_ALLOCATION_FAILURE -100
+#define CLK_OUT_OF_RESOURCES -5
+
+#define CLK_NULL_QUEUE 0
+#define CLK_NULL_EVENT (__builtin_astype(((__SIZE_MAX__)), clk_event_t))
+
+// execution model related definitions
+#define CLK_ENQUEUE_FLAGS_NO_WAIT 0x0
+#define CLK_ENQUEUE_FLAGS_WAIT_KERNEL 0x1
+#define CLK_ENQUEUE_FLAGS_WAIT_WORK_GROUP 0x2
+
+typedef int kernel_enqueue_flags_t;
+typedef int clk_profiling_info;
+
+// Profiling info name (see capture_event_profiling_info)
+#define CLK_PROFILING_COMMAND_EXEC_TIME 0x1
+
+#define MAX_WORK_DIM 3
+
+typedef struct {
+ unsigned int workDimension;
+ size_t globalWorkOffset[MAX_WORK_DIM];
+ size_t globalWorkSize[MAX_WORK_DIM];
+ size_t localWorkSize[MAX_WORK_DIM];
+} ndrange_t;
+
+#endif // defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+
+#ifdef cl_intel_device_side_avc_motion_estimation
+#pragma OPENCL EXTENSION cl_intel_device_side_avc_motion_estimation : begin
+
+#define CLK_AVC_ME_MAJOR_16x16_INTEL 0x0
+#define CLK_AVC_ME_MAJOR_16x8_INTEL 0x1
+#define CLK_AVC_ME_MAJOR_8x16_INTEL 0x2
+#define CLK_AVC_ME_MAJOR_8x8_INTEL 0x3
+
+#define CLK_AVC_ME_MINOR_8x8_INTEL 0x0
+#define CLK_AVC_ME_MINOR_8x4_INTEL 0x1
+#define CLK_AVC_ME_MINOR_4x8_INTEL 0x2
+#define CLK_AVC_ME_MINOR_4x4_INTEL 0x3
+
+#define CLK_AVC_ME_MAJOR_FORWARD_INTEL 0x0
+#define CLK_AVC_ME_MAJOR_BACKWARD_INTEL 0x1
+#define CLK_AVC_ME_MAJOR_BIDIRECTIONAL_INTEL 0x2
+
+#define CLK_AVC_ME_PARTITION_MASK_ALL_INTEL 0x0
+#define CLK_AVC_ME_PARTITION_MASK_16x16_INTEL 0x7E
+#define CLK_AVC_ME_PARTITION_MASK_16x8_INTEL 0x7D
+#define CLK_AVC_ME_PARTITION_MASK_8x16_INTEL 0x7B
+#define CLK_AVC_ME_PARTITION_MASK_8x8_INTEL 0x77
+#define CLK_AVC_ME_PARTITION_MASK_8x4_INTEL 0x6F
+#define CLK_AVC_ME_PARTITION_MASK_4x8_INTEL 0x5F
+#define CLK_AVC_ME_PARTITION_MASK_4x4_INTEL 0x3F
+
+#define CLK_AVC_ME_SLICE_TYPE_PRED_INTEL 0x0
+#define CLK_AVC_ME_SLICE_TYPE_BPRED_INTEL 0x1
+#define CLK_AVC_ME_SLICE_TYPE_INTRA_INTEL 0x2
+
+#define CLK_AVC_ME_SEARCH_WINDOW_EXHAUSTIVE_INTEL 0x0
+#define CLK_AVC_ME_SEARCH_WINDOW_SMALL_INTEL 0x1
+#define CLK_AVC_ME_SEARCH_WINDOW_TINY_INTEL 0x2
+#define CLK_AVC_ME_SEARCH_WINDOW_EXTRA_TINY_INTEL 0x3
+#define CLK_AVC_ME_SEARCH_WINDOW_DIAMOND_INTEL 0x4
+#define CLK_AVC_ME_SEARCH_WINDOW_LARGE_DIAMOND_INTEL 0x5
+#define CLK_AVC_ME_SEARCH_WINDOW_RESERVED0_INTEL 0x6
+#define CLK_AVC_ME_SEARCH_WINDOW_RESERVED1_INTEL 0x7
+#define CLK_AVC_ME_SEARCH_WINDOW_CUSTOM_INTEL 0x8
+
+#define CLK_AVC_ME_SAD_ADJUST_MODE_NONE_INTEL 0x0
+#define CLK_AVC_ME_SAD_ADJUST_MODE_HAAR_INTEL 0x2
+
+#define CLK_AVC_ME_SUBPIXEL_MODE_INTEGER_INTEL 0x0
+#define CLK_AVC_ME_SUBPIXEL_MODE_HPEL_INTEL 0x1
+#define CLK_AVC_ME_SUBPIXEL_MODE_QPEL_INTEL 0x3
+
+#define CLK_AVC_ME_COST_PRECISION_QPEL_INTEL 0x0
+#define CLK_AVC_ME_COST_PRECISION_HPEL_INTEL 0x1
+#define CLK_AVC_ME_COST_PRECISION_PEL_INTEL 0x2
+#define CLK_AVC_ME_COST_PRECISION_DPEL_INTEL 0x3
+
+#define CLK_AVC_ME_BIDIR_WEIGHT_QUARTER_INTEL 0x10
+#define CLK_AVC_ME_BIDIR_WEIGHT_THIRD_INTEL 0x15
+#define CLK_AVC_ME_BIDIR_WEIGHT_HALF_INTEL 0x20
+#define CLK_AVC_ME_BIDIR_WEIGHT_TWO_THIRD_INTEL 0x2B
+#define CLK_AVC_ME_BIDIR_WEIGHT_THREE_QUARTER_INTEL 0x30
+
+#define CLK_AVC_ME_BORDER_REACHED_LEFT_INTEL 0x0
+#define CLK_AVC_ME_BORDER_REACHED_RIGHT_INTEL 0x2
+#define CLK_AVC_ME_BORDER_REACHED_TOP_INTEL 0x4
+#define CLK_AVC_ME_BORDER_REACHED_BOTTOM_INTEL 0x8
+
+#define CLK_AVC_ME_INTRA_16x16_INTEL 0x0
+#define CLK_AVC_ME_INTRA_8x8_INTEL 0x1
+#define CLK_AVC_ME_INTRA_4x4_INTEL 0x2
+
+#define CLK_AVC_ME_SKIP_BLOCK_PARTITION_16x16_INTEL 0x0
+#define CLK_AVC_ME_SKIP_BLOCK_PARTITION_8x8_INTEL 0x4000
+
+#define CLK_AVC_ME_SKIP_BLOCK_16x16_FORWARD_ENABLE_INTEL (0x1 << 24)
+#define CLK_AVC_ME_SKIP_BLOCK_16x16_BACKWARD_ENABLE_INTEL (0x2 << 24)
+#define CLK_AVC_ME_SKIP_BLOCK_16x16_DUAL_ENABLE_INTEL (0x3 << 24)
+#define CLK_AVC_ME_SKIP_BLOCK_8x8_FORWARD_ENABLE_INTEL (0x55 << 24)
+#define CLK_AVC_ME_SKIP_BLOCK_8x8_BACKWARD_ENABLE_INTEL (0xAA << 24)
+#define CLK_AVC_ME_SKIP_BLOCK_8x8_DUAL_ENABLE_INTEL (0xFF << 24)
+#define CLK_AVC_ME_SKIP_BLOCK_8x8_0_FORWARD_ENABLE_INTEL (0x1 << 24)
+#define CLK_AVC_ME_SKIP_BLOCK_8x8_0_BACKWARD_ENABLE_INTEL (0x2 << 24)
+#define CLK_AVC_ME_SKIP_BLOCK_8x8_1_FORWARD_ENABLE_INTEL (0x1 << 26)
+#define CLK_AVC_ME_SKIP_BLOCK_8x8_1_BACKWARD_ENABLE_INTEL (0x2 << 26)
+#define CLK_AVC_ME_SKIP_BLOCK_8x8_2_FORWARD_ENABLE_INTEL (0x1 << 28)
+#define CLK_AVC_ME_SKIP_BLOCK_8x8_2_BACKWARD_ENABLE_INTEL (0x2 << 28)
+#define CLK_AVC_ME_SKIP_BLOCK_8x8_3_FORWARD_ENABLE_INTEL (0x1 << 30)
+#define CLK_AVC_ME_SKIP_BLOCK_8x8_3_BACKWARD_ENABLE_INTEL (0x2 << 30)
+
+#define CLK_AVC_ME_BLOCK_BASED_SKIP_4x4_INTEL 0x00
+#define CLK_AVC_ME_BLOCK_BASED_SKIP_8x8_INTEL 0x80
+
+#define CLK_AVC_ME_INTRA_LUMA_PARTITION_MASK_ALL_INTEL 0x0
+#define CLK_AVC_ME_INTRA_LUMA_PARTITION_MASK_16x16_INTEL 0x6
+#define CLK_AVC_ME_INTRA_LUMA_PARTITION_MASK_8x8_INTEL 0x5
+#define CLK_AVC_ME_INTRA_LUMA_PARTITION_MASK_4x4_INTEL 0x3
+
+#define CLK_AVC_ME_INTRA_NEIGHBOR_LEFT_MASK_ENABLE_INTEL 0x60
+#define CLK_AVC_ME_INTRA_NEIGHBOR_UPPER_MASK_ENABLE_INTEL 0x10
+#define CLK_AVC_ME_INTRA_NEIGHBOR_UPPER_RIGHT_MASK_ENABLE_INTEL 0x8
+#define CLK_AVC_ME_INTRA_NEIGHBOR_UPPER_LEFT_MASK_ENABLE_INTEL 0x4
+
+#define CLK_AVC_ME_LUMA_PREDICTOR_MODE_VERTICAL_INTEL 0x0
+#define CLK_AVC_ME_LUMA_PREDICTOR_MODE_HORIZONTAL_INTEL 0x1
+#define CLK_AVC_ME_LUMA_PREDICTOR_MODE_DC_INTEL 0x2
+#define CLK_AVC_ME_LUMA_PREDICTOR_MODE_DIAGONAL_DOWN_LEFT_INTEL 0x3
+#define CLK_AVC_ME_LUMA_PREDICTOR_MODE_DIAGONAL_DOWN_RIGHT_INTEL 0x4
+#define CLK_AVC_ME_LUMA_PREDICTOR_MODE_PLANE_INTEL 0x4
+#define CLK_AVC_ME_LUMA_PREDICTOR_MODE_VERTICAL_RIGHT_INTEL 0x5
+#define CLK_AVC_ME_LUMA_PREDICTOR_MODE_HORIZONTAL_DOWN_INTEL 0x6
+#define CLK_AVC_ME_LUMA_PREDICTOR_MODE_VERTICAL_LEFT_INTEL 0x7
+#define CLK_AVC_ME_LUMA_PREDICTOR_MODE_HORIZONTAL_UP_INTEL 0x8
+#define CLK_AVC_ME_CHROMA_PREDICTOR_MODE_DC_INTEL 0x0
+#define CLK_AVC_ME_CHROMA_PREDICTOR_MODE_HORIZONTAL_INTEL 0x1
+#define CLK_AVC_ME_CHROMA_PREDICTOR_MODE_VERTICAL_INTEL 0x2
+#define CLK_AVC_ME_CHROMA_PREDICTOR_MODE_PLANE_INTEL 0x3
+
+#define CLK_AVC_ME_FRAME_FORWARD_INTEL 0x1
+#define CLK_AVC_ME_FRAME_BACKWARD_INTEL 0x2
+#define CLK_AVC_ME_FRAME_DUAL_INTEL 0x3
+
+#define CLK_AVC_ME_INTERLACED_SCAN_TOP_FIELD_INTEL 0x0
+#define CLK_AVC_ME_INTERLACED_SCAN_BOTTOM_FIELD_INTEL 0x1
+
+#define CLK_AVC_ME_INITIALIZE_INTEL 0x0
+
+#define CLK_AVC_IME_PAYLOAD_INITIALIZE_INTEL 0x0
+#define CLK_AVC_REF_PAYLOAD_INITIALIZE_INTEL 0x0
+#define CLK_AVC_SIC_PAYLOAD_INITIALIZE_INTEL 0x0
+
+#define CLK_AVC_IME_RESULT_INITIALIZE_INTEL 0x0
+#define CLK_AVC_REF_RESULT_INITIALIZE_INTEL 0x0
+#define CLK_AVC_SIC_RESULT_INITIALIZE_INTEL 0x0
+
+#define CLK_AVC_IME_RESULT_SINGLE_REFERENCE_STREAMOUT_INITIALIZE_INTEL 0x0
+#define CLK_AVC_IME_RESULT_SINGLE_REFERENCE_STREAMIN_INITIALIZE_INTEL 0x0
+#define CLK_AVC_IME_RESULT_DUAL_REFERENCE_STREAMOUT_INITIALIZE_INTEL 0x0
+#define CLK_AVC_IME_RESULT_DUAL_REFERENCE_STREAMIN_INITIALIZE_INTEL 0x0
+
+#pragma OPENCL EXTENSION cl_intel_device_side_avc_motion_estimation : end
+#endif // cl_intel_device_side_avc_motion_estimation
+
+#endif //_OPENCL_BASE_H_
//===--- opencl-c.h - OpenCL C language builtin function header -----------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
#ifndef _OPENCL_H_
#define _OPENCL_H_
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#include "opencl-c-base.h"
+
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
#ifndef cl_khr_depth_images
#define cl_khr_depth_images
#endif //cl_khr_depth_images
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
#if __OPENCL_C_VERSION__ < CL_VERSION_2_0
#ifdef cl_khr_3d_image_writes
#endif //cl_khr_3d_image_writes
#endif //__OPENCL_C_VERSION__ < CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_1_2)
+#pragma OPENCL EXTENSION cl_intel_planar_yuv : begin
+#pragma OPENCL EXTENSION cl_intel_planar_yuv : end
+#endif // defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_1_2)
+
#define __ovld __attribute__((overloadable))
#define __conv __attribute__((convergent))
#define __purefn __attribute__((pure))
#define __cnfn __attribute__((const))
-// built-in scalar data types:
-
-/**
- * An unsigned 8-bit integer.
- */
-typedef unsigned char uchar;
-
-/**
- * An unsigned 16-bit integer.
- */
-typedef unsigned short ushort;
-
-/**
- * An unsigned 32-bit integer.
- */
-typedef unsigned int uint;
-
-/**
- * An unsigned 64-bit integer.
- */
-typedef unsigned long ulong;
-
-/**
- * The unsigned integer type of the result of the sizeof operator. This
- * is a 32-bit unsigned integer if CL_DEVICE_ADDRESS_BITS
- * defined in table 4.3 is 32-bits and is a 64-bit unsigned integer if
- * CL_DEVICE_ADDRESS_BITS is 64-bits.
- */
-typedef __SIZE_TYPE__ size_t;
-
-/**
- * A signed integer type that is the result of subtracting two pointers.
- * This is a 32-bit signed integer if CL_DEVICE_ADDRESS_BITS
- * defined in table 4.3 is 32-bits and is a 64-bit signed integer if
- * CL_DEVICE_ADDRESS_BITS is 64-bits.
- */
-typedef __PTRDIFF_TYPE__ ptrdiff_t;
-
-/**
-* A signed integer type with the property that any valid pointer to
-* void can be converted to this type, then converted back to pointer
-* to void, and the result will compare equal to the original pointer.
-*/
-typedef __INTPTR_TYPE__ intptr_t;
-
-/**
-* An unsigned integer type with the property that any valid pointer to
-* void can be converted to this type, then converted back to pointer
-* to void, and the result will compare equal to the original pointer.
-*/
-typedef __UINTPTR_TYPE__ uintptr_t;
-
-// built-in vector data types:
-typedef char char2 __attribute__((ext_vector_type(2)));
-typedef char char3 __attribute__((ext_vector_type(3)));
-typedef char char4 __attribute__((ext_vector_type(4)));
-typedef char char8 __attribute__((ext_vector_type(8)));
-typedef char char16 __attribute__((ext_vector_type(16)));
-typedef uchar uchar2 __attribute__((ext_vector_type(2)));
-typedef uchar uchar3 __attribute__((ext_vector_type(3)));
-typedef uchar uchar4 __attribute__((ext_vector_type(4)));
-typedef uchar uchar8 __attribute__((ext_vector_type(8)));
-typedef uchar uchar16 __attribute__((ext_vector_type(16)));
-typedef short short2 __attribute__((ext_vector_type(2)));
-typedef short short3 __attribute__((ext_vector_type(3)));
-typedef short short4 __attribute__((ext_vector_type(4)));
-typedef short short8 __attribute__((ext_vector_type(8)));
-typedef short short16 __attribute__((ext_vector_type(16)));
-typedef ushort ushort2 __attribute__((ext_vector_type(2)));
-typedef ushort ushort3 __attribute__((ext_vector_type(3)));
-typedef ushort ushort4 __attribute__((ext_vector_type(4)));
-typedef ushort ushort8 __attribute__((ext_vector_type(8)));
-typedef ushort ushort16 __attribute__((ext_vector_type(16)));
-typedef int int2 __attribute__((ext_vector_type(2)));
-typedef int int3 __attribute__((ext_vector_type(3)));
-typedef int int4 __attribute__((ext_vector_type(4)));
-typedef int int8 __attribute__((ext_vector_type(8)));
-typedef int int16 __attribute__((ext_vector_type(16)));
-typedef uint uint2 __attribute__((ext_vector_type(2)));
-typedef uint uint3 __attribute__((ext_vector_type(3)));
-typedef uint uint4 __attribute__((ext_vector_type(4)));
-typedef uint uint8 __attribute__((ext_vector_type(8)));
-typedef uint uint16 __attribute__((ext_vector_type(16)));
-typedef long long2 __attribute__((ext_vector_type(2)));
-typedef long long3 __attribute__((ext_vector_type(3)));
-typedef long long4 __attribute__((ext_vector_type(4)));
-typedef long long8 __attribute__((ext_vector_type(8)));
-typedef long long16 __attribute__((ext_vector_type(16)));
-typedef ulong ulong2 __attribute__((ext_vector_type(2)));
-typedef ulong ulong3 __attribute__((ext_vector_type(3)));
-typedef ulong ulong4 __attribute__((ext_vector_type(4)));
-typedef ulong ulong8 __attribute__((ext_vector_type(8)));
-typedef ulong ulong16 __attribute__((ext_vector_type(16)));
-typedef float float2 __attribute__((ext_vector_type(2)));
-typedef float float3 __attribute__((ext_vector_type(3)));
-typedef float float4 __attribute__((ext_vector_type(4)));
-typedef float float8 __attribute__((ext_vector_type(8)));
-typedef float float16 __attribute__((ext_vector_type(16)));
-#ifdef cl_khr_fp16
-#pragma OPENCL EXTENSION cl_khr_fp16 : enable
-typedef half half2 __attribute__((ext_vector_type(2)));
-typedef half half3 __attribute__((ext_vector_type(3)));
-typedef half half4 __attribute__((ext_vector_type(4)));
-typedef half half8 __attribute__((ext_vector_type(8)));
-typedef half half16 __attribute__((ext_vector_type(16)));
-#endif
-#ifdef cl_khr_fp64
-#if __OPENCL_C_VERSION__ < CL_VERSION_1_2
-#pragma OPENCL EXTENSION cl_khr_fp64 : enable
-#endif
-typedef double double2 __attribute__((ext_vector_type(2)));
-typedef double double3 __attribute__((ext_vector_type(3)));
-typedef double double4 __attribute__((ext_vector_type(4)));
-typedef double double8 __attribute__((ext_vector_type(8)));
-typedef double double16 __attribute__((ext_vector_type(16)));
-#endif
-
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
-#define NULL ((void*)0)
-#endif
-
-/**
- * Value of maximum non-infinite single-precision floating-point
- * number.
- */
-#define MAXFLOAT 0x1.fffffep127f
-
-/**
- * A positive float constant expression. HUGE_VALF evaluates
- * to +infinity. Used as an error value returned by the built-in
- * math functions.
- */
-#define HUGE_VALF (__builtin_huge_valf())
-
-/**
- * A positive double constant expression. HUGE_VAL evaluates
- * to +infinity. Used as an error value returned by the built-in
- * math functions.
- */
-#define HUGE_VAL (__builtin_huge_val())
-
-/**
- * A constant expression of type float representing positive or
- * unsigned infinity.
- */
-#define INFINITY (__builtin_inff())
-
-/**
- * A constant expression of type float representing a quiet NaN.
- */
-#define NAN as_float(INT_MAX)
-
-#define FP_ILOGB0 INT_MIN
-#define FP_ILOGBNAN INT_MAX
-
-#define FLT_DIG 6
-#define FLT_MANT_DIG 24
-#define FLT_MAX_10_EXP +38
-#define FLT_MAX_EXP +128
-#define FLT_MIN_10_EXP -37
-#define FLT_MIN_EXP -125
-#define FLT_RADIX 2
-#define FLT_MAX 0x1.fffffep127f
-#define FLT_MIN 0x1.0p-126f
-#define FLT_EPSILON 0x1.0p-23f
-
-#define M_E_F 2.71828182845904523536028747135266250f
-#define M_LOG2E_F 1.44269504088896340735992468100189214f
-#define M_LOG10E_F 0.434294481903251827651128918916605082f
-#define M_LN2_F 0.693147180559945309417232121458176568f
-#define M_LN10_F 2.30258509299404568401799145468436421f
-#define M_PI_F 3.14159265358979323846264338327950288f
-#define M_PI_2_F 1.57079632679489661923132169163975144f
-#define M_PI_4_F 0.785398163397448309615660845819875721f
-#define M_1_PI_F 0.318309886183790671537767526745028724f
-#define M_2_PI_F 0.636619772367581343075535053490057448f
-#define M_2_SQRTPI_F 1.12837916709551257389615890312154517f
-#define M_SQRT2_F 1.41421356237309504880168872420969808f
-#define M_SQRT1_2_F 0.707106781186547524400844362104849039f
-
-#define DBL_DIG 15
-#define DBL_MANT_DIG 53
-#define DBL_MAX_10_EXP +308
-#define DBL_MAX_EXP +1024
-#define DBL_MIN_10_EXP -307
-#define DBL_MIN_EXP -1021
-#define DBL_RADIX 2
-#define DBL_MAX 0x1.fffffffffffffp1023
-#define DBL_MIN 0x1.0p-1022
-#define DBL_EPSILON 0x1.0p-52
-
-#define M_E 0x1.5bf0a8b145769p+1
-#define M_LOG2E 0x1.71547652b82fep+0
-#define M_LOG10E 0x1.bcb7b1526e50ep-2
-#define M_LN2 0x1.62e42fefa39efp-1
-#define M_LN10 0x1.26bb1bbb55516p+1
-#define M_PI 0x1.921fb54442d18p+1
-#define M_PI_2 0x1.921fb54442d18p+0
-#define M_PI_4 0x1.921fb54442d18p-1
-#define M_1_PI 0x1.45f306dc9c883p-2
-#define M_2_PI 0x1.45f306dc9c883p-1
-#define M_2_SQRTPI 0x1.20dd750429b6dp+0
-#define M_SQRT2 0x1.6a09e667f3bcdp+0
-#define M_SQRT1_2 0x1.6a09e667f3bcdp-1
-
-#ifdef cl_khr_fp16
-
-#define HALF_DIG 3
-#define HALF_MANT_DIG 11
-#define HALF_MAX_10_EXP +4
-#define HALF_MAX_EXP +16
-#define HALF_MIN_10_EXP -4
-#define HALF_MIN_EXP -13
-#define HALF_RADIX 2
-#define HALF_MAX ((0x1.ffcp15h))
-#define HALF_MIN ((0x1.0p-14h))
-#define HALF_EPSILON ((0x1.0p-10h))
-
-#define M_E_H 2.71828182845904523536028747135266250h
-#define M_LOG2E_H 1.44269504088896340735992468100189214h
-#define M_LOG10E_H 0.434294481903251827651128918916605082h
-#define M_LN2_H 0.693147180559945309417232121458176568h
-#define M_LN10_H 2.30258509299404568401799145468436421h
-#define M_PI_H 3.14159265358979323846264338327950288h
-#define M_PI_2_H 1.57079632679489661923132169163975144h
-#define M_PI_4_H 0.785398163397448309615660845819875721h
-#define M_1_PI_H 0.318309886183790671537767526745028724h
-#define M_2_PI_H 0.636619772367581343075535053490057448h
-#define M_2_SQRTPI_H 1.12837916709551257389615890312154517h
-#define M_SQRT2_H 1.41421356237309504880168872420969808h
-#define M_SQRT1_2_H 0.707106781186547524400844362104849039h
-
-#endif //cl_khr_fp16
-
-#define CHAR_BIT 8
-#define SCHAR_MAX 127
-#define SCHAR_MIN (-128)
-#define UCHAR_MAX 255
-#define CHAR_MAX SCHAR_MAX
-#define CHAR_MIN SCHAR_MIN
-#define USHRT_MAX 65535
-#define SHRT_MAX 32767
-#define SHRT_MIN (-32768)
-#define UINT_MAX 0xffffffff
-#define INT_MAX 2147483647
-#define INT_MIN (-2147483647-1)
-#define ULONG_MAX 0xffffffffffffffffUL
-#define LONG_MAX 0x7fffffffffffffffL
-#define LONG_MIN (-0x7fffffffffffffffL-1)
// OpenCL v1.1/1.2/2.0 s6.2.3 - Explicit conversions
*/
size_t __ovld __cnfn get_global_offset(uint dimindx);
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
size_t __ovld get_enqueued_local_size(uint dimindx);
size_t __ovld get_global_linear_id(void);
size_t __ovld get_local_linear_id(void);
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
// OpenCL v1.1 s6.11.2, v1.2 s6.12.2, v2.0 s6.13.2 - Math functions
* Returns fmin(x - floor (x), 0x1.fffffep-1f ).
* floor(x) is returned in iptr.
*/
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
float __ovld fract(float x, float *iptr);
float2 __ovld fract(float2 x, float2 *iptr);
float3 __ovld fract(float3 x, float3 *iptr);
half8 __ovld fract(half8 x, __private half8 *iptr);
half16 __ovld fract(half16 x, __private half16 *iptr);
#endif //cl_khr_fp16
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Extract mantissa and exponent from x. For each
* magnitude in the interval [1/2, 1) or 0. Each
* component of x equals mantissa returned * 2^exp.
*/
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
float __ovld frexp(float x, int *exp);
float2 __ovld frexp(float2 x, int2 *exp);
float3 __ovld frexp(float3 x, int3 *exp);
half8 __ovld frexp(half8 x, __private int8 *exp);
half16 __ovld frexp(half16 x, __private int16 *exp);
#endif //cl_khr_fp16
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Compute the value of the square root of x^2 + y^2
half16 __ovld __cnfn lgamma(half16 x);
#endif //cl_khr_fp16
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
float __ovld lgamma_r(float x, int *signp);
float2 __ovld lgamma_r(float2 x, int2 *signp);
float3 __ovld lgamma_r(float3 x, int3 *signp);
half8 __ovld lgamma_r(half8 x, __private int8 *signp);
half16 __ovld lgamma_r(half16 x, __private int16 *signp);
#endif //cl_khr_fp16
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Compute natural logarithm.
* the argument. It stores the integral part in the object
* pointed to by iptr.
*/
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
float __ovld modf(float x, float *iptr);
float2 __ovld modf(float2 x, float2 *iptr);
float3 __ovld modf(float3 x, float3 *iptr);
half8 __ovld modf(half8 x, __private half8 *iptr);
half16 __ovld modf(half16 x, __private half16 *iptr);
#endif //cl_khr_fp16
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Returns a quiet NaN. The nancode may be placed
* sign as x/y. It stores this signed value in the object
* pointed to by quo.
*/
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
float __ovld remquo(float x, float y, int *quo);
float2 __ovld remquo(float2 x, float2 y, int2 *quo);
float3 __ovld remquo(float3 x, float3 y, int3 *quo);
half8 __ovld remquo(half8 x, half8 y, __private int8 *quo);
half16 __ovld remquo(half16 x, half16 y, __private int16 *quo);
#endif //cl_khr_fp16
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Round to integral value (using round to nearest
* even rounding mode) in floating-point format.
* is the return value and computed cosine is returned
* in cosval.
*/
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
float __ovld sincos(float x, float *cosval);
float2 __ovld sincos(float2 x, float2 *cosval);
float3 __ovld sincos(float3 x, float3 *cosval);
half8 __ovld sincos(half8 x, __private half8 *cosval);
half16 __ovld sincos(half16 x, __private half16 *cosval);
#endif //cl_khr_fp16
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Compute hyperbolic sine.
ulong8 __ovld __cnfn clamp(ulong8 x, ulong8 minval, ulong8 maxval);
long16 __ovld __cnfn clamp(long16 x, long16 minval, long16 maxval);
ulong16 __ovld __cnfn clamp(ulong16 x, ulong16 minval, ulong16 maxval);
-char __ovld __cnfn clamp(char x, char minval, char maxval);
-uchar __ovld __cnfn clamp(uchar x, uchar minval, uchar maxval);
char2 __ovld __cnfn clamp(char2 x, char minval, char maxval);
uchar2 __ovld __cnfn clamp(uchar2 x, uchar minval, uchar maxval);
char3 __ovld __cnfn clamp(char3 x, char minval, char maxval);
uchar8 __ovld __cnfn clamp(uchar8 x, uchar minval, uchar maxval);
char16 __ovld __cnfn clamp(char16 x, char minval, char maxval);
uchar16 __ovld __cnfn clamp(uchar16 x, uchar minval, uchar maxval);
-short __ovld __cnfn clamp(short x, short minval, short maxval);
-ushort __ovld __cnfn clamp(ushort x, ushort minval, ushort maxval);
short2 __ovld __cnfn clamp(short2 x, short minval, short maxval);
ushort2 __ovld __cnfn clamp(ushort2 x, ushort minval, ushort maxval);
short3 __ovld __cnfn clamp(short3 x, short minval, short maxval);
ushort8 __ovld __cnfn clamp(ushort8 x, ushort minval, ushort maxval);
short16 __ovld __cnfn clamp(short16 x, short minval, short maxval);
ushort16 __ovld __cnfn clamp(ushort16 x, ushort minval, ushort maxval);
-int __ovld __cnfn clamp(int x, int minval, int maxval);
-uint __ovld __cnfn clamp(uint x, uint minval, uint maxval);
int2 __ovld __cnfn clamp(int2 x, int minval, int maxval);
uint2 __ovld __cnfn clamp(uint2 x, uint minval, uint maxval);
int3 __ovld __cnfn clamp(int3 x, int minval, int maxval);
uint8 __ovld __cnfn clamp(uint8 x, uint minval, uint maxval);
int16 __ovld __cnfn clamp(int16 x, int minval, int maxval);
uint16 __ovld __cnfn clamp(uint16 x, uint minval, uint maxval);
-long __ovld __cnfn clamp(long x, long minval, long maxval);
-ulong __ovld __cnfn clamp(ulong x, ulong minval, ulong maxval);
long2 __ovld __cnfn clamp(long2 x, long minval, long maxval);
ulong2 __ovld __cnfn clamp(ulong2 x, ulong minval, ulong maxval);
long3 __ovld __cnfn clamp(long3 x, long minval, long maxval);
* returns the size in bits of the type of x or
* component type of x, if x is a vector.
*/
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
char __ovld ctz(char x);
uchar __ovld ctz(uchar x);
char2 __ovld ctz(char2 x);
ulong8 __ovld ctz(ulong8 x);
long16 __ovld ctz(long16 x);
ulong16 __ovld ctz(ulong16 x);
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Returns mul_hi(a, b) + c.
ulong8 __ovld __cnfn max(ulong8 x, ulong8 y);
long16 __ovld __cnfn max(long16 x, long16 y);
ulong16 __ovld __cnfn max(ulong16 x, ulong16 y);
-char __ovld __cnfn max(char x, char y);
-uchar __ovld __cnfn max(uchar x, uchar y);
char2 __ovld __cnfn max(char2 x, char y);
uchar2 __ovld __cnfn max(uchar2 x, uchar y);
char3 __ovld __cnfn max(char3 x, char y);
uchar8 __ovld __cnfn max(uchar8 x, uchar y);
char16 __ovld __cnfn max(char16 x, char y);
uchar16 __ovld __cnfn max(uchar16 x, uchar y);
-short __ovld __cnfn max(short x, short y);
-ushort __ovld __cnfn max(ushort x, ushort y);
short2 __ovld __cnfn max(short2 x, short y);
ushort2 __ovld __cnfn max(ushort2 x, ushort y);
short3 __ovld __cnfn max(short3 x, short y);
ushort8 __ovld __cnfn max(ushort8 x, ushort y);
short16 __ovld __cnfn max(short16 x, short y);
ushort16 __ovld __cnfn max(ushort16 x, ushort y);
-int __ovld __cnfn max(int x, int y);
-uint __ovld __cnfn max(uint x, uint y);
int2 __ovld __cnfn max(int2 x, int y);
uint2 __ovld __cnfn max(uint2 x, uint y);
int3 __ovld __cnfn max(int3 x, int y);
uint8 __ovld __cnfn max(uint8 x, uint y);
int16 __ovld __cnfn max(int16 x, int y);
uint16 __ovld __cnfn max(uint16 x, uint y);
-long __ovld __cnfn max(long x, long y);
-ulong __ovld __cnfn max(ulong x, ulong y);
long2 __ovld __cnfn max(long2 x, long y);
ulong2 __ovld __cnfn max(ulong2 x, ulong y);
long3 __ovld __cnfn max(long3 x, long y);
ulong8 __ovld __cnfn min(ulong8 x, ulong8 y);
long16 __ovld __cnfn min(long16 x, long16 y);
ulong16 __ovld __cnfn min(ulong16 x, ulong16 y);
-char __ovld __cnfn min(char x, char y);
-uchar __ovld __cnfn min(uchar x, uchar y);
char2 __ovld __cnfn min(char2 x, char y);
uchar2 __ovld __cnfn min(uchar2 x, uchar y);
char3 __ovld __cnfn min(char3 x, char y);
uchar8 __ovld __cnfn min(uchar8 x, uchar y);
char16 __ovld __cnfn min(char16 x, char y);
uchar16 __ovld __cnfn min(uchar16 x, uchar y);
-short __ovld __cnfn min(short x, short y);
-ushort __ovld __cnfn min(ushort x, ushort y);
short2 __ovld __cnfn min(short2 x, short y);
ushort2 __ovld __cnfn min(ushort2 x, ushort y);
short3 __ovld __cnfn min(short3 x, short y);
ushort8 __ovld __cnfn min(ushort8 x, ushort y);
short16 __ovld __cnfn min(short16 x, short y);
ushort16 __ovld __cnfn min(ushort16 x, ushort y);
-int __ovld __cnfn min(int x, int y);
-uint __ovld __cnfn min(uint x, uint y);
int2 __ovld __cnfn min(int2 x, int y);
uint2 __ovld __cnfn min(uint2 x, uint y);
int3 __ovld __cnfn min(int3 x, int y);
uint8 __ovld __cnfn min(uint8 x, uint y);
int16 __ovld __cnfn min(int16 x, int y);
uint16 __ovld __cnfn min(uint16 x, uint y);
-long __ovld __cnfn min(long x, long y);
-ulong __ovld __cnfn min(ulong x, ulong y);
long2 __ovld __cnfn min(long2 x, long y);
ulong2 __ovld __cnfn min(ulong2 x, ulong y);
long3 __ovld __cnfn min(long3 x, long y);
half4 __ovld __cnfn step(half4 edge, half4 x);
half8 __ovld __cnfn step(half8 edge, half8 x);
half16 __ovld __cnfn step(half16 edge, half16 x);
-half __ovld __cnfn step(half edge, half x);
half2 __ovld __cnfn step(half edge, half2 x);
half3 __ovld __cnfn step(half edge, half3 x);
half4 __ovld __cnfn step(half edge, half4 x);
half4 __ovld __cnfn smoothstep(half4 edge0, half4 edge1, half4 x);
half8 __ovld __cnfn smoothstep(half8 edge0, half8 edge1, half8 x);
half16 __ovld __cnfn smoothstep(half16 edge0, half16 edge1, half16 x);
-half __ovld __cnfn smoothstep(half edge0, half edge1, half x);
half2 __ovld __cnfn smoothstep(half edge0, half edge1, half2 x);
half3 __ovld __cnfn smoothstep(half edge0, half edge1, half3 x);
half4 __ovld __cnfn smoothstep(half edge0, half edge1, half4 x);
* For each component of a vector type,
* result[i] = if MSB of c[i] is set ? b[i] : a[i].
* For a scalar type, result = c ? b : a.
+ * b and a must have the same type.
+ * c must have the same number of elements and bits as a.
*/
char __ovld __cnfn select(char a, char b, char c);
uchar __ovld __cnfn select(uchar a, uchar b, char c);
uchar8 __ovld __cnfn select(uchar8 a, uchar8 b, char8 c);
char16 __ovld __cnfn select(char16 a, char16 b, char16 c);
uchar16 __ovld __cnfn select(uchar16 a, uchar16 b, char16 c);
-short __ovld __cnfn select(short a, short b, char c);
-ushort __ovld __cnfn select(ushort a, ushort b, char c);
-short2 __ovld __cnfn select(short2 a, short2 b, char2 c);
-ushort2 __ovld __cnfn select(ushort2 a, ushort2 b, char2 c);
-short3 __ovld __cnfn select(short3 a, short3 b, char3 c);
-ushort3 __ovld __cnfn select(ushort3 a, ushort3 b, char3 c);
-short4 __ovld __cnfn select(short4 a, short4 b, char4 c);
-ushort4 __ovld __cnfn select(ushort4 a, ushort4 b, char4 c);
-short8 __ovld __cnfn select(short8 a, short8 b, char8 c);
-ushort8 __ovld __cnfn select(ushort8 a, ushort8 b, char8 c);
-short16 __ovld __cnfn select(short16 a, short16 b, char16 c);
-ushort16 __ovld __cnfn select(ushort16 a, ushort16 b, char16 c);
-int __ovld __cnfn select(int a, int b, char c);
-uint __ovld __cnfn select(uint a, uint b, char c);
-int2 __ovld __cnfn select(int2 a, int2 b, char2 c);
-uint2 __ovld __cnfn select(uint2 a, uint2 b, char2 c);
-int3 __ovld __cnfn select(int3 a, int3 b, char3 c);
-uint3 __ovld __cnfn select(uint3 a, uint3 b, char3 c);
-int4 __ovld __cnfn select(int4 a, int4 b, char4 c);
-uint4 __ovld __cnfn select(uint4 a, uint4 b, char4 c);
-int8 __ovld __cnfn select(int8 a, int8 b, char8 c);
-uint8 __ovld __cnfn select(uint8 a, uint8 b, char8 c);
-int16 __ovld __cnfn select(int16 a, int16 b, char16 c);
-uint16 __ovld __cnfn select(uint16 a, uint16 b, char16 c);
-long __ovld __cnfn select(long a, long b, char c);
-ulong __ovld __cnfn select(ulong a, ulong b, char c);
-long2 __ovld __cnfn select(long2 a, long2 b, char2 c);
-ulong2 __ovld __cnfn select(ulong2 a, ulong2 b, char2 c);
-long3 __ovld __cnfn select(long3 a, long3 b, char3 c);
-ulong3 __ovld __cnfn select(ulong3 a, ulong3 b, char3 c);
-long4 __ovld __cnfn select(long4 a, long4 b, char4 c);
-ulong4 __ovld __cnfn select(ulong4 a, ulong4 b, char4 c);
-long8 __ovld __cnfn select(long8 a, long8 b, char8 c);
-ulong8 __ovld __cnfn select(ulong8 a, ulong8 b, char8 c);
-long16 __ovld __cnfn select(long16 a, long16 b, char16 c);
-ulong16 __ovld __cnfn select(ulong16 a, ulong16 b, char16 c);
-float __ovld __cnfn select(float a, float b, char c);
-float2 __ovld __cnfn select(float2 a, float2 b, char2 c);
-float3 __ovld __cnfn select(float3 a, float3 b, char3 c);
-float4 __ovld __cnfn select(float4 a, float4 b, char4 c);
-float8 __ovld __cnfn select(float8 a, float8 b, char8 c);
-float16 __ovld __cnfn select(float16 a, float16 b, char16 c);
-char __ovld __cnfn select(char a, char b, short c);
-uchar __ovld __cnfn select(uchar a, uchar b, short c);
-char2 __ovld __cnfn select(char2 a, char2 b, short2 c);
-uchar2 __ovld __cnfn select(uchar2 a, uchar2 b, short2 c);
-char3 __ovld __cnfn select(char3 a, char3 b, short3 c);
-uchar3 __ovld __cnfn select(uchar3 a, uchar3 b, short3 c);
-char4 __ovld __cnfn select(char4 a, char4 b, short4 c);
-uchar4 __ovld __cnfn select(uchar4 a, uchar4 b, short4 c);
-char8 __ovld __cnfn select(char8 a, char8 b, short8 c);
-uchar8 __ovld __cnfn select(uchar8 a, uchar8 b, short8 c);
-char16 __ovld __cnfn select(char16 a, char16 b, short16 c);
-uchar16 __ovld __cnfn select(uchar16 a, uchar16 b, short16 c);
+
short __ovld __cnfn select(short a, short b, short c);
ushort __ovld __cnfn select(ushort a, ushort b, short c);
short2 __ovld __cnfn select(short2 a, short2 b, short2 c);
ushort8 __ovld __cnfn select(ushort8 a, ushort8 b, short8 c);
short16 __ovld __cnfn select(short16 a, short16 b, short16 c);
ushort16 __ovld __cnfn select(ushort16 a, ushort16 b, short16 c);
-int __ovld __cnfn select(int a, int b, short c);
-uint __ovld __cnfn select(uint a, uint b, short c);
-int2 __ovld __cnfn select(int2 a, int2 b, short2 c);
-uint2 __ovld __cnfn select(uint2 a, uint2 b, short2 c);
-int3 __ovld __cnfn select(int3 a, int3 b, short3 c);
-uint3 __ovld __cnfn select(uint3 a, uint3 b, short3 c);
-int4 __ovld __cnfn select(int4 a, int4 b, short4 c);
-uint4 __ovld __cnfn select(uint4 a, uint4 b, short4 c);
-int8 __ovld __cnfn select(int8 a, int8 b, short8 c);
-uint8 __ovld __cnfn select(uint8 a, uint8 b, short8 c);
-int16 __ovld __cnfn select(int16 a, int16 b, short16 c);
-uint16 __ovld __cnfn select(uint16 a, uint16 b, short16 c);
-long __ovld __cnfn select(long a, long b, short c);
-ulong __ovld __cnfn select(ulong a, ulong b, short c);
-long2 __ovld __cnfn select(long2 a, long2 b, short2 c);
-ulong2 __ovld __cnfn select(ulong2 a, ulong2 b, short2 c);
-long3 __ovld __cnfn select(long3 a, long3 b, short3 c);
-ulong3 __ovld __cnfn select(ulong3 a, ulong3 b, short3 c);
-long4 __ovld __cnfn select(long4 a, long4 b, short4 c);
-ulong4 __ovld __cnfn select(ulong4 a, ulong4 b, short4 c);
-long8 __ovld __cnfn select(long8 a, long8 b, short8 c);
-ulong8 __ovld __cnfn select(ulong8 a, ulong8 b, short8 c);
-long16 __ovld __cnfn select(long16 a, long16 b, short16 c);
-ulong16 __ovld __cnfn select(ulong16 a, ulong16 b, short16 c);
-float __ovld __cnfn select(float a, float b, short c);
-float2 __ovld __cnfn select(float2 a, float2 b, short2 c);
-float3 __ovld __cnfn select(float3 a, float3 b, short3 c);
-float4 __ovld __cnfn select(float4 a, float4 b, short4 c);
-float8 __ovld __cnfn select(float8 a, float8 b, short8 c);
-float16 __ovld __cnfn select(float16 a, float16 b, short16 c);
-char __ovld __cnfn select(char a, char b, int c);
-uchar __ovld __cnfn select(uchar a, uchar b, int c);
-char2 __ovld __cnfn select(char2 a, char2 b, int2 c);
-uchar2 __ovld __cnfn select(uchar2 a, uchar2 b, int2 c);
-char3 __ovld __cnfn select(char3 a, char3 b, int3 c);
-uchar3 __ovld __cnfn select(uchar3 a, uchar3 b, int3 c);
-char4 __ovld __cnfn select(char4 a, char4 b, int4 c);
-uchar4 __ovld __cnfn select(uchar4 a, uchar4 b, int4 c);
-char8 __ovld __cnfn select(char8 a, char8 b, int8 c);
-uchar8 __ovld __cnfn select(uchar8 a, uchar8 b, int8 c);
-char16 __ovld __cnfn select(char16 a, char16 b, int16 c);
-uchar16 __ovld __cnfn select(uchar16 a, uchar16 b, int16 c);
-short __ovld __cnfn select(short a, short b, int c);
-ushort __ovld __cnfn select(ushort a, ushort b, int c);
-short2 __ovld __cnfn select(short2 a, short2 b, int2 c);
-ushort2 __ovld __cnfn select(ushort2 a, ushort2 b, int2 c);
-short3 __ovld __cnfn select(short3 a, short3 b, int3 c);
-ushort3 __ovld __cnfn select(ushort3 a, ushort3 b, int3 c);
-short4 __ovld __cnfn select(short4 a, short4 b, int4 c);
-ushort4 __ovld __cnfn select(ushort4 a, ushort4 b, int4 c);
-short8 __ovld __cnfn select(short8 a, short8 b, int8 c);
-ushort8 __ovld __cnfn select(ushort8 a, ushort8 b, int8 c);
-short16 __ovld __cnfn select(short16 a, short16 b, int16 c);
-ushort16 __ovld __cnfn select(ushort16 a, ushort16 b, int16 c);
+
int __ovld __cnfn select(int a, int b, int c);
uint __ovld __cnfn select(uint a, uint b, int c);
int2 __ovld __cnfn select(int2 a, int2 b, int2 c);
uint8 __ovld __cnfn select(uint8 a, uint8 b, int8 c);
int16 __ovld __cnfn select(int16 a, int16 b, int16 c);
uint16 __ovld __cnfn select(uint16 a, uint16 b, int16 c);
-long __ovld __cnfn select(long a, long b, int c);
-ulong __ovld __cnfn select(ulong a, ulong b, int c);
-long2 __ovld __cnfn select(long2 a, long2 b, int2 c);
-ulong2 __ovld __cnfn select(ulong2 a, ulong2 b, int2 c);
-long3 __ovld __cnfn select(long3 a, long3 b, int3 c);
-ulong3 __ovld __cnfn select(ulong3 a, ulong3 b, int3 c);
-long4 __ovld __cnfn select(long4 a, long4 b, int4 c);
-ulong4 __ovld __cnfn select(ulong4 a, ulong4 b, int4 c);
-long8 __ovld __cnfn select(long8 a, long8 b, int8 c);
-ulong8 __ovld __cnfn select(ulong8 a, ulong8 b, int8 c);
-long16 __ovld __cnfn select(long16 a, long16 b, int16 c);
-ulong16 __ovld __cnfn select(ulong16 a, ulong16 b, int16 c);
float __ovld __cnfn select(float a, float b, int c);
float2 __ovld __cnfn select(float2 a, float2 b, int2 c);
float3 __ovld __cnfn select(float3 a, float3 b, int3 c);
float4 __ovld __cnfn select(float4 a, float4 b, int4 c);
float8 __ovld __cnfn select(float8 a, float8 b, int8 c);
float16 __ovld __cnfn select(float16 a, float16 b, int16 c);
-char __ovld __cnfn select(char a, char b, long c);
-uchar __ovld __cnfn select(uchar a, uchar b, long c);
-char2 __ovld __cnfn select(char2 a, char2 b, long2 c);
-uchar2 __ovld __cnfn select(uchar2 a, uchar2 b, long2 c);
-char3 __ovld __cnfn select(char3 a, char3 b, long3 c);
-uchar3 __ovld __cnfn select(uchar3 a, uchar3 b, long3 c);
-char4 __ovld __cnfn select(char4 a, char4 b, long4 c);
-uchar4 __ovld __cnfn select(uchar4 a, uchar4 b, long4 c);
-char8 __ovld __cnfn select(char8 a, char8 b, long8 c);
-uchar8 __ovld __cnfn select(uchar8 a, uchar8 b, long8 c);
-char16 __ovld __cnfn select(char16 a, char16 b, long16 c);
-uchar16 __ovld __cnfn select(uchar16 a, uchar16 b, long16 c);
-short __ovld __cnfn select(short a, short b, long c);
-ushort __ovld __cnfn select(ushort a, ushort b, long c);
-short2 __ovld __cnfn select(short2 a, short2 b, long2 c);
-ushort2 __ovld __cnfn select(ushort2 a, ushort2 b, long2 c);
-short3 __ovld __cnfn select(short3 a, short3 b, long3 c);
-ushort3 __ovld __cnfn select(ushort3 a, ushort3 b, long3 c);
-short4 __ovld __cnfn select(short4 a, short4 b, long4 c);
-ushort4 __ovld __cnfn select(ushort4 a, ushort4 b, long4 c);
-short8 __ovld __cnfn select(short8 a, short8 b, long8 c);
-ushort8 __ovld __cnfn select(ushort8 a, ushort8 b, long8 c);
-short16 __ovld __cnfn select(short16 a, short16 b, long16 c);
-ushort16 __ovld __cnfn select(ushort16 a, ushort16 b, long16 c);
-int __ovld __cnfn select(int a, int b, long c);
-uint __ovld __cnfn select(uint a, uint b, long c);
-int2 __ovld __cnfn select(int2 a, int2 b, long2 c);
-uint2 __ovld __cnfn select(uint2 a, uint2 b, long2 c);
-int3 __ovld __cnfn select(int3 a, int3 b, long3 c);
-uint3 __ovld __cnfn select(uint3 a, uint3 b, long3 c);
-int4 __ovld __cnfn select(int4 a, int4 b, long4 c);
-uint4 __ovld __cnfn select(uint4 a, uint4 b, long4 c);
-int8 __ovld __cnfn select(int8 a, int8 b, long8 c);
-uint8 __ovld __cnfn select(uint8 a, uint8 b, long8 c);
-int16 __ovld __cnfn select(int16 a, int16 b, long16 c);
-uint16 __ovld __cnfn select(uint16 a, uint16 b, long16 c);
+
long __ovld __cnfn select(long a, long b, long c);
ulong __ovld __cnfn select(ulong a, ulong b, long c);
long2 __ovld __cnfn select(long2 a, long2 b, long2 c);
ulong8 __ovld __cnfn select(ulong8 a, ulong8 b, long8 c);
long16 __ovld __cnfn select(long16 a, long16 b, long16 c);
ulong16 __ovld __cnfn select(ulong16 a, ulong16 b, long16 c);
-float __ovld __cnfn select(float a, float b, long c);
-float2 __ovld __cnfn select(float2 a, float2 b, long2 c);
-float3 __ovld __cnfn select(float3 a, float3 b, long3 c);
-float4 __ovld __cnfn select(float4 a, float4 b, long4 c);
-float8 __ovld __cnfn select(float8 a, float8 b, long8 c);
-float16 __ovld __cnfn select(float16 a, float16 b, long16 c);
+
char __ovld __cnfn select(char a, char b, uchar c);
uchar __ovld __cnfn select(uchar a, uchar b, uchar c);
char2 __ovld __cnfn select(char2 a, char2 b, uchar2 c);
uchar8 __ovld __cnfn select(uchar8 a, uchar8 b, uchar8 c);
char16 __ovld __cnfn select(char16 a, char16 b, uchar16 c);
uchar16 __ovld __cnfn select(uchar16 a, uchar16 b, uchar16 c);
-short __ovld __cnfn select(short a, short b, uchar c);
-ushort __ovld __cnfn select(ushort a, ushort b, uchar c);
-short2 __ovld __cnfn select(short2 a, short2 b, uchar2 c);
-ushort2 __ovld __cnfn select(ushort2 a, ushort2 b, uchar2 c);
-short3 __ovld __cnfn select(short3 a, short3 b, uchar3 c);
-ushort3 __ovld __cnfn select(ushort3 a, ushort3 b, uchar3 c);
-short4 __ovld __cnfn select(short4 a, short4 b, uchar4 c);
-ushort4 __ovld __cnfn select(ushort4 a, ushort4 b, uchar4 c);
-short8 __ovld __cnfn select(short8 a, short8 b, uchar8 c);
-ushort8 __ovld __cnfn select(ushort8 a, ushort8 b, uchar8 c);
-short16 __ovld __cnfn select(short16 a, short16 b, uchar16 c);
-ushort16 __ovld __cnfn select(ushort16 a, ushort16 b, uchar16 c);
-int __ovld __cnfn select(int a, int b, uchar c);
-uint __ovld __cnfn select(uint a, uint b, uchar c);
-int2 __ovld __cnfn select(int2 a, int2 b, uchar2 c);
-uint2 __ovld __cnfn select(uint2 a, uint2 b, uchar2 c);
-int3 __ovld __cnfn select(int3 a, int3 b, uchar3 c);
-uint3 __ovld __cnfn select(uint3 a, uint3 b, uchar3 c);
-int4 __ovld __cnfn select(int4 a, int4 b, uchar4 c);
-uint4 __ovld __cnfn select(uint4 a, uint4 b, uchar4 c);
-int8 __ovld __cnfn select(int8 a, int8 b, uchar8 c);
-uint8 __ovld __cnfn select(uint8 a, uint8 b, uchar8 c);
-int16 __ovld __cnfn select(int16 a, int16 b, uchar16 c);
-uint16 __ovld __cnfn select(uint16 a, uint16 b, uchar16 c);
-long __ovld __cnfn select(long a, long b, uchar c);
-ulong __ovld __cnfn select(ulong a, ulong b, uchar c);
-long2 __ovld __cnfn select(long2 a, long2 b, uchar2 c);
-ulong2 __ovld __cnfn select(ulong2 a, ulong2 b, uchar2 c);
-long3 __ovld __cnfn select(long3 a, long3 b, uchar3 c);
-ulong3 __ovld __cnfn select(ulong3 a, ulong3 b, uchar3 c);
-long4 __ovld __cnfn select(long4 a, long4 b, uchar4 c);
-ulong4 __ovld __cnfn select(ulong4 a, ulong4 b, uchar4 c);
-long8 __ovld __cnfn select(long8 a, long8 b, uchar8 c);
-ulong8 __ovld __cnfn select(ulong8 a, ulong8 b, uchar8 c);
-long16 __ovld __cnfn select(long16 a, long16 b, uchar16 c);
-ulong16 __ovld __cnfn select(ulong16 a, ulong16 b, uchar16 c);
-float __ovld __cnfn select(float a, float b, uchar c);
-float2 __ovld __cnfn select(float2 a, float2 b, uchar2 c);
-float3 __ovld __cnfn select(float3 a, float3 b, uchar3 c);
-float4 __ovld __cnfn select(float4 a, float4 b, uchar4 c);
-float8 __ovld __cnfn select(float8 a, float8 b, uchar8 c);
-float16 __ovld __cnfn select(float16 a, float16 b, uchar16 c);
-char __ovld __cnfn select(char a, char b, ushort c);
-uchar __ovld __cnfn select(uchar a, uchar b, ushort c);
-char2 __ovld __cnfn select(char2 a, char2 b, ushort2 c);
-uchar2 __ovld __cnfn select(uchar2 a, uchar2 b, ushort2 c);
-char3 __ovld __cnfn select(char3 a, char3 b, ushort3 c);
-uchar3 __ovld __cnfn select(uchar3 a, uchar3 b, ushort3 c);
-char4 __ovld __cnfn select(char4 a, char4 b, ushort4 c);
-uchar4 __ovld __cnfn select(uchar4 a, uchar4 b, ushort4 c);
-char8 __ovld __cnfn select(char8 a, char8 b, ushort8 c);
-uchar8 __ovld __cnfn select(uchar8 a, uchar8 b, ushort8 c);
-char16 __ovld __cnfn select(char16 a, char16 b, ushort16 c);
-uchar16 __ovld __cnfn select(uchar16 a, uchar16 b, ushort16 c);
+
short __ovld __cnfn select(short a, short b, ushort c);
ushort __ovld __cnfn select(ushort a, ushort b, ushort c);
short2 __ovld __cnfn select(short2 a, short2 b, ushort2 c);
ushort8 __ovld __cnfn select(ushort8 a, ushort8 b, ushort8 c);
short16 __ovld __cnfn select(short16 a, short16 b, ushort16 c);
ushort16 __ovld __cnfn select(ushort16 a, ushort16 b, ushort16 c);
-int __ovld __cnfn select(int a, int b, ushort c);
-uint __ovld __cnfn select(uint a, uint b, ushort c);
-int2 __ovld __cnfn select(int2 a, int2 b, ushort2 c);
-uint2 __ovld __cnfn select(uint2 a, uint2 b, ushort2 c);
-int3 __ovld __cnfn select(int3 a, int3 b, ushort3 c);
-uint3 __ovld __cnfn select(uint3 a, uint3 b, ushort3 c);
-int4 __ovld __cnfn select(int4 a, int4 b, ushort4 c);
-uint4 __ovld __cnfn select(uint4 a, uint4 b, ushort4 c);
-int8 __ovld __cnfn select(int8 a, int8 b, ushort8 c);
-uint8 __ovld __cnfn select(uint8 a, uint8 b, ushort8 c);
-int16 __ovld __cnfn select(int16 a, int16 b, ushort16 c);
-uint16 __ovld __cnfn select(uint16 a, uint16 b, ushort16 c);
-long __ovld __cnfn select(long a, long b, ushort c);
-ulong __ovld __cnfn select(ulong a, ulong b, ushort c);
-long2 __ovld __cnfn select(long2 a, long2 b, ushort2 c);
-ulong2 __ovld __cnfn select(ulong2 a, ulong2 b, ushort2 c);
-long3 __ovld __cnfn select(long3 a, long3 b, ushort3 c);
-ulong3 __ovld __cnfn select(ulong3 a, ulong3 b, ushort3 c);
-long4 __ovld __cnfn select(long4 a, long4 b, ushort4 c);
-ulong4 __ovld __cnfn select(ulong4 a, ulong4 b, ushort4 c);
-long8 __ovld __cnfn select(long8 a, long8 b, ushort8 c);
-ulong8 __ovld __cnfn select(ulong8 a, ulong8 b, ushort8 c);
-long16 __ovld __cnfn select(long16 a, long16 b, ushort16 c);
-ulong16 __ovld __cnfn select(ulong16 a, ulong16 b, ushort16 c);
-float __ovld __cnfn select(float a, float b, ushort c);
-float2 __ovld __cnfn select(float2 a, float2 b, ushort2 c);
-float3 __ovld __cnfn select(float3 a, float3 b, ushort3 c);
-float4 __ovld __cnfn select(float4 a, float4 b, ushort4 c);
-float8 __ovld __cnfn select(float8 a, float8 b, ushort8 c);
-float16 __ovld __cnfn select(float16 a, float16 b, ushort16 c);
-char __ovld __cnfn select(char a, char b, uint c);
-uchar __ovld __cnfn select(uchar a, uchar b, uint c);
-char2 __ovld __cnfn select(char2 a, char2 b, uint2 c);
-uchar2 __ovld __cnfn select(uchar2 a, uchar2 b, uint2 c);
-char3 __ovld __cnfn select(char3 a, char3 b, uint3 c);
-uchar3 __ovld __cnfn select(uchar3 a, uchar3 b, uint3 c);
-char4 __ovld __cnfn select(char4 a, char4 b, uint4 c);
-uchar4 __ovld __cnfn select(uchar4 a, uchar4 b, uint4 c);
-char8 __ovld __cnfn select(char8 a, char8 b, uint8 c);
-uchar8 __ovld __cnfn select(uchar8 a, uchar8 b, uint8 c);
-char16 __ovld __cnfn select(char16 a, char16 b, uint16 c);
-uchar16 __ovld __cnfn select(uchar16 a, uchar16 b, uint16 c);
-short __ovld __cnfn select(short a, short b, uint c);
-ushort __ovld __cnfn select(ushort a, ushort b, uint c);
-short2 __ovld __cnfn select(short2 a, short2 b, uint2 c);
-ushort2 __ovld __cnfn select(ushort2 a, ushort2 b, uint2 c);
-short3 __ovld __cnfn select(short3 a, short3 b, uint3 c);
-ushort3 __ovld __cnfn select(ushort3 a, ushort3 b, uint3 c);
-short4 __ovld __cnfn select(short4 a, short4 b, uint4 c);
-ushort4 __ovld __cnfn select(ushort4 a, ushort4 b, uint4 c);
-short8 __ovld __cnfn select(short8 a, short8 b, uint8 c);
-ushort8 __ovld __cnfn select(ushort8 a, ushort8 b, uint8 c);
-short16 __ovld __cnfn select(short16 a, short16 b, uint16 c);
-ushort16 __ovld __cnfn select(ushort16 a, ushort16 b, uint16 c);
+
int __ovld __cnfn select(int a, int b, uint c);
uint __ovld __cnfn select(uint a, uint b, uint c);
int2 __ovld __cnfn select(int2 a, int2 b, uint2 c);
uint8 __ovld __cnfn select(uint8 a, uint8 b, uint8 c);
int16 __ovld __cnfn select(int16 a, int16 b, uint16 c);
uint16 __ovld __cnfn select(uint16 a, uint16 b, uint16 c);
-long __ovld __cnfn select(long a, long b, uint c);
-ulong __ovld __cnfn select(ulong a, ulong b, uint c);
-long2 __ovld __cnfn select(long2 a, long2 b, uint2 c);
-ulong2 __ovld __cnfn select(ulong2 a, ulong2 b, uint2 c);
-long3 __ovld __cnfn select(long3 a, long3 b, uint3 c);
-ulong3 __ovld __cnfn select(ulong3 a, ulong3 b, uint3 c);
-long4 __ovld __cnfn select(long4 a, long4 b, uint4 c);
-ulong4 __ovld __cnfn select(ulong4 a, ulong4 b, uint4 c);
-long8 __ovld __cnfn select(long8 a, long8 b, uint8 c);
-ulong8 __ovld __cnfn select(ulong8 a, ulong8 b, uint8 c);
-long16 __ovld __cnfn select(long16 a, long16 b, uint16 c);
-ulong16 __ovld __cnfn select(ulong16 a, ulong16 b, uint16 c);
float __ovld __cnfn select(float a, float b, uint c);
float2 __ovld __cnfn select(float2 a, float2 b, uint2 c);
float3 __ovld __cnfn select(float3 a, float3 b, uint3 c);
float4 __ovld __cnfn select(float4 a, float4 b, uint4 c);
float8 __ovld __cnfn select(float8 a, float8 b, uint8 c);
float16 __ovld __cnfn select(float16 a, float16 b, uint16 c);
-char __ovld __cnfn select(char a, char b, ulong c);
-uchar __ovld __cnfn select(uchar a, uchar b, ulong c);
-char2 __ovld __cnfn select(char2 a, char2 b, ulong2 c);
-uchar2 __ovld __cnfn select(uchar2 a, uchar2 b, ulong2 c);
-char3 __ovld __cnfn select(char3 a, char3 b, ulong3 c);
-uchar3 __ovld __cnfn select(uchar3 a, uchar3 b, ulong3 c);
-char4 __ovld __cnfn select(char4 a, char4 b, ulong4 c);
-uchar4 __ovld __cnfn select(uchar4 a, uchar4 b, ulong4 c);
-char8 __ovld __cnfn select(char8 a, char8 b, ulong8 c);
-uchar8 __ovld __cnfn select(uchar8 a, uchar8 b, ulong8 c);
-char16 __ovld __cnfn select(char16 a, char16 b, ulong16 c);
-uchar16 __ovld __cnfn select(uchar16 a, uchar16 b, ulong16 c);
-short __ovld __cnfn select(short a, short b, ulong c);
-ushort __ovld __cnfn select(ushort a, ushort b, ulong c);
-short2 __ovld __cnfn select(short2 a, short2 b, ulong2 c);
-ushort2 __ovld __cnfn select(ushort2 a, ushort2 b, ulong2 c);
-short3 __ovld __cnfn select(short3 a, short3 b, ulong3 c);
-ushort3 __ovld __cnfn select(ushort3 a, ushort3 b, ulong3 c);
-short4 __ovld __cnfn select(short4 a, short4 b, ulong4 c);
-ushort4 __ovld __cnfn select(ushort4 a, ushort4 b, ulong4 c);
-short8 __ovld __cnfn select(short8 a, short8 b, ulong8 c);
-ushort8 __ovld __cnfn select(ushort8 a, ushort8 b, ulong8 c);
-short16 __ovld __cnfn select(short16 a, short16 b, ulong16 c);
-ushort16 __ovld __cnfn select(ushort16 a, ushort16 b, ulong16 c);
-int __ovld __cnfn select(int a, int b, ulong c);
-uint __ovld __cnfn select(uint a, uint b, ulong c);
-int2 __ovld __cnfn select(int2 a, int2 b, ulong2 c);
-uint2 __ovld __cnfn select(uint2 a, uint2 b, ulong2 c);
-int3 __ovld __cnfn select(int3 a, int3 b, ulong3 c);
-uint3 __ovld __cnfn select(uint3 a, uint3 b, ulong3 c);
-int4 __ovld __cnfn select(int4 a, int4 b, ulong4 c);
-uint4 __ovld __cnfn select(uint4 a, uint4 b, ulong4 c);
-int8 __ovld __cnfn select(int8 a, int8 b, ulong8 c);
-uint8 __ovld __cnfn select(uint8 a, uint8 b, ulong8 c);
-int16 __ovld __cnfn select(int16 a, int16 b, ulong16 c);
-uint16 __ovld __cnfn select(uint16 a, uint16 b, ulong16 c);
+
long __ovld __cnfn select(long a, long b, ulong c);
ulong __ovld __cnfn select(ulong a, ulong b, ulong c);
long2 __ovld __cnfn select(long2 a, long2 b, ulong2 c);
ulong8 __ovld __cnfn select(ulong8 a, ulong8 b, ulong8 c);
long16 __ovld __cnfn select(long16 a, long16 b, ulong16 c);
ulong16 __ovld __cnfn select(ulong16 a, ulong16 b, ulong16 c);
-float __ovld __cnfn select(float a, float b, ulong c);
-float2 __ovld __cnfn select(float2 a, float2 b, ulong2 c);
-float3 __ovld __cnfn select(float3 a, float3 b, ulong3 c);
-float4 __ovld __cnfn select(float4 a, float4 b, ulong4 c);
-float8 __ovld __cnfn select(float8 a, float8 b, ulong8 c);
-float16 __ovld __cnfn select(float16 a, float16 b, ulong16 c);
+
#ifdef cl_khr_fp64
double __ovld __cnfn select(double a, double b, long c);
double2 __ovld __cnfn select(double2 a, double2 b, long2 c);
*
* vstoren write sizeof (gentypen) bytes given by data to address (p + (offset * n)).
*
- * The address computed as (p + (offset * n)) must be
+ * The address computed as (p + (offset * n)) must be
* 8-bit aligned if gentype is char, uchar;
* 16-bit aligned if gentype is short, ushort, half;
* 32-bit aligned if gentype is int, uint, float;
half16 __ovld vload16(size_t offset, const __constant half *p);
#endif //cl_khr_fp16
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
char2 __ovld vload2(size_t offset, const char *p);
uchar2 __ovld vload2(size_t offset, const uchar *p);
short2 __ovld vload2(size_t offset, const short *p);
half8 __ovld vload8(size_t offset, const __private half *p);
half16 __ovld vload16(size_t offset, const __private half *p);
#endif //cl_khr_fp16
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
void __ovld vstore2(char2 data, size_t offset, char *p);
void __ovld vstore2(uchar2 data, size_t offset, uchar *p);
void __ovld vstore2(short2 data, size_t offset, short *p);
void __ovld vstore8(half8 data, size_t offset, __private half *p);
void __ovld vstore16(half16 data, size_t offset, __private half *p);
#endif //cl_khr_fp16
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Read sizeof (half) bytes of data from address
* must be 16-bit aligned.
*/
float __ovld vload_half(size_t offset, const __constant half *p);
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
float __ovld vload_half(size_t offset, const half *p);
#else
float __ovld vload_half(size_t offset, const __global half *p);
float __ovld vload_half(size_t offset, const __local half *p);
float __ovld vload_half(size_t offset, const __private half *p);
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Read sizeof (halfn) bytes of data from address
float4 __ovld vload_half4(size_t offset, const __constant half *p);
float8 __ovld vload_half8(size_t offset, const __constant half *p);
float16 __ovld vload_half16(size_t offset, const __constant half *p);
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
float2 __ovld vload_half2(size_t offset, const half *p);
float3 __ovld vload_half3(size_t offset, const half *p);
float4 __ovld vload_half4(size_t offset, const half *p);
float4 __ovld vload_half4(size_t offset, const __private half *p);
float8 __ovld vload_half8(size_t offset, const __private half *p);
float16 __ovld vload_half16(size_t offset, const __private half *p);
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* The float value given by data is first
* The default current rounding mode is round to
* nearest even.
*/
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
void __ovld vstore_half(float data, size_t offset, half *p);
void __ovld vstore_half_rte(float data, size_t offset, half *p);
void __ovld vstore_half_rtz(float data, size_t offset, half *p);
void __ovld vstore_half_rtp(double data, size_t offset, __private half *p);
void __ovld vstore_half_rtn(double data, size_t offset, __private half *p);
#endif //cl_khr_fp64
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* The floatn value given by data is converted to
* The default current rounding mode is round to
* nearest even.
*/
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
void __ovld vstore_half2(float2 data, size_t offset, half *p);
void __ovld vstore_half3(float3 data, size_t offset, half *p);
void __ovld vstore_half4(float4 data, size_t offset, half *p);
void __ovld vstore_half8_rtn(double8 data, size_t offset, __private half *p);
void __ovld vstore_half16_rtn(double16 data, size_t offset, __private half *p);
#endif //cl_khr_fp64
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* For n = 1, 2, 4, 8 and 16 read sizeof (halfn)
float4 __ovld vloada_half4(size_t offset, const __constant half *p);
float8 __ovld vloada_half8(size_t offset, const __constant half *p);
float16 __ovld vloada_half16(size_t offset, const __constant half *p);
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
float __ovld vloada_half(size_t offset, const half *p);
float2 __ovld vloada_half2(size_t offset, const half *p);
float3 __ovld vloada_half3(size_t offset, const half *p);
float4 __ovld vloada_half4(size_t offset, const __private half *p);
float8 __ovld vloada_half8(size_t offset, const __private half *p);
float16 __ovld vloada_half16(size_t offset, const __private half *p);
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* The floatn value given by data is converted to
* mode. The default current rounding mode is
* round to nearest even.
*/
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
void __ovld vstorea_half(float data, size_t offset, half *p);
void __ovld vstorea_half2(float2 data, size_t offset, half *p);
void __ovld vstorea_half3(float3 data, size_t offset, half *p);
void __ovld vstorea_half8_rtn(double8 data,size_t offset, __private half *p);
void __ovld vstorea_half16_rtn(double16 data,size_t offset, __private half *p);
#endif //cl_khr_fp64
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
// OpenCL v1.1 s6.11.8, v1.2 s6.12.8, v2.0 s6.13.8 - Synchronization Functions
-// Flag type and values for barrier, mem_fence, read_mem_fence, write_mem_fence
-typedef uint cl_mem_fence_flags;
-
-/**
- * Queue a memory fence to ensure correct
- * ordering of memory operations to local memory
- */
-#define CLK_LOCAL_MEM_FENCE 0x01
-
-/**
- * Queue a memory fence to ensure correct
- * ordering of memory operations to global memory
- */
-#define CLK_GLOBAL_MEM_FENCE 0x02
-
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
-/**
- * Queue a memory fence to ensure correct ordering of memory
- * operations between work-items of a work-group to
- * image memory.
- */
-#define CLK_IMAGE_MEM_FENCE 0x04
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
-
/**
* All work-items in a work-group executing the kernel
* on a processor must execute this function before any
void __ovld __conv barrier(cl_mem_fence_flags flags);
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
-
-typedef enum memory_scope
-{
- memory_scope_work_item,
- memory_scope_work_group,
- memory_scope_device,
- memory_scope_all_svm_devices,
- memory_scope_sub_group
-} memory_scope;
-
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
void __ovld __conv work_group_barrier(cl_mem_fence_flags flags, memory_scope scope);
void __ovld __conv work_group_barrier(cl_mem_fence_flags flags);
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
// OpenCL v1.1 s6.11.9, v1.2 s6.12.9 - Explicit Memory Fence Functions
* Read memory barrier that orders only
* loads.
* The flags argument specifies the memory
- * address space and can be set to to a
+ * address space and can be set to a
* combination of the following literal
* values:
* CLK_LOCAL_MEM_FENCE
* Write memory barrier that orders only
* stores.
* The flags argument specifies the memory
- * address space and can be set to to a
+ * address space and can be set to a
* combination of the following literal
* values:
* CLK_LOCAL_MEM_FENCE
// OpenCL v2.0 s6.13.9 - Address Space Qualifier Functions
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
cl_mem_fence_flags __ovld get_fence(const void *ptr);
cl_mem_fence_flags __ovld get_fence(void *ptr);
-/**
+/**
* Builtin functions to_global, to_local, and to_private need to be declared as Clang builtin functions
* and checked in Sema since they should be declared as
* addr gentype* to_addr (gentype*);
* where gentype is builtin type or user defined type.
*/
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
// OpenCL v1.1 s6.11.10, v1.2 s6.12.10, v2.0 s6.13.10 - Async Copies from Global to Local Memory, Local to Global Memory, and Prefetch
unsigned int __ovld atomic_add(volatile __global unsigned int *p, unsigned int val);
int __ovld atomic_add(volatile __local int *p, int val);
unsigned int __ovld atomic_add(volatile __local unsigned int *p, unsigned int val);
+#ifdef __OPENCL_CPP_VERSION__
+int __ovld atomic_add(volatile int *p, int val);
+unsigned int __ovld atomic_add(volatile unsigned int *p, unsigned int val);
+#endif
#if defined(cl_khr_global_int32_base_atomics)
int __ovld atom_add(volatile __global int *p, int val);
unsigned int __ovld atomic_sub(volatile __global unsigned int *p, unsigned int val);
int __ovld atomic_sub(volatile __local int *p, int val);
unsigned int __ovld atomic_sub(volatile __local unsigned int *p, unsigned int val);
+#ifdef __OPENCL_CPP_VERSION__
+int __ovld atomic_sub(volatile int *p, int val);
+unsigned int __ovld atomic_sub(volatile unsigned int *p, unsigned int val);
+#endif
#if defined(cl_khr_global_int32_base_atomics)
int __ovld atom_sub(volatile __global int *p, int val);
unsigned int __ovld atomic_xchg(volatile __local unsigned int *p, unsigned int val);
float __ovld atomic_xchg(volatile __global float *p, float val);
float __ovld atomic_xchg(volatile __local float *p, float val);
+#ifdef __OPENCL_CPP_VERSION__
+int __ovld atomic_xchg(volatile int *p, int val);
+unsigned int __ovld atomic_xchg(volatile unsigned int *p, unsigned int val);
+float __ovld atomic_xchg(volatile float *p, float val);
+#endif
#if defined(cl_khr_global_int32_base_atomics)
int __ovld atom_xchg(volatile __global int *p, int val);
unsigned int __ovld atomic_inc(volatile __global unsigned int *p);
int __ovld atomic_inc(volatile __local int *p);
unsigned int __ovld atomic_inc(volatile __local unsigned int *p);
+#ifdef __OPENCL_CPP_VERSION__
+int __ovld atomic_inc(volatile int *p);
+unsigned int __ovld atomic_inc(volatile unsigned int *p);
+#endif
#if defined(cl_khr_global_int32_base_atomics)
int __ovld atom_inc(volatile __global int *p);
unsigned int __ovld atomic_dec(volatile __global unsigned int *p);
int __ovld atomic_dec(volatile __local int *p);
unsigned int __ovld atomic_dec(volatile __local unsigned int *p);
+#ifdef __OPENCL_CPP_VERSION__
+int __ovld atomic_dec(volatile int *p);
+unsigned int __ovld atomic_dec(volatile unsigned int *p);
+#endif
#if defined(cl_khr_global_int32_base_atomics)
int __ovld atom_dec(volatile __global int *p);
unsigned int __ovld atomic_cmpxchg(volatile __global unsigned int *p, unsigned int cmp, unsigned int val);
int __ovld atomic_cmpxchg(volatile __local int *p, int cmp, int val);
unsigned int __ovld atomic_cmpxchg(volatile __local unsigned int *p, unsigned int cmp, unsigned int val);
+#ifdef __OPENCL_CPP_VERSION__
+int __ovld atomic_cmpxchg(volatile int *p, int cmp, int val);
+unsigned int __ovld atomic_cmpxchg(volatile unsigned int *p, unsigned int cmp, unsigned int val);
+#endif
#if defined(cl_khr_global_int32_base_atomics)
int __ovld atom_cmpxchg(volatile __global int *p, int cmp, int val);
unsigned int __ovld atomic_min(volatile __global unsigned int *p, unsigned int val);
int __ovld atomic_min(volatile __local int *p, int val);
unsigned int __ovld atomic_min(volatile __local unsigned int *p, unsigned int val);
+#ifdef __OPENCL_CPP_VERSION__
+int __ovld atomic_min(volatile int *p, int val);
+unsigned int __ovld atomic_min(volatile unsigned int *p, unsigned int val);
+#endif
#if defined(cl_khr_global_int32_extended_atomics)
int __ovld atom_min(volatile __global int *p, int val);
unsigned int __ovld atomic_max(volatile __global unsigned int *p, unsigned int val);
int __ovld atomic_max(volatile __local int *p, int val);
unsigned int __ovld atomic_max(volatile __local unsigned int *p, unsigned int val);
+#ifdef __OPENCL_CPP_VERSION__
+int __ovld atomic_max(volatile int *p, int val);
+unsigned int __ovld atomic_max(volatile unsigned int *p, unsigned int val);
+#endif
#if defined(cl_khr_global_int32_extended_atomics)
int __ovld atom_max(volatile __global int *p, int val);
unsigned int __ovld atomic_and(volatile __global unsigned int *p, unsigned int val);
int __ovld atomic_and(volatile __local int *p, int val);
unsigned int __ovld atomic_and(volatile __local unsigned int *p, unsigned int val);
+#ifdef __OPENCL_CPP_VERSION__
+int __ovld atomic_and(volatile int *p, int val);
+unsigned int __ovld atomic_and(volatile unsigned int *p, unsigned int val);
+#endif
#if defined(cl_khr_global_int32_extended_atomics)
int __ovld atom_and(volatile __global int *p, int val);
unsigned int __ovld atomic_or(volatile __global unsigned int *p, unsigned int val);
int __ovld atomic_or(volatile __local int *p, int val);
unsigned int __ovld atomic_or(volatile __local unsigned int *p, unsigned int val);
+#ifdef __OPENCL_CPP_VERSION__
+int __ovld atomic_or(volatile int *p, int val);
+unsigned int __ovld atomic_or(volatile unsigned int *p, unsigned int val);
+#endif
#if defined(cl_khr_global_int32_extended_atomics)
int __ovld atom_or(volatile __global int *p, int val);
unsigned int __ovld atomic_xor(volatile __global unsigned int *p, unsigned int val);
int __ovld atomic_xor(volatile __local int *p, int val);
unsigned int __ovld atomic_xor(volatile __local unsigned int *p, unsigned int val);
+#ifdef __OPENCL_CPP_VERSION__
+int __ovld atomic_xor(volatile int *p, int val);
+unsigned int __ovld atomic_xor(volatile unsigned int *p, unsigned int val);
+#endif
#if defined(cl_khr_global_int32_extended_atomics)
int __ovld atom_xor(volatile __global int *p, int val);
// OpenCL v2.0 s6.13.11 - Atomics Functions
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
-#ifndef ATOMIC_VAR_INIT
-#define ATOMIC_VAR_INIT(x) (x)
-#endif //ATOMIC_VAR_INIT
-#define ATOMIC_FLAG_INIT 0
-
-// enum values aligned with what clang uses in EmitAtomicExpr()
-typedef enum memory_order
-{
- memory_order_relaxed,
- memory_order_acquire,
- memory_order_release,
- memory_order_acq_rel,
- memory_order_seq_cst
-} memory_order;
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
// double atomics support requires extensions cl_khr_int64_base_atomics and cl_khr_int64_extended_atomics
#if defined(cl_khr_int64_base_atomics) && defined(cl_khr_int64_extended_atomics)
uint __ovld atomic_fetch_min(volatile atomic_uint *object, uint operand);
uint __ovld atomic_fetch_min_explicit(volatile atomic_uint *object, uint operand, memory_order order);
uint __ovld atomic_fetch_min_explicit(volatile atomic_uint *object, uint operand, memory_order order, memory_scope scope);
-uint __ovld atomic_fetch_min(volatile atomic_uint *object, int operand);
-uint __ovld atomic_fetch_min_explicit(volatile atomic_uint *object, int operand, memory_order order);
-uint __ovld atomic_fetch_min_explicit(volatile atomic_uint *object, int operand, memory_order order, memory_scope scope);
int __ovld atomic_fetch_max(volatile atomic_int *object, int operand);
int __ovld atomic_fetch_max_explicit(volatile atomic_int *object, int operand, memory_order order);
int __ovld atomic_fetch_max_explicit(volatile atomic_int *object, int operand, memory_order order, memory_scope scope);
uint __ovld atomic_fetch_max(volatile atomic_uint *object, uint operand);
uint __ovld atomic_fetch_max_explicit(volatile atomic_uint *object, uint operand, memory_order order);
uint __ovld atomic_fetch_max_explicit(volatile atomic_uint *object, uint operand, memory_order order, memory_scope scope);
-uint __ovld atomic_fetch_max(volatile atomic_uint *object, int operand);
-uint __ovld atomic_fetch_max_explicit(volatile atomic_uint *object, int operand, memory_order order);
-uint __ovld atomic_fetch_max_explicit(volatile atomic_uint *object, int operand, memory_order order, memory_scope scope);
#if defined(cl_khr_int64_base_atomics) && defined(cl_khr_int64_extended_atomics)
long __ovld atomic_fetch_add(volatile atomic_long *object, long operand);
ulong __ovld atomic_fetch_min(volatile atomic_ulong *object, ulong operand);
ulong __ovld atomic_fetch_min_explicit(volatile atomic_ulong *object, ulong operand, memory_order order);
ulong __ovld atomic_fetch_min_explicit(volatile atomic_ulong *object, ulong operand, memory_order order, memory_scope scope);
-ulong __ovld atomic_fetch_min(volatile atomic_ulong *object, long operand);
-ulong __ovld atomic_fetch_min_explicit(volatile atomic_ulong *object, long operand, memory_order order);
-ulong __ovld atomic_fetch_min_explicit(volatile atomic_ulong *object, long operand, memory_order order, memory_scope scope);
long __ovld atomic_fetch_max(volatile atomic_long *object, long operand);
long __ovld atomic_fetch_max_explicit(volatile atomic_long *object, long operand, memory_order order);
long __ovld atomic_fetch_max_explicit(volatile atomic_long *object, long operand, memory_order order, memory_scope scope);
ulong __ovld atomic_fetch_max(volatile atomic_ulong *object, ulong operand);
ulong __ovld atomic_fetch_max_explicit(volatile atomic_ulong *object, ulong operand, memory_order order);
ulong __ovld atomic_fetch_max_explicit(volatile atomic_ulong *object, ulong operand, memory_order order, memory_scope scope);
-ulong __ovld atomic_fetch_max(volatile atomic_ulong *object, long operand);
-ulong __ovld atomic_fetch_max_explicit(volatile atomic_ulong *object, long operand, memory_order order);
-ulong __ovld atomic_fetch_max_explicit(volatile atomic_ulong *object, long operand, memory_order order, memory_scope scope);
#endif //defined(cl_khr_int64_base_atomics) && defined(cl_khr_int64_extended_atomics)
// OpenCL v2.0 s6.13.11.7.5:
// add/sub: atomic type argument can be uintptr_t/intptr_t, value type argument can be ptrdiff_t.
// or/xor/and/min/max: atomic type argument can be intptr_t/uintptr_t, value type argument can be intptr_t/uintptr_t.
-#if defined(cl_khr_int64_base_atomics) && defined(cl_khr_int64_extended_atomics)
+#if defined(cl_khr_int64_base_atomics) && defined(cl_khr_int64_extended_atomics)
uintptr_t __ovld atomic_fetch_add(volatile atomic_uintptr_t *object, ptrdiff_t operand);
uintptr_t __ovld atomic_fetch_add_explicit(volatile atomic_uintptr_t *object, ptrdiff_t operand, memory_order order);
uintptr_t __ovld atomic_fetch_add_explicit(volatile atomic_uintptr_t *object, ptrdiff_t operand, memory_order order, memory_scope scope);
void __ovld atomic_flag_clear_explicit(volatile atomic_flag *object, memory_order order);
void __ovld atomic_flag_clear_explicit(volatile atomic_flag *object, memory_order order, memory_scope scope);
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
// OpenCL v1.1 s6.11.12, v1.2 s6.12.12, v2.0 s6.13.12 - Miscellaneous Vector Functions
half16 __ovld __cnfn shuffle2(half16 x, half16 y, ushort16 mask);
#endif //cl_khr_fp16
-#if __OPENCL_C_VERSION__ >= CL_VERSION_1_2
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_1_2)
// OpenCL v1.2 s6.12.13, v2.0 s6.13.13 - printf
-int printf(__constant const char* st, ...);
+int printf(__constant const char* st, ...) __attribute__((format(printf, 1, 2)));
#endif
// OpenCL v1.1 s6.11.3, v1.2 s6.12.14, v2.0 s6.13.14 - Image Read and Write Functions
-// These values need to match the runtime equivalent
-//
-// Addressing Mode.
-//
-#define CLK_ADDRESS_NONE 0
-#define CLK_ADDRESS_CLAMP_TO_EDGE 2
-#define CLK_ADDRESS_CLAMP 4
-#define CLK_ADDRESS_REPEAT 6
-#define CLK_ADDRESS_MIRRORED_REPEAT 8
-
-//
-// Coordination Normalization
-//
-#define CLK_NORMALIZED_COORDS_FALSE 0
-#define CLK_NORMALIZED_COORDS_TRUE 1
-
-//
-// Filtering Mode.
-//
-#define CLK_FILTER_NEAREST 0x10
-#define CLK_FILTER_LINEAR 0x20
-
#ifdef cl_khr_gl_msaa_sharing
#pragma OPENCL EXTENSION cl_khr_gl_msaa_sharing : enable
#endif //cl_khr_gl_msaa_sharing
* only. The filter_mode specified in sampler
* must be set to CLK_FILTER_NEAREST; otherwise
* the values returned are undefined.
-
+
* The read_image{f|i|ui} calls that take
* integer coordinates must use a sampler with
* normalized coordinates set to
uint4 __purefn __ovld read_imageui(read_only image3d_t image, sampler_t sampler, int4 coord);
uint4 __purefn __ovld read_imageui(read_only image3d_t image, sampler_t sampler, float4 coord);
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_1_2)
float4 __purefn __ovld read_imagef(read_only image2d_array_t image_array, sampler_t sampler, int4 coord);
float4 __purefn __ovld read_imagef(read_only image2d_array_t image_array, sampler_t sampler, float4 coord);
int4 __purefn __ovld read_imagei(read_only image2d_array_t image_array, sampler_t sampler, float4 coord);
uint4 __purefn __ovld read_imageui(read_only image2d_array_t image_array, sampler_t sampler, int4 coord);
uint4 __purefn __ovld read_imageui(read_only image2d_array_t image_array, sampler_t sampler, float4 coord);
+#endif // defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_1_2)
float4 __purefn __ovld read_imagef(read_only image1d_t image, sampler_t sampler, int coord);
float4 __purefn __ovld read_imagef(read_only image1d_t image, sampler_t sampler, float coord);
uint4 __purefn __ovld read_imageui(read_only image1d_t image, sampler_t sampler, int coord);
uint4 __purefn __ovld read_imageui(read_only image1d_t image, sampler_t sampler, float coord);
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_1_2)
float4 __purefn __ovld read_imagef(read_only image1d_array_t image_array, sampler_t sampler, int2 coord);
float4 __purefn __ovld read_imagef(read_only image1d_array_t image_array, sampler_t sampler, float2 coord);
int4 __purefn __ovld read_imagei(read_only image1d_array_t image_array, sampler_t sampler, float2 coord);
uint4 __purefn __ovld read_imageui(read_only image1d_array_t image_array, sampler_t sampler, int2 coord);
uint4 __purefn __ovld read_imageui(read_only image1d_array_t image_array, sampler_t sampler, float2 coord);
+#endif // defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_1_2)
#ifdef cl_khr_depth_images
float __purefn __ovld read_imagef(read_only image2d_depth_t image, sampler_t sampler, float2 coord);
#endif //cl_khr_gl_msaa_sharing
// OpenCL Extension v2.0 s9.18 - Mipmaps
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
#ifdef cl_khr_mipmap_image
float4 __purefn __ovld read_imagef(read_only image1d_t image, sampler_t sampler, float coord, float lod);
int4 __purefn __ovld read_imagei(read_only image3d_t image, sampler_t sampler, float4 coord, float4 gradientX, float4 gradientY);
uint4 __purefn __ovld read_imageui(read_only image3d_t image, sampler_t sampler, float4 coord, float4 gradientX, float4 gradientY);
-float4 __purefn __ovld read_imagef(read_only image1d_t image, sampler_t sampler, float coord, float lod);
-int4 __purefn __ovld read_imagei(read_only image1d_t image, sampler_t sampler, float coord, float lod);
-uint4 __purefn __ovld read_imageui(read_only image1d_t image, sampler_t sampler, float coord, float lod);
-
-float4 __purefn __ovld read_imagef(read_only image1d_array_t image_array, sampler_t sampler, float2 coord, float lod);
-int4 __purefn __ovld read_imagei(read_only image1d_array_t image_array, sampler_t sampler, float2 coord, float lod);
-uint4 __purefn __ovld read_imageui(read_only image1d_array_t image_array, sampler_t sampler, float2 coord, float lod);
-
-float4 __purefn __ovld read_imagef(read_only image2d_t image, sampler_t sampler, float2 coord, float lod);
-int4 __purefn __ovld read_imagei(read_only image2d_t image, sampler_t sampler, float2 coord, float lod);
-uint4 __purefn __ovld read_imageui(read_only image2d_t image, sampler_t sampler, float2 coord, float lod);
-
-float __purefn __ovld read_imagef(read_only image2d_depth_t image, sampler_t sampler, float2 coord, float lod);
-
-float4 __purefn __ovld read_imagef(read_only image2d_array_t image_array, sampler_t sampler, float4 coord, float lod);
-int4 __purefn __ovld read_imagei(read_only image2d_array_t image_array, sampler_t sampler, float4 coord, float lod);
-uint4 __purefn __ovld read_imageui(read_only image2d_array_t image_array, sampler_t sampler, float4 coord, float lod);
-
-float __purefn __ovld read_imagef(read_only image2d_array_depth_t image, sampler_t sampler, float4 coord, float lod);
-
-float4 __purefn __ovld read_imagef(read_only image3d_t image, sampler_t sampler, float4 coord, float lod);
-int4 __purefn __ovld read_imagei(read_only image3d_t image, sampler_t sampler, float4 coord, float lod);
-uint4 __purefn __ovld read_imageui(read_only image3d_t image, sampler_t sampler, float4 coord, float lod);
-
#endif //cl_khr_mipmap_image
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_1_2)
/**
* Sampler-less Image Access
int4 __purefn __ovld read_imagei(read_only image3d_t image, int4 coord);
uint4 __purefn __ovld read_imageui(read_only image3d_t image, int4 coord);
+#endif // defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_1_2)
+
// Image read functions returning half4 type
#ifdef cl_khr_fp16
half4 __purefn __ovld read_imageh(read_only image1d_t image, sampler_t sampler, int coord);
half4 __purefn __ovld read_imageh(read_only image1d_t image, sampler_t sampler, float coord);
-half4 __purefn __ovld read_imageh(read_only image1d_array_t image, sampler_t sampler, int2 coord);
-half4 __purefn __ovld read_imageh(read_only image1d_array_t image, sampler_t sampler, float2 coord);
half4 __purefn __ovld read_imageh(read_only image2d_t image, sampler_t sampler, int2 coord);
half4 __purefn __ovld read_imageh(read_only image2d_t image, sampler_t sampler, float2 coord);
half4 __purefn __ovld read_imageh(read_only image3d_t image, sampler_t sampler, int4 coord);
half4 __purefn __ovld read_imageh(read_only image3d_t image, sampler_t sampler, float4 coord);
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_1_2)
+half4 __purefn __ovld read_imageh(read_only image1d_array_t image, sampler_t sampler, int2 coord);
+half4 __purefn __ovld read_imageh(read_only image1d_array_t image, sampler_t sampler, float2 coord);
half4 __purefn __ovld read_imageh(read_only image2d_array_t image, sampler_t sampler, int4 coord);
half4 __purefn __ovld read_imageh(read_only image2d_array_t image, sampler_t sampler, float4 coord);
+/**
+ * Sampler-less Image Access
+ */
half4 __purefn __ovld read_imageh(read_only image1d_t image, int coord);
half4 __purefn __ovld read_imageh(read_only image2d_t image, int2 coord);
half4 __purefn __ovld read_imageh(read_only image3d_t image, int4 coord);
half4 __purefn __ovld read_imageh(read_only image1d_array_t image, int2 coord);
half4 __purefn __ovld read_imageh(read_only image2d_array_t image, int4 coord);
half4 __purefn __ovld read_imageh(read_only image1d_buffer_t image, int coord);
+#endif // defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_1_2)
#endif //cl_khr_fp16
// Image read functions for read_write images
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
float4 __purefn __ovld read_imagef(read_write image1d_t image, int coord);
int4 __purefn __ovld read_imagei(read_write image1d_t image, int coord);
uint4 __purefn __ovld read_imageui(read_write image1d_t image, int coord);
float __purefn __ovld read_imagef(read_write image2d_array_msaa_depth_t image, int4 coord, int sample);
#endif //cl_khr_gl_msaa_sharing
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
#ifdef cl_khr_mipmap_image
float4 __purefn __ovld read_imagef(read_write image1d_t image, sampler_t sampler, float coord, float lod);
int4 __purefn __ovld read_imagei(read_write image1d_t image, sampler_t sampler, float coord, float lod);
int4 __purefn __ovld read_imagei(read_write image3d_t image, sampler_t sampler, float4 coord, float4 gradientX, float4 gradientY);
uint4 __purefn __ovld read_imageui(read_write image3d_t image, sampler_t sampler, float4 coord, float4 gradientX, float4 gradientY);
-float4 __purefn __ovld read_imagef(read_write image1d_t image, sampler_t sampler, float coord, float lod);
-int4 __purefn __ovld read_imagei(read_write image1d_t image, sampler_t sampler, float coord, float lod);
-uint4 __purefn __ovld read_imageui(read_write image1d_t image, sampler_t sampler, float coord, float lod);
-
-float4 __purefn __ovld read_imagef(read_write image1d_array_t image_array, sampler_t sampler, float2 coord, float lod);
-int4 __purefn __ovld read_imagei(read_write image1d_array_t image_array, sampler_t sampler, float2 coord, float lod);
-uint4 __purefn __ovld read_imageui(read_write image1d_array_t image_array, sampler_t sampler, float2 coord, float lod);
-
-float4 __purefn __ovld read_imagef(read_write image2d_t image, sampler_t sampler, float2 coord, float lod);
-int4 __purefn __ovld read_imagei(read_write image2d_t image, sampler_t sampler, float2 coord, float lod);
-uint4 __purefn __ovld read_imageui(read_write image2d_t image, sampler_t sampler, float2 coord, float lod);
-
-float __purefn __ovld read_imagef(read_write image2d_depth_t image, sampler_t sampler, float2 coord, float lod);
-
-float4 __purefn __ovld read_imagef(read_write image2d_array_t image_array, sampler_t sampler, float4 coord, float lod);
-int4 __purefn __ovld read_imagei(read_write image2d_array_t image_array, sampler_t sampler, float4 coord, float lod);
-uint4 __purefn __ovld read_imageui(read_write image2d_array_t image_array, sampler_t sampler, float4 coord, float lod);
-
-float __purefn __ovld read_imagef(read_write image2d_array_depth_t image, sampler_t sampler, float4 coord, float lod);
-
-float4 __purefn __ovld read_imagef(read_write image3d_t image, sampler_t sampler, float4 coord, float lod);
-int4 __purefn __ovld read_imagei(read_write image3d_t image, sampler_t sampler, float4 coord, float lod);
-uint4 __purefn __ovld read_imageui(read_write image3d_t image, sampler_t sampler, float4 coord, float lod);
#endif //cl_khr_mipmap_image
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
// Image read functions returning half4 type
#ifdef cl_khr_fp16
half4 __purefn __ovld read_imageh(read_write image2d_array_t image, int4 coord);
half4 __purefn __ovld read_imageh(read_write image1d_buffer_t image, int coord);
#endif //cl_khr_fp16
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Write color value to location specified by coordinate
#endif //cl_khr_depth_images
// OpenCL Extension v2.0 s9.18 - Mipmaps
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
-#ifdef cl_khr_mipmap_image
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+#if defined(cl_khr_mipmap_image_writes)
void __ovld write_imagef(write_only image1d_t image, int coord, int lod, float4 color);
void __ovld write_imagei(write_only image1d_t image, int coord, int lod, int4 color);
void __ovld write_imageui(write_only image1d_t image, int coord, int lod, uint4 color);
void __ovld write_imagei(write_only image2d_array_t image_array, int4 coord, int lod, int4 color);
void __ovld write_imageui(write_only image2d_array_t image_array, int4 coord, int lod, uint4 color);
-void __ovld write_imagef(write_only image2d_depth_t image, int2 coord, int lod, float color);
-void __ovld write_imagef(write_only image2d_array_depth_t image, int4 coord, int lod, float color);
+void __ovld write_imagef(write_only image2d_depth_t image, int2 coord, int lod, float depth);
+void __ovld write_imagef(write_only image2d_array_depth_t image, int4 coord, int lod, float depth);
#ifdef cl_khr_3d_image_writes
void __ovld write_imagef(write_only image3d_t image, int4 coord, int lod, float4 color);
void __ovld write_imagei(write_only image3d_t image, int4 coord, int lod, int4 color);
void __ovld write_imageui(write_only image3d_t image, int4 coord, int lod, uint4 color);
-#endif
-#endif //cl_khr_mipmap_image
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //cl_khr_3d_image_writes
+
+#endif //defined(cl_khr_mipmap_image_writes)
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
// Image write functions for half4 type
#ifdef cl_khr_fp16
#endif //cl_khr_fp16
// Image write functions for read_write images
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
void __ovld write_imagef(read_write image2d_t image, int2 coord, float4 color);
void __ovld write_imagei(read_write image2d_t image, int2 coord, int4 color);
void __ovld write_imageui(read_write image2d_t image, int2 coord, uint4 color);
void __ovld write_imagef(read_write image2d_array_depth_t image, int4 coord, float color);
#endif //cl_khr_depth_images
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
-#ifdef cl_khr_mipmap_image
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+#if defined(cl_khr_mipmap_image_writes)
void __ovld write_imagef(read_write image1d_t image, int coord, int lod, float4 color);
void __ovld write_imagei(read_write image1d_t image, int coord, int lod, int4 color);
void __ovld write_imageui(read_write image1d_t image, int coord, int lod, uint4 color);
void __ovld write_imagef(read_write image3d_t image, int4 coord, int lod, float4 color);
void __ovld write_imagei(read_write image3d_t image, int4 coord, int lod, int4 color);
void __ovld write_imageui(read_write image3d_t image, int4 coord, int lod, uint4 color);
-#endif
-#endif //cl_khr_mipmap_image
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //cl_khr_3d_image_writes
+
+#endif //cl_khr_mipmap_image_writes
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
// Image write functions for half4 type
#ifdef cl_khr_fp16
void __ovld write_imageh(read_write image2d_array_t image, int4 coord, half4 color);
void __ovld write_imageh(read_write image1d_buffer_t image, int coord, half4 color);
#endif //cl_khr_fp16
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
// Note: In OpenCL v1.0/1.1/1.2, image argument of image query builtin functions does not have
// access qualifier, which by default assume read_only access qualifier. Image query builtin
int __ovld __cnfn get_image_width(write_only image2d_array_msaa_depth_t image);
#endif //cl_khr_gl_msaa_sharing
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
int __ovld __cnfn get_image_width(read_write image1d_t image);
int __ovld __cnfn get_image_width(read_write image1d_buffer_t image);
int __ovld __cnfn get_image_width(read_write image2d_t image);
int __ovld __cnfn get_image_width(read_write image2d_array_msaa_t image);
int __ovld __cnfn get_image_width(read_write image2d_array_msaa_depth_t image);
#endif //cl_khr_gl_msaa_sharing
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Return the image height in pixels.
int __ovld __cnfn get_image_height(write_only image2d_array_msaa_depth_t image);
#endif //cl_khr_gl_msaa_sharing
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
int __ovld __cnfn get_image_height(read_write image2d_t image);
int __ovld __cnfn get_image_height(read_write image3d_t image);
int __ovld __cnfn get_image_height(read_write image2d_array_t image);
int __ovld __cnfn get_image_height(read_write image2d_array_msaa_t image);
int __ovld __cnfn get_image_height(read_write image2d_array_msaa_depth_t image);
#endif //cl_khr_gl_msaa_sharing
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Return the image depth in pixels.
int __ovld __cnfn get_image_depth(write_only image3d_t image);
#endif
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
int __ovld __cnfn get_image_depth(read_write image3d_t image);
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
// OpenCL Extension v2.0 s9.18 - Mipmaps
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
#ifdef cl_khr_mipmap_image
/**
* Return the image miplevels.
int __ovld get_image_num_mip_levels(read_write image2d_depth_t image);
#endif //cl_khr_mipmap_image
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Return the channel data type. Valid values are:
* CLK_FLOAT
*/
-//
-// Channel Datatype.
-//
-#define CLK_SNORM_INT8 0x10D0
-#define CLK_SNORM_INT16 0x10D1
-#define CLK_UNORM_INT8 0x10D2
-#define CLK_UNORM_INT16 0x10D3
-#define CLK_UNORM_SHORT_565 0x10D4
-#define CLK_UNORM_SHORT_555 0x10D5
-#define CLK_UNORM_INT_101010 0x10D6
-#define CLK_SIGNED_INT8 0x10D7
-#define CLK_SIGNED_INT16 0x10D8
-#define CLK_SIGNED_INT32 0x10D9
-#define CLK_UNSIGNED_INT8 0x10DA
-#define CLK_UNSIGNED_INT16 0x10DB
-#define CLK_UNSIGNED_INT32 0x10DC
-#define CLK_HALF_FLOAT 0x10DD
-#define CLK_FLOAT 0x10DE
-#define CLK_UNORM_INT24 0x10DF
-
int __ovld __cnfn get_image_channel_data_type(read_only image1d_t image);
int __ovld __cnfn get_image_channel_data_type(read_only image1d_buffer_t image);
int __ovld __cnfn get_image_channel_data_type(read_only image2d_t image);
int __ovld __cnfn get_image_channel_data_type(write_only image2d_array_msaa_depth_t image);
#endif //cl_khr_gl_msaa_sharing
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
int __ovld __cnfn get_image_channel_data_type(read_write image1d_t image);
int __ovld __cnfn get_image_channel_data_type(read_write image1d_buffer_t image);
int __ovld __cnfn get_image_channel_data_type(read_write image2d_t image);
int __ovld __cnfn get_image_channel_data_type(read_write image2d_array_msaa_t image);
int __ovld __cnfn get_image_channel_data_type(read_write image2d_array_msaa_depth_t image);
#endif //cl_khr_gl_msaa_sharing
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Return the image channel order. Valid values are:
* CLK_INTENSITY
* CLK_LUMINANCE
*/
-// Channel order, numbering must be aligned with cl_channel_order in cl.h
-//
-#define CLK_R 0x10B0
-#define CLK_A 0x10B1
-#define CLK_RG 0x10B2
-#define CLK_RA 0x10B3
-#define CLK_RGB 0x10B4
-#define CLK_RGBA 0x10B5
-#define CLK_BGRA 0x10B6
-#define CLK_ARGB 0x10B7
-#define CLK_INTENSITY 0x10B8
-#define CLK_LUMINANCE 0x10B9
-#define CLK_Rx 0x10BA
-#define CLK_RGx 0x10BB
-#define CLK_RGBx 0x10BC
-#define CLK_DEPTH 0x10BD
-#define CLK_DEPTH_STENCIL 0x10BE
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
-#define CLK_sRGB 0x10BF
-#define CLK_sRGBA 0x10C1
-#define CLK_sRGBx 0x10C0
-#define CLK_sBGRA 0x10C2
-#define CLK_ABGR 0x10C3
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
int __ovld __cnfn get_image_channel_order(read_only image1d_t image);
int __ovld __cnfn get_image_channel_order(read_only image1d_buffer_t image);
int __ovld __cnfn get_image_channel_order(write_only image2d_array_msaa_depth_t image);
#endif //cl_khr_gl_msaa_sharing
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
int __ovld __cnfn get_image_channel_order(read_write image1d_t image);
int __ovld __cnfn get_image_channel_order(read_write image1d_buffer_t image);
int __ovld __cnfn get_image_channel_order(read_write image2d_t image);
int __ovld __cnfn get_image_channel_order(read_write image2d_array_msaa_t image);
int __ovld __cnfn get_image_channel_order(read_write image2d_array_msaa_depth_t image);
#endif //cl_khr_gl_msaa_sharing
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Return the 2D image width and height as an int2
int2 __ovld __cnfn get_image_dim(write_only image2d_array_msaa_depth_t image);
#endif //cl_khr_gl_msaa_sharing
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
int2 __ovld __cnfn get_image_dim(read_write image2d_t image);
int2 __ovld __cnfn get_image_dim(read_write image2d_array_t image);
#ifdef cl_khr_depth_images
int2 __ovld __cnfn get_image_dim(read_write image2d_array_msaa_t image);
int2 __ovld __cnfn get_image_dim(read_write image2d_array_msaa_depth_t image);
#endif //cl_khr_gl_msaa_sharing
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Return the 3D image width, height, and depth as an
#ifdef cl_khr_3d_image_writes
int4 __ovld __cnfn get_image_dim(write_only image3d_t image);
#endif
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
int4 __ovld __cnfn get_image_dim(read_write image3d_t image);
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Return the image array size.
size_t __ovld __cnfn get_image_array_size(write_only image2d_array_msaa_depth_t image_array);
#endif //cl_khr_gl_msaa_sharing
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
size_t __ovld __cnfn get_image_array_size(read_write image1d_array_t image_array);
size_t __ovld __cnfn get_image_array_size(read_write image2d_array_t image_array);
#ifdef cl_khr_depth_images
size_t __ovld __cnfn get_image_array_size(read_write image2d_array_msaa_t image_array);
size_t __ovld __cnfn get_image_array_size(read_write image2d_array_msaa_depth_t image_array);
#endif //cl_khr_gl_msaa_sharing
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
/**
* Return the number of samples associated with image
#if defined(cl_khr_gl_msaa_sharing)
int __ovld get_image_num_samples(read_only image2d_msaa_t image);
int __ovld get_image_num_samples(read_only image2d_msaa_depth_t image);
-int __ovld get_image_num_samples(read_only image2d_array_msaa_depth_t image);
int __ovld get_image_num_samples(read_only image2d_array_msaa_t image);
int __ovld get_image_num_samples(read_only image2d_array_msaa_depth_t image);
int __ovld get_image_num_samples(write_only image2d_msaa_t image);
int __ovld get_image_num_samples(write_only image2d_msaa_depth_t image);
-int __ovld get_image_num_samples(write_only image2d_array_msaa_depth_t image);
int __ovld get_image_num_samples(write_only image2d_array_msaa_t image);
int __ovld get_image_num_samples(write_only image2d_array_msaa_depth_t image);
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
int __ovld get_image_num_samples(read_write image2d_msaa_t image);
int __ovld get_image_num_samples(read_write image2d_msaa_depth_t image);
-int __ovld get_image_num_samples(read_write image2d_array_msaa_depth_t image);
int __ovld get_image_num_samples(read_write image2d_array_msaa_t image);
int __ovld get_image_num_samples(read_write image2d_array_msaa_depth_t image);
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
#endif
// OpenCL v2.0 s6.13.15 - Work-group Functions
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
int __ovld __conv work_group_all(int predicate);
int __ovld __conv work_group_any(int predicate);
double __ovld __conv work_group_scan_inclusive_max(double x);
#endif //cl_khr_fp64
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
// OpenCL v2.0 s6.13.16 - Pipe Functions
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
-#define PIPE_RESERVE_ID_VALID_BIT (1U << 30)
-#define CLK_NULL_RESERVE_ID (__builtin_astype(((void*)(__SIZE_MAX__)), reserve_id_t))
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
bool __ovld is_valid_reserve_id(reserve_id_t reserve_id);
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
// OpenCL v2.0 s6.13.17 - Enqueue Kernels
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
-
-#define CL_COMPLETE 0x0
-#define CL_RUNNING 0x1
-#define CL_SUBMITTED 0x2
-#define CL_QUEUED 0x3
-
-#define CLK_SUCCESS 0
-#define CLK_ENQUEUE_FAILURE -101
-#define CLK_INVALID_QUEUE -102
-#define CLK_INVALID_NDRANGE -160
-#define CLK_INVALID_EVENT_WAIT_LIST -57
-#define CLK_DEVICE_QUEUE_FULL -161
-#define CLK_INVALID_ARG_SIZE -51
-#define CLK_EVENT_ALLOCATION_FAILURE -100
-#define CLK_OUT_OF_RESOURCES -5
-
-#define CLK_NULL_QUEUE 0
-#define CLK_NULL_EVENT (__builtin_astype(((void*)(__SIZE_MAX__)), clk_event_t))
-
-// execution model related definitions
-#define CLK_ENQUEUE_FLAGS_NO_WAIT 0x0
-#define CLK_ENQUEUE_FLAGS_WAIT_KERNEL 0x1
-#define CLK_ENQUEUE_FLAGS_WAIT_WORK_GROUP 0x2
-
-typedef int kernel_enqueue_flags_t;
-typedef int clk_profiling_info;
-
-// Profiling info name (see capture_event_profiling_info)
-#define CLK_PROFILING_COMMAND_EXEC_TIME 0x1
-
-#define MAX_WORK_DIM 3
-
-typedef struct {
- unsigned int workDimension;
- size_t globalWorkOffset[MAX_WORK_DIM];
- size_t globalWorkSize[MAX_WORK_DIM];
- size_t localWorkSize[MAX_WORK_DIM];
-} ndrange_t;
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
ndrange_t __ovld ndrange_1D(size_t);
ndrange_t __ovld ndrange_1D(size_t, size_t);
ndrange_t __ovld ndrange_3D(const size_t[3], const size_t[3]);
ndrange_t __ovld ndrange_3D(const size_t[3], const size_t[3], const size_t[3]);
-int __ovld enqueue_marker(queue_t, uint, const __private clk_event_t*, __private clk_event_t*);
+int __ovld enqueue_marker(queue_t, uint, const clk_event_t*, clk_event_t*);
void __ovld retain_event(clk_event_t);
void __ovld capture_event_profiling_info(clk_event_t, clk_profiling_info, __global void* value);
queue_t __ovld get_default_queue(void);
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
// OpenCL Extension v2.0 s9.17 - Sub-groups
uint __ovld get_sub_group_size(void);
uint __ovld get_max_sub_group_size(void);
uint __ovld get_num_sub_groups(void);
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
uint __ovld get_enqueued_num_sub_groups(void);
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
uint __ovld get_sub_group_id(void);
uint __ovld get_sub_group_local_id(void);
void __ovld __conv sub_group_barrier(cl_mem_fence_flags flags);
-#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
void __ovld __conv sub_group_barrier(cl_mem_fence_flags flags, memory_scope scope);
-#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0
+#endif //defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
int __ovld __conv sub_group_all(int predicate);
int __ovld __conv sub_group_any(int predicate);
#endif //cl_khr_subgroups cl_intel_subgroups
+#if defined(cl_khr_subgroup_extended_types)
+char __ovld __conv sub_group_broadcast( char value, uint index );
+char2 __ovld __conv sub_group_broadcast( char2 value, uint index );
+char3 __ovld __conv sub_group_broadcast( char3 value, uint index );
+char4 __ovld __conv sub_group_broadcast( char4 value, uint index );
+char8 __ovld __conv sub_group_broadcast( char8 value, uint index );
+char16 __ovld __conv sub_group_broadcast( char16 value, uint index );
+
+uchar __ovld __conv sub_group_broadcast( uchar value, uint index );
+uchar2 __ovld __conv sub_group_broadcast( uchar2 value, uint index );
+uchar3 __ovld __conv sub_group_broadcast( uchar3 value, uint index );
+uchar4 __ovld __conv sub_group_broadcast( uchar4 value, uint index );
+uchar8 __ovld __conv sub_group_broadcast( uchar8 value, uint index );
+uchar16 __ovld __conv sub_group_broadcast( uchar16 value, uint index );
+
+short __ovld __conv sub_group_broadcast( short value, uint index );
+short2 __ovld __conv sub_group_broadcast( short2 value, uint index );
+short3 __ovld __conv sub_group_broadcast( short3 value, uint index );
+short4 __ovld __conv sub_group_broadcast( short4 value, uint index );
+short8 __ovld __conv sub_group_broadcast( short8 value, uint index );
+short16 __ovld __conv sub_group_broadcast( short16 value, uint index );
+
+ushort __ovld __conv sub_group_broadcast( ushort value, uint index );
+ushort2 __ovld __conv sub_group_broadcast( ushort2 value, uint index );
+ushort3 __ovld __conv sub_group_broadcast( ushort3 value, uint index );
+ushort4 __ovld __conv sub_group_broadcast( ushort4 value, uint index );
+ushort8 __ovld __conv sub_group_broadcast( ushort8 value, uint index );
+ushort16 __ovld __conv sub_group_broadcast( ushort16 value, uint index );
+
+// scalar int broadcast is part of cl_khr_subgroups
+int2 __ovld __conv sub_group_broadcast( int2 value, uint index );
+int3 __ovld __conv sub_group_broadcast( int3 value, uint index );
+int4 __ovld __conv sub_group_broadcast( int4 value, uint index );
+int8 __ovld __conv sub_group_broadcast( int8 value, uint index );
+int16 __ovld __conv sub_group_broadcast( int16 value, uint index );
+
+// scalar uint broadcast is part of cl_khr_subgroups
+uint2 __ovld __conv sub_group_broadcast( uint2 value, uint index );
+uint3 __ovld __conv sub_group_broadcast( uint3 value, uint index );
+uint4 __ovld __conv sub_group_broadcast( uint4 value, uint index );
+uint8 __ovld __conv sub_group_broadcast( uint8 value, uint index );
+uint16 __ovld __conv sub_group_broadcast( uint16 value, uint index );
+
+// scalar long broadcast is part of cl_khr_subgroups
+long2 __ovld __conv sub_group_broadcast( long2 value, uint index );
+long3 __ovld __conv sub_group_broadcast( long3 value, uint index );
+long4 __ovld __conv sub_group_broadcast( long4 value, uint index );
+long8 __ovld __conv sub_group_broadcast( long8 value, uint index );
+long16 __ovld __conv sub_group_broadcast( long16 value, uint index );
+
+// scalar ulong broadcast is part of cl_khr_subgroups
+ulong2 __ovld __conv sub_group_broadcast( ulong2 value, uint index );
+ulong3 __ovld __conv sub_group_broadcast( ulong3 value, uint index );
+ulong4 __ovld __conv sub_group_broadcast( ulong4 value, uint index );
+ulong8 __ovld __conv sub_group_broadcast( ulong8 value, uint index );
+ulong16 __ovld __conv sub_group_broadcast( ulong16 value, uint index );
+
+// scalar float broadcast is part of cl_khr_subgroups
+float2 __ovld __conv sub_group_broadcast( float2 value, uint index );
+float3 __ovld __conv sub_group_broadcast( float3 value, uint index );
+float4 __ovld __conv sub_group_broadcast( float4 value, uint index );
+float8 __ovld __conv sub_group_broadcast( float8 value, uint index );
+float16 __ovld __conv sub_group_broadcast( float16 value, uint index );
+
+char __ovld __conv sub_group_reduce_add( char value );
+uchar __ovld __conv sub_group_reduce_add( uchar value );
+short __ovld __conv sub_group_reduce_add( short value );
+ushort __ovld __conv sub_group_reduce_add( ushort value );
+
+char __ovld __conv sub_group_reduce_min( char value );
+uchar __ovld __conv sub_group_reduce_min( uchar value );
+short __ovld __conv sub_group_reduce_min( short value );
+ushort __ovld __conv sub_group_reduce_min( ushort value );
+
+char __ovld __conv sub_group_reduce_max( char value );
+uchar __ovld __conv sub_group_reduce_max( uchar value );
+short __ovld __conv sub_group_reduce_max( short value );
+ushort __ovld __conv sub_group_reduce_max( ushort value );
+
+char __ovld __conv sub_group_scan_inclusive_add( char value );
+uchar __ovld __conv sub_group_scan_inclusive_add( uchar value );
+short __ovld __conv sub_group_scan_inclusive_add( short value );
+ushort __ovld __conv sub_group_scan_inclusive_add( ushort value );
+
+char __ovld __conv sub_group_scan_inclusive_min( char value );
+uchar __ovld __conv sub_group_scan_inclusive_min( uchar value );
+short __ovld __conv sub_group_scan_inclusive_min( short value );
+ushort __ovld __conv sub_group_scan_inclusive_min( ushort value );
+
+char __ovld __conv sub_group_scan_inclusive_max( char value );
+uchar __ovld __conv sub_group_scan_inclusive_max( uchar value );
+short __ovld __conv sub_group_scan_inclusive_max( short value );
+ushort __ovld __conv sub_group_scan_inclusive_max( ushort value );
+
+char __ovld __conv sub_group_scan_exclusive_add( char value );
+uchar __ovld __conv sub_group_scan_exclusive_add( uchar value );
+short __ovld __conv sub_group_scan_exclusive_add( short value );
+ushort __ovld __conv sub_group_scan_exclusive_add( ushort value );
+
+char __ovld __conv sub_group_scan_exclusive_min( char value );
+uchar __ovld __conv sub_group_scan_exclusive_min( uchar value );
+short __ovld __conv sub_group_scan_exclusive_min( short value );
+ushort __ovld __conv sub_group_scan_exclusive_min( ushort value );
+
+char __ovld __conv sub_group_scan_exclusive_max( char value );
+uchar __ovld __conv sub_group_scan_exclusive_max( uchar value );
+short __ovld __conv sub_group_scan_exclusive_max( short value );
+ushort __ovld __conv sub_group_scan_exclusive_max( ushort value );
+
+#if defined(cl_khr_fp16)
+// scalar half broadcast is part of cl_khr_subgroups
+half2 __ovld __conv sub_group_broadcast( half2 value, uint index );
+half3 __ovld __conv sub_group_broadcast( half3 value, uint index );
+half4 __ovld __conv sub_group_broadcast( half4 value, uint index );
+half8 __ovld __conv sub_group_broadcast( half8 value, uint index );
+half16 __ovld __conv sub_group_broadcast( half16 value, uint index );
+#endif // cl_khr_fp16
+
+#if defined(cl_khr_fp64)
+// scalar double broadcast is part of cl_khr_subgroups
+double2 __ovld __conv sub_group_broadcast( double2 value, uint index );
+double3 __ovld __conv sub_group_broadcast( double3 value, uint index );
+double4 __ovld __conv sub_group_broadcast( double4 value, uint index );
+double8 __ovld __conv sub_group_broadcast( double8 value, uint index );
+double16 __ovld __conv sub_group_broadcast( double16 value, uint index );
+#endif // cl_khr_fp64
+
+#endif // cl_khr_subgroup_extended_types
+
+#if defined(cl_khr_subgroup_non_uniform_vote)
+int __ovld sub_group_elect(void);
+int __ovld sub_group_non_uniform_all( int predicate );
+int __ovld sub_group_non_uniform_any( int predicate );
+
+int __ovld sub_group_non_uniform_all_equal( char value );
+int __ovld sub_group_non_uniform_all_equal( uchar value );
+int __ovld sub_group_non_uniform_all_equal( short value );
+int __ovld sub_group_non_uniform_all_equal( ushort value );
+int __ovld sub_group_non_uniform_all_equal( int value );
+int __ovld sub_group_non_uniform_all_equal( uint value );
+int __ovld sub_group_non_uniform_all_equal( long value );
+int __ovld sub_group_non_uniform_all_equal( ulong value );
+int __ovld sub_group_non_uniform_all_equal( float value );
+
+#if defined(cl_khr_fp16)
+int __ovld sub_group_non_uniform_all_equal( half value );
+#endif // cl_khr_fp16
+
+#if defined(cl_khr_fp64)
+int __ovld sub_group_non_uniform_all_equal( double value );
+#endif // cl_khr_fp64
+
+#endif // cl_khr_subgroup_non_uniform_vote
+
+#if defined(cl_khr_subgroup_ballot)
+char __ovld sub_group_non_uniform_broadcast( char value, uint index );
+char2 __ovld sub_group_non_uniform_broadcast( char2 value, uint index );
+char3 __ovld sub_group_non_uniform_broadcast( char3 value, uint index );
+char4 __ovld sub_group_non_uniform_broadcast( char4 value, uint index );
+char8 __ovld sub_group_non_uniform_broadcast( char8 value, uint index );
+char16 __ovld sub_group_non_uniform_broadcast( char16 value, uint index );
+
+uchar __ovld sub_group_non_uniform_broadcast( uchar value, uint index );
+uchar2 __ovld sub_group_non_uniform_broadcast( uchar2 value, uint index );
+uchar3 __ovld sub_group_non_uniform_broadcast( uchar3 value, uint index );
+uchar4 __ovld sub_group_non_uniform_broadcast( uchar4 value, uint index );
+uchar8 __ovld sub_group_non_uniform_broadcast( uchar8 value, uint index );
+uchar16 __ovld sub_group_non_uniform_broadcast( uchar16 value, uint index );
+
+short __ovld sub_group_non_uniform_broadcast( short value, uint index );
+short2 __ovld sub_group_non_uniform_broadcast( short2 value, uint index );
+short3 __ovld sub_group_non_uniform_broadcast( short3 value, uint index );
+short4 __ovld sub_group_non_uniform_broadcast( short4 value, uint index );
+short8 __ovld sub_group_non_uniform_broadcast( short8 value, uint index );
+short16 __ovld sub_group_non_uniform_broadcast( short16 value, uint index );
+
+ushort __ovld sub_group_non_uniform_broadcast( ushort value, uint index );
+ushort2 __ovld sub_group_non_uniform_broadcast( ushort2 value, uint index );
+ushort3 __ovld sub_group_non_uniform_broadcast( ushort3 value, uint index );
+ushort4 __ovld sub_group_non_uniform_broadcast( ushort4 value, uint index );
+ushort8 __ovld sub_group_non_uniform_broadcast( ushort8 value, uint index );
+ushort16 __ovld sub_group_non_uniform_broadcast( ushort16 value, uint index );
+
+int __ovld sub_group_non_uniform_broadcast( int value, uint index );
+int2 __ovld sub_group_non_uniform_broadcast( int2 value, uint index );
+int3 __ovld sub_group_non_uniform_broadcast( int3 value, uint index );
+int4 __ovld sub_group_non_uniform_broadcast( int4 value, uint index );
+int8 __ovld sub_group_non_uniform_broadcast( int8 value, uint index );
+int16 __ovld sub_group_non_uniform_broadcast( int16 value, uint index );
+
+uint __ovld sub_group_non_uniform_broadcast( uint value, uint index );
+uint2 __ovld sub_group_non_uniform_broadcast( uint2 value, uint index );
+uint3 __ovld sub_group_non_uniform_broadcast( uint3 value, uint index );
+uint4 __ovld sub_group_non_uniform_broadcast( uint4 value, uint index );
+uint8 __ovld sub_group_non_uniform_broadcast( uint8 value, uint index );
+uint16 __ovld sub_group_non_uniform_broadcast( uint16 value, uint index );
+
+long __ovld sub_group_non_uniform_broadcast( long value, uint index );
+long2 __ovld sub_group_non_uniform_broadcast( long2 value, uint index );
+long3 __ovld sub_group_non_uniform_broadcast( long3 value, uint index );
+long4 __ovld sub_group_non_uniform_broadcast( long4 value, uint index );
+long8 __ovld sub_group_non_uniform_broadcast( long8 value, uint index );
+long16 __ovld sub_group_non_uniform_broadcast( long16 value, uint index );
+
+ulong __ovld sub_group_non_uniform_broadcast( ulong value, uint index );
+ulong2 __ovld sub_group_non_uniform_broadcast( ulong2 value, uint index );
+ulong3 __ovld sub_group_non_uniform_broadcast( ulong3 value, uint index );
+ulong4 __ovld sub_group_non_uniform_broadcast( ulong4 value, uint index );
+ulong8 __ovld sub_group_non_uniform_broadcast( ulong8 value, uint index );
+ulong16 __ovld sub_group_non_uniform_broadcast( ulong16 value, uint index );
+
+float __ovld sub_group_non_uniform_broadcast( float value, uint index );
+float2 __ovld sub_group_non_uniform_broadcast( float2 value, uint index );
+float3 __ovld sub_group_non_uniform_broadcast( float3 value, uint index );
+float4 __ovld sub_group_non_uniform_broadcast( float4 value, uint index );
+float8 __ovld sub_group_non_uniform_broadcast( float8 value, uint index );
+float16 __ovld sub_group_non_uniform_broadcast( float16 value, uint index );
+
+char __ovld sub_group_broadcast_first( char value );
+uchar __ovld sub_group_broadcast_first( uchar value );
+short __ovld sub_group_broadcast_first( short value );
+ushort __ovld sub_group_broadcast_first( ushort value );
+int __ovld sub_group_broadcast_first( int value );
+uint __ovld sub_group_broadcast_first( uint value );
+long __ovld sub_group_broadcast_first( long value );
+ulong __ovld sub_group_broadcast_first( ulong value );
+float __ovld sub_group_broadcast_first( float value );
+
+uint4 __ovld sub_group_ballot( int predicate );
+int __ovld __cnfn sub_group_inverse_ballot( uint4 value );
+int __ovld __cnfn sub_group_ballot_bit_extract( uint4 value, uint index );
+uint __ovld __cnfn sub_group_ballot_bit_count( uint4 value );
+
+uint __ovld sub_group_ballot_inclusive_scan( uint4 value );
+uint __ovld sub_group_ballot_exclusive_scan( uint4 value );
+uint __ovld sub_group_ballot_find_lsb( uint4 value );
+uint __ovld sub_group_ballot_find_msb( uint4 value );
+
+uint4 __ovld __cnfn get_sub_group_eq_mask(void);
+uint4 __ovld __cnfn get_sub_group_ge_mask(void);
+uint4 __ovld __cnfn get_sub_group_gt_mask(void);
+uint4 __ovld __cnfn get_sub_group_le_mask(void);
+uint4 __ovld __cnfn get_sub_group_lt_mask(void);
+
+#if defined(cl_khr_fp16)
+half __ovld sub_group_non_uniform_broadcast( half value, uint index );
+half2 __ovld sub_group_non_uniform_broadcast( half2 value, uint index );
+half3 __ovld sub_group_non_uniform_broadcast( half3 value, uint index );
+half4 __ovld sub_group_non_uniform_broadcast( half4 value, uint index );
+half8 __ovld sub_group_non_uniform_broadcast( half8 value, uint index );
+half16 __ovld sub_group_non_uniform_broadcast( half16 value, uint index );
+
+half __ovld sub_group_broadcast_first( half value );
+#endif // cl_khr_fp16
+
+#if defined(cl_khr_fp64)
+double __ovld sub_group_non_uniform_broadcast( double value, uint index );
+double2 __ovld sub_group_non_uniform_broadcast( double2 value, uint index );
+double3 __ovld sub_group_non_uniform_broadcast( double3 value, uint index );
+double4 __ovld sub_group_non_uniform_broadcast( double4 value, uint index );
+double8 __ovld sub_group_non_uniform_broadcast( double8 value, uint index );
+double16 __ovld sub_group_non_uniform_broadcast( double16 value, uint index );
+
+double __ovld sub_group_broadcast_first( double value );
+#endif // cl_khr_fp64
+
+#endif // cl_khr_subgroup_ballot
+
+#if defined(cl_khr_subgroup_non_uniform_arithmetic)
+char __ovld sub_group_non_uniform_reduce_add( char value );
+uchar __ovld sub_group_non_uniform_reduce_add( uchar value );
+short __ovld sub_group_non_uniform_reduce_add( short value );
+ushort __ovld sub_group_non_uniform_reduce_add( ushort value );
+int __ovld sub_group_non_uniform_reduce_add( int value );
+uint __ovld sub_group_non_uniform_reduce_add( uint value );
+long __ovld sub_group_non_uniform_reduce_add( long value );
+ulong __ovld sub_group_non_uniform_reduce_add( ulong value );
+float __ovld sub_group_non_uniform_reduce_add( float value );
+
+char __ovld sub_group_non_uniform_reduce_mul( char value );
+uchar __ovld sub_group_non_uniform_reduce_mul( uchar value );
+short __ovld sub_group_non_uniform_reduce_mul( short value );
+ushort __ovld sub_group_non_uniform_reduce_mul( ushort value );
+int __ovld sub_group_non_uniform_reduce_mul( int value );
+uint __ovld sub_group_non_uniform_reduce_mul( uint value );
+long __ovld sub_group_non_uniform_reduce_mul( long value );
+ulong __ovld sub_group_non_uniform_reduce_mul( ulong value );
+float __ovld sub_group_non_uniform_reduce_mul( float value );
+
+char __ovld sub_group_non_uniform_reduce_min( char value );
+uchar __ovld sub_group_non_uniform_reduce_min( uchar value );
+short __ovld sub_group_non_uniform_reduce_min( short value );
+ushort __ovld sub_group_non_uniform_reduce_min( ushort value );
+int __ovld sub_group_non_uniform_reduce_min( int value );
+uint __ovld sub_group_non_uniform_reduce_min( uint value );
+long __ovld sub_group_non_uniform_reduce_min( long value );
+ulong __ovld sub_group_non_uniform_reduce_min( ulong value );
+float __ovld sub_group_non_uniform_reduce_min( float value );
+
+char __ovld sub_group_non_uniform_reduce_max( char value );
+uchar __ovld sub_group_non_uniform_reduce_max( uchar value );
+short __ovld sub_group_non_uniform_reduce_max( short value );
+ushort __ovld sub_group_non_uniform_reduce_max( ushort value );
+int __ovld sub_group_non_uniform_reduce_max( int value );
+uint __ovld sub_group_non_uniform_reduce_max( uint value );
+long __ovld sub_group_non_uniform_reduce_max( long value );
+ulong __ovld sub_group_non_uniform_reduce_max( ulong value );
+float __ovld sub_group_non_uniform_reduce_max( float value );
+
+char __ovld sub_group_non_uniform_scan_inclusive_add( char value );
+uchar __ovld sub_group_non_uniform_scan_inclusive_add( uchar value );
+short __ovld sub_group_non_uniform_scan_inclusive_add( short value );
+ushort __ovld sub_group_non_uniform_scan_inclusive_add( ushort value );
+int __ovld sub_group_non_uniform_scan_inclusive_add( int value );
+uint __ovld sub_group_non_uniform_scan_inclusive_add( uint value );
+long __ovld sub_group_non_uniform_scan_inclusive_add( long value );
+ulong __ovld sub_group_non_uniform_scan_inclusive_add( ulong value );
+float __ovld sub_group_non_uniform_scan_inclusive_add( float value );
+
+char __ovld sub_group_non_uniform_scan_inclusive_mul( char value );
+uchar __ovld sub_group_non_uniform_scan_inclusive_mul( uchar value );
+short __ovld sub_group_non_uniform_scan_inclusive_mul( short value );
+ushort __ovld sub_group_non_uniform_scan_inclusive_mul( ushort value );
+int __ovld sub_group_non_uniform_scan_inclusive_mul( int value );
+uint __ovld sub_group_non_uniform_scan_inclusive_mul( uint value );
+long __ovld sub_group_non_uniform_scan_inclusive_mul( long value );
+ulong __ovld sub_group_non_uniform_scan_inclusive_mul( ulong value );
+float __ovld sub_group_non_uniform_scan_inclusive_mul( float value );
+
+char __ovld sub_group_non_uniform_scan_inclusive_min( char value );
+uchar __ovld sub_group_non_uniform_scan_inclusive_min( uchar value );
+short __ovld sub_group_non_uniform_scan_inclusive_min( short value );
+ushort __ovld sub_group_non_uniform_scan_inclusive_min( ushort value );
+int __ovld sub_group_non_uniform_scan_inclusive_min( int value );
+uint __ovld sub_group_non_uniform_scan_inclusive_min( uint value );
+long __ovld sub_group_non_uniform_scan_inclusive_min( long value );
+ulong __ovld sub_group_non_uniform_scan_inclusive_min( ulong value );
+float __ovld sub_group_non_uniform_scan_inclusive_min( float value );
+
+char __ovld sub_group_non_uniform_scan_inclusive_max( char value );
+uchar __ovld sub_group_non_uniform_scan_inclusive_max( uchar value );
+short __ovld sub_group_non_uniform_scan_inclusive_max( short value );
+ushort __ovld sub_group_non_uniform_scan_inclusive_max( ushort value );
+int __ovld sub_group_non_uniform_scan_inclusive_max( int value );
+uint __ovld sub_group_non_uniform_scan_inclusive_max( uint value );
+long __ovld sub_group_non_uniform_scan_inclusive_max( long value );
+ulong __ovld sub_group_non_uniform_scan_inclusive_max( ulong value );
+float __ovld sub_group_non_uniform_scan_inclusive_max( float value );
+
+char __ovld sub_group_non_uniform_scan_exclusive_add( char value );
+uchar __ovld sub_group_non_uniform_scan_exclusive_add( uchar value );
+short __ovld sub_group_non_uniform_scan_exclusive_add( short value );
+ushort __ovld sub_group_non_uniform_scan_exclusive_add( ushort value );
+int __ovld sub_group_non_uniform_scan_exclusive_add( int value );
+uint __ovld sub_group_non_uniform_scan_exclusive_add( uint value );
+long __ovld sub_group_non_uniform_scan_exclusive_add( long value );
+ulong __ovld sub_group_non_uniform_scan_exclusive_add( ulong value );
+float __ovld sub_group_non_uniform_scan_exclusive_add( float value );
+
+char __ovld sub_group_non_uniform_scan_exclusive_mul( char value );
+uchar __ovld sub_group_non_uniform_scan_exclusive_mul( uchar value );
+short __ovld sub_group_non_uniform_scan_exclusive_mul( short value );
+ushort __ovld sub_group_non_uniform_scan_exclusive_mul( ushort value );
+int __ovld sub_group_non_uniform_scan_exclusive_mul( int value );
+uint __ovld sub_group_non_uniform_scan_exclusive_mul( uint value );
+long __ovld sub_group_non_uniform_scan_exclusive_mul( long value );
+ulong __ovld sub_group_non_uniform_scan_exclusive_mul( ulong value );
+float __ovld sub_group_non_uniform_scan_exclusive_mul( float value );
+
+char __ovld sub_group_non_uniform_scan_exclusive_min( char value );
+uchar __ovld sub_group_non_uniform_scan_exclusive_min( uchar value );
+short __ovld sub_group_non_uniform_scan_exclusive_min( short value );
+ushort __ovld sub_group_non_uniform_scan_exclusive_min( ushort value );
+int __ovld sub_group_non_uniform_scan_exclusive_min( int value );
+uint __ovld sub_group_non_uniform_scan_exclusive_min( uint value );
+long __ovld sub_group_non_uniform_scan_exclusive_min( long value );
+ulong __ovld sub_group_non_uniform_scan_exclusive_min( ulong value );
+float __ovld sub_group_non_uniform_scan_exclusive_min( float value );
+
+char __ovld sub_group_non_uniform_scan_exclusive_max( char value );
+uchar __ovld sub_group_non_uniform_scan_exclusive_max( uchar value );
+short __ovld sub_group_non_uniform_scan_exclusive_max( short value );
+ushort __ovld sub_group_non_uniform_scan_exclusive_max( ushort value );
+int __ovld sub_group_non_uniform_scan_exclusive_max( int value );
+uint __ovld sub_group_non_uniform_scan_exclusive_max( uint value );
+long __ovld sub_group_non_uniform_scan_exclusive_max( long value );
+ulong __ovld sub_group_non_uniform_scan_exclusive_max( ulong value );
+float __ovld sub_group_non_uniform_scan_exclusive_max( float value );
+
+char __ovld sub_group_non_uniform_reduce_and( char value );
+uchar __ovld sub_group_non_uniform_reduce_and( uchar value );
+short __ovld sub_group_non_uniform_reduce_and( short value );
+ushort __ovld sub_group_non_uniform_reduce_and( ushort value );
+int __ovld sub_group_non_uniform_reduce_and( int value );
+uint __ovld sub_group_non_uniform_reduce_and( uint value );
+long __ovld sub_group_non_uniform_reduce_and( long value );
+ulong __ovld sub_group_non_uniform_reduce_and( ulong value );
+
+char __ovld sub_group_non_uniform_reduce_or( char value );
+uchar __ovld sub_group_non_uniform_reduce_or( uchar value );
+short __ovld sub_group_non_uniform_reduce_or( short value );
+ushort __ovld sub_group_non_uniform_reduce_or( ushort value );
+int __ovld sub_group_non_uniform_reduce_or( int value );
+uint __ovld sub_group_non_uniform_reduce_or( uint value );
+long __ovld sub_group_non_uniform_reduce_or( long value );
+ulong __ovld sub_group_non_uniform_reduce_or( ulong value );
+
+char __ovld sub_group_non_uniform_reduce_xor( char value );
+uchar __ovld sub_group_non_uniform_reduce_xor( uchar value );
+short __ovld sub_group_non_uniform_reduce_xor( short value );
+ushort __ovld sub_group_non_uniform_reduce_xor( ushort value );
+int __ovld sub_group_non_uniform_reduce_xor( int value );
+uint __ovld sub_group_non_uniform_reduce_xor( uint value );
+long __ovld sub_group_non_uniform_reduce_xor( long value );
+ulong __ovld sub_group_non_uniform_reduce_xor( ulong value );
+
+char __ovld sub_group_non_uniform_scan_inclusive_and( char value );
+uchar __ovld sub_group_non_uniform_scan_inclusive_and( uchar value );
+short __ovld sub_group_non_uniform_scan_inclusive_and( short value );
+ushort __ovld sub_group_non_uniform_scan_inclusive_and( ushort value );
+int __ovld sub_group_non_uniform_scan_inclusive_and( int value );
+uint __ovld sub_group_non_uniform_scan_inclusive_and( uint value );
+long __ovld sub_group_non_uniform_scan_inclusive_and( long value );
+ulong __ovld sub_group_non_uniform_scan_inclusive_and( ulong value );
+
+char __ovld sub_group_non_uniform_scan_inclusive_or( char value );
+uchar __ovld sub_group_non_uniform_scan_inclusive_or( uchar value );
+short __ovld sub_group_non_uniform_scan_inclusive_or( short value );
+ushort __ovld sub_group_non_uniform_scan_inclusive_or( ushort value );
+int __ovld sub_group_non_uniform_scan_inclusive_or( int value );
+uint __ovld sub_group_non_uniform_scan_inclusive_or( uint value );
+long __ovld sub_group_non_uniform_scan_inclusive_or( long value );
+ulong __ovld sub_group_non_uniform_scan_inclusive_or( ulong value );
+
+char __ovld sub_group_non_uniform_scan_inclusive_xor( char value );
+uchar __ovld sub_group_non_uniform_scan_inclusive_xor( uchar value );
+short __ovld sub_group_non_uniform_scan_inclusive_xor( short value );
+ushort __ovld sub_group_non_uniform_scan_inclusive_xor( ushort value );
+int __ovld sub_group_non_uniform_scan_inclusive_xor( int value );
+uint __ovld sub_group_non_uniform_scan_inclusive_xor( uint value );
+long __ovld sub_group_non_uniform_scan_inclusive_xor( long value );
+ulong __ovld sub_group_non_uniform_scan_inclusive_xor( ulong value );
+
+char __ovld sub_group_non_uniform_scan_exclusive_and( char value );
+uchar __ovld sub_group_non_uniform_scan_exclusive_and( uchar value );
+short __ovld sub_group_non_uniform_scan_exclusive_and( short value );
+ushort __ovld sub_group_non_uniform_scan_exclusive_and( ushort value );
+int __ovld sub_group_non_uniform_scan_exclusive_and( int value );
+uint __ovld sub_group_non_uniform_scan_exclusive_and( uint value );
+long __ovld sub_group_non_uniform_scan_exclusive_and( long value );
+ulong __ovld sub_group_non_uniform_scan_exclusive_and( ulong value );
+
+char __ovld sub_group_non_uniform_scan_exclusive_or( char value );
+uchar __ovld sub_group_non_uniform_scan_exclusive_or( uchar value );
+short __ovld sub_group_non_uniform_scan_exclusive_or( short value );
+ushort __ovld sub_group_non_uniform_scan_exclusive_or( ushort value );
+int __ovld sub_group_non_uniform_scan_exclusive_or( int value );
+uint __ovld sub_group_non_uniform_scan_exclusive_or( uint value );
+long __ovld sub_group_non_uniform_scan_exclusive_or( long value );
+ulong __ovld sub_group_non_uniform_scan_exclusive_or( ulong value );
+
+char __ovld sub_group_non_uniform_scan_exclusive_xor( char value );
+uchar __ovld sub_group_non_uniform_scan_exclusive_xor( uchar value );
+short __ovld sub_group_non_uniform_scan_exclusive_xor( short value );
+ushort __ovld sub_group_non_uniform_scan_exclusive_xor( ushort value );
+int __ovld sub_group_non_uniform_scan_exclusive_xor( int value );
+uint __ovld sub_group_non_uniform_scan_exclusive_xor( uint value );
+long __ovld sub_group_non_uniform_scan_exclusive_xor( long value );
+ulong __ovld sub_group_non_uniform_scan_exclusive_xor( ulong value );
+
+int __ovld sub_group_non_uniform_reduce_logical_and( int predicate );
+int __ovld sub_group_non_uniform_reduce_logical_or( int predicate );
+int __ovld sub_group_non_uniform_reduce_logical_xor( int predicate );
+
+int __ovld sub_group_non_uniform_scan_inclusive_logical_and( int predicate );
+int __ovld sub_group_non_uniform_scan_inclusive_logical_or( int predicate );
+int __ovld sub_group_non_uniform_scan_inclusive_logical_xor( int predicate );
+
+int __ovld sub_group_non_uniform_scan_exclusive_logical_and( int predicate );
+int __ovld sub_group_non_uniform_scan_exclusive_logical_or( int predicate );
+int __ovld sub_group_non_uniform_scan_exclusive_logical_xor( int predicate );
+
+#if defined(cl_khr_fp16)
+half __ovld sub_group_non_uniform_reduce_add( half value );
+half __ovld sub_group_non_uniform_reduce_mul( half value );
+half __ovld sub_group_non_uniform_reduce_min( half value );
+half __ovld sub_group_non_uniform_reduce_max( half value );
+half __ovld sub_group_non_uniform_scan_inclusive_add( half value );
+half __ovld sub_group_non_uniform_scan_inclusive_mul( half value );
+half __ovld sub_group_non_uniform_scan_inclusive_min( half value );
+half __ovld sub_group_non_uniform_scan_inclusive_max( half value );
+half __ovld sub_group_non_uniform_scan_exclusive_add( half value );
+half __ovld sub_group_non_uniform_scan_exclusive_mul( half value );
+half __ovld sub_group_non_uniform_scan_exclusive_min( half value );
+half __ovld sub_group_non_uniform_scan_exclusive_max( half value );
+#endif // cl_khr_fp16
+
+#if defined(cl_khr_fp64)
+double __ovld sub_group_non_uniform_reduce_add( double value );
+double __ovld sub_group_non_uniform_reduce_mul( double value );
+double __ovld sub_group_non_uniform_reduce_min( double value );
+double __ovld sub_group_non_uniform_reduce_max( double value );
+double __ovld sub_group_non_uniform_scan_inclusive_add( double value );
+double __ovld sub_group_non_uniform_scan_inclusive_mul( double value );
+double __ovld sub_group_non_uniform_scan_inclusive_min( double value );
+double __ovld sub_group_non_uniform_scan_inclusive_max( double value );
+double __ovld sub_group_non_uniform_scan_exclusive_add( double value );
+double __ovld sub_group_non_uniform_scan_exclusive_mul( double value );
+double __ovld sub_group_non_uniform_scan_exclusive_min( double value );
+double __ovld sub_group_non_uniform_scan_exclusive_max( double value );
+#endif // cl_khr_fp64
+
+#endif // cl_khr_subgroup_non_uniform_arithmetic
+
+#if defined(cl_khr_subgroup_shuffle)
+char __ovld sub_group_shuffle( char value, uint index );
+uchar __ovld sub_group_shuffle( uchar value, uint index );
+short __ovld sub_group_shuffle( short value, uint index );
+ushort __ovld sub_group_shuffle( ushort value, uint index );
+int __ovld sub_group_shuffle( int value, uint index );
+uint __ovld sub_group_shuffle( uint value, uint index );
+long __ovld sub_group_shuffle( long value, uint index );
+ulong __ovld sub_group_shuffle( ulong value, uint index );
+float __ovld sub_group_shuffle( float value, uint index );
+
+char __ovld sub_group_shuffle_xor( char value, uint mask );
+uchar __ovld sub_group_shuffle_xor( uchar value, uint mask );
+short __ovld sub_group_shuffle_xor( short value, uint mask );
+ushort __ovld sub_group_shuffle_xor( ushort value, uint mask );
+int __ovld sub_group_shuffle_xor( int value, uint mask );
+uint __ovld sub_group_shuffle_xor( uint value, uint mask );
+long __ovld sub_group_shuffle_xor( long value, uint mask );
+ulong __ovld sub_group_shuffle_xor( ulong value, uint mask );
+float __ovld sub_group_shuffle_xor( float value, uint mask );
+
+#if defined(cl_khr_fp16)
+half __ovld sub_group_shuffle( half value, uint index );
+half __ovld sub_group_shuffle_xor( half value, uint mask );
+#endif // cl_khr_fp16
+
+#if defined(cl_khr_fp64)
+double __ovld sub_group_shuffle( double value, uint index );
+double __ovld sub_group_shuffle_xor( double value, uint mask );
+#endif // cl_khr_fp64
+
+#endif // cl_khr_subgroup_shuffle
+
+#if defined(cl_khr_subgroup_shuffle_relative)
+char __ovld sub_group_shuffle_up( char value, uint delta );
+uchar __ovld sub_group_shuffle_up( uchar value, uint delta );
+short __ovld sub_group_shuffle_up( short value, uint delta );
+ushort __ovld sub_group_shuffle_up( ushort value, uint delta );
+int __ovld sub_group_shuffle_up( int value, uint delta );
+uint __ovld sub_group_shuffle_up( uint value, uint delta );
+long __ovld sub_group_shuffle_up( long value, uint delta );
+ulong __ovld sub_group_shuffle_up( ulong value, uint delta );
+float __ovld sub_group_shuffle_up( float value, uint delta );
+
+char __ovld sub_group_shuffle_down( char value, uint delta );
+uchar __ovld sub_group_shuffle_down( uchar value, uint delta );
+short __ovld sub_group_shuffle_down( short value, uint delta );
+ushort __ovld sub_group_shuffle_down( ushort value, uint delta );
+int __ovld sub_group_shuffle_down( int value, uint delta );
+uint __ovld sub_group_shuffle_down( uint value, uint delta );
+long __ovld sub_group_shuffle_down( long value, uint delta );
+ulong __ovld sub_group_shuffle_down( ulong value, uint delta );
+float __ovld sub_group_shuffle_down( float value, uint delta );
+
+#if defined(cl_khr_fp16)
+half __ovld sub_group_shuffle_up( half value, uint delta );
+half __ovld sub_group_shuffle_down( half value, uint delta );
+#endif // cl_khr_fp16
+
+#if defined(cl_khr_fp64)
+double __ovld sub_group_shuffle_up( double value, uint delta );
+double __ovld sub_group_shuffle_down( double value, uint delta );
+#endif // cl_khr_fp64
+
+#endif // cl_khr_subgroup_shuffle_relative
+
+#if defined(cl_khr_subgroup_clustered_reduce)
+char __ovld sub_group_clustered_reduce_add( char value, uint clustersize );
+uchar __ovld sub_group_clustered_reduce_add( uchar value, uint clustersize );
+short __ovld sub_group_clustered_reduce_add( short value, uint clustersize );
+ushort __ovld sub_group_clustered_reduce_add( ushort value, uint clustersize );
+int __ovld sub_group_clustered_reduce_add( int value, uint clustersize );
+uint __ovld sub_group_clustered_reduce_add( uint value, uint clustersize );
+long __ovld sub_group_clustered_reduce_add( long value, uint clustersize );
+ulong __ovld sub_group_clustered_reduce_add( ulong value, uint clustersize );
+float __ovld sub_group_clustered_reduce_add( float value, uint clustersize );
+
+char __ovld sub_group_clustered_reduce_mul( char value, uint clustersize );
+uchar __ovld sub_group_clustered_reduce_mul( uchar value, uint clustersize );
+short __ovld sub_group_clustered_reduce_mul( short value, uint clustersize );
+ushort __ovld sub_group_clustered_reduce_mul( ushort value, uint clustersize );
+int __ovld sub_group_clustered_reduce_mul( int value, uint clustersize );
+uint __ovld sub_group_clustered_reduce_mul( uint value, uint clustersize );
+long __ovld sub_group_clustered_reduce_mul( long value, uint clustersize );
+ulong __ovld sub_group_clustered_reduce_mul( ulong value, uint clustersize );
+float __ovld sub_group_clustered_reduce_mul( float value, uint clustersize );
+
+char __ovld sub_group_clustered_reduce_min( char value, uint clustersize );
+uchar __ovld sub_group_clustered_reduce_min( uchar value, uint clustersize );
+short __ovld sub_group_clustered_reduce_min( short value, uint clustersize );
+ushort __ovld sub_group_clustered_reduce_min( ushort value, uint clustersize );
+int __ovld sub_group_clustered_reduce_min( int value, uint clustersize );
+uint __ovld sub_group_clustered_reduce_min( uint value, uint clustersize );
+long __ovld sub_group_clustered_reduce_min( long value, uint clustersize );
+ulong __ovld sub_group_clustered_reduce_min( ulong value, uint clustersize );
+float __ovld sub_group_clustered_reduce_min( float value, uint clustersize );
+
+char __ovld sub_group_clustered_reduce_max( char value, uint clustersize );
+uchar __ovld sub_group_clustered_reduce_max( uchar value, uint clustersize );
+short __ovld sub_group_clustered_reduce_max( short value, uint clustersize );
+ushort __ovld sub_group_clustered_reduce_max( ushort value, uint clustersize );
+int __ovld sub_group_clustered_reduce_max( int value, uint clustersize );
+uint __ovld sub_group_clustered_reduce_max( uint value, uint clustersize );
+long __ovld sub_group_clustered_reduce_max( long value, uint clustersize );
+ulong __ovld sub_group_clustered_reduce_max( ulong value, uint clustersize );
+float __ovld sub_group_clustered_reduce_max( float value, uint clustersize );
+
+char __ovld sub_group_clustered_reduce_and( char value, uint clustersize );
+uchar __ovld sub_group_clustered_reduce_and( uchar value, uint clustersize );
+short __ovld sub_group_clustered_reduce_and( short value, uint clustersize );
+ushort __ovld sub_group_clustered_reduce_and( ushort value, uint clustersize );
+int __ovld sub_group_clustered_reduce_and( int value, uint clustersize );
+uint __ovld sub_group_clustered_reduce_and( uint value, uint clustersize );
+long __ovld sub_group_clustered_reduce_and( long value, uint clustersize );
+ulong __ovld sub_group_clustered_reduce_and( ulong value, uint clustersize );
+
+char __ovld sub_group_clustered_reduce_or( char value, uint clustersize );
+uchar __ovld sub_group_clustered_reduce_or( uchar value, uint clustersize );
+short __ovld sub_group_clustered_reduce_or( short value, uint clustersize );
+ushort __ovld sub_group_clustered_reduce_or( ushort value, uint clustersize );
+int __ovld sub_group_clustered_reduce_or( int value, uint clustersize );
+uint __ovld sub_group_clustered_reduce_or( uint value, uint clustersize );
+long __ovld sub_group_clustered_reduce_or( long value, uint clustersize );
+ulong __ovld sub_group_clustered_reduce_or( ulong value, uint clustersize );
+
+char __ovld sub_group_clustered_reduce_xor( char value, uint clustersize );
+uchar __ovld sub_group_clustered_reduce_xor( uchar value, uint clustersize );
+short __ovld sub_group_clustered_reduce_xor( short value, uint clustersize );
+ushort __ovld sub_group_clustered_reduce_xor( ushort value, uint clustersize );
+int __ovld sub_group_clustered_reduce_xor( int value, uint clustersize );
+uint __ovld sub_group_clustered_reduce_xor( uint value, uint clustersize );
+long __ovld sub_group_clustered_reduce_xor( long value, uint clustersize );
+ulong __ovld sub_group_clustered_reduce_xor( ulong value, uint clustersize );
+
+int __ovld sub_group_clustered_reduce_logical_and( int predicate, uint clustersize );
+int __ovld sub_group_clustered_reduce_logical_or( int predicate, uint clustersize );
+int __ovld sub_group_clustered_reduce_logical_xor( int predicate, uint clustersize );
+
+#if defined(cl_khr_fp16)
+half __ovld sub_group_clustered_reduce_add( half value, uint clustersize );
+half __ovld sub_group_clustered_reduce_mul( half value, uint clustersize );
+half __ovld sub_group_clustered_reduce_min( half value, uint clustersize );
+half __ovld sub_group_clustered_reduce_max( half value, uint clustersize );
+#endif // cl_khr_fp16
+
+#if defined(cl_khr_fp64)
+double __ovld sub_group_clustered_reduce_add( double value, uint clustersize );
+double __ovld sub_group_clustered_reduce_mul( double value, uint clustersize );
+double __ovld sub_group_clustered_reduce_min( double value, uint clustersize );
+double __ovld sub_group_clustered_reduce_max( double value, uint clustersize );
+#endif // cl_khr_fp64
+
+#endif // cl_khr_subgroup_clustered_reduce
+
+#if defined(cl_intel_subgroups)
+// Intel-Specific Sub Group Functions
+float __ovld __conv intel_sub_group_shuffle( float x, uint c );
+float2 __ovld __conv intel_sub_group_shuffle( float2 x, uint c );
+float3 __ovld __conv intel_sub_group_shuffle( float3 x, uint c );
+float4 __ovld __conv intel_sub_group_shuffle( float4 x, uint c );
+float8 __ovld __conv intel_sub_group_shuffle( float8 x, uint c );
+float16 __ovld __conv intel_sub_group_shuffle( float16 x, uint c );
+
+int __ovld __conv intel_sub_group_shuffle( int x, uint c );
+int2 __ovld __conv intel_sub_group_shuffle( int2 x, uint c );
+int3 __ovld __conv intel_sub_group_shuffle( int3 x, uint c );
+int4 __ovld __conv intel_sub_group_shuffle( int4 x, uint c );
+int8 __ovld __conv intel_sub_group_shuffle( int8 x, uint c );
+int16 __ovld __conv intel_sub_group_shuffle( int16 x, uint c );
+
+uint __ovld __conv intel_sub_group_shuffle( uint x, uint c );
+uint2 __ovld __conv intel_sub_group_shuffle( uint2 x, uint c );
+uint3 __ovld __conv intel_sub_group_shuffle( uint3 x, uint c );
+uint4 __ovld __conv intel_sub_group_shuffle( uint4 x, uint c );
+uint8 __ovld __conv intel_sub_group_shuffle( uint8 x, uint c );
+uint16 __ovld __conv intel_sub_group_shuffle( uint16 x, uint c );
+
+long __ovld __conv intel_sub_group_shuffle( long x, uint c );
+ulong __ovld __conv intel_sub_group_shuffle( ulong x, uint c );
+
+float __ovld __conv intel_sub_group_shuffle_down( float cur, float next, uint c );
+float2 __ovld __conv intel_sub_group_shuffle_down( float2 cur, float2 next, uint c );
+float3 __ovld __conv intel_sub_group_shuffle_down( float3 cur, float3 next, uint c );
+float4 __ovld __conv intel_sub_group_shuffle_down( float4 cur, float4 next, uint c );
+float8 __ovld __conv intel_sub_group_shuffle_down( float8 cur, float8 next, uint c );
+float16 __ovld __conv intel_sub_group_shuffle_down( float16 cur, float16 next, uint c );
+
+int __ovld __conv intel_sub_group_shuffle_down( int cur, int next, uint c );
+int2 __ovld __conv intel_sub_group_shuffle_down( int2 cur, int2 next, uint c );
+int3 __ovld __conv intel_sub_group_shuffle_down( int3 cur, int3 next, uint c );
+int4 __ovld __conv intel_sub_group_shuffle_down( int4 cur, int4 next, uint c );
+int8 __ovld __conv intel_sub_group_shuffle_down( int8 cur, int8 next, uint c );
+int16 __ovld __conv intel_sub_group_shuffle_down( int16 cur, int16 next, uint c );
+
+uint __ovld __conv intel_sub_group_shuffle_down( uint cur, uint next, uint c );
+uint2 __ovld __conv intel_sub_group_shuffle_down( uint2 cur, uint2 next, uint c );
+uint3 __ovld __conv intel_sub_group_shuffle_down( uint3 cur, uint3 next, uint c );
+uint4 __ovld __conv intel_sub_group_shuffle_down( uint4 cur, uint4 next, uint c );
+uint8 __ovld __conv intel_sub_group_shuffle_down( uint8 cur, uint8 next, uint c );
+uint16 __ovld __conv intel_sub_group_shuffle_down( uint16 cur, uint16 next, uint c );
+
+long __ovld __conv intel_sub_group_shuffle_down( long prev, long cur, uint c );
+ulong __ovld __conv intel_sub_group_shuffle_down( ulong prev, ulong cur, uint c );
+
+float __ovld __conv intel_sub_group_shuffle_up( float prev, float cur, uint c );
+float2 __ovld __conv intel_sub_group_shuffle_up( float2 prev, float2 cur, uint c );
+float3 __ovld __conv intel_sub_group_shuffle_up( float3 prev, float3 cur, uint c );
+float4 __ovld __conv intel_sub_group_shuffle_up( float4 prev, float4 cur, uint c );
+float8 __ovld __conv intel_sub_group_shuffle_up( float8 prev, float8 cur, uint c );
+float16 __ovld __conv intel_sub_group_shuffle_up( float16 prev, float16 cur, uint c );
+
+int __ovld __conv intel_sub_group_shuffle_up( int prev, int cur, uint c );
+int2 __ovld __conv intel_sub_group_shuffle_up( int2 prev, int2 cur, uint c );
+int3 __ovld __conv intel_sub_group_shuffle_up( int3 prev, int3 cur, uint c );
+int4 __ovld __conv intel_sub_group_shuffle_up( int4 prev, int4 cur, uint c );
+int8 __ovld __conv intel_sub_group_shuffle_up( int8 prev, int8 cur, uint c );
+int16 __ovld __conv intel_sub_group_shuffle_up( int16 prev, int16 cur, uint c );
+
+uint __ovld __conv intel_sub_group_shuffle_up( uint prev, uint cur, uint c );
+uint2 __ovld __conv intel_sub_group_shuffle_up( uint2 prev, uint2 cur, uint c );
+uint3 __ovld __conv intel_sub_group_shuffle_up( uint3 prev, uint3 cur, uint c );
+uint4 __ovld __conv intel_sub_group_shuffle_up( uint4 prev, uint4 cur, uint c );
+uint8 __ovld __conv intel_sub_group_shuffle_up( uint8 prev, uint8 cur, uint c );
+uint16 __ovld __conv intel_sub_group_shuffle_up( uint16 prev, uint16 cur, uint c );
+
+long __ovld __conv intel_sub_group_shuffle_up( long prev, long cur, uint c );
+ulong __ovld __conv intel_sub_group_shuffle_up( ulong prev, ulong cur, uint c );
+
+float __ovld __conv intel_sub_group_shuffle_xor( float x, uint c );
+float2 __ovld __conv intel_sub_group_shuffle_xor( float2 x, uint c );
+float3 __ovld __conv intel_sub_group_shuffle_xor( float3 x, uint c );
+float4 __ovld __conv intel_sub_group_shuffle_xor( float4 x, uint c );
+float8 __ovld __conv intel_sub_group_shuffle_xor( float8 x, uint c );
+float16 __ovld __conv intel_sub_group_shuffle_xor( float16 x, uint c );
+
+int __ovld __conv intel_sub_group_shuffle_xor( int x, uint c );
+int2 __ovld __conv intel_sub_group_shuffle_xor( int2 x, uint c );
+int3 __ovld __conv intel_sub_group_shuffle_xor( int3 x, uint c );
+int4 __ovld __conv intel_sub_group_shuffle_xor( int4 x, uint c );
+int8 __ovld __conv intel_sub_group_shuffle_xor( int8 x, uint c );
+int16 __ovld __conv intel_sub_group_shuffle_xor( int16 x, uint c );
+
+uint __ovld __conv intel_sub_group_shuffle_xor( uint x, uint c );
+uint2 __ovld __conv intel_sub_group_shuffle_xor( uint2 x, uint c );
+uint3 __ovld __conv intel_sub_group_shuffle_xor( uint3 x, uint c );
+uint4 __ovld __conv intel_sub_group_shuffle_xor( uint4 x, uint c );
+uint8 __ovld __conv intel_sub_group_shuffle_xor( uint8 x, uint c );
+uint16 __ovld __conv intel_sub_group_shuffle_xor( uint16 x, uint c );
+
+long __ovld __conv intel_sub_group_shuffle_xor( long x, uint c );
+ulong __ovld __conv intel_sub_group_shuffle_xor( ulong x, uint c );
+
+uint __ovld __conv intel_sub_group_block_read( read_only image2d_t image, int2 coord );
+uint2 __ovld __conv intel_sub_group_block_read2( read_only image2d_t image, int2 coord );
+uint4 __ovld __conv intel_sub_group_block_read4( read_only image2d_t image, int2 coord );
+uint8 __ovld __conv intel_sub_group_block_read8( read_only image2d_t image, int2 coord );
+
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+uint __ovld __conv intel_sub_group_block_read(read_write image2d_t image, int2 coord);
+uint2 __ovld __conv intel_sub_group_block_read2(read_write image2d_t image, int2 coord);
+uint4 __ovld __conv intel_sub_group_block_read4(read_write image2d_t image, int2 coord);
+uint8 __ovld __conv intel_sub_group_block_read8(read_write image2d_t image, int2 coord);
+#endif // defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+
+uint __ovld __conv intel_sub_group_block_read( const __global uint* p );
+uint2 __ovld __conv intel_sub_group_block_read2( const __global uint* p );
+uint4 __ovld __conv intel_sub_group_block_read4( const __global uint* p );
+uint8 __ovld __conv intel_sub_group_block_read8( const __global uint* p );
+
+void __ovld __conv intel_sub_group_block_write(write_only image2d_t image, int2 coord, uint data);
+void __ovld __conv intel_sub_group_block_write2(write_only image2d_t image, int2 coord, uint2 data);
+void __ovld __conv intel_sub_group_block_write4(write_only image2d_t image, int2 coord, uint4 data);
+void __ovld __conv intel_sub_group_block_write8(write_only image2d_t image, int2 coord, uint8 data);
+
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+void __ovld __conv intel_sub_group_block_write(read_write image2d_t image, int2 coord, uint data);
+void __ovld __conv intel_sub_group_block_write2(read_write image2d_t image, int2 coord, uint2 data);
+void __ovld __conv intel_sub_group_block_write4(read_write image2d_t image, int2 coord, uint4 data);
+void __ovld __conv intel_sub_group_block_write8(read_write image2d_t image, int2 coord, uint8 data);
+#endif // defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+
+void __ovld __conv intel_sub_group_block_write( __global uint* p, uint data );
+void __ovld __conv intel_sub_group_block_write2( __global uint* p, uint2 data );
+void __ovld __conv intel_sub_group_block_write4( __global uint* p, uint4 data );
+void __ovld __conv intel_sub_group_block_write8( __global uint* p, uint8 data );
+
+#ifdef cl_khr_fp16
+half __ovld __conv intel_sub_group_shuffle( half x, uint c );
+half __ovld __conv intel_sub_group_shuffle_down( half prev, half cur, uint c );
+half __ovld __conv intel_sub_group_shuffle_up( half prev, half cur, uint c );
+half __ovld __conv intel_sub_group_shuffle_xor( half x, uint c );
+#endif
+
+#if defined(cl_khr_fp64)
+double __ovld __conv intel_sub_group_shuffle( double x, uint c );
+double __ovld __conv intel_sub_group_shuffle_down( double prev, double cur, uint c );
+double __ovld __conv intel_sub_group_shuffle_up( double prev, double cur, uint c );
+double __ovld __conv intel_sub_group_shuffle_xor( double x, uint c );
+#endif
+
+#endif //cl_intel_subgroups
+
+#if defined(cl_intel_subgroups_short)
+short __ovld __conv intel_sub_group_broadcast( short x, uint sub_group_local_id );
+short2 __ovld __conv intel_sub_group_broadcast( short2 x, uint sub_group_local_id );
+short3 __ovld __conv intel_sub_group_broadcast( short3 x, uint sub_group_local_id );
+short4 __ovld __conv intel_sub_group_broadcast( short4 x, uint sub_group_local_id );
+short8 __ovld __conv intel_sub_group_broadcast( short8 x, uint sub_group_local_id );
+
+ushort __ovld __conv intel_sub_group_broadcast( ushort x, uint sub_group_local_id );
+ushort2 __ovld __conv intel_sub_group_broadcast( ushort2 x, uint sub_group_local_id );
+ushort3 __ovld __conv intel_sub_group_broadcast( ushort3 x, uint sub_group_local_id );
+ushort4 __ovld __conv intel_sub_group_broadcast( ushort4 x, uint sub_group_local_id );
+ushort8 __ovld __conv intel_sub_group_broadcast( ushort8 x, uint sub_group_local_id );
+
+short __ovld __conv intel_sub_group_shuffle( short x, uint c );
+short2 __ovld __conv intel_sub_group_shuffle( short2 x, uint c );
+short3 __ovld __conv intel_sub_group_shuffle( short3 x, uint c );
+short4 __ovld __conv intel_sub_group_shuffle( short4 x, uint c );
+short8 __ovld __conv intel_sub_group_shuffle( short8 x, uint c );
+short16 __ovld __conv intel_sub_group_shuffle( short16 x, uint c);
+
+ushort __ovld __conv intel_sub_group_shuffle( ushort x, uint c );
+ushort2 __ovld __conv intel_sub_group_shuffle( ushort2 x, uint c );
+ushort3 __ovld __conv intel_sub_group_shuffle( ushort3 x, uint c );
+ushort4 __ovld __conv intel_sub_group_shuffle( ushort4 x, uint c );
+ushort8 __ovld __conv intel_sub_group_shuffle( ushort8 x, uint c );
+ushort16 __ovld __conv intel_sub_group_shuffle( ushort16 x, uint c );
+
+short __ovld __conv intel_sub_group_shuffle_down( short cur, short next, uint c );
+short2 __ovld __conv intel_sub_group_shuffle_down( short2 cur, short2 next, uint c );
+short3 __ovld __conv intel_sub_group_shuffle_down( short3 cur, short3 next, uint c );
+short4 __ovld __conv intel_sub_group_shuffle_down( short4 cur, short4 next, uint c );
+short8 __ovld __conv intel_sub_group_shuffle_down( short8 cur, short8 next, uint c );
+short16 __ovld __conv intel_sub_group_shuffle_down( short16 cur, short16 next, uint c );
+
+ushort __ovld __conv intel_sub_group_shuffle_down( ushort cur, ushort next, uint c );
+ushort2 __ovld __conv intel_sub_group_shuffle_down( ushort2 cur, ushort2 next, uint c );
+ushort3 __ovld __conv intel_sub_group_shuffle_down( ushort3 cur, ushort3 next, uint c );
+ushort4 __ovld __conv intel_sub_group_shuffle_down( ushort4 cur, ushort4 next, uint c );
+ushort8 __ovld __conv intel_sub_group_shuffle_down( ushort8 cur, ushort8 next, uint c );
+ushort16 __ovld __conv intel_sub_group_shuffle_down( ushort16 cur, ushort16 next, uint c );
+
+short __ovld __conv intel_sub_group_shuffle_up( short cur, short next, uint c );
+short2 __ovld __conv intel_sub_group_shuffle_up( short2 cur, short2 next, uint c );
+short3 __ovld __conv intel_sub_group_shuffle_up( short3 cur, short3 next, uint c );
+short4 __ovld __conv intel_sub_group_shuffle_up( short4 cur, short4 next, uint c );
+short8 __ovld __conv intel_sub_group_shuffle_up( short8 cur, short8 next, uint c );
+short16 __ovld __conv intel_sub_group_shuffle_up( short16 cur, short16 next, uint c );
+
+ushort __ovld __conv intel_sub_group_shuffle_up( ushort cur, ushort next, uint c );
+ushort2 __ovld __conv intel_sub_group_shuffle_up( ushort2 cur, ushort2 next, uint c );
+ushort3 __ovld __conv intel_sub_group_shuffle_up( ushort3 cur, ushort3 next, uint c );
+ushort4 __ovld __conv intel_sub_group_shuffle_up( ushort4 cur, ushort4 next, uint c );
+ushort8 __ovld __conv intel_sub_group_shuffle_up( ushort8 cur, ushort8 next, uint c );
+ushort16 __ovld __conv intel_sub_group_shuffle_up( ushort16 cur, ushort16 next, uint c );
+
+short __ovld __conv intel_sub_group_shuffle_xor( short x, uint c );
+short2 __ovld __conv intel_sub_group_shuffle_xor( short2 x, uint c );
+short3 __ovld __conv intel_sub_group_shuffle_xor( short3 x, uint c );
+short4 __ovld __conv intel_sub_group_shuffle_xor( short4 x, uint c );
+short8 __ovld __conv intel_sub_group_shuffle_xor( short8 x, uint c );
+short16 __ovld __conv intel_sub_group_shuffle_xor( short16 x, uint c );
+
+ushort __ovld __conv intel_sub_group_shuffle_xor( ushort x, uint c );
+ushort2 __ovld __conv intel_sub_group_shuffle_xor( ushort2 x, uint c );
+ushort3 __ovld __conv intel_sub_group_shuffle_xor( ushort3 x, uint c );
+ushort4 __ovld __conv intel_sub_group_shuffle_xor( ushort4 x, uint c );
+ushort8 __ovld __conv intel_sub_group_shuffle_xor( ushort8 x, uint c );
+ushort16 __ovld __conv intel_sub_group_shuffle_xor( ushort16 x, uint c );
+
+short __ovld __conv intel_sub_group_reduce_add( short x );
+ushort __ovld __conv intel_sub_group_reduce_add( ushort x );
+short __ovld __conv intel_sub_group_reduce_min( short x );
+ushort __ovld __conv intel_sub_group_reduce_min( ushort x );
+short __ovld __conv intel_sub_group_reduce_max( short x );
+ushort __ovld __conv intel_sub_group_reduce_max( ushort x );
+
+short __ovld __conv intel_sub_group_scan_exclusive_add( short x );
+ushort __ovld __conv intel_sub_group_scan_exclusive_add( ushort x );
+short __ovld __conv intel_sub_group_scan_exclusive_min( short x );
+ushort __ovld __conv intel_sub_group_scan_exclusive_min( ushort x );
+short __ovld __conv intel_sub_group_scan_exclusive_max( short x );
+ushort __ovld __conv intel_sub_group_scan_exclusive_max( ushort x );
+
+short __ovld __conv intel_sub_group_scan_inclusive_add( short x );
+ushort __ovld __conv intel_sub_group_scan_inclusive_add( ushort x );
+short __ovld __conv intel_sub_group_scan_inclusive_min( short x );
+ushort __ovld __conv intel_sub_group_scan_inclusive_min( ushort x );
+short __ovld __conv intel_sub_group_scan_inclusive_max( short x );
+ushort __ovld __conv intel_sub_group_scan_inclusive_max( ushort x );
+
+uint __ovld __conv intel_sub_group_block_read_ui( read_only image2d_t image, int2 byte_coord );
+uint2 __ovld __conv intel_sub_group_block_read_ui2( read_only image2d_t image, int2 byte_coord );
+uint4 __ovld __conv intel_sub_group_block_read_ui4( read_only image2d_t image, int2 byte_coord );
+uint8 __ovld __conv intel_sub_group_block_read_ui8( read_only image2d_t image, int2 byte_coord );
+
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+uint __ovld __conv intel_sub_group_block_read_ui( read_write image2d_t image, int2 byte_coord );
+uint2 __ovld __conv intel_sub_group_block_read_ui2( read_write image2d_t image, int2 byte_coord );
+uint4 __ovld __conv intel_sub_group_block_read_ui4( read_write image2d_t image, int2 byte_coord );
+uint8 __ovld __conv intel_sub_group_block_read_ui8( read_write image2d_t image, int2 byte_coord );
+#endif // defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+
+uint __ovld __conv intel_sub_group_block_read_ui( const __global uint* p );
+uint2 __ovld __conv intel_sub_group_block_read_ui2( const __global uint* p );
+uint4 __ovld __conv intel_sub_group_block_read_ui4( const __global uint* p );
+uint8 __ovld __conv intel_sub_group_block_read_ui8( const __global uint* p );
+
+void __ovld __conv intel_sub_group_block_write_ui( read_only image2d_t image, int2 byte_coord, uint data );
+void __ovld __conv intel_sub_group_block_write_ui2( read_only image2d_t image, int2 byte_coord, uint2 data );
+void __ovld __conv intel_sub_group_block_write_ui4( read_only image2d_t image, int2 byte_coord, uint4 data );
+void __ovld __conv intel_sub_group_block_write_ui8( read_only image2d_t image, int2 byte_coord, uint8 data );
+
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+void __ovld __conv intel_sub_group_block_write_ui( read_write image2d_t image, int2 byte_coord, uint data );
+void __ovld __conv intel_sub_group_block_write_ui2( read_write image2d_t image, int2 byte_coord, uint2 data );
+void __ovld __conv intel_sub_group_block_write_ui4( read_write image2d_t image, int2 byte_coord, uint4 data );
+void __ovld __conv intel_sub_group_block_write_ui8( read_write image2d_t image, int2 byte_coord, uint8 data );
+#endif // defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+
+void __ovld __conv intel_sub_group_block_write_ui( __global uint* p, uint data );
+void __ovld __conv intel_sub_group_block_write_ui2( __global uint* p, uint2 data );
+void __ovld __conv intel_sub_group_block_write_ui4( __global uint* p, uint4 data );
+void __ovld __conv intel_sub_group_block_write_ui8( __global uint* p, uint8 data );
+
+ushort __ovld __conv intel_sub_group_block_read_us( read_only image2d_t image, int2 coord );
+ushort2 __ovld __conv intel_sub_group_block_read_us2( read_only image2d_t image, int2 coord );
+ushort4 __ovld __conv intel_sub_group_block_read_us4( read_only image2d_t image, int2 coord );
+ushort8 __ovld __conv intel_sub_group_block_read_us8( read_only image2d_t image, int2 coord );
+
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+ushort __ovld __conv intel_sub_group_block_read_us(read_write image2d_t image, int2 coord);
+ushort2 __ovld __conv intel_sub_group_block_read_us2(read_write image2d_t image, int2 coord);
+ushort4 __ovld __conv intel_sub_group_block_read_us4(read_write image2d_t image, int2 coord);
+ushort8 __ovld __conv intel_sub_group_block_read_us8(read_write image2d_t image, int2 coord);
+#endif // defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+
+ushort __ovld __conv intel_sub_group_block_read_us( const __global ushort* p );
+ushort2 __ovld __conv intel_sub_group_block_read_us2( const __global ushort* p );
+ushort4 __ovld __conv intel_sub_group_block_read_us4( const __global ushort* p );
+ushort8 __ovld __conv intel_sub_group_block_read_us8( const __global ushort* p );
+
+void __ovld __conv intel_sub_group_block_write_us(write_only image2d_t image, int2 coord, ushort data);
+void __ovld __conv intel_sub_group_block_write_us2(write_only image2d_t image, int2 coord, ushort2 data);
+void __ovld __conv intel_sub_group_block_write_us4(write_only image2d_t image, int2 coord, ushort4 data);
+void __ovld __conv intel_sub_group_block_write_us8(write_only image2d_t image, int2 coord, ushort8 data);
+
+#if defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+void __ovld __conv intel_sub_group_block_write_us(read_write image2d_t image, int2 coord, ushort data);
+void __ovld __conv intel_sub_group_block_write_us2(read_write image2d_t image, int2 coord, ushort2 data);
+void __ovld __conv intel_sub_group_block_write_us4(read_write image2d_t image, int2 coord, ushort4 data);
+void __ovld __conv intel_sub_group_block_write_us8(read_write image2d_t image, int2 coord, ushort8 data);
+#endif // defined(__OPENCL_CPP_VERSION__) || (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
+
+void __ovld __conv intel_sub_group_block_write_us( __global ushort* p, ushort data );
+void __ovld __conv intel_sub_group_block_write_us2( __global ushort* p, ushort2 data );
+void __ovld __conv intel_sub_group_block_write_us4( __global ushort* p, ushort4 data );
+void __ovld __conv intel_sub_group_block_write_us8( __global ushort* p, ushort8 data );
+#endif // cl_intel_subgroups_short
+
+#ifdef cl_intel_device_side_avc_motion_estimation
+#pragma OPENCL EXTENSION cl_intel_device_side_avc_motion_estimation : begin
+
+// MCE built-in functions
+uchar __ovld
+intel_sub_group_avc_mce_get_default_inter_base_multi_reference_penalty(
+ uchar slice_type, uchar qp);
+ulong __ovld intel_sub_group_avc_mce_get_default_inter_shape_penalty(
+ uchar slice_type, uchar qp);
+uchar __ovld intel_sub_group_avc_mce_get_default_inter_direction_penalty(
+ uchar slice_type, uchar qp);
+uint __ovld intel_sub_group_avc_mce_get_default_intra_luma_shape_penalty(
+ uchar slice_type, uchar qp);
+uint2 __ovld
+intel_sub_group_avc_mce_get_default_inter_motion_vector_cost_table(
+ uchar slice_type, uchar qp);
+uchar __ovld intel_sub_group_avc_mce_get_default_intra_luma_mode_penalty(
+ uchar slice_type, uchar qp);
+
+uint2 __ovld intel_sub_group_avc_mce_get_default_high_penalty_cost_table();
+uint2 __ovld intel_sub_group_avc_mce_get_default_medium_penalty_cost_table();
+uint2 __ovld intel_sub_group_avc_mce_get_default_low_penalty_cost_table();
+uint __ovld intel_sub_group_avc_mce_get_default_non_dc_luma_intra_penalty();
+uchar __ovld
+intel_sub_group_avc_mce_get_default_intra_chroma_mode_base_penalty();
+
+intel_sub_group_avc_mce_payload_t __ovld
+intel_sub_group_avc_mce_set_inter_base_multi_reference_penalty(
+ uchar reference_base_penalty, intel_sub_group_avc_mce_payload_t payload);
+intel_sub_group_avc_mce_payload_t __ovld
+intel_sub_group_avc_mce_set_inter_shape_penalty(
+ ulong packed_shape_penalty, intel_sub_group_avc_mce_payload_t payload);
+intel_sub_group_avc_mce_payload_t __ovld
+intel_sub_group_avc_mce_set_inter_direction_penalty(
+ uchar direction_cost, intel_sub_group_avc_mce_payload_t payload);
+intel_sub_group_avc_mce_payload_t __ovld
+intel_sub_group_avc_mce_set_motion_vector_cost_function(
+ ulong packed_cost_center_delta, uint2 packed_cost_table,
+ uchar cost_precision, intel_sub_group_avc_mce_payload_t payload);
+intel_sub_group_avc_mce_payload_t __ovld
+intel_sub_group_avc_mce_set_ac_only_haar(
+ intel_sub_group_avc_mce_payload_t payload);
+intel_sub_group_avc_mce_payload_t __ovld
+intel_sub_group_avc_mce_set_source_interlaced_field_polarity(
+ uchar src_field_polarity, intel_sub_group_avc_mce_payload_t payload);
+intel_sub_group_avc_mce_payload_t __ovld
+intel_sub_group_avc_mce_set_single_reference_interlaced_field_polarity(
+ uchar ref_field_polarity, intel_sub_group_avc_mce_payload_t payload);
+intel_sub_group_avc_mce_payload_t __ovld
+intel_sub_group_avc_mce_set_dual_reference_interlaced_field_polarities(
+ uchar fwd_ref_field_polarity, uchar bwd_ref_field_polarity,
+ intel_sub_group_avc_mce_payload_t payload);
+
+ulong __ovld intel_sub_group_avc_mce_get_motion_vectors(
+ intel_sub_group_avc_mce_result_t result);
+ushort __ovld intel_sub_group_avc_mce_get_inter_distortions(
+ intel_sub_group_avc_mce_result_t result);
+ushort __ovld intel_sub_group_avc_mce_get_best_inter_distortion(
+ intel_sub_group_avc_mce_result_t result);
+uchar __ovld intel_sub_group_avc_mce_get_inter_major_shape(
+ intel_sub_group_avc_mce_result_t result);
+uchar __ovld intel_sub_group_avc_mce_get_inter_minor_shapes(
+ intel_sub_group_avc_mce_result_t result);
+uchar __ovld intel_sub_group_avc_mce_get_inter_directions(
+ intel_sub_group_avc_mce_result_t result);
+uchar __ovld intel_sub_group_avc_mce_get_inter_motion_vector_count(
+ intel_sub_group_avc_mce_result_t result);
+uint __ovld intel_sub_group_avc_mce_get_inter_reference_ids(
+ intel_sub_group_avc_mce_result_t result);
+uchar __ovld
+intel_sub_group_avc_mce_get_inter_reference_interlaced_field_polarities(
+ uint packed_reference_ids, uint packed_reference_parameter_field_polarities,
+ intel_sub_group_avc_mce_result_t result);
+
+// IME built-in functions
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_ime_initialize(
+ ushort2 src_coord, uchar partition_mask, uchar sad_adjustment);
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_ime_set_single_reference(
+ short2 ref_offset, uchar search_window_config,
+ intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_ime_set_dual_reference(
+ short2 fwd_ref_offset, short2 bwd_ref_offset, uchar search_window_config,
+ intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_ime_set_max_motion_vector_count(
+ uchar max_motion_vector_count, intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_ime_set_unidirectional_mix_disable(
+ intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_ime_set_early_search_termination_threshold(
+ uchar threshold, intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_ime_set_weighted_sad(
+ uint packed_sad_weights, intel_sub_group_avc_ime_payload_t payload);
+
+__attribute__((deprecated("If you use the latest Intel driver, please use "
+ "intel_sub_group_avc_ime_ref_window_size instead",
+ "intel_sub_group_avc_ime_ref_window_size")))
+ushort2 __ovld
+intel_sub_group_ime_ref_window_size(uchar search_window_config, char dual_ref);
+ushort2 __ovld intel_sub_group_avc_ime_ref_window_size(
+ uchar search_window_config, char dual_ref);
+short2 __ovld intel_sub_group_avc_ime_adjust_ref_offset(
+ short2 ref_offset, ushort2 src_coord, ushort2 ref_window_size,
+ ushort2 image_size);
+
+intel_sub_group_avc_ime_result_t __ovld
+intel_sub_group_avc_ime_evaluate_with_single_reference(
+ read_only image2d_t src_image, read_only image2d_t ref_image,
+ sampler_t vme_media_sampler, intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ime_result_t __ovld
+intel_sub_group_avc_ime_evaluate_with_dual_reference(
+ read_only image2d_t src_image, read_only image2d_t fwd_ref_image,
+ read_only image2d_t bwd_ref_image, sampler_t vme_media_sampler,
+ intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ime_result_single_reference_streamout_t __ovld
+intel_sub_group_avc_ime_evaluate_with_single_reference_streamout(
+ read_only image2d_t src_image, read_only image2d_t ref_image,
+ sampler_t vme_media_sampler, intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ime_result_dual_reference_streamout_t __ovld
+intel_sub_group_avc_ime_evaluate_with_dual_reference_streamout(
+ read_only image2d_t src_image, read_only image2d_t fwd_ref_image,
+ read_only image2d_t bwd_ref_image, sampler_t vme_media_sampler,
+ intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ime_result_t __ovld
+intel_sub_group_avc_ime_evaluate_with_single_reference_streamin(
+ read_only image2d_t src_image, read_only image2d_t ref_image,
+ sampler_t vme_media_sampler, intel_sub_group_avc_ime_payload_t payload,
+ intel_sub_group_avc_ime_single_reference_streamin_t streamin_components);
+intel_sub_group_avc_ime_result_t __ovld
+intel_sub_group_avc_ime_evaluate_with_dual_reference_streamin(
+ read_only image2d_t src_image, read_only image2d_t fwd_ref_image,
+ read_only image2d_t bwd_ref_image, sampler_t vme_media_sampler,
+ intel_sub_group_avc_ime_payload_t payload,
+ intel_sub_group_avc_ime_dual_reference_streamin_t streamin_components);
+intel_sub_group_avc_ime_result_single_reference_streamout_t __ovld
+intel_sub_group_avc_ime_evaluate_with_single_reference_streaminout(
+ read_only image2d_t src_image, read_only image2d_t ref_image,
+ sampler_t vme_media_sampler, intel_sub_group_avc_ime_payload_t payload,
+ intel_sub_group_avc_ime_single_reference_streamin_t streamin_components);
+intel_sub_group_avc_ime_result_dual_reference_streamout_t __ovld
+intel_sub_group_avc_ime_evaluate_with_dual_reference_streaminout(
+ read_only image2d_t src_image, read_only image2d_t fwd_ref_image,
+ read_only image2d_t bwd_ref_image, sampler_t vme_media_sampler,
+ intel_sub_group_avc_ime_payload_t payload,
+ intel_sub_group_avc_ime_dual_reference_streamin_t streamin_components);
+
+intel_sub_group_avc_ime_single_reference_streamin_t __ovld
+intel_sub_group_avc_ime_get_single_reference_streamin(
+ intel_sub_group_avc_ime_result_single_reference_streamout_t result);
+intel_sub_group_avc_ime_dual_reference_streamin_t __ovld
+intel_sub_group_avc_ime_get_dual_reference_streamin(
+ intel_sub_group_avc_ime_result_dual_reference_streamout_t result);
+intel_sub_group_avc_ime_result_t __ovld
+intel_sub_group_avc_ime_strip_single_reference_streamout(
+ intel_sub_group_avc_ime_result_single_reference_streamout_t result);
+intel_sub_group_avc_ime_result_t __ovld
+intel_sub_group_avc_ime_strip_dual_reference_streamout(
+ intel_sub_group_avc_ime_result_dual_reference_streamout_t result);
+
+uint __ovld intel_sub_group_avc_ime_get_streamout_major_shape_motion_vectors(
+ intel_sub_group_avc_ime_result_single_reference_streamout_t result,
+ uchar major_shape);
+ushort __ovld intel_sub_group_avc_ime_get_streamout_major_shape_distortions(
+ intel_sub_group_avc_ime_result_single_reference_streamout_t result,
+ uchar major_shape);
+uchar __ovld intel_sub_group_avc_ime_get_streamout_major_shape_reference_ids(
+ intel_sub_group_avc_ime_result_single_reference_streamout_t result,
+ uchar major_shape);
+uint __ovld intel_sub_group_avc_ime_get_streamout_major_shape_motion_vectors(
+ intel_sub_group_avc_ime_result_dual_reference_streamout_t result,
+ uchar major_shape, uchar direction);
+ushort __ovld intel_sub_group_avc_ime_get_streamout_major_shape_distortions(
+ intel_sub_group_avc_ime_result_dual_reference_streamout_t result,
+ uchar major_shape, uchar direction);
+uchar __ovld intel_sub_group_avc_ime_get_streamout_major_shape_reference_ids(
+ intel_sub_group_avc_ime_result_dual_reference_streamout_t result,
+ uchar major_shape, uchar direction);
+
+uchar __ovld intel_sub_group_avc_ime_get_border_reached(
+ uchar image_select, intel_sub_group_avc_ime_result_t result);
+uchar __ovld intel_sub_group_avc_ime_get_truncated_search_indication(
+ intel_sub_group_avc_ime_result_t result);
+uchar __ovld
+intel_sub_group_avc_ime_get_unidirectional_early_search_termination(
+ intel_sub_group_avc_ime_result_t result);
+uint __ovld intel_sub_group_avc_ime_get_weighting_pattern_minimum_motion_vector(
+ intel_sub_group_avc_ime_result_t result);
+ushort __ovld intel_sub_group_avc_ime_get_weighting_pattern_minimum_distortion(
+ intel_sub_group_avc_ime_result_t result);
+
+// REF built-in functions
+intel_sub_group_avc_ref_payload_t __ovld
+intel_sub_group_avc_fme_initialize(
+ ushort2 src_coord, ulong motion_vectors, uchar major_shapes,
+ uchar minor_shapes, uchar directions, uchar pixel_resolution,
+ uchar sad_adjustment);
+intel_sub_group_avc_ref_payload_t __ovld
+intel_sub_group_avc_bme_initialize(
+ ushort2 src_coord, ulong motion_vectors, uchar major_shapes,
+ uchar minor_shapes, uchar directions, uchar pixel_resolution,
+ uchar bidirectional_weight, uchar sad_adjustment);
+
+intel_sub_group_avc_ref_payload_t __ovld
+intel_sub_group_avc_ref_set_bidirectional_mix_disable(
+ intel_sub_group_avc_ref_payload_t payload);
+intel_sub_group_avc_ref_payload_t __ovld
+intel_sub_group_avc_ref_set_bilinear_filter_enable(
+ intel_sub_group_avc_ref_payload_t payload);
+
+intel_sub_group_avc_ref_result_t __ovld
+intel_sub_group_avc_ref_evaluate_with_single_reference(
+ read_only image2d_t src_image, read_only image2d_t ref_image,
+ sampler_t vme_media_sampler, intel_sub_group_avc_ref_payload_t payload);
+intel_sub_group_avc_ref_result_t __ovld
+intel_sub_group_avc_ref_evaluate_with_dual_reference(
+ read_only image2d_t src_image, read_only image2d_t fwd_ref_image,
+ read_only image2d_t bwd_ref_image, sampler_t vme_media_sampler,
+ intel_sub_group_avc_ref_payload_t payload);
+intel_sub_group_avc_ref_result_t __ovld
+intel_sub_group_avc_ref_evaluate_with_multi_reference(
+ read_only image2d_t src_image, uint packed_reference_ids,
+ sampler_t vme_media_sampler, intel_sub_group_avc_ref_payload_t payload);
+intel_sub_group_avc_ref_result_t __ovld
+intel_sub_group_avc_ref_evaluate_with_multi_reference(
+ read_only image2d_t src_image, uint packed_reference_ids,
+ uchar packed_reference_field_polarities, sampler_t vme_media_sampler,
+ intel_sub_group_avc_ref_payload_t payload);
+
+// SIC built-in functions
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_initialize(
+ ushort2 src_coord);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_configure_skc(
+ uint skip_block_partition_type, uint skip_motion_vector_mask,
+ ulong motion_vectors, uchar bidirectional_weight, uchar skip_sad_adjustment,
+ intel_sub_group_avc_sic_payload_t payload);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_configure_ipe(
+ uchar luma_intra_partition_mask, uchar intra_neighbour_availabilty,
+ uchar left_edge_luma_pixels, uchar upper_left_corner_luma_pixel,
+ uchar upper_edge_luma_pixels, uchar upper_right_edge_luma_pixels,
+ uchar intra_sad_adjustment, intel_sub_group_avc_sic_payload_t payload);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_configure_ipe(
+ uchar luma_intra_partition_mask, uchar intra_neighbour_availabilty,
+ uchar left_edge_luma_pixels, uchar upper_left_corner_luma_pixel,
+ uchar upper_edge_luma_pixels, uchar upper_right_edge_luma_pixels,
+ ushort left_edge_chroma_pixels, ushort upper_left_corner_chroma_pixel,
+ ushort upper_edge_chroma_pixels, uchar intra_sad_adjustment,
+ intel_sub_group_avc_sic_payload_t payload);
+uint __ovld
+intel_sub_group_avc_sic_get_motion_vector_mask(
+ uint skip_block_partition_type, uchar direction);
+
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_set_intra_luma_shape_penalty(
+ uint packed_shape_cost, intel_sub_group_avc_sic_payload_t payload);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_set_intra_luma_mode_cost_function(
+ uchar luma_mode_penalty, uint luma_packed_neighbor_modes,
+ uint luma_packed_non_dc_penalty, intel_sub_group_avc_sic_payload_t payload);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_set_intra_chroma_mode_cost_function(
+ uchar chroma_mode_penalty, intel_sub_group_avc_sic_payload_t payload);
+
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_set_skc_bilinear_filter_enable(
+ intel_sub_group_avc_sic_payload_t payload);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_set_skc_forward_transform_enable(
+ ulong packed_sad_coefficients, intel_sub_group_avc_sic_payload_t payload);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_set_block_based_raw_skip_sad(
+ uchar block_based_skip_type,
+ intel_sub_group_avc_sic_payload_t payload);
+
+intel_sub_group_avc_sic_result_t __ovld
+intel_sub_group_avc_sic_evaluate_ipe(
+ read_only image2d_t src_image, sampler_t vme_media_sampler,
+ intel_sub_group_avc_sic_payload_t payload);
+intel_sub_group_avc_sic_result_t __ovld
+intel_sub_group_avc_sic_evaluate_with_single_reference(
+ read_only image2d_t src_image, read_only image2d_t ref_image,
+ sampler_t vme_media_sampler, intel_sub_group_avc_sic_payload_t payload);
+intel_sub_group_avc_sic_result_t __ovld
+intel_sub_group_avc_sic_evaluate_with_dual_reference(
+ read_only image2d_t src_image, read_only image2d_t fwd_ref_image,
+ read_only image2d_t bwd_ref_image, sampler_t vme_media_sampler,
+ intel_sub_group_avc_sic_payload_t payload);
+intel_sub_group_avc_sic_result_t __ovld
+intel_sub_group_avc_sic_evaluate_with_multi_reference(
+ read_only image2d_t src_image, uint packed_reference_ids,
+ sampler_t vme_media_sampler, intel_sub_group_avc_sic_payload_t payload);
+intel_sub_group_avc_sic_result_t __ovld
+intel_sub_group_avc_sic_evaluate_with_multi_reference(
+ read_only image2d_t src_image, uint packed_reference_ids,
+ uchar packed_reference_field_polarities, sampler_t vme_media_sampler,
+ intel_sub_group_avc_sic_payload_t payload);
+
+uchar __ovld intel_sub_group_avc_sic_get_ipe_luma_shape(
+ intel_sub_group_avc_sic_result_t result);
+ushort __ovld intel_sub_group_avc_sic_get_best_ipe_luma_distortion(
+ intel_sub_group_avc_sic_result_t result);
+ushort __ovld intel_sub_group_avc_sic_get_best_ipe_chroma_distortion(
+ intel_sub_group_avc_sic_result_t result);
+ulong __ovld intel_sub_group_avc_sic_get_packed_ipe_luma_modes(
+ intel_sub_group_avc_sic_result_t result);
+uchar __ovld intel_sub_group_avc_sic_get_ipe_chroma_mode(
+ intel_sub_group_avc_sic_result_t result);
+uint __ovld intel_sub_group_avc_sic_get_packed_skc_luma_count_threshold(
+ intel_sub_group_avc_sic_result_t result);
+ulong __ovld intel_sub_group_avc_sic_get_packed_skc_luma_sum_threshold(
+ intel_sub_group_avc_sic_result_t result);
+ushort __ovld intel_sub_group_avc_sic_get_inter_raw_sads(
+ intel_sub_group_avc_sic_result_t result);
+
+// Wrappers
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_ime_set_inter_base_multi_reference_penalty(
+ uchar reference_base_penalty, intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ref_payload_t __ovld
+intel_sub_group_avc_ref_set_inter_base_multi_reference_penalty(
+ uchar reference_base_penalty, intel_sub_group_avc_ref_payload_t payload);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_set_inter_base_multi_reference_penalty(
+ uchar reference_base_penalty, intel_sub_group_avc_sic_payload_t payload);
+
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_ime_set_inter_shape_penalty(
+ ulong packed_shape_cost, intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ref_payload_t __ovld
+intel_sub_group_avc_ref_set_inter_shape_penalty(
+ ulong packed_shape_cost, intel_sub_group_avc_ref_payload_t payload);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_set_inter_shape_penalty(
+ ulong packed_shape_cost, intel_sub_group_avc_sic_payload_t payload);
+
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_ime_set_inter_direction_penalty(
+ uchar direction_cost, intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ref_payload_t __ovld
+intel_sub_group_avc_ref_set_inter_direction_penalty(
+ uchar direction_cost, intel_sub_group_avc_ref_payload_t payload);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_set_inter_direction_penalty(
+ uchar direction_cost, intel_sub_group_avc_sic_payload_t payload);
+
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_ime_set_motion_vector_cost_function(
+ ulong packed_cost_center_delta, uint2 packed_cost_table,
+ uchar cost_precision, intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ref_payload_t __ovld
+intel_sub_group_avc_ref_set_motion_vector_cost_function(
+ ulong packed_cost_center_delta, uint2 packed_cost_table,
+ uchar cost_precision, intel_sub_group_avc_ref_payload_t payload);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_set_motion_vector_cost_function(
+ ulong packed_cost_center_delta, uint2 packed_cost_table,
+ uchar cost_precision, intel_sub_group_avc_sic_payload_t payload);
+
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_ime_set_source_interlaced_field_polarity(
+ uchar src_field_polarity, intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ref_payload_t __ovld
+intel_sub_group_avc_ref_set_source_interlaced_field_polarity(
+ uchar src_field_polarity, intel_sub_group_avc_ref_payload_t payload);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_set_source_interlaced_field_polarity(
+ uchar src_field_polarity, intel_sub_group_avc_sic_payload_t payload);
+
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_ime_set_single_reference_interlaced_field_polarity(
+ uchar ref_field_polarity, intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ref_payload_t __ovld
+intel_sub_group_avc_ref_set_single_reference_interlaced_field_polarity(
+ uchar ref_field_polarity, intel_sub_group_avc_ref_payload_t payload);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_set_single_reference_interlaced_field_polarity(
+ uchar ref_field_polarity, intel_sub_group_avc_sic_payload_t payload);
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_ime_set_dual_reference_interlaced_field_polarities(
+ uchar fwd_ref_field_polarity, uchar bwd_ref_field_polarity,
+ intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ref_payload_t __ovld
+intel_sub_group_avc_ref_set_dual_reference_interlaced_field_polarities(
+ uchar fwd_ref_field_polarity, uchar bwd_ref_field_polarity,
+ intel_sub_group_avc_ref_payload_t payload);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_set_dual_reference_interlaced_field_polarities(
+ uchar fwd_ref_field_polarity, uchar bwd_ref_field_polarity,
+ intel_sub_group_avc_sic_payload_t payload);
+
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_ime_set_ac_only_haar(
+ intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ref_payload_t __ovld
+intel_sub_group_avc_ref_set_ac_only_haar(
+ intel_sub_group_avc_ref_payload_t payload);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_sic_set_ac_only_haar(
+ intel_sub_group_avc_sic_payload_t payload);
+
+ulong __ovld intel_sub_group_avc_ime_get_motion_vectors(
+ intel_sub_group_avc_ime_result_t result);
+ulong __ovld intel_sub_group_avc_ref_get_motion_vectors(
+ intel_sub_group_avc_ref_result_t result);
+
+ushort __ovld intel_sub_group_avc_ime_get_inter_distortions(
+ intel_sub_group_avc_ime_result_t result);
+ushort __ovld intel_sub_group_avc_ref_get_inter_distortions(
+ intel_sub_group_avc_ref_result_t result);
+ushort __ovld intel_sub_group_avc_sic_get_inter_distortions(
+ intel_sub_group_avc_sic_result_t result);
+
+ushort __ovld intel_sub_group_avc_ime_get_best_inter_distortion(
+ intel_sub_group_avc_ime_result_t result);
+ushort __ovld intel_sub_group_avc_ref_get_best_inter_distortion(
+ intel_sub_group_avc_ref_result_t result);
+
+uchar __ovld intel_sub_group_avc_ime_get_inter_major_shape(
+ intel_sub_group_avc_ime_result_t result);
+uchar __ovld intel_sub_group_avc_ref_get_inter_major_shape(
+ intel_sub_group_avc_ref_result_t result);
+uchar __ovld intel_sub_group_avc_ime_get_inter_minor_shapes(
+ intel_sub_group_avc_ime_result_t result);
+uchar __ovld intel_sub_group_avc_ref_get_inter_minor_shapes(
+ intel_sub_group_avc_ref_result_t result);
+
+uchar __ovld intel_sub_group_avc_ime_get_inter_directions(
+ intel_sub_group_avc_ime_result_t result);
+uchar __ovld intel_sub_group_avc_ref_get_inter_directions(
+ intel_sub_group_avc_ref_result_t result);
+
+uchar __ovld intel_sub_group_avc_ime_get_inter_motion_vector_count(
+ intel_sub_group_avc_ime_result_t result);
+uchar __ovld intel_sub_group_avc_ref_get_inter_motion_vector_count(
+ intel_sub_group_avc_ref_result_t result);
+
+uint __ovld intel_sub_group_avc_ime_get_inter_reference_ids(
+ intel_sub_group_avc_ime_result_t result);
+uint __ovld intel_sub_group_avc_ref_get_inter_reference_ids(
+ intel_sub_group_avc_ref_result_t result);
+
+uchar __ovld
+intel_sub_group_avc_ime_get_inter_reference_interlaced_field_polarities(
+ uint packed_reference_ids, uint packed_reference_parameter_field_polarities,
+ intel_sub_group_avc_ime_result_t result);
+uchar __ovld
+intel_sub_group_avc_ref_get_inter_reference_interlaced_field_polarities(
+ uint packed_reference_ids, uint packed_reference_parameter_field_polarities,
+ intel_sub_group_avc_ref_result_t result);
+
+// Type conversion functions
+intel_sub_group_avc_mce_payload_t __ovld
+intel_sub_group_avc_ime_convert_to_mce_payload(
+ intel_sub_group_avc_ime_payload_t payload);
+intel_sub_group_avc_ime_payload_t __ovld
+intel_sub_group_avc_mce_convert_to_ime_payload(
+ intel_sub_group_avc_mce_payload_t payload);
+intel_sub_group_avc_mce_payload_t __ovld
+intel_sub_group_avc_ref_convert_to_mce_payload(
+ intel_sub_group_avc_ref_payload_t payload);
+intel_sub_group_avc_ref_payload_t __ovld
+intel_sub_group_avc_mce_convert_to_ref_payload(
+ intel_sub_group_avc_mce_payload_t payload);
+intel_sub_group_avc_mce_payload_t __ovld
+intel_sub_group_avc_sic_convert_to_mce_payload(
+ intel_sub_group_avc_sic_payload_t payload);
+intel_sub_group_avc_sic_payload_t __ovld
+intel_sub_group_avc_mce_convert_to_sic_payload(
+ intel_sub_group_avc_mce_payload_t payload);
+
+intel_sub_group_avc_mce_result_t __ovld
+intel_sub_group_avc_ime_convert_to_mce_result(
+ intel_sub_group_avc_ime_result_t result);
+intel_sub_group_avc_ime_result_t __ovld
+intel_sub_group_avc_mce_convert_to_ime_result(
+ intel_sub_group_avc_mce_result_t result);
+intel_sub_group_avc_mce_result_t __ovld
+intel_sub_group_avc_ref_convert_to_mce_result(
+ intel_sub_group_avc_ref_result_t result);
+intel_sub_group_avc_ref_result_t __ovld
+intel_sub_group_avc_mce_convert_to_ref_result(
+ intel_sub_group_avc_mce_result_t result);
+intel_sub_group_avc_mce_result_t __ovld
+intel_sub_group_avc_sic_convert_to_mce_result(
+ intel_sub_group_avc_sic_result_t result);
+intel_sub_group_avc_sic_result_t __ovld
+intel_sub_group_avc_mce_convert_to_sic_result(
+ intel_sub_group_avc_mce_result_t result);
+#pragma OPENCL EXTENSION cl_intel_device_side_avc_motion_estimation : end
+#endif // cl_intel_device_side_avc_motion_estimation
+
#ifdef cl_amd_media_ops
uint __ovld amd_bitalign(uint a, uint b, uint c);
uint2 __ovld amd_bitalign(uint2 a, uint2 b, uint2 c);
uint16 __ovld amd_sadw(uint16 src0, uint16 src1, uint16 src2);
#endif // cl_amd_media_ops2
+#if defined(cl_arm_integer_dot_product_int8)
+#pragma OPENCL EXTENSION cl_arm_integer_dot_product_int8 : begin
+uint __ovld arm_dot(uchar4 a, uchar4 b);
+int __ovld arm_dot(char4 a, char4 b);
+#pragma OPENCL EXTENSION cl_arm_integer_dot_product_int8 : end
+#endif // defined(cl_arm_integer_dot_product_int8)
+
+#if defined(cl_arm_integer_dot_product_accumulate_int8)
+#pragma OPENCL EXTENSION cl_arm_integer_dot_product_accumulate_int8 : begin
+uint __ovld arm_dot_acc(uchar4 a, uchar4 b, uint c);
+int __ovld arm_dot_acc(char4 a, char4 b, int c);
+#pragma OPENCL EXTENSION cl_arm_integer_dot_product_accumulate_int8 : end
+#endif // defined(cl_arm_integer_dot_product_accumulate_int8)
+
+#if defined(cl_arm_integer_dot_product_accumulate_int16)
+#pragma OPENCL EXTENSION cl_arm_integer_dot_product_accumulate_int16 : begin
+uint __ovld arm_dot_acc(ushort2 a, ushort2 b, uint c);
+int __ovld arm_dot_acc(short2 a, short2 b, int c);
+#pragma OPENCL EXTENSION cl_arm_integer_dot_product_accumulate_int16 : end
+#endif // defined(cl_arm_integer_dot_product_accumulate_int16)
+
+#if defined(cl_arm_integer_dot_product_accumulate_saturate_int8)
+#pragma OPENCL EXTENSION cl_arm_integer_dot_product_accumulate_saturate_int8 : begin
+uint __ovld arm_dot_acc_sat(uchar4 a, uchar4 b, uint c);
+int __ovld arm_dot_acc_sat(char4 a, char4 b, int c);
+#pragma OPENCL EXTENSION cl_arm_integer_dot_product_accumulate_saturate_int8 : end
+#endif // defined(cl_arm_integer_dot_product_accumulate_saturate_int8)
+
// Disable any extensions we may have enabled previously.
#pragma OPENCL EXTENSION all : disable
--- /dev/null
+/*===- __clang_openmp_device_functions.h - OpenMP device function declares -===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __CLANG_OPENMP_DEVICE_FUNCTIONS_H__
+#define __CLANG_OPENMP_DEVICE_FUNCTIONS_H__
+
+#ifndef _OPENMP
+#error "This file is for OpenMP compilation only."
+#endif
+
+#pragma omp begin declare variant match( \
+ device = {arch(nvptx, nvptx64)}, implementation = {extension(match_any)})
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#define __CUDA__
+#define __OPENMP_NVPTX__
+
+/// Include declarations for libdevice functions.
+#include <__clang_cuda_libdevice_declares.h>
+
+/// Provide definitions for these functions.
+#include <__clang_cuda_device_functions.h>
+
+#undef __OPENMP_NVPTX__
+#undef __CUDA__
+
+#ifdef __cplusplus
+} // extern "C"
+#endif
+
+#pragma omp end declare variant
+
+#endif
--- /dev/null
+/*===-- __clang_openmp_device_functions.h - OpenMP math declares ------ c++ -===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __CLANG_OPENMP_CMATH_H__
+#define __CLANG_OPENMP_CMATH_H__
+
+#ifndef _OPENMP
+#error "This file is for OpenMP compilation only."
+#endif
+
+#include_next <cmath>
+
+// Make sure we include our math.h overlay, it probably happend already but we
+// need to be sure.
+#include <math.h>
+
+// We (might) need cstdlib because __clang_cuda_cmath.h below declares `abs`
+// which might live in cstdlib.
+#include <cstdlib>
+
+#pragma omp begin declare variant match( \
+ device = {arch(nvptx, nvptx64)}, implementation = {extension(match_any)})
+
+#define __CUDA__
+#define __OPENMP_NVPTX__
+#include <__clang_cuda_cmath.h>
+#undef __OPENMP_NVPTX__
+#undef __CUDA__
+
+// Overloads not provided by the CUDA wrappers but by the CUDA system headers.
+// Since we do not include the latter we define them ourselves.
+#define __DEVICE__ static constexpr __attribute__((always_inline, nothrow))
+
+__DEVICE__ float acosh(float __x) { return ::acoshf(__x); }
+__DEVICE__ float asinh(float __x) { return ::asinhf(__x); }
+__DEVICE__ float atanh(float __x) { return ::atanhf(__x); }
+__DEVICE__ float cbrt(float __x) { return ::cbrtf(__x); }
+__DEVICE__ float erf(float __x) { return ::erff(__x); }
+__DEVICE__ float erfc(float __x) { return ::erfcf(__x); }
+__DEVICE__ float exp2(float __x) { return ::exp2f(__x); }
+__DEVICE__ float expm1(float __x) { return ::expm1f(__x); }
+__DEVICE__ float fdim(float __x, float __y) { return ::fdimf(__x, __y); }
+__DEVICE__ float hypot(float __x, float __y) { return ::hypotf(__x, __y); }
+__DEVICE__ int ilogb(float __x) { return ::ilogbf(__x); }
+__DEVICE__ float lgamma(float __x) { return ::lgammaf(__x); }
+__DEVICE__ long long int llrint(float __x) { return ::llrintf(__x); }
+__DEVICE__ long long int llround(float __x) { return ::llroundf(__x); }
+__DEVICE__ float log1p(float __x) { return ::log1pf(__x); }
+__DEVICE__ float log2(float __x) { return ::log2f(__x); }
+__DEVICE__ float logb(float __x) { return ::logbf(__x); }
+__DEVICE__ long int lrint(float __x) { return ::lrintf(__x); }
+__DEVICE__ long int lround(float __x) { return ::lroundf(__x); }
+__DEVICE__ float nextafter(float __x, float __y) {
+ return ::nextafterf(__x, __y);
+}
+__DEVICE__ float remainder(float __x, float __y) {
+ return ::remainderf(__x, __y);
+}
+__DEVICE__ float scalbln(float __x, long int __y) {
+ return ::scalblnf(__x, __y);
+}
+__DEVICE__ float scalbn(float __x, int __y) { return ::scalbnf(__x, __y); }
+__DEVICE__ float tgamma(float __x) { return ::tgammaf(__x); }
+
+#undef __DEVICE__
+
+#pragma omp end declare variant
+
+#endif
--- /dev/null
+/*===-- complex --- OpenMP complex wrapper for target regions --------- c++ -===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __CLANG_OPENMP_COMPLEX__
+#define __CLANG_OPENMP_COMPLEX__
+
+#ifndef _OPENMP
+#error "This file is for OpenMP compilation only."
+#endif
+
+// We require std::math functions in the complex builtins below.
+#include <cmath>
+
+#define __CUDA__
+#include <__clang_cuda_complex_builtins.h>
+#endif
+
+// Grab the host header too.
+#include_next <complex>
--- /dev/null
+/*===-- complex --- OpenMP complex wrapper for target regions --------- c++ -===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __CLANG_OPENMP_COMPLEX_H__
+#define __CLANG_OPENMP_COMPLEX_H__
+
+#ifndef _OPENMP
+#error "This file is for OpenMP compilation only."
+#endif
+
+// We require math functions in the complex builtins below.
+#include <math.h>
+
+#define __CUDA__
+#include <__clang_cuda_complex_builtins.h>
+#endif
+
+// Grab the host header too.
+#include_next <complex.h>
--- /dev/null
+/*===---- openmp_wrapper/math.h -------- OpenMP math.h intercept ------ c++ -===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+// If we are in C++ mode and include <math.h> (not <cmath>) first, we still need
+// to make sure <cmath> is read first. The problem otherwise is that we haven't
+// seen the declarations of the math.h functions when the system math.h includes
+// our cmath overlay. However, our cmath overlay, or better the underlying
+// overlay, e.g. CUDA, uses the math.h functions. Since we haven't declared them
+// yet we get errors. CUDA avoids this by eagerly declaring all math functions
+// (in the __device__ space) but we cannot do this. Instead we break the
+// dependence by forcing cmath to go first. While our cmath will in turn include
+// this file, the cmath guards will prevent recursion.
+#ifdef __cplusplus
+#include <cmath>
+#endif
+
+#ifndef __CLANG_OPENMP_MATH_H__
+#define __CLANG_OPENMP_MATH_H__
+
+#ifndef _OPENMP
+#error "This file is for OpenMP compilation only."
+#endif
+
+#include_next <math.h>
+
+// We need limits.h for __clang_cuda_math.h below and because it should not hurt
+// we include it eagerly here.
+#include <limits.h>
+
+// We need stdlib.h because (for now) __clang_cuda_math.h below declares `abs`
+// which should live in stdlib.h.
+#include <stdlib.h>
+
+#pragma omp begin declare variant match( \
+ device = {arch(nvptx, nvptx64)}, implementation = {extension(match_any)})
+
+#define __CUDA__
+#define __OPENMP_NVPTX__
+#include <__clang_cuda_math.h>
+#undef __OPENMP_NVPTX__
+#undef __CUDA__
+
+#pragma omp end declare variant
+
+#endif
--- /dev/null
+//===--------- new - OPENMP wrapper for <new> ------------------------------===
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===-----------------------------------------------------------------------===
+
+#ifndef __CLANG_OPENMP_WRAPPERS_NEW
+#define __CLANG_OPENMP_WRAPPERS_NEW
+
+#include_next <new>
+
+#if defined(__NVPTX__) && defined(_OPENMP)
+
+#include <cstdlib>
+
+#pragma push_macro("OPENMP_NOEXCEPT")
+#if __cplusplus >= 201103L
+#define OPENMP_NOEXCEPT noexcept
+#else
+#define OPENMP_NOEXCEPT
+#endif
+
+// Device overrides for non-placement new and delete.
+inline void *operator new(__SIZE_TYPE__ size) {
+ if (size == 0)
+ size = 1;
+ return ::malloc(size);
+}
+inline void *operator new(__SIZE_TYPE__ size,
+ const std::nothrow_t &) OPENMP_NOEXCEPT {
+ return ::operator new(size);
+}
+
+inline void *operator new[](__SIZE_TYPE__ size) { return ::operator new(size); }
+inline void *operator new[](__SIZE_TYPE__ size, const std::nothrow_t &) {
+ return ::operator new(size);
+}
+
+inline void operator delete(void *ptr)OPENMP_NOEXCEPT {
+ if (ptr)
+ ::free(ptr);
+}
+inline void operator delete(void *ptr, const std::nothrow_t &)OPENMP_NOEXCEPT {
+ ::operator delete(ptr);
+}
+
+inline void operator delete[](void *ptr) OPENMP_NOEXCEPT {
+ ::operator delete(ptr);
+}
+inline void operator delete[](void *ptr,
+ const std::nothrow_t &) OPENMP_NOEXCEPT {
+ ::operator delete(ptr);
+}
+
+// Sized delete, C++14 only.
+#if __cplusplus >= 201402L
+inline void operator delete(void *ptr, __SIZE_TYPE__ size)OPENMP_NOEXCEPT {
+ ::operator delete(ptr);
+}
+inline void operator delete[](void *ptr, __SIZE_TYPE__ size) OPENMP_NOEXCEPT {
+ ::operator delete(ptr);
+}
+#endif
+
+#pragma pop_macro("OPENMP_NOEXCEPT")
+#endif
+
+#endif // include guard
--- /dev/null
+/*===---- pconfigintrin.h - X86 platform configuration ---------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#if !defined __X86INTRIN_H && !defined __IMMINTRIN_H
+#error "Never use <pconfigintrin.h> directly; include <x86intrin.h> instead."
+#endif
+
+#ifndef __PCONFIGINTRIN_H
+#define __PCONFIGINTRIN_H
+
+#define __PCONFIG_KEY_PROGRAM 0x00000001
+
+#if __has_extension(gnu_asm)
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("pconfig")))
+
+static __inline unsigned int __DEFAULT_FN_ATTRS
+_pconfig_u32(unsigned int __leaf, __SIZE_TYPE__ __d[])
+{
+ unsigned int __result;
+ __asm__ ("pconfig"
+ : "=a" (__result), "=b" (__d[0]), "=c" (__d[1]), "=d" (__d[2])
+ : "a" (__leaf), "b" (__d[0]), "c" (__d[1]), "d" (__d[2])
+ : "cc");
+ return __result;
+}
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif /* __has_extension(gnu_asm) */
+
+#endif
--- /dev/null
+/*===---- pkuintrin.h - PKU intrinsics -------------------------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <pkuintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __PKUINTRIN_H
+#define __PKUINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("pku")))
+
+static __inline__ unsigned int __DEFAULT_FN_ATTRS
+_rdpkru_u32(void)
+{
+ return __builtin_ia32_rdpkru();
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_wrpkru(unsigned int __val)
+{
+ __builtin_ia32_wrpkru(__val);
+}
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif
/*===---- pmmintrin.h - SSE3 intrinsics ------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
/* Define the default attributes for the functions in this file. */
#define __DEFAULT_FN_ATTRS \
- __attribute__((__always_inline__, __nodebug__, __target__("sse3")))
+ __attribute__((__always_inline__, __nodebug__, __target__("sse3"), __min_vector_width__(128)))
-/// \brief Loads data from an unaligned memory location to elements in a 128-bit
+/// Loads data from an unaligned memory location to elements in a 128-bit
/// vector.
///
/// If the address of the data is not 16-byte aligned, the instruction may
return (__m128i)__builtin_ia32_lddqu((char const *)__p);
}
-/// \brief Adds the even-indexed values and subtracts the odd-indexed values of
+/// Adds the even-indexed values and subtracts the odd-indexed values of
/// two 128-bit vectors of [4 x float].
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_addsubps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Horizontally adds the adjacent pairs of values contained in two
+/// Horizontally adds the adjacent pairs of values contained in two
/// 128-bit vectors of [4 x float].
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_haddps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Horizontally subtracts the adjacent pairs of values contained in two
+/// Horizontally subtracts the adjacent pairs of values contained in two
/// 128-bit vectors of [4 x float].
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_hsubps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Moves and duplicates high-order (odd-indexed) values from a 128-bit
-/// vector of [4 x float] to float values stored in a 128-bit vector of
+/// Moves and duplicates odd-indexed values from a 128-bit vector
+/// of [4 x float] to float values stored in a 128-bit vector of
/// [4 x float].
///
/// \headerfile <x86intrin.h>
return __builtin_shufflevector((__v4sf)__a, (__v4sf)__a, 1, 1, 3, 3);
}
-/// \brief Duplicates low-order (even-indexed) values from a 128-bit vector of
+/// Duplicates even-indexed values from a 128-bit vector of
/// [4 x float] to float values stored in a 128-bit vector of [4 x float].
///
/// \headerfile <x86intrin.h>
return __builtin_shufflevector((__v4sf)__a, (__v4sf)__a, 0, 0, 2, 2);
}
-/// \brief Adds the even-indexed values and subtracts the odd-indexed values of
+/// Adds the even-indexed values and subtracts the odd-indexed values of
/// two 128-bit vectors of [2 x double].
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_addsubpd((__v2df)__a, (__v2df)__b);
}
-/// \brief Horizontally adds the pairs of values contained in two 128-bit
+/// Horizontally adds the pairs of values contained in two 128-bit
/// vectors of [2 x double].
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_haddpd((__v2df)__a, (__v2df)__b);
}
-/// \brief Horizontally subtracts the pairs of values contained in two 128-bit
+/// Horizontally subtracts the pairs of values contained in two 128-bit
/// vectors of [2 x double].
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_hsubpd((__v2df)__a, (__v2df)__b);
}
-/// \brief Moves and duplicates one double-precision value to double-precision
+/// Moves and duplicates one double-precision value to double-precision
/// values stored in a 128-bit vector of [2 x double].
///
/// \headerfile <x86intrin.h>
///
/// \code
-/// __m128d _mm_loaddup_pd(double const * dp);
+/// __m128d _mm_loaddup_pd(double const *dp);
/// \endcode
///
/// This intrinsic corresponds to the <c> VMOVDDUP </c> instruction.
/// duplicated values.
#define _mm_loaddup_pd(dp) _mm_load1_pd(dp)
-/// \brief Moves and duplicates the double-precision value in the lower bits of
+/// Moves and duplicates the double-precision value in the lower bits of
/// a 128-bit vector of [2 x double] to double-precision values stored in a
/// 128-bit vector of [2 x double].
///
return __builtin_shufflevector((__v2df)__a, (__v2df)__a, 0, 0);
}
-/// \brief Establishes a linear address memory range to be monitored and puts
+/// Establishes a linear address memory range to be monitored and puts
/// the processor in the monitor event pending state. Data stored in the
/// monitored address range causes the processor to exit the pending state.
///
static __inline__ void __DEFAULT_FN_ATTRS
_mm_monitor(void const *__p, unsigned __extensions, unsigned __hints)
{
- __builtin_ia32_monitor((void *)__p, __extensions, __hints);
+ __builtin_ia32_monitor(__p, __extensions, __hints);
}
-/// \brief Used with the MONITOR instruction to wait while the processor is in
+/// Used with the MONITOR instruction to wait while the processor is in
/// the monitor event pending state. Data stored in the monitored address
/// range causes the processor to exit the pending state.
///
--- /dev/null
+/*===---- popcntintrin.h - POPCNT intrinsics -------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __POPCNTINTRIN_H
+#define __POPCNTINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("popcnt")))
+
+/// Counts the number of bits in the source operand having a value of 1.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> POPCNT </c> instruction.
+///
+/// \param __A
+/// An unsigned 32-bit integer operand.
+/// \returns A 32-bit integer containing the number of bits with value 1 in the
+/// source operand.
+static __inline__ int __DEFAULT_FN_ATTRS
+_mm_popcnt_u32(unsigned int __A)
+{
+ return __builtin_popcount(__A);
+}
+
+#ifdef __x86_64__
+/// Counts the number of bits in the source operand having a value of 1.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> POPCNT </c> instruction.
+///
+/// \param __A
+/// An unsigned 64-bit integer operand.
+/// \returns A 64-bit integer containing the number of bits with value 1 in the
+/// source operand.
+static __inline__ long long __DEFAULT_FN_ATTRS
+_mm_popcnt_u64(unsigned long long __A)
+{
+ return __builtin_popcountll(__A);
+}
+#endif /* __x86_64__ */
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif /* __POPCNTINTRIN_H */
--- /dev/null
+/*===---- emmintrin.h - Implementation of SSE2 intrinsics on PowerPC -------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+/* Implemented from the specification included in the Intel C++ Compiler
+ User Guide and Reference, version 9.0. */
+
+#ifndef NO_WARN_X86_INTRINSICS
+/* This header file is to help porting code using Intel intrinsics
+ explicitly from x86_64 to powerpc64/powerpc64le.
+
+ Since X86 SSE2 intrinsics mainly handles __m128i and __m128d type,
+ PowerPC VMX/VSX ISA is a good match for vector float SIMD operations.
+ However scalar float operations in vector (XMM) registers require
+ the POWER8 VSX ISA (2.07) level. There are differences for data
+ format and placement of float scalars in the vector register, which
+ require extra steps to match SSE2 scalar float semantics on POWER.
+
+ It should be noted that there's much difference between X86_64's
+ MXSCR and PowerISA's FPSCR/VSCR registers. It's recommended to use
+ portable <fenv.h> instead of access MXSCR directly.
+
+ Most SSE2 scalar float intrinsic operations can be performed more
+ efficiently as C language float scalar operations or optimized to
+ use vector SIMD operations. We recommend this for new applications.
+*/
+#error "Please read comment above. Use -DNO_WARN_X86_INTRINSICS to disable this error."
+#endif
+
+#ifndef EMMINTRIN_H_
+#define EMMINTRIN_H_
+
+#if defined(__linux__) && defined(__ppc64__)
+
+#include <altivec.h>
+
+/* We need definitions from the SSE header files. */
+#include <xmmintrin.h>
+
+/* SSE2 */
+typedef __vector double __v2df;
+typedef __vector long long __v2di;
+typedef __vector unsigned long long __v2du;
+typedef __vector int __v4si;
+typedef __vector unsigned int __v4su;
+typedef __vector short __v8hi;
+typedef __vector unsigned short __v8hu;
+typedef __vector signed char __v16qi;
+typedef __vector unsigned char __v16qu;
+
+/* The Intel API is flexible enough that we must allow aliasing with other
+ vector types, and their scalar components. */
+typedef long long __m128i __attribute__ ((__vector_size__ (16), __may_alias__));
+typedef double __m128d __attribute__ ((__vector_size__ (16), __may_alias__));
+
+/* Unaligned version of the same types. */
+typedef long long __m128i_u __attribute__ ((__vector_size__ (16), __may_alias__, __aligned__ (1)));
+typedef double __m128d_u __attribute__ ((__vector_size__ (16), __may_alias__, __aligned__ (1)));
+
+/* Define two value permute mask. */
+#define _MM_SHUFFLE2(x,y) (((x) << 1) | (y))
+
+/* Create a vector with element 0 as F and the rest zero. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_sd (double __F)
+{
+ return __extension__ (__m128d){ __F, 0.0 };
+}
+
+/* Create a vector with both elements equal to F. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set1_pd (double __F)
+{
+ return __extension__ (__m128d){ __F, __F };
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_pd1 (double __F)
+{
+ return _mm_set1_pd (__F);
+}
+
+/* Create a vector with the lower value X and upper value W. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_pd (double __W, double __X)
+{
+ return __extension__ (__m128d){ __X, __W };
+}
+
+/* Create a vector with the lower value W and upper value X. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_setr_pd (double __W, double __X)
+{
+ return __extension__ (__m128d){ __W, __X };
+}
+
+/* Create an undefined vector. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_undefined_pd (void)
+{
+ __m128d __Y = __Y;
+ return __Y;
+}
+
+/* Create a vector of zeros. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_setzero_pd (void)
+{
+ return (__m128d) vec_splats (0);
+}
+
+/* Sets the low DPFP value of A from the low value of B. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_move_sd (__m128d __A, __m128d __B)
+{
+ __v2df result = (__v2df) __A;
+ result [0] = ((__v2df) __B)[0];
+ return (__m128d) result;
+}
+
+/* Load two DPFP values from P. The address must be 16-byte aligned. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_load_pd (double const *__P)
+{
+ return ((__m128d)vec_ld(0, (__v16qu*)__P));
+}
+
+/* Load two DPFP values from P. The address need not be 16-byte aligned. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loadu_pd (double const *__P)
+{
+ return (vec_vsx_ld(0, __P));
+}
+
+/* Create a vector with all two elements equal to *P. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_load1_pd (double const *__P)
+{
+ return (vec_splats (*__P));
+}
+
+/* Create a vector with element 0 as *P and the rest zero. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_load_sd (double const *__P)
+{
+ return _mm_set_sd (*__P);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_load_pd1 (double const *__P)
+{
+ return _mm_load1_pd (__P);
+}
+
+/* Load two DPFP values in reverse order. The address must be aligned. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loadr_pd (double const *__P)
+{
+ __v2df __tmp = _mm_load_pd (__P);
+ return (__m128d)vec_xxpermdi (__tmp, __tmp, 2);
+}
+
+/* Store two DPFP values. The address must be 16-byte aligned. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_store_pd (double *__P, __m128d __A)
+{
+ vec_st((__v16qu)__A, 0, (__v16qu*)__P);
+}
+
+/* Store two DPFP values. The address need not be 16-byte aligned. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_storeu_pd (double *__P, __m128d __A)
+{
+ *(__m128d_u *)__P = __A;
+}
+
+/* Stores the lower DPFP value. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_store_sd (double *__P, __m128d __A)
+{
+ *__P = ((__v2df)__A)[0];
+}
+
+extern __inline double __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsd_f64 (__m128d __A)
+{
+ return ((__v2df)__A)[0];
+}
+
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_storel_pd (double *__P, __m128d __A)
+{
+ _mm_store_sd (__P, __A);
+}
+
+/* Stores the upper DPFP value. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_storeh_pd (double *__P, __m128d __A)
+{
+ *__P = ((__v2df)__A)[1];
+}
+/* Store the lower DPFP value across two words.
+ The address must be 16-byte aligned. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_store1_pd (double *__P, __m128d __A)
+{
+ _mm_store_pd (__P, vec_splat (__A, 0));
+}
+
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_store_pd1 (double *__P, __m128d __A)
+{
+ _mm_store1_pd (__P, __A);
+}
+
+/* Store two DPFP values in reverse order. The address must be aligned. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_storer_pd (double *__P, __m128d __A)
+{
+ _mm_store_pd (__P, vec_xxpermdi (__A, __A, 2));
+}
+
+/* Intel intrinsic. */
+extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi128_si64 (__m128i __A)
+{
+ return ((__v2di)__A)[0];
+}
+
+/* Microsoft intrinsic. */
+extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi128_si64x (__m128i __A)
+{
+ return ((__v2di)__A)[0];
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_add_pd (__m128d __A, __m128d __B)
+{
+ return (__m128d) ((__v2df)__A + (__v2df)__B);
+}
+
+/* Add the lower double-precision (64-bit) floating-point element in
+ a and b, store the result in the lower element of dst, and copy
+ the upper element from a to the upper element of dst. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_add_sd (__m128d __A, __m128d __B)
+{
+ __A[0] = __A[0] + __B[0];
+ return (__A);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sub_pd (__m128d __A, __m128d __B)
+{
+ return (__m128d) ((__v2df)__A - (__v2df)__B);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sub_sd (__m128d __A, __m128d __B)
+{
+ __A[0] = __A[0] - __B[0];
+ return (__A);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mul_pd (__m128d __A, __m128d __B)
+{
+ return (__m128d) ((__v2df)__A * (__v2df)__B);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mul_sd (__m128d __A, __m128d __B)
+{
+ __A[0] = __A[0] * __B[0];
+ return (__A);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_div_pd (__m128d __A, __m128d __B)
+{
+ return (__m128d) ((__v2df)__A / (__v2df)__B);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_div_sd (__m128d __A, __m128d __B)
+{
+ __A[0] = __A[0] / __B[0];
+ return (__A);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sqrt_pd (__m128d __A)
+{
+ return (vec_sqrt (__A));
+}
+
+/* Return pair {sqrt (B[0]), A[1]}. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sqrt_sd (__m128d __A, __m128d __B)
+{
+ __v2df c;
+ c = vec_sqrt ((__v2df) _mm_set1_pd (__B[0]));
+ return (__m128d) _mm_setr_pd (c[0], __A[1]);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_min_pd (__m128d __A, __m128d __B)
+{
+ return (vec_min (__A, __B));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_min_sd (__m128d __A, __m128d __B)
+{
+ __v2df a, b, c;
+ a = vec_splats (__A[0]);
+ b = vec_splats (__B[0]);
+ c = vec_min (a, b);
+ return (__m128d) _mm_setr_pd (c[0], __A[1]);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_max_pd (__m128d __A, __m128d __B)
+{
+ return (vec_max (__A, __B));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_max_sd (__m128d __A, __m128d __B)
+{
+ __v2df a, b, c;
+ a = vec_splats (__A[0]);
+ b = vec_splats (__B[0]);
+ c = vec_max (a, b);
+ return (__m128d) _mm_setr_pd (c[0], __A[1]);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpeq_pd (__m128d __A, __m128d __B)
+{
+ return ((__m128d)vec_cmpeq ((__v2df) __A, (__v2df) __B));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmplt_pd (__m128d __A, __m128d __B)
+{
+ return ((__m128d)vec_cmplt ((__v2df) __A, (__v2df) __B));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmple_pd (__m128d __A, __m128d __B)
+{
+ return ((__m128d)vec_cmple ((__v2df) __A, (__v2df) __B));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpgt_pd (__m128d __A, __m128d __B)
+{
+ return ((__m128d)vec_cmpgt ((__v2df) __A, (__v2df) __B));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpge_pd (__m128d __A, __m128d __B)
+{
+ return ((__m128d)vec_cmpge ((__v2df) __A,(__v2df) __B));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpneq_pd (__m128d __A, __m128d __B)
+{
+ __v2df temp = (__v2df) vec_cmpeq ((__v2df) __A, (__v2df)__B);
+ return ((__m128d)vec_nor (temp, temp));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnlt_pd (__m128d __A, __m128d __B)
+{
+ return ((__m128d)vec_cmpge ((__v2df) __A, (__v2df) __B));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnle_pd (__m128d __A, __m128d __B)
+{
+ return ((__m128d)vec_cmpgt ((__v2df) __A, (__v2df) __B));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpngt_pd (__m128d __A, __m128d __B)
+{
+ return ((__m128d)vec_cmple ((__v2df) __A, (__v2df) __B));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnge_pd (__m128d __A, __m128d __B)
+{
+ return ((__m128d)vec_cmplt ((__v2df) __A, (__v2df) __B));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpord_pd (__m128d __A, __m128d __B)
+{
+#if _ARCH_PWR8
+ __v2du c, d;
+ /* Compare against self will return false (0's) if NAN. */
+ c = (__v2du)vec_cmpeq (__A, __A);
+ d = (__v2du)vec_cmpeq (__B, __B);
+#else
+ __v2du a, b;
+ __v2du c, d;
+ const __v2du double_exp_mask = {0x7ff0000000000000, 0x7ff0000000000000};
+ a = (__v2du)vec_abs ((__v2df)__A);
+ b = (__v2du)vec_abs ((__v2df)__B);
+ c = (__v2du)vec_cmpgt (double_exp_mask, a);
+ d = (__v2du)vec_cmpgt (double_exp_mask, b);
+#endif
+ /* A != NAN and B != NAN. */
+ return ((__m128d)vec_and(c, d));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpunord_pd (__m128d __A, __m128d __B)
+{
+#if _ARCH_PWR8
+ __v2du c, d;
+ /* Compare against self will return false (0's) if NAN. */
+ c = (__v2du)vec_cmpeq ((__v2df)__A, (__v2df)__A);
+ d = (__v2du)vec_cmpeq ((__v2df)__B, (__v2df)__B);
+ /* A == NAN OR B == NAN converts too:
+ NOT(A != NAN) OR NOT(B != NAN). */
+ c = vec_nor (c, c);
+ return ((__m128d)vec_orc(c, d));
+#else
+ __v2du c, d;
+ /* Compare against self will return false (0's) if NAN. */
+ c = (__v2du)vec_cmpeq ((__v2df)__A, (__v2df)__A);
+ d = (__v2du)vec_cmpeq ((__v2df)__B, (__v2df)__B);
+ /* Convert the true ('1's) is NAN. */
+ c = vec_nor (c, c);
+ d = vec_nor (d, d);
+ return ((__m128d)vec_or(c, d));
+#endif
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpeq_sd(__m128d __A, __m128d __B)
+{
+ __v2df a, b, c;
+ /* PowerISA VSX does not allow partial (for just lower double)
+ results. So to insure we don't generate spurious exceptions
+ (from the upper double values) we splat the lower double
+ before we do the operation. */
+ a = vec_splats (__A[0]);
+ b = vec_splats (__B[0]);
+ c = (__v2df) vec_cmpeq(a, b);
+ /* Then we merge the lower double result with the original upper
+ double from __A. */
+ return (__m128d) _mm_setr_pd (c[0], __A[1]);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmplt_sd (__m128d __A, __m128d __B)
+{
+ __v2df a, b, c;
+ a = vec_splats (__A[0]);
+ b = vec_splats (__B[0]);
+ c = (__v2df) vec_cmplt(a, b);
+ return (__m128d) _mm_setr_pd (c[0], __A[1]);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmple_sd (__m128d __A, __m128d __B)
+{
+ __v2df a, b, c;
+ a = vec_splats (__A[0]);
+ b = vec_splats (__B[0]);
+ c = (__v2df) vec_cmple(a, b);
+ return (__m128d) _mm_setr_pd (c[0], __A[1]);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpgt_sd (__m128d __A, __m128d __B)
+{
+ __v2df a, b, c;
+ a = vec_splats (__A[0]);
+ b = vec_splats (__B[0]);
+ c = (__v2df) vec_cmpgt(a, b);
+ return (__m128d) _mm_setr_pd (c[0], __A[1]);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpge_sd (__m128d __A, __m128d __B)
+{
+ __v2df a, b, c;
+ a = vec_splats (__A[0]);
+ b = vec_splats (__B[0]);
+ c = (__v2df) vec_cmpge(a, b);
+ return (__m128d) _mm_setr_pd (c[0], __A[1]);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpneq_sd (__m128d __A, __m128d __B)
+{
+ __v2df a, b, c;
+ a = vec_splats (__A[0]);
+ b = vec_splats (__B[0]);
+ c = (__v2df) vec_cmpeq(a, b);
+ c = vec_nor (c, c);
+ return (__m128d) _mm_setr_pd (c[0], __A[1]);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnlt_sd (__m128d __A, __m128d __B)
+{
+ __v2df a, b, c;
+ a = vec_splats (__A[0]);
+ b = vec_splats (__B[0]);
+ /* Not less than is just greater than or equal. */
+ c = (__v2df) vec_cmpge(a, b);
+ return (__m128d) _mm_setr_pd (c[0], __A[1]);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnle_sd (__m128d __A, __m128d __B)
+{
+ __v2df a, b, c;
+ a = vec_splats (__A[0]);
+ b = vec_splats (__B[0]);
+ /* Not less than or equal is just greater than. */
+ c = (__v2df) vec_cmpge(a, b);
+ return (__m128d) _mm_setr_pd (c[0], __A[1]);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpngt_sd (__m128d __A, __m128d __B)
+{
+ __v2df a, b, c;
+ a = vec_splats (__A[0]);
+ b = vec_splats (__B[0]);
+ /* Not greater than is just less than or equal. */
+ c = (__v2df) vec_cmple(a, b);
+ return (__m128d) _mm_setr_pd (c[0], __A[1]);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnge_sd (__m128d __A, __m128d __B)
+{
+ __v2df a, b, c;
+ a = vec_splats (__A[0]);
+ b = vec_splats (__B[0]);
+ /* Not greater than or equal is just less than. */
+ c = (__v2df) vec_cmplt(a, b);
+ return (__m128d) _mm_setr_pd (c[0], __A[1]);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpord_sd (__m128d __A, __m128d __B)
+{
+ __v2df r;
+ r = (__v2df)_mm_cmpord_pd (vec_splats (__A[0]), vec_splats (__B[0]));
+ return (__m128d) _mm_setr_pd (r[0], ((__v2df)__A)[1]);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpunord_sd (__m128d __A, __m128d __B)
+{
+ __v2df r;
+ r = _mm_cmpunord_pd (vec_splats (__A[0]), vec_splats (__B[0]));
+ return (__m128d) _mm_setr_pd (r[0], __A[1]);
+}
+
+/* FIXME
+ The __mm_comi??_sd and __mm_ucomi??_sd implementations below are
+ exactly the same because GCC for PowerPC only generates unordered
+ compares (scalar and vector).
+ Technically __mm_comieq_sp et all should be using the ordered
+ compare and signal for QNaNs. The __mm_ucomieq_sd et all should
+ be OK. */
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comieq_sd (__m128d __A, __m128d __B)
+{
+ return (__A[0] == __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comilt_sd (__m128d __A, __m128d __B)
+{
+ return (__A[0] < __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comile_sd (__m128d __A, __m128d __B)
+{
+ return (__A[0] <= __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comigt_sd (__m128d __A, __m128d __B)
+{
+ return (__A[0] > __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comige_sd (__m128d __A, __m128d __B)
+{
+ return (__A[0] >= __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comineq_sd (__m128d __A, __m128d __B)
+{
+ return (__A[0] != __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomieq_sd (__m128d __A, __m128d __B)
+{
+ return (__A[0] == __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomilt_sd (__m128d __A, __m128d __B)
+{
+ return (__A[0] < __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomile_sd (__m128d __A, __m128d __B)
+{
+ return (__A[0] <= __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomigt_sd (__m128d __A, __m128d __B)
+{
+ return (__A[0] > __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomige_sd (__m128d __A, __m128d __B)
+{
+ return (__A[0] >= __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomineq_sd (__m128d __A, __m128d __B)
+{
+ return (__A[0] != __B[0]);
+}
+
+/* Create a vector of Qi, where i is the element number. */
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_epi64x (long long __q1, long long __q0)
+{
+ return __extension__ (__m128i)(__v2di){ __q0, __q1 };
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_epi64 (__m64 __q1, __m64 __q0)
+{
+ return _mm_set_epi64x ((long long)__q1, (long long)__q0);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_epi32 (int __q3, int __q2, int __q1, int __q0)
+{
+ return __extension__ (__m128i)(__v4si){ __q0, __q1, __q2, __q3 };
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_epi16 (short __q7, short __q6, short __q5, short __q4,
+ short __q3, short __q2, short __q1, short __q0)
+{
+ return __extension__ (__m128i)(__v8hi){
+ __q0, __q1, __q2, __q3, __q4, __q5, __q6, __q7 };
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_epi8 (char __q15, char __q14, char __q13, char __q12,
+ char __q11, char __q10, char __q09, char __q08,
+ char __q07, char __q06, char __q05, char __q04,
+ char __q03, char __q02, char __q01, char __q00)
+{
+ return __extension__ (__m128i)(__v16qi){
+ __q00, __q01, __q02, __q03, __q04, __q05, __q06, __q07,
+ __q08, __q09, __q10, __q11, __q12, __q13, __q14, __q15
+ };
+}
+
+/* Set all of the elements of the vector to A. */
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set1_epi64x (long long __A)
+{
+ return _mm_set_epi64x (__A, __A);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set1_epi64 (__m64 __A)
+{
+ return _mm_set_epi64 (__A, __A);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set1_epi32 (int __A)
+{
+ return _mm_set_epi32 (__A, __A, __A, __A);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set1_epi16 (short __A)
+{
+ return _mm_set_epi16 (__A, __A, __A, __A, __A, __A, __A, __A);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set1_epi8 (char __A)
+{
+ return _mm_set_epi8 (__A, __A, __A, __A, __A, __A, __A, __A,
+ __A, __A, __A, __A, __A, __A, __A, __A);
+}
+
+/* Create a vector of Qi, where i is the element number.
+ The parameter order is reversed from the _mm_set_epi* functions. */
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_setr_epi64 (__m64 __q0, __m64 __q1)
+{
+ return _mm_set_epi64 (__q1, __q0);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_setr_epi32 (int __q0, int __q1, int __q2, int __q3)
+{
+ return _mm_set_epi32 (__q3, __q2, __q1, __q0);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_setr_epi16 (short __q0, short __q1, short __q2, short __q3,
+ short __q4, short __q5, short __q6, short __q7)
+{
+ return _mm_set_epi16 (__q7, __q6, __q5, __q4, __q3, __q2, __q1, __q0);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_setr_epi8 (char __q00, char __q01, char __q02, char __q03,
+ char __q04, char __q05, char __q06, char __q07,
+ char __q08, char __q09, char __q10, char __q11,
+ char __q12, char __q13, char __q14, char __q15)
+{
+ return _mm_set_epi8 (__q15, __q14, __q13, __q12, __q11, __q10, __q09, __q08,
+ __q07, __q06, __q05, __q04, __q03, __q02, __q01, __q00);
+}
+
+/* Create a vector with element 0 as *P and the rest zero. */
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_load_si128 (__m128i const *__P)
+{
+ return *__P;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loadu_si128 (__m128i_u const *__P)
+{
+ return (__m128i) (vec_vsx_ld(0, (signed int const *)__P));
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loadl_epi64 (__m128i_u const *__P)
+{
+ return _mm_set_epi64 ((__m64)0LL, *(__m64 *)__P);
+}
+
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_store_si128 (__m128i *__P, __m128i __B)
+{
+ vec_st ((__v16qu) __B, 0, (__v16qu*)__P);
+}
+
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_storeu_si128 (__m128i_u *__P, __m128i __B)
+{
+ *__P = __B;
+}
+
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_storel_epi64 (__m128i_u *__P, __m128i __B)
+{
+ *(long long *)__P = ((__v2di)__B)[0];
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_movepi64_pi64 (__m128i_u __B)
+{
+ return (__m64) ((__v2di)__B)[0];
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_movpi64_epi64 (__m64 __A)
+{
+ return _mm_set_epi64 ((__m64)0LL, __A);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_move_epi64 (__m128i __A)
+{
+ return _mm_set_epi64 ((__m64)0LL, (__m64)__A[0]);
+}
+
+/* Create an undefined vector. */
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_undefined_si128 (void)
+{
+ __m128i __Y = __Y;
+ return __Y;
+}
+
+/* Create a vector of zeros. */
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_setzero_si128 (void)
+{
+ return __extension__ (__m128i)(__v4si){ 0, 0, 0, 0 };
+}
+
+#ifdef _ARCH_PWR8
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtepi32_pd (__m128i __A)
+{
+ __v2di val;
+ /* For LE need to generate Vector Unpack Low Signed Word.
+ Which is generated from unpackh. */
+ val = (__v2di)vec_unpackh ((__v4si)__A);
+
+ return (__m128d)vec_ctf (val, 0);
+}
+#endif
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtepi32_ps (__m128i __A)
+{
+ return ((__m128)vec_ctf((__v4si)__A, 0));
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpd_epi32 (__m128d __A)
+{
+ __v2df rounded = vec_rint (__A);
+ __v4si result, temp;
+ const __v4si vzero =
+ { 0, 0, 0, 0 };
+
+ /* VSX Vector truncate Double-Precision to integer and Convert to
+ Signed Integer Word format with Saturate. */
+ __asm__(
+ "xvcvdpsxws %x0,%x1"
+ : "=wa" (temp)
+ : "wa" (rounded)
+ : );
+
+#ifdef _ARCH_PWR8
+ temp = vec_mergeo (temp, temp);
+ result = (__v4si) vec_vpkudum ((__vector long long) temp,
+ (__vector long long) vzero);
+#else
+ {
+ const __v16qu pkperm = {0x00, 0x01, 0x02, 0x03, 0x08, 0x09, 0x0a, 0x0b,
+ 0x14, 0x15, 0x16, 0x17, 0x1c, 0x1d, 0x1e, 0x1f };
+ result = (__v4si) vec_perm ((__v16qu) temp, (__v16qu) vzero, pkperm);
+ }
+#endif
+ return (__m128i) result;
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpd_pi32 (__m128d __A)
+{
+ __m128i result = _mm_cvtpd_epi32(__A);
+
+ return (__m64) result[0];
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpd_ps (__m128d __A)
+{
+ __v4sf result;
+ __v4si temp;
+ const __v4si vzero = { 0, 0, 0, 0 };
+
+ __asm__(
+ "xvcvdpsp %x0,%x1"
+ : "=wa" (temp)
+ : "wa" (__A)
+ : );
+
+#ifdef _ARCH_PWR8
+ temp = vec_mergeo (temp, temp);
+ result = (__v4sf) vec_vpkudum ((__vector long long) temp,
+ (__vector long long) vzero);
+#else
+ {
+ const __v16qu pkperm = {0x00, 0x01, 0x02, 0x03, 0x08, 0x09, 0x0a, 0x0b,
+ 0x14, 0x15, 0x16, 0x17, 0x1c, 0x1d, 0x1e, 0x1f };
+ result = (__v4sf) vec_perm ((__v16qu) temp, (__v16qu) vzero, pkperm);
+ }
+#endif
+ return ((__m128)result);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttpd_epi32 (__m128d __A)
+{
+ __v4si result;
+ __v4si temp;
+ const __v4si vzero = { 0, 0, 0, 0 };
+
+ /* VSX Vector truncate Double-Precision to integer and Convert to
+ Signed Integer Word format with Saturate. */
+ __asm__(
+ "xvcvdpsxws %x0,%x1"
+ : "=wa" (temp)
+ : "wa" (__A)
+ : );
+
+#ifdef _ARCH_PWR8
+ temp = vec_mergeo (temp, temp);
+ result = (__v4si) vec_vpkudum ((__vector long long) temp,
+ (__vector long long) vzero);
+#else
+ {
+ const __v16qu pkperm = {0x00, 0x01, 0x02, 0x03, 0x08, 0x09, 0x0a, 0x0b,
+ 0x14, 0x15, 0x16, 0x17, 0x1c, 0x1d, 0x1e, 0x1f };
+ result = (__v4si) vec_perm ((__v16qu) temp, (__v16qu) vzero, pkperm);
+ }
+#endif
+
+ return ((__m128i) result);
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttpd_pi32 (__m128d __A)
+{
+ __m128i result = _mm_cvttpd_epi32 (__A);
+
+ return (__m64) result[0];
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi128_si32 (__m128i __A)
+{
+ return ((__v4si)__A)[0];
+}
+
+#ifdef _ARCH_PWR8
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpi32_pd (__m64 __A)
+{
+ __v4si temp;
+ __v2di tmp2;
+ __v2df result;
+
+ temp = (__v4si)vec_splats (__A);
+ tmp2 = (__v2di)vec_unpackl (temp);
+ result = vec_ctf ((__vector signed long long) tmp2, 0);
+ return (__m128d)result;
+}
+#endif
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtps_epi32 (__m128 __A)
+{
+ __v4sf rounded;
+ __v4si result;
+
+ rounded = vec_rint((__v4sf) __A);
+ result = vec_cts (rounded, 0);
+ return (__m128i) result;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttps_epi32 (__m128 __A)
+{
+ __v4si result;
+
+ result = vec_cts ((__v4sf) __A, 0);
+ return (__m128i) result;
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtps_pd (__m128 __A)
+{
+ /* Check if vec_doubleh is defined by <altivec.h>. If so use that. */
+#ifdef vec_doubleh
+ return (__m128d) vec_doubleh ((__v4sf)__A);
+#else
+ /* Otherwise the compiler is not current and so need to generate the
+ equivalent code. */
+ __v4sf a = (__v4sf)__A;
+ __v4sf temp;
+ __v2df result;
+#ifdef __LITTLE_ENDIAN__
+ /* The input float values are in elements {[0], [1]} but the convert
+ instruction needs them in elements {[1], [3]}, So we use two
+ shift left double vector word immediates to get the elements
+ lined up. */
+ temp = __builtin_vsx_xxsldwi (a, a, 3);
+ temp = __builtin_vsx_xxsldwi (a, temp, 2);
+#else
+ /* The input float values are in elements {[0], [1]} but the convert
+ instruction needs them in elements {[0], [2]}, So we use two
+ shift left double vector word immediates to get the elements
+ lined up. */
+ temp = vec_vmrghw (a, a);
+#endif
+ __asm__(
+ " xvcvspdp %x0,%x1"
+ : "=wa" (result)
+ : "wa" (temp)
+ : );
+ return (__m128d) result;
+#endif
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsd_si32 (__m128d __A)
+{
+ __v2df rounded = vec_rint((__v2df) __A);
+ int result = ((__v2df)rounded)[0];
+
+ return result;
+}
+/* Intel intrinsic. */
+extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsd_si64 (__m128d __A)
+{
+ __v2df rounded = vec_rint ((__v2df) __A );
+ long long result = ((__v2df) rounded)[0];
+
+ return result;
+}
+
+/* Microsoft intrinsic. */
+extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsd_si64x (__m128d __A)
+{
+ return _mm_cvtsd_si64 ((__v2df)__A);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttsd_si32 (__m128d __A)
+{
+ int result = ((__v2df)__A)[0];
+
+ return result;
+}
+
+/* Intel intrinsic. */
+extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttsd_si64 (__m128d __A)
+{
+ long long result = ((__v2df)__A)[0];
+
+ return result;
+}
+
+/* Microsoft intrinsic. */
+extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttsd_si64x (__m128d __A)
+{
+ return _mm_cvttsd_si64 (__A);
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsd_ss (__m128 __A, __m128d __B)
+{
+ __v4sf result = (__v4sf)__A;
+
+#ifdef __LITTLE_ENDIAN__
+ __v4sf temp_s;
+ /* Copy double element[0] to element [1] for conversion. */
+ __v2df temp_b = vec_splat((__v2df)__B, 0);
+
+ /* Pre-rotate __A left 3 (logically right 1) elements. */
+ result = __builtin_vsx_xxsldwi (result, result, 3);
+ /* Convert double to single float scalar in a vector. */
+ __asm__(
+ "xscvdpsp %x0,%x1"
+ : "=wa" (temp_s)
+ : "wa" (temp_b)
+ : );
+ /* Shift the resulting scalar into vector element [0]. */
+ result = __builtin_vsx_xxsldwi (result, temp_s, 1);
+#else
+ result [0] = ((__v2df)__B)[0];
+#endif
+ return (__m128) result;
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi32_sd (__m128d __A, int __B)
+{
+ __v2df result = (__v2df)__A;
+ double db = __B;
+ result [0] = db;
+ return (__m128d)result;
+}
+
+/* Intel intrinsic. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi64_sd (__m128d __A, long long __B)
+{
+ __v2df result = (__v2df)__A;
+ double db = __B;
+ result [0] = db;
+ return (__m128d)result;
+}
+
+/* Microsoft intrinsic. */
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi64x_sd (__m128d __A, long long __B)
+{
+ return _mm_cvtsi64_sd (__A, __B);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtss_sd (__m128d __A, __m128 __B)
+{
+#ifdef __LITTLE_ENDIAN__
+ /* Use splat to move element [0] into position for the convert. */
+ __v4sf temp = vec_splat ((__v4sf)__B, 0);
+ __v2df res;
+ /* Convert single float scalar to double in a vector. */
+ __asm__(
+ "xscvspdp %x0,%x1"
+ : "=wa" (res)
+ : "wa" (temp)
+ : );
+ return (__m128d) vec_mergel (res, (__v2df)__A);
+#else
+ __v2df res = (__v2df)__A;
+ res [0] = ((__v4sf)__B) [0];
+ return (__m128d) res;
+#endif
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_shuffle_pd(__m128d __A, __m128d __B, const int __mask)
+{
+ __vector double result;
+ const int litmsk = __mask & 0x3;
+
+ if (litmsk == 0)
+ result = vec_mergeh (__A, __B);
+#if __GNUC__ < 6
+ else if (litmsk == 1)
+ result = vec_xxpermdi (__B, __A, 2);
+ else if (litmsk == 2)
+ result = vec_xxpermdi (__B, __A, 1);
+#else
+ else if (litmsk == 1)
+ result = vec_xxpermdi (__A, __B, 2);
+ else if (litmsk == 2)
+ result = vec_xxpermdi (__A, __B, 1);
+#endif
+ else
+ result = vec_mergel (__A, __B);
+
+ return result;
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_unpackhi_pd (__m128d __A, __m128d __B)
+{
+ return (__m128d) vec_mergel ((__v2df)__A, (__v2df)__B);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_unpacklo_pd (__m128d __A, __m128d __B)
+{
+ return (__m128d) vec_mergeh ((__v2df)__A, (__v2df)__B);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loadh_pd (__m128d __A, double const *__B)
+{
+ __v2df result = (__v2df)__A;
+ result [1] = *__B;
+ return (__m128d)result;
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loadl_pd (__m128d __A, double const *__B)
+{
+ __v2df result = (__v2df)__A;
+ result [0] = *__B;
+ return (__m128d)result;
+}
+
+#ifdef _ARCH_PWR8
+/* Intrinsic functions that require PowerISA 2.07 minimum. */
+
+/* Creates a 2-bit mask from the most significant bits of the DPFP values. */
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_movemask_pd (__m128d __A)
+{
+ __vector unsigned long long result;
+ static const __vector unsigned int perm_mask =
+ {
+#ifdef __LITTLE_ENDIAN__
+ 0x80800040, 0x80808080, 0x80808080, 0x80808080
+#else
+ 0x80808080, 0x80808080, 0x80808080, 0x80804000
+#endif
+ };
+
+ result = ((__vector unsigned long long)
+ vec_vbpermq ((__vector unsigned char) __A,
+ (__vector unsigned char) perm_mask));
+
+#ifdef __LITTLE_ENDIAN__
+ return result[1];
+#else
+ return result[0];
+#endif
+}
+#endif /* _ARCH_PWR8 */
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_packs_epi16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_packs ((__v8hi) __A, (__v8hi)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_packs_epi32 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_packs ((__v4si)__A, (__v4si)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_packus_epi16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_packsu ((__v8hi) __A, (__v8hi)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_unpackhi_epi8 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_mergel ((__v16qu)__A, (__v16qu)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_unpackhi_epi16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_mergel ((__v8hu)__A, (__v8hu)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_unpackhi_epi32 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_mergel ((__v4su)__A, (__v4su)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_unpackhi_epi64 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_mergel ((__vector long long) __A,
+ (__vector long long) __B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_unpacklo_epi8 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_mergeh ((__v16qu)__A, (__v16qu)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_unpacklo_epi16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_mergeh ((__v8hi)__A, (__v8hi)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_unpacklo_epi32 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_mergeh ((__v4si)__A, (__v4si)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_unpacklo_epi64 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_mergeh ((__vector long long) __A,
+ (__vector long long) __B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_add_epi8 (__m128i __A, __m128i __B)
+{
+ return (__m128i) ((__v16qu)__A + (__v16qu)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_add_epi16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) ((__v8hu)__A + (__v8hu)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_add_epi32 (__m128i __A, __m128i __B)
+{
+ return (__m128i) ((__v4su)__A + (__v4su)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_add_epi64 (__m128i __A, __m128i __B)
+{
+ return (__m128i) ((__v2du)__A + (__v2du)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_adds_epi8 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_adds ((__v16qi)__A, (__v16qi)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_adds_epi16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_adds ((__v8hi)__A, (__v8hi)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_adds_epu8 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_adds ((__v16qu)__A, (__v16qu)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_adds_epu16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_adds ((__v8hu)__A, (__v8hu)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sub_epi8 (__m128i __A, __m128i __B)
+{
+ return (__m128i) ((__v16qu)__A - (__v16qu)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sub_epi16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) ((__v8hu)__A - (__v8hu)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sub_epi32 (__m128i __A, __m128i __B)
+{
+ return (__m128i) ((__v4su)__A - (__v4su)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sub_epi64 (__m128i __A, __m128i __B)
+{
+ return (__m128i) ((__v2du)__A - (__v2du)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_subs_epi8 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_subs ((__v16qi)__A, (__v16qi)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_subs_epi16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_subs ((__v8hi)__A, (__v8hi)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_subs_epu8 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_subs ((__v16qu)__A, (__v16qu)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_subs_epu16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_subs ((__v8hu)__A, (__v8hu)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_madd_epi16 (__m128i __A, __m128i __B)
+{
+ __vector signed int zero = {0, 0, 0, 0};
+
+ return (__m128i) vec_vmsumshm ((__v8hi)__A, (__v8hi)__B, zero);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mulhi_epi16 (__m128i __A, __m128i __B)
+{
+ __vector signed int w0, w1;
+
+ __vector unsigned char xform1 = {
+#ifdef __LITTLE_ENDIAN__
+ 0x02, 0x03, 0x12, 0x13, 0x06, 0x07, 0x16, 0x17,
+ 0x0A, 0x0B, 0x1A, 0x1B, 0x0E, 0x0F, 0x1E, 0x1F
+#else
+ 0x00, 0x01, 0x10, 0x11, 0x04, 0x05, 0x14, 0x15,
+ 0x08, 0x09, 0x18, 0x19, 0x0C, 0x0D, 0x1C, 0x1D
+#endif
+ };
+
+ w0 = vec_vmulesh ((__v8hi)__A, (__v8hi)__B);
+ w1 = vec_vmulosh ((__v8hi)__A, (__v8hi)__B);
+ return (__m128i) vec_perm (w0, w1, xform1);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mullo_epi16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) ((__v8hi)__A * (__v8hi)__B);
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mul_su32 (__m64 __A, __m64 __B)
+{
+ unsigned int a = __A;
+ unsigned int b = __B;
+
+ return ((__m64)a * (__m64)b);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mul_epu32 (__m128i __A, __m128i __B)
+{
+#if __GNUC__ < 8
+ __v2du result;
+
+#ifdef __LITTLE_ENDIAN__
+ /* VMX Vector Multiply Odd Unsigned Word. */
+ __asm__(
+ "vmulouw %0,%1,%2"
+ : "=v" (result)
+ : "v" (__A), "v" (__B)
+ : );
+#else
+ /* VMX Vector Multiply Even Unsigned Word. */
+ __asm__(
+ "vmuleuw %0,%1,%2"
+ : "=v" (result)
+ : "v" (__A), "v" (__B)
+ : );
+#endif
+ return (__m128i) result;
+#else
+ return (__m128i) vec_mule ((__v4su)__A, (__v4su)__B);
+#endif
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_slli_epi16 (__m128i __A, int __B)
+{
+ __v8hu lshift;
+ __v8hi result = { 0, 0, 0, 0, 0, 0, 0, 0 };
+
+ if (__B >= 0 && __B < 16)
+ {
+ if (__builtin_constant_p(__B))
+ lshift = (__v8hu) vec_splat_s16(__B);
+ else
+ lshift = vec_splats ((unsigned short) __B);
+
+ result = vec_sl ((__v8hi) __A, lshift);
+ }
+
+ return (__m128i) result;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_slli_epi32 (__m128i __A, int __B)
+{
+ __v4su lshift;
+ __v4si result = { 0, 0, 0, 0 };
+
+ if (__B >= 0 && __B < 32)
+ {
+ if (__builtin_constant_p(__B) && __B < 16)
+ lshift = (__v4su) vec_splat_s32(__B);
+ else
+ lshift = vec_splats ((unsigned int) __B);
+
+ result = vec_sl ((__v4si) __A, lshift);
+ }
+
+ return (__m128i) result;
+}
+
+#ifdef _ARCH_PWR8
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_slli_epi64 (__m128i __A, int __B)
+{
+ __v2du lshift;
+ __v2di result = { 0, 0 };
+
+ if (__B >= 0 && __B < 64)
+ {
+ if (__builtin_constant_p(__B) && __B < 16)
+ lshift = (__v2du) vec_splat_s32(__B);
+ else
+ lshift = (__v2du) vec_splats ((unsigned int) __B);
+
+ result = vec_sl ((__v2di) __A, lshift);
+ }
+
+ return (__m128i) result;
+}
+#endif
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_srai_epi16 (__m128i __A, int __B)
+{
+ __v8hu rshift = { 15, 15, 15, 15, 15, 15, 15, 15 };
+ __v8hi result;
+
+ if (__B < 16)
+ {
+ if (__builtin_constant_p(__B))
+ rshift = (__v8hu) vec_splat_s16(__B);
+ else
+ rshift = vec_splats ((unsigned short) __B);
+ }
+ result = vec_sra ((__v8hi) __A, rshift);
+
+ return (__m128i) result;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_srai_epi32 (__m128i __A, int __B)
+{
+ __v4su rshift = { 31, 31, 31, 31 };
+ __v4si result;
+
+ if (__B < 32)
+ {
+ if (__builtin_constant_p(__B))
+ {
+ if (__B < 16)
+ rshift = (__v4su) vec_splat_s32(__B);
+ else
+ rshift = (__v4su) vec_splats((unsigned int)__B);
+ }
+ else
+ rshift = vec_splats ((unsigned int) __B);
+ }
+ result = vec_sra ((__v4si) __A, rshift);
+
+ return (__m128i) result;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_bslli_si128 (__m128i __A, const int __N)
+{
+ __v16qu result;
+ const __v16qu zeros = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
+
+ if (__N < 16)
+ result = vec_sld ((__v16qu) __A, zeros, __N);
+ else
+ result = zeros;
+
+ return (__m128i) result;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_bsrli_si128 (__m128i __A, const int __N)
+{
+ __v16qu result;
+ const __v16qu zeros = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
+
+ if (__N < 16)
+#ifdef __LITTLE_ENDIAN__
+ if (__builtin_constant_p(__N))
+ /* Would like to use Vector Shift Left Double by Octet
+ Immediate here to use the immediate form and avoid
+ load of __N * 8 value into a separate VR. */
+ result = vec_sld (zeros, (__v16qu) __A, (16 - __N));
+ else
+#endif
+ {
+ __v16qu shift = vec_splats((unsigned char)(__N*8));
+#ifdef __LITTLE_ENDIAN__
+ result = vec_sro ((__v16qu)__A, shift);
+#else
+ result = vec_slo ((__v16qu)__A, shift);
+#endif
+ }
+ else
+ result = zeros;
+
+ return (__m128i) result;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_srli_si128 (__m128i __A, const int __N)
+{
+ return _mm_bsrli_si128 (__A, __N);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_slli_si128 (__m128i __A, const int _imm5)
+{
+ __v16qu result;
+ const __v16qu zeros = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
+
+ if (_imm5 < 16)
+#ifdef __LITTLE_ENDIAN__
+ result = vec_sld ((__v16qu) __A, zeros, _imm5);
+#else
+ result = vec_sld (zeros, (__v16qu) __A, (16 - _imm5));
+#endif
+ else
+ result = zeros;
+
+ return (__m128i) result;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+
+_mm_srli_epi16 (__m128i __A, int __B)
+{
+ __v8hu rshift;
+ __v8hi result = { 0, 0, 0, 0, 0, 0, 0, 0 };
+
+ if (__B < 16)
+ {
+ if (__builtin_constant_p(__B))
+ rshift = (__v8hu) vec_splat_s16(__B);
+ else
+ rshift = vec_splats ((unsigned short) __B);
+
+ result = vec_sr ((__v8hi) __A, rshift);
+ }
+
+ return (__m128i) result;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_srli_epi32 (__m128i __A, int __B)
+{
+ __v4su rshift;
+ __v4si result = { 0, 0, 0, 0 };
+
+ if (__B < 32)
+ {
+ if (__builtin_constant_p(__B))
+ {
+ if (__B < 16)
+ rshift = (__v4su) vec_splat_s32(__B);
+ else
+ rshift = (__v4su) vec_splats((unsigned int)__B);
+ }
+ else
+ rshift = vec_splats ((unsigned int) __B);
+
+ result = vec_sr ((__v4si) __A, rshift);
+ }
+
+ return (__m128i) result;
+}
+
+#ifdef _ARCH_PWR8
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_srli_epi64 (__m128i __A, int __B)
+{
+ __v2du rshift;
+ __v2di result = { 0, 0 };
+
+ if (__B < 64)
+ {
+ if (__builtin_constant_p(__B))
+ {
+ if (__B < 16)
+ rshift = (__v2du) vec_splat_s32(__B);
+ else
+ rshift = (__v2du) vec_splats((unsigned long long)__B);
+ }
+ else
+ rshift = (__v2du) vec_splats ((unsigned int) __B);
+
+ result = vec_sr ((__v2di) __A, rshift);
+ }
+
+ return (__m128i) result;
+}
+#endif
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sll_epi16 (__m128i __A, __m128i __B)
+{
+ __v8hu lshift;
+ __vector __bool short shmask;
+ const __v8hu shmax = { 15, 15, 15, 15, 15, 15, 15, 15 };
+ __v8hu result;
+
+#ifdef __LITTLE_ENDIAN__
+ lshift = vec_splat ((__v8hu) __B, 0);
+#else
+ lshift = vec_splat ((__v8hu) __B, 3);
+#endif
+ shmask = vec_cmple (lshift, shmax);
+ result = vec_sl ((__v8hu) __A, lshift);
+ result = vec_sel ((__v8hu) shmask, result, shmask);
+
+ return (__m128i) result;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sll_epi32 (__m128i __A, __m128i __B)
+{
+ __v4su lshift;
+ __vector __bool int shmask;
+ const __v4su shmax = { 32, 32, 32, 32 };
+ __v4su result;
+#ifdef __LITTLE_ENDIAN__
+ lshift = vec_splat ((__v4su) __B, 0);
+#else
+ lshift = vec_splat ((__v4su) __B, 1);
+#endif
+ shmask = vec_cmplt (lshift, shmax);
+ result = vec_sl ((__v4su) __A, lshift);
+ result = vec_sel ((__v4su) shmask, result, shmask);
+
+ return (__m128i) result;
+}
+
+#ifdef _ARCH_PWR8
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sll_epi64 (__m128i __A, __m128i __B)
+{
+ __v2du lshift;
+ __vector __bool long long shmask;
+ const __v2du shmax = { 64, 64 };
+ __v2du result;
+
+ lshift = vec_splat ((__v2du) __B, 0);
+ shmask = vec_cmplt (lshift, shmax);
+ result = vec_sl ((__v2du) __A, lshift);
+ result = (__v2du)vec_sel ((__v2df) shmask, (__v2df)result, shmask);
+
+ return (__m128i) result;
+}
+#endif
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sra_epi16 (__m128i __A, __m128i __B)
+{
+ const __v8hu rshmax = { 15, 15, 15, 15, 15, 15, 15, 15 };
+ __v8hu rshift;
+ __v8hi result;
+
+#ifdef __LITTLE_ENDIAN__
+ rshift = vec_splat ((__v8hu)__B, 0);
+#else
+ rshift = vec_splat ((__v8hu)__B, 3);
+#endif
+ rshift = vec_min (rshift, rshmax);
+ result = vec_sra ((__v8hi) __A, rshift);
+
+ return (__m128i) result;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sra_epi32 (__m128i __A, __m128i __B)
+{
+ const __v4su rshmax = { 31, 31, 31, 31 };
+ __v4su rshift;
+ __v4si result;
+
+#ifdef __LITTLE_ENDIAN__
+ rshift = vec_splat ((__v4su)__B, 0);
+#else
+ rshift = vec_splat ((__v4su)__B, 1);
+#endif
+ rshift = vec_min (rshift, rshmax);
+ result = vec_sra ((__v4si) __A, rshift);
+
+ return (__m128i) result;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_srl_epi16 (__m128i __A, __m128i __B)
+{
+ __v8hu rshift;
+ __vector __bool short shmask;
+ const __v8hu shmax = { 15, 15, 15, 15, 15, 15, 15, 15 };
+ __v8hu result;
+
+#ifdef __LITTLE_ENDIAN__
+ rshift = vec_splat ((__v8hu) __B, 0);
+#else
+ rshift = vec_splat ((__v8hu) __B, 3);
+#endif
+ shmask = vec_cmple (rshift, shmax);
+ result = vec_sr ((__v8hu) __A, rshift);
+ result = vec_sel ((__v8hu) shmask, result, shmask);
+
+ return (__m128i) result;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_srl_epi32 (__m128i __A, __m128i __B)
+{
+ __v4su rshift;
+ __vector __bool int shmask;
+ const __v4su shmax = { 32, 32, 32, 32 };
+ __v4su result;
+
+#ifdef __LITTLE_ENDIAN__
+ rshift = vec_splat ((__v4su) __B, 0);
+#else
+ rshift = vec_splat ((__v4su) __B, 1);
+#endif
+ shmask = vec_cmplt (rshift, shmax);
+ result = vec_sr ((__v4su) __A, rshift);
+ result = vec_sel ((__v4su) shmask, result, shmask);
+
+ return (__m128i) result;
+}
+
+#ifdef _ARCH_PWR8
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_srl_epi64 (__m128i __A, __m128i __B)
+{
+ __v2du rshift;
+ __vector __bool long long shmask;
+ const __v2du shmax = { 64, 64 };
+ __v2du result;
+
+ rshift = vec_splat ((__v2du) __B, 0);
+ shmask = vec_cmplt (rshift, shmax);
+ result = vec_sr ((__v2du) __A, rshift);
+ result = (__v2du)vec_sel ((__v2df) shmask, (__v2df)result, shmask);
+
+ return (__m128i) result;
+}
+#endif
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_and_pd (__m128d __A, __m128d __B)
+{
+ return (vec_and ((__v2df) __A, (__v2df) __B));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_andnot_pd (__m128d __A, __m128d __B)
+{
+ return (vec_andc ((__v2df) __B, (__v2df) __A));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_or_pd (__m128d __A, __m128d __B)
+{
+ return (vec_or ((__v2df) __A, (__v2df) __B));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_xor_pd (__m128d __A, __m128d __B)
+{
+ return (vec_xor ((__v2df) __A, (__v2df) __B));
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_and_si128 (__m128i __A, __m128i __B)
+{
+ return (__m128i)vec_and ((__v2di) __A, (__v2di) __B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_andnot_si128 (__m128i __A, __m128i __B)
+{
+ return (__m128i)vec_andc ((__v2di) __B, (__v2di) __A);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_or_si128 (__m128i __A, __m128i __B)
+{
+ return (__m128i)vec_or ((__v2di) __A, (__v2di) __B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_xor_si128 (__m128i __A, __m128i __B)
+{
+ return (__m128i)vec_xor ((__v2di) __A, (__v2di) __B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpeq_epi8 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_cmpeq ((__v16qi) __A, (__v16qi)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpeq_epi16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_cmpeq ((__v8hi) __A, (__v8hi)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpeq_epi32 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_cmpeq ((__v4si) __A, (__v4si)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmplt_epi8 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_cmplt ((__v16qi) __A, (__v16qi)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmplt_epi16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_cmplt ((__v8hi) __A, (__v8hi)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmplt_epi32 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_cmplt ((__v4si) __A, (__v4si)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpgt_epi8 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_cmpgt ((__v16qi) __A, (__v16qi)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpgt_epi16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_cmpgt ((__v8hi) __A, (__v8hi)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpgt_epi32 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_cmpgt ((__v4si) __A, (__v4si)__B);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_extract_epi16 (__m128i const __A, int const __N)
+{
+ return (unsigned short) ((__v8hi)__A)[__N & 7];
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_insert_epi16 (__m128i const __A, int const __D, int const __N)
+{
+ __v8hi result = (__v8hi)__A;
+
+ result [(__N & 7)] = __D;
+
+ return (__m128i) result;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_max_epi16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_max ((__v8hi)__A, (__v8hi)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_max_epu8 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_max ((__v16qu) __A, (__v16qu)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_min_epi16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_min ((__v8hi) __A, (__v8hi)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_min_epu8 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_min ((__v16qu) __A, (__v16qu)__B);
+}
+
+
+#ifdef _ARCH_PWR8
+/* Intrinsic functions that require PowerISA 2.07 minimum. */
+
+/* Creates a 4-bit mask from the most significant bits of the SPFP values. */
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_movemask_epi8 (__m128i __A)
+{
+ __vector unsigned long long result;
+ static const __vector unsigned char perm_mask =
+ {
+ 0x78, 0x70, 0x68, 0x60, 0x58, 0x50, 0x48, 0x40,
+ 0x38, 0x30, 0x28, 0x20, 0x18, 0x10, 0x08, 0x00
+ };
+
+ result = ((__vector unsigned long long)
+ vec_vbpermq ((__vector unsigned char) __A,
+ (__vector unsigned char) perm_mask));
+
+#ifdef __LITTLE_ENDIAN__
+ return result[1];
+#else
+ return result[0];
+#endif
+}
+#endif /* _ARCH_PWR8 */
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mulhi_epu16 (__m128i __A, __m128i __B)
+{
+ __v4su w0, w1;
+ __v16qu xform1 = {
+#ifdef __LITTLE_ENDIAN__
+ 0x02, 0x03, 0x12, 0x13, 0x06, 0x07, 0x16, 0x17,
+ 0x0A, 0x0B, 0x1A, 0x1B, 0x0E, 0x0F, 0x1E, 0x1F
+#else
+ 0x00, 0x01, 0x10, 0x11, 0x04, 0x05, 0x14, 0x15,
+ 0x08, 0x09, 0x18, 0x19, 0x0C, 0x0D, 0x1C, 0x1D
+#endif
+ };
+
+ w0 = vec_vmuleuh ((__v8hu)__A, (__v8hu)__B);
+ w1 = vec_vmulouh ((__v8hu)__A, (__v8hu)__B);
+ return (__m128i) vec_perm (w0, w1, xform1);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_shufflehi_epi16 (__m128i __A, const int __mask)
+{
+ unsigned long element_selector_98 = __mask & 0x03;
+ unsigned long element_selector_BA = (__mask >> 2) & 0x03;
+ unsigned long element_selector_DC = (__mask >> 4) & 0x03;
+ unsigned long element_selector_FE = (__mask >> 6) & 0x03;
+ static const unsigned short permute_selectors[4] =
+ {
+#ifdef __LITTLE_ENDIAN__
+ 0x0908, 0x0B0A, 0x0D0C, 0x0F0E
+#else
+ 0x0809, 0x0A0B, 0x0C0D, 0x0E0F
+#endif
+ };
+ __v2du pmask =
+#ifdef __LITTLE_ENDIAN__
+ { 0x1716151413121110UL, 0UL};
+#else
+ { 0x1011121314151617UL, 0UL};
+#endif
+ __m64_union t;
+ __v2du a, r;
+
+ t.as_short[0] = permute_selectors[element_selector_98];
+ t.as_short[1] = permute_selectors[element_selector_BA];
+ t.as_short[2] = permute_selectors[element_selector_DC];
+ t.as_short[3] = permute_selectors[element_selector_FE];
+ pmask[1] = t.as_m64;
+ a = (__v2du)__A;
+ r = vec_perm (a, a, (__vector unsigned char)pmask);
+ return (__m128i) r;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_shufflelo_epi16 (__m128i __A, const int __mask)
+{
+ unsigned long element_selector_10 = __mask & 0x03;
+ unsigned long element_selector_32 = (__mask >> 2) & 0x03;
+ unsigned long element_selector_54 = (__mask >> 4) & 0x03;
+ unsigned long element_selector_76 = (__mask >> 6) & 0x03;
+ static const unsigned short permute_selectors[4] =
+ {
+#ifdef __LITTLE_ENDIAN__
+ 0x0100, 0x0302, 0x0504, 0x0706
+#else
+ 0x0001, 0x0203, 0x0405, 0x0607
+#endif
+ };
+ __v2du pmask =
+#ifdef __LITTLE_ENDIAN__
+ { 0UL, 0x1f1e1d1c1b1a1918UL};
+#else
+ { 0UL, 0x18191a1b1c1d1e1fUL};
+#endif
+ __m64_union t;
+ __v2du a, r;
+ t.as_short[0] = permute_selectors[element_selector_10];
+ t.as_short[1] = permute_selectors[element_selector_32];
+ t.as_short[2] = permute_selectors[element_selector_54];
+ t.as_short[3] = permute_selectors[element_selector_76];
+ pmask[0] = t.as_m64;
+ a = (__v2du)__A;
+ r = vec_perm (a, a, (__vector unsigned char)pmask);
+ return (__m128i) r;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_shuffle_epi32 (__m128i __A, const int __mask)
+{
+ unsigned long element_selector_10 = __mask & 0x03;
+ unsigned long element_selector_32 = (__mask >> 2) & 0x03;
+ unsigned long element_selector_54 = (__mask >> 4) & 0x03;
+ unsigned long element_selector_76 = (__mask >> 6) & 0x03;
+ static const unsigned int permute_selectors[4] =
+ {
+#ifdef __LITTLE_ENDIAN__
+ 0x03020100, 0x07060504, 0x0B0A0908, 0x0F0E0D0C
+#else
+ 0x00010203, 0x04050607, 0x08090A0B, 0x0C0D0E0F
+#endif
+ };
+ __v4su t;
+
+ t[0] = permute_selectors[element_selector_10];
+ t[1] = permute_selectors[element_selector_32];
+ t[2] = permute_selectors[element_selector_54] + 0x10101010;
+ t[3] = permute_selectors[element_selector_76] + 0x10101010;
+ return (__m128i)vec_perm ((__v4si) __A, (__v4si)__A, (__vector unsigned char)t);
+}
+
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskmoveu_si128 (__m128i __A, __m128i __B, char *__C)
+{
+ __v2du hibit = { 0x7f7f7f7f7f7f7f7fUL, 0x7f7f7f7f7f7f7f7fUL};
+ __v16qu mask, tmp;
+ __m128i_u *p = (__m128i_u*)__C;
+
+ tmp = (__v16qu)_mm_loadu_si128(p);
+ mask = (__v16qu)vec_cmpgt ((__v16qu)__B, (__v16qu)hibit);
+ tmp = vec_sel (tmp, (__v16qu)__A, mask);
+ _mm_storeu_si128 (p, (__m128i)tmp);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_avg_epu8 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_avg ((__v16qu)__A, (__v16qu)__B);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_avg_epu16 (__m128i __A, __m128i __B)
+{
+ return (__m128i) vec_avg ((__v8hu)__A, (__v8hu)__B);
+}
+
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sad_epu8 (__m128i __A, __m128i __B)
+{
+ __v16qu a, b;
+ __v16qu vmin, vmax, vabsdiff;
+ __v4si vsum;
+ const __v4su zero = { 0, 0, 0, 0 };
+ __v4si result;
+
+ a = (__v16qu) __A;
+ b = (__v16qu) __B;
+ vmin = vec_min (a, b);
+ vmax = vec_max (a, b);
+ vabsdiff = vec_sub (vmax, vmin);
+ /* Sum four groups of bytes into integers. */
+ vsum = (__vector signed int) vec_sum4s (vabsdiff, zero);
+ /* Sum across four integers with two integer results. */
+ result = vec_sum2s (vsum, (__vector signed int) zero);
+ /* Rotate the sums into the correct position. */
+#ifdef __LITTLE_ENDIAN__
+ result = vec_sld (result, result, 4);
+#else
+ result = vec_sld (result, result, 6);
+#endif
+ /* Rotate the sums into the correct position. */
+ return (__m128i) result;
+}
+
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_stream_si32 (int *__A, int __B)
+{
+ /* Use the data cache block touch for store transient. */
+ __asm__ (
+ "dcbtstt 0,%0"
+ :
+ : "b" (__A)
+ : "memory"
+ );
+ *__A = __B;
+}
+
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_stream_si64 (long long int *__A, long long int __B)
+{
+ /* Use the data cache block touch for store transient. */
+ __asm__ (
+ " dcbtstt 0,%0"
+ :
+ : "b" (__A)
+ : "memory"
+ );
+ *__A = __B;
+}
+
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_stream_si128 (__m128i *__A, __m128i __B)
+{
+ /* Use the data cache block touch for store transient. */
+ __asm__ (
+ "dcbtstt 0,%0"
+ :
+ : "b" (__A)
+ : "memory"
+ );
+ *__A = __B;
+}
+
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_stream_pd (double *__A, __m128d __B)
+{
+ /* Use the data cache block touch for store transient. */
+ __asm__ (
+ "dcbtstt 0,%0"
+ :
+ : "b" (__A)
+ : "memory"
+ );
+ *(__m128d*)__A = __B;
+}
+
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_clflush (void const *__A)
+{
+ /* Use the data cache block flush. */
+ __asm__ (
+ "dcbf 0,%0"
+ :
+ : "b" (__A)
+ : "memory"
+ );
+}
+
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_lfence (void)
+{
+ /* Use light weight sync for load to load ordering. */
+ __atomic_thread_fence (__ATOMIC_RELEASE);
+}
+
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mfence (void)
+{
+ /* Use heavy weight sync for any to any ordering. */
+ __atomic_thread_fence (__ATOMIC_SEQ_CST);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi32_si128 (int __A)
+{
+ return _mm_set_epi32 (0, 0, 0, __A);
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi64_si128 (long long __A)
+{
+ return __extension__ (__m128i)(__v2di){ __A, 0LL };
+}
+
+/* Microsoft intrinsic. */
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi64x_si128 (long long __A)
+{
+ return __extension__ (__m128i)(__v2di){ __A, 0LL };
+}
+
+/* Casts between various SP, DP, INT vector types. Note that these do no
+ conversion of values, they just change the type. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_castpd_ps(__m128d __A)
+{
+ return (__m128) __A;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_castpd_si128(__m128d __A)
+{
+ return (__m128i) __A;
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_castps_pd(__m128 __A)
+{
+ return (__m128d) __A;
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_castps_si128(__m128 __A)
+{
+ return (__m128i) __A;
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_castsi128_ps(__m128i __A)
+{
+ return (__m128) __A;
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_castsi128_pd(__m128i __A)
+{
+ return (__m128d) __A;
+}
+
+#else
+#include_next <emmintrin.h>
+#endif /* defined(__linux__) && defined(__ppc64__) */
+
+#endif /* EMMINTRIN_H_ */
--- /dev/null
+/*===---- mm_malloc.h - Implementation of _mm_malloc and _mm_free ----------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef _MM_MALLOC_H_INCLUDED
+#define _MM_MALLOC_H_INCLUDED
+
+#if defined(__linux__) && defined(__ppc64__)
+
+#include <stdlib.h>
+
+/* We can't depend on <stdlib.h> since the prototype of posix_memalign
+ may not be visible. */
+#ifndef __cplusplus
+extern int posix_memalign (void **, size_t, size_t);
+#else
+extern "C" int posix_memalign (void **, size_t, size_t) throw ();
+#endif
+
+static __inline void *
+_mm_malloc (size_t size, size_t alignment)
+{
+ /* PowerPC64 ELF V2 ABI requires quadword alignment. */
+ size_t vec_align = sizeof (__vector float);
+ void *ptr;
+
+ if (alignment < vec_align)
+ alignment = vec_align;
+ if (posix_memalign (&ptr, alignment, size) == 0)
+ return ptr;
+ else
+ return NULL;
+}
+
+static __inline void
+_mm_free (void * ptr)
+{
+ free (ptr);
+}
+
+#else
+#include_next <mm_malloc.h>
+#endif
+
+#endif /* _MM_MALLOC_H_INCLUDED */
--- /dev/null
+/*===---- mmintrin.h - Implementation of MMX intrinsics on PowerPC ---------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+/* Implemented from the specification included in the Intel C++ Compiler
+ User Guide and Reference, version 9.0. */
+
+#ifndef NO_WARN_X86_INTRINSICS
+/* This header file is to help porting code using Intel intrinsics
+ explicitly from x86_64 to powerpc64/powerpc64le.
+
+ Since PowerPC target doesn't support native 64-bit vector type, we
+ typedef __m64 to 64-bit unsigned long long in MMX intrinsics, which
+ works well for _si64 and some _pi32 operations.
+
+ For _pi16 and _pi8 operations, it's better to transfer __m64 into
+ 128-bit PowerPC vector first. Power8 introduced direct register
+ move instructions which helps for more efficient implementation.
+
+ It's user's responsibility to determine if the results of such port
+ are acceptable or further changes are needed. Please note that much
+ code using Intel intrinsics CAN BE REWRITTEN in more portable and
+ efficient standard C or GNU C extensions with 64-bit scalar
+ operations, or 128-bit SSE/Altivec operations, which are more
+ recommended. */
+#error \
+ "Please read comment above. Use -DNO_WARN_X86_INTRINSICS to disable this error."
+#endif
+
+#ifndef _MMINTRIN_H_INCLUDED
+#define _MMINTRIN_H_INCLUDED
+
+#if defined(__linux__) && defined(__ppc64__)
+
+#include <altivec.h>
+/* The Intel API is flexible enough that we must allow aliasing with other
+ vector types, and their scalar components. */
+typedef __attribute__((__aligned__(8))) unsigned long long __m64;
+
+typedef __attribute__((__aligned__(8))) union {
+ __m64 as_m64;
+ char as_char[8];
+ signed char as_signed_char[8];
+ short as_short[4];
+ int as_int[2];
+ long long as_long_long;
+ float as_float[2];
+ double as_double;
+} __m64_union;
+
+/* Empty the multimedia state. */
+extern __inline void
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_empty(void) {
+ /* nothing to do on PowerPC. */
+}
+
+extern __inline void
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_empty(void) {
+ /* nothing to do on PowerPC. */
+}
+
+/* Convert I to a __m64 object. The integer is zero-extended to 64-bits. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_cvtsi32_si64(int __i) {
+ return (__m64)(unsigned int)__i;
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_from_int(int __i) {
+ return _mm_cvtsi32_si64(__i);
+}
+
+/* Convert the lower 32 bits of the __m64 object into an integer. */
+extern __inline int
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_cvtsi64_si32(__m64 __i) {
+ return ((int)__i);
+}
+
+extern __inline int
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_to_int(__m64 __i) {
+ return _mm_cvtsi64_si32(__i);
+}
+
+/* Convert I to a __m64 object. */
+
+/* Intel intrinsic. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_from_int64(long long __i) {
+ return (__m64)__i;
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_cvtsi64_m64(long long __i) {
+ return (__m64)__i;
+}
+
+/* Microsoft intrinsic. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_cvtsi64x_si64(long long __i) {
+ return (__m64)__i;
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_set_pi64x(long long __i) {
+ return (__m64)__i;
+}
+
+/* Convert the __m64 object to a 64bit integer. */
+
+/* Intel intrinsic. */
+extern __inline long long
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_to_int64(__m64 __i) {
+ return (long long)__i;
+}
+
+extern __inline long long
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_cvtm64_si64(__m64 __i) {
+ return (long long)__i;
+}
+
+/* Microsoft intrinsic. */
+extern __inline long long
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_cvtsi64_si64x(__m64 __i) {
+ return (long long)__i;
+}
+
+#ifdef _ARCH_PWR8
+/* Pack the four 16-bit values from M1 into the lower four 8-bit values of
+ the result, and the four 16-bit values from M2 into the upper four 8-bit
+ values of the result, all with signed saturation. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_packs_pi16(__m64 __m1, __m64 __m2) {
+ __vector signed short vm1;
+ __vector signed char vresult;
+
+ vm1 = (__vector signed short)(__vector unsigned long long)
+#ifdef __LITTLE_ENDIAN__
+ {__m1, __m2};
+#else
+ {__m2, __m1};
+#endif
+ vresult = vec_packs(vm1, vm1);
+ return (__m64)((__vector long long)vresult)[0];
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_packsswb(__m64 __m1, __m64 __m2) {
+ return _mm_packs_pi16(__m1, __m2);
+}
+
+/* Pack the two 32-bit values from M1 in to the lower two 16-bit values of
+ the result, and the two 32-bit values from M2 into the upper two 16-bit
+ values of the result, all with signed saturation. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_packs_pi32(__m64 __m1, __m64 __m2) {
+ __vector signed int vm1;
+ __vector signed short vresult;
+
+ vm1 = (__vector signed int)(__vector unsigned long long)
+#ifdef __LITTLE_ENDIAN__
+ {__m1, __m2};
+#else
+ {__m2, __m1};
+#endif
+ vresult = vec_packs(vm1, vm1);
+ return (__m64)((__vector long long)vresult)[0];
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_packssdw(__m64 __m1, __m64 __m2) {
+ return _mm_packs_pi32(__m1, __m2);
+}
+
+/* Pack the four 16-bit values from M1 into the lower four 8-bit values of
+ the result, and the four 16-bit values from M2 into the upper four 8-bit
+ values of the result, all with unsigned saturation. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_packs_pu16(__m64 __m1, __m64 __m2) {
+ __vector unsigned char r;
+ __vector signed short vm1 = (__vector signed short)(__vector long long)
+#ifdef __LITTLE_ENDIAN__
+ {__m1, __m2};
+#else
+ {__m2, __m1};
+#endif
+ const __vector signed short __zero = {0};
+ __vector __bool short __select = vec_cmplt(vm1, __zero);
+ r = vec_packs((__vector unsigned short)vm1, (__vector unsigned short)vm1);
+ __vector __bool char packsel = vec_pack(__select, __select);
+ r = vec_sel(r, (const __vector unsigned char)__zero, packsel);
+ return (__m64)((__vector long long)r)[0];
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_packuswb(__m64 __m1, __m64 __m2) {
+ return _mm_packs_pu16(__m1, __m2);
+}
+#endif /* end ARCH_PWR8 */
+
+/* Interleave the four 8-bit values from the high half of M1 with the four
+ 8-bit values from the high half of M2. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_unpackhi_pi8(__m64 __m1, __m64 __m2) {
+#if _ARCH_PWR8
+ __vector unsigned char a, b, c;
+
+ a = (__vector unsigned char)vec_splats(__m1);
+ b = (__vector unsigned char)vec_splats(__m2);
+ c = vec_mergel(a, b);
+ return (__m64)((__vector long long)c)[1];
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_char[0] = m1.as_char[4];
+ res.as_char[1] = m2.as_char[4];
+ res.as_char[2] = m1.as_char[5];
+ res.as_char[3] = m2.as_char[5];
+ res.as_char[4] = m1.as_char[6];
+ res.as_char[5] = m2.as_char[6];
+ res.as_char[6] = m1.as_char[7];
+ res.as_char[7] = m2.as_char[7];
+
+ return (__m64)res.as_m64;
+#endif
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_punpckhbw(__m64 __m1, __m64 __m2) {
+ return _mm_unpackhi_pi8(__m1, __m2);
+}
+
+/* Interleave the two 16-bit values from the high half of M1 with the two
+ 16-bit values from the high half of M2. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_unpackhi_pi16(__m64 __m1, __m64 __m2) {
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_short[0] = m1.as_short[2];
+ res.as_short[1] = m2.as_short[2];
+ res.as_short[2] = m1.as_short[3];
+ res.as_short[3] = m2.as_short[3];
+
+ return (__m64)res.as_m64;
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_punpckhwd(__m64 __m1, __m64 __m2) {
+ return _mm_unpackhi_pi16(__m1, __m2);
+}
+/* Interleave the 32-bit value from the high half of M1 with the 32-bit
+ value from the high half of M2. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_unpackhi_pi32(__m64 __m1, __m64 __m2) {
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_int[0] = m1.as_int[1];
+ res.as_int[1] = m2.as_int[1];
+
+ return (__m64)res.as_m64;
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_punpckhdq(__m64 __m1, __m64 __m2) {
+ return _mm_unpackhi_pi32(__m1, __m2);
+}
+/* Interleave the four 8-bit values from the low half of M1 with the four
+ 8-bit values from the low half of M2. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_unpacklo_pi8(__m64 __m1, __m64 __m2) {
+#if _ARCH_PWR8
+ __vector unsigned char a, b, c;
+
+ a = (__vector unsigned char)vec_splats(__m1);
+ b = (__vector unsigned char)vec_splats(__m2);
+ c = vec_mergel(a, b);
+ return (__m64)((__vector long long)c)[0];
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_char[0] = m1.as_char[0];
+ res.as_char[1] = m2.as_char[0];
+ res.as_char[2] = m1.as_char[1];
+ res.as_char[3] = m2.as_char[1];
+ res.as_char[4] = m1.as_char[2];
+ res.as_char[5] = m2.as_char[2];
+ res.as_char[6] = m1.as_char[3];
+ res.as_char[7] = m2.as_char[3];
+
+ return (__m64)res.as_m64;
+#endif
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_punpcklbw(__m64 __m1, __m64 __m2) {
+ return _mm_unpacklo_pi8(__m1, __m2);
+}
+/* Interleave the two 16-bit values from the low half of M1 with the two
+ 16-bit values from the low half of M2. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_unpacklo_pi16(__m64 __m1, __m64 __m2) {
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_short[0] = m1.as_short[0];
+ res.as_short[1] = m2.as_short[0];
+ res.as_short[2] = m1.as_short[1];
+ res.as_short[3] = m2.as_short[1];
+
+ return (__m64)res.as_m64;
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_punpcklwd(__m64 __m1, __m64 __m2) {
+ return _mm_unpacklo_pi16(__m1, __m2);
+}
+
+/* Interleave the 32-bit value from the low half of M1 with the 32-bit
+ value from the low half of M2. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_unpacklo_pi32(__m64 __m1, __m64 __m2) {
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_int[0] = m1.as_int[0];
+ res.as_int[1] = m2.as_int[0];
+
+ return (__m64)res.as_m64;
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_punpckldq(__m64 __m1, __m64 __m2) {
+ return _mm_unpacklo_pi32(__m1, __m2);
+}
+
+/* Add the 8-bit values in M1 to the 8-bit values in M2. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_add_pi8(__m64 __m1, __m64 __m2) {
+#if _ARCH_PWR8
+ __vector signed char a, b, c;
+
+ a = (__vector signed char)vec_splats(__m1);
+ b = (__vector signed char)vec_splats(__m2);
+ c = vec_add(a, b);
+ return (__m64)((__vector long long)c)[0];
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_char[0] = m1.as_char[0] + m2.as_char[0];
+ res.as_char[1] = m1.as_char[1] + m2.as_char[1];
+ res.as_char[2] = m1.as_char[2] + m2.as_char[2];
+ res.as_char[3] = m1.as_char[3] + m2.as_char[3];
+ res.as_char[4] = m1.as_char[4] + m2.as_char[4];
+ res.as_char[5] = m1.as_char[5] + m2.as_char[5];
+ res.as_char[6] = m1.as_char[6] + m2.as_char[6];
+ res.as_char[7] = m1.as_char[7] + m2.as_char[7];
+
+ return (__m64)res.as_m64;
+#endif
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_paddb(__m64 __m1, __m64 __m2) {
+ return _mm_add_pi8(__m1, __m2);
+}
+
+/* Add the 16-bit values in M1 to the 16-bit values in M2. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_add_pi16(__m64 __m1, __m64 __m2) {
+#if _ARCH_PWR8
+ __vector signed short a, b, c;
+
+ a = (__vector signed short)vec_splats(__m1);
+ b = (__vector signed short)vec_splats(__m2);
+ c = vec_add(a, b);
+ return (__m64)((__vector long long)c)[0];
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_short[0] = m1.as_short[0] + m2.as_short[0];
+ res.as_short[1] = m1.as_short[1] + m2.as_short[1];
+ res.as_short[2] = m1.as_short[2] + m2.as_short[2];
+ res.as_short[3] = m1.as_short[3] + m2.as_short[3];
+
+ return (__m64)res.as_m64;
+#endif
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_paddw(__m64 __m1, __m64 __m2) {
+ return _mm_add_pi16(__m1, __m2);
+}
+
+/* Add the 32-bit values in M1 to the 32-bit values in M2. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_add_pi32(__m64 __m1, __m64 __m2) {
+#if _ARCH_PWR9
+ __vector signed int a, b, c;
+
+ a = (__vector signed int)vec_splats(__m1);
+ b = (__vector signed int)vec_splats(__m2);
+ c = vec_add(a, b);
+ return (__m64)((__vector long long)c)[0];
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_int[0] = m1.as_int[0] + m2.as_int[0];
+ res.as_int[1] = m1.as_int[1] + m2.as_int[1];
+
+ return (__m64)res.as_m64;
+#endif
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_paddd(__m64 __m1, __m64 __m2) {
+ return _mm_add_pi32(__m1, __m2);
+}
+
+/* Subtract the 8-bit values in M2 from the 8-bit values in M1. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_sub_pi8(__m64 __m1, __m64 __m2) {
+#if _ARCH_PWR8
+ __vector signed char a, b, c;
+
+ a = (__vector signed char)vec_splats(__m1);
+ b = (__vector signed char)vec_splats(__m2);
+ c = vec_sub(a, b);
+ return (__m64)((__vector long long)c)[0];
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_char[0] = m1.as_char[0] - m2.as_char[0];
+ res.as_char[1] = m1.as_char[1] - m2.as_char[1];
+ res.as_char[2] = m1.as_char[2] - m2.as_char[2];
+ res.as_char[3] = m1.as_char[3] - m2.as_char[3];
+ res.as_char[4] = m1.as_char[4] - m2.as_char[4];
+ res.as_char[5] = m1.as_char[5] - m2.as_char[5];
+ res.as_char[6] = m1.as_char[6] - m2.as_char[6];
+ res.as_char[7] = m1.as_char[7] - m2.as_char[7];
+
+ return (__m64)res.as_m64;
+#endif
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psubb(__m64 __m1, __m64 __m2) {
+ return _mm_sub_pi8(__m1, __m2);
+}
+
+/* Subtract the 16-bit values in M2 from the 16-bit values in M1. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_sub_pi16(__m64 __m1, __m64 __m2) {
+#if _ARCH_PWR8
+ __vector signed short a, b, c;
+
+ a = (__vector signed short)vec_splats(__m1);
+ b = (__vector signed short)vec_splats(__m2);
+ c = vec_sub(a, b);
+ return (__m64)((__vector long long)c)[0];
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_short[0] = m1.as_short[0] - m2.as_short[0];
+ res.as_short[1] = m1.as_short[1] - m2.as_short[1];
+ res.as_short[2] = m1.as_short[2] - m2.as_short[2];
+ res.as_short[3] = m1.as_short[3] - m2.as_short[3];
+
+ return (__m64)res.as_m64;
+#endif
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psubw(__m64 __m1, __m64 __m2) {
+ return _mm_sub_pi16(__m1, __m2);
+}
+
+/* Subtract the 32-bit values in M2 from the 32-bit values in M1. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_sub_pi32(__m64 __m1, __m64 __m2) {
+#if _ARCH_PWR9
+ __vector signed int a, b, c;
+
+ a = (__vector signed int)vec_splats(__m1);
+ b = (__vector signed int)vec_splats(__m2);
+ c = vec_sub(a, b);
+ return (__m64)((__vector long long)c)[0];
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_int[0] = m1.as_int[0] - m2.as_int[0];
+ res.as_int[1] = m1.as_int[1] - m2.as_int[1];
+
+ return (__m64)res.as_m64;
+#endif
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psubd(__m64 __m1, __m64 __m2) {
+ return _mm_sub_pi32(__m1, __m2);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_add_si64(__m64 __m1, __m64 __m2) {
+ return (__m1 + __m2);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_sub_si64(__m64 __m1, __m64 __m2) {
+ return (__m1 - __m2);
+}
+
+/* Shift the 64-bit value in M left by COUNT. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_sll_si64(__m64 __m, __m64 __count) {
+ return (__m << __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psllq(__m64 __m, __m64 __count) {
+ return _mm_sll_si64(__m, __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_slli_si64(__m64 __m, const int __count) {
+ return (__m << __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psllqi(__m64 __m, const int __count) {
+ return _mm_slli_si64(__m, __count);
+}
+
+/* Shift the 64-bit value in M left by COUNT; shift in zeros. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_srl_si64(__m64 __m, __m64 __count) {
+ return (__m >> __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psrlq(__m64 __m, __m64 __count) {
+ return _mm_srl_si64(__m, __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_srli_si64(__m64 __m, const int __count) {
+ return (__m >> __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psrlqi(__m64 __m, const int __count) {
+ return _mm_srli_si64(__m, __count);
+}
+
+/* Bit-wise AND the 64-bit values in M1 and M2. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_and_si64(__m64 __m1, __m64 __m2) {
+ return (__m1 & __m2);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_pand(__m64 __m1, __m64 __m2) {
+ return _mm_and_si64(__m1, __m2);
+}
+
+/* Bit-wise complement the 64-bit value in M1 and bit-wise AND it with the
+ 64-bit value in M2. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_andnot_si64(__m64 __m1, __m64 __m2) {
+ return (~__m1 & __m2);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_pandn(__m64 __m1, __m64 __m2) {
+ return _mm_andnot_si64(__m1, __m2);
+}
+
+/* Bit-wise inclusive OR the 64-bit values in M1 and M2. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_or_si64(__m64 __m1, __m64 __m2) {
+ return (__m1 | __m2);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_por(__m64 __m1, __m64 __m2) {
+ return _mm_or_si64(__m1, __m2);
+}
+
+/* Bit-wise exclusive OR the 64-bit values in M1 and M2. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_xor_si64(__m64 __m1, __m64 __m2) {
+ return (__m1 ^ __m2);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_pxor(__m64 __m1, __m64 __m2) {
+ return _mm_xor_si64(__m1, __m2);
+}
+
+/* Creates a 64-bit zero. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_setzero_si64(void) {
+ return (__m64)0;
+}
+
+/* Compare eight 8-bit values. The result of the comparison is 0xFF if the
+ test is true and zero if false. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_cmpeq_pi8(__m64 __m1, __m64 __m2) {
+#if defined(_ARCH_PWR6) && defined(__powerpc64__)
+ __m64 res;
+ __asm__("cmpb %0,%1,%2;\n" : "=r"(res) : "r"(__m1), "r"(__m2) :);
+ return (res);
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_char[0] = (m1.as_char[0] == m2.as_char[0]) ? -1 : 0;
+ res.as_char[1] = (m1.as_char[1] == m2.as_char[1]) ? -1 : 0;
+ res.as_char[2] = (m1.as_char[2] == m2.as_char[2]) ? -1 : 0;
+ res.as_char[3] = (m1.as_char[3] == m2.as_char[3]) ? -1 : 0;
+ res.as_char[4] = (m1.as_char[4] == m2.as_char[4]) ? -1 : 0;
+ res.as_char[5] = (m1.as_char[5] == m2.as_char[5]) ? -1 : 0;
+ res.as_char[6] = (m1.as_char[6] == m2.as_char[6]) ? -1 : 0;
+ res.as_char[7] = (m1.as_char[7] == m2.as_char[7]) ? -1 : 0;
+
+ return (__m64)res.as_m64;
+#endif
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_pcmpeqb(__m64 __m1, __m64 __m2) {
+ return _mm_cmpeq_pi8(__m1, __m2);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_cmpgt_pi8(__m64 __m1, __m64 __m2) {
+#if _ARCH_PWR8
+ __vector signed char a, b, c;
+
+ a = (__vector signed char)vec_splats(__m1);
+ b = (__vector signed char)vec_splats(__m2);
+ c = (__vector signed char)vec_cmpgt(a, b);
+ return (__m64)((__vector long long)c)[0];
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_char[0] = (m1.as_char[0] > m2.as_char[0]) ? -1 : 0;
+ res.as_char[1] = (m1.as_char[1] > m2.as_char[1]) ? -1 : 0;
+ res.as_char[2] = (m1.as_char[2] > m2.as_char[2]) ? -1 : 0;
+ res.as_char[3] = (m1.as_char[3] > m2.as_char[3]) ? -1 : 0;
+ res.as_char[4] = (m1.as_char[4] > m2.as_char[4]) ? -1 : 0;
+ res.as_char[5] = (m1.as_char[5] > m2.as_char[5]) ? -1 : 0;
+ res.as_char[6] = (m1.as_char[6] > m2.as_char[6]) ? -1 : 0;
+ res.as_char[7] = (m1.as_char[7] > m2.as_char[7]) ? -1 : 0;
+
+ return (__m64)res.as_m64;
+#endif
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_pcmpgtb(__m64 __m1, __m64 __m2) {
+ return _mm_cmpgt_pi8(__m1, __m2);
+}
+
+/* Compare four 16-bit values. The result of the comparison is 0xFFFF if
+ the test is true and zero if false. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_cmpeq_pi16(__m64 __m1, __m64 __m2) {
+#if _ARCH_PWR8
+ __vector signed short a, b, c;
+
+ a = (__vector signed short)vec_splats(__m1);
+ b = (__vector signed short)vec_splats(__m2);
+ c = (__vector signed short)vec_cmpeq(a, b);
+ return (__m64)((__vector long long)c)[0];
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_short[0] = (m1.as_short[0] == m2.as_short[0]) ? -1 : 0;
+ res.as_short[1] = (m1.as_short[1] == m2.as_short[1]) ? -1 : 0;
+ res.as_short[2] = (m1.as_short[2] == m2.as_short[2]) ? -1 : 0;
+ res.as_short[3] = (m1.as_short[3] == m2.as_short[3]) ? -1 : 0;
+
+ return (__m64)res.as_m64;
+#endif
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_pcmpeqw(__m64 __m1, __m64 __m2) {
+ return _mm_cmpeq_pi16(__m1, __m2);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_cmpgt_pi16(__m64 __m1, __m64 __m2) {
+#if _ARCH_PWR8
+ __vector signed short a, b, c;
+
+ a = (__vector signed short)vec_splats(__m1);
+ b = (__vector signed short)vec_splats(__m2);
+ c = (__vector signed short)vec_cmpgt(a, b);
+ return (__m64)((__vector long long)c)[0];
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_short[0] = (m1.as_short[0] > m2.as_short[0]) ? -1 : 0;
+ res.as_short[1] = (m1.as_short[1] > m2.as_short[1]) ? -1 : 0;
+ res.as_short[2] = (m1.as_short[2] > m2.as_short[2]) ? -1 : 0;
+ res.as_short[3] = (m1.as_short[3] > m2.as_short[3]) ? -1 : 0;
+
+ return (__m64)res.as_m64;
+#endif
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_pcmpgtw(__m64 __m1, __m64 __m2) {
+ return _mm_cmpgt_pi16(__m1, __m2);
+}
+
+/* Compare two 32-bit values. The result of the comparison is 0xFFFFFFFF if
+ the test is true and zero if false. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_cmpeq_pi32(__m64 __m1, __m64 __m2) {
+#if _ARCH_PWR9
+ __vector signed int a, b, c;
+
+ a = (__vector signed int)vec_splats(__m1);
+ b = (__vector signed int)vec_splats(__m2);
+ c = (__vector signed int)vec_cmpeq(a, b);
+ return (__m64)((__vector long long)c)[0];
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_int[0] = (m1.as_int[0] == m2.as_int[0]) ? -1 : 0;
+ res.as_int[1] = (m1.as_int[1] == m2.as_int[1]) ? -1 : 0;
+
+ return (__m64)res.as_m64;
+#endif
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_pcmpeqd(__m64 __m1, __m64 __m2) {
+ return _mm_cmpeq_pi32(__m1, __m2);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_cmpgt_pi32(__m64 __m1, __m64 __m2) {
+#if _ARCH_PWR9
+ __vector signed int a, b, c;
+
+ a = (__vector signed int)vec_splats(__m1);
+ b = (__vector signed int)vec_splats(__m2);
+ c = (__vector signed int)vec_cmpgt(a, b);
+ return (__m64)((__vector long long)c)[0];
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __m1;
+ m2.as_m64 = __m2;
+
+ res.as_int[0] = (m1.as_int[0] > m2.as_int[0]) ? -1 : 0;
+ res.as_int[1] = (m1.as_int[1] > m2.as_int[1]) ? -1 : 0;
+
+ return (__m64)res.as_m64;
+#endif
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_pcmpgtd(__m64 __m1, __m64 __m2) {
+ return _mm_cmpgt_pi32(__m1, __m2);
+}
+
+#if _ARCH_PWR8
+/* Add the 8-bit values in M1 to the 8-bit values in M2 using signed
+ saturated arithmetic. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_adds_pi8(__m64 __m1, __m64 __m2) {
+ __vector signed char a, b, c;
+
+ a = (__vector signed char)vec_splats(__m1);
+ b = (__vector signed char)vec_splats(__m2);
+ c = vec_adds(a, b);
+ return (__m64)((__vector long long)c)[0];
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_paddsb(__m64 __m1, __m64 __m2) {
+ return _mm_adds_pi8(__m1, __m2);
+}
+/* Add the 16-bit values in M1 to the 16-bit values in M2 using signed
+ saturated arithmetic. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_adds_pi16(__m64 __m1, __m64 __m2) {
+ __vector signed short a, b, c;
+
+ a = (__vector signed short)vec_splats(__m1);
+ b = (__vector signed short)vec_splats(__m2);
+ c = vec_adds(a, b);
+ return (__m64)((__vector long long)c)[0];
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_paddsw(__m64 __m1, __m64 __m2) {
+ return _mm_adds_pi16(__m1, __m2);
+}
+/* Add the 8-bit values in M1 to the 8-bit values in M2 using unsigned
+ saturated arithmetic. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_adds_pu8(__m64 __m1, __m64 __m2) {
+ __vector unsigned char a, b, c;
+
+ a = (__vector unsigned char)vec_splats(__m1);
+ b = (__vector unsigned char)vec_splats(__m2);
+ c = vec_adds(a, b);
+ return (__m64)((__vector long long)c)[0];
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_paddusb(__m64 __m1, __m64 __m2) {
+ return _mm_adds_pu8(__m1, __m2);
+}
+
+/* Add the 16-bit values in M1 to the 16-bit values in M2 using unsigned
+ saturated arithmetic. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_adds_pu16(__m64 __m1, __m64 __m2) {
+ __vector unsigned short a, b, c;
+
+ a = (__vector unsigned short)vec_splats(__m1);
+ b = (__vector unsigned short)vec_splats(__m2);
+ c = vec_adds(a, b);
+ return (__m64)((__vector long long)c)[0];
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_paddusw(__m64 __m1, __m64 __m2) {
+ return _mm_adds_pu16(__m1, __m2);
+}
+
+/* Subtract the 8-bit values in M2 from the 8-bit values in M1 using signed
+ saturating arithmetic. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_subs_pi8(__m64 __m1, __m64 __m2) {
+ __vector signed char a, b, c;
+
+ a = (__vector signed char)vec_splats(__m1);
+ b = (__vector signed char)vec_splats(__m2);
+ c = vec_subs(a, b);
+ return (__m64)((__vector long long)c)[0];
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psubsb(__m64 __m1, __m64 __m2) {
+ return _mm_subs_pi8(__m1, __m2);
+}
+
+/* Subtract the 16-bit values in M2 from the 16-bit values in M1 using
+ signed saturating arithmetic. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_subs_pi16(__m64 __m1, __m64 __m2) {
+ __vector signed short a, b, c;
+
+ a = (__vector signed short)vec_splats(__m1);
+ b = (__vector signed short)vec_splats(__m2);
+ c = vec_subs(a, b);
+ return (__m64)((__vector long long)c)[0];
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psubsw(__m64 __m1, __m64 __m2) {
+ return _mm_subs_pi16(__m1, __m2);
+}
+
+/* Subtract the 8-bit values in M2 from the 8-bit values in M1 using
+ unsigned saturating arithmetic. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_subs_pu8(__m64 __m1, __m64 __m2) {
+ __vector unsigned char a, b, c;
+
+ a = (__vector unsigned char)vec_splats(__m1);
+ b = (__vector unsigned char)vec_splats(__m2);
+ c = vec_subs(a, b);
+ return (__m64)((__vector long long)c)[0];
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psubusb(__m64 __m1, __m64 __m2) {
+ return _mm_subs_pu8(__m1, __m2);
+}
+
+/* Subtract the 16-bit values in M2 from the 16-bit values in M1 using
+ unsigned saturating arithmetic. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_subs_pu16(__m64 __m1, __m64 __m2) {
+ __vector unsigned short a, b, c;
+
+ a = (__vector unsigned short)vec_splats(__m1);
+ b = (__vector unsigned short)vec_splats(__m2);
+ c = vec_subs(a, b);
+ return (__m64)((__vector long long)c)[0];
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psubusw(__m64 __m1, __m64 __m2) {
+ return _mm_subs_pu16(__m1, __m2);
+}
+
+/* Multiply four 16-bit values in M1 by four 16-bit values in M2 producing
+ four 32-bit intermediate results, which are then summed by pairs to
+ produce two 32-bit results. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_madd_pi16(__m64 __m1, __m64 __m2) {
+ __vector signed short a, b;
+ __vector signed int c;
+ __vector signed int zero = {0, 0, 0, 0};
+
+ a = (__vector signed short)vec_splats(__m1);
+ b = (__vector signed short)vec_splats(__m2);
+ c = vec_vmsumshm(a, b, zero);
+ return (__m64)((__vector long long)c)[0];
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_pmaddwd(__m64 __m1, __m64 __m2) {
+ return _mm_madd_pi16(__m1, __m2);
+}
+/* Multiply four signed 16-bit values in M1 by four signed 16-bit values in
+ M2 and produce the high 16 bits of the 32-bit results. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_mulhi_pi16(__m64 __m1, __m64 __m2) {
+ __vector signed short a, b;
+ __vector signed short c;
+ __vector signed int w0, w1;
+ __vector unsigned char xform1 = {
+#ifdef __LITTLE_ENDIAN__
+ 0x02, 0x03, 0x12, 0x13, 0x06, 0x07, 0x16, 0x17, 0x0A,
+ 0x0B, 0x1A, 0x1B, 0x0E, 0x0F, 0x1E, 0x1F
+#else
+ 0x00, 0x01, 0x10, 0x11, 0x04, 0x05, 0x14, 0x15, 0x00,
+ 0x01, 0x10, 0x11, 0x04, 0x05, 0x14, 0x15
+#endif
+ };
+
+ a = (__vector signed short)vec_splats(__m1);
+ b = (__vector signed short)vec_splats(__m2);
+
+ w0 = vec_vmulesh(a, b);
+ w1 = vec_vmulosh(a, b);
+ c = (__vector signed short)vec_perm(w0, w1, xform1);
+
+ return (__m64)((__vector long long)c)[0];
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_pmulhw(__m64 __m1, __m64 __m2) {
+ return _mm_mulhi_pi16(__m1, __m2);
+}
+
+/* Multiply four 16-bit values in M1 by four 16-bit values in M2 and produce
+ the low 16 bits of the results. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_mullo_pi16(__m64 __m1, __m64 __m2) {
+ __vector signed short a, b, c;
+
+ a = (__vector signed short)vec_splats(__m1);
+ b = (__vector signed short)vec_splats(__m2);
+ c = a * b;
+ return (__m64)((__vector long long)c)[0];
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_pmullw(__m64 __m1, __m64 __m2) {
+ return _mm_mullo_pi16(__m1, __m2);
+}
+
+/* Shift four 16-bit values in M left by COUNT. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_sll_pi16(__m64 __m, __m64 __count) {
+ __vector signed short m, r;
+ __vector unsigned short c;
+
+ if (__count <= 15) {
+ m = (__vector signed short)vec_splats(__m);
+ c = (__vector unsigned short)vec_splats((unsigned short)__count);
+ r = vec_sl(m, (__vector unsigned short)c);
+ return (__m64)((__vector long long)r)[0];
+ } else
+ return (0);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psllw(__m64 __m, __m64 __count) {
+ return _mm_sll_pi16(__m, __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_slli_pi16(__m64 __m, int __count) {
+ /* Promote int to long then invoke mm_sll_pi16. */
+ return _mm_sll_pi16(__m, __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psllwi(__m64 __m, int __count) {
+ return _mm_slli_pi16(__m, __count);
+}
+
+/* Shift two 32-bit values in M left by COUNT. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_sll_pi32(__m64 __m, __m64 __count) {
+ __m64_union m, res;
+
+ m.as_m64 = __m;
+
+ res.as_int[0] = m.as_int[0] << __count;
+ res.as_int[1] = m.as_int[1] << __count;
+ return (res.as_m64);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_pslld(__m64 __m, __m64 __count) {
+ return _mm_sll_pi32(__m, __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_slli_pi32(__m64 __m, int __count) {
+ /* Promote int to long then invoke mm_sll_pi32. */
+ return _mm_sll_pi32(__m, __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_pslldi(__m64 __m, int __count) {
+ return _mm_slli_pi32(__m, __count);
+}
+
+/* Shift four 16-bit values in M right by COUNT; shift in the sign bit. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_sra_pi16(__m64 __m, __m64 __count) {
+ __vector signed short m, r;
+ __vector unsigned short c;
+
+ if (__count <= 15) {
+ m = (__vector signed short)vec_splats(__m);
+ c = (__vector unsigned short)vec_splats((unsigned short)__count);
+ r = vec_sra(m, (__vector unsigned short)c);
+ return (__m64)((__vector long long)r)[0];
+ } else
+ return (0);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psraw(__m64 __m, __m64 __count) {
+ return _mm_sra_pi16(__m, __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_srai_pi16(__m64 __m, int __count) {
+ /* Promote int to long then invoke mm_sra_pi32. */
+ return _mm_sra_pi16(__m, __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psrawi(__m64 __m, int __count) {
+ return _mm_srai_pi16(__m, __count);
+}
+
+/* Shift two 32-bit values in M right by COUNT; shift in the sign bit. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_sra_pi32(__m64 __m, __m64 __count) {
+ __m64_union m, res;
+
+ m.as_m64 = __m;
+
+ res.as_int[0] = m.as_int[0] >> __count;
+ res.as_int[1] = m.as_int[1] >> __count;
+ return (res.as_m64);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psrad(__m64 __m, __m64 __count) {
+ return _mm_sra_pi32(__m, __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_srai_pi32(__m64 __m, int __count) {
+ /* Promote int to long then invoke mm_sra_pi32. */
+ return _mm_sra_pi32(__m, __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psradi(__m64 __m, int __count) {
+ return _mm_srai_pi32(__m, __count);
+}
+
+/* Shift four 16-bit values in M right by COUNT; shift in zeros. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_srl_pi16(__m64 __m, __m64 __count) {
+ __vector unsigned short m, r;
+ __vector unsigned short c;
+
+ if (__count <= 15) {
+ m = (__vector unsigned short)vec_splats(__m);
+ c = (__vector unsigned short)vec_splats((unsigned short)__count);
+ r = vec_sr(m, (__vector unsigned short)c);
+ return (__m64)((__vector long long)r)[0];
+ } else
+ return (0);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psrlw(__m64 __m, __m64 __count) {
+ return _mm_srl_pi16(__m, __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_srli_pi16(__m64 __m, int __count) {
+ /* Promote int to long then invoke mm_sra_pi32. */
+ return _mm_srl_pi16(__m, __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psrlwi(__m64 __m, int __count) {
+ return _mm_srli_pi16(__m, __count);
+}
+
+/* Shift two 32-bit values in M right by COUNT; shift in zeros. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_srl_pi32(__m64 __m, __m64 __count) {
+ __m64_union m, res;
+
+ m.as_m64 = __m;
+
+ res.as_int[0] = (unsigned int)m.as_int[0] >> __count;
+ res.as_int[1] = (unsigned int)m.as_int[1] >> __count;
+ return (res.as_m64);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psrld(__m64 __m, __m64 __count) {
+ return _mm_srl_pi32(__m, __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_srli_pi32(__m64 __m, int __count) {
+ /* Promote int to long then invoke mm_srl_pi32. */
+ return _mm_srl_pi32(__m, __count);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _m_psrldi(__m64 __m, int __count) {
+ return _mm_srli_pi32(__m, __count);
+}
+#endif /* _ARCH_PWR8 */
+
+/* Creates a vector of two 32-bit values; I0 is least significant. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_set_pi32(int __i1, int __i0) {
+ __m64_union res;
+
+ res.as_int[0] = __i0;
+ res.as_int[1] = __i1;
+ return (res.as_m64);
+}
+
+/* Creates a vector of four 16-bit values; W0 is least significant. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_set_pi16(short __w3, short __w2, short __w1, short __w0) {
+ __m64_union res;
+
+ res.as_short[0] = __w0;
+ res.as_short[1] = __w1;
+ res.as_short[2] = __w2;
+ res.as_short[3] = __w3;
+ return (res.as_m64);
+}
+
+/* Creates a vector of eight 8-bit values; B0 is least significant. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_set_pi8(char __b7, char __b6, char __b5, char __b4, char __b3,
+ char __b2, char __b1, char __b0) {
+ __m64_union res;
+
+ res.as_char[0] = __b0;
+ res.as_char[1] = __b1;
+ res.as_char[2] = __b2;
+ res.as_char[3] = __b3;
+ res.as_char[4] = __b4;
+ res.as_char[5] = __b5;
+ res.as_char[6] = __b6;
+ res.as_char[7] = __b7;
+ return (res.as_m64);
+}
+
+/* Similar, but with the arguments in reverse order. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_setr_pi32(int __i0, int __i1) {
+ __m64_union res;
+
+ res.as_int[0] = __i0;
+ res.as_int[1] = __i1;
+ return (res.as_m64);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_setr_pi16(short __w0, short __w1, short __w2, short __w3) {
+ return _mm_set_pi16(__w3, __w2, __w1, __w0);
+}
+
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_setr_pi8(char __b0, char __b1, char __b2, char __b3, char __b4,
+ char __b5, char __b6, char __b7) {
+ return _mm_set_pi8(__b7, __b6, __b5, __b4, __b3, __b2, __b1, __b0);
+}
+
+/* Creates a vector of two 32-bit values, both elements containing I. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_set1_pi32(int __i) {
+ __m64_union res;
+
+ res.as_int[0] = __i;
+ res.as_int[1] = __i;
+ return (res.as_m64);
+}
+
+/* Creates a vector of four 16-bit values, all elements containing W. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_set1_pi16(short __w) {
+#if _ARCH_PWR9
+ __vector signed short w;
+
+ w = (__vector signed short)vec_splats(__w);
+ return (__m64)((__vector long long)w)[0];
+#else
+ __m64_union res;
+
+ res.as_short[0] = __w;
+ res.as_short[1] = __w;
+ res.as_short[2] = __w;
+ res.as_short[3] = __w;
+ return (res.as_m64);
+#endif
+}
+
+/* Creates a vector of eight 8-bit values, all elements containing B. */
+extern __inline __m64
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_set1_pi8(signed char __b) {
+#if _ARCH_PWR8
+ __vector signed char b;
+
+ b = (__vector signed char)vec_splats(__b);
+ return (__m64)((__vector long long)b)[0];
+#else
+ __m64_union res;
+
+ res.as_char[0] = __b;
+ res.as_char[1] = __b;
+ res.as_char[2] = __b;
+ res.as_char[3] = __b;
+ res.as_char[4] = __b;
+ res.as_char[5] = __b;
+ res.as_char[6] = __b;
+ res.as_char[7] = __b;
+ return (res.as_m64);
+#endif
+}
+
+#else
+#include_next <mmintrin.h>
+#endif /* defined(__linux__) && defined(__ppc64__) */
+
+#endif /* _MMINTRIN_H_INCLUDED */
--- /dev/null
+/*===---- pmmintrin.h - Implementation of SSE3 intrinsics on PowerPC -------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+/* Implemented from the specification included in the Intel C++ Compiler
+ User Guide and Reference, version 9.0. */
+
+#ifndef NO_WARN_X86_INTRINSICS
+/* This header is distributed to simplify porting x86_64 code that
+ makes explicit use of Intel intrinsics to powerpc64le.
+ It is the user's responsibility to determine if the results are
+ acceptable and make additional changes as necessary.
+ Note that much code that uses Intel intrinsics can be rewritten in
+ standard C or GNU C extensions, which are more portable and better
+ optimized across multiple targets.
+
+ In the specific case of X86 SSE3 intrinsics, the PowerPC VMX/VSX ISA
+ is a good match for most SIMD operations. However the Horizontal
+ add/sub requires the data pairs be permuted into a separate
+ registers with vertical even/odd alignment for the operation.
+ And the addsub operation requires the sign of only the even numbered
+ elements be flipped (xored with -0.0).
+ For larger blocks of code using these intrinsic implementations,
+ the compiler be should be able to schedule instructions to avoid
+ additional latency.
+
+ In the specific case of the monitor and mwait instructions there are
+ no direct equivalent in the PowerISA at this time. So those
+ intrinsics are not implemented. */
+#error "Please read comment above. Use -DNO_WARN_X86_INTRINSICS to disable this warning."
+#endif
+
+#ifndef PMMINTRIN_H_
+#define PMMINTRIN_H_
+
+#if defined(__linux__) && defined(__ppc64__)
+
+/* We need definitions from the SSE2 and SSE header files*/
+#include <emmintrin.h>
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_addsub_ps (__m128 __X, __m128 __Y)
+{
+ const __v4sf even_n0 = {-0.0, 0.0, -0.0, 0.0};
+ __v4sf even_neg_Y = vec_xor(__Y, even_n0);
+ return (__m128) vec_add (__X, even_neg_Y);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_addsub_pd (__m128d __X, __m128d __Y)
+{
+ const __v2df even_n0 = {-0.0, 0.0};
+ __v2df even_neg_Y = vec_xor(__Y, even_n0);
+ return (__m128d) vec_add (__X, even_neg_Y);
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hadd_ps (__m128 __X, __m128 __Y)
+{
+ __vector unsigned char xform2 = {
+ 0x00, 0x01, 0x02, 0x03,
+ 0x08, 0x09, 0x0A, 0x0B,
+ 0x10, 0x11, 0x12, 0x13,
+ 0x18, 0x19, 0x1A, 0x1B
+ };
+ __vector unsigned char xform1 = {
+ 0x04, 0x05, 0x06, 0x07,
+ 0x0C, 0x0D, 0x0E, 0x0F,
+ 0x14, 0x15, 0x16, 0x17,
+ 0x1C, 0x1D, 0x1E, 0x1F
+ };
+ return (__m128) vec_add (vec_perm ((__v4sf) __X, (__v4sf) __Y, xform2),
+ vec_perm ((__v4sf) __X, (__v4sf) __Y, xform1));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hsub_ps (__m128 __X, __m128 __Y)
+{
+ __vector unsigned char xform2 = {
+ 0x00, 0x01, 0x02, 0x03,
+ 0x08, 0x09, 0x0A, 0x0B,
+ 0x10, 0x11, 0x12, 0x13,
+ 0x18, 0x19, 0x1A, 0x1B
+ };
+ __vector unsigned char xform1 = {
+ 0x04, 0x05, 0x06, 0x07,
+ 0x0C, 0x0D, 0x0E, 0x0F,
+ 0x14, 0x15, 0x16, 0x17,
+ 0x1C, 0x1D, 0x1E, 0x1F
+ };
+ return (__m128) vec_sub (vec_perm ((__v4sf) __X, (__v4sf) __Y, xform2),
+ vec_perm ((__v4sf) __X, (__v4sf) __Y, xform1));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hadd_pd (__m128d __X, __m128d __Y)
+{
+ return (__m128d) vec_add (vec_mergeh ((__v2df) __X, (__v2df)__Y),
+ vec_mergel ((__v2df) __X, (__v2df)__Y));
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hsub_pd (__m128d __X, __m128d __Y)
+{
+ return (__m128d) vec_sub (vec_mergeh ((__v2df) __X, (__v2df)__Y),
+ vec_mergel ((__v2df) __X, (__v2df)__Y));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_movehdup_ps (__m128 __X)
+{
+ return (__m128)vec_mergeo ((__v4su)__X, (__v4su)__X);
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_moveldup_ps (__m128 __X)
+{
+ return (__m128)vec_mergee ((__v4su)__X, (__v4su)__X);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loaddup_pd (double const *__P)
+{
+ return (__m128d) vec_splats (*__P);
+}
+
+extern __inline __m128d __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_movedup_pd (__m128d __X)
+{
+ return _mm_shuffle_pd (__X, __X, _MM_SHUFFLE2 (0,0));
+}
+
+extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_lddqu_si128 (__m128i const *__P)
+{
+ return (__m128i) (vec_vsx_ld(0, (signed int const *)__P));
+}
+
+/* POWER8 / POWER9 have no equivalent for _mm_monitor nor _mm_wait. */
+
+#else
+#include_next <pmmintrin.h>
+#endif /* defined(__linux__) && defined(__ppc64__) */
+
+#endif /* PMMINTRIN_H_ */
--- /dev/null
+/*===---- smmintrin.h - Implementation of SSE4 intrinsics on PowerPC -------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+/* Implemented from the specification included in the Intel C++ Compiler
+ User Guide and Reference, version 9.0.
+
+ NOTE: This is NOT a complete implementation of the SSE4 intrinsics! */
+
+#ifndef NO_WARN_X86_INTRINSICS
+/* This header is distributed to simplify porting x86_64 code that
+ makes explicit use of Intel intrinsics to powerp64/powerpc64le.
+
+ It is the user's responsibility to determine if the results are
+ acceptable and make additional changes as necessary.
+
+ Note that much code that uses Intel intrinsics can be rewritten in
+ standard C or GNU C extensions, which are more portable and better
+ optimized across multiple targets. */
+#error \
+ "Please read comment above. Use -DNO_WARN_X86_INTRINSICS to disable this error."
+#endif
+
+#ifndef SMMINTRIN_H_
+#define SMMINTRIN_H_
+
+#if defined(__linux__) && defined(__ppc64__)
+
+#include <altivec.h>
+#include <emmintrin.h>
+
+extern __inline int
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_extract_epi8(__m128i __X, const int __N) {
+ return (unsigned char)((__v16qi)__X)[__N & 15];
+}
+
+extern __inline int
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_extract_epi32(__m128i __X, const int __N) {
+ return ((__v4si)__X)[__N & 3];
+}
+
+extern __inline int
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_extract_epi64(__m128i __X, const int __N) {
+ return ((__v2di)__X)[__N & 1];
+}
+
+extern __inline int
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_extract_ps(__m128 __X, const int __N) {
+ return ((__v4si)__X)[__N & 3];
+}
+
+extern __inline __m128i
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_blend_epi16(__m128i __A, __m128i __B, const int __imm8) {
+ __v16qi __charmask = vec_splats((signed char)__imm8);
+ __charmask = vec_gb(__charmask);
+ __v8hu __shortmask = (__v8hu)vec_unpackh(__charmask);
+#ifdef __BIG_ENDIAN__
+ __shortmask = vec_reve(__shortmask);
+#endif
+ return (__m128i)vec_sel((__v8hu)__A, (__v8hu)__B, __shortmask);
+}
+
+extern __inline __m128i
+ __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+ _mm_blendv_epi8(__m128i __A, __m128i __B, __m128i __mask) {
+ const __v16qu __seven = vec_splats((unsigned char)0x07);
+ __v16qu __lmask = vec_sra((__v16qu)__mask, __seven);
+ return (__m128i)vec_sel((__v16qu)__A, (__v16qu)__B, __lmask);
+}
+
+#else
+#include_next <smmintrin.h>
+#endif /* defined(__linux__) && defined(__ppc64__) */
+
+#endif /* _SMMINTRIN_H_ */
--- /dev/null
+/*===---- tmmintrin.h - Implementation of SSSE3 intrinsics on PowerPC ------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+/* Implemented from the specification included in the Intel C++ Compiler
+ User Guide and Reference, version 9.0. */
+
+#ifndef NO_WARN_X86_INTRINSICS
+/* This header is distributed to simplify porting x86_64 code that
+ makes explicit use of Intel intrinsics to powerpc64le.
+
+ It is the user's responsibility to determine if the results are
+ acceptable and make additional changes as necessary.
+
+ Note that much code that uses Intel intrinsics can be rewritten in
+ standard C or GNU C extensions, which are more portable and better
+ optimized across multiple targets. */
+#endif
+
+#ifndef TMMINTRIN_H_
+#define TMMINTRIN_H_
+
+#if defined(__linux__) && defined(__ppc64__)
+
+#include <altivec.h>
+
+/* We need definitions from the SSE header files. */
+#include <pmmintrin.h>
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_abs_epi16 (__m128i __A)
+{
+ return (__m128i) vec_abs ((__v8hi) __A);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_abs_epi32 (__m128i __A)
+{
+ return (__m128i) vec_abs ((__v4si) __A);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_abs_epi8 (__m128i __A)
+{
+ return (__m128i) vec_abs ((__v16qi) __A);
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_abs_pi16 (__m64 __A)
+{
+ __v8hi __B = (__v8hi) (__v2du) { __A, __A };
+ return (__m64) ((__v2du) vec_abs (__B))[0];
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_abs_pi32 (__m64 __A)
+{
+ __v4si __B = (__v4si) (__v2du) { __A, __A };
+ return (__m64) ((__v2du) vec_abs (__B))[0];
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_abs_pi8 (__m64 __A)
+{
+ __v16qi __B = (__v16qi) (__v2du) { __A, __A };
+ return (__m64) ((__v2du) vec_abs (__B))[0];
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_alignr_epi8 (__m128i __A, __m128i __B, const unsigned int __count)
+{
+ if (__builtin_constant_p (__count) && __count < 16)
+ {
+#ifdef __LITTLE_ENDIAN__
+ __A = (__m128i) vec_reve ((__v16qu) __A);
+ __B = (__m128i) vec_reve ((__v16qu) __B);
+#endif
+ __A = (__m128i) vec_sld ((__v16qu) __B, (__v16qu) __A, __count);
+#ifdef __LITTLE_ENDIAN__
+ __A = (__m128i) vec_reve ((__v16qu) __A);
+#endif
+ return __A;
+ }
+
+ if (__count == 0)
+ return __B;
+
+ if (__count >= 16)
+ {
+ if (__count >= 32)
+ {
+ const __v16qu zero = { 0 };
+ return (__m128i) zero;
+ }
+ else
+ {
+ const __v16qu __shift =
+ vec_splats ((unsigned char) ((__count - 16) * 8));
+#ifdef __LITTLE_ENDIAN__
+ return (__m128i) vec_sro ((__v16qu) __A, __shift);
+#else
+ return (__m128i) vec_slo ((__v16qu) __A, __shift);
+#endif
+ }
+ }
+ else
+ {
+ const __v16qu __shiftA =
+ vec_splats ((unsigned char) ((16 - __count) * 8));
+ const __v16qu __shiftB = vec_splats ((unsigned char) (__count * 8));
+#ifdef __LITTLE_ENDIAN__
+ __A = (__m128i) vec_slo ((__v16qu) __A, __shiftA);
+ __B = (__m128i) vec_sro ((__v16qu) __B, __shiftB);
+#else
+ __A = (__m128i) vec_sro ((__v16qu) __A, __shiftA);
+ __B = (__m128i) vec_slo ((__v16qu) __B, __shiftB);
+#endif
+ return (__m128i) vec_or ((__v16qu) __A, (__v16qu) __B);
+ }
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_alignr_pi8 (__m64 __A, __m64 __B, unsigned int __count)
+{
+ if (__count < 16)
+ {
+ __v2du __C = { __B, __A };
+#ifdef __LITTLE_ENDIAN__
+ const __v4su __shift = { __count << 3, 0, 0, 0 };
+ __C = (__v2du) vec_sro ((__v16qu) __C, (__v16qu) __shift);
+#else
+ const __v4su __shift = { 0, 0, 0, __count << 3 };
+ __C = (__v2du) vec_slo ((__v16qu) __C, (__v16qu) __shift);
+#endif
+ return (__m64) __C[0];
+ }
+ else
+ {
+ const __m64 __zero = { 0 };
+ return __zero;
+ }
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hadd_epi16 (__m128i __A, __m128i __B)
+{
+ const __v16qu __P =
+ { 0, 1, 4, 5, 8, 9, 12, 13, 16, 17, 20, 21, 24, 25, 28, 29 };
+ const __v16qu __Q =
+ { 2, 3, 6, 7, 10, 11, 14, 15, 18, 19, 22, 23, 26, 27, 30, 31 };
+ __v8hi __C = vec_perm ((__v8hi) __A, (__v8hi) __B, __P);
+ __v8hi __D = vec_perm ((__v8hi) __A, (__v8hi) __B, __Q);
+ return (__m128i) vec_add (__C, __D);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hadd_epi32 (__m128i __A, __m128i __B)
+{
+ const __v16qu __P =
+ { 0, 1, 2, 3, 8, 9, 10, 11, 16, 17, 18, 19, 24, 25, 26, 27 };
+ const __v16qu __Q =
+ { 4, 5, 6, 7, 12, 13, 14, 15, 20, 21, 22, 23, 28, 29, 30, 31 };
+ __v4si __C = vec_perm ((__v4si) __A, (__v4si) __B, __P);
+ __v4si __D = vec_perm ((__v4si) __A, (__v4si) __B, __Q);
+ return (__m128i) vec_add (__C, __D);
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hadd_pi16 (__m64 __A, __m64 __B)
+{
+ __v8hi __C = (__v8hi) (__v2du) { __A, __B };
+ const __v16qu __P =
+ { 0, 1, 4, 5, 8, 9, 12, 13, 0, 1, 4, 5, 8, 9, 12, 13 };
+ const __v16qu __Q =
+ { 2, 3, 6, 7, 10, 11, 14, 15, 2, 3, 6, 7, 10, 11, 14, 15 };
+ __v8hi __D = vec_perm (__C, __C, __Q);
+ __C = vec_perm (__C, __C, __P);
+ __C = vec_add (__C, __D);
+ return (__m64) ((__v2du) __C)[1];
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hadd_pi32 (__m64 __A, __m64 __B)
+{
+ __v4si __C = (__v4si) (__v2du) { __A, __B };
+ const __v16qu __P =
+ { 0, 1, 2, 3, 8, 9, 10, 11, 0, 1, 2, 3, 8, 9, 10, 11 };
+ const __v16qu __Q =
+ { 4, 5, 6, 7, 12, 13, 14, 15, 4, 5, 6, 7, 12, 13, 14, 15 };
+ __v4si __D = vec_perm (__C, __C, __Q);
+ __C = vec_perm (__C, __C, __P);
+ __C = vec_add (__C, __D);
+ return (__m64) ((__v2du) __C)[1];
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hadds_epi16 (__m128i __A, __m128i __B)
+{
+ __v4si __C = { 0 }, __D = { 0 };
+ __C = vec_sum4s ((__v8hi) __A, __C);
+ __D = vec_sum4s ((__v8hi) __B, __D);
+ __C = (__v4si) vec_packs (__C, __D);
+ return (__m128i) __C;
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hadds_pi16 (__m64 __A, __m64 __B)
+{
+ const __v4si __zero = { 0 };
+ __v8hi __C = (__v8hi) (__v2du) { __A, __B };
+ __v4si __D = vec_sum4s (__C, __zero);
+ __C = vec_packs (__D, __D);
+ return (__m64) ((__v2du) __C)[1];
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hsub_epi16 (__m128i __A, __m128i __B)
+{
+ const __v16qu __P =
+ { 0, 1, 4, 5, 8, 9, 12, 13, 16, 17, 20, 21, 24, 25, 28, 29 };
+ const __v16qu __Q =
+ { 2, 3, 6, 7, 10, 11, 14, 15, 18, 19, 22, 23, 26, 27, 30, 31 };
+ __v8hi __C = vec_perm ((__v8hi) __A, (__v8hi) __B, __P);
+ __v8hi __D = vec_perm ((__v8hi) __A, (__v8hi) __B, __Q);
+ return (__m128i) vec_sub (__C, __D);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hsub_epi32 (__m128i __A, __m128i __B)
+{
+ const __v16qu __P =
+ { 0, 1, 2, 3, 8, 9, 10, 11, 16, 17, 18, 19, 24, 25, 26, 27 };
+ const __v16qu __Q =
+ { 4, 5, 6, 7, 12, 13, 14, 15, 20, 21, 22, 23, 28, 29, 30, 31 };
+ __v4si __C = vec_perm ((__v4si) __A, (__v4si) __B, __P);
+ __v4si __D = vec_perm ((__v4si) __A, (__v4si) __B, __Q);
+ return (__m128i) vec_sub (__C, __D);
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hsub_pi16 (__m64 __A, __m64 __B)
+{
+ const __v16qu __P =
+ { 0, 1, 4, 5, 8, 9, 12, 13, 0, 1, 4, 5, 8, 9, 12, 13 };
+ const __v16qu __Q =
+ { 2, 3, 6, 7, 10, 11, 14, 15, 2, 3, 6, 7, 10, 11, 14, 15 };
+ __v8hi __C = (__v8hi) (__v2du) { __A, __B };
+ __v8hi __D = vec_perm (__C, __C, __Q);
+ __C = vec_perm (__C, __C, __P);
+ __C = vec_sub (__C, __D);
+ return (__m64) ((__v2du) __C)[1];
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hsub_pi32 (__m64 __A, __m64 __B)
+{
+ const __v16qu __P =
+ { 0, 1, 2, 3, 8, 9, 10, 11, 0, 1, 2, 3, 8, 9, 10, 11 };
+ const __v16qu __Q =
+ { 4, 5, 6, 7, 12, 13, 14, 15, 4, 5, 6, 7, 12, 13, 14, 15 };
+ __v4si __C = (__v4si) (__v2du) { __A, __B };
+ __v4si __D = vec_perm (__C, __C, __Q);
+ __C = vec_perm (__C, __C, __P);
+ __C = vec_sub (__C, __D);
+ return (__m64) ((__v2du) __C)[1];
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hsubs_epi16 (__m128i __A, __m128i __B)
+{
+ const __v16qu __P =
+ { 0, 1, 4, 5, 8, 9, 12, 13, 16, 17, 20, 21, 24, 25, 28, 29 };
+ const __v16qu __Q =
+ { 2, 3, 6, 7, 10, 11, 14, 15, 18, 19, 22, 23, 26, 27, 30, 31 };
+ __v8hi __C = vec_perm ((__v8hi) __A, (__v8hi) __B, __P);
+ __v8hi __D = vec_perm ((__v8hi) __A, (__v8hi) __B, __Q);
+ return (__m128i) vec_subs (__C, __D);
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_hsubs_pi16 (__m64 __A, __m64 __B)
+{
+ const __v16qu __P =
+ { 0, 1, 4, 5, 8, 9, 12, 13, 0, 1, 4, 5, 8, 9, 12, 13 };
+ const __v16qu __Q =
+ { 2, 3, 6, 7, 10, 11, 14, 15, 2, 3, 6, 7, 10, 11, 14, 15 };
+ __v8hi __C = (__v8hi) (__v2du) { __A, __B };
+ __v8hi __D = vec_perm (__C, __C, __P);
+ __v8hi __E = vec_perm (__C, __C, __Q);
+ __C = vec_subs (__D, __E);
+ return (__m64) ((__v2du) __C)[1];
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_shuffle_epi8 (__m128i __A, __m128i __B)
+{
+ const __v16qi __zero = { 0 };
+ __vector __bool char __select = vec_cmplt ((__v16qi) __B, __zero);
+ __v16qi __C = vec_perm ((__v16qi) __A, (__v16qi) __A, (__v16qu) __B);
+ return (__m128i) vec_sel (__C, __zero, __select);
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_shuffle_pi8 (__m64 __A, __m64 __B)
+{
+ const __v16qi __zero = { 0 };
+ __v16qi __C = (__v16qi) (__v2du) { __A, __A };
+ __v16qi __D = (__v16qi) (__v2du) { __B, __B };
+ __vector __bool char __select = vec_cmplt ((__v16qi) __D, __zero);
+ __C = vec_perm ((__v16qi) __C, (__v16qi) __C, (__v16qu) __D);
+ __C = vec_sel (__C, __zero, __select);
+ return (__m64) ((__v2du) (__C))[0];
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sign_epi8 (__m128i __A, __m128i __B)
+{
+ const __v16qi __zero = { 0 };
+ __v16qi __selectneg = (__v16qi) vec_cmplt ((__v16qi) __B, __zero);
+ __v16qi __selectpos =
+ (__v16qi) vec_neg ((__v16qi) vec_cmpgt ((__v16qi) __B, __zero));
+ __v16qi __conv = vec_add (__selectneg, __selectpos);
+ return (__m128i) vec_mul ((__v16qi) __A, (__v16qi) __conv);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sign_epi16 (__m128i __A, __m128i __B)
+{
+ const __v8hi __zero = { 0 };
+ __v8hi __selectneg = (__v8hi) vec_cmplt ((__v8hi) __B, __zero);
+ __v8hi __selectpos =
+ (__v8hi) vec_neg ((__v8hi) vec_cmpgt ((__v8hi) __B, __zero));
+ __v8hi __conv = vec_add (__selectneg, __selectpos);
+ return (__m128i) vec_mul ((__v8hi) __A, (__v8hi) __conv);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sign_epi32 (__m128i __A, __m128i __B)
+{
+ const __v4si __zero = { 0 };
+ __v4si __selectneg = (__v4si) vec_cmplt ((__v4si) __B, __zero);
+ __v4si __selectpos =
+ (__v4si) vec_neg ((__v4si) vec_cmpgt ((__v4si) __B, __zero));
+ __v4si __conv = vec_add (__selectneg, __selectpos);
+ return (__m128i) vec_mul ((__v4si) __A, (__v4si) __conv);
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sign_pi8 (__m64 __A, __m64 __B)
+{
+ const __v16qi __zero = { 0 };
+ __v16qi __C = (__v16qi) (__v2du) { __A, __A };
+ __v16qi __D = (__v16qi) (__v2du) { __B, __B };
+ __C = (__v16qi) _mm_sign_epi8 ((__m128i) __C, (__m128i) __D);
+ return (__m64) ((__v2du) (__C))[0];
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sign_pi16 (__m64 __A, __m64 __B)
+{
+ const __v8hi __zero = { 0 };
+ __v8hi __C = (__v8hi) (__v2du) { __A, __A };
+ __v8hi __D = (__v8hi) (__v2du) { __B, __B };
+ __C = (__v8hi) _mm_sign_epi16 ((__m128i) __C, (__m128i) __D);
+ return (__m64) ((__v2du) (__C))[0];
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sign_pi32 (__m64 __A, __m64 __B)
+{
+ const __v4si __zero = { 0 };
+ __v4si __C = (__v4si) (__v2du) { __A, __A };
+ __v4si __D = (__v4si) (__v2du) { __B, __B };
+ __C = (__v4si) _mm_sign_epi32 ((__m128i) __C, (__m128i) __D);
+ return (__m64) ((__v2du) (__C))[0];
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maddubs_epi16 (__m128i __A, __m128i __B)
+{
+ __v8hi __unsigned = vec_splats ((signed short) 0x00ff);
+ __v8hi __C = vec_and (vec_unpackh ((__v16qi) __A), __unsigned);
+ __v8hi __D = vec_and (vec_unpackl ((__v16qi) __A), __unsigned);
+ __v8hi __E = vec_unpackh ((__v16qi) __B);
+ __v8hi __F = vec_unpackl ((__v16qi) __B);
+ __C = vec_mul (__C, __E);
+ __D = vec_mul (__D, __F);
+ const __v16qu __odds =
+ { 0, 1, 4, 5, 8, 9, 12, 13, 16, 17, 20, 21, 24, 25, 28, 29 };
+ const __v16qu __evens =
+ { 2, 3, 6, 7, 10, 11, 14, 15, 18, 19, 22, 23, 26, 27, 30, 31 };
+ __E = vec_perm (__C, __D, __odds);
+ __F = vec_perm (__C, __D, __evens);
+ return (__m128i) vec_adds (__E, __F);
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maddubs_pi16 (__m64 __A, __m64 __B)
+{
+ __v8hi __C = (__v8hi) (__v2du) { __A, __A };
+ __C = vec_unpackl ((__v16qi) __C);
+ const __v8hi __unsigned = vec_splats ((signed short) 0x00ff);
+ __C = vec_and (__C, __unsigned);
+ __v8hi __D = (__v8hi) (__v2du) { __B, __B };
+ __D = vec_unpackl ((__v16qi) __D);
+ __D = vec_mul (__C, __D);
+ const __v16qu __odds =
+ { 0, 1, 4, 5, 8, 9, 12, 13, 16, 17, 20, 21, 24, 25, 28, 29 };
+ const __v16qu __evens =
+ { 2, 3, 6, 7, 10, 11, 14, 15, 18, 19, 22, 23, 26, 27, 30, 31 };
+ __C = vec_perm (__D, __D, __odds);
+ __D = vec_perm (__D, __D, __evens);
+ __C = vec_adds (__C, __D);
+ return (__m64) ((__v2du) (__C))[0];
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mulhrs_epi16 (__m128i __A, __m128i __B)
+{
+ __v4si __C = vec_unpackh ((__v8hi) __A);
+ __v4si __D = vec_unpackh ((__v8hi) __B);
+ __C = vec_mul (__C, __D);
+ __D = vec_unpackl ((__v8hi) __A);
+ __v4si __E = vec_unpackl ((__v8hi) __B);
+ __D = vec_mul (__D, __E);
+ const __v4su __shift = vec_splats ((unsigned int) 14);
+ __C = vec_sr (__C, __shift);
+ __D = vec_sr (__D, __shift);
+ const __v4si __ones = vec_splats ((signed int) 1);
+ __C = vec_add (__C, __ones);
+ __C = vec_sr (__C, (__v4su) __ones);
+ __D = vec_add (__D, __ones);
+ __D = vec_sr (__D, (__v4su) __ones);
+ return (__m128i) vec_pack (__C, __D);
+}
+
+extern __inline __m64
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mulhrs_pi16 (__m64 __A, __m64 __B)
+{
+ __v4si __C = (__v4si) (__v2du) { __A, __A };
+ __C = vec_unpackh ((__v8hi) __C);
+ __v4si __D = (__v4si) (__v2du) { __B, __B };
+ __D = vec_unpackh ((__v8hi) __D);
+ __C = vec_mul (__C, __D);
+ const __v4su __shift = vec_splats ((unsigned int) 14);
+ __C = vec_sr (__C, __shift);
+ const __v4si __ones = vec_splats ((signed int) 1);
+ __C = vec_add (__C, __ones);
+ __C = vec_sr (__C, (__v4su) __ones);
+ __v8hi __E = vec_pack (__C, __D);
+ return (__m64) ((__v2du) (__E))[0];
+}
+
+#else
+#include_next <tmmintrin.h>
+#endif /* defined(__linux__) && defined(__ppc64__) */
+
+#endif /* TMMINTRIN_H_ */
--- /dev/null
+/*===---- xmmintrin.h - Implementation of SSE intrinsics on PowerPC --------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+/* Implemented from the specification included in the Intel C++ Compiler
+ User Guide and Reference, version 9.0. */
+
+#ifndef NO_WARN_X86_INTRINSICS
+/* This header file is to help porting code using Intel intrinsics
+ explicitly from x86_64 to powerpc64/powerpc64le.
+
+ Since X86 SSE intrinsics mainly handles __m128 type, PowerPC
+ VMX/VSX ISA is a good match for vector float SIMD operations.
+ However scalar float operations in vector (XMM) registers require
+ the POWER8 VSX ISA (2.07) level. There are differences for data
+ format and placement of float scalars in the vector register, which
+ require extra steps to match SSE scalar float semantics on POWER.
+
+ It should be noted that there's much difference between X86_64's
+ MXSCR and PowerISA's FPSCR/VSCR registers. It's recommended to use
+ portable <fenv.h> instead of access MXSCR directly.
+
+ Most SSE scalar float intrinsic operations can be performed more
+ efficiently as C language float scalar operations or optimized to
+ use vector SIMD operations. We recommend this for new applications. */
+#error "Please read comment above. Use -DNO_WARN_X86_INTRINSICS to disable this error."
+#endif
+
+#ifndef _XMMINTRIN_H_INCLUDED
+#define _XMMINTRIN_H_INCLUDED
+
+#if defined(__linux__) && defined(__ppc64__)
+
+/* Define four value permute mask */
+#define _MM_SHUFFLE(w,x,y,z) (((w) << 6) | ((x) << 4) | ((y) << 2) | (z))
+
+#include <altivec.h>
+
+/* Avoid collisions between altivec.h and strict adherence to C++ and
+ C11 standards. This should eventually be done inside altivec.h itself,
+ but only after testing a full distro build. */
+#if defined(__STRICT_ANSI__) && (defined(__cplusplus) || \
+ (defined(__STDC_VERSION__) && \
+ __STDC_VERSION__ >= 201112L))
+#undef vector
+#undef pixel
+#undef bool
+#endif
+
+/* We need type definitions from the MMX header file. */
+#include <mmintrin.h>
+
+/* Get _mm_malloc () and _mm_free (). */
+#if __STDC_HOSTED__
+#include <mm_malloc.h>
+#endif
+
+/* The Intel API is flexible enough that we must allow aliasing with other
+ vector types, and their scalar components. */
+typedef float __m128 __attribute__ ((__vector_size__ (16), __may_alias__));
+
+/* Unaligned version of the same type. */
+typedef float __m128_u __attribute__ ((__vector_size__ (16), __may_alias__,
+ __aligned__ (1)));
+
+/* Internal data types for implementing the intrinsics. */
+typedef float __v4sf __attribute__ ((__vector_size__ (16)));
+
+/* Create an undefined vector. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_undefined_ps (void)
+{
+ __m128 __Y = __Y;
+ return __Y;
+}
+
+/* Create a vector of zeros. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_setzero_ps (void)
+{
+ return __extension__ (__m128){ 0.0f, 0.0f, 0.0f, 0.0f };
+}
+
+/* Load four SPFP values from P. The address must be 16-byte aligned. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_load_ps (float const *__P)
+{
+ return ((__m128)vec_ld(0, (__v4sf*)__P));
+}
+
+/* Load four SPFP values from P. The address need not be 16-byte aligned. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loadu_ps (float const *__P)
+{
+ return (vec_vsx_ld(0, __P));
+}
+
+/* Load four SPFP values in reverse order. The address must be aligned. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loadr_ps (float const *__P)
+{
+ __v4sf __tmp;
+ __m128 result;
+ static const __vector unsigned char permute_vector =
+ { 0x1C, 0x1D, 0x1E, 0x1F, 0x18, 0x19, 0x1A, 0x1B, 0x14, 0x15, 0x16,
+ 0x17, 0x10, 0x11, 0x12, 0x13 };
+
+ __tmp = vec_ld (0, (__v4sf *) __P);
+ result = (__m128) vec_perm (__tmp, __tmp, permute_vector);
+ return result;
+}
+
+/* Create a vector with all four elements equal to F. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set1_ps (float __F)
+{
+ return __extension__ (__m128)(__v4sf){ __F, __F, __F, __F };
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_ps1 (float __F)
+{
+ return _mm_set1_ps (__F);
+}
+
+/* Create the vector [Z Y X W]. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_ps (const float __Z, const float __Y, const float __X, const float __W)
+{
+ return __extension__ (__m128)(__v4sf){ __W, __X, __Y, __Z };
+}
+
+/* Create the vector [W X Y Z]. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_setr_ps (float __Z, float __Y, float __X, float __W)
+{
+ return __extension__ (__m128)(__v4sf){ __Z, __Y, __X, __W };
+}
+
+/* Store four SPFP values. The address must be 16-byte aligned. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_store_ps (float *__P, __m128 __A)
+{
+ vec_st((__v4sf)__A, 0, (__v4sf*)__P);
+}
+
+/* Store four SPFP values. The address need not be 16-byte aligned. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_storeu_ps (float *__P, __m128 __A)
+{
+ *(__m128_u *)__P = __A;
+}
+
+/* Store four SPFP values in reverse order. The address must be aligned. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_storer_ps (float *__P, __m128 __A)
+{
+ __v4sf __tmp;
+ static const __vector unsigned char permute_vector =
+ { 0x1C, 0x1D, 0x1E, 0x1F, 0x18, 0x19, 0x1A, 0x1B, 0x14, 0x15, 0x16,
+ 0x17, 0x10, 0x11, 0x12, 0x13 };
+
+ __tmp = (__m128) vec_perm (__A, __A, permute_vector);
+
+ _mm_store_ps (__P, __tmp);
+}
+
+/* Store the lower SPFP value across four words. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_store1_ps (float *__P, __m128 __A)
+{
+ __v4sf __va = vec_splat((__v4sf)__A, 0);
+ _mm_store_ps (__P, __va);
+}
+
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_store_ps1 (float *__P, __m128 __A)
+{
+ _mm_store1_ps (__P, __A);
+}
+
+/* Create a vector with element 0 as F and the rest zero. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_ss (float __F)
+{
+ return __extension__ (__m128)(__v4sf){ __F, 0.0f, 0.0f, 0.0f };
+}
+
+/* Sets the low SPFP value of A from the low value of B. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_move_ss (__m128 __A, __m128 __B)
+{
+ static const __vector unsigned int mask = {0xffffffff, 0, 0, 0};
+
+ return (vec_sel ((__v4sf)__A, (__v4sf)__B, mask));
+}
+
+/* Create a vector with element 0 as *P and the rest zero. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_load_ss (float const *__P)
+{
+ return _mm_set_ss (*__P);
+}
+
+/* Stores the lower SPFP value. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_store_ss (float *__P, __m128 __A)
+{
+ *__P = ((__v4sf)__A)[0];
+}
+
+/* Perform the respective operation on the lower SPFP (single-precision
+ floating-point) values of A and B; the upper three SPFP values are
+ passed through from A. */
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_add_ss (__m128 __A, __m128 __B)
+{
+#ifdef _ARCH_PWR7
+ __m128 a, b, c;
+ static const __vector unsigned int mask = {0xffffffff, 0, 0, 0};
+ /* PowerISA VSX does not allow partial (for just lower double)
+ results. So to insure we don't generate spurious exceptions
+ (from the upper double values) we splat the lower double
+ before we to the operation. */
+ a = vec_splat (__A, 0);
+ b = vec_splat (__B, 0);
+ c = a + b;
+ /* Then we merge the lower float result with the original upper
+ float elements from __A. */
+ return (vec_sel (__A, c, mask));
+#else
+ __A[0] = __A[0] + __B[0];
+ return (__A);
+#endif
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sub_ss (__m128 __A, __m128 __B)
+{
+#ifdef _ARCH_PWR7
+ __m128 a, b, c;
+ static const __vector unsigned int mask = {0xffffffff, 0, 0, 0};
+ /* PowerISA VSX does not allow partial (for just lower double)
+ results. So to insure we don't generate spurious exceptions
+ (from the upper double values) we splat the lower double
+ before we to the operation. */
+ a = vec_splat (__A, 0);
+ b = vec_splat (__B, 0);
+ c = a - b;
+ /* Then we merge the lower float result with the original upper
+ float elements from __A. */
+ return (vec_sel (__A, c, mask));
+#else
+ __A[0] = __A[0] - __B[0];
+ return (__A);
+#endif
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mul_ss (__m128 __A, __m128 __B)
+{
+#ifdef _ARCH_PWR7
+ __m128 a, b, c;
+ static const __vector unsigned int mask = {0xffffffff, 0, 0, 0};
+ /* PowerISA VSX does not allow partial (for just lower double)
+ results. So to insure we don't generate spurious exceptions
+ (from the upper double values) we splat the lower double
+ before we to the operation. */
+ a = vec_splat (__A, 0);
+ b = vec_splat (__B, 0);
+ c = a * b;
+ /* Then we merge the lower float result with the original upper
+ float elements from __A. */
+ return (vec_sel (__A, c, mask));
+#else
+ __A[0] = __A[0] * __B[0];
+ return (__A);
+#endif
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_div_ss (__m128 __A, __m128 __B)
+{
+#ifdef _ARCH_PWR7
+ __m128 a, b, c;
+ static const __vector unsigned int mask = {0xffffffff, 0, 0, 0};
+ /* PowerISA VSX does not allow partial (for just lower double)
+ results. So to insure we don't generate spurious exceptions
+ (from the upper double values) we splat the lower double
+ before we to the operation. */
+ a = vec_splat (__A, 0);
+ b = vec_splat (__B, 0);
+ c = a / b;
+ /* Then we merge the lower float result with the original upper
+ float elements from __A. */
+ return (vec_sel (__A, c, mask));
+#else
+ __A[0] = __A[0] / __B[0];
+ return (__A);
+#endif
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sqrt_ss (__m128 __A)
+{
+ __m128 a, c;
+ static const __vector unsigned int mask = {0xffffffff, 0, 0, 0};
+ /* PowerISA VSX does not allow partial (for just lower double)
+ * results. So to insure we don't generate spurious exceptions
+ * (from the upper double values) we splat the lower double
+ * before we to the operation. */
+ a = vec_splat (__A, 0);
+ c = vec_sqrt (a);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return (vec_sel (__A, c, mask));
+}
+
+/* Perform the respective operation on the four SPFP values in A and B. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_add_ps (__m128 __A, __m128 __B)
+{
+ return (__m128) ((__v4sf)__A + (__v4sf)__B);
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sub_ps (__m128 __A, __m128 __B)
+{
+ return (__m128) ((__v4sf)__A - (__v4sf)__B);
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mul_ps (__m128 __A, __m128 __B)
+{
+ return (__m128) ((__v4sf)__A * (__v4sf)__B);
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_div_ps (__m128 __A, __m128 __B)
+{
+ return (__m128) ((__v4sf)__A / (__v4sf)__B);
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sqrt_ps (__m128 __A)
+{
+ return (vec_sqrt ((__v4sf)__A));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_rcp_ps (__m128 __A)
+{
+ return (vec_re ((__v4sf)__A));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_rsqrt_ps (__m128 __A)
+{
+ return (vec_rsqrte (__A));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_rcp_ss (__m128 __A)
+{
+ __m128 a, c;
+ static const __vector unsigned int mask = {0xffffffff, 0, 0, 0};
+ /* PowerISA VSX does not allow partial (for just lower double)
+ * results. So to insure we don't generate spurious exceptions
+ * (from the upper double values) we splat the lower double
+ * before we to the operation. */
+ a = vec_splat (__A, 0);
+ c = _mm_rcp_ps (a);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return (vec_sel (__A, c, mask));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_rsqrt_ss (__m128 __A)
+{
+ __m128 a, c;
+ static const __vector unsigned int mask = {0xffffffff, 0, 0, 0};
+ /* PowerISA VSX does not allow partial (for just lower double)
+ * results. So to insure we don't generate spurious exceptions
+ * (from the upper double values) we splat the lower double
+ * before we to the operation. */
+ a = vec_splat (__A, 0);
+ c = vec_rsqrte (a);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return (vec_sel (__A, c, mask));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_min_ss (__m128 __A, __m128 __B)
+{
+ __v4sf a, b, c;
+ static const __vector unsigned int mask = {0xffffffff, 0, 0, 0};
+ /* PowerISA VSX does not allow partial (for just lower float)
+ * results. So to insure we don't generate spurious exceptions
+ * (from the upper float values) we splat the lower float
+ * before we to the operation. */
+ a = vec_splat ((__v4sf)__A, 0);
+ b = vec_splat ((__v4sf)__B, 0);
+ c = vec_min (a, b);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return (vec_sel ((__v4sf)__A, c, mask));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_max_ss (__m128 __A, __m128 __B)
+{
+ __v4sf a, b, c;
+ static const __vector unsigned int mask = {0xffffffff, 0, 0, 0};
+ /* PowerISA VSX does not allow partial (for just lower float)
+ * results. So to insure we don't generate spurious exceptions
+ * (from the upper float values) we splat the lower float
+ * before we to the operation. */
+ a = vec_splat (__A, 0);
+ b = vec_splat (__B, 0);
+ c = vec_max (a, b);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return (vec_sel ((__v4sf)__A, c, mask));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_min_ps (__m128 __A, __m128 __B)
+{
+ __vector __bool int m = vec_cmpgt ((__v4sf) __B, (__v4sf) __A);
+ return vec_sel (__B, __A, m);
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_max_ps (__m128 __A, __m128 __B)
+{
+ __vector __bool int m = vec_cmpgt ((__v4sf) __A, (__v4sf) __B);
+ return vec_sel (__B, __A, m);
+}
+
+/* Perform logical bit-wise operations on 128-bit values. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_and_ps (__m128 __A, __m128 __B)
+{
+ return ((__m128)vec_and ((__v4sf)__A, (__v4sf)__B));
+// return __builtin_ia32_andps (__A, __B);
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_andnot_ps (__m128 __A, __m128 __B)
+{
+ return ((__m128)vec_andc ((__v4sf)__B, (__v4sf)__A));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_or_ps (__m128 __A, __m128 __B)
+{
+ return ((__m128)vec_or ((__v4sf)__A, (__v4sf)__B));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_xor_ps (__m128 __A, __m128 __B)
+{
+ return ((__m128)vec_xor ((__v4sf)__A, (__v4sf)__B));
+}
+
+/* Perform a comparison on the four SPFP values of A and B. For each
+ element, if the comparison is true, place a mask of all ones in the
+ result, otherwise a mask of zeros. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpeq_ps (__m128 __A, __m128 __B)
+{
+ return ((__m128)vec_cmpeq ((__v4sf)__A,(__v4sf) __B));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmplt_ps (__m128 __A, __m128 __B)
+{
+ return ((__m128)vec_cmplt ((__v4sf)__A, (__v4sf)__B));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmple_ps (__m128 __A, __m128 __B)
+{
+ return ((__m128)vec_cmple ((__v4sf)__A, (__v4sf)__B));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpgt_ps (__m128 __A, __m128 __B)
+{
+ return ((__m128)vec_cmpgt ((__v4sf)__A, (__v4sf)__B));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpge_ps (__m128 __A, __m128 __B)
+{
+ return ((__m128)vec_cmpge ((__v4sf)__A, (__v4sf)__B));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpneq_ps (__m128 __A, __m128 __B)
+{
+ __v4sf temp = (__v4sf ) vec_cmpeq ((__v4sf) __A, (__v4sf)__B);
+ return ((__m128)vec_nor (temp, temp));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnlt_ps (__m128 __A, __m128 __B)
+{
+ return ((__m128)vec_cmpge ((__v4sf)__A, (__v4sf)__B));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnle_ps (__m128 __A, __m128 __B)
+{
+ return ((__m128)vec_cmpgt ((__v4sf)__A, (__v4sf)__B));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpngt_ps (__m128 __A, __m128 __B)
+{
+ return ((__m128)vec_cmple ((__v4sf)__A, (__v4sf)__B));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnge_ps (__m128 __A, __m128 __B)
+{
+ return ((__m128)vec_cmplt ((__v4sf)__A, (__v4sf)__B));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpord_ps (__m128 __A, __m128 __B)
+{
+ __vector unsigned int a, b;
+ __vector unsigned int c, d;
+ static const __vector unsigned int float_exp_mask =
+ { 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000 };
+
+ a = (__vector unsigned int) vec_abs ((__v4sf)__A);
+ b = (__vector unsigned int) vec_abs ((__v4sf)__B);
+ c = (__vector unsigned int) vec_cmpgt (float_exp_mask, a);
+ d = (__vector unsigned int) vec_cmpgt (float_exp_mask, b);
+ return ((__m128 ) vec_and (c, d));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpunord_ps (__m128 __A, __m128 __B)
+{
+ __vector unsigned int a, b;
+ __vector unsigned int c, d;
+ static const __vector unsigned int float_exp_mask =
+ { 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000 };
+
+ a = (__vector unsigned int) vec_abs ((__v4sf)__A);
+ b = (__vector unsigned int) vec_abs ((__v4sf)__B);
+ c = (__vector unsigned int) vec_cmpgt (a, float_exp_mask);
+ d = (__vector unsigned int) vec_cmpgt (b, float_exp_mask);
+ return ((__m128 ) vec_or (c, d));
+}
+
+/* Perform a comparison on the lower SPFP values of A and B. If the
+ comparison is true, place a mask of all ones in the result, otherwise a
+ mask of zeros. The upper three SPFP values are passed through from A. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpeq_ss (__m128 __A, __m128 __B)
+{
+ static const __vector unsigned int mask =
+ { 0xffffffff, 0, 0, 0 };
+ __v4sf a, b, c;
+ /* PowerISA VMX does not allow partial (for just element 0)
+ * results. So to insure we don't generate spurious exceptions
+ * (from the upper elements) we splat the lower float
+ * before we to the operation. */
+ a = vec_splat ((__v4sf) __A, 0);
+ b = vec_splat ((__v4sf) __B, 0);
+ c = (__v4sf) vec_cmpeq(a, b);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return ((__m128)vec_sel ((__v4sf)__A, c, mask));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmplt_ss (__m128 __A, __m128 __B)
+{
+ static const __vector unsigned int mask =
+ { 0xffffffff, 0, 0, 0 };
+ __v4sf a, b, c;
+ /* PowerISA VMX does not allow partial (for just element 0)
+ * results. So to insure we don't generate spurious exceptions
+ * (from the upper elements) we splat the lower float
+ * before we to the operation. */
+ a = vec_splat ((__v4sf) __A, 0);
+ b = vec_splat ((__v4sf) __B, 0);
+ c = (__v4sf) vec_cmplt(a, b);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return ((__m128)vec_sel ((__v4sf)__A, c, mask));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmple_ss (__m128 __A, __m128 __B)
+{
+ static const __vector unsigned int mask =
+ { 0xffffffff, 0, 0, 0 };
+ __v4sf a, b, c;
+ /* PowerISA VMX does not allow partial (for just element 0)
+ * results. So to insure we don't generate spurious exceptions
+ * (from the upper elements) we splat the lower float
+ * before we to the operation. */
+ a = vec_splat ((__v4sf) __A, 0);
+ b = vec_splat ((__v4sf) __B, 0);
+ c = (__v4sf) vec_cmple(a, b);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return ((__m128)vec_sel ((__v4sf)__A, c, mask));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpgt_ss (__m128 __A, __m128 __B)
+{
+ static const __vector unsigned int mask =
+ { 0xffffffff, 0, 0, 0 };
+ __v4sf a, b, c;
+ /* PowerISA VMX does not allow partial (for just element 0)
+ * results. So to insure we don't generate spurious exceptions
+ * (from the upper elements) we splat the lower float
+ * before we to the operation. */
+ a = vec_splat ((__v4sf) __A, 0);
+ b = vec_splat ((__v4sf) __B, 0);
+ c = (__v4sf) vec_cmpgt(a, b);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return ((__m128)vec_sel ((__v4sf)__A, c, mask));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpge_ss (__m128 __A, __m128 __B)
+{
+ static const __vector unsigned int mask =
+ { 0xffffffff, 0, 0, 0 };
+ __v4sf a, b, c;
+ /* PowerISA VMX does not allow partial (for just element 0)
+ * results. So to insure we don't generate spurious exceptions
+ * (from the upper elements) we splat the lower float
+ * before we to the operation. */
+ a = vec_splat ((__v4sf) __A, 0);
+ b = vec_splat ((__v4sf) __B, 0);
+ c = (__v4sf) vec_cmpge(a, b);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return ((__m128)vec_sel ((__v4sf)__A, c, mask));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpneq_ss (__m128 __A, __m128 __B)
+{
+ static const __vector unsigned int mask =
+ { 0xffffffff, 0, 0, 0 };
+ __v4sf a, b, c;
+ /* PowerISA VMX does not allow partial (for just element 0)
+ * results. So to insure we don't generate spurious exceptions
+ * (from the upper elements) we splat the lower float
+ * before we to the operation. */
+ a = vec_splat ((__v4sf) __A, 0);
+ b = vec_splat ((__v4sf) __B, 0);
+ c = (__v4sf) vec_cmpeq(a, b);
+ c = vec_nor (c, c);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return ((__m128)vec_sel ((__v4sf)__A, c, mask));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnlt_ss (__m128 __A, __m128 __B)
+{
+ static const __vector unsigned int mask =
+ { 0xffffffff, 0, 0, 0 };
+ __v4sf a, b, c;
+ /* PowerISA VMX does not allow partial (for just element 0)
+ * results. So to insure we don't generate spurious exceptions
+ * (from the upper elements) we splat the lower float
+ * before we to the operation. */
+ a = vec_splat ((__v4sf) __A, 0);
+ b = vec_splat ((__v4sf) __B, 0);
+ c = (__v4sf) vec_cmpge(a, b);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return ((__m128)vec_sel ((__v4sf)__A, c, mask));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnle_ss (__m128 __A, __m128 __B)
+{
+ static const __vector unsigned int mask =
+ { 0xffffffff, 0, 0, 0 };
+ __v4sf a, b, c;
+ /* PowerISA VMX does not allow partial (for just element 0)
+ * results. So to insure we don't generate spurious exceptions
+ * (from the upper elements) we splat the lower float
+ * before we to the operation. */
+ a = vec_splat ((__v4sf) __A, 0);
+ b = vec_splat ((__v4sf) __B, 0);
+ c = (__v4sf) vec_cmpgt(a, b);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return ((__m128)vec_sel ((__v4sf)__A, c, mask));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpngt_ss (__m128 __A, __m128 __B)
+{
+ static const __vector unsigned int mask =
+ { 0xffffffff, 0, 0, 0 };
+ __v4sf a, b, c;
+ /* PowerISA VMX does not allow partial (for just element 0)
+ * results. So to insure we don't generate spurious exceptions
+ * (from the upper elements) we splat the lower float
+ * before we to the operation. */
+ a = vec_splat ((__v4sf) __A, 0);
+ b = vec_splat ((__v4sf) __B, 0);
+ c = (__v4sf) vec_cmple(a, b);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return ((__m128)vec_sel ((__v4sf)__A, c, mask));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnge_ss (__m128 __A, __m128 __B)
+{
+ static const __vector unsigned int mask =
+ { 0xffffffff, 0, 0, 0 };
+ __v4sf a, b, c;
+ /* PowerISA VMX does not allow partial (for just element 0)
+ * results. So to insure we don't generate spurious exceptions
+ * (from the upper elements) we splat the lower float
+ * before we do the operation. */
+ a = vec_splat ((__v4sf) __A, 0);
+ b = vec_splat ((__v4sf) __B, 0);
+ c = (__v4sf) vec_cmplt(a, b);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return ((__m128)vec_sel ((__v4sf)__A, c, mask));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpord_ss (__m128 __A, __m128 __B)
+{
+ __vector unsigned int a, b;
+ __vector unsigned int c, d;
+ static const __vector unsigned int float_exp_mask =
+ { 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000 };
+ static const __vector unsigned int mask =
+ { 0xffffffff, 0, 0, 0 };
+
+ a = (__vector unsigned int) vec_abs ((__v4sf)__A);
+ b = (__vector unsigned int) vec_abs ((__v4sf)__B);
+ c = (__vector unsigned int) vec_cmpgt (float_exp_mask, a);
+ d = (__vector unsigned int) vec_cmpgt (float_exp_mask, b);
+ c = vec_and (c, d);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return ((__m128)vec_sel ((__v4sf)__A, (__v4sf)c, mask));
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpunord_ss (__m128 __A, __m128 __B)
+{
+ __vector unsigned int a, b;
+ __vector unsigned int c, d;
+ static const __vector unsigned int float_exp_mask =
+ { 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000 };
+ static const __vector unsigned int mask =
+ { 0xffffffff, 0, 0, 0 };
+
+ a = (__vector unsigned int) vec_abs ((__v4sf)__A);
+ b = (__vector unsigned int) vec_abs ((__v4sf)__B);
+ c = (__vector unsigned int) vec_cmpgt (a, float_exp_mask);
+ d = (__vector unsigned int) vec_cmpgt (b, float_exp_mask);
+ c = vec_or (c, d);
+ /* Then we merge the lower float result with the original upper
+ * float elements from __A. */
+ return ((__m128)vec_sel ((__v4sf)__A, (__v4sf)c, mask));
+}
+
+/* Compare the lower SPFP values of A and B and return 1 if true
+ and 0 if false. */
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comieq_ss (__m128 __A, __m128 __B)
+{
+ return (__A[0] == __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comilt_ss (__m128 __A, __m128 __B)
+{
+ return (__A[0] < __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comile_ss (__m128 __A, __m128 __B)
+{
+ return (__A[0] <= __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comigt_ss (__m128 __A, __m128 __B)
+{
+ return (__A[0] > __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comige_ss (__m128 __A, __m128 __B)
+{
+ return (__A[0] >= __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comineq_ss (__m128 __A, __m128 __B)
+{
+ return (__A[0] != __B[0]);
+}
+
+/* FIXME
+ * The __mm_ucomi??_ss implementations below are exactly the same as
+ * __mm_comi??_ss because GCC for PowerPC only generates unordered
+ * compares (scalar and vector).
+ * Technically __mm_comieq_ss et al should be using the ordered
+ * compare and signal for QNaNs.
+ * The __mm_ucomieq_sd et all should be OK, as is.
+ */
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomieq_ss (__m128 __A, __m128 __B)
+{
+ return (__A[0] == __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomilt_ss (__m128 __A, __m128 __B)
+{
+ return (__A[0] < __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomile_ss (__m128 __A, __m128 __B)
+{
+ return (__A[0] <= __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomigt_ss (__m128 __A, __m128 __B)
+{
+ return (__A[0] > __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomige_ss (__m128 __A, __m128 __B)
+{
+ return (__A[0] >= __B[0]);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomineq_ss (__m128 __A, __m128 __B)
+{
+ return (__A[0] != __B[0]);
+}
+
+extern __inline float __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtss_f32 (__m128 __A)
+{
+ return ((__v4sf)__A)[0];
+}
+
+/* Convert the lower SPFP value to a 32-bit integer according to the current
+ rounding mode. */
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtss_si32 (__m128 __A)
+{
+ __m64 res = 0;
+#ifdef _ARCH_PWR8
+ double dtmp;
+ __asm__(
+#ifdef __LITTLE_ENDIAN__
+ "xxsldwi %x0,%x0,%x0,3;\n"
+#endif
+ "xscvspdp %x2,%x0;\n"
+ "fctiw %2,%2;\n"
+ "mfvsrd %1,%x2;\n"
+ : "+wa" (__A),
+ "=r" (res),
+ "=f" (dtmp)
+ : );
+#else
+ res = __builtin_rint(__A[0]);
+#endif
+ return (res);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_ss2si (__m128 __A)
+{
+ return _mm_cvtss_si32 (__A);
+}
+
+/* Convert the lower SPFP value to a 32-bit integer according to the
+ current rounding mode. */
+
+/* Intel intrinsic. */
+extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtss_si64 (__m128 __A)
+{
+ __m64 res = 0;
+#ifdef _ARCH_PWR8
+ double dtmp;
+ __asm__(
+#ifdef __LITTLE_ENDIAN__
+ "xxsldwi %x0,%x0,%x0,3;\n"
+#endif
+ "xscvspdp %x2,%x0;\n"
+ "fctid %2,%2;\n"
+ "mfvsrd %1,%x2;\n"
+ : "+wa" (__A),
+ "=r" (res),
+ "=f" (dtmp)
+ : );
+#else
+ res = __builtin_llrint(__A[0]);
+#endif
+ return (res);
+}
+
+/* Microsoft intrinsic. */
+extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtss_si64x (__m128 __A)
+{
+ return _mm_cvtss_si64 ((__v4sf) __A);
+}
+
+/* Constants for use with _mm_prefetch. */
+enum _mm_hint
+{
+ /* _MM_HINT_ET is _MM_HINT_T with set 3rd bit. */
+ _MM_HINT_ET0 = 7,
+ _MM_HINT_ET1 = 6,
+ _MM_HINT_T0 = 3,
+ _MM_HINT_T1 = 2,
+ _MM_HINT_T2 = 1,
+ _MM_HINT_NTA = 0
+};
+
+/* Loads one cache line from address P to a location "closer" to the
+ processor. The selector I specifies the type of prefetch operation. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_prefetch (const void *__P, enum _mm_hint __I)
+{
+ /* Current PowerPC will ignores the hint parameters. */
+ __builtin_prefetch (__P);
+}
+
+/* Convert the two lower SPFP values to 32-bit integers according to the
+ current rounding mode. Return the integers in packed form. */
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtps_pi32 (__m128 __A)
+{
+ /* Splat two lower SPFP values to both halves. */
+ __v4sf temp, rounded;
+ __vector unsigned long long result;
+
+ /* Splat two lower SPFP values to both halves. */
+ temp = (__v4sf) vec_splat ((__vector long long)__A, 0);
+ rounded = vec_rint(temp);
+ result = (__vector unsigned long long) vec_cts (rounded, 0);
+
+ return (__m64) ((__vector long long) result)[0];
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_ps2pi (__m128 __A)
+{
+ return _mm_cvtps_pi32 (__A);
+}
+
+/* Truncate the lower SPFP value to a 32-bit integer. */
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttss_si32 (__m128 __A)
+{
+ /* Extract the lower float element. */
+ float temp = __A[0];
+ /* truncate to 32-bit integer and return. */
+ return temp;
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtt_ss2si (__m128 __A)
+{
+ return _mm_cvttss_si32 (__A);
+}
+
+/* Intel intrinsic. */
+extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttss_si64 (__m128 __A)
+{
+ /* Extract the lower float element. */
+ float temp = __A[0];
+ /* truncate to 32-bit integer and return. */
+ return temp;
+}
+
+/* Microsoft intrinsic. */
+extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttss_si64x (__m128 __A)
+{
+ /* Extract the lower float element. */
+ float temp = __A[0];
+ /* truncate to 32-bit integer and return. */
+ return temp;
+}
+
+/* Truncate the two lower SPFP values to 32-bit integers. Return the
+ integers in packed form. */
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttps_pi32 (__m128 __A)
+{
+ __v4sf temp;
+ __vector unsigned long long result;
+
+ /* Splat two lower SPFP values to both halves. */
+ temp = (__v4sf) vec_splat ((__vector long long)__A, 0);
+ result = (__vector unsigned long long) vec_cts (temp, 0);
+
+ return (__m64) ((__vector long long) result)[0];
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtt_ps2pi (__m128 __A)
+{
+ return _mm_cvttps_pi32 (__A);
+}
+
+/* Convert B to a SPFP value and insert it as element zero in A. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi32_ss (__m128 __A, int __B)
+{
+ float temp = __B;
+ __A[0] = temp;
+
+ return __A;
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_si2ss (__m128 __A, int __B)
+{
+ return _mm_cvtsi32_ss (__A, __B);
+}
+
+/* Convert B to a SPFP value and insert it as element zero in A. */
+/* Intel intrinsic. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi64_ss (__m128 __A, long long __B)
+{
+ float temp = __B;
+ __A[0] = temp;
+
+ return __A;
+}
+
+/* Microsoft intrinsic. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi64x_ss (__m128 __A, long long __B)
+{
+ return _mm_cvtsi64_ss (__A, __B);
+}
+
+/* Convert the two 32-bit values in B to SPFP form and insert them
+ as the two lower elements in A. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpi32_ps (__m128 __A, __m64 __B)
+{
+ __vector signed int vm1;
+ __vector float vf1;
+
+ vm1 = (__vector signed int) (__vector unsigned long long) {__B, __B};
+ vf1 = (__vector float) vec_ctf (vm1, 0);
+
+ return ((__m128) (__vector unsigned long long)
+ { ((__vector unsigned long long)vf1) [0],
+ ((__vector unsigned long long)__A) [1]});
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_pi2ps (__m128 __A, __m64 __B)
+{
+ return _mm_cvtpi32_ps (__A, __B);
+}
+
+/* Convert the four signed 16-bit values in A to SPFP form. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpi16_ps (__m64 __A)
+{
+ __vector signed short vs8;
+ __vector signed int vi4;
+ __vector float vf1;
+
+ vs8 = (__vector signed short) (__vector unsigned long long) { __A, __A };
+ vi4 = vec_vupklsh (vs8);
+ vf1 = (__vector float) vec_ctf (vi4, 0);
+
+ return (__m128) vf1;
+}
+
+/* Convert the four unsigned 16-bit values in A to SPFP form. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpu16_ps (__m64 __A)
+{
+ const __vector unsigned short zero =
+ { 0, 0, 0, 0, 0, 0, 0, 0 };
+ __vector unsigned short vs8;
+ __vector unsigned int vi4;
+ __vector float vf1;
+
+ vs8 = (__vector unsigned short) (__vector unsigned long long) { __A, __A };
+ vi4 = (__vector unsigned int) vec_mergel
+#ifdef __LITTLE_ENDIAN__
+ (vs8, zero);
+#else
+ (zero, vs8);
+#endif
+ vf1 = (__vector float) vec_ctf (vi4, 0);
+
+ return (__m128) vf1;
+}
+
+/* Convert the low four signed 8-bit values in A to SPFP form. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpi8_ps (__m64 __A)
+{
+ __vector signed char vc16;
+ __vector signed short vs8;
+ __vector signed int vi4;
+ __vector float vf1;
+
+ vc16 = (__vector signed char) (__vector unsigned long long) { __A, __A };
+ vs8 = vec_vupkhsb (vc16);
+ vi4 = vec_vupkhsh (vs8);
+ vf1 = (__vector float) vec_ctf (vi4, 0);
+
+ return (__m128) vf1;
+}
+
+/* Convert the low four unsigned 8-bit values in A to SPFP form. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+
+_mm_cvtpu8_ps (__m64 __A)
+{
+ const __vector unsigned char zero =
+ { 0, 0, 0, 0, 0, 0, 0, 0 };
+ __vector unsigned char vc16;
+ __vector unsigned short vs8;
+ __vector unsigned int vi4;
+ __vector float vf1;
+
+ vc16 = (__vector unsigned char) (__vector unsigned long long) { __A, __A };
+#ifdef __LITTLE_ENDIAN__
+ vs8 = (__vector unsigned short) vec_mergel (vc16, zero);
+ vi4 = (__vector unsigned int) vec_mergeh (vs8,
+ (__vector unsigned short) zero);
+#else
+ vs8 = (__vector unsigned short) vec_mergel (zero, vc16);
+ vi4 = (__vector unsigned int) vec_mergeh ((__vector unsigned short) zero,
+ vs8);
+#endif
+ vf1 = (__vector float) vec_ctf (vi4, 0);
+
+ return (__m128) vf1;
+}
+
+/* Convert the four signed 32-bit values in A and B to SPFP form. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpi32x2_ps (__m64 __A, __m64 __B)
+{
+ __vector signed int vi4;
+ __vector float vf4;
+
+ vi4 = (__vector signed int) (__vector unsigned long long) { __A, __B };
+ vf4 = (__vector float) vec_ctf (vi4, 0);
+ return (__m128) vf4;
+}
+
+/* Convert the four SPFP values in A to four signed 16-bit integers. */
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtps_pi16 (__m128 __A)
+{
+ __v4sf rounded;
+ __vector signed int temp;
+ __vector unsigned long long result;
+
+ rounded = vec_rint(__A);
+ temp = vec_cts (rounded, 0);
+ result = (__vector unsigned long long) vec_pack (temp, temp);
+
+ return (__m64) ((__vector long long) result)[0];
+}
+
+/* Convert the four SPFP values in A to four signed 8-bit integers. */
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtps_pi8 (__m128 __A)
+{
+ __v4sf rounded;
+ __vector signed int tmp_i;
+ static const __vector signed int zero = {0, 0, 0, 0};
+ __vector signed short tmp_s;
+ __vector signed char res_v;
+
+ rounded = vec_rint(__A);
+ tmp_i = vec_cts (rounded, 0);
+ tmp_s = vec_pack (tmp_i, zero);
+ res_v = vec_pack (tmp_s, tmp_s);
+ return (__m64) ((__vector long long) res_v)[0];
+}
+
+/* Selects four specific SPFP values from A and B based on MASK. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+
+_mm_shuffle_ps (__m128 __A, __m128 __B, int const __mask)
+{
+ unsigned long element_selector_10 = __mask & 0x03;
+ unsigned long element_selector_32 = (__mask >> 2) & 0x03;
+ unsigned long element_selector_54 = (__mask >> 4) & 0x03;
+ unsigned long element_selector_76 = (__mask >> 6) & 0x03;
+ static const unsigned int permute_selectors[4] =
+ {
+#ifdef __LITTLE_ENDIAN__
+ 0x03020100, 0x07060504, 0x0B0A0908, 0x0F0E0D0C
+#else
+ 0x00010203, 0x04050607, 0x08090A0B, 0x0C0D0E0F
+#endif
+ };
+ __vector unsigned int t;
+
+ t[0] = permute_selectors[element_selector_10];
+ t[1] = permute_selectors[element_selector_32];
+ t[2] = permute_selectors[element_selector_54] + 0x10101010;
+ t[3] = permute_selectors[element_selector_76] + 0x10101010;
+ return vec_perm ((__v4sf) __A, (__v4sf)__B, (__vector unsigned char)t);
+}
+
+/* Selects and interleaves the upper two SPFP values from A and B. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_unpackhi_ps (__m128 __A, __m128 __B)
+{
+ return (__m128) vec_vmrglw ((__v4sf) __A, (__v4sf)__B);
+}
+
+/* Selects and interleaves the lower two SPFP values from A and B. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_unpacklo_ps (__m128 __A, __m128 __B)
+{
+ return (__m128) vec_vmrghw ((__v4sf) __A, (__v4sf)__B);
+}
+
+/* Sets the upper two SPFP values with 64-bits of data loaded from P;
+ the lower two values are passed through from A. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loadh_pi (__m128 __A, __m64 const *__P)
+{
+ __vector unsigned long long __a = (__vector unsigned long long)__A;
+ __vector unsigned long long __p = vec_splats(*__P);
+ __a [1] = __p [1];
+
+ return (__m128)__a;
+}
+
+/* Stores the upper two SPFP values of A into P. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_storeh_pi (__m64 *__P, __m128 __A)
+{
+ __vector unsigned long long __a = (__vector unsigned long long) __A;
+
+ *__P = __a[1];
+}
+
+/* Moves the upper two values of B into the lower two values of A. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_movehl_ps (__m128 __A, __m128 __B)
+{
+ return (__m128) vec_mergel ((__vector unsigned long long)__B,
+ (__vector unsigned long long)__A);
+}
+
+/* Moves the lower two values of B into the upper two values of A. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_movelh_ps (__m128 __A, __m128 __B)
+{
+ return (__m128) vec_mergeh ((__vector unsigned long long)__A,
+ (__vector unsigned long long)__B);
+}
+
+/* Sets the lower two SPFP values with 64-bits of data loaded from P;
+ the upper two values are passed through from A. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loadl_pi (__m128 __A, __m64 const *__P)
+{
+ __vector unsigned long long __a = (__vector unsigned long long)__A;
+ __vector unsigned long long __p = vec_splats(*__P);
+ __a [0] = __p [0];
+
+ return (__m128)__a;
+}
+
+/* Stores the lower two SPFP values of A into P. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_storel_pi (__m64 *__P, __m128 __A)
+{
+ __vector unsigned long long __a = (__vector unsigned long long) __A;
+
+ *__P = __a[0];
+}
+
+#ifdef _ARCH_PWR8
+/* Intrinsic functions that require PowerISA 2.07 minimum. */
+
+/* Creates a 4-bit mask from the most significant bits of the SPFP values. */
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_movemask_ps (__m128 __A)
+{
+ __vector unsigned long long result;
+ static const __vector unsigned int perm_mask =
+ {
+#ifdef __LITTLE_ENDIAN__
+ 0x00204060, 0x80808080, 0x80808080, 0x80808080
+#else
+ 0x80808080, 0x80808080, 0x80808080, 0x00204060
+#endif
+ };
+
+ result = ((__vector unsigned long long)
+ vec_vbpermq ((__vector unsigned char) __A,
+ (__vector unsigned char) perm_mask));
+
+#ifdef __LITTLE_ENDIAN__
+ return result[1];
+#else
+ return result[0];
+#endif
+}
+#endif /* _ARCH_PWR8 */
+
+/* Create a vector with all four elements equal to *P. */
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_load1_ps (float const *__P)
+{
+ return _mm_set1_ps (*__P);
+}
+
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_load_ps1 (float const *__P)
+{
+ return _mm_load1_ps (__P);
+}
+
+/* Extracts one of the four words of A. The selector N must be immediate. */
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_extract_pi16 (__m64 const __A, int const __N)
+{
+ unsigned int shiftr = __N & 3;
+#ifdef __BIG_ENDIAN__
+ shiftr = 3 - shiftr;
+#endif
+
+ return ((__A >> (shiftr * 16)) & 0xffff);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pextrw (__m64 const __A, int const __N)
+{
+ return _mm_extract_pi16 (__A, __N);
+}
+
+/* Inserts word D into one of four words of A. The selector N must be
+ immediate. */
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_insert_pi16 (__m64 const __A, int const __D, int const __N)
+{
+ const int shiftl = (__N & 3) * 16;
+ const __m64 shiftD = (const __m64) __D << shiftl;
+ const __m64 mask = 0xffffUL << shiftl;
+ __m64 result = (__A & (~mask)) | (shiftD & mask);
+
+ return (result);
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pinsrw (__m64 const __A, int const __D, int const __N)
+{
+ return _mm_insert_pi16 (__A, __D, __N);
+}
+
+/* Compute the element-wise maximum of signed 16-bit values. */
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+
+_mm_max_pi16 (__m64 __A, __m64 __B)
+{
+#if _ARCH_PWR8
+ __vector signed short a, b, r;
+ __vector __bool short c;
+
+ a = (__vector signed short)vec_splats (__A);
+ b = (__vector signed short)vec_splats (__B);
+ c = (__vector __bool short)vec_cmpgt (a, b);
+ r = vec_sel (b, a, c);
+ return (__m64) ((__vector long long) r)[0];
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __A;
+ m2.as_m64 = __B;
+
+ res.as_short[0] =
+ (m1.as_short[0] > m2.as_short[0]) ? m1.as_short[0] : m2.as_short[0];
+ res.as_short[1] =
+ (m1.as_short[1] > m2.as_short[1]) ? m1.as_short[1] : m2.as_short[1];
+ res.as_short[2] =
+ (m1.as_short[2] > m2.as_short[2]) ? m1.as_short[2] : m2.as_short[2];
+ res.as_short[3] =
+ (m1.as_short[3] > m2.as_short[3]) ? m1.as_short[3] : m2.as_short[3];
+
+ return (__m64) res.as_m64;
+#endif
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pmaxsw (__m64 __A, __m64 __B)
+{
+ return _mm_max_pi16 (__A, __B);
+}
+
+/* Compute the element-wise maximum of unsigned 8-bit values. */
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_max_pu8 (__m64 __A, __m64 __B)
+{
+#if _ARCH_PWR8
+ __vector unsigned char a, b, r;
+ __vector __bool char c;
+
+ a = (__vector unsigned char)vec_splats (__A);
+ b = (__vector unsigned char)vec_splats (__B);
+ c = (__vector __bool char)vec_cmpgt (a, b);
+ r = vec_sel (b, a, c);
+ return (__m64) ((__vector long long) r)[0];
+#else
+ __m64_union m1, m2, res;
+ long i;
+
+ m1.as_m64 = __A;
+ m2.as_m64 = __B;
+
+
+ for (i = 0; i < 8; i++)
+ res.as_char[i] =
+ ((unsigned char) m1.as_char[i] > (unsigned char) m2.as_char[i]) ?
+ m1.as_char[i] : m2.as_char[i];
+
+ return (__m64) res.as_m64;
+#endif
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pmaxub (__m64 __A, __m64 __B)
+{
+ return _mm_max_pu8 (__A, __B);
+}
+
+/* Compute the element-wise minimum of signed 16-bit values. */
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_min_pi16 (__m64 __A, __m64 __B)
+{
+#if _ARCH_PWR8
+ __vector signed short a, b, r;
+ __vector __bool short c;
+
+ a = (__vector signed short)vec_splats (__A);
+ b = (__vector signed short)vec_splats (__B);
+ c = (__vector __bool short)vec_cmplt (a, b);
+ r = vec_sel (b, a, c);
+ return (__m64) ((__vector long long) r)[0];
+#else
+ __m64_union m1, m2, res;
+
+ m1.as_m64 = __A;
+ m2.as_m64 = __B;
+
+ res.as_short[0] =
+ (m1.as_short[0] < m2.as_short[0]) ? m1.as_short[0] : m2.as_short[0];
+ res.as_short[1] =
+ (m1.as_short[1] < m2.as_short[1]) ? m1.as_short[1] : m2.as_short[1];
+ res.as_short[2] =
+ (m1.as_short[2] < m2.as_short[2]) ? m1.as_short[2] : m2.as_short[2];
+ res.as_short[3] =
+ (m1.as_short[3] < m2.as_short[3]) ? m1.as_short[3] : m2.as_short[3];
+
+ return (__m64) res.as_m64;
+#endif
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pminsw (__m64 __A, __m64 __B)
+{
+ return _mm_min_pi16 (__A, __B);
+}
+
+/* Compute the element-wise minimum of unsigned 8-bit values. */
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_min_pu8 (__m64 __A, __m64 __B)
+{
+#if _ARCH_PWR8
+ __vector unsigned char a, b, r;
+ __vector __bool char c;
+
+ a = (__vector unsigned char)vec_splats (__A);
+ b = (__vector unsigned char)vec_splats (__B);
+ c = (__vector __bool char)vec_cmplt (a, b);
+ r = vec_sel (b, a, c);
+ return (__m64) ((__vector long long) r)[0];
+#else
+ __m64_union m1, m2, res;
+ long i;
+
+ m1.as_m64 = __A;
+ m2.as_m64 = __B;
+
+
+ for (i = 0; i < 8; i++)
+ res.as_char[i] =
+ ((unsigned char) m1.as_char[i] < (unsigned char) m2.as_char[i]) ?
+ m1.as_char[i] : m2.as_char[i];
+
+ return (__m64) res.as_m64;
+#endif
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pminub (__m64 __A, __m64 __B)
+{
+ return _mm_min_pu8 (__A, __B);
+}
+
+/* Create an 8-bit mask of the signs of 8-bit values. */
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_movemask_pi8 (__m64 __A)
+{
+ unsigned long long p =
+#ifdef __LITTLE_ENDIAN__
+ 0x0008101820283038UL; // permute control for sign bits
+#else
+ 0x3830282018100800UL; // permute control for sign bits
+#endif
+ return __builtin_bpermd (p, __A);
+}
+
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pmovmskb (__m64 __A)
+{
+ return _mm_movemask_pi8 (__A);
+}
+
+/* Multiply four unsigned 16-bit values in A by four unsigned 16-bit values
+ in B and produce the high 16 bits of the 32-bit results. */
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mulhi_pu16 (__m64 __A, __m64 __B)
+{
+ __vector unsigned short a, b;
+ __vector unsigned short c;
+ __vector unsigned int w0, w1;
+ __vector unsigned char xform1 = {
+#ifdef __LITTLE_ENDIAN__
+ 0x02, 0x03, 0x12, 0x13, 0x06, 0x07, 0x16, 0x17,
+ 0x0A, 0x0B, 0x1A, 0x1B, 0x0E, 0x0F, 0x1E, 0x1F
+#else
+ 0x00, 0x01, 0x10, 0x11, 0x04, 0x05, 0x14, 0x15,
+ 0x00, 0x01, 0x10, 0x11, 0x04, 0x05, 0x14, 0x15
+#endif
+ };
+
+ a = (__vector unsigned short)vec_splats (__A);
+ b = (__vector unsigned short)vec_splats (__B);
+
+ w0 = vec_vmuleuh (a, b);
+ w1 = vec_vmulouh (a, b);
+ c = (__vector unsigned short)vec_perm (w0, w1, xform1);
+
+ return (__m64) ((__vector long long) c)[0];
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pmulhuw (__m64 __A, __m64 __B)
+{
+ return _mm_mulhi_pu16 (__A, __B);
+}
+
+/* Return a combination of the four 16-bit values in A. The selector
+ must be an immediate. */
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_shuffle_pi16 (__m64 __A, int const __N)
+{
+ unsigned long element_selector_10 = __N & 0x03;
+ unsigned long element_selector_32 = (__N >> 2) & 0x03;
+ unsigned long element_selector_54 = (__N >> 4) & 0x03;
+ unsigned long element_selector_76 = (__N >> 6) & 0x03;
+ static const unsigned short permute_selectors[4] =
+ {
+#ifdef __LITTLE_ENDIAN__
+ 0x0908, 0x0B0A, 0x0D0C, 0x0F0E
+#else
+ 0x0607, 0x0405, 0x0203, 0x0001
+#endif
+ };
+ __m64_union t;
+ __vector unsigned long long a, p, r;
+
+#ifdef __LITTLE_ENDIAN__
+ t.as_short[0] = permute_selectors[element_selector_10];
+ t.as_short[1] = permute_selectors[element_selector_32];
+ t.as_short[2] = permute_selectors[element_selector_54];
+ t.as_short[3] = permute_selectors[element_selector_76];
+#else
+ t.as_short[3] = permute_selectors[element_selector_10];
+ t.as_short[2] = permute_selectors[element_selector_32];
+ t.as_short[1] = permute_selectors[element_selector_54];
+ t.as_short[0] = permute_selectors[element_selector_76];
+#endif
+ p = vec_splats (t.as_m64);
+ a = vec_splats (__A);
+ r = vec_perm (a, a, (__vector unsigned char)p);
+ return (__m64) ((__vector long long) r)[0];
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pshufw (__m64 __A, int const __N)
+{
+ return _mm_shuffle_pi16 (__A, __N);
+}
+
+/* Conditionally store byte elements of A into P. The high bit of each
+ byte in the selector N determines whether the corresponding byte from
+ A is stored. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskmove_si64 (__m64 __A, __m64 __N, char *__P)
+{
+ __m64 hibit = 0x8080808080808080UL;
+ __m64 mask, tmp;
+ __m64 *p = (__m64*)__P;
+
+ tmp = *p;
+ mask = _mm_cmpeq_pi8 ((__N & hibit), hibit);
+ tmp = (tmp & (~mask)) | (__A & mask);
+ *p = tmp;
+}
+
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_maskmovq (__m64 __A, __m64 __N, char *__P)
+{
+ _mm_maskmove_si64 (__A, __N, __P);
+}
+
+/* Compute the rounded averages of the unsigned 8-bit values in A and B. */
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_avg_pu8 (__m64 __A, __m64 __B)
+{
+ __vector unsigned char a, b, c;
+
+ a = (__vector unsigned char)vec_splats (__A);
+ b = (__vector unsigned char)vec_splats (__B);
+ c = vec_avg (a, b);
+ return (__m64) ((__vector long long) c)[0];
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pavgb (__m64 __A, __m64 __B)
+{
+ return _mm_avg_pu8 (__A, __B);
+}
+
+/* Compute the rounded averages of the unsigned 16-bit values in A and B. */
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_avg_pu16 (__m64 __A, __m64 __B)
+{
+ __vector unsigned short a, b, c;
+
+ a = (__vector unsigned short)vec_splats (__A);
+ b = (__vector unsigned short)vec_splats (__B);
+ c = vec_avg (a, b);
+ return (__m64) ((__vector long long) c)[0];
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pavgw (__m64 __A, __m64 __B)
+{
+ return _mm_avg_pu16 (__A, __B);
+}
+
+/* Compute the sum of the absolute differences of the unsigned 8-bit
+ values in A and B. Return the value in the lower 16-bit word; the
+ upper words are cleared. */
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sad_pu8 (__m64 __A, __m64 __B)
+{
+ __vector unsigned char a, b;
+ __vector unsigned char vmin, vmax, vabsdiff;
+ __vector signed int vsum;
+ const __vector unsigned int zero =
+ { 0, 0, 0, 0 };
+ __m64_union result = {0};
+
+ a = (__vector unsigned char) (__vector unsigned long long) { 0UL, __A };
+ b = (__vector unsigned char) (__vector unsigned long long) { 0UL, __B };
+ vmin = vec_min (a, b);
+ vmax = vec_max (a, b);
+ vabsdiff = vec_sub (vmax, vmin);
+ /* Sum four groups of bytes into integers. */
+ vsum = (__vector signed int) vec_sum4s (vabsdiff, zero);
+ /* Sum across four integers with integer result. */
+ vsum = vec_sums (vsum, (__vector signed int) zero);
+ /* The sum is in the right most 32-bits of the vector result.
+ Transfer to a GPR and truncate to 16 bits. */
+ result.as_short[0] = vsum[3];
+ return result.as_m64;
+}
+
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_psadbw (__m64 __A, __m64 __B)
+{
+ return _mm_sad_pu8 (__A, __B);
+}
+
+/* Stores the data in A to the address P without polluting the caches. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_stream_pi (__m64 *__P, __m64 __A)
+{
+ /* Use the data cache block touch for store transient. */
+ __asm__ (
+ " dcbtstt 0,%0"
+ :
+ : "b" (__P)
+ : "memory"
+ );
+ *__P = __A;
+}
+
+/* Likewise. The address must be 16-byte aligned. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_stream_ps (float *__P, __m128 __A)
+{
+ /* Use the data cache block touch for store transient. */
+ __asm__ (
+ " dcbtstt 0,%0"
+ :
+ : "b" (__P)
+ : "memory"
+ );
+ _mm_store_ps (__P, __A);
+}
+
+/* Guarantees that every preceding store is globally visible before
+ any subsequent store. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sfence (void)
+{
+ /* Generate a light weight sync. */
+ __atomic_thread_fence (__ATOMIC_RELEASE);
+}
+
+/* The execution of the next instruction is delayed by an implementation
+ specific amount of time. The instruction does not modify the
+ architectural state. This is after the pop_options pragma because
+ it does not require SSE support in the processor--the encoding is a
+ nop on processors that do not support it. */
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_pause (void)
+{
+ /* There is no exact match with this construct, but the following is
+ close to the desired effect. */
+#if _ARCH_PWR8
+ /* On power8 and later processors we can depend on Program Priority
+ (PRI) and associated "very low" PPI setting. Since we don't know
+ what PPI this thread is running at we: 1) save the current PRI
+ from the PPR SPR into a local GRP, 2) set the PRI to "very low*
+ via the special or 31,31,31 encoding. 3) issue an "isync" to
+ insure the PRI change takes effect before we execute any more
+ instructions.
+ Now we can execute a lwsync (release barrier) while we execute
+ this thread at "very low" PRI. Finally we restore the original
+ PRI and continue execution. */
+ unsigned long __PPR;
+
+ __asm__ volatile (
+ " mfppr %0;"
+ " or 31,31,31;"
+ " isync;"
+ " lwsync;"
+ " isync;"
+ " mtppr %0;"
+ : "=r" (__PPR)
+ :
+ : "memory"
+ );
+#else
+ /* For older processor where we may not even have Program Priority
+ controls we can only depend on Heavy Weight Sync. */
+ __atomic_thread_fence (__ATOMIC_SEQ_CST);
+#endif
+}
+
+/* Transpose the 4x4 matrix composed of row[0-3]. */
+#define _MM_TRANSPOSE4_PS(row0, row1, row2, row3) \
+do { \
+ __v4sf __r0 = (row0), __r1 = (row1), __r2 = (row2), __r3 = (row3); \
+ __v4sf __t0 = vec_vmrghw (__r0, __r1); \
+ __v4sf __t1 = vec_vmrghw (__r2, __r3); \
+ __v4sf __t2 = vec_vmrglw (__r0, __r1); \
+ __v4sf __t3 = vec_vmrglw (__r2, __r3); \
+ (row0) = (__v4sf)vec_mergeh ((__vector long long)__t0, \
+ (__vector long long)__t1); \
+ (row1) = (__v4sf)vec_mergel ((__vector long long)__t0, \
+ (__vector long long)__t1); \
+ (row2) = (__v4sf)vec_mergeh ((__vector long long)__t2, \
+ (__vector long long)__t3); \
+ (row3) = (__v4sf)vec_mergel ((__vector long long)__t2, \
+ (__vector long long)__t3); \
+} while (0)
+
+/* For backward source compatibility. */
+//# include <emmintrin.h>
+
+#else
+#include_next <xmmintrin.h>
+#endif /* defined(__linux__) && defined(__ppc64__) */
+
+#endif /* _XMMINTRIN_H_INCLUDED */
/*===---- prfchwintrin.h - PREFETCHW intrinsic -----------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#ifndef __PRFCHWINTRIN_H
#define __PRFCHWINTRIN_H
-#if defined(__PRFCHW__) || defined(__3dNOW__)
-/// \brief Loads a memory sequence containing the specified memory address into
+/// Loads a memory sequence containing the specified memory address into
/// all data cache levels. The cache-coherency state is set to exclusive.
/// Data can be read from and written to the cache line without additional
/// delay.
__builtin_prefetch (__P, 0, 3 /* _MM_HINT_T0 */);
}
-/// \brief Loads a memory sequence containing the specified memory address into
+/// Loads a memory sequence containing the specified memory address into
/// the L1 data cache and sets the cache-coherency to modified. This
/// provides a hint to the processor that the cache line will be modified.
/// It is intended for use when the cache line will be written to shortly
{
__builtin_prefetch (__P, 1, 3 /* _MM_HINT_T0 */);
}
-#endif
#endif /* __PRFCHWINTRIN_H */
--- /dev/null
+/*===-- InstrProfData.inc - instr profiling runtime structures -*- C++ -*-=== *\
+|*
+|* Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+|* See https://llvm.org/LICENSE.txt for license information.
+|* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+|*
+\*===----------------------------------------------------------------------===*/
+/*
+ * This is the master file that defines all the data structure, signature,
+ * constant literals that are shared across profiling runtime library,
+ * compiler (instrumentation), and host tools (reader/writer). The entities
+ * defined in this file affect the profile runtime ABI, the raw profile format,
+ * or both.
+ *
+ * The file has two identical copies. The master copy lives in LLVM and
+ * the other one sits in compiler-rt/lib/profile directory. To make changes
+ * in this file, first modify the master copy and copy it over to compiler-rt.
+ * Testing of any change in this file can start only after the two copies are
+ * synced up.
+ *
+ * The first part of the file includes macros that defines types, names, and
+ * initializers for the member fields of the core data structures. The field
+ * declarations for one structure is enabled by defining the field activation
+ * macro associated with that structure. Only one field activation record
+ * can be defined at one time and the rest definitions will be filtered out by
+ * the preprocessor.
+ *
+ * Examples of how the template is used to instantiate structure definition:
+ * 1. To declare a structure:
+ *
+ * struct ProfData {
+ * #define INSTR_PROF_DATA(Type, LLVMType, Name, Initializer) \
+ * Type Name;
+ * #include "llvm/ProfileData/InstrProfData.inc"
+ * };
+ *
+ * 2. To construct LLVM type arrays for the struct type:
+ *
+ * Type *DataTypes[] = {
+ * #define INSTR_PROF_DATA(Type, LLVMType, Name, Initializer) \
+ * LLVMType,
+ * #include "llvm/ProfileData/InstrProfData.inc"
+ * };
+ *
+ * 4. To construct constant array for the initializers:
+ * #define INSTR_PROF_DATA(Type, LLVMType, Name, Initializer) \
+ * Initializer,
+ * Constant *ConstantVals[] = {
+ * #include "llvm/ProfileData/InstrProfData.inc"
+ * };
+ *
+ *
+ * The second part of the file includes definitions all other entities that
+ * are related to runtime ABI and format. When no field activation macro is
+ * defined, this file can be included to introduce the definitions.
+ *
+\*===----------------------------------------------------------------------===*/
+
+/* Functions marked with INSTR_PROF_VISIBILITY must have hidden visibility in
+ * the compiler runtime. */
+#ifndef INSTR_PROF_VISIBILITY
+#define INSTR_PROF_VISIBILITY
+#endif
+
+/* INSTR_PROF_DATA start. */
+/* Definition of member fields of the per-function control structure. */
+#ifndef INSTR_PROF_DATA
+#define INSTR_PROF_DATA(Type, LLVMType, Name, Initializer)
+#else
+#define INSTR_PROF_DATA_DEFINED
+#endif
+INSTR_PROF_DATA(const uint64_t, llvm::Type::getInt64Ty(Ctx), NameRef, \
+ ConstantInt::get(llvm::Type::getInt64Ty(Ctx), \
+ IndexedInstrProf::ComputeHash(getPGOFuncNameVarInitializer(Inc->getName()))))
+INSTR_PROF_DATA(const uint64_t, llvm::Type::getInt64Ty(Ctx), FuncHash, \
+ ConstantInt::get(llvm::Type::getInt64Ty(Ctx), \
+ Inc->getHash()->getZExtValue()))
+INSTR_PROF_DATA(const IntPtrT, llvm::Type::getInt64PtrTy(Ctx), CounterPtr, \
+ ConstantExpr::getBitCast(CounterPtr, \
+ llvm::Type::getInt64PtrTy(Ctx)))
+/* This is used to map function pointers for the indirect call targets to
+ * function name hashes during the conversion from raw to merged profile
+ * data.
+ */
+INSTR_PROF_DATA(const IntPtrT, llvm::Type::getInt8PtrTy(Ctx), FunctionPointer, \
+ FunctionAddr)
+INSTR_PROF_DATA(IntPtrT, llvm::Type::getInt8PtrTy(Ctx), Values, \
+ ValuesPtrExpr)
+INSTR_PROF_DATA(const uint32_t, llvm::Type::getInt32Ty(Ctx), NumCounters, \
+ ConstantInt::get(llvm::Type::getInt32Ty(Ctx), NumCounters))
+INSTR_PROF_DATA(const uint16_t, Int16ArrayTy, NumValueSites[IPVK_Last+1], \
+ ConstantArray::get(Int16ArrayTy, Int16ArrayVals))
+#undef INSTR_PROF_DATA
+/* INSTR_PROF_DATA end. */
+
+
+/* This is an internal data structure used by value profiler. It
+ * is defined here to allow serialization code sharing by LLVM
+ * to be used in unit test.
+ *
+ * typedef struct ValueProfNode {
+ * // InstrProfValueData VData;
+ * uint64_t Value;
+ * uint64_t Count;
+ * struct ValueProfNode *Next;
+ * } ValueProfNode;
+ */
+/* INSTR_PROF_VALUE_NODE start. */
+#ifndef INSTR_PROF_VALUE_NODE
+#define INSTR_PROF_VALUE_NODE(Type, LLVMType, Name, Initializer)
+#else
+#define INSTR_PROF_DATA_DEFINED
+#endif
+INSTR_PROF_VALUE_NODE(uint64_t, llvm::Type::getInt64Ty(Ctx), Value, \
+ ConstantInt::get(llvm::Type::GetInt64Ty(Ctx), 0))
+INSTR_PROF_VALUE_NODE(uint64_t, llvm::Type::getInt64Ty(Ctx), Count, \
+ ConstantInt::get(llvm::Type::GetInt64Ty(Ctx), 0))
+INSTR_PROF_VALUE_NODE(PtrToNodeT, llvm::Type::getInt8PtrTy(Ctx), Next, \
+ ConstantInt::get(llvm::Type::GetInt8PtrTy(Ctx), 0))
+#undef INSTR_PROF_VALUE_NODE
+/* INSTR_PROF_VALUE_NODE end. */
+
+/* INSTR_PROF_RAW_HEADER start */
+/* Definition of member fields of the raw profile header data structure. */
+#ifndef INSTR_PROF_RAW_HEADER
+#define INSTR_PROF_RAW_HEADER(Type, Name, Initializer)
+#else
+#define INSTR_PROF_DATA_DEFINED
+#endif
+INSTR_PROF_RAW_HEADER(uint64_t, Magic, __llvm_profile_get_magic())
+INSTR_PROF_RAW_HEADER(uint64_t, Version, __llvm_profile_get_version())
+INSTR_PROF_RAW_HEADER(uint64_t, DataSize, DataSize)
+INSTR_PROF_RAW_HEADER(uint64_t, PaddingBytesBeforeCounters, PaddingBytesBeforeCounters)
+INSTR_PROF_RAW_HEADER(uint64_t, CountersSize, CountersSize)
+INSTR_PROF_RAW_HEADER(uint64_t, PaddingBytesAfterCounters, PaddingBytesAfterCounters)
+INSTR_PROF_RAW_HEADER(uint64_t, NamesSize, NamesSize)
+INSTR_PROF_RAW_HEADER(uint64_t, CountersDelta, (uintptr_t)CountersBegin)
+INSTR_PROF_RAW_HEADER(uint64_t, NamesDelta, (uintptr_t)NamesBegin)
+INSTR_PROF_RAW_HEADER(uint64_t, ValueKindLast, IPVK_Last)
+#undef INSTR_PROF_RAW_HEADER
+/* INSTR_PROF_RAW_HEADER end */
+
+/* VALUE_PROF_FUNC_PARAM start */
+/* Definition of parameter types of the runtime API used to do value profiling
+ * for a given value site.
+ */
+#ifndef VALUE_PROF_FUNC_PARAM
+#define VALUE_PROF_FUNC_PARAM(ArgType, ArgName, ArgLLVMType)
+#define INSTR_PROF_COMMA
+#else
+#define INSTR_PROF_DATA_DEFINED
+#define INSTR_PROF_COMMA ,
+#endif
+VALUE_PROF_FUNC_PARAM(uint64_t, TargetValue, Type::getInt64Ty(Ctx)) \
+ INSTR_PROF_COMMA
+VALUE_PROF_FUNC_PARAM(void *, Data, Type::getInt8PtrTy(Ctx)) INSTR_PROF_COMMA
+#ifndef VALUE_RANGE_PROF
+VALUE_PROF_FUNC_PARAM(uint32_t, CounterIndex, Type::getInt32Ty(Ctx))
+#else /* VALUE_RANGE_PROF */
+VALUE_PROF_FUNC_PARAM(uint32_t, CounterIndex, Type::getInt32Ty(Ctx)) \
+ INSTR_PROF_COMMA
+VALUE_PROF_FUNC_PARAM(uint64_t, PreciseRangeStart, Type::getInt64Ty(Ctx)) \
+ INSTR_PROF_COMMA
+VALUE_PROF_FUNC_PARAM(uint64_t, PreciseRangeLast, Type::getInt64Ty(Ctx)) \
+ INSTR_PROF_COMMA
+VALUE_PROF_FUNC_PARAM(uint64_t, LargeValue, Type::getInt64Ty(Ctx))
+#endif /*VALUE_RANGE_PROF */
+#undef VALUE_PROF_FUNC_PARAM
+#undef INSTR_PROF_COMMA
+/* VALUE_PROF_FUNC_PARAM end */
+
+/* VALUE_PROF_KIND start */
+#ifndef VALUE_PROF_KIND
+#define VALUE_PROF_KIND(Enumerator, Value, Descr)
+#else
+#define INSTR_PROF_DATA_DEFINED
+#endif
+/* For indirect function call value profiling, the addresses of the target
+ * functions are profiled by the instrumented code. The target addresses are
+ * written in the raw profile data and converted to target function name's MD5
+ * hash by the profile reader during deserialization. Typically, this happens
+ * when the raw profile data is read during profile merging.
+ *
+ * For this remapping the ProfData is used. ProfData contains both the function
+ * name hash and the function address.
+ */
+VALUE_PROF_KIND(IPVK_IndirectCallTarget, 0, "indirect call target")
+/* For memory intrinsic functions size profiling. */
+VALUE_PROF_KIND(IPVK_MemOPSize, 1, "memory intrinsic functions size")
+/* These two kinds must be the last to be
+ * declared. This is to make sure the string
+ * array created with the template can be
+ * indexed with the kind value.
+ */
+VALUE_PROF_KIND(IPVK_First, IPVK_IndirectCallTarget, "first")
+VALUE_PROF_KIND(IPVK_Last, IPVK_MemOPSize, "last")
+
+#undef VALUE_PROF_KIND
+/* VALUE_PROF_KIND end */
+
+#undef COVMAP_V2_OR_V3
+#ifdef COVMAP_V2
+#define COVMAP_V2_OR_V3
+#endif
+#ifdef COVMAP_V3
+#define COVMAP_V2_OR_V3
+#endif
+
+/* COVMAP_FUNC_RECORD start */
+/* Definition of member fields of the function record structure in coverage
+ * map.
+ */
+#ifndef COVMAP_FUNC_RECORD
+#define COVMAP_FUNC_RECORD(Type, LLVMType, Name, Initializer)
+#else
+#define INSTR_PROF_DATA_DEFINED
+#endif
+#ifdef COVMAP_V1
+COVMAP_FUNC_RECORD(const IntPtrT, llvm::Type::getInt8PtrTy(Ctx), \
+ NamePtr, llvm::ConstantExpr::getBitCast(NamePtr, \
+ llvm::Type::getInt8PtrTy(Ctx)))
+COVMAP_FUNC_RECORD(const uint32_t, llvm::Type::getInt32Ty(Ctx), NameSize, \
+ llvm::ConstantInt::get(llvm::Type::getInt32Ty(Ctx), \
+ NameValue.size()))
+#endif
+#ifdef COVMAP_V2_OR_V3
+COVMAP_FUNC_RECORD(const int64_t, llvm::Type::getInt64Ty(Ctx), NameRef, \
+ llvm::ConstantInt::get( \
+ llvm::Type::getInt64Ty(Ctx), NameHash))
+#endif
+COVMAP_FUNC_RECORD(const uint32_t, llvm::Type::getInt32Ty(Ctx), DataSize, \
+ llvm::ConstantInt::get( \
+ llvm::Type::getInt32Ty(Ctx), CoverageMapping.size()))
+COVMAP_FUNC_RECORD(const uint64_t, llvm::Type::getInt64Ty(Ctx), FuncHash, \
+ llvm::ConstantInt::get( \
+ llvm::Type::getInt64Ty(Ctx), FuncHash))
+#ifdef COVMAP_V3
+COVMAP_FUNC_RECORD(const uint64_t, llvm::Type::getInt64Ty(Ctx), FilenamesRef, \
+ llvm::ConstantInt::get( \
+ llvm::Type::getInt64Ty(Ctx), FilenamesRef))
+COVMAP_FUNC_RECORD(const char, \
+ llvm::ArrayType::get(llvm::Type::getInt8Ty(Ctx), \
+ CoverageMapping.size()), \
+ CoverageMapping,
+ llvm::ConstantDataArray::getRaw( \
+ CoverageMapping, CoverageMapping.size(), \
+ llvm::Type::getInt8Ty(Ctx)))
+#endif
+#undef COVMAP_FUNC_RECORD
+/* COVMAP_FUNC_RECORD end. */
+
+/* COVMAP_HEADER start */
+/* Definition of member fields of coverage map header.
+ */
+#ifndef COVMAP_HEADER
+#define COVMAP_HEADER(Type, LLVMType, Name, Initializer)
+#else
+#define INSTR_PROF_DATA_DEFINED
+#endif
+COVMAP_HEADER(uint32_t, Int32Ty, NRecords, \
+ llvm::ConstantInt::get(Int32Ty, NRecords))
+COVMAP_HEADER(uint32_t, Int32Ty, FilenamesSize, \
+ llvm::ConstantInt::get(Int32Ty, FilenamesSize))
+COVMAP_HEADER(uint32_t, Int32Ty, CoverageSize, \
+ llvm::ConstantInt::get(Int32Ty, CoverageMappingSize))
+COVMAP_HEADER(uint32_t, Int32Ty, Version, \
+ llvm::ConstantInt::get(Int32Ty, CovMapVersion::CurrentVersion))
+#undef COVMAP_HEADER
+/* COVMAP_HEADER end. */
+
+
+#ifdef INSTR_PROF_SECT_ENTRY
+#define INSTR_PROF_DATA_DEFINED
+INSTR_PROF_SECT_ENTRY(IPSK_data, \
+ INSTR_PROF_QUOTE(INSTR_PROF_DATA_COMMON), \
+ INSTR_PROF_DATA_COFF, "__DATA,")
+INSTR_PROF_SECT_ENTRY(IPSK_cnts, \
+ INSTR_PROF_QUOTE(INSTR_PROF_CNTS_COMMON), \
+ INSTR_PROF_CNTS_COFF, "__DATA,")
+INSTR_PROF_SECT_ENTRY(IPSK_name, \
+ INSTR_PROF_QUOTE(INSTR_PROF_NAME_COMMON), \
+ INSTR_PROF_NAME_COFF, "__DATA,")
+INSTR_PROF_SECT_ENTRY(IPSK_vals, \
+ INSTR_PROF_QUOTE(INSTR_PROF_VALS_COMMON), \
+ INSTR_PROF_VALS_COFF, "__DATA,")
+INSTR_PROF_SECT_ENTRY(IPSK_vnodes, \
+ INSTR_PROF_QUOTE(INSTR_PROF_VNODES_COMMON), \
+ INSTR_PROF_VNODES_COFF, "__DATA,")
+INSTR_PROF_SECT_ENTRY(IPSK_covmap, \
+ INSTR_PROF_QUOTE(INSTR_PROF_COVMAP_COMMON), \
+ INSTR_PROF_COVMAP_COFF, "__LLVM_COV,")
+INSTR_PROF_SECT_ENTRY(IPSK_covfun, \
+ INSTR_PROF_QUOTE(INSTR_PROF_COVFUN_COMMON), \
+ INSTR_PROF_COVFUN_COFF, "__LLVM_COV,")
+INSTR_PROF_SECT_ENTRY(IPSK_orderfile, \
+ INSTR_PROF_QUOTE(INSTR_PROF_ORDERFILE_COMMON), \
+ INSTR_PROF_QUOTE(INSTR_PROF_ORDERFILE_COFF), "__DATA,")
+
+#undef INSTR_PROF_SECT_ENTRY
+#endif
+
+
+#ifdef INSTR_PROF_VALUE_PROF_DATA
+#define INSTR_PROF_DATA_DEFINED
+
+#define INSTR_PROF_MAX_NUM_VAL_PER_SITE 255
+/*!
+ * This is the header of the data structure that defines the on-disk
+ * layout of the value profile data of a particular kind for one function.
+ */
+typedef struct ValueProfRecord {
+ /* The kind of the value profile record. */
+ uint32_t Kind;
+ /*
+ * The number of value profile sites. It is guaranteed to be non-zero;
+ * otherwise the record for this kind won't be emitted.
+ */
+ uint32_t NumValueSites;
+ /*
+ * The first element of the array that stores the number of profiled
+ * values for each value site. The size of the array is NumValueSites.
+ * Since NumValueSites is greater than zero, there is at least one
+ * element in the array.
+ */
+ uint8_t SiteCountArray[1];
+
+ /*
+ * The fake declaration is for documentation purpose only.
+ * Align the start of next field to be on 8 byte boundaries.
+ uint8_t Padding[X];
+ */
+
+ /* The array of value profile data. The size of the array is the sum
+ * of all elements in SiteCountArray[].
+ InstrProfValueData ValueData[];
+ */
+
+#ifdef __cplusplus
+ /*!
+ * Return the number of value sites.
+ */
+ uint32_t getNumValueSites() const { return NumValueSites; }
+ /*!
+ * Read data from this record and save it to Record.
+ */
+ void deserializeTo(InstrProfRecord &Record,
+ InstrProfSymtab *SymTab);
+ /*
+ * In-place byte swap:
+ * Do byte swap for this instance. \c Old is the original order before
+ * the swap, and \c New is the New byte order.
+ */
+ void swapBytes(support::endianness Old, support::endianness New);
+#endif
+} ValueProfRecord;
+
+/*!
+ * Per-function header/control data structure for value profiling
+ * data in indexed format.
+ */
+typedef struct ValueProfData {
+ /*
+ * Total size in bytes including this field. It must be a multiple
+ * of sizeof(uint64_t).
+ */
+ uint32_t TotalSize;
+ /*
+ *The number of value profile kinds that has value profile data.
+ * In this implementation, a value profile kind is considered to
+ * have profile data if the number of value profile sites for the
+ * kind is not zero. More aggressively, the implementation can
+ * choose to check the actual data value: if none of the value sites
+ * has any profiled values, the kind can be skipped.
+ */
+ uint32_t NumValueKinds;
+
+ /*
+ * Following are a sequence of variable length records. The prefix/header
+ * of each record is defined by ValueProfRecord type. The number of
+ * records is NumValueKinds.
+ * ValueProfRecord Record_1;
+ * ValueProfRecord Record_N;
+ */
+
+#if __cplusplus
+ /*!
+ * Return the total size in bytes of the on-disk value profile data
+ * given the data stored in Record.
+ */
+ static uint32_t getSize(const InstrProfRecord &Record);
+ /*!
+ * Return a pointer to \c ValueProfData instance ready to be streamed.
+ */
+ static std::unique_ptr<ValueProfData>
+ serializeFrom(const InstrProfRecord &Record);
+ /*!
+ * Check the integrity of the record.
+ */
+ Error checkIntegrity();
+ /*!
+ * Return a pointer to \c ValueProfileData instance ready to be read.
+ * All data in the instance are properly byte swapped. The input
+ * data is assumed to be in little endian order.
+ */
+ static Expected<std::unique_ptr<ValueProfData>>
+ getValueProfData(const unsigned char *SrcBuffer,
+ const unsigned char *const SrcBufferEnd,
+ support::endianness SrcDataEndianness);
+ /*!
+ * Swap byte order from \c Endianness order to host byte order.
+ */
+ void swapBytesToHost(support::endianness Endianness);
+ /*!
+ * Swap byte order from host byte order to \c Endianness order.
+ */
+ void swapBytesFromHost(support::endianness Endianness);
+ /*!
+ * Return the total size of \c ValueProfileData.
+ */
+ uint32_t getSize() const { return TotalSize; }
+ /*!
+ * Read data from this data and save it to \c Record.
+ */
+ void deserializeTo(InstrProfRecord &Record,
+ InstrProfSymtab *SymTab);
+ void operator delete(void *ptr) { ::operator delete(ptr); }
+#endif
+} ValueProfData;
+
+/*
+ * The closure is designed to abstact away two types of value profile data:
+ * - InstrProfRecord which is the primary data structure used to
+ * represent profile data in host tools (reader, writer, and profile-use)
+ * - value profile runtime data structure suitable to be used by C
+ * runtime library.
+ *
+ * Both sources of data need to serialize to disk/memory-buffer in common
+ * format: ValueProfData. The abstraction allows compiler-rt's raw profiler
+ * writer to share the same format and code with indexed profile writer.
+ *
+ * For documentation of the member methods below, refer to corresponding methods
+ * in class InstrProfRecord.
+ */
+typedef struct ValueProfRecordClosure {
+ const void *Record;
+ uint32_t (*GetNumValueKinds)(const void *Record);
+ uint32_t (*GetNumValueSites)(const void *Record, uint32_t VKind);
+ uint32_t (*GetNumValueData)(const void *Record, uint32_t VKind);
+ uint32_t (*GetNumValueDataForSite)(const void *R, uint32_t VK, uint32_t S);
+
+ /*
+ * After extracting the value profile data from the value profile record,
+ * this method is used to map the in-memory value to on-disk value. If
+ * the method is null, value will be written out untranslated.
+ */
+ uint64_t (*RemapValueData)(uint32_t, uint64_t Value);
+ void (*GetValueForSite)(const void *R, InstrProfValueData *Dst, uint32_t K,
+ uint32_t S);
+ ValueProfData *(*AllocValueProfData)(size_t TotalSizeInBytes);
+} ValueProfRecordClosure;
+
+INSTR_PROF_VISIBILITY ValueProfRecord *
+getFirstValueProfRecord(ValueProfData *VPD);
+INSTR_PROF_VISIBILITY ValueProfRecord *
+getValueProfRecordNext(ValueProfRecord *VPR);
+INSTR_PROF_VISIBILITY InstrProfValueData *
+getValueProfRecordValueData(ValueProfRecord *VPR);
+INSTR_PROF_VISIBILITY uint32_t
+getValueProfRecordHeaderSize(uint32_t NumValueSites);
+
+#undef INSTR_PROF_VALUE_PROF_DATA
+#endif /* INSTR_PROF_VALUE_PROF_DATA */
+
+
+#ifdef INSTR_PROF_COMMON_API_IMPL
+#define INSTR_PROF_DATA_DEFINED
+#ifdef __cplusplus
+#define INSTR_PROF_INLINE inline
+#define INSTR_PROF_NULLPTR nullptr
+#else
+#define INSTR_PROF_INLINE
+#define INSTR_PROF_NULLPTR NULL
+#endif
+
+#ifndef offsetof
+#define offsetof(TYPE, MEMBER) ((size_t) &((TYPE *)0)->MEMBER)
+#endif
+
+/*!
+ * Return the \c ValueProfRecord header size including the
+ * padding bytes.
+ */
+INSTR_PROF_VISIBILITY INSTR_PROF_INLINE
+uint32_t getValueProfRecordHeaderSize(uint32_t NumValueSites) {
+ uint32_t Size = offsetof(ValueProfRecord, SiteCountArray) +
+ sizeof(uint8_t) * NumValueSites;
+ /* Round the size to multiple of 8 bytes. */
+ Size = (Size + 7) & ~7;
+ return Size;
+}
+
+/*!
+ * Return the total size of the value profile record including the
+ * header and the value data.
+ */
+INSTR_PROF_VISIBILITY INSTR_PROF_INLINE
+uint32_t getValueProfRecordSize(uint32_t NumValueSites,
+ uint32_t NumValueData) {
+ return getValueProfRecordHeaderSize(NumValueSites) +
+ sizeof(InstrProfValueData) * NumValueData;
+}
+
+/*!
+ * Return the pointer to the start of value data array.
+ */
+INSTR_PROF_VISIBILITY INSTR_PROF_INLINE
+InstrProfValueData *getValueProfRecordValueData(ValueProfRecord *This) {
+ return (InstrProfValueData *)((char *)This + getValueProfRecordHeaderSize(
+ This->NumValueSites));
+}
+
+/*!
+ * Return the total number of value data for \c This record.
+ */
+INSTR_PROF_VISIBILITY INSTR_PROF_INLINE
+uint32_t getValueProfRecordNumValueData(ValueProfRecord *This) {
+ uint32_t NumValueData = 0;
+ uint32_t I;
+ for (I = 0; I < This->NumValueSites; I++)
+ NumValueData += This->SiteCountArray[I];
+ return NumValueData;
+}
+
+/*!
+ * Use this method to advance to the next \c This \c ValueProfRecord.
+ */
+INSTR_PROF_VISIBILITY INSTR_PROF_INLINE
+ValueProfRecord *getValueProfRecordNext(ValueProfRecord *This) {
+ uint32_t NumValueData = getValueProfRecordNumValueData(This);
+ return (ValueProfRecord *)((char *)This +
+ getValueProfRecordSize(This->NumValueSites,
+ NumValueData));
+}
+
+/*!
+ * Return the first \c ValueProfRecord instance.
+ */
+INSTR_PROF_VISIBILITY INSTR_PROF_INLINE
+ValueProfRecord *getFirstValueProfRecord(ValueProfData *This) {
+ return (ValueProfRecord *)((char *)This + sizeof(ValueProfData));
+}
+
+/* Closure based interfaces. */
+
+/*!
+ * Return the total size in bytes of the on-disk value profile data
+ * given the data stored in Record.
+ */
+INSTR_PROF_VISIBILITY uint32_t
+getValueProfDataSize(ValueProfRecordClosure *Closure) {
+ uint32_t Kind;
+ uint32_t TotalSize = sizeof(ValueProfData);
+ const void *Record = Closure->Record;
+
+ for (Kind = IPVK_First; Kind <= IPVK_Last; Kind++) {
+ uint32_t NumValueSites = Closure->GetNumValueSites(Record, Kind);
+ if (!NumValueSites)
+ continue;
+ TotalSize += getValueProfRecordSize(NumValueSites,
+ Closure->GetNumValueData(Record, Kind));
+ }
+ return TotalSize;
+}
+
+/*!
+ * Extract value profile data of a function for the profile kind \c ValueKind
+ * from the \c Closure and serialize the data into \c This record instance.
+ */
+INSTR_PROF_VISIBILITY void
+serializeValueProfRecordFrom(ValueProfRecord *This,
+ ValueProfRecordClosure *Closure,
+ uint32_t ValueKind, uint32_t NumValueSites) {
+ uint32_t S;
+ const void *Record = Closure->Record;
+ This->Kind = ValueKind;
+ This->NumValueSites = NumValueSites;
+ InstrProfValueData *DstVD = getValueProfRecordValueData(This);
+
+ for (S = 0; S < NumValueSites; S++) {
+ uint32_t ND = Closure->GetNumValueDataForSite(Record, ValueKind, S);
+ This->SiteCountArray[S] = ND;
+ Closure->GetValueForSite(Record, DstVD, ValueKind, S);
+ DstVD += ND;
+ }
+}
+
+/*!
+ * Extract value profile data of a function from the \c Closure
+ * and serialize the data into \c DstData if it is not NULL or heap
+ * memory allocated by the \c Closure's allocator method. If \c
+ * DstData is not null, the caller is expected to set the TotalSize
+ * in DstData.
+ */
+INSTR_PROF_VISIBILITY ValueProfData *
+serializeValueProfDataFrom(ValueProfRecordClosure *Closure,
+ ValueProfData *DstData) {
+ uint32_t Kind;
+ uint32_t TotalSize =
+ DstData ? DstData->TotalSize : getValueProfDataSize(Closure);
+
+ ValueProfData *VPD =
+ DstData ? DstData : Closure->AllocValueProfData(TotalSize);
+
+ VPD->TotalSize = TotalSize;
+ VPD->NumValueKinds = Closure->GetNumValueKinds(Closure->Record);
+ ValueProfRecord *VR = getFirstValueProfRecord(VPD);
+ for (Kind = IPVK_First; Kind <= IPVK_Last; Kind++) {
+ uint32_t NumValueSites = Closure->GetNumValueSites(Closure->Record, Kind);
+ if (!NumValueSites)
+ continue;
+ serializeValueProfRecordFrom(VR, Closure, Kind, NumValueSites);
+ VR = getValueProfRecordNext(VR);
+ }
+ return VPD;
+}
+
+#undef INSTR_PROF_COMMON_API_IMPL
+#endif /* INSTR_PROF_COMMON_API_IMPL */
+
+/*============================================================================*/
+
+#ifndef INSTR_PROF_DATA_DEFINED
+
+#ifndef INSTR_PROF_DATA_INC
+#define INSTR_PROF_DATA_INC
+
+/* Helper macros. */
+#define INSTR_PROF_SIMPLE_QUOTE(x) #x
+#define INSTR_PROF_QUOTE(x) INSTR_PROF_SIMPLE_QUOTE(x)
+#define INSTR_PROF_SIMPLE_CONCAT(x,y) x ## y
+#define INSTR_PROF_CONCAT(x,y) INSTR_PROF_SIMPLE_CONCAT(x,y)
+
+/* Magic number to detect file format and endianness.
+ * Use 255 at one end, since no UTF-8 file can use that character. Avoid 0,
+ * so that utilities, like strings, don't grab it as a string. 129 is also
+ * invalid UTF-8, and high enough to be interesting.
+ * Use "lprofr" in the centre to stand for "LLVM Profile Raw", or "lprofR"
+ * for 32-bit platforms.
+ */
+#define INSTR_PROF_RAW_MAGIC_64 (uint64_t)255 << 56 | (uint64_t)'l' << 48 | \
+ (uint64_t)'p' << 40 | (uint64_t)'r' << 32 | (uint64_t)'o' << 24 | \
+ (uint64_t)'f' << 16 | (uint64_t)'r' << 8 | (uint64_t)129
+#define INSTR_PROF_RAW_MAGIC_32 (uint64_t)255 << 56 | (uint64_t)'l' << 48 | \
+ (uint64_t)'p' << 40 | (uint64_t)'r' << 32 | (uint64_t)'o' << 24 | \
+ (uint64_t)'f' << 16 | (uint64_t)'R' << 8 | (uint64_t)129
+
+/* Raw profile format version (start from 1). */
+#define INSTR_PROF_RAW_VERSION 5
+/* Indexed profile format version (start from 1). */
+#define INSTR_PROF_INDEX_VERSION 6
+/* Coverage mapping format version (start from 0). */
+#define INSTR_PROF_COVMAP_VERSION 3
+
+/* Profile version is always of type uint64_t. Reserve the upper 8 bits in the
+ * version for other variants of profile. We set the lowest bit of the upper 8
+ * bits (i.e. bit 56) to 1 to indicate if this is an IR-level instrumentaiton
+ * generated profile, and 0 if this is a Clang FE generated profile.
+ * 1 in bit 57 indicates there are context-sensitive records in the profile.
+ */
+#define VARIANT_MASKS_ALL 0xff00000000000000ULL
+#define GET_VERSION(V) ((V) & ~VARIANT_MASKS_ALL)
+#define VARIANT_MASK_IR_PROF (0x1ULL << 56)
+#define VARIANT_MASK_CSIR_PROF (0x1ULL << 57)
+#define INSTR_PROF_RAW_VERSION_VAR __llvm_profile_raw_version
+#define INSTR_PROF_PROFILE_RUNTIME_VAR __llvm_profile_runtime
+
+/* The variable that holds the name of the profile data
+ * specified via command line. */
+#define INSTR_PROF_PROFILE_NAME_VAR __llvm_profile_filename
+
+/* section name strings common to all targets other
+ than WIN32 */
+#define INSTR_PROF_DATA_COMMON __llvm_prf_data
+#define INSTR_PROF_NAME_COMMON __llvm_prf_names
+#define INSTR_PROF_CNTS_COMMON __llvm_prf_cnts
+#define INSTR_PROF_VALS_COMMON __llvm_prf_vals
+#define INSTR_PROF_VNODES_COMMON __llvm_prf_vnds
+#define INSTR_PROF_COVMAP_COMMON __llvm_covmap
+#define INSTR_PROF_COVFUN_COMMON __llvm_covfun
+#define INSTR_PROF_ORDERFILE_COMMON __llvm_orderfile
+/* Windows section names. Because these section names contain dollar characters,
+ * they must be quoted.
+ */
+#define INSTR_PROF_DATA_COFF ".lprfd$M"
+#define INSTR_PROF_NAME_COFF ".lprfn$M"
+#define INSTR_PROF_CNTS_COFF ".lprfc$M"
+#define INSTR_PROF_VALS_COFF ".lprfv$M"
+#define INSTR_PROF_VNODES_COFF ".lprfnd$M"
+#define INSTR_PROF_COVMAP_COFF ".lcovmap$M"
+#define INSTR_PROF_COVFUN_COFF ".lcovfun$M"
+#define INSTR_PROF_ORDERFILE_COFF ".lorderfile$M"
+
+#ifdef _WIN32
+/* Runtime section names and name strings. */
+#define INSTR_PROF_DATA_SECT_NAME INSTR_PROF_DATA_COFF
+#define INSTR_PROF_NAME_SECT_NAME INSTR_PROF_NAME_COFF
+#define INSTR_PROF_CNTS_SECT_NAME INSTR_PROF_CNTS_COFF
+/* Array of pointers. Each pointer points to a list
+ * of value nodes associated with one value site.
+ */
+#define INSTR_PROF_VALS_SECT_NAME INSTR_PROF_VALS_COFF
+/* Value profile nodes section. */
+#define INSTR_PROF_VNODES_SECT_NAME INSTR_PROF_VNODES_COFF
+#define INSTR_PROF_COVMAP_SECT_NAME INSTR_PROF_COVMAP_COFF
+#define INSTR_PROF_COVFUN_SECT_NAME INSTR_PROF_COVFUN_COFF
+#define INSTR_PROF_ORDERFILE_SECT_NAME INSTR_PROF_ORDERFILE_COFF
+#else
+/* Runtime section names and name strings. */
+#define INSTR_PROF_DATA_SECT_NAME INSTR_PROF_QUOTE(INSTR_PROF_DATA_COMMON)
+#define INSTR_PROF_NAME_SECT_NAME INSTR_PROF_QUOTE(INSTR_PROF_NAME_COMMON)
+#define INSTR_PROF_CNTS_SECT_NAME INSTR_PROF_QUOTE(INSTR_PROF_CNTS_COMMON)
+/* Array of pointers. Each pointer points to a list
+ * of value nodes associated with one value site.
+ */
+#define INSTR_PROF_VALS_SECT_NAME INSTR_PROF_QUOTE(INSTR_PROF_VALS_COMMON)
+/* Value profile nodes section. */
+#define INSTR_PROF_VNODES_SECT_NAME INSTR_PROF_QUOTE(INSTR_PROF_VNODES_COMMON)
+#define INSTR_PROF_COVMAP_SECT_NAME INSTR_PROF_QUOTE(INSTR_PROF_COVMAP_COMMON)
+#define INSTR_PROF_COVFUN_SECT_NAME INSTR_PROF_QUOTE(INSTR_PROF_COVFUN_COMMON)
+/* Order file instrumentation. */
+#define INSTR_PROF_ORDERFILE_SECT_NAME \
+ INSTR_PROF_QUOTE(INSTR_PROF_ORDERFILE_COMMON)
+#endif
+
+#define INSTR_PROF_ORDERFILE_BUFFER_NAME _llvm_order_file_buffer
+#define INSTR_PROF_ORDERFILE_BUFFER_NAME_STR \
+ INSTR_PROF_QUOTE(INSTR_PROF_ORDERFILE_BUFFER_NAME)
+#define INSTR_PROF_ORDERFILE_BUFFER_IDX_NAME _llvm_order_file_buffer_idx
+#define INSTR_PROF_ORDERFILE_BUFFER_IDX_NAME_STR \
+ INSTR_PROF_QUOTE(INSTR_PROF_ORDERFILE_BUFFER_IDX_NAME)
+
+/* Macros to define start/stop section symbol for a given
+ * section on Linux. For instance
+ * INSTR_PROF_SECT_START(INSTR_PROF_DATA_SECT_NAME) will
+ * expand to __start___llvm_prof_data
+ */
+#define INSTR_PROF_SECT_START(Sect) \
+ INSTR_PROF_CONCAT(__start_,Sect)
+#define INSTR_PROF_SECT_STOP(Sect) \
+ INSTR_PROF_CONCAT(__stop_,Sect)
+
+/* Value Profiling API linkage name. */
+#define INSTR_PROF_VALUE_PROF_FUNC __llvm_profile_instrument_target
+#define INSTR_PROF_VALUE_PROF_FUNC_STR \
+ INSTR_PROF_QUOTE(INSTR_PROF_VALUE_PROF_FUNC)
+#define INSTR_PROF_VALUE_RANGE_PROF_FUNC __llvm_profile_instrument_range
+#define INSTR_PROF_VALUE_RANGE_PROF_FUNC_STR \
+ INSTR_PROF_QUOTE(INSTR_PROF_VALUE_RANGE_PROF_FUNC)
+
+/* InstrProfile per-function control data alignment. */
+#define INSTR_PROF_DATA_ALIGNMENT 8
+
+/* The data structure that represents a tracked value by the
+ * value profiler.
+ */
+typedef struct InstrProfValueData {
+ /* Profiled value. */
+ uint64_t Value;
+ /* Number of times the value appears in the training run. */
+ uint64_t Count;
+} InstrProfValueData;
+
+#endif /* INSTR_PROF_DATA_INC */
+
+#ifndef INSTR_ORDER_FILE_INC
+/* The maximal # of functions: 128*1024 (the buffer size will be 128*4 KB). */
+#define INSTR_ORDER_FILE_BUFFER_SIZE 131072
+#define INSTR_ORDER_FILE_BUFFER_BITS 17
+#define INSTR_ORDER_FILE_BUFFER_MASK 0x1ffff
+#endif /* INSTR_ORDER_FILE_INC */
+#else
+#undef INSTR_PROF_DATA_DEFINED
+#endif
+
+#undef COVMAP_V2_OR_V3
--- /dev/null
+/*===------------ ptwriteintrin.h - PTWRITE intrinsic --------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#if !defined __X86INTRIN_H && !defined __IMMINTRIN_H
+#error "Never use <ptwriteintrin.h> directly; include <x86intrin.h> instead."
+#endif
+
+#ifndef __PTWRITEINTRIN_H
+#define __PTWRITEINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("ptwrite")))
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_ptwrite32(unsigned int __value) {
+ __builtin_ia32_ptwrite32(__value);
+}
+
+#ifdef __x86_64__
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_ptwrite64(unsigned long long __value) {
+ __builtin_ia32_ptwrite64(__value);
+}
+
+#endif /* __x86_64__ */
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif /* __PTWRITEINTRIN_H */
--- /dev/null
+/*===---- rdseedintrin.h - RDSEED intrinsics -------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#if !defined __X86INTRIN_H && !defined __IMMINTRIN_H
+#error "Never use <rdseedintrin.h> directly; include <x86intrin.h> instead."
+#endif
+
+#ifndef __RDSEEDINTRIN_H
+#define __RDSEEDINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("rdseed")))
+
+static __inline__ int __DEFAULT_FN_ATTRS
+_rdseed16_step(unsigned short *__p)
+{
+ return __builtin_ia32_rdseed16_step(__p);
+}
+
+static __inline__ int __DEFAULT_FN_ATTRS
+_rdseed32_step(unsigned int *__p)
+{
+ return __builtin_ia32_rdseed32_step(__p);
+}
+
+#ifdef __x86_64__
+static __inline__ int __DEFAULT_FN_ATTRS
+_rdseed64_step(unsigned long long *__p)
+{
+ return __builtin_ia32_rdseed64_step(__p);
+}
+#endif
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif /* __RDSEEDINTRIN_H */
--- /dev/null
+/*===---- rtmintrin.h - RTM intrinsics -------------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <rtmintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __RTMINTRIN_H
+#define __RTMINTRIN_H
+
+#define _XBEGIN_STARTED (~0u)
+#define _XABORT_EXPLICIT (1 << 0)
+#define _XABORT_RETRY (1 << 1)
+#define _XABORT_CONFLICT (1 << 2)
+#define _XABORT_CAPACITY (1 << 3)
+#define _XABORT_DEBUG (1 << 4)
+#define _XABORT_NESTED (1 << 5)
+#define _XABORT_CODE(x) (((x) >> 24) & 0xFF)
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("rtm")))
+
+static __inline__ unsigned int __DEFAULT_FN_ATTRS
+_xbegin(void)
+{
+ return __builtin_ia32_xbegin();
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_xend(void)
+{
+ __builtin_ia32_xend();
+}
+
+#define _xabort(imm) __builtin_ia32_xabort((imm))
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif /* __RTMINTRIN_H */
--- /dev/null
+/*===---- s390intrin.h - SystemZ intrinsics --------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __S390INTRIN_H
+#define __S390INTRIN_H
+
+#ifndef __s390__
+#error "<s390intrin.h> is for s390 only"
+#endif
+
+#ifdef __HTM__
+#include <htmintrin.h>
+#endif
+
+#ifdef __VEC__
+#include <vecintrin.h>
+#endif
+
+#endif /* __S390INTRIN_H*/
--- /dev/null
+//===-- allocator_interface.h ---------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// Public interface header for allocator used in sanitizers (ASan/TSan/MSan).
+//===----------------------------------------------------------------------===//
+#ifndef SANITIZER_ALLOCATOR_INTERFACE_H
+#define SANITIZER_ALLOCATOR_INTERFACE_H
+
+#include <stddef.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+ /* Returns the estimated number of bytes that will be reserved by allocator
+ for request of "size" bytes. If allocator can't allocate that much
+ memory, returns the maximal possible allocation size, otherwise returns
+ "size". */
+ size_t __sanitizer_get_estimated_allocated_size(size_t size);
+
+ /* Returns true if p was returned by the allocator and
+ is not yet freed. */
+ int __sanitizer_get_ownership(const volatile void *p);
+
+ /* Returns the number of bytes reserved for the pointer p.
+ Requires (get_ownership(p) == true) or (p == 0). */
+ size_t __sanitizer_get_allocated_size(const volatile void *p);
+
+ /* Number of bytes, allocated and not yet freed by the application. */
+ size_t __sanitizer_get_current_allocated_bytes(void);
+
+ /* Number of bytes, mmaped by the allocator to fulfill allocation requests.
+ Generally, for request of X bytes, allocator can reserve and add to free
+ lists a large number of chunks of size X to use them for future requests.
+ All these chunks count toward the heap size. Currently, allocator never
+ releases memory to OS (instead, it just puts freed chunks to free
+ lists). */
+ size_t __sanitizer_get_heap_size(void);
+
+ /* Number of bytes, mmaped by the allocator, which can be used to fulfill
+ allocation requests. When a user program frees memory chunk, it can first
+ fall into quarantine and will count toward __sanitizer_get_free_bytes()
+ later. */
+ size_t __sanitizer_get_free_bytes(void);
+
+ /* Number of bytes in unmapped pages, that are released to OS. Currently,
+ always returns 0. */
+ size_t __sanitizer_get_unmapped_bytes(void);
+
+ /* Malloc hooks that may be optionally provided by user.
+ __sanitizer_malloc_hook(ptr, size) is called immediately after
+ allocation of "size" bytes, which returned "ptr".
+ __sanitizer_free_hook(ptr) is called immediately before
+ deallocation of "ptr". */
+ void __sanitizer_malloc_hook(const volatile void *ptr, size_t size);
+ void __sanitizer_free_hook(const volatile void *ptr);
+
+ /* Installs a pair of hooks for malloc/free.
+ Several (currently, 5) hook pairs may be installed, they are executed
+ in the order they were installed and after calling
+ __sanitizer_malloc_hook/__sanitizer_free_hook.
+ Unlike __sanitizer_malloc_hook/__sanitizer_free_hook these hooks can be
+ chained and do not rely on weak symbols working on the platform, but
+ require __sanitizer_install_malloc_and_free_hooks to be called at startup
+ and thus will not be called on malloc/free very early in the process.
+ Returns the number of hooks currently installed or 0 on failure.
+ Not thread-safe, should be called in the main thread before starting
+ other threads.
+ */
+ int __sanitizer_install_malloc_and_free_hooks(
+ void (*malloc_hook)(const volatile void *, size_t),
+ void (*free_hook)(const volatile void *));
+
+ /* Drains allocator quarantines (calling thread's and global ones), returns
+ freed memory back to OS and releases other non-essential internal allocator
+ resources in attempt to reduce process RSS.
+ Currently available with ASan only.
+ */
+ void __sanitizer_purge_allocator(void);
+#ifdef __cplusplus
+} // extern "C"
+#endif
+
+#endif
--- /dev/null
+//===-- sanitizer/asan_interface.h ------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file is a part of AddressSanitizer (ASan).
+//
+// Public interface header.
+//===----------------------------------------------------------------------===//
+#ifndef SANITIZER_ASAN_INTERFACE_H
+#define SANITIZER_ASAN_INTERFACE_H
+
+#include <sanitizer/common_interface_defs.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+/// Marks a memory region (<c>[addr, addr+size)</c>) as unaddressable.
+///
+/// This memory must be previously allocated by your program. Instrumented
+/// code is forbidden from accessing addresses in this region until it is
+/// unpoisoned. This function is not guaranteed to poison the entire region -
+/// it could poison only a subregion of <c>[addr, addr+size)</c> due to ASan
+/// alignment restrictions.
+///
+/// \note This function is not thread-safe because no two threads can poison or
+/// unpoison memory in the same memory region simultaneously.
+///
+/// \param addr Start of memory region.
+/// \param size Size of memory region.
+void __asan_poison_memory_region(void const volatile *addr, size_t size);
+
+/// Marks a memory region (<c>[addr, addr+size)</c>) as addressable.
+///
+/// This memory must be previously allocated by your program. Accessing
+/// addresses in this region is allowed until this region is poisoned again.
+/// This function could unpoison a super-region of <c>[addr, addr+size)</c> due
+/// to ASan alignment restrictions.
+///
+/// \note This function is not thread-safe because no two threads can
+/// poison or unpoison memory in the same memory region simultaneously.
+///
+/// \param addr Start of memory region.
+/// \param size Size of memory region.
+void __asan_unpoison_memory_region(void const volatile *addr, size_t size);
+
+// Macros provided for convenience.
+#if __has_feature(address_sanitizer) || defined(__SANITIZE_ADDRESS__)
+/// Marks a memory region as unaddressable.
+///
+/// \note Macro provided for convenience; defined as a no-op if ASan is not
+/// enabled.
+///
+/// \param addr Start of memory region.
+/// \param size Size of memory region.
+#define ASAN_POISON_MEMORY_REGION(addr, size) \
+ __asan_poison_memory_region((addr), (size))
+
+/// Marks a memory region as addressable.
+///
+/// \note Macro provided for convenience; defined as a no-op if ASan is not
+/// enabled.
+///
+/// \param addr Start of memory region.
+/// \param size Size of memory region.
+#define ASAN_UNPOISON_MEMORY_REGION(addr, size) \
+ __asan_unpoison_memory_region((addr), (size))
+#else
+#define ASAN_POISON_MEMORY_REGION(addr, size) \
+ ((void)(addr), (void)(size))
+#define ASAN_UNPOISON_MEMORY_REGION(addr, size) \
+ ((void)(addr), (void)(size))
+#endif
+
+/// Checks if an address is poisoned.
+///
+/// Returns 1 if <c><i>addr</i></c> is poisoned (that is, 1-byte read/write
+/// access to this address would result in an error report from ASan).
+/// Otherwise returns 0.
+///
+/// \param addr Address to check.
+///
+/// \retval 1 Address is poisoned.
+/// \retval 0 Address is not poisoned.
+int __asan_address_is_poisoned(void const volatile *addr);
+
+/// Checks if a region is poisoned.
+///
+/// If at least one byte in <c>[beg, beg+size)</c> is poisoned, returns the
+/// address of the first such byte. Otherwise returns 0.
+///
+/// \param beg Start of memory region.
+/// \param size Start of memory region.
+/// \returns Address of first poisoned byte.
+void *__asan_region_is_poisoned(void *beg, size_t size);
+
+/// Describes an address (useful for calling from the debugger).
+///
+/// Prints the description of <c><i>addr</i></c>.
+///
+/// \param addr Address to describe.
+void __asan_describe_address(void *addr);
+
+/// Checks if an error has been or is being reported (useful for calling from
+/// the debugger to get information about an ASan error).
+///
+/// Returns 1 if an error has been (or is being) reported. Otherwise returns 0.
+///
+/// \returns 1 if an error has been (or is being) reported. Otherwise returns
+/// 0.
+int __asan_report_present(void);
+
+/// Gets the PC (program counter) register value of an ASan error (useful for
+/// calling from the debugger).
+///
+/// Returns PC if an error has been (or is being) reported.
+/// Otherwise returns 0.
+///
+/// \returns PC value.
+void *__asan_get_report_pc(void);
+
+/// Gets the BP (base pointer) register value of an ASan error (useful for
+/// calling from the debugger).
+///
+/// Returns BP if an error has been (or is being) reported.
+/// Otherwise returns 0.
+///
+/// \returns BP value.
+void *__asan_get_report_bp(void);
+
+/// Gets the SP (stack pointer) register value of an ASan error (useful for
+/// calling from the debugger).
+///
+/// If an error has been (or is being) reported, returns SP.
+/// Otherwise returns 0.
+///
+/// \returns SP value.
+void *__asan_get_report_sp(void);
+
+/// Gets the address of the report buffer of an ASan error (useful for calling
+/// from the debugger).
+///
+/// Returns the address of the report buffer if an error has been (or is being)
+/// reported. Otherwise returns 0.
+///
+/// \returns Address of report buffer.
+void *__asan_get_report_address(void);
+
+/// Gets access type of an ASan error (useful for calling from the debugger).
+///
+/// Returns access type (read or write) if an error has been (or is being)
+/// reported. Otherwise returns 0.
+///
+/// \returns Access type (0 = read, 1 = write).
+int __asan_get_report_access_type(void);
+
+/// Gets access size of an ASan error (useful for calling from the debugger).
+///
+/// Returns access size if an error has been (or is being) reported. Otherwise
+/// returns 0.
+///
+/// \returns Access size in bytes.
+size_t __asan_get_report_access_size(void);
+
+/// Gets the bug description of an ASan error (useful for calling from a
+/// debugger).
+///
+/// \returns Returns a bug description if an error has been (or is being)
+/// reported - for example, "heap-use-after-free". Otherwise returns an empty
+/// string.
+const char *__asan_get_report_description(void);
+
+/// Gets information about a pointer (useful for calling from the debugger).
+///
+/// Returns the category of the given pointer as a constant string.
+/// Possible return values are <c>global</c>, <c>stack</c>, <c>stack-fake</c>,
+/// <c>heap</c>, <c>heap-invalid</c>, <c>shadow-low</c>, <c>shadow-gap</c>,
+/// <c>shadow-high</c>, and <c>unknown</c>.
+///
+/// If the return value is <c>global</c> or <c>stack</c>, tries to also return
+/// the variable name, address, and size. If the return value is <c>heap</c>,
+/// tries to return the chunk address and size. <c><i>name</i></c> should point
+/// to an allocated buffer of size <c><i>name_size</i></c>.
+///
+/// \param addr Address to locate.
+/// \param name Buffer to store the variable's name.
+/// \param name_size Size in bytes of the variable's name buffer.
+/// \param region_address [out] Address of the region.
+/// \param region_size [out] Size of the region in bytes.
+///
+/// \returns Returns the category of the given pointer as a constant string.
+const char *__asan_locate_address(void *addr, char *name, size_t name_size,
+ void **region_address, size_t *region_size);
+
+/// Gets the allocation stack trace and thread ID for a heap address (useful
+/// for calling from the debugger).
+///
+/// Stores up to <c><i>size</i></c> frames in <c><i>trace</i></c>. Returns
+/// the number of stored frames or 0 on error.
+///
+/// \param addr A heap address.
+/// \param trace A buffer to store the stack trace.
+/// \param size Size in bytes of the trace buffer.
+/// \param thread_id [out] The thread ID of the address.
+///
+/// \returns Returns the number of stored frames or 0 on error.
+size_t __asan_get_alloc_stack(void *addr, void **trace, size_t size,
+ int *thread_id);
+
+/// Gets the free stack trace and thread ID for a heap address (useful for
+/// calling from the debugger).
+///
+/// Stores up to <c><i>size</i></c> frames in <c><i>trace</i></c>. Returns
+/// the number of stored frames or 0 on error.
+///
+/// \param addr A heap address.
+/// \param trace A buffer to store the stack trace.
+/// \param size Size in bytes of the trace buffer.
+/// \param thread_id [out] The thread ID of the address.
+///
+/// \returns Returns the number of stored frames or 0 on error.
+size_t __asan_get_free_stack(void *addr, void **trace, size_t size,
+ int *thread_id);
+
+/// Gets the current shadow memory mapping (useful for calling from the
+/// debugger).
+///
+/// \param shadow_scale [out] Shadow scale value.
+/// \param shadow_offset [out] Offset value.
+void __asan_get_shadow_mapping(size_t *shadow_scale, size_t *shadow_offset);
+
+/// This is an internal function that is called to report an error. However,
+/// it is still a part of the interface because you might want to set a
+/// breakpoint on this function in the debugger.
+///
+/// \param pc <c><i>pc</i></c> value of the ASan error.
+/// \param bp <c><i>bp</i></c> value of the ASan error.
+/// \param sp <c><i>sp</i></c> value of the ASan error.
+/// \param addr Address of the ASan error.
+/// \param is_write True if the error is a write error; false otherwise.
+/// \param access_size Size of the memory access of the ASan error.
+void __asan_report_error(void *pc, void *bp, void *sp,
+ void *addr, int is_write, size_t access_size);
+
+// Deprecated. Call __sanitizer_set_death_callback instead.
+void __asan_set_death_callback(void (*callback)(void));
+
+/// Sets the callback function to be called during ASan error reporting.
+///
+/// The callback provides a string pointer to the report.
+///
+/// \param callback User-provided function.
+void __asan_set_error_report_callback(void (*callback)(const char *));
+
+/// User-provided callback on ASan errors.
+///
+/// You can provide a function that would be called immediately when ASan
+/// detects an error. This is useful in cases when ASan detects an error but
+/// your program crashes before the ASan report is printed.
+void __asan_on_error(void);
+
+/// Prints accumulated statistics to <c>stderr</c> (useful for calling from the
+/// debugger).
+void __asan_print_accumulated_stats(void);
+
+/// User-provided default option settings.
+///
+/// You can provide your own implementation of this function to return a string
+/// containing ASan runtime options (for example,
+/// <c>verbosity=1:halt_on_error=0</c>).
+///
+/// \returns Default options string.
+const char* __asan_default_options(void);
+
+// The following two functions facilitate garbage collection in presence of
+// ASan's fake stack.
+
+/// Gets an opaque handler to the current thread's fake stack.
+///
+/// Returns an opaque handler to be used by
+/// <c>__asan_addr_is_in_fake_stack()</c>. Returns NULL if the current thread
+/// does not have a fake stack.
+///
+/// \returns An opaque handler to the fake stack or NULL.
+void *__asan_get_current_fake_stack(void);
+
+/// Checks if an address belongs to a given fake stack.
+///
+/// If <c><i>fake_stack</i></c> is non-NULL and <c><i>addr</i></c> belongs to a
+/// fake frame in <c><i>fake_stack</i></c>, returns the address of the real
+/// stack that corresponds to the fake frame and sets <c><i>beg</i></c> and
+/// <c><i>end</i></c> to the boundaries of this fake frame. Otherwise returns
+/// NULL and does not touch <c><i>beg</i></c> and <c><i>end</i></c>.
+///
+/// If <c><i>beg</i></c> or <c><i>end</i></c> are NULL, they are not touched.
+///
+/// \note This function can be called from a thread other than the owner of
+/// <c><i>fake_stack</i></c>, but the owner thread needs to be alive.
+///
+/// \param fake_stack An opaque handler to a fake stack.
+/// \param addr Address to test.
+/// \param beg [out] Beginning of fake frame.
+/// \param end [out] End of fake frame.
+/// \returns Stack address or NULL.
+void *__asan_addr_is_in_fake_stack(void *fake_stack, void *addr, void **beg,
+ void **end);
+
+/// Performs shadow memory cleanup of the current thread's stack before a
+/// function marked with the <c>[[noreturn]]</c> attribute is called.
+///
+/// To avoid false positives on the stack, must be called before no-return
+/// functions like <c>_exit()</c> and <c>execl()</c>.
+void __asan_handle_no_return(void);
+
+/// Update allocation stack trace for the given allocation to the current stack
+/// trace. Returns 1 if successfull, 0 if not.
+int __asan_update_allocation_context(void* addr);
+
+#ifdef __cplusplus
+} // extern "C"
+#endif
+
+#endif // SANITIZER_ASAN_INTERFACE_H
--- /dev/null
+//===-- sanitizer/common_interface_defs.h -----------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// Common part of the public sanitizer interface.
+//===----------------------------------------------------------------------===//
+
+#ifndef SANITIZER_COMMON_INTERFACE_DEFS_H
+#define SANITIZER_COMMON_INTERFACE_DEFS_H
+
+#include <stddef.h>
+#include <stdint.h>
+
+// GCC does not understand __has_feature.
+#if !defined(__has_feature)
+#define __has_feature(x) 0
+#endif
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+// Arguments for __sanitizer_sandbox_on_notify() below.
+typedef struct {
+ // Enable sandbox support in sanitizer coverage.
+ int coverage_sandboxed;
+ // File descriptor to write coverage data to. If -1 is passed, a file will
+ // be pre-opened by __sanitizer_sandobx_on_notify(). This field has no
+ // effect if coverage_sandboxed == 0.
+ intptr_t coverage_fd;
+ // If non-zero, split the coverage data into well-formed blocks. This is
+ // useful when coverage_fd is a socket descriptor. Each block will contain
+ // a header, allowing data from multiple processes to be sent over the same
+ // socket.
+ unsigned int coverage_max_block_size;
+} __sanitizer_sandbox_arguments;
+
+// Tell the tools to write their reports to "path.<pid>" instead of stderr.
+void __sanitizer_set_report_path(const char *path);
+// Tell the tools to write their reports to the provided file descriptor
+// (casted to void *).
+void __sanitizer_set_report_fd(void *fd);
+
+// Notify the tools that the sandbox is going to be turned on. The reserved
+// parameter will be used in the future to hold a structure with functions
+// that the tools may call to bypass the sandbox.
+void __sanitizer_sandbox_on_notify(__sanitizer_sandbox_arguments *args);
+
+// This function is called by the tool when it has just finished reporting
+// an error. 'error_summary' is a one-line string that summarizes
+// the error message. This function can be overridden by the client.
+void __sanitizer_report_error_summary(const char *error_summary);
+
+// Some of the sanitizers (for example ASan/TSan) could miss bugs that happen
+// in unaligned loads/stores. To find such bugs reliably, you need to replace
+// plain unaligned loads/stores with these calls.
+
+/// Loads a 16-bit unaligned value.
+///
+/// \param p Pointer to unaligned memory.
+///
+/// \returns Loaded value.
+uint16_t __sanitizer_unaligned_load16(const void *p);
+
+/// Loads a 32-bit unaligned value.
+///
+/// \param p Pointer to unaligned memory.
+///
+/// \returns Loaded value.
+uint32_t __sanitizer_unaligned_load32(const void *p);
+
+/// Loads a 64-bit unaligned value.
+///
+/// \param p Pointer to unaligned memory.
+///
+/// \returns Loaded value.
+uint64_t __sanitizer_unaligned_load64(const void *p);
+
+/// Stores a 16-bit unaligned value.
+///
+/// \param p Pointer to unaligned memory.
+/// \param x 16-bit value to store.
+void __sanitizer_unaligned_store16(void *p, uint16_t x);
+
+/// Stores a 32-bit unaligned value.
+///
+/// \param p Pointer to unaligned memory.
+/// \param x 32-bit value to store.
+void __sanitizer_unaligned_store32(void *p, uint32_t x);
+
+/// Stores a 64-bit unaligned value.
+///
+/// \param p Pointer to unaligned memory.
+/// \param x 64-bit value to store.
+void __sanitizer_unaligned_store64(void *p, uint64_t x);
+
+// Returns 1 on the first call, then returns 0 thereafter. Called by the tool
+// to ensure only one report is printed when multiple errors occur
+// simultaneously.
+int __sanitizer_acquire_crash_state();
+
+/// Annotates the current state of a contiguous container, such as
+/// <c>std::vector</c>, <c>std::string</c>, or similar.
+///
+/// A contiguous container is a container that keeps all of its elements
+/// in a contiguous region of memory. The container owns the region of memory
+/// <c>[beg, end)</c>; the memory <c>[beg, mid)</c> is used to store the
+/// current elements, and the memory <c>[mid, end)</c> is reserved for future
+/// elements (<c>beg <= mid <= end</c>). For example, in
+/// <c>std::vector<> v</c>:
+///
+/// \code
+/// beg = &v[0];
+/// end = beg + v.capacity() * sizeof(v[0]);
+/// mid = beg + v.size() * sizeof(v[0]);
+/// \endcode
+///
+/// This annotation tells the Sanitizer tool about the current state of the
+/// container so that the tool can report errors when memory from
+/// <c>[mid, end)</c> is accessed. Insert this annotation into methods like
+/// <c>push_back()</c> or <c>pop_back()</c>. Supply the old and new values of
+/// <c>mid</c>(<c><i>old_mid</i></c> and <c><i>new_mid</i></c>). In the initial
+/// state <c>mid == end</c>, so that should be the final state when the
+/// container is destroyed or when the container reallocates the storage.
+///
+/// For ASan, <c><i>beg</i></c> should be 8-aligned and <c><i>end</i></c>
+/// should be either 8-aligned or it should point to the end of a separate
+/// heap-, stack-, or global-allocated buffer. So the following example will
+/// not work:
+///
+/// \code
+/// int64_t x[2]; // 16 bytes, 8-aligned
+/// char *beg = (char *)&x[0];
+/// char *end = beg + 12; // Not 8-aligned, not the end of the buffer
+/// \endcode
+///
+/// The following, however, will work:
+/// \code
+/// int32_t x[3]; // 12 bytes, but 8-aligned under ASan.
+/// char *beg = (char*)&x[0];
+/// char *end = beg + 12; // Not 8-aligned, but is the end of the buffer
+/// \endcode
+///
+/// \note Use this function with caution and do not use for anything other
+/// than vector-like classes.
+///
+/// \param beg Beginning of memory region.
+/// \param end End of memory region.
+/// \param old_mid Old middle of memory region.
+/// \param new_mid New middle of memory region.
+void __sanitizer_annotate_contiguous_container(const void *beg,
+ const void *end,
+ const void *old_mid,
+ const void *new_mid);
+
+/// Returns true if the contiguous container <c>[beg, end)</c> is properly
+/// poisoned.
+///
+/// Proper poisoning could occur, for example, with
+/// <c>__sanitizer_annotate_contiguous_container</c>), that is, if
+/// <c>[beg, mid)</c> is addressable and <c>[mid, end)</c> is unaddressable.
+/// Full verification requires O (<c>end - beg</c>) time; this function tries
+/// to avoid such complexity by touching only parts of the container around
+/// <c><i>beg</i></c>, <c><i>mid</i></c>, and <c><i>end</i></c>.
+///
+/// \param beg Beginning of memory region.
+/// \param mid Middle of memory region.
+/// \param end Old end of memory region.
+///
+/// \returns True if the contiguous container <c>[beg, end)</c> is properly
+/// poisoned.
+int __sanitizer_verify_contiguous_container(const void *beg, const void *mid,
+ const void *end);
+
+/// Similar to <c>__sanitizer_verify_contiguous_container()</c> but also
+/// returns the address of the first improperly poisoned byte.
+///
+/// Returns NULL if the area is poisoned properly.
+///
+/// \param beg Beginning of memory region.
+/// \param mid Middle of memory region.
+/// \param end Old end of memory region.
+///
+/// \returns The bad address or NULL.
+const void *__sanitizer_contiguous_container_find_bad_address(const void *beg,
+ const void *mid,
+ const void *end);
+
+/// Prints the stack trace leading to this call (useful for calling from the
+/// debugger).
+void __sanitizer_print_stack_trace(void);
+
+// Symbolizes the supplied 'pc' using the format string 'fmt'.
+// Outputs at most 'out_buf_size' bytes into 'out_buf'.
+// If 'out_buf' is not empty then output is zero or more non empty C strings
+// followed by single empty C string. Multiple strings can be returned if PC
+// corresponds to inlined function. Inlined frames are printed in the order
+// from "most-inlined" to the "least-inlined", so the last frame should be the
+// not inlined function.
+// Inlined frames can be removed with 'symbolize_inline_frames=0'.
+// The format syntax is described in
+// lib/sanitizer_common/sanitizer_stacktrace_printer.h.
+void __sanitizer_symbolize_pc(void *pc, const char *fmt, char *out_buf,
+ size_t out_buf_size);
+// Same as __sanitizer_symbolize_pc, but for data section (i.e. globals).
+void __sanitizer_symbolize_global(void *data_ptr, const char *fmt,
+ char *out_buf, size_t out_buf_size);
+
+/// Sets the callback to be called immediately before death on error.
+///
+/// Passing 0 will unset the callback.
+///
+/// \param callback User-provided callback.
+void __sanitizer_set_death_callback(void (*callback)(void));
+
+
+// Interceptor hooks.
+// Whenever a libc function interceptor is called, it checks if the
+// corresponding weak hook is defined, and calls it if it is indeed defined.
+// The primary use-case is data-flow-guided fuzzing, where the fuzzer needs
+// to know what is being passed to libc functions (for example memcmp).
+// FIXME: implement more hooks.
+
+/// Interceptor hook for <c>memcmp()</c>.
+///
+/// \param called_pc PC (program counter) address of the original call.
+/// \param s1 Pointer to block of memory.
+/// \param s2 Pointer to block of memory.
+/// \param n Number of bytes to compare.
+/// \param result Value returned by the intercepted function.
+void __sanitizer_weak_hook_memcmp(void *called_pc, const void *s1,
+ const void *s2, size_t n, int result);
+
+/// Interceptor hook for <c>strncmp()</c>.
+///
+/// \param called_pc PC (program counter) address of the original call.
+/// \param s1 Pointer to block of memory.
+/// \param s2 Pointer to block of memory.
+/// \param n Number of bytes to compare.
+/// \param result Value returned by the intercepted function.
+void __sanitizer_weak_hook_strncmp(void *called_pc, const char *s1,
+ const char *s2, size_t n, int result);
+
+/// Interceptor hook for <c>strncasecmp()</c>.
+///
+/// \param called_pc PC (program counter) address of the original call.
+/// \param s1 Pointer to block of memory.
+/// \param s2 Pointer to block of memory.
+/// \param n Number of bytes to compare.
+/// \param result Value returned by the intercepted function.
+void __sanitizer_weak_hook_strncasecmp(void *called_pc, const char *s1,
+ const char *s2, size_t n, int result);
+
+/// Interceptor hook for <c>strcmp()</c>.
+///
+/// \param called_pc PC (program counter) address of the original call.
+/// \param s1 Pointer to block of memory.
+/// \param s2 Pointer to block of memory.
+/// \param result Value returned by the intercepted function.
+void __sanitizer_weak_hook_strcmp(void *called_pc, const char *s1,
+ const char *s2, int result);
+
+/// Interceptor hook for <c>strcasecmp()</c>.
+///
+/// \param called_pc PC (program counter) address of the original call.
+/// \param s1 Pointer to block of memory.
+/// \param s2 Pointer to block of memory.
+/// \param result Value returned by the intercepted function.
+void __sanitizer_weak_hook_strcasecmp(void *called_pc, const char *s1,
+ const char *s2, int result);
+
+/// Interceptor hook for <c>strstr()</c>.
+///
+/// \param called_pc PC (program counter) address of the original call.
+/// \param s1 Pointer to block of memory.
+/// \param s2 Pointer to block of memory.
+/// \param result Value returned by the intercepted function.
+void __sanitizer_weak_hook_strstr(void *called_pc, const char *s1,
+ const char *s2, char *result);
+
+void __sanitizer_weak_hook_strcasestr(void *called_pc, const char *s1,
+ const char *s2, char *result);
+
+void __sanitizer_weak_hook_memmem(void *called_pc,
+ const void *s1, size_t len1,
+ const void *s2, size_t len2, void *result);
+
+// Prints stack traces for all live heap allocations ordered by total
+// allocation size until top_percent of total live heap is shown. top_percent
+// should be between 1 and 100. At most max_number_of_contexts contexts
+// (stack traces) are printed.
+// Experimental feature currently available only with ASan on Linux/x86_64.
+void __sanitizer_print_memory_profile(size_t top_percent,
+ size_t max_number_of_contexts);
+
+/// Notify ASan that a fiber switch has started (required only if implementing
+/// your own fiber library).
+///
+/// Before switching to a different stack, you must call
+/// <c>__sanitizer_start_switch_fiber()</c> with a pointer to the bottom of the
+/// destination stack and with its size. When code starts running on the new
+/// stack, it must call <c>__sanitizer_finish_switch_fiber()</c> to finalize
+/// the switch. The <c>__sanitizer_start_switch_fiber()</c> function takes a
+/// <c>void**</c> pointer argument to store the current fake stack if there is
+/// one (it is necessary when the runtime option
+/// <c>detect_stack_use_after_return</c> is enabled).
+///
+/// When restoring a stack, this <c>void**</c> pointer must be given to the
+/// <c>__sanitizer_finish_switch_fiber()</c> function. In most cases, this
+/// pointer can be stored on the stack immediately before switching. When
+/// leaving a fiber definitely, NULL must be passed as the first argument to
+/// the <c>__sanitizer_start_switch_fiber()</c> function so that the fake stack
+/// is destroyed. If your program does not need stack use-after-return
+/// detection, you can always pass NULL to these two functions.
+///
+/// \note The fake stack mechanism is disabled during fiber switch, so if a
+/// signal callback runs during the switch, it will not benefit from stack
+/// use-after-return detection.
+///
+/// \param fake_stack_save [out] Fake stack save location.
+/// \param bottom Bottom address of stack.
+/// \param size Size of stack in bytes.
+void __sanitizer_start_switch_fiber(void **fake_stack_save,
+ const void *bottom, size_t size);
+
+/// Notify ASan that a fiber switch has completed (required only if
+/// implementing your own fiber library).
+///
+/// When code starts running on the new stack, it must call
+/// <c>__sanitizer_finish_switch_fiber()</c> to finalize
+/// the switch. For usage details, see the description of
+/// <c>__sanitizer_start_switch_fiber()</c>.
+///
+/// \param fake_stack_save Fake stack save location.
+/// \param bottom_old [out] Bottom address of old stack.
+/// \param size_old [out] Size of old stack in bytes.
+void __sanitizer_finish_switch_fiber(void *fake_stack_save,
+ const void **bottom_old,
+ size_t *size_old);
+
+// Get full module name and calculate pc offset within it.
+// Returns 1 if pc belongs to some module, 0 if module was not found.
+int __sanitizer_get_module_and_offset_for_pc(void *pc, char *module_path,
+ size_t module_path_len,
+ void **pc_offset);
+
+#ifdef __cplusplus
+} // extern "C"
+#endif
+
+#endif // SANITIZER_COMMON_INTERFACE_DEFS_H
--- /dev/null
+//===-- sanitizer/coverage_interface.h --------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// Public interface for sanitizer coverage.
+//===----------------------------------------------------------------------===//
+
+#ifndef SANITIZER_COVERAG_INTERFACE_H
+#define SANITIZER_COVERAG_INTERFACE_H
+
+#include <sanitizer/common_interface_defs.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+ // Record and dump coverage info.
+ void __sanitizer_cov_dump(void);
+
+ // Clear collected coverage info.
+ void __sanitizer_cov_reset(void);
+
+ // Dump collected coverage info. Sorts pcs by module into individual .sancov
+ // files.
+ void __sanitizer_dump_coverage(const uintptr_t *pcs, uintptr_t len);
+
+#ifdef __cplusplus
+} // extern "C"
+#endif
+
+#endif // SANITIZER_COVERAG_INTERFACE_H
--- /dev/null
+//===-- dfsan_interface.h -------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file is a part of DataFlowSanitizer.
+//
+// Public interface header.
+//===----------------------------------------------------------------------===//
+#ifndef DFSAN_INTERFACE_H
+#define DFSAN_INTERFACE_H
+
+#include <stddef.h>
+#include <stdint.h>
+#include <sanitizer/common_interface_defs.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+typedef uint16_t dfsan_label;
+
+/// Stores information associated with a specific label identifier. A label
+/// may be a base label created using dfsan_create_label, with associated
+/// text description and user data, or an automatically created union label,
+/// which represents the union of two label identifiers (which may themselves
+/// be base or union labels).
+struct dfsan_label_info {
+ // Fields for union labels, set to 0 for base labels.
+ dfsan_label l1;
+ dfsan_label l2;
+
+ // Fields for base labels.
+ const char *desc;
+ void *userdata;
+};
+
+/// Signature of the callback argument to dfsan_set_write_callback().
+typedef void (*dfsan_write_callback_t)(int fd, const void *buf, size_t count);
+
+/// Computes the union of \c l1 and \c l2, possibly creating a union label in
+/// the process.
+dfsan_label dfsan_union(dfsan_label l1, dfsan_label l2);
+
+/// Creates and returns a base label with the given description and user data.
+dfsan_label dfsan_create_label(const char *desc, void *userdata);
+
+/// Sets the label for each address in [addr,addr+size) to \c label.
+void dfsan_set_label(dfsan_label label, void *addr, size_t size);
+
+/// Sets the label for each address in [addr,addr+size) to the union of the
+/// current label for that address and \c label.
+void dfsan_add_label(dfsan_label label, void *addr, size_t size);
+
+/// Retrieves the label associated with the given data.
+///
+/// The type of 'data' is arbitrary. The function accepts a value of any type,
+/// which can be truncated or extended (implicitly or explicitly) as necessary.
+/// The truncation/extension operations will preserve the label of the original
+/// value.
+dfsan_label dfsan_get_label(long data);
+
+/// Retrieves the label associated with the data at the given address.
+dfsan_label dfsan_read_label(const void *addr, size_t size);
+
+/// Retrieves a pointer to the dfsan_label_info struct for the given label.
+const struct dfsan_label_info *dfsan_get_label_info(dfsan_label label);
+
+/// Returns whether the given label label contains the label elem.
+int dfsan_has_label(dfsan_label label, dfsan_label elem);
+
+/// If the given label label contains a label with the description desc, returns
+/// that label, else returns 0.
+dfsan_label dfsan_has_label_with_desc(dfsan_label label, const char *desc);
+
+/// Returns the number of labels allocated.
+size_t dfsan_get_label_count(void);
+
+/// Flushes the DFSan shadow, i.e. forgets about all labels currently associated
+/// with the application memory. Will work only if there are no other
+/// threads executing DFSan-instrumented code concurrently.
+/// Use this call to start over the taint tracking within the same procces.
+void dfsan_flush(void);
+
+/// Sets a callback to be invoked on calls to write(). The callback is invoked
+/// before the write is done. The write is not guaranteed to succeed when the
+/// callback executes. Pass in NULL to remove any callback.
+void dfsan_set_write_callback(dfsan_write_callback_t labeled_write_callback);
+
+/// Writes the labels currently used by the program to the given file
+/// descriptor. The lines of the output have the following format:
+///
+/// <label> <parent label 1> <parent label 2> <label description if any>
+void dfsan_dump_labels(int fd);
+
+/// Interceptor hooks.
+/// Whenever a dfsan's custom function is called the corresponding
+/// hook is called it non-zero. The hooks should be defined by the user.
+/// The primary use case is taint-guided fuzzing, where the fuzzer
+/// needs to see the parameters of the function and the labels.
+/// FIXME: implement more hooks.
+void dfsan_weak_hook_memcmp(void *caller_pc, const void *s1, const void *s2,
+ size_t n, dfsan_label s1_label,
+ dfsan_label s2_label, dfsan_label n_label);
+void dfsan_weak_hook_strncmp(void *caller_pc, const char *s1, const char *s2,
+ size_t n, dfsan_label s1_label,
+ dfsan_label s2_label, dfsan_label n_label);
+#ifdef __cplusplus
+} // extern "C"
+
+template <typename T>
+void dfsan_set_label(dfsan_label label, T &data) { // NOLINT
+ dfsan_set_label(label, (void *)&data, sizeof(T));
+}
+
+#endif
+
+#endif // DFSAN_INTERFACE_H
--- /dev/null
+//===-- sanitizer/asan_interface.h ------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file is a part of HWAddressSanitizer.
+//
+// Public interface header.
+//===----------------------------------------------------------------------===//
+#ifndef SANITIZER_HWASAN_INTERFACE_H
+#define SANITIZER_HWASAN_INTERFACE_H
+
+#include <sanitizer/common_interface_defs.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+ // Libc hook for program startup in statically linked executables.
+ // Initializes enough of the runtime to run instrumented code. This function
+ // should only be called in statically linked executables because it modifies
+ // the GOT, which won't work in regular binaries because RELRO will already
+ // have been applied by the time the function is called. This also means that
+ // the function should be called before libc applies RELRO.
+ // Does not call libc unless there is an error.
+ // Can be called multiple times.
+ void __hwasan_init_static(void);
+
+ // This function may be optionally provided by user and should return
+ // a string containing HWASan runtime options. See asan_flags.h for details.
+ const char* __hwasan_default_options(void);
+
+ void __hwasan_enable_allocator_tagging(void);
+ void __hwasan_disable_allocator_tagging(void);
+
+ // Mark region of memory with the given tag. Both address and size need to be
+ // 16-byte aligned.
+ void __hwasan_tag_memory(const volatile void *p, unsigned char tag,
+ size_t size);
+
+ /// Set pointer tag. Previous tag is lost.
+ void *__hwasan_tag_pointer(const volatile void *p, unsigned char tag);
+
+ // Set memory tag from the current SP address to the given address to zero.
+ // This is meant to annotate longjmp and other non-local jumps.
+ // This function needs to know the (almost) exact destination frame address;
+ // clearing shadow for the entire thread stack like __asan_handle_no_return
+ // does would cause false reports.
+ void __hwasan_handle_longjmp(const void *sp_dst);
+
+ // Set memory tag for the part of the current thread stack below sp_dst to
+ // zero. Call this in vfork() before returning in the parent process.
+ void __hwasan_handle_vfork(const void *sp_dst);
+
+ // Libc hook for thread creation. Should be called in the child thread before
+ // any instrumented code.
+ void __hwasan_thread_enter();
+
+ // Libc hook for thread destruction. No instrumented code should run after
+ // this call.
+ void __hwasan_thread_exit();
+
+ // Print shadow and origin for the memory range to stderr in a human-readable
+ // format.
+ void __hwasan_print_shadow(const volatile void *x, size_t size);
+
+ // Print one-line report about the memory usage of the current process.
+ void __hwasan_print_memory_usage();
+
+ /* Returns the offset of the first byte in the memory range that can not be
+ * accessed through the pointer in x, or -1 if the whole range is good. */
+ intptr_t __hwasan_test_shadow(const volatile void *x, size_t size);
+
+ int __sanitizer_posix_memalign(void **memptr, size_t alignment, size_t size);
+ void * __sanitizer_memalign(size_t alignment, size_t size);
+ void * __sanitizer_aligned_alloc(size_t alignment, size_t size);
+ void * __sanitizer___libc_memalign(size_t alignment, size_t size);
+ void * __sanitizer_valloc(size_t size);
+ void * __sanitizer_pvalloc(size_t size);
+ void __sanitizer_free(void *ptr);
+ void __sanitizer_cfree(void *ptr);
+ size_t __sanitizer_malloc_usable_size(const void *ptr);
+ struct mallinfo __sanitizer_mallinfo();
+ int __sanitizer_mallopt(int cmd, int value);
+ void __sanitizer_malloc_stats(void);
+ void * __sanitizer_calloc(size_t nmemb, size_t size);
+ void * __sanitizer_realloc(void *ptr, size_t size);
+ void * __sanitizer_reallocarray(void *ptr, size_t nmemb, size_t size);
+ void * __sanitizer_malloc(size_t size);
+#ifdef __cplusplus
+} // extern "C"
+#endif
+
+#endif // SANITIZER_HWASAN_INTERFACE_H
--- /dev/null
+//===-- linux_syscall_hooks.h ---------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file is a part of public sanitizer interface.
+//
+// System call handlers.
+//
+// Interface methods declared in this header implement pre- and post- syscall
+// actions for the active sanitizer.
+// Usage:
+// __sanitizer_syscall_pre_getfoo(...args...);
+// long res = syscall(__NR_getfoo, ...args...);
+// __sanitizer_syscall_post_getfoo(res, ...args...);
+//===----------------------------------------------------------------------===//
+#ifndef SANITIZER_LINUX_SYSCALL_HOOKS_H
+#define SANITIZER_LINUX_SYSCALL_HOOKS_H
+
+#define __sanitizer_syscall_pre_time(tloc) \
+ __sanitizer_syscall_pre_impl_time((long)(tloc))
+#define __sanitizer_syscall_post_time(res, tloc) \
+ __sanitizer_syscall_post_impl_time(res, (long)(tloc))
+#define __sanitizer_syscall_pre_stime(tptr) \
+ __sanitizer_syscall_pre_impl_stime((long)(tptr))
+#define __sanitizer_syscall_post_stime(res, tptr) \
+ __sanitizer_syscall_post_impl_stime(res, (long)(tptr))
+#define __sanitizer_syscall_pre_gettimeofday(tv, tz) \
+ __sanitizer_syscall_pre_impl_gettimeofday((long)(tv), (long)(tz))
+#define __sanitizer_syscall_post_gettimeofday(res, tv, tz) \
+ __sanitizer_syscall_post_impl_gettimeofday(res, (long)(tv), (long)(tz))
+#define __sanitizer_syscall_pre_settimeofday(tv, tz) \
+ __sanitizer_syscall_pre_impl_settimeofday((long)(tv), (long)(tz))
+#define __sanitizer_syscall_post_settimeofday(res, tv, tz) \
+ __sanitizer_syscall_post_impl_settimeofday(res, (long)(tv), (long)(tz))
+#define __sanitizer_syscall_pre_adjtimex(txc_p) \
+ __sanitizer_syscall_pre_impl_adjtimex((long)(txc_p))
+#define __sanitizer_syscall_post_adjtimex(res, txc_p) \
+ __sanitizer_syscall_post_impl_adjtimex(res, (long)(txc_p))
+#define __sanitizer_syscall_pre_times(tbuf) \
+ __sanitizer_syscall_pre_impl_times((long)(tbuf))
+#define __sanitizer_syscall_post_times(res, tbuf) \
+ __sanitizer_syscall_post_impl_times(res, (long)(tbuf))
+#define __sanitizer_syscall_pre_gettid() __sanitizer_syscall_pre_impl_gettid()
+#define __sanitizer_syscall_post_gettid(res) \
+ __sanitizer_syscall_post_impl_gettid(res)
+#define __sanitizer_syscall_pre_nanosleep(rqtp, rmtp) \
+ __sanitizer_syscall_pre_impl_nanosleep((long)(rqtp), (long)(rmtp))
+#define __sanitizer_syscall_post_nanosleep(res, rqtp, rmtp) \
+ __sanitizer_syscall_post_impl_nanosleep(res, (long)(rqtp), (long)(rmtp))
+#define __sanitizer_syscall_pre_alarm(seconds) \
+ __sanitizer_syscall_pre_impl_alarm((long)(seconds))
+#define __sanitizer_syscall_post_alarm(res, seconds) \
+ __sanitizer_syscall_post_impl_alarm(res, (long)(seconds))
+#define __sanitizer_syscall_pre_getpid() __sanitizer_syscall_pre_impl_getpid()
+#define __sanitizer_syscall_post_getpid(res) \
+ __sanitizer_syscall_post_impl_getpid(res)
+#define __sanitizer_syscall_pre_getppid() __sanitizer_syscall_pre_impl_getppid()
+#define __sanitizer_syscall_post_getppid(res) \
+ __sanitizer_syscall_post_impl_getppid(res)
+#define __sanitizer_syscall_pre_getuid() __sanitizer_syscall_pre_impl_getuid()
+#define __sanitizer_syscall_post_getuid(res) \
+ __sanitizer_syscall_post_impl_getuid(res)
+#define __sanitizer_syscall_pre_geteuid() __sanitizer_syscall_pre_impl_geteuid()
+#define __sanitizer_syscall_post_geteuid(res) \
+ __sanitizer_syscall_post_impl_geteuid(res)
+#define __sanitizer_syscall_pre_getgid() __sanitizer_syscall_pre_impl_getgid()
+#define __sanitizer_syscall_post_getgid(res) \
+ __sanitizer_syscall_post_impl_getgid(res)
+#define __sanitizer_syscall_pre_getegid() __sanitizer_syscall_pre_impl_getegid()
+#define __sanitizer_syscall_post_getegid(res) \
+ __sanitizer_syscall_post_impl_getegid(res)
+#define __sanitizer_syscall_pre_getresuid(ruid, euid, suid) \
+ __sanitizer_syscall_pre_impl_getresuid((long)(ruid), (long)(euid), \
+ (long)(suid))
+#define __sanitizer_syscall_post_getresuid(res, ruid, euid, suid) \
+ __sanitizer_syscall_post_impl_getresuid(res, (long)(ruid), (long)(euid), \
+ (long)(suid))
+#define __sanitizer_syscall_pre_getresgid(rgid, egid, sgid) \
+ __sanitizer_syscall_pre_impl_getresgid((long)(rgid), (long)(egid), \
+ (long)(sgid))
+#define __sanitizer_syscall_post_getresgid(res, rgid, egid, sgid) \
+ __sanitizer_syscall_post_impl_getresgid(res, (long)(rgid), (long)(egid), \
+ (long)(sgid))
+#define __sanitizer_syscall_pre_getpgid(pid) \
+ __sanitizer_syscall_pre_impl_getpgid((long)(pid))
+#define __sanitizer_syscall_post_getpgid(res, pid) \
+ __sanitizer_syscall_post_impl_getpgid(res, (long)(pid))
+#define __sanitizer_syscall_pre_getpgrp() __sanitizer_syscall_pre_impl_getpgrp()
+#define __sanitizer_syscall_post_getpgrp(res) \
+ __sanitizer_syscall_post_impl_getpgrp(res)
+#define __sanitizer_syscall_pre_getsid(pid) \
+ __sanitizer_syscall_pre_impl_getsid((long)(pid))
+#define __sanitizer_syscall_post_getsid(res, pid) \
+ __sanitizer_syscall_post_impl_getsid(res, (long)(pid))
+#define __sanitizer_syscall_pre_getgroups(gidsetsize, grouplist) \
+ __sanitizer_syscall_pre_impl_getgroups((long)(gidsetsize), (long)(grouplist))
+#define __sanitizer_syscall_post_getgroups(res, gidsetsize, grouplist) \
+ __sanitizer_syscall_post_impl_getgroups(res, (long)(gidsetsize), \
+ (long)(grouplist))
+#define __sanitizer_syscall_pre_setregid(rgid, egid) \
+ __sanitizer_syscall_pre_impl_setregid((long)(rgid), (long)(egid))
+#define __sanitizer_syscall_post_setregid(res, rgid, egid) \
+ __sanitizer_syscall_post_impl_setregid(res, (long)(rgid), (long)(egid))
+#define __sanitizer_syscall_pre_setgid(gid) \
+ __sanitizer_syscall_pre_impl_setgid((long)(gid))
+#define __sanitizer_syscall_post_setgid(res, gid) \
+ __sanitizer_syscall_post_impl_setgid(res, (long)(gid))
+#define __sanitizer_syscall_pre_setreuid(ruid, euid) \
+ __sanitizer_syscall_pre_impl_setreuid((long)(ruid), (long)(euid))
+#define __sanitizer_syscall_post_setreuid(res, ruid, euid) \
+ __sanitizer_syscall_post_impl_setreuid(res, (long)(ruid), (long)(euid))
+#define __sanitizer_syscall_pre_setuid(uid) \
+ __sanitizer_syscall_pre_impl_setuid((long)(uid))
+#define __sanitizer_syscall_post_setuid(res, uid) \
+ __sanitizer_syscall_post_impl_setuid(res, (long)(uid))
+#define __sanitizer_syscall_pre_setresuid(ruid, euid, suid) \
+ __sanitizer_syscall_pre_impl_setresuid((long)(ruid), (long)(euid), \
+ (long)(suid))
+#define __sanitizer_syscall_post_setresuid(res, ruid, euid, suid) \
+ __sanitizer_syscall_post_impl_setresuid(res, (long)(ruid), (long)(euid), \
+ (long)(suid))
+#define __sanitizer_syscall_pre_setresgid(rgid, egid, sgid) \
+ __sanitizer_syscall_pre_impl_setresgid((long)(rgid), (long)(egid), \
+ (long)(sgid))
+#define __sanitizer_syscall_post_setresgid(res, rgid, egid, sgid) \
+ __sanitizer_syscall_post_impl_setresgid(res, (long)(rgid), (long)(egid), \
+ (long)(sgid))
+#define __sanitizer_syscall_pre_setfsuid(uid) \
+ __sanitizer_syscall_pre_impl_setfsuid((long)(uid))
+#define __sanitizer_syscall_post_setfsuid(res, uid) \
+ __sanitizer_syscall_post_impl_setfsuid(res, (long)(uid))
+#define __sanitizer_syscall_pre_setfsgid(gid) \
+ __sanitizer_syscall_pre_impl_setfsgid((long)(gid))
+#define __sanitizer_syscall_post_setfsgid(res, gid) \
+ __sanitizer_syscall_post_impl_setfsgid(res, (long)(gid))
+#define __sanitizer_syscall_pre_setpgid(pid, pgid) \
+ __sanitizer_syscall_pre_impl_setpgid((long)(pid), (long)(pgid))
+#define __sanitizer_syscall_post_setpgid(res, pid, pgid) \
+ __sanitizer_syscall_post_impl_setpgid(res, (long)(pid), (long)(pgid))
+#define __sanitizer_syscall_pre_setsid() __sanitizer_syscall_pre_impl_setsid()
+#define __sanitizer_syscall_post_setsid(res) \
+ __sanitizer_syscall_post_impl_setsid(res)
+#define __sanitizer_syscall_pre_setgroups(gidsetsize, grouplist) \
+ __sanitizer_syscall_pre_impl_setgroups((long)(gidsetsize), (long)(grouplist))
+#define __sanitizer_syscall_post_setgroups(res, gidsetsize, grouplist) \
+ __sanitizer_syscall_post_impl_setgroups(res, (long)(gidsetsize), \
+ (long)(grouplist))
+#define __sanitizer_syscall_pre_acct(name) \
+ __sanitizer_syscall_pre_impl_acct((long)(name))
+#define __sanitizer_syscall_post_acct(res, name) \
+ __sanitizer_syscall_post_impl_acct(res, (long)(name))
+#define __sanitizer_syscall_pre_capget(header, dataptr) \
+ __sanitizer_syscall_pre_impl_capget((long)(header), (long)(dataptr))
+#define __sanitizer_syscall_post_capget(res, header, dataptr) \
+ __sanitizer_syscall_post_impl_capget(res, (long)(header), (long)(dataptr))
+#define __sanitizer_syscall_pre_capset(header, data) \
+ __sanitizer_syscall_pre_impl_capset((long)(header), (long)(data))
+#define __sanitizer_syscall_post_capset(res, header, data) \
+ __sanitizer_syscall_post_impl_capset(res, (long)(header), (long)(data))
+#define __sanitizer_syscall_pre_personality(personality) \
+ __sanitizer_syscall_pre_impl_personality((long)(personality))
+#define __sanitizer_syscall_post_personality(res, personality) \
+ __sanitizer_syscall_post_impl_personality(res, (long)(personality))
+#define __sanitizer_syscall_pre_sigpending(set) \
+ __sanitizer_syscall_pre_impl_sigpending((long)(set))
+#define __sanitizer_syscall_post_sigpending(res, set) \
+ __sanitizer_syscall_post_impl_sigpending(res, (long)(set))
+#define __sanitizer_syscall_pre_sigprocmask(how, set, oset) \
+ __sanitizer_syscall_pre_impl_sigprocmask((long)(how), (long)(set), \
+ (long)(oset))
+#define __sanitizer_syscall_post_sigprocmask(res, how, set, oset) \
+ __sanitizer_syscall_post_impl_sigprocmask(res, (long)(how), (long)(set), \
+ (long)(oset))
+#define __sanitizer_syscall_pre_getitimer(which, value) \
+ __sanitizer_syscall_pre_impl_getitimer((long)(which), (long)(value))
+#define __sanitizer_syscall_post_getitimer(res, which, value) \
+ __sanitizer_syscall_post_impl_getitimer(res, (long)(which), (long)(value))
+#define __sanitizer_syscall_pre_setitimer(which, value, ovalue) \
+ __sanitizer_syscall_pre_impl_setitimer((long)(which), (long)(value), \
+ (long)(ovalue))
+#define __sanitizer_syscall_post_setitimer(res, which, value, ovalue) \
+ __sanitizer_syscall_post_impl_setitimer(res, (long)(which), (long)(value), \
+ (long)(ovalue))
+#define __sanitizer_syscall_pre_timer_create(which_clock, timer_event_spec, \
+ created_timer_id) \
+ __sanitizer_syscall_pre_impl_timer_create( \
+ (long)(which_clock), (long)(timer_event_spec), (long)(created_timer_id))
+#define __sanitizer_syscall_post_timer_create( \
+ res, which_clock, timer_event_spec, created_timer_id) \
+ __sanitizer_syscall_post_impl_timer_create(res, (long)(which_clock), \
+ (long)(timer_event_spec), \
+ (long)(created_timer_id))
+#define __sanitizer_syscall_pre_timer_gettime(timer_id, setting) \
+ __sanitizer_syscall_pre_impl_timer_gettime((long)(timer_id), (long)(setting))
+#define __sanitizer_syscall_post_timer_gettime(res, timer_id, setting) \
+ __sanitizer_syscall_post_impl_timer_gettime(res, (long)(timer_id), \
+ (long)(setting))
+#define __sanitizer_syscall_pre_timer_getoverrun(timer_id) \
+ __sanitizer_syscall_pre_impl_timer_getoverrun((long)(timer_id))
+#define __sanitizer_syscall_post_timer_getoverrun(res, timer_id) \
+ __sanitizer_syscall_post_impl_timer_getoverrun(res, (long)(timer_id))
+#define __sanitizer_syscall_pre_timer_settime(timer_id, flags, new_setting, \
+ old_setting) \
+ __sanitizer_syscall_pre_impl_timer_settime((long)(timer_id), (long)(flags), \
+ (long)(new_setting), \
+ (long)(old_setting))
+#define __sanitizer_syscall_post_timer_settime(res, timer_id, flags, \
+ new_setting, old_setting) \
+ __sanitizer_syscall_post_impl_timer_settime( \
+ res, (long)(timer_id), (long)(flags), (long)(new_setting), \
+ (long)(old_setting))
+#define __sanitizer_syscall_pre_timer_delete(timer_id) \
+ __sanitizer_syscall_pre_impl_timer_delete((long)(timer_id))
+#define __sanitizer_syscall_post_timer_delete(res, timer_id) \
+ __sanitizer_syscall_post_impl_timer_delete(res, (long)(timer_id))
+#define __sanitizer_syscall_pre_clock_settime(which_clock, tp) \
+ __sanitizer_syscall_pre_impl_clock_settime((long)(which_clock), (long)(tp))
+#define __sanitizer_syscall_post_clock_settime(res, which_clock, tp) \
+ __sanitizer_syscall_post_impl_clock_settime(res, (long)(which_clock), \
+ (long)(tp))
+#define __sanitizer_syscall_pre_clock_gettime(which_clock, tp) \
+ __sanitizer_syscall_pre_impl_clock_gettime((long)(which_clock), (long)(tp))
+#define __sanitizer_syscall_post_clock_gettime(res, which_clock, tp) \
+ __sanitizer_syscall_post_impl_clock_gettime(res, (long)(which_clock), \
+ (long)(tp))
+#define __sanitizer_syscall_pre_clock_adjtime(which_clock, tx) \
+ __sanitizer_syscall_pre_impl_clock_adjtime((long)(which_clock), (long)(tx))
+#define __sanitizer_syscall_post_clock_adjtime(res, which_clock, tx) \
+ __sanitizer_syscall_post_impl_clock_adjtime(res, (long)(which_clock), \
+ (long)(tx))
+#define __sanitizer_syscall_pre_clock_getres(which_clock, tp) \
+ __sanitizer_syscall_pre_impl_clock_getres((long)(which_clock), (long)(tp))
+#define __sanitizer_syscall_post_clock_getres(res, which_clock, tp) \
+ __sanitizer_syscall_post_impl_clock_getres(res, (long)(which_clock), \
+ (long)(tp))
+#define __sanitizer_syscall_pre_clock_nanosleep(which_clock, flags, rqtp, \
+ rmtp) \
+ __sanitizer_syscall_pre_impl_clock_nanosleep( \
+ (long)(which_clock), (long)(flags), (long)(rqtp), (long)(rmtp))
+#define __sanitizer_syscall_post_clock_nanosleep(res, which_clock, flags, \
+ rqtp, rmtp) \
+ __sanitizer_syscall_post_impl_clock_nanosleep( \
+ res, (long)(which_clock), (long)(flags), (long)(rqtp), (long)(rmtp))
+#define __sanitizer_syscall_pre_nice(increment) \
+ __sanitizer_syscall_pre_impl_nice((long)(increment))
+#define __sanitizer_syscall_post_nice(res, increment) \
+ __sanitizer_syscall_post_impl_nice(res, (long)(increment))
+#define __sanitizer_syscall_pre_sched_setscheduler(pid, policy, param) \
+ __sanitizer_syscall_pre_impl_sched_setscheduler((long)(pid), (long)(policy), \
+ (long)(param))
+#define __sanitizer_syscall_post_sched_setscheduler(res, pid, policy, param) \
+ __sanitizer_syscall_post_impl_sched_setscheduler( \
+ res, (long)(pid), (long)(policy), (long)(param))
+#define __sanitizer_syscall_pre_sched_setparam(pid, param) \
+ __sanitizer_syscall_pre_impl_sched_setparam((long)(pid), (long)(param))
+#define __sanitizer_syscall_post_sched_setparam(res, pid, param) \
+ __sanitizer_syscall_post_impl_sched_setparam(res, (long)(pid), (long)(param))
+#define __sanitizer_syscall_pre_sched_getscheduler(pid) \
+ __sanitizer_syscall_pre_impl_sched_getscheduler((long)(pid))
+#define __sanitizer_syscall_post_sched_getscheduler(res, pid) \
+ __sanitizer_syscall_post_impl_sched_getscheduler(res, (long)(pid))
+#define __sanitizer_syscall_pre_sched_getparam(pid, param) \
+ __sanitizer_syscall_pre_impl_sched_getparam((long)(pid), (long)(param))
+#define __sanitizer_syscall_post_sched_getparam(res, pid, param) \
+ __sanitizer_syscall_post_impl_sched_getparam(res, (long)(pid), (long)(param))
+#define __sanitizer_syscall_pre_sched_setaffinity(pid, len, user_mask_ptr) \
+ __sanitizer_syscall_pre_impl_sched_setaffinity((long)(pid), (long)(len), \
+ (long)(user_mask_ptr))
+#define __sanitizer_syscall_post_sched_setaffinity(res, pid, len, \
+ user_mask_ptr) \
+ __sanitizer_syscall_post_impl_sched_setaffinity( \
+ res, (long)(pid), (long)(len), (long)(user_mask_ptr))
+#define __sanitizer_syscall_pre_sched_getaffinity(pid, len, user_mask_ptr) \
+ __sanitizer_syscall_pre_impl_sched_getaffinity((long)(pid), (long)(len), \
+ (long)(user_mask_ptr))
+#define __sanitizer_syscall_post_sched_getaffinity(res, pid, len, \
+ user_mask_ptr) \
+ __sanitizer_syscall_post_impl_sched_getaffinity( \
+ res, (long)(pid), (long)(len), (long)(user_mask_ptr))
+#define __sanitizer_syscall_pre_sched_yield() \
+ __sanitizer_syscall_pre_impl_sched_yield()
+#define __sanitizer_syscall_post_sched_yield(res) \
+ __sanitizer_syscall_post_impl_sched_yield(res)
+#define __sanitizer_syscall_pre_sched_get_priority_max(policy) \
+ __sanitizer_syscall_pre_impl_sched_get_priority_max((long)(policy))
+#define __sanitizer_syscall_post_sched_get_priority_max(res, policy) \
+ __sanitizer_syscall_post_impl_sched_get_priority_max(res, (long)(policy))
+#define __sanitizer_syscall_pre_sched_get_priority_min(policy) \
+ __sanitizer_syscall_pre_impl_sched_get_priority_min((long)(policy))
+#define __sanitizer_syscall_post_sched_get_priority_min(res, policy) \
+ __sanitizer_syscall_post_impl_sched_get_priority_min(res, (long)(policy))
+#define __sanitizer_syscall_pre_sched_rr_get_interval(pid, interval) \
+ __sanitizer_syscall_pre_impl_sched_rr_get_interval((long)(pid), \
+ (long)(interval))
+#define __sanitizer_syscall_post_sched_rr_get_interval(res, pid, interval) \
+ __sanitizer_syscall_post_impl_sched_rr_get_interval(res, (long)(pid), \
+ (long)(interval))
+#define __sanitizer_syscall_pre_setpriority(which, who, niceval) \
+ __sanitizer_syscall_pre_impl_setpriority((long)(which), (long)(who), \
+ (long)(niceval))
+#define __sanitizer_syscall_post_setpriority(res, which, who, niceval) \
+ __sanitizer_syscall_post_impl_setpriority(res, (long)(which), (long)(who), \
+ (long)(niceval))
+#define __sanitizer_syscall_pre_getpriority(which, who) \
+ __sanitizer_syscall_pre_impl_getpriority((long)(which), (long)(who))
+#define __sanitizer_syscall_post_getpriority(res, which, who) \
+ __sanitizer_syscall_post_impl_getpriority(res, (long)(which), (long)(who))
+#define __sanitizer_syscall_pre_shutdown(arg0, arg1) \
+ __sanitizer_syscall_pre_impl_shutdown((long)(arg0), (long)(arg1))
+#define __sanitizer_syscall_post_shutdown(res, arg0, arg1) \
+ __sanitizer_syscall_post_impl_shutdown(res, (long)(arg0), (long)(arg1))
+#define __sanitizer_syscall_pre_reboot(magic1, magic2, cmd, arg) \
+ __sanitizer_syscall_pre_impl_reboot((long)(magic1), (long)(magic2), \
+ (long)(cmd), (long)(arg))
+#define __sanitizer_syscall_post_reboot(res, magic1, magic2, cmd, arg) \
+ __sanitizer_syscall_post_impl_reboot(res, (long)(magic1), (long)(magic2), \
+ (long)(cmd), (long)(arg))
+#define __sanitizer_syscall_pre_restart_syscall() \
+ __sanitizer_syscall_pre_impl_restart_syscall()
+#define __sanitizer_syscall_post_restart_syscall(res) \
+ __sanitizer_syscall_post_impl_restart_syscall(res)
+#define __sanitizer_syscall_pre_kexec_load(entry, nr_segments, segments, \
+ flags) \
+ __sanitizer_syscall_pre_impl_kexec_load((long)(entry), (long)(nr_segments), \
+ (long)(segments), (long)(flags))
+#define __sanitizer_syscall_post_kexec_load(res, entry, nr_segments, segments, \
+ flags) \
+ __sanitizer_syscall_post_impl_kexec_load(res, (long)(entry), \
+ (long)(nr_segments), \
+ (long)(segments), (long)(flags))
+#define __sanitizer_syscall_pre_exit(error_code) \
+ __sanitizer_syscall_pre_impl_exit((long)(error_code))
+#define __sanitizer_syscall_post_exit(res, error_code) \
+ __sanitizer_syscall_post_impl_exit(res, (long)(error_code))
+#define __sanitizer_syscall_pre_exit_group(error_code) \
+ __sanitizer_syscall_pre_impl_exit_group((long)(error_code))
+#define __sanitizer_syscall_post_exit_group(res, error_code) \
+ __sanitizer_syscall_post_impl_exit_group(res, (long)(error_code))
+#define __sanitizer_syscall_pre_wait4(pid, stat_addr, options, ru) \
+ __sanitizer_syscall_pre_impl_wait4((long)(pid), (long)(stat_addr), \
+ (long)(options), (long)(ru))
+#define __sanitizer_syscall_post_wait4(res, pid, stat_addr, options, ru) \
+ __sanitizer_syscall_post_impl_wait4(res, (long)(pid), (long)(stat_addr), \
+ (long)(options), (long)(ru))
+#define __sanitizer_syscall_pre_waitid(which, pid, infop, options, ru) \
+ __sanitizer_syscall_pre_impl_waitid( \
+ (long)(which), (long)(pid), (long)(infop), (long)(options), (long)(ru))
+#define __sanitizer_syscall_post_waitid(res, which, pid, infop, options, ru) \
+ __sanitizer_syscall_post_impl_waitid(res, (long)(which), (long)(pid), \
+ (long)(infop), (long)(options), \
+ (long)(ru))
+#define __sanitizer_syscall_pre_waitpid(pid, stat_addr, options) \
+ __sanitizer_syscall_pre_impl_waitpid((long)(pid), (long)(stat_addr), \
+ (long)(options))
+#define __sanitizer_syscall_post_waitpid(res, pid, stat_addr, options) \
+ __sanitizer_syscall_post_impl_waitpid(res, (long)(pid), (long)(stat_addr), \
+ (long)(options))
+#define __sanitizer_syscall_pre_set_tid_address(tidptr) \
+ __sanitizer_syscall_pre_impl_set_tid_address((long)(tidptr))
+#define __sanitizer_syscall_post_set_tid_address(res, tidptr) \
+ __sanitizer_syscall_post_impl_set_tid_address(res, (long)(tidptr))
+#define __sanitizer_syscall_pre_init_module(umod, len, uargs) \
+ __sanitizer_syscall_pre_impl_init_module((long)(umod), (long)(len), \
+ (long)(uargs))
+#define __sanitizer_syscall_post_init_module(res, umod, len, uargs) \
+ __sanitizer_syscall_post_impl_init_module(res, (long)(umod), (long)(len), \
+ (long)(uargs))
+#define __sanitizer_syscall_pre_delete_module(name_user, flags) \
+ __sanitizer_syscall_pre_impl_delete_module((long)(name_user), (long)(flags))
+#define __sanitizer_syscall_post_delete_module(res, name_user, flags) \
+ __sanitizer_syscall_post_impl_delete_module(res, (long)(name_user), \
+ (long)(flags))
+#define __sanitizer_syscall_pre_rt_sigprocmask(how, set, oset, sigsetsize) \
+ __sanitizer_syscall_pre_impl_rt_sigprocmask( \
+ (long)(how), (long)(set), (long)(oset), (long)(sigsetsize))
+#define __sanitizer_syscall_post_rt_sigprocmask(res, how, set, oset, \
+ sigsetsize) \
+ __sanitizer_syscall_post_impl_rt_sigprocmask( \
+ res, (long)(how), (long)(set), (long)(oset), (long)(sigsetsize))
+#define __sanitizer_syscall_pre_rt_sigpending(set, sigsetsize) \
+ __sanitizer_syscall_pre_impl_rt_sigpending((long)(set), (long)(sigsetsize))
+#define __sanitizer_syscall_post_rt_sigpending(res, set, sigsetsize) \
+ __sanitizer_syscall_post_impl_rt_sigpending(res, (long)(set), \
+ (long)(sigsetsize))
+#define __sanitizer_syscall_pre_rt_sigtimedwait(uthese, uinfo, uts, \
+ sigsetsize) \
+ __sanitizer_syscall_pre_impl_rt_sigtimedwait( \
+ (long)(uthese), (long)(uinfo), (long)(uts), (long)(sigsetsize))
+#define __sanitizer_syscall_post_rt_sigtimedwait(res, uthese, uinfo, uts, \
+ sigsetsize) \
+ __sanitizer_syscall_post_impl_rt_sigtimedwait( \
+ res, (long)(uthese), (long)(uinfo), (long)(uts), (long)(sigsetsize))
+#define __sanitizer_syscall_pre_rt_tgsigqueueinfo(tgid, pid, sig, uinfo) \
+ __sanitizer_syscall_pre_impl_rt_tgsigqueueinfo((long)(tgid), (long)(pid), \
+ (long)(sig), (long)(uinfo))
+#define __sanitizer_syscall_post_rt_tgsigqueueinfo(res, tgid, pid, sig, uinfo) \
+ __sanitizer_syscall_post_impl_rt_tgsigqueueinfo( \
+ res, (long)(tgid), (long)(pid), (long)(sig), (long)(uinfo))
+#define __sanitizer_syscall_pre_kill(pid, sig) \
+ __sanitizer_syscall_pre_impl_kill((long)(pid), (long)(sig))
+#define __sanitizer_syscall_post_kill(res, pid, sig) \
+ __sanitizer_syscall_post_impl_kill(res, (long)(pid), (long)(sig))
+#define __sanitizer_syscall_pre_tgkill(tgid, pid, sig) \
+ __sanitizer_syscall_pre_impl_tgkill((long)(tgid), (long)(pid), (long)(sig))
+#define __sanitizer_syscall_post_tgkill(res, tgid, pid, sig) \
+ __sanitizer_syscall_post_impl_tgkill(res, (long)(tgid), (long)(pid), \
+ (long)(sig))
+#define __sanitizer_syscall_pre_tkill(pid, sig) \
+ __sanitizer_syscall_pre_impl_tkill((long)(pid), (long)(sig))
+#define __sanitizer_syscall_post_tkill(res, pid, sig) \
+ __sanitizer_syscall_post_impl_tkill(res, (long)(pid), (long)(sig))
+#define __sanitizer_syscall_pre_rt_sigqueueinfo(pid, sig, uinfo) \
+ __sanitizer_syscall_pre_impl_rt_sigqueueinfo((long)(pid), (long)(sig), \
+ (long)(uinfo))
+#define __sanitizer_syscall_post_rt_sigqueueinfo(res, pid, sig, uinfo) \
+ __sanitizer_syscall_post_impl_rt_sigqueueinfo(res, (long)(pid), (long)(sig), \
+ (long)(uinfo))
+#define __sanitizer_syscall_pre_sgetmask() \
+ __sanitizer_syscall_pre_impl_sgetmask()
+#define __sanitizer_syscall_post_sgetmask(res) \
+ __sanitizer_syscall_post_impl_sgetmask(res)
+#define __sanitizer_syscall_pre_ssetmask(newmask) \
+ __sanitizer_syscall_pre_impl_ssetmask((long)(newmask))
+#define __sanitizer_syscall_post_ssetmask(res, newmask) \
+ __sanitizer_syscall_post_impl_ssetmask(res, (long)(newmask))
+#define __sanitizer_syscall_pre_signal(sig, handler) \
+ __sanitizer_syscall_pre_impl_signal((long)(sig), (long)(handler))
+#define __sanitizer_syscall_post_signal(res, sig, handler) \
+ __sanitizer_syscall_post_impl_signal(res, (long)(sig), (long)(handler))
+#define __sanitizer_syscall_pre_pause() __sanitizer_syscall_pre_impl_pause()
+#define __sanitizer_syscall_post_pause(res) \
+ __sanitizer_syscall_post_impl_pause(res)
+#define __sanitizer_syscall_pre_sync() __sanitizer_syscall_pre_impl_sync()
+#define __sanitizer_syscall_post_sync(res) \
+ __sanitizer_syscall_post_impl_sync(res)
+#define __sanitizer_syscall_pre_fsync(fd) \
+ __sanitizer_syscall_pre_impl_fsync((long)(fd))
+#define __sanitizer_syscall_post_fsync(res, fd) \
+ __sanitizer_syscall_post_impl_fsync(res, (long)(fd))
+#define __sanitizer_syscall_pre_fdatasync(fd) \
+ __sanitizer_syscall_pre_impl_fdatasync((long)(fd))
+#define __sanitizer_syscall_post_fdatasync(res, fd) \
+ __sanitizer_syscall_post_impl_fdatasync(res, (long)(fd))
+#define __sanitizer_syscall_pre_bdflush(func, data) \
+ __sanitizer_syscall_pre_impl_bdflush((long)(func), (long)(data))
+#define __sanitizer_syscall_post_bdflush(res, func, data) \
+ __sanitizer_syscall_post_impl_bdflush(res, (long)(func), (long)(data))
+#define __sanitizer_syscall_pre_mount(dev_name, dir_name, type, flags, data) \
+ __sanitizer_syscall_pre_impl_mount((long)(dev_name), (long)(dir_name), \
+ (long)(type), (long)(flags), \
+ (long)(data))
+#define __sanitizer_syscall_post_mount(res, dev_name, dir_name, type, flags, \
+ data) \
+ __sanitizer_syscall_post_impl_mount(res, (long)(dev_name), (long)(dir_name), \
+ (long)(type), (long)(flags), \
+ (long)(data))
+#define __sanitizer_syscall_pre_umount(name, flags) \
+ __sanitizer_syscall_pre_impl_umount((long)(name), (long)(flags))
+#define __sanitizer_syscall_post_umount(res, name, flags) \
+ __sanitizer_syscall_post_impl_umount(res, (long)(name), (long)(flags))
+#define __sanitizer_syscall_pre_oldumount(name) \
+ __sanitizer_syscall_pre_impl_oldumount((long)(name))
+#define __sanitizer_syscall_post_oldumount(res, name) \
+ __sanitizer_syscall_post_impl_oldumount(res, (long)(name))
+#define __sanitizer_syscall_pre_truncate(path, length) \
+ __sanitizer_syscall_pre_impl_truncate((long)(path), (long)(length))
+#define __sanitizer_syscall_post_truncate(res, path, length) \
+ __sanitizer_syscall_post_impl_truncate(res, (long)(path), (long)(length))
+#define __sanitizer_syscall_pre_ftruncate(fd, length) \
+ __sanitizer_syscall_pre_impl_ftruncate((long)(fd), (long)(length))
+#define __sanitizer_syscall_post_ftruncate(res, fd, length) \
+ __sanitizer_syscall_post_impl_ftruncate(res, (long)(fd), (long)(length))
+#define __sanitizer_syscall_pre_stat(filename, statbuf) \
+ __sanitizer_syscall_pre_impl_stat((long)(filename), (long)(statbuf))
+#define __sanitizer_syscall_post_stat(res, filename, statbuf) \
+ __sanitizer_syscall_post_impl_stat(res, (long)(filename), (long)(statbuf))
+#define __sanitizer_syscall_pre_statfs(path, buf) \
+ __sanitizer_syscall_pre_impl_statfs((long)(path), (long)(buf))
+#define __sanitizer_syscall_post_statfs(res, path, buf) \
+ __sanitizer_syscall_post_impl_statfs(res, (long)(path), (long)(buf))
+#define __sanitizer_syscall_pre_statfs64(path, sz, buf) \
+ __sanitizer_syscall_pre_impl_statfs64((long)(path), (long)(sz), (long)(buf))
+#define __sanitizer_syscall_post_statfs64(res, path, sz, buf) \
+ __sanitizer_syscall_post_impl_statfs64(res, (long)(path), (long)(sz), \
+ (long)(buf))
+#define __sanitizer_syscall_pre_fstatfs(fd, buf) \
+ __sanitizer_syscall_pre_impl_fstatfs((long)(fd), (long)(buf))
+#define __sanitizer_syscall_post_fstatfs(res, fd, buf) \
+ __sanitizer_syscall_post_impl_fstatfs(res, (long)(fd), (long)(buf))
+#define __sanitizer_syscall_pre_fstatfs64(fd, sz, buf) \
+ __sanitizer_syscall_pre_impl_fstatfs64((long)(fd), (long)(sz), (long)(buf))
+#define __sanitizer_syscall_post_fstatfs64(res, fd, sz, buf) \
+ __sanitizer_syscall_post_impl_fstatfs64(res, (long)(fd), (long)(sz), \
+ (long)(buf))
+#define __sanitizer_syscall_pre_lstat(filename, statbuf) \
+ __sanitizer_syscall_pre_impl_lstat((long)(filename), (long)(statbuf))
+#define __sanitizer_syscall_post_lstat(res, filename, statbuf) \
+ __sanitizer_syscall_post_impl_lstat(res, (long)(filename), (long)(statbuf))
+#define __sanitizer_syscall_pre_fstat(fd, statbuf) \
+ __sanitizer_syscall_pre_impl_fstat((long)(fd), (long)(statbuf))
+#define __sanitizer_syscall_post_fstat(res, fd, statbuf) \
+ __sanitizer_syscall_post_impl_fstat(res, (long)(fd), (long)(statbuf))
+#define __sanitizer_syscall_pre_newstat(filename, statbuf) \
+ __sanitizer_syscall_pre_impl_newstat((long)(filename), (long)(statbuf))
+#define __sanitizer_syscall_post_newstat(res, filename, statbuf) \
+ __sanitizer_syscall_post_impl_newstat(res, (long)(filename), (long)(statbuf))
+#define __sanitizer_syscall_pre_newlstat(filename, statbuf) \
+ __sanitizer_syscall_pre_impl_newlstat((long)(filename), (long)(statbuf))
+#define __sanitizer_syscall_post_newlstat(res, filename, statbuf) \
+ __sanitizer_syscall_post_impl_newlstat(res, (long)(filename), (long)(statbuf))
+#define __sanitizer_syscall_pre_newfstat(fd, statbuf) \
+ __sanitizer_syscall_pre_impl_newfstat((long)(fd), (long)(statbuf))
+#define __sanitizer_syscall_post_newfstat(res, fd, statbuf) \
+ __sanitizer_syscall_post_impl_newfstat(res, (long)(fd), (long)(statbuf))
+#define __sanitizer_syscall_pre_ustat(dev, ubuf) \
+ __sanitizer_syscall_pre_impl_ustat((long)(dev), (long)(ubuf))
+#define __sanitizer_syscall_post_ustat(res, dev, ubuf) \
+ __sanitizer_syscall_post_impl_ustat(res, (long)(dev), (long)(ubuf))
+#define __sanitizer_syscall_pre_stat64(filename, statbuf) \
+ __sanitizer_syscall_pre_impl_stat64((long)(filename), (long)(statbuf))
+#define __sanitizer_syscall_post_stat64(res, filename, statbuf) \
+ __sanitizer_syscall_post_impl_stat64(res, (long)(filename), (long)(statbuf))
+#define __sanitizer_syscall_pre_fstat64(fd, statbuf) \
+ __sanitizer_syscall_pre_impl_fstat64((long)(fd), (long)(statbuf))
+#define __sanitizer_syscall_post_fstat64(res, fd, statbuf) \
+ __sanitizer_syscall_post_impl_fstat64(res, (long)(fd), (long)(statbuf))
+#define __sanitizer_syscall_pre_lstat64(filename, statbuf) \
+ __sanitizer_syscall_pre_impl_lstat64((long)(filename), (long)(statbuf))
+#define __sanitizer_syscall_post_lstat64(res, filename, statbuf) \
+ __sanitizer_syscall_post_impl_lstat64(res, (long)(filename), (long)(statbuf))
+#define __sanitizer_syscall_pre_setxattr(path, name, value, size, flags) \
+ __sanitizer_syscall_pre_impl_setxattr( \
+ (long)(path), (long)(name), (long)(value), (long)(size), (long)(flags))
+#define __sanitizer_syscall_post_setxattr(res, path, name, value, size, flags) \
+ __sanitizer_syscall_post_impl_setxattr(res, (long)(path), (long)(name), \
+ (long)(value), (long)(size), \
+ (long)(flags))
+#define __sanitizer_syscall_pre_lsetxattr(path, name, value, size, flags) \
+ __sanitizer_syscall_pre_impl_lsetxattr( \
+ (long)(path), (long)(name), (long)(value), (long)(size), (long)(flags))
+#define __sanitizer_syscall_post_lsetxattr(res, path, name, value, size, \
+ flags) \
+ __sanitizer_syscall_post_impl_lsetxattr(res, (long)(path), (long)(name), \
+ (long)(value), (long)(size), \
+ (long)(flags))
+#define __sanitizer_syscall_pre_fsetxattr(fd, name, value, size, flags) \
+ __sanitizer_syscall_pre_impl_fsetxattr( \
+ (long)(fd), (long)(name), (long)(value), (long)(size), (long)(flags))
+#define __sanitizer_syscall_post_fsetxattr(res, fd, name, value, size, flags) \
+ __sanitizer_syscall_post_impl_fsetxattr(res, (long)(fd), (long)(name), \
+ (long)(value), (long)(size), \
+ (long)(flags))
+#define __sanitizer_syscall_pre_getxattr(path, name, value, size) \
+ __sanitizer_syscall_pre_impl_getxattr((long)(path), (long)(name), \
+ (long)(value), (long)(size))
+#define __sanitizer_syscall_post_getxattr(res, path, name, value, size) \
+ __sanitizer_syscall_post_impl_getxattr(res, (long)(path), (long)(name), \
+ (long)(value), (long)(size))
+#define __sanitizer_syscall_pre_lgetxattr(path, name, value, size) \
+ __sanitizer_syscall_pre_impl_lgetxattr((long)(path), (long)(name), \
+ (long)(value), (long)(size))
+#define __sanitizer_syscall_post_lgetxattr(res, path, name, value, size) \
+ __sanitizer_syscall_post_impl_lgetxattr(res, (long)(path), (long)(name), \
+ (long)(value), (long)(size))
+#define __sanitizer_syscall_pre_fgetxattr(fd, name, value, size) \
+ __sanitizer_syscall_pre_impl_fgetxattr((long)(fd), (long)(name), \
+ (long)(value), (long)(size))
+#define __sanitizer_syscall_post_fgetxattr(res, fd, name, value, size) \
+ __sanitizer_syscall_post_impl_fgetxattr(res, (long)(fd), (long)(name), \
+ (long)(value), (long)(size))
+#define __sanitizer_syscall_pre_listxattr(path, list, size) \
+ __sanitizer_syscall_pre_impl_listxattr((long)(path), (long)(list), \
+ (long)(size))
+#define __sanitizer_syscall_post_listxattr(res, path, list, size) \
+ __sanitizer_syscall_post_impl_listxattr(res, (long)(path), (long)(list), \
+ (long)(size))
+#define __sanitizer_syscall_pre_llistxattr(path, list, size) \
+ __sanitizer_syscall_pre_impl_llistxattr((long)(path), (long)(list), \
+ (long)(size))
+#define __sanitizer_syscall_post_llistxattr(res, path, list, size) \
+ __sanitizer_syscall_post_impl_llistxattr(res, (long)(path), (long)(list), \
+ (long)(size))
+#define __sanitizer_syscall_pre_flistxattr(fd, list, size) \
+ __sanitizer_syscall_pre_impl_flistxattr((long)(fd), (long)(list), \
+ (long)(size))
+#define __sanitizer_syscall_post_flistxattr(res, fd, list, size) \
+ __sanitizer_syscall_post_impl_flistxattr(res, (long)(fd), (long)(list), \
+ (long)(size))
+#define __sanitizer_syscall_pre_removexattr(path, name) \
+ __sanitizer_syscall_pre_impl_removexattr((long)(path), (long)(name))
+#define __sanitizer_syscall_post_removexattr(res, path, name) \
+ __sanitizer_syscall_post_impl_removexattr(res, (long)(path), (long)(name))
+#define __sanitizer_syscall_pre_lremovexattr(path, name) \
+ __sanitizer_syscall_pre_impl_lremovexattr((long)(path), (long)(name))
+#define __sanitizer_syscall_post_lremovexattr(res, path, name) \
+ __sanitizer_syscall_post_impl_lremovexattr(res, (long)(path), (long)(name))
+#define __sanitizer_syscall_pre_fremovexattr(fd, name) \
+ __sanitizer_syscall_pre_impl_fremovexattr((long)(fd), (long)(name))
+#define __sanitizer_syscall_post_fremovexattr(res, fd, name) \
+ __sanitizer_syscall_post_impl_fremovexattr(res, (long)(fd), (long)(name))
+#define __sanitizer_syscall_pre_brk(brk) \
+ __sanitizer_syscall_pre_impl_brk((long)(brk))
+#define __sanitizer_syscall_post_brk(res, brk) \
+ __sanitizer_syscall_post_impl_brk(res, (long)(brk))
+#define __sanitizer_syscall_pre_mprotect(start, len, prot) \
+ __sanitizer_syscall_pre_impl_mprotect((long)(start), (long)(len), \
+ (long)(prot))
+#define __sanitizer_syscall_post_mprotect(res, start, len, prot) \
+ __sanitizer_syscall_post_impl_mprotect(res, (long)(start), (long)(len), \
+ (long)(prot))
+#define __sanitizer_syscall_pre_mremap(addr, old_len, new_len, flags, \
+ new_addr) \
+ __sanitizer_syscall_pre_impl_mremap((long)(addr), (long)(old_len), \
+ (long)(new_len), (long)(flags), \
+ (long)(new_addr))
+#define __sanitizer_syscall_post_mremap(res, addr, old_len, new_len, flags, \
+ new_addr) \
+ __sanitizer_syscall_post_impl_mremap(res, (long)(addr), (long)(old_len), \
+ (long)(new_len), (long)(flags), \
+ (long)(new_addr))
+#define __sanitizer_syscall_pre_remap_file_pages(start, size, prot, pgoff, \
+ flags) \
+ __sanitizer_syscall_pre_impl_remap_file_pages( \
+ (long)(start), (long)(size), (long)(prot), (long)(pgoff), (long)(flags))
+#define __sanitizer_syscall_post_remap_file_pages(res, start, size, prot, \
+ pgoff, flags) \
+ __sanitizer_syscall_post_impl_remap_file_pages(res, (long)(start), \
+ (long)(size), (long)(prot), \
+ (long)(pgoff), (long)(flags))
+#define __sanitizer_syscall_pre_msync(start, len, flags) \
+ __sanitizer_syscall_pre_impl_msync((long)(start), (long)(len), (long)(flags))
+#define __sanitizer_syscall_post_msync(res, start, len, flags) \
+ __sanitizer_syscall_post_impl_msync(res, (long)(start), (long)(len), \
+ (long)(flags))
+#define __sanitizer_syscall_pre_munmap(addr, len) \
+ __sanitizer_syscall_pre_impl_munmap((long)(addr), (long)(len))
+#define __sanitizer_syscall_post_munmap(res, addr, len) \
+ __sanitizer_syscall_post_impl_munmap(res, (long)(addr), (long)(len))
+#define __sanitizer_syscall_pre_mlock(start, len) \
+ __sanitizer_syscall_pre_impl_mlock((long)(start), (long)(len))
+#define __sanitizer_syscall_post_mlock(res, start, len) \
+ __sanitizer_syscall_post_impl_mlock(res, (long)(start), (long)(len))
+#define __sanitizer_syscall_pre_munlock(start, len) \
+ __sanitizer_syscall_pre_impl_munlock((long)(start), (long)(len))
+#define __sanitizer_syscall_post_munlock(res, start, len) \
+ __sanitizer_syscall_post_impl_munlock(res, (long)(start), (long)(len))
+#define __sanitizer_syscall_pre_mlockall(flags) \
+ __sanitizer_syscall_pre_impl_mlockall((long)(flags))
+#define __sanitizer_syscall_post_mlockall(res, flags) \
+ __sanitizer_syscall_post_impl_mlockall(res, (long)(flags))
+#define __sanitizer_syscall_pre_munlockall() \
+ __sanitizer_syscall_pre_impl_munlockall()
+#define __sanitizer_syscall_post_munlockall(res) \
+ __sanitizer_syscall_post_impl_munlockall(res)
+#define __sanitizer_syscall_pre_madvise(start, len, behavior) \
+ __sanitizer_syscall_pre_impl_madvise((long)(start), (long)(len), \
+ (long)(behavior))
+#define __sanitizer_syscall_post_madvise(res, start, len, behavior) \
+ __sanitizer_syscall_post_impl_madvise(res, (long)(start), (long)(len), \
+ (long)(behavior))
+#define __sanitizer_syscall_pre_mincore(start, len, vec) \
+ __sanitizer_syscall_pre_impl_mincore((long)(start), (long)(len), (long)(vec))
+#define __sanitizer_syscall_post_mincore(res, start, len, vec) \
+ __sanitizer_syscall_post_impl_mincore(res, (long)(start), (long)(len), \
+ (long)(vec))
+#define __sanitizer_syscall_pre_pivot_root(new_root, put_old) \
+ __sanitizer_syscall_pre_impl_pivot_root((long)(new_root), (long)(put_old))
+#define __sanitizer_syscall_post_pivot_root(res, new_root, put_old) \
+ __sanitizer_syscall_post_impl_pivot_root(res, (long)(new_root), \
+ (long)(put_old))
+#define __sanitizer_syscall_pre_chroot(filename) \
+ __sanitizer_syscall_pre_impl_chroot((long)(filename))
+#define __sanitizer_syscall_post_chroot(res, filename) \
+ __sanitizer_syscall_post_impl_chroot(res, (long)(filename))
+#define __sanitizer_syscall_pre_mknod(filename, mode, dev) \
+ __sanitizer_syscall_pre_impl_mknod((long)(filename), (long)(mode), \
+ (long)(dev))
+#define __sanitizer_syscall_post_mknod(res, filename, mode, dev) \
+ __sanitizer_syscall_post_impl_mknod(res, (long)(filename), (long)(mode), \
+ (long)(dev))
+#define __sanitizer_syscall_pre_link(oldname, newname) \
+ __sanitizer_syscall_pre_impl_link((long)(oldname), (long)(newname))
+#define __sanitizer_syscall_post_link(res, oldname, newname) \
+ __sanitizer_syscall_post_impl_link(res, (long)(oldname), (long)(newname))
+#define __sanitizer_syscall_pre_symlink(old, new_) \
+ __sanitizer_syscall_pre_impl_symlink((long)(old), (long)(new_))
+#define __sanitizer_syscall_post_symlink(res, old, new_) \
+ __sanitizer_syscall_post_impl_symlink(res, (long)(old), (long)(new_))
+#define __sanitizer_syscall_pre_unlink(pathname) \
+ __sanitizer_syscall_pre_impl_unlink((long)(pathname))
+#define __sanitizer_syscall_post_unlink(res, pathname) \
+ __sanitizer_syscall_post_impl_unlink(res, (long)(pathname))
+#define __sanitizer_syscall_pre_rename(oldname, newname) \
+ __sanitizer_syscall_pre_impl_rename((long)(oldname), (long)(newname))
+#define __sanitizer_syscall_post_rename(res, oldname, newname) \
+ __sanitizer_syscall_post_impl_rename(res, (long)(oldname), (long)(newname))
+#define __sanitizer_syscall_pre_chmod(filename, mode) \
+ __sanitizer_syscall_pre_impl_chmod((long)(filename), (long)(mode))
+#define __sanitizer_syscall_post_chmod(res, filename, mode) \
+ __sanitizer_syscall_post_impl_chmod(res, (long)(filename), (long)(mode))
+#define __sanitizer_syscall_pre_fchmod(fd, mode) \
+ __sanitizer_syscall_pre_impl_fchmod((long)(fd), (long)(mode))
+#define __sanitizer_syscall_post_fchmod(res, fd, mode) \
+ __sanitizer_syscall_post_impl_fchmod(res, (long)(fd), (long)(mode))
+#define __sanitizer_syscall_pre_fcntl(fd, cmd, arg) \
+ __sanitizer_syscall_pre_impl_fcntl((long)(fd), (long)(cmd), (long)(arg))
+#define __sanitizer_syscall_post_fcntl(res, fd, cmd, arg) \
+ __sanitizer_syscall_post_impl_fcntl(res, (long)(fd), (long)(cmd), (long)(arg))
+#define __sanitizer_syscall_pre_fcntl64(fd, cmd, arg) \
+ __sanitizer_syscall_pre_impl_fcntl64((long)(fd), (long)(cmd), (long)(arg))
+#define __sanitizer_syscall_post_fcntl64(res, fd, cmd, arg) \
+ __sanitizer_syscall_post_impl_fcntl64(res, (long)(fd), (long)(cmd), \
+ (long)(arg))
+#define __sanitizer_syscall_pre_pipe(fildes) \
+ __sanitizer_syscall_pre_impl_pipe((long)(fildes))
+#define __sanitizer_syscall_post_pipe(res, fildes) \
+ __sanitizer_syscall_post_impl_pipe(res, (long)(fildes))
+#define __sanitizer_syscall_pre_pipe2(fildes, flags) \
+ __sanitizer_syscall_pre_impl_pipe2((long)(fildes), (long)(flags))
+#define __sanitizer_syscall_post_pipe2(res, fildes, flags) \
+ __sanitizer_syscall_post_impl_pipe2(res, (long)(fildes), (long)(flags))
+#define __sanitizer_syscall_pre_dup(fildes) \
+ __sanitizer_syscall_pre_impl_dup((long)(fildes))
+#define __sanitizer_syscall_post_dup(res, fildes) \
+ __sanitizer_syscall_post_impl_dup(res, (long)(fildes))
+#define __sanitizer_syscall_pre_dup2(oldfd, newfd) \
+ __sanitizer_syscall_pre_impl_dup2((long)(oldfd), (long)(newfd))
+#define __sanitizer_syscall_post_dup2(res, oldfd, newfd) \
+ __sanitizer_syscall_post_impl_dup2(res, (long)(oldfd), (long)(newfd))
+#define __sanitizer_syscall_pre_dup3(oldfd, newfd, flags) \
+ __sanitizer_syscall_pre_impl_dup3((long)(oldfd), (long)(newfd), (long)(flags))
+#define __sanitizer_syscall_post_dup3(res, oldfd, newfd, flags) \
+ __sanitizer_syscall_post_impl_dup3(res, (long)(oldfd), (long)(newfd), \
+ (long)(flags))
+#define __sanitizer_syscall_pre_ioperm(from, num, on) \
+ __sanitizer_syscall_pre_impl_ioperm((long)(from), (long)(num), (long)(on))
+#define __sanitizer_syscall_post_ioperm(res, from, num, on) \
+ __sanitizer_syscall_post_impl_ioperm(res, (long)(from), (long)(num), \
+ (long)(on))
+#define __sanitizer_syscall_pre_ioctl(fd, cmd, arg) \
+ __sanitizer_syscall_pre_impl_ioctl((long)(fd), (long)(cmd), (long)(arg))
+#define __sanitizer_syscall_post_ioctl(res, fd, cmd, arg) \
+ __sanitizer_syscall_post_impl_ioctl(res, (long)(fd), (long)(cmd), (long)(arg))
+#define __sanitizer_syscall_pre_flock(fd, cmd) \
+ __sanitizer_syscall_pre_impl_flock((long)(fd), (long)(cmd))
+#define __sanitizer_syscall_post_flock(res, fd, cmd) \
+ __sanitizer_syscall_post_impl_flock(res, (long)(fd), (long)(cmd))
+#define __sanitizer_syscall_pre_io_setup(nr_reqs, ctx) \
+ __sanitizer_syscall_pre_impl_io_setup((long)(nr_reqs), (long)(ctx))
+#define __sanitizer_syscall_post_io_setup(res, nr_reqs, ctx) \
+ __sanitizer_syscall_post_impl_io_setup(res, (long)(nr_reqs), (long)(ctx))
+#define __sanitizer_syscall_pre_io_destroy(ctx) \
+ __sanitizer_syscall_pre_impl_io_destroy((long)(ctx))
+#define __sanitizer_syscall_post_io_destroy(res, ctx) \
+ __sanitizer_syscall_post_impl_io_destroy(res, (long)(ctx))
+#define __sanitizer_syscall_pre_io_getevents(ctx_id, min_nr, nr, events, \
+ timeout) \
+ __sanitizer_syscall_pre_impl_io_getevents((long)(ctx_id), (long)(min_nr), \
+ (long)(nr), (long)(events), \
+ (long)(timeout))
+#define __sanitizer_syscall_post_io_getevents(res, ctx_id, min_nr, nr, events, \
+ timeout) \
+ __sanitizer_syscall_post_impl_io_getevents(res, (long)(ctx_id), \
+ (long)(min_nr), (long)(nr), \
+ (long)(events), (long)(timeout))
+#define __sanitizer_syscall_pre_io_submit(ctx_id, arg1, arg2) \
+ __sanitizer_syscall_pre_impl_io_submit((long)(ctx_id), (long)(arg1), \
+ (long)(arg2))
+#define __sanitizer_syscall_post_io_submit(res, ctx_id, arg1, arg2) \
+ __sanitizer_syscall_post_impl_io_submit(res, (long)(ctx_id), (long)(arg1), \
+ (long)(arg2))
+#define __sanitizer_syscall_pre_io_cancel(ctx_id, iocb, result) \
+ __sanitizer_syscall_pre_impl_io_cancel((long)(ctx_id), (long)(iocb), \
+ (long)(result))
+#define __sanitizer_syscall_post_io_cancel(res, ctx_id, iocb, result) \
+ __sanitizer_syscall_post_impl_io_cancel(res, (long)(ctx_id), (long)(iocb), \
+ (long)(result))
+#define __sanitizer_syscall_pre_sendfile(out_fd, in_fd, offset, count) \
+ __sanitizer_syscall_pre_impl_sendfile((long)(out_fd), (long)(in_fd), \
+ (long)(offset), (long)(count))
+#define __sanitizer_syscall_post_sendfile(res, out_fd, in_fd, offset, count) \
+ __sanitizer_syscall_post_impl_sendfile(res, (long)(out_fd), (long)(in_fd), \
+ (long)(offset), (long)(count))
+#define __sanitizer_syscall_pre_sendfile64(out_fd, in_fd, offset, count) \
+ __sanitizer_syscall_pre_impl_sendfile64((long)(out_fd), (long)(in_fd), \
+ (long)(offset), (long)(count))
+#define __sanitizer_syscall_post_sendfile64(res, out_fd, in_fd, offset, count) \
+ __sanitizer_syscall_post_impl_sendfile64(res, (long)(out_fd), (long)(in_fd), \
+ (long)(offset), (long)(count))
+#define __sanitizer_syscall_pre_readlink(path, buf, bufsiz) \
+ __sanitizer_syscall_pre_impl_readlink((long)(path), (long)(buf), \
+ (long)(bufsiz))
+#define __sanitizer_syscall_post_readlink(res, path, buf, bufsiz) \
+ __sanitizer_syscall_post_impl_readlink(res, (long)(path), (long)(buf), \
+ (long)(bufsiz))
+#define __sanitizer_syscall_pre_creat(pathname, mode) \
+ __sanitizer_syscall_pre_impl_creat((long)(pathname), (long)(mode))
+#define __sanitizer_syscall_post_creat(res, pathname, mode) \
+ __sanitizer_syscall_post_impl_creat(res, (long)(pathname), (long)(mode))
+#define __sanitizer_syscall_pre_open(filename, flags, mode) \
+ __sanitizer_syscall_pre_impl_open((long)(filename), (long)(flags), \
+ (long)(mode))
+#define __sanitizer_syscall_post_open(res, filename, flags, mode) \
+ __sanitizer_syscall_post_impl_open(res, (long)(filename), (long)(flags), \
+ (long)(mode))
+#define __sanitizer_syscall_pre_close(fd) \
+ __sanitizer_syscall_pre_impl_close((long)(fd))
+#define __sanitizer_syscall_post_close(res, fd) \
+ __sanitizer_syscall_post_impl_close(res, (long)(fd))
+#define __sanitizer_syscall_pre_access(filename, mode) \
+ __sanitizer_syscall_pre_impl_access((long)(filename), (long)(mode))
+#define __sanitizer_syscall_post_access(res, filename, mode) \
+ __sanitizer_syscall_post_impl_access(res, (long)(filename), (long)(mode))
+#define __sanitizer_syscall_pre_vhangup() __sanitizer_syscall_pre_impl_vhangup()
+#define __sanitizer_syscall_post_vhangup(res) \
+ __sanitizer_syscall_post_impl_vhangup(res)
+#define __sanitizer_syscall_pre_chown(filename, user, group) \
+ __sanitizer_syscall_pre_impl_chown((long)(filename), (long)(user), \
+ (long)(group))
+#define __sanitizer_syscall_post_chown(res, filename, user, group) \
+ __sanitizer_syscall_post_impl_chown(res, (long)(filename), (long)(user), \
+ (long)(group))
+#define __sanitizer_syscall_pre_lchown(filename, user, group) \
+ __sanitizer_syscall_pre_impl_lchown((long)(filename), (long)(user), \
+ (long)(group))
+#define __sanitizer_syscall_post_lchown(res, filename, user, group) \
+ __sanitizer_syscall_post_impl_lchown(res, (long)(filename), (long)(user), \
+ (long)(group))
+#define __sanitizer_syscall_pre_fchown(fd, user, group) \
+ __sanitizer_syscall_pre_impl_fchown((long)(fd), (long)(user), (long)(group))
+#define __sanitizer_syscall_post_fchown(res, fd, user, group) \
+ __sanitizer_syscall_post_impl_fchown(res, (long)(fd), (long)(user), \
+ (long)(group))
+#define __sanitizer_syscall_pre_chown16(filename, user, group) \
+ __sanitizer_syscall_pre_impl_chown16((long)(filename), (long)user, \
+ (long)group)
+#define __sanitizer_syscall_post_chown16(res, filename, user, group) \
+ __sanitizer_syscall_post_impl_chown16(res, (long)(filename), (long)user, \
+ (long)group)
+#define __sanitizer_syscall_pre_lchown16(filename, user, group) \
+ __sanitizer_syscall_pre_impl_lchown16((long)(filename), (long)user, \
+ (long)group)
+#define __sanitizer_syscall_post_lchown16(res, filename, user, group) \
+ __sanitizer_syscall_post_impl_lchown16(res, (long)(filename), (long)user, \
+ (long)group)
+#define __sanitizer_syscall_pre_fchown16(fd, user, group) \
+ __sanitizer_syscall_pre_impl_fchown16((long)(fd), (long)user, (long)group)
+#define __sanitizer_syscall_post_fchown16(res, fd, user, group) \
+ __sanitizer_syscall_post_impl_fchown16(res, (long)(fd), (long)user, \
+ (long)group)
+#define __sanitizer_syscall_pre_setregid16(rgid, egid) \
+ __sanitizer_syscall_pre_impl_setregid16((long)rgid, (long)egid)
+#define __sanitizer_syscall_post_setregid16(res, rgid, egid) \
+ __sanitizer_syscall_post_impl_setregid16(res, (long)rgid, (long)egid)
+#define __sanitizer_syscall_pre_setgid16(gid) \
+ __sanitizer_syscall_pre_impl_setgid16((long)gid)
+#define __sanitizer_syscall_post_setgid16(res, gid) \
+ __sanitizer_syscall_post_impl_setgid16(res, (long)gid)
+#define __sanitizer_syscall_pre_setreuid16(ruid, euid) \
+ __sanitizer_syscall_pre_impl_setreuid16((long)ruid, (long)euid)
+#define __sanitizer_syscall_post_setreuid16(res, ruid, euid) \
+ __sanitizer_syscall_post_impl_setreuid16(res, (long)ruid, (long)euid)
+#define __sanitizer_syscall_pre_setuid16(uid) \
+ __sanitizer_syscall_pre_impl_setuid16((long)uid)
+#define __sanitizer_syscall_post_setuid16(res, uid) \
+ __sanitizer_syscall_post_impl_setuid16(res, (long)uid)
+#define __sanitizer_syscall_pre_setresuid16(ruid, euid, suid) \
+ __sanitizer_syscall_pre_impl_setresuid16((long)ruid, (long)euid, (long)suid)
+#define __sanitizer_syscall_post_setresuid16(res, ruid, euid, suid) \
+ __sanitizer_syscall_post_impl_setresuid16(res, (long)ruid, (long)euid, \
+ (long)suid)
+#define __sanitizer_syscall_pre_getresuid16(ruid, euid, suid) \
+ __sanitizer_syscall_pre_impl_getresuid16((long)(ruid), (long)(euid), \
+ (long)(suid))
+#define __sanitizer_syscall_post_getresuid16(res, ruid, euid, suid) \
+ __sanitizer_syscall_post_impl_getresuid16(res, (long)(ruid), (long)(euid), \
+ (long)(suid))
+#define __sanitizer_syscall_pre_setresgid16(rgid, egid, sgid) \
+ __sanitizer_syscall_pre_impl_setresgid16((long)rgid, (long)egid, (long)sgid)
+#define __sanitizer_syscall_post_setresgid16(res, rgid, egid, sgid) \
+ __sanitizer_syscall_post_impl_setresgid16(res, (long)rgid, (long)egid, \
+ (long)sgid)
+#define __sanitizer_syscall_pre_getresgid16(rgid, egid, sgid) \
+ __sanitizer_syscall_pre_impl_getresgid16((long)(rgid), (long)(egid), \
+ (long)(sgid))
+#define __sanitizer_syscall_post_getresgid16(res, rgid, egid, sgid) \
+ __sanitizer_syscall_post_impl_getresgid16(res, (long)(rgid), (long)(egid), \
+ (long)(sgid))
+#define __sanitizer_syscall_pre_setfsuid16(uid) \
+ __sanitizer_syscall_pre_impl_setfsuid16((long)uid)
+#define __sanitizer_syscall_post_setfsuid16(res, uid) \
+ __sanitizer_syscall_post_impl_setfsuid16(res, (long)uid)
+#define __sanitizer_syscall_pre_setfsgid16(gid) \
+ __sanitizer_syscall_pre_impl_setfsgid16((long)gid)
+#define __sanitizer_syscall_post_setfsgid16(res, gid) \
+ __sanitizer_syscall_post_impl_setfsgid16(res, (long)gid)
+#define __sanitizer_syscall_pre_getgroups16(gidsetsize, grouplist) \
+ __sanitizer_syscall_pre_impl_getgroups16((long)(gidsetsize), \
+ (long)(grouplist))
+#define __sanitizer_syscall_post_getgroups16(res, gidsetsize, grouplist) \
+ __sanitizer_syscall_post_impl_getgroups16(res, (long)(gidsetsize), \
+ (long)(grouplist))
+#define __sanitizer_syscall_pre_setgroups16(gidsetsize, grouplist) \
+ __sanitizer_syscall_pre_impl_setgroups16((long)(gidsetsize), \
+ (long)(grouplist))
+#define __sanitizer_syscall_post_setgroups16(res, gidsetsize, grouplist) \
+ __sanitizer_syscall_post_impl_setgroups16(res, (long)(gidsetsize), \
+ (long)(grouplist))
+#define __sanitizer_syscall_pre_getuid16() \
+ __sanitizer_syscall_pre_impl_getuid16()
+#define __sanitizer_syscall_post_getuid16(res) \
+ __sanitizer_syscall_post_impl_getuid16(res)
+#define __sanitizer_syscall_pre_geteuid16() \
+ __sanitizer_syscall_pre_impl_geteuid16()
+#define __sanitizer_syscall_post_geteuid16(res) \
+ __sanitizer_syscall_post_impl_geteuid16(res)
+#define __sanitizer_syscall_pre_getgid16() \
+ __sanitizer_syscall_pre_impl_getgid16()
+#define __sanitizer_syscall_post_getgid16(res) \
+ __sanitizer_syscall_post_impl_getgid16(res)
+#define __sanitizer_syscall_pre_getegid16() \
+ __sanitizer_syscall_pre_impl_getegid16()
+#define __sanitizer_syscall_post_getegid16(res) \
+ __sanitizer_syscall_post_impl_getegid16(res)
+#define __sanitizer_syscall_pre_utime(filename, times) \
+ __sanitizer_syscall_pre_impl_utime((long)(filename), (long)(times))
+#define __sanitizer_syscall_post_utime(res, filename, times) \
+ __sanitizer_syscall_post_impl_utime(res, (long)(filename), (long)(times))
+#define __sanitizer_syscall_pre_utimes(filename, utimes) \
+ __sanitizer_syscall_pre_impl_utimes((long)(filename), (long)(utimes))
+#define __sanitizer_syscall_post_utimes(res, filename, utimes) \
+ __sanitizer_syscall_post_impl_utimes(res, (long)(filename), (long)(utimes))
+#define __sanitizer_syscall_pre_lseek(fd, offset, origin) \
+ __sanitizer_syscall_pre_impl_lseek((long)(fd), (long)(offset), (long)(origin))
+#define __sanitizer_syscall_post_lseek(res, fd, offset, origin) \
+ __sanitizer_syscall_post_impl_lseek(res, (long)(fd), (long)(offset), \
+ (long)(origin))
+#define __sanitizer_syscall_pre_llseek(fd, offset_high, offset_low, result, \
+ origin) \
+ __sanitizer_syscall_pre_impl_llseek((long)(fd), (long)(offset_high), \
+ (long)(offset_low), (long)(result), \
+ (long)(origin))
+#define __sanitizer_syscall_post_llseek(res, fd, offset_high, offset_low, \
+ result, origin) \
+ __sanitizer_syscall_post_impl_llseek(res, (long)(fd), (long)(offset_high), \
+ (long)(offset_low), (long)(result), \
+ (long)(origin))
+#define __sanitizer_syscall_pre_read(fd, buf, count) \
+ __sanitizer_syscall_pre_impl_read((long)(fd), (long)(buf), (long)(count))
+#define __sanitizer_syscall_post_read(res, fd, buf, count) \
+ __sanitizer_syscall_post_impl_read(res, (long)(fd), (long)(buf), \
+ (long)(count))
+#define __sanitizer_syscall_pre_readv(fd, vec, vlen) \
+ __sanitizer_syscall_pre_impl_readv((long)(fd), (long)(vec), (long)(vlen))
+#define __sanitizer_syscall_post_readv(res, fd, vec, vlen) \
+ __sanitizer_syscall_post_impl_readv(res, (long)(fd), (long)(vec), \
+ (long)(vlen))
+#define __sanitizer_syscall_pre_write(fd, buf, count) \
+ __sanitizer_syscall_pre_impl_write((long)(fd), (long)(buf), (long)(count))
+#define __sanitizer_syscall_post_write(res, fd, buf, count) \
+ __sanitizer_syscall_post_impl_write(res, (long)(fd), (long)(buf), \
+ (long)(count))
+#define __sanitizer_syscall_pre_writev(fd, vec, vlen) \
+ __sanitizer_syscall_pre_impl_writev((long)(fd), (long)(vec), (long)(vlen))
+#define __sanitizer_syscall_post_writev(res, fd, vec, vlen) \
+ __sanitizer_syscall_post_impl_writev(res, (long)(fd), (long)(vec), \
+ (long)(vlen))
+
+#ifdef _LP64
+#define __sanitizer_syscall_pre_pread64(fd, buf, count, pos) \
+ __sanitizer_syscall_pre_impl_pread64((long)(fd), (long)(buf), (long)(count), \
+ (long)(pos))
+#define __sanitizer_syscall_post_pread64(res, fd, buf, count, pos) \
+ __sanitizer_syscall_post_impl_pread64(res, (long)(fd), (long)(buf), \
+ (long)(count), (long)(pos))
+#define __sanitizer_syscall_pre_pwrite64(fd, buf, count, pos) \
+ __sanitizer_syscall_pre_impl_pwrite64((long)(fd), (long)(buf), \
+ (long)(count), (long)(pos))
+#define __sanitizer_syscall_post_pwrite64(res, fd, buf, count, pos) \
+ __sanitizer_syscall_post_impl_pwrite64(res, (long)(fd), (long)(buf), \
+ (long)(count), (long)(pos))
+#else
+#define __sanitizer_syscall_pre_pread64(fd, buf, count, pos0, pos1) \
+ __sanitizer_syscall_pre_impl_pread64((long)(fd), (long)(buf), (long)(count), \
+ (long)(pos0), (long)(pos1))
+#define __sanitizer_syscall_post_pread64(res, fd, buf, count, pos0, pos1) \
+ __sanitizer_syscall_post_impl_pread64(res, (long)(fd), (long)(buf), \
+ (long)(count), (long)(pos0), \
+ (long)(pos1))
+#define __sanitizer_syscall_pre_pwrite64(fd, buf, count, pos0, pos1) \
+ __sanitizer_syscall_pre_impl_pwrite64( \
+ (long)(fd), (long)(buf), (long)(count), (long)(pos0), (long)(pos1))
+#define __sanitizer_syscall_post_pwrite64(res, fd, buf, count, pos0, pos1) \
+ __sanitizer_syscall_post_impl_pwrite64( \
+ res, (long)(fd), (long)(buf), (long)(count), (long)(pos0), (long)(pos1))
+#endif
+
+#define __sanitizer_syscall_pre_preadv(fd, vec, vlen, pos_l, pos_h) \
+ __sanitizer_syscall_pre_impl_preadv((long)(fd), (long)(vec), (long)(vlen), \
+ (long)(pos_l), (long)(pos_h))
+#define __sanitizer_syscall_post_preadv(res, fd, vec, vlen, pos_l, pos_h) \
+ __sanitizer_syscall_post_impl_preadv(res, (long)(fd), (long)(vec), \
+ (long)(vlen), (long)(pos_l), \
+ (long)(pos_h))
+#define __sanitizer_syscall_pre_pwritev(fd, vec, vlen, pos_l, pos_h) \
+ __sanitizer_syscall_pre_impl_pwritev((long)(fd), (long)(vec), (long)(vlen), \
+ (long)(pos_l), (long)(pos_h))
+#define __sanitizer_syscall_post_pwritev(res, fd, vec, vlen, pos_l, pos_h) \
+ __sanitizer_syscall_post_impl_pwritev(res, (long)(fd), (long)(vec), \
+ (long)(vlen), (long)(pos_l), \
+ (long)(pos_h))
+#define __sanitizer_syscall_pre_getcwd(buf, size) \
+ __sanitizer_syscall_pre_impl_getcwd((long)(buf), (long)(size))
+#define __sanitizer_syscall_post_getcwd(res, buf, size) \
+ __sanitizer_syscall_post_impl_getcwd(res, (long)(buf), (long)(size))
+#define __sanitizer_syscall_pre_mkdir(pathname, mode) \
+ __sanitizer_syscall_pre_impl_mkdir((long)(pathname), (long)(mode))
+#define __sanitizer_syscall_post_mkdir(res, pathname, mode) \
+ __sanitizer_syscall_post_impl_mkdir(res, (long)(pathname), (long)(mode))
+#define __sanitizer_syscall_pre_chdir(filename) \
+ __sanitizer_syscall_pre_impl_chdir((long)(filename))
+#define __sanitizer_syscall_post_chdir(res, filename) \
+ __sanitizer_syscall_post_impl_chdir(res, (long)(filename))
+#define __sanitizer_syscall_pre_fchdir(fd) \
+ __sanitizer_syscall_pre_impl_fchdir((long)(fd))
+#define __sanitizer_syscall_post_fchdir(res, fd) \
+ __sanitizer_syscall_post_impl_fchdir(res, (long)(fd))
+#define __sanitizer_syscall_pre_rmdir(pathname) \
+ __sanitizer_syscall_pre_impl_rmdir((long)(pathname))
+#define __sanitizer_syscall_post_rmdir(res, pathname) \
+ __sanitizer_syscall_post_impl_rmdir(res, (long)(pathname))
+#define __sanitizer_syscall_pre_lookup_dcookie(cookie64, buf, len) \
+ __sanitizer_syscall_pre_impl_lookup_dcookie((long)(cookie64), (long)(buf), \
+ (long)(len))
+#define __sanitizer_syscall_post_lookup_dcookie(res, cookie64, buf, len) \
+ __sanitizer_syscall_post_impl_lookup_dcookie(res, (long)(cookie64), \
+ (long)(buf), (long)(len))
+#define __sanitizer_syscall_pre_quotactl(cmd, special, id, addr) \
+ __sanitizer_syscall_pre_impl_quotactl((long)(cmd), (long)(special), \
+ (long)(id), (long)(addr))
+#define __sanitizer_syscall_post_quotactl(res, cmd, special, id, addr) \
+ __sanitizer_syscall_post_impl_quotactl(res, (long)(cmd), (long)(special), \
+ (long)(id), (long)(addr))
+#define __sanitizer_syscall_pre_getdents(fd, dirent, count) \
+ __sanitizer_syscall_pre_impl_getdents((long)(fd), (long)(dirent), \
+ (long)(count))
+#define __sanitizer_syscall_post_getdents(res, fd, dirent, count) \
+ __sanitizer_syscall_post_impl_getdents(res, (long)(fd), (long)(dirent), \
+ (long)(count))
+#define __sanitizer_syscall_pre_getdents64(fd, dirent, count) \
+ __sanitizer_syscall_pre_impl_getdents64((long)(fd), (long)(dirent), \
+ (long)(count))
+#define __sanitizer_syscall_post_getdents64(res, fd, dirent, count) \
+ __sanitizer_syscall_post_impl_getdents64(res, (long)(fd), (long)(dirent), \
+ (long)(count))
+#define __sanitizer_syscall_pre_setsockopt(fd, level, optname, optval, optlen) \
+ __sanitizer_syscall_pre_impl_setsockopt((long)(fd), (long)(level), \
+ (long)(optname), (long)(optval), \
+ (long)(optlen))
+#define __sanitizer_syscall_post_setsockopt(res, fd, level, optname, optval, \
+ optlen) \
+ __sanitizer_syscall_post_impl_setsockopt(res, (long)(fd), (long)(level), \
+ (long)(optname), (long)(optval), \
+ (long)(optlen))
+#define __sanitizer_syscall_pre_getsockopt(fd, level, optname, optval, optlen) \
+ __sanitizer_syscall_pre_impl_getsockopt((long)(fd), (long)(level), \
+ (long)(optname), (long)(optval), \
+ (long)(optlen))
+#define __sanitizer_syscall_post_getsockopt(res, fd, level, optname, optval, \
+ optlen) \
+ __sanitizer_syscall_post_impl_getsockopt(res, (long)(fd), (long)(level), \
+ (long)(optname), (long)(optval), \
+ (long)(optlen))
+#define __sanitizer_syscall_pre_bind(arg0, arg1, arg2) \
+ __sanitizer_syscall_pre_impl_bind((long)(arg0), (long)(arg1), (long)(arg2))
+#define __sanitizer_syscall_post_bind(res, arg0, arg1, arg2) \
+ __sanitizer_syscall_post_impl_bind(res, (long)(arg0), (long)(arg1), \
+ (long)(arg2))
+#define __sanitizer_syscall_pre_connect(arg0, arg1, arg2) \
+ __sanitizer_syscall_pre_impl_connect((long)(arg0), (long)(arg1), (long)(arg2))
+#define __sanitizer_syscall_post_connect(res, arg0, arg1, arg2) \
+ __sanitizer_syscall_post_impl_connect(res, (long)(arg0), (long)(arg1), \
+ (long)(arg2))
+#define __sanitizer_syscall_pre_accept(arg0, arg1, arg2) \
+ __sanitizer_syscall_pre_impl_accept((long)(arg0), (long)(arg1), (long)(arg2))
+#define __sanitizer_syscall_post_accept(res, arg0, arg1, arg2) \
+ __sanitizer_syscall_post_impl_accept(res, (long)(arg0), (long)(arg1), \
+ (long)(arg2))
+#define __sanitizer_syscall_pre_accept4(arg0, arg1, arg2, arg3) \
+ __sanitizer_syscall_pre_impl_accept4((long)(arg0), (long)(arg1), \
+ (long)(arg2), (long)(arg3))
+#define __sanitizer_syscall_post_accept4(res, arg0, arg1, arg2, arg3) \
+ __sanitizer_syscall_post_impl_accept4(res, (long)(arg0), (long)(arg1), \
+ (long)(arg2), (long)(arg3))
+#define __sanitizer_syscall_pre_getsockname(arg0, arg1, arg2) \
+ __sanitizer_syscall_pre_impl_getsockname((long)(arg0), (long)(arg1), \
+ (long)(arg2))
+#define __sanitizer_syscall_post_getsockname(res, arg0, arg1, arg2) \
+ __sanitizer_syscall_post_impl_getsockname(res, (long)(arg0), (long)(arg1), \
+ (long)(arg2))
+#define __sanitizer_syscall_pre_getpeername(arg0, arg1, arg2) \
+ __sanitizer_syscall_pre_impl_getpeername((long)(arg0), (long)(arg1), \
+ (long)(arg2))
+#define __sanitizer_syscall_post_getpeername(res, arg0, arg1, arg2) \
+ __sanitizer_syscall_post_impl_getpeername(res, (long)(arg0), (long)(arg1), \
+ (long)(arg2))
+#define __sanitizer_syscall_pre_send(arg0, arg1, arg2, arg3) \
+ __sanitizer_syscall_pre_impl_send((long)(arg0), (long)(arg1), (long)(arg2), \
+ (long)(arg3))
+#define __sanitizer_syscall_post_send(res, arg0, arg1, arg2, arg3) \
+ __sanitizer_syscall_post_impl_send(res, (long)(arg0), (long)(arg1), \
+ (long)(arg2), (long)(arg3))
+#define __sanitizer_syscall_pre_sendto(arg0, arg1, arg2, arg3, arg4, arg5) \
+ __sanitizer_syscall_pre_impl_sendto((long)(arg0), (long)(arg1), \
+ (long)(arg2), (long)(arg3), \
+ (long)(arg4), (long)(arg5))
+#define __sanitizer_syscall_post_sendto(res, arg0, arg1, arg2, arg3, arg4, \
+ arg5) \
+ __sanitizer_syscall_post_impl_sendto(res, (long)(arg0), (long)(arg1), \
+ (long)(arg2), (long)(arg3), \
+ (long)(arg4), (long)(arg5))
+#define __sanitizer_syscall_pre_sendmsg(fd, msg, flags) \
+ __sanitizer_syscall_pre_impl_sendmsg((long)(fd), (long)(msg), (long)(flags))
+#define __sanitizer_syscall_post_sendmsg(res, fd, msg, flags) \
+ __sanitizer_syscall_post_impl_sendmsg(res, (long)(fd), (long)(msg), \
+ (long)(flags))
+#define __sanitizer_syscall_pre_sendmmsg(fd, msg, vlen, flags) \
+ __sanitizer_syscall_pre_impl_sendmmsg((long)(fd), (long)(msg), (long)(vlen), \
+ (long)(flags))
+#define __sanitizer_syscall_post_sendmmsg(res, fd, msg, vlen, flags) \
+ __sanitizer_syscall_post_impl_sendmmsg(res, (long)(fd), (long)(msg), \
+ (long)(vlen), (long)(flags))
+#define __sanitizer_syscall_pre_recv(arg0, arg1, arg2, arg3) \
+ __sanitizer_syscall_pre_impl_recv((long)(arg0), (long)(arg1), (long)(arg2), \
+ (long)(arg3))
+#define __sanitizer_syscall_post_recv(res, arg0, arg1, arg2, arg3) \
+ __sanitizer_syscall_post_impl_recv(res, (long)(arg0), (long)(arg1), \
+ (long)(arg2), (long)(arg3))
+#define __sanitizer_syscall_pre_recvfrom(arg0, arg1, arg2, arg3, arg4, arg5) \
+ __sanitizer_syscall_pre_impl_recvfrom((long)(arg0), (long)(arg1), \
+ (long)(arg2), (long)(arg3), \
+ (long)(arg4), (long)(arg5))
+#define __sanitizer_syscall_post_recvfrom(res, arg0, arg1, arg2, arg3, arg4, \
+ arg5) \
+ __sanitizer_syscall_post_impl_recvfrom(res, (long)(arg0), (long)(arg1), \
+ (long)(arg2), (long)(arg3), \
+ (long)(arg4), (long)(arg5))
+#define __sanitizer_syscall_pre_recvmsg(fd, msg, flags) \
+ __sanitizer_syscall_pre_impl_recvmsg((long)(fd), (long)(msg), (long)(flags))
+#define __sanitizer_syscall_post_recvmsg(res, fd, msg, flags) \
+ __sanitizer_syscall_post_impl_recvmsg(res, (long)(fd), (long)(msg), \
+ (long)(flags))
+#define __sanitizer_syscall_pre_recvmmsg(fd, msg, vlen, flags, timeout) \
+ __sanitizer_syscall_pre_impl_recvmmsg((long)(fd), (long)(msg), (long)(vlen), \
+ (long)(flags), (long)(timeout))
+#define __sanitizer_syscall_post_recvmmsg(res, fd, msg, vlen, flags, timeout) \
+ __sanitizer_syscall_post_impl_recvmmsg(res, (long)(fd), (long)(msg), \
+ (long)(vlen), (long)(flags), \
+ (long)(timeout))
+#define __sanitizer_syscall_pre_socket(arg0, arg1, arg2) \
+ __sanitizer_syscall_pre_impl_socket((long)(arg0), (long)(arg1), (long)(arg2))
+#define __sanitizer_syscall_post_socket(res, arg0, arg1, arg2) \
+ __sanitizer_syscall_post_impl_socket(res, (long)(arg0), (long)(arg1), \
+ (long)(arg2))
+#define __sanitizer_syscall_pre_socketpair(arg0, arg1, arg2, arg3) \
+ __sanitizer_syscall_pre_impl_socketpair((long)(arg0), (long)(arg1), \
+ (long)(arg2), (long)(arg3))
+#define __sanitizer_syscall_post_socketpair(res, arg0, arg1, arg2, arg3) \
+ __sanitizer_syscall_post_impl_socketpair(res, (long)(arg0), (long)(arg1), \
+ (long)(arg2), (long)(arg3))
+#define __sanitizer_syscall_pre_socketcall(call, args) \
+ __sanitizer_syscall_pre_impl_socketcall((long)(call), (long)(args))
+#define __sanitizer_syscall_post_socketcall(res, call, args) \
+ __sanitizer_syscall_post_impl_socketcall(res, (long)(call), (long)(args))
+#define __sanitizer_syscall_pre_listen(arg0, arg1) \
+ __sanitizer_syscall_pre_impl_listen((long)(arg0), (long)(arg1))
+#define __sanitizer_syscall_post_listen(res, arg0, arg1) \
+ __sanitizer_syscall_post_impl_listen(res, (long)(arg0), (long)(arg1))
+#define __sanitizer_syscall_pre_poll(ufds, nfds, timeout) \
+ __sanitizer_syscall_pre_impl_poll((long)(ufds), (long)(nfds), (long)(timeout))
+#define __sanitizer_syscall_post_poll(res, ufds, nfds, timeout) \
+ __sanitizer_syscall_post_impl_poll(res, (long)(ufds), (long)(nfds), \
+ (long)(timeout))
+#define __sanitizer_syscall_pre_select(n, inp, outp, exp, tvp) \
+ __sanitizer_syscall_pre_impl_select((long)(n), (long)(inp), (long)(outp), \
+ (long)(exp), (long)(tvp))
+#define __sanitizer_syscall_post_select(res, n, inp, outp, exp, tvp) \
+ __sanitizer_syscall_post_impl_select(res, (long)(n), (long)(inp), \
+ (long)(outp), (long)(exp), (long)(tvp))
+#define __sanitizer_syscall_pre_old_select(arg) \
+ __sanitizer_syscall_pre_impl_old_select((long)(arg))
+#define __sanitizer_syscall_post_old_select(res, arg) \
+ __sanitizer_syscall_post_impl_old_select(res, (long)(arg))
+#define __sanitizer_syscall_pre_epoll_create(size) \
+ __sanitizer_syscall_pre_impl_epoll_create((long)(size))
+#define __sanitizer_syscall_post_epoll_create(res, size) \
+ __sanitizer_syscall_post_impl_epoll_create(res, (long)(size))
+#define __sanitizer_syscall_pre_epoll_create1(flags) \
+ __sanitizer_syscall_pre_impl_epoll_create1((long)(flags))
+#define __sanitizer_syscall_post_epoll_create1(res, flags) \
+ __sanitizer_syscall_post_impl_epoll_create1(res, (long)(flags))
+#define __sanitizer_syscall_pre_epoll_ctl(epfd, op, fd, event) \
+ __sanitizer_syscall_pre_impl_epoll_ctl((long)(epfd), (long)(op), (long)(fd), \
+ (long)(event))
+#define __sanitizer_syscall_post_epoll_ctl(res, epfd, op, fd, event) \
+ __sanitizer_syscall_post_impl_epoll_ctl(res, (long)(epfd), (long)(op), \
+ (long)(fd), (long)(event))
+#define __sanitizer_syscall_pre_epoll_wait(epfd, events, maxevents, timeout) \
+ __sanitizer_syscall_pre_impl_epoll_wait((long)(epfd), (long)(events), \
+ (long)(maxevents), (long)(timeout))
+#define __sanitizer_syscall_post_epoll_wait(res, epfd, events, maxevents, \
+ timeout) \
+ __sanitizer_syscall_post_impl_epoll_wait(res, (long)(epfd), (long)(events), \
+ (long)(maxevents), (long)(timeout))
+#define __sanitizer_syscall_pre_epoll_pwait(epfd, events, maxevents, timeout, \
+ sigmask, sigsetsize) \
+ __sanitizer_syscall_pre_impl_epoll_pwait( \
+ (long)(epfd), (long)(events), (long)(maxevents), (long)(timeout), \
+ (long)(sigmask), (long)(sigsetsize))
+#define __sanitizer_syscall_post_epoll_pwait(res, epfd, events, maxevents, \
+ timeout, sigmask, sigsetsize) \
+ __sanitizer_syscall_post_impl_epoll_pwait( \
+ res, (long)(epfd), (long)(events), (long)(maxevents), (long)(timeout), \
+ (long)(sigmask), (long)(sigsetsize))
+#define __sanitizer_syscall_pre_gethostname(name, len) \
+ __sanitizer_syscall_pre_impl_gethostname((long)(name), (long)(len))
+#define __sanitizer_syscall_post_gethostname(res, name, len) \
+ __sanitizer_syscall_post_impl_gethostname(res, (long)(name), (long)(len))
+#define __sanitizer_syscall_pre_sethostname(name, len) \
+ __sanitizer_syscall_pre_impl_sethostname((long)(name), (long)(len))
+#define __sanitizer_syscall_post_sethostname(res, name, len) \
+ __sanitizer_syscall_post_impl_sethostname(res, (long)(name), (long)(len))
+#define __sanitizer_syscall_pre_setdomainname(name, len) \
+ __sanitizer_syscall_pre_impl_setdomainname((long)(name), (long)(len))
+#define __sanitizer_syscall_post_setdomainname(res, name, len) \
+ __sanitizer_syscall_post_impl_setdomainname(res, (long)(name), (long)(len))
+#define __sanitizer_syscall_pre_newuname(name) \
+ __sanitizer_syscall_pre_impl_newuname((long)(name))
+#define __sanitizer_syscall_post_newuname(res, name) \
+ __sanitizer_syscall_post_impl_newuname(res, (long)(name))
+#define __sanitizer_syscall_pre_uname(arg0) \
+ __sanitizer_syscall_pre_impl_uname((long)(arg0))
+#define __sanitizer_syscall_post_uname(res, arg0) \
+ __sanitizer_syscall_post_impl_uname(res, (long)(arg0))
+#define __sanitizer_syscall_pre_olduname(arg0) \
+ __sanitizer_syscall_pre_impl_olduname((long)(arg0))
+#define __sanitizer_syscall_post_olduname(res, arg0) \
+ __sanitizer_syscall_post_impl_olduname(res, (long)(arg0))
+#define __sanitizer_syscall_pre_getrlimit(resource, rlim) \
+ __sanitizer_syscall_pre_impl_getrlimit((long)(resource), (long)(rlim))
+#define __sanitizer_syscall_post_getrlimit(res, resource, rlim) \
+ __sanitizer_syscall_post_impl_getrlimit(res, (long)(resource), (long)(rlim))
+#define __sanitizer_syscall_pre_old_getrlimit(resource, rlim) \
+ __sanitizer_syscall_pre_impl_old_getrlimit((long)(resource), (long)(rlim))
+#define __sanitizer_syscall_post_old_getrlimit(res, resource, rlim) \
+ __sanitizer_syscall_post_impl_old_getrlimit(res, (long)(resource), \
+ (long)(rlim))
+#define __sanitizer_syscall_pre_setrlimit(resource, rlim) \
+ __sanitizer_syscall_pre_impl_setrlimit((long)(resource), (long)(rlim))
+#define __sanitizer_syscall_post_setrlimit(res, resource, rlim) \
+ __sanitizer_syscall_post_impl_setrlimit(res, (long)(resource), (long)(rlim))
+#define __sanitizer_syscall_pre_prlimit64(pid, resource, new_rlim, old_rlim) \
+ __sanitizer_syscall_pre_impl_prlimit64((long)(pid), (long)(resource), \
+ (long)(new_rlim), (long)(old_rlim))
+#define __sanitizer_syscall_post_prlimit64(res, pid, resource, new_rlim, \
+ old_rlim) \
+ __sanitizer_syscall_post_impl_prlimit64(res, (long)(pid), (long)(resource), \
+ (long)(new_rlim), (long)(old_rlim))
+#define __sanitizer_syscall_pre_getrusage(who, ru) \
+ __sanitizer_syscall_pre_impl_getrusage((long)(who), (long)(ru))
+#define __sanitizer_syscall_post_getrusage(res, who, ru) \
+ __sanitizer_syscall_post_impl_getrusage(res, (long)(who), (long)(ru))
+#define __sanitizer_syscall_pre_umask(mask) \
+ __sanitizer_syscall_pre_impl_umask((long)(mask))
+#define __sanitizer_syscall_post_umask(res, mask) \
+ __sanitizer_syscall_post_impl_umask(res, (long)(mask))
+#define __sanitizer_syscall_pre_msgget(key, msgflg) \
+ __sanitizer_syscall_pre_impl_msgget((long)(key), (long)(msgflg))
+#define __sanitizer_syscall_post_msgget(res, key, msgflg) \
+ __sanitizer_syscall_post_impl_msgget(res, (long)(key), (long)(msgflg))
+#define __sanitizer_syscall_pre_msgsnd(msqid, msgp, msgsz, msgflg) \
+ __sanitizer_syscall_pre_impl_msgsnd((long)(msqid), (long)(msgp), \
+ (long)(msgsz), (long)(msgflg))
+#define __sanitizer_syscall_post_msgsnd(res, msqid, msgp, msgsz, msgflg) \
+ __sanitizer_syscall_post_impl_msgsnd(res, (long)(msqid), (long)(msgp), \
+ (long)(msgsz), (long)(msgflg))
+#define __sanitizer_syscall_pre_msgrcv(msqid, msgp, msgsz, msgtyp, msgflg) \
+ __sanitizer_syscall_pre_impl_msgrcv((long)(msqid), (long)(msgp), \
+ (long)(msgsz), (long)(msgtyp), \
+ (long)(msgflg))
+#define __sanitizer_syscall_post_msgrcv(res, msqid, msgp, msgsz, msgtyp, \
+ msgflg) \
+ __sanitizer_syscall_post_impl_msgrcv(res, (long)(msqid), (long)(msgp), \
+ (long)(msgsz), (long)(msgtyp), \
+ (long)(msgflg))
+#define __sanitizer_syscall_pre_msgctl(msqid, cmd, buf) \
+ __sanitizer_syscall_pre_impl_msgctl((long)(msqid), (long)(cmd), (long)(buf))
+#define __sanitizer_syscall_post_msgctl(res, msqid, cmd, buf) \
+ __sanitizer_syscall_post_impl_msgctl(res, (long)(msqid), (long)(cmd), \
+ (long)(buf))
+#define __sanitizer_syscall_pre_semget(key, nsems, semflg) \
+ __sanitizer_syscall_pre_impl_semget((long)(key), (long)(nsems), \
+ (long)(semflg))
+#define __sanitizer_syscall_post_semget(res, key, nsems, semflg) \
+ __sanitizer_syscall_post_impl_semget(res, (long)(key), (long)(nsems), \
+ (long)(semflg))
+#define __sanitizer_syscall_pre_semop(semid, sops, nsops) \
+ __sanitizer_syscall_pre_impl_semop((long)(semid), (long)(sops), (long)(nsops))
+#define __sanitizer_syscall_post_semop(res, semid, sops, nsops) \
+ __sanitizer_syscall_post_impl_semop(res, (long)(semid), (long)(sops), \
+ (long)(nsops))
+#define __sanitizer_syscall_pre_semctl(semid, semnum, cmd, arg) \
+ __sanitizer_syscall_pre_impl_semctl((long)(semid), (long)(semnum), \
+ (long)(cmd), (long)(arg))
+#define __sanitizer_syscall_post_semctl(res, semid, semnum, cmd, arg) \
+ __sanitizer_syscall_post_impl_semctl(res, (long)(semid), (long)(semnum), \
+ (long)(cmd), (long)(arg))
+#define __sanitizer_syscall_pre_semtimedop(semid, sops, nsops, timeout) \
+ __sanitizer_syscall_pre_impl_semtimedop((long)(semid), (long)(sops), \
+ (long)(nsops), (long)(timeout))
+#define __sanitizer_syscall_post_semtimedop(res, semid, sops, nsops, timeout) \
+ __sanitizer_syscall_post_impl_semtimedop(res, (long)(semid), (long)(sops), \
+ (long)(nsops), (long)(timeout))
+#define __sanitizer_syscall_pre_shmat(shmid, shmaddr, shmflg) \
+ __sanitizer_syscall_pre_impl_shmat((long)(shmid), (long)(shmaddr), \
+ (long)(shmflg))
+#define __sanitizer_syscall_post_shmat(res, shmid, shmaddr, shmflg) \
+ __sanitizer_syscall_post_impl_shmat(res, (long)(shmid), (long)(shmaddr), \
+ (long)(shmflg))
+#define __sanitizer_syscall_pre_shmget(key, size, flag) \
+ __sanitizer_syscall_pre_impl_shmget((long)(key), (long)(size), (long)(flag))
+#define __sanitizer_syscall_post_shmget(res, key, size, flag) \
+ __sanitizer_syscall_post_impl_shmget(res, (long)(key), (long)(size), \
+ (long)(flag))
+#define __sanitizer_syscall_pre_shmdt(shmaddr) \
+ __sanitizer_syscall_pre_impl_shmdt((long)(shmaddr))
+#define __sanitizer_syscall_post_shmdt(res, shmaddr) \
+ __sanitizer_syscall_post_impl_shmdt(res, (long)(shmaddr))
+#define __sanitizer_syscall_pre_shmctl(shmid, cmd, buf) \
+ __sanitizer_syscall_pre_impl_shmctl((long)(shmid), (long)(cmd), (long)(buf))
+#define __sanitizer_syscall_post_shmctl(res, shmid, cmd, buf) \
+ __sanitizer_syscall_post_impl_shmctl(res, (long)(shmid), (long)(cmd), \
+ (long)(buf))
+#define __sanitizer_syscall_pre_ipc(call, first, second, third, ptr, fifth) \
+ __sanitizer_syscall_pre_impl_ipc((long)(call), (long)(first), \
+ (long)(second), (long)(third), (long)(ptr), \
+ (long)(fifth))
+#define __sanitizer_syscall_post_ipc(res, call, first, second, third, ptr, \
+ fifth) \
+ __sanitizer_syscall_post_impl_ipc(res, (long)(call), (long)(first), \
+ (long)(second), (long)(third), \
+ (long)(ptr), (long)(fifth))
+#define __sanitizer_syscall_pre_mq_open(name, oflag, mode, attr) \
+ __sanitizer_syscall_pre_impl_mq_open((long)(name), (long)(oflag), \
+ (long)(mode), (long)(attr))
+#define __sanitizer_syscall_post_mq_open(res, name, oflag, mode, attr) \
+ __sanitizer_syscall_post_impl_mq_open(res, (long)(name), (long)(oflag), \
+ (long)(mode), (long)(attr))
+#define __sanitizer_syscall_pre_mq_unlink(name) \
+ __sanitizer_syscall_pre_impl_mq_unlink((long)(name))
+#define __sanitizer_syscall_post_mq_unlink(res, name) \
+ __sanitizer_syscall_post_impl_mq_unlink(res, (long)(name))
+#define __sanitizer_syscall_pre_mq_timedsend(mqdes, msg_ptr, msg_len, \
+ msg_prio, abs_timeout) \
+ __sanitizer_syscall_pre_impl_mq_timedsend((long)(mqdes), (long)(msg_ptr), \
+ (long)(msg_len), (long)(msg_prio), \
+ (long)(abs_timeout))
+#define __sanitizer_syscall_post_mq_timedsend(res, mqdes, msg_ptr, msg_len, \
+ msg_prio, abs_timeout) \
+ __sanitizer_syscall_post_impl_mq_timedsend( \
+ res, (long)(mqdes), (long)(msg_ptr), (long)(msg_len), (long)(msg_prio), \
+ (long)(abs_timeout))
+#define __sanitizer_syscall_pre_mq_timedreceive(mqdes, msg_ptr, msg_len, \
+ msg_prio, abs_timeout) \
+ __sanitizer_syscall_pre_impl_mq_timedreceive( \
+ (long)(mqdes), (long)(msg_ptr), (long)(msg_len), (long)(msg_prio), \
+ (long)(abs_timeout))
+#define __sanitizer_syscall_post_mq_timedreceive(res, mqdes, msg_ptr, msg_len, \
+ msg_prio, abs_timeout) \
+ __sanitizer_syscall_post_impl_mq_timedreceive( \
+ res, (long)(mqdes), (long)(msg_ptr), (long)(msg_len), (long)(msg_prio), \
+ (long)(abs_timeout))
+#define __sanitizer_syscall_pre_mq_notify(mqdes, notification) \
+ __sanitizer_syscall_pre_impl_mq_notify((long)(mqdes), (long)(notification))
+#define __sanitizer_syscall_post_mq_notify(res, mqdes, notification) \
+ __sanitizer_syscall_post_impl_mq_notify(res, (long)(mqdes), \
+ (long)(notification))
+#define __sanitizer_syscall_pre_mq_getsetattr(mqdes, mqstat, omqstat) \
+ __sanitizer_syscall_pre_impl_mq_getsetattr((long)(mqdes), (long)(mqstat), \
+ (long)(omqstat))
+#define __sanitizer_syscall_post_mq_getsetattr(res, mqdes, mqstat, omqstat) \
+ __sanitizer_syscall_post_impl_mq_getsetattr(res, (long)(mqdes), \
+ (long)(mqstat), (long)(omqstat))
+#define __sanitizer_syscall_pre_pciconfig_iobase(which, bus, devfn) \
+ __sanitizer_syscall_pre_impl_pciconfig_iobase((long)(which), (long)(bus), \
+ (long)(devfn))
+#define __sanitizer_syscall_post_pciconfig_iobase(res, which, bus, devfn) \
+ __sanitizer_syscall_post_impl_pciconfig_iobase(res, (long)(which), \
+ (long)(bus), (long)(devfn))
+#define __sanitizer_syscall_pre_pciconfig_read(bus, dfn, off, len, buf) \
+ __sanitizer_syscall_pre_impl_pciconfig_read( \
+ (long)(bus), (long)(dfn), (long)(off), (long)(len), (long)(buf))
+#define __sanitizer_syscall_post_pciconfig_read(res, bus, dfn, off, len, buf) \
+ __sanitizer_syscall_post_impl_pciconfig_read( \
+ res, (long)(bus), (long)(dfn), (long)(off), (long)(len), (long)(buf))
+#define __sanitizer_syscall_pre_pciconfig_write(bus, dfn, off, len, buf) \
+ __sanitizer_syscall_pre_impl_pciconfig_write( \
+ (long)(bus), (long)(dfn), (long)(off), (long)(len), (long)(buf))
+#define __sanitizer_syscall_post_pciconfig_write(res, bus, dfn, off, len, buf) \
+ __sanitizer_syscall_post_impl_pciconfig_write( \
+ res, (long)(bus), (long)(dfn), (long)(off), (long)(len), (long)(buf))
+#define __sanitizer_syscall_pre_swapon(specialfile, swap_flags) \
+ __sanitizer_syscall_pre_impl_swapon((long)(specialfile), (long)(swap_flags))
+#define __sanitizer_syscall_post_swapon(res, specialfile, swap_flags) \
+ __sanitizer_syscall_post_impl_swapon(res, (long)(specialfile), \
+ (long)(swap_flags))
+#define __sanitizer_syscall_pre_swapoff(specialfile) \
+ __sanitizer_syscall_pre_impl_swapoff((long)(specialfile))
+#define __sanitizer_syscall_post_swapoff(res, specialfile) \
+ __sanitizer_syscall_post_impl_swapoff(res, (long)(specialfile))
+#define __sanitizer_syscall_pre_sysctl(args) \
+ __sanitizer_syscall_pre_impl_sysctl((long)(args))
+#define __sanitizer_syscall_post_sysctl(res, args) \
+ __sanitizer_syscall_post_impl_sysctl(res, (long)(args))
+#define __sanitizer_syscall_pre_sysinfo(info) \
+ __sanitizer_syscall_pre_impl_sysinfo((long)(info))
+#define __sanitizer_syscall_post_sysinfo(res, info) \
+ __sanitizer_syscall_post_impl_sysinfo(res, (long)(info))
+#define __sanitizer_syscall_pre_sysfs(option, arg1, arg2) \
+ __sanitizer_syscall_pre_impl_sysfs((long)(option), (long)(arg1), (long)(arg2))
+#define __sanitizer_syscall_post_sysfs(res, option, arg1, arg2) \
+ __sanitizer_syscall_post_impl_sysfs(res, (long)(option), (long)(arg1), \
+ (long)(arg2))
+#define __sanitizer_syscall_pre_syslog(type, buf, len) \
+ __sanitizer_syscall_pre_impl_syslog((long)(type), (long)(buf), (long)(len))
+#define __sanitizer_syscall_post_syslog(res, type, buf, len) \
+ __sanitizer_syscall_post_impl_syslog(res, (long)(type), (long)(buf), \
+ (long)(len))
+#define __sanitizer_syscall_pre_uselib(library) \
+ __sanitizer_syscall_pre_impl_uselib((long)(library))
+#define __sanitizer_syscall_post_uselib(res, library) \
+ __sanitizer_syscall_post_impl_uselib(res, (long)(library))
+#define __sanitizer_syscall_pre_ni_syscall() \
+ __sanitizer_syscall_pre_impl_ni_syscall()
+#define __sanitizer_syscall_post_ni_syscall(res) \
+ __sanitizer_syscall_post_impl_ni_syscall(res)
+#define __sanitizer_syscall_pre_ptrace(request, pid, addr, data) \
+ __sanitizer_syscall_pre_impl_ptrace((long)(request), (long)(pid), \
+ (long)(addr), (long)(data))
+#define __sanitizer_syscall_post_ptrace(res, request, pid, addr, data) \
+ __sanitizer_syscall_post_impl_ptrace(res, (long)(request), (long)(pid), \
+ (long)(addr), (long)(data))
+#define __sanitizer_syscall_pre_add_key(_type, _description, _payload, plen, \
+ destringid) \
+ __sanitizer_syscall_pre_impl_add_key((long)(_type), (long)(_description), \
+ (long)(_payload), (long)(plen), \
+ (long)(destringid))
+#define __sanitizer_syscall_post_add_key(res, _type, _description, _payload, \
+ plen, destringid) \
+ __sanitizer_syscall_post_impl_add_key( \
+ res, (long)(_type), (long)(_description), (long)(_payload), \
+ (long)(plen), (long)(destringid))
+#define __sanitizer_syscall_pre_request_key(_type, _description, \
+ _callout_info, destringid) \
+ __sanitizer_syscall_pre_impl_request_key( \
+ (long)(_type), (long)(_description), (long)(_callout_info), \
+ (long)(destringid))
+#define __sanitizer_syscall_post_request_key(res, _type, _description, \
+ _callout_info, destringid) \
+ __sanitizer_syscall_post_impl_request_key( \
+ res, (long)(_type), (long)(_description), (long)(_callout_info), \
+ (long)(destringid))
+#define __sanitizer_syscall_pre_keyctl(cmd, arg2, arg3, arg4, arg5) \
+ __sanitizer_syscall_pre_impl_keyctl((long)(cmd), (long)(arg2), (long)(arg3), \
+ (long)(arg4), (long)(arg5))
+#define __sanitizer_syscall_post_keyctl(res, cmd, arg2, arg3, arg4, arg5) \
+ __sanitizer_syscall_post_impl_keyctl(res, (long)(cmd), (long)(arg2), \
+ (long)(arg3), (long)(arg4), \
+ (long)(arg5))
+#define __sanitizer_syscall_pre_ioprio_set(which, who, ioprio) \
+ __sanitizer_syscall_pre_impl_ioprio_set((long)(which), (long)(who), \
+ (long)(ioprio))
+#define __sanitizer_syscall_post_ioprio_set(res, which, who, ioprio) \
+ __sanitizer_syscall_post_impl_ioprio_set(res, (long)(which), (long)(who), \
+ (long)(ioprio))
+#define __sanitizer_syscall_pre_ioprio_get(which, who) \
+ __sanitizer_syscall_pre_impl_ioprio_get((long)(which), (long)(who))
+#define __sanitizer_syscall_post_ioprio_get(res, which, who) \
+ __sanitizer_syscall_post_impl_ioprio_get(res, (long)(which), (long)(who))
+#define __sanitizer_syscall_pre_set_mempolicy(mode, nmask, maxnode) \
+ __sanitizer_syscall_pre_impl_set_mempolicy((long)(mode), (long)(nmask), \
+ (long)(maxnode))
+#define __sanitizer_syscall_post_set_mempolicy(res, mode, nmask, maxnode) \
+ __sanitizer_syscall_post_impl_set_mempolicy(res, (long)(mode), \
+ (long)(nmask), (long)(maxnode))
+#define __sanitizer_syscall_pre_migrate_pages(pid, maxnode, from, to) \
+ __sanitizer_syscall_pre_impl_migrate_pages((long)(pid), (long)(maxnode), \
+ (long)(from), (long)(to))
+#define __sanitizer_syscall_post_migrate_pages(res, pid, maxnode, from, to) \
+ __sanitizer_syscall_post_impl_migrate_pages( \
+ res, (long)(pid), (long)(maxnode), (long)(from), (long)(to))
+#define __sanitizer_syscall_pre_move_pages(pid, nr_pages, pages, nodes, \
+ status, flags) \
+ __sanitizer_syscall_pre_impl_move_pages((long)(pid), (long)(nr_pages), \
+ (long)(pages), (long)(nodes), \
+ (long)(status), (long)(flags))
+#define __sanitizer_syscall_post_move_pages(res, pid, nr_pages, pages, nodes, \
+ status, flags) \
+ __sanitizer_syscall_post_impl_move_pages(res, (long)(pid), (long)(nr_pages), \
+ (long)(pages), (long)(nodes), \
+ (long)(status), (long)(flags))
+#define __sanitizer_syscall_pre_mbind(start, len, mode, nmask, maxnode, flags) \
+ __sanitizer_syscall_pre_impl_mbind((long)(start), (long)(len), (long)(mode), \
+ (long)(nmask), (long)(maxnode), \
+ (long)(flags))
+#define __sanitizer_syscall_post_mbind(res, start, len, mode, nmask, maxnode, \
+ flags) \
+ __sanitizer_syscall_post_impl_mbind(res, (long)(start), (long)(len), \
+ (long)(mode), (long)(nmask), \
+ (long)(maxnode), (long)(flags))
+#define __sanitizer_syscall_pre_get_mempolicy(policy, nmask, maxnode, addr, \
+ flags) \
+ __sanitizer_syscall_pre_impl_get_mempolicy((long)(policy), (long)(nmask), \
+ (long)(maxnode), (long)(addr), \
+ (long)(flags))
+#define __sanitizer_syscall_post_get_mempolicy(res, policy, nmask, maxnode, \
+ addr, flags) \
+ __sanitizer_syscall_post_impl_get_mempolicy(res, (long)(policy), \
+ (long)(nmask), (long)(maxnode), \
+ (long)(addr), (long)(flags))
+#define __sanitizer_syscall_pre_inotify_init() \
+ __sanitizer_syscall_pre_impl_inotify_init()
+#define __sanitizer_syscall_post_inotify_init(res) \
+ __sanitizer_syscall_post_impl_inotify_init(res)
+#define __sanitizer_syscall_pre_inotify_init1(flags) \
+ __sanitizer_syscall_pre_impl_inotify_init1((long)(flags))
+#define __sanitizer_syscall_post_inotify_init1(res, flags) \
+ __sanitizer_syscall_post_impl_inotify_init1(res, (long)(flags))
+#define __sanitizer_syscall_pre_inotify_add_watch(fd, path, mask) \
+ __sanitizer_syscall_pre_impl_inotify_add_watch((long)(fd), (long)(path), \
+ (long)(mask))
+#define __sanitizer_syscall_post_inotify_add_watch(res, fd, path, mask) \
+ __sanitizer_syscall_post_impl_inotify_add_watch(res, (long)(fd), \
+ (long)(path), (long)(mask))
+#define __sanitizer_syscall_pre_inotify_rm_watch(fd, wd) \
+ __sanitizer_syscall_pre_impl_inotify_rm_watch((long)(fd), (long)(wd))
+#define __sanitizer_syscall_post_inotify_rm_watch(res, fd, wd) \
+ __sanitizer_syscall_post_impl_inotify_rm_watch(res, (long)(fd), (long)(wd))
+#define __sanitizer_syscall_pre_spu_run(fd, unpc, ustatus) \
+ __sanitizer_syscall_pre_impl_spu_run((long)(fd), (long)(unpc), \
+ (long)(ustatus))
+#define __sanitizer_syscall_post_spu_run(res, fd, unpc, ustatus) \
+ __sanitizer_syscall_post_impl_spu_run(res, (long)(fd), (long)(unpc), \
+ (long)(ustatus))
+#define __sanitizer_syscall_pre_spu_create(name, flags, mode, fd) \
+ __sanitizer_syscall_pre_impl_spu_create((long)(name), (long)(flags), \
+ (long)(mode), (long)(fd))
+#define __sanitizer_syscall_post_spu_create(res, name, flags, mode, fd) \
+ __sanitizer_syscall_post_impl_spu_create(res, (long)(name), (long)(flags), \
+ (long)(mode), (long)(fd))
+#define __sanitizer_syscall_pre_mknodat(dfd, filename, mode, dev) \
+ __sanitizer_syscall_pre_impl_mknodat((long)(dfd), (long)(filename), \
+ (long)(mode), (long)(dev))
+#define __sanitizer_syscall_post_mknodat(res, dfd, filename, mode, dev) \
+ __sanitizer_syscall_post_impl_mknodat(res, (long)(dfd), (long)(filename), \
+ (long)(mode), (long)(dev))
+#define __sanitizer_syscall_pre_mkdirat(dfd, pathname, mode) \
+ __sanitizer_syscall_pre_impl_mkdirat((long)(dfd), (long)(pathname), \
+ (long)(mode))
+#define __sanitizer_syscall_post_mkdirat(res, dfd, pathname, mode) \
+ __sanitizer_syscall_post_impl_mkdirat(res, (long)(dfd), (long)(pathname), \
+ (long)(mode))
+#define __sanitizer_syscall_pre_unlinkat(dfd, pathname, flag) \
+ __sanitizer_syscall_pre_impl_unlinkat((long)(dfd), (long)(pathname), \
+ (long)(flag))
+#define __sanitizer_syscall_post_unlinkat(res, dfd, pathname, flag) \
+ __sanitizer_syscall_post_impl_unlinkat(res, (long)(dfd), (long)(pathname), \
+ (long)(flag))
+#define __sanitizer_syscall_pre_symlinkat(oldname, newdfd, newname) \
+ __sanitizer_syscall_pre_impl_symlinkat((long)(oldname), (long)(newdfd), \
+ (long)(newname))
+#define __sanitizer_syscall_post_symlinkat(res, oldname, newdfd, newname) \
+ __sanitizer_syscall_post_impl_symlinkat(res, (long)(oldname), \
+ (long)(newdfd), (long)(newname))
+#define __sanitizer_syscall_pre_linkat(olddfd, oldname, newdfd, newname, \
+ flags) \
+ __sanitizer_syscall_pre_impl_linkat((long)(olddfd), (long)(oldname), \
+ (long)(newdfd), (long)(newname), \
+ (long)(flags))
+#define __sanitizer_syscall_post_linkat(res, olddfd, oldname, newdfd, newname, \
+ flags) \
+ __sanitizer_syscall_post_impl_linkat(res, (long)(olddfd), (long)(oldname), \
+ (long)(newdfd), (long)(newname), \
+ (long)(flags))
+#define __sanitizer_syscall_pre_renameat(olddfd, oldname, newdfd, newname) \
+ __sanitizer_syscall_pre_impl_renameat((long)(olddfd), (long)(oldname), \
+ (long)(newdfd), (long)(newname))
+#define __sanitizer_syscall_post_renameat(res, olddfd, oldname, newdfd, \
+ newname) \
+ __sanitizer_syscall_post_impl_renameat(res, (long)(olddfd), (long)(oldname), \
+ (long)(newdfd), (long)(newname))
+#define __sanitizer_syscall_pre_futimesat(dfd, filename, utimes) \
+ __sanitizer_syscall_pre_impl_futimesat((long)(dfd), (long)(filename), \
+ (long)(utimes))
+#define __sanitizer_syscall_post_futimesat(res, dfd, filename, utimes) \
+ __sanitizer_syscall_post_impl_futimesat(res, (long)(dfd), (long)(filename), \
+ (long)(utimes))
+#define __sanitizer_syscall_pre_faccessat(dfd, filename, mode) \
+ __sanitizer_syscall_pre_impl_faccessat((long)(dfd), (long)(filename), \
+ (long)(mode))
+#define __sanitizer_syscall_post_faccessat(res, dfd, filename, mode) \
+ __sanitizer_syscall_post_impl_faccessat(res, (long)(dfd), (long)(filename), \
+ (long)(mode))
+#define __sanitizer_syscall_pre_fchmodat(dfd, filename, mode) \
+ __sanitizer_syscall_pre_impl_fchmodat((long)(dfd), (long)(filename), \
+ (long)(mode))
+#define __sanitizer_syscall_post_fchmodat(res, dfd, filename, mode) \
+ __sanitizer_syscall_post_impl_fchmodat(res, (long)(dfd), (long)(filename), \
+ (long)(mode))
+#define __sanitizer_syscall_pre_fchownat(dfd, filename, user, group, flag) \
+ __sanitizer_syscall_pre_impl_fchownat((long)(dfd), (long)(filename), \
+ (long)(user), (long)(group), \
+ (long)(flag))
+#define __sanitizer_syscall_post_fchownat(res, dfd, filename, user, group, \
+ flag) \
+ __sanitizer_syscall_post_impl_fchownat(res, (long)(dfd), (long)(filename), \
+ (long)(user), (long)(group), \
+ (long)(flag))
+#define __sanitizer_syscall_pre_openat(dfd, filename, flags, mode) \
+ __sanitizer_syscall_pre_impl_openat((long)(dfd), (long)(filename), \
+ (long)(flags), (long)(mode))
+#define __sanitizer_syscall_post_openat(res, dfd, filename, flags, mode) \
+ __sanitizer_syscall_post_impl_openat(res, (long)(dfd), (long)(filename), \
+ (long)(flags), (long)(mode))
+#define __sanitizer_syscall_pre_newfstatat(dfd, filename, statbuf, flag) \
+ __sanitizer_syscall_pre_impl_newfstatat((long)(dfd), (long)(filename), \
+ (long)(statbuf), (long)(flag))
+#define __sanitizer_syscall_post_newfstatat(res, dfd, filename, statbuf, flag) \
+ __sanitizer_syscall_post_impl_newfstatat(res, (long)(dfd), (long)(filename), \
+ (long)(statbuf), (long)(flag))
+#define __sanitizer_syscall_pre_fstatat64(dfd, filename, statbuf, flag) \
+ __sanitizer_syscall_pre_impl_fstatat64((long)(dfd), (long)(filename), \
+ (long)(statbuf), (long)(flag))
+#define __sanitizer_syscall_post_fstatat64(res, dfd, filename, statbuf, flag) \
+ __sanitizer_syscall_post_impl_fstatat64(res, (long)(dfd), (long)(filename), \
+ (long)(statbuf), (long)(flag))
+#define __sanitizer_syscall_pre_readlinkat(dfd, path, buf, bufsiz) \
+ __sanitizer_syscall_pre_impl_readlinkat((long)(dfd), (long)(path), \
+ (long)(buf), (long)(bufsiz))
+#define __sanitizer_syscall_post_readlinkat(res, dfd, path, buf, bufsiz) \
+ __sanitizer_syscall_post_impl_readlinkat(res, (long)(dfd), (long)(path), \
+ (long)(buf), (long)(bufsiz))
+#define __sanitizer_syscall_pre_utimensat(dfd, filename, utimes, flags) \
+ __sanitizer_syscall_pre_impl_utimensat((long)(dfd), (long)(filename), \
+ (long)(utimes), (long)(flags))
+#define __sanitizer_syscall_post_utimensat(res, dfd, filename, utimes, flags) \
+ __sanitizer_syscall_post_impl_utimensat(res, (long)(dfd), (long)(filename), \
+ (long)(utimes), (long)(flags))
+#define __sanitizer_syscall_pre_unshare(unshare_flags) \
+ __sanitizer_syscall_pre_impl_unshare((long)(unshare_flags))
+#define __sanitizer_syscall_post_unshare(res, unshare_flags) \
+ __sanitizer_syscall_post_impl_unshare(res, (long)(unshare_flags))
+#define __sanitizer_syscall_pre_splice(fd_in, off_in, fd_out, off_out, len, \
+ flags) \
+ __sanitizer_syscall_pre_impl_splice((long)(fd_in), (long)(off_in), \
+ (long)(fd_out), (long)(off_out), \
+ (long)(len), (long)(flags))
+#define __sanitizer_syscall_post_splice(res, fd_in, off_in, fd_out, off_out, \
+ len, flags) \
+ __sanitizer_syscall_post_impl_splice(res, (long)(fd_in), (long)(off_in), \
+ (long)(fd_out), (long)(off_out), \
+ (long)(len), (long)(flags))
+#define __sanitizer_syscall_pre_vmsplice(fd, iov, nr_segs, flags) \
+ __sanitizer_syscall_pre_impl_vmsplice((long)(fd), (long)(iov), \
+ (long)(nr_segs), (long)(flags))
+#define __sanitizer_syscall_post_vmsplice(res, fd, iov, nr_segs, flags) \
+ __sanitizer_syscall_post_impl_vmsplice(res, (long)(fd), (long)(iov), \
+ (long)(nr_segs), (long)(flags))
+#define __sanitizer_syscall_pre_tee(fdin, fdout, len, flags) \
+ __sanitizer_syscall_pre_impl_tee((long)(fdin), (long)(fdout), (long)(len), \
+ (long)(flags))
+#define __sanitizer_syscall_post_tee(res, fdin, fdout, len, flags) \
+ __sanitizer_syscall_post_impl_tee(res, (long)(fdin), (long)(fdout), \
+ (long)(len), (long)(flags))
+#define __sanitizer_syscall_pre_get_robust_list(pid, head_ptr, len_ptr) \
+ __sanitizer_syscall_pre_impl_get_robust_list((long)(pid), (long)(head_ptr), \
+ (long)(len_ptr))
+#define __sanitizer_syscall_post_get_robust_list(res, pid, head_ptr, len_ptr) \
+ __sanitizer_syscall_post_impl_get_robust_list( \
+ res, (long)(pid), (long)(head_ptr), (long)(len_ptr))
+#define __sanitizer_syscall_pre_set_robust_list(head, len) \
+ __sanitizer_syscall_pre_impl_set_robust_list((long)(head), (long)(len))
+#define __sanitizer_syscall_post_set_robust_list(res, head, len) \
+ __sanitizer_syscall_post_impl_set_robust_list(res, (long)(head), (long)(len))
+#define __sanitizer_syscall_pre_getcpu(cpu, node, cache) \
+ __sanitizer_syscall_pre_impl_getcpu((long)(cpu), (long)(node), (long)(cache))
+#define __sanitizer_syscall_post_getcpu(res, cpu, node, cache) \
+ __sanitizer_syscall_post_impl_getcpu(res, (long)(cpu), (long)(node), \
+ (long)(cache))
+#define __sanitizer_syscall_pre_signalfd(ufd, user_mask, sizemask) \
+ __sanitizer_syscall_pre_impl_signalfd((long)(ufd), (long)(user_mask), \
+ (long)(sizemask))
+#define __sanitizer_syscall_post_signalfd(res, ufd, user_mask, sizemask) \
+ __sanitizer_syscall_post_impl_signalfd(res, (long)(ufd), (long)(user_mask), \
+ (long)(sizemask))
+#define __sanitizer_syscall_pre_signalfd4(ufd, user_mask, sizemask, flags) \
+ __sanitizer_syscall_pre_impl_signalfd4((long)(ufd), (long)(user_mask), \
+ (long)(sizemask), (long)(flags))
+#define __sanitizer_syscall_post_signalfd4(res, ufd, user_mask, sizemask, \
+ flags) \
+ __sanitizer_syscall_post_impl_signalfd4(res, (long)(ufd), (long)(user_mask), \
+ (long)(sizemask), (long)(flags))
+#define __sanitizer_syscall_pre_timerfd_create(clockid, flags) \
+ __sanitizer_syscall_pre_impl_timerfd_create((long)(clockid), (long)(flags))
+#define __sanitizer_syscall_post_timerfd_create(res, clockid, flags) \
+ __sanitizer_syscall_post_impl_timerfd_create(res, (long)(clockid), \
+ (long)(flags))
+#define __sanitizer_syscall_pre_timerfd_settime(ufd, flags, utmr, otmr) \
+ __sanitizer_syscall_pre_impl_timerfd_settime((long)(ufd), (long)(flags), \
+ (long)(utmr), (long)(otmr))
+#define __sanitizer_syscall_post_timerfd_settime(res, ufd, flags, utmr, otmr) \
+ __sanitizer_syscall_post_impl_timerfd_settime( \
+ res, (long)(ufd), (long)(flags), (long)(utmr), (long)(otmr))
+#define __sanitizer_syscall_pre_timerfd_gettime(ufd, otmr) \
+ __sanitizer_syscall_pre_impl_timerfd_gettime((long)(ufd), (long)(otmr))
+#define __sanitizer_syscall_post_timerfd_gettime(res, ufd, otmr) \
+ __sanitizer_syscall_post_impl_timerfd_gettime(res, (long)(ufd), (long)(otmr))
+#define __sanitizer_syscall_pre_eventfd(count) \
+ __sanitizer_syscall_pre_impl_eventfd((long)(count))
+#define __sanitizer_syscall_post_eventfd(res, count) \
+ __sanitizer_syscall_post_impl_eventfd(res, (long)(count))
+#define __sanitizer_syscall_pre_eventfd2(count, flags) \
+ __sanitizer_syscall_pre_impl_eventfd2((long)(count), (long)(flags))
+#define __sanitizer_syscall_post_eventfd2(res, count, flags) \
+ __sanitizer_syscall_post_impl_eventfd2(res, (long)(count), (long)(flags))
+#define __sanitizer_syscall_pre_old_readdir(arg0, arg1, arg2) \
+ __sanitizer_syscall_pre_impl_old_readdir((long)(arg0), (long)(arg1), \
+ (long)(arg2))
+#define __sanitizer_syscall_post_old_readdir(res, arg0, arg1, arg2) \
+ __sanitizer_syscall_post_impl_old_readdir(res, (long)(arg0), (long)(arg1), \
+ (long)(arg2))
+#define __sanitizer_syscall_pre_pselect6(arg0, arg1, arg2, arg3, arg4, arg5) \
+ __sanitizer_syscall_pre_impl_pselect6((long)(arg0), (long)(arg1), \
+ (long)(arg2), (long)(arg3), \
+ (long)(arg4), (long)(arg5))
+#define __sanitizer_syscall_post_pselect6(res, arg0, arg1, arg2, arg3, arg4, \
+ arg5) \
+ __sanitizer_syscall_post_impl_pselect6(res, (long)(arg0), (long)(arg1), \
+ (long)(arg2), (long)(arg3), \
+ (long)(arg4), (long)(arg5))
+#define __sanitizer_syscall_pre_ppoll(arg0, arg1, arg2, arg3, arg4) \
+ __sanitizer_syscall_pre_impl_ppoll((long)(arg0), (long)(arg1), (long)(arg2), \
+ (long)(arg3), (long)(arg4))
+#define __sanitizer_syscall_post_ppoll(res, arg0, arg1, arg2, arg3, arg4) \
+ __sanitizer_syscall_post_impl_ppoll(res, (long)(arg0), (long)(arg1), \
+ (long)(arg2), (long)(arg3), \
+ (long)(arg4))
+#define __sanitizer_syscall_pre_syncfs(fd) \
+ __sanitizer_syscall_pre_impl_syncfs((long)(fd))
+#define __sanitizer_syscall_post_syncfs(res, fd) \
+ __sanitizer_syscall_post_impl_syncfs(res, (long)(fd))
+#define __sanitizer_syscall_pre_perf_event_open(attr_uptr, pid, cpu, group_fd, \
+ flags) \
+ __sanitizer_syscall_pre_impl_perf_event_open((long)(attr_uptr), (long)(pid), \
+ (long)(cpu), (long)(group_fd), \
+ (long)(flags))
+#define __sanitizer_syscall_post_perf_event_open(res, attr_uptr, pid, cpu, \
+ group_fd, flags) \
+ __sanitizer_syscall_post_impl_perf_event_open( \
+ res, (long)(attr_uptr), (long)(pid), (long)(cpu), (long)(group_fd), \
+ (long)(flags))
+#define __sanitizer_syscall_pre_mmap_pgoff(addr, len, prot, flags, fd, pgoff) \
+ __sanitizer_syscall_pre_impl_mmap_pgoff((long)(addr), (long)(len), \
+ (long)(prot), (long)(flags), \
+ (long)(fd), (long)(pgoff))
+#define __sanitizer_syscall_post_mmap_pgoff(res, addr, len, prot, flags, fd, \
+ pgoff) \
+ __sanitizer_syscall_post_impl_mmap_pgoff(res, (long)(addr), (long)(len), \
+ (long)(prot), (long)(flags), \
+ (long)(fd), (long)(pgoff))
+#define __sanitizer_syscall_pre_old_mmap(arg) \
+ __sanitizer_syscall_pre_impl_old_mmap((long)(arg))
+#define __sanitizer_syscall_post_old_mmap(res, arg) \
+ __sanitizer_syscall_post_impl_old_mmap(res, (long)(arg))
+#define __sanitizer_syscall_pre_name_to_handle_at(dfd, name, handle, mnt_id, \
+ flag) \
+ __sanitizer_syscall_pre_impl_name_to_handle_at( \
+ (long)(dfd), (long)(name), (long)(handle), (long)(mnt_id), (long)(flag))
+#define __sanitizer_syscall_post_name_to_handle_at(res, dfd, name, handle, \
+ mnt_id, flag) \
+ __sanitizer_syscall_post_impl_name_to_handle_at( \
+ res, (long)(dfd), (long)(name), (long)(handle), (long)(mnt_id), \
+ (long)(flag))
+#define __sanitizer_syscall_pre_open_by_handle_at(mountdirfd, handle, flags) \
+ __sanitizer_syscall_pre_impl_open_by_handle_at( \
+ (long)(mountdirfd), (long)(handle), (long)(flags))
+#define __sanitizer_syscall_post_open_by_handle_at(res, mountdirfd, handle, \
+ flags) \
+ __sanitizer_syscall_post_impl_open_by_handle_at( \
+ res, (long)(mountdirfd), (long)(handle), (long)(flags))
+#define __sanitizer_syscall_pre_setns(fd, nstype) \
+ __sanitizer_syscall_pre_impl_setns((long)(fd), (long)(nstype))
+#define __sanitizer_syscall_post_setns(res, fd, nstype) \
+ __sanitizer_syscall_post_impl_setns(res, (long)(fd), (long)(nstype))
+#define __sanitizer_syscall_pre_process_vm_readv(pid, lvec, liovcnt, rvec, \
+ riovcnt, flags) \
+ __sanitizer_syscall_pre_impl_process_vm_readv( \
+ (long)(pid), (long)(lvec), (long)(liovcnt), (long)(rvec), \
+ (long)(riovcnt), (long)(flags))
+#define __sanitizer_syscall_post_process_vm_readv(res, pid, lvec, liovcnt, \
+ rvec, riovcnt, flags) \
+ __sanitizer_syscall_post_impl_process_vm_readv( \
+ res, (long)(pid), (long)(lvec), (long)(liovcnt), (long)(rvec), \
+ (long)(riovcnt), (long)(flags))
+#define __sanitizer_syscall_pre_process_vm_writev(pid, lvec, liovcnt, rvec, \
+ riovcnt, flags) \
+ __sanitizer_syscall_pre_impl_process_vm_writev( \
+ (long)(pid), (long)(lvec), (long)(liovcnt), (long)(rvec), \
+ (long)(riovcnt), (long)(flags))
+#define __sanitizer_syscall_post_process_vm_writev(res, pid, lvec, liovcnt, \
+ rvec, riovcnt, flags) \
+ __sanitizer_syscall_post_impl_process_vm_writev( \
+ res, (long)(pid), (long)(lvec), (long)(liovcnt), (long)(rvec), \
+ (long)(riovcnt), (long)(flags))
+#define __sanitizer_syscall_pre_fork() \
+ __sanitizer_syscall_pre_impl_fork()
+#define __sanitizer_syscall_post_fork(res) \
+ __sanitizer_syscall_post_impl_fork(res)
+#define __sanitizer_syscall_pre_vfork() \
+ __sanitizer_syscall_pre_impl_vfork()
+#define __sanitizer_syscall_post_vfork(res) \
+ __sanitizer_syscall_post_impl_vfork(res)
+#define __sanitizer_syscall_pre_sigaction(signum, act, oldact) \
+ __sanitizer_syscall_pre_impl_sigaction((long)signum, (long)act, (long)oldact)
+#define __sanitizer_syscall_post_sigaction(res, signum, act, oldact) \
+ __sanitizer_syscall_post_impl_sigaction(res, (long)signum, (long)act, \
+ (long)oldact)
+#define __sanitizer_syscall_pre_rt_sigaction(signum, act, oldact, sz) \
+ __sanitizer_syscall_pre_impl_rt_sigaction((long)signum, (long)act, \
+ (long)oldact, (long)sz)
+#define __sanitizer_syscall_post_rt_sigaction(res, signum, act, oldact, sz) \
+ __sanitizer_syscall_post_impl_rt_sigaction(res, (long)signum, (long)act, \
+ (long)oldact, (long)sz)
+#define __sanitizer_syscall_pre_sigaltstack(ss, oss) \
+ __sanitizer_syscall_pre_impl_sigaltstack((long)ss, (long)oss)
+#define __sanitizer_syscall_post_sigaltstack(res, ss, oss) \
+ __sanitizer_syscall_post_impl_sigaltstack(res, (long)ss, (long)oss)
+
+// And now a few syscalls we don't handle yet.
+#define __sanitizer_syscall_pre_afs_syscall(...)
+#define __sanitizer_syscall_pre_arch_prctl(...)
+#define __sanitizer_syscall_pre_break(...)
+#define __sanitizer_syscall_pre_chown32(...)
+#define __sanitizer_syscall_pre_clone(...)
+#define __sanitizer_syscall_pre_create_module(...)
+#define __sanitizer_syscall_pre_epoll_ctl_old(...)
+#define __sanitizer_syscall_pre_epoll_wait_old(...)
+#define __sanitizer_syscall_pre_execve(...)
+#define __sanitizer_syscall_pre_fadvise64(...)
+#define __sanitizer_syscall_pre_fadvise64_64(...)
+#define __sanitizer_syscall_pre_fallocate(...)
+#define __sanitizer_syscall_pre_fanotify_init(...)
+#define __sanitizer_syscall_pre_fanotify_mark(...)
+#define __sanitizer_syscall_pre_fchown32(...)
+#define __sanitizer_syscall_pre_ftime(...)
+#define __sanitizer_syscall_pre_ftruncate64(...)
+#define __sanitizer_syscall_pre_futex(...)
+#define __sanitizer_syscall_pre_getegid32(...)
+#define __sanitizer_syscall_pre_geteuid32(...)
+#define __sanitizer_syscall_pre_getgid32(...)
+#define __sanitizer_syscall_pre_getgroups32(...)
+#define __sanitizer_syscall_pre_get_kernel_syms(...)
+#define __sanitizer_syscall_pre_getpmsg(...)
+#define __sanitizer_syscall_pre_getresgid32(...)
+#define __sanitizer_syscall_pre_getresuid32(...)
+#define __sanitizer_syscall_pre_get_thread_area(...)
+#define __sanitizer_syscall_pre_getuid32(...)
+#define __sanitizer_syscall_pre_gtty(...)
+#define __sanitizer_syscall_pre_idle(...)
+#define __sanitizer_syscall_pre_iopl(...)
+#define __sanitizer_syscall_pre_lchown32(...)
+#define __sanitizer_syscall_pre__llseek(...)
+#define __sanitizer_syscall_pre_lock(...)
+#define __sanitizer_syscall_pre_madvise1(...)
+#define __sanitizer_syscall_pre_mmap(...)
+#define __sanitizer_syscall_pre_mmap2(...)
+#define __sanitizer_syscall_pre_modify_ldt(...)
+#define __sanitizer_syscall_pre_mpx(...)
+#define __sanitizer_syscall_pre__newselect(...)
+#define __sanitizer_syscall_pre_nfsservctl(...)
+#define __sanitizer_syscall_pre_oldfstat(...)
+#define __sanitizer_syscall_pre_oldlstat(...)
+#define __sanitizer_syscall_pre_oldolduname(...)
+#define __sanitizer_syscall_pre_oldstat(...)
+#define __sanitizer_syscall_pre_prctl(...)
+#define __sanitizer_syscall_pre_prof(...)
+#define __sanitizer_syscall_pre_profil(...)
+#define __sanitizer_syscall_pre_putpmsg(...)
+#define __sanitizer_syscall_pre_query_module(...)
+#define __sanitizer_syscall_pre_readahead(...)
+#define __sanitizer_syscall_pre_readdir(...)
+#define __sanitizer_syscall_pre_rt_sigreturn(...)
+#define __sanitizer_syscall_pre_rt_sigsuspend(...)
+#define __sanitizer_syscall_pre_security(...)
+#define __sanitizer_syscall_pre_setfsgid32(...)
+#define __sanitizer_syscall_pre_setfsuid32(...)
+#define __sanitizer_syscall_pre_setgid32(...)
+#define __sanitizer_syscall_pre_setgroups32(...)
+#define __sanitizer_syscall_pre_setregid32(...)
+#define __sanitizer_syscall_pre_setresgid32(...)
+#define __sanitizer_syscall_pre_setresuid32(...)
+#define __sanitizer_syscall_pre_setreuid32(...)
+#define __sanitizer_syscall_pre_set_thread_area(...)
+#define __sanitizer_syscall_pre_setuid32(...)
+#define __sanitizer_syscall_pre_sigreturn(...)
+#define __sanitizer_syscall_pre_sigsuspend(...)
+#define __sanitizer_syscall_pre_stty(...)
+#define __sanitizer_syscall_pre_sync_file_range(...)
+#define __sanitizer_syscall_pre__sysctl(...)
+#define __sanitizer_syscall_pre_truncate64(...)
+#define __sanitizer_syscall_pre_tuxcall(...)
+#define __sanitizer_syscall_pre_ugetrlimit(...)
+#define __sanitizer_syscall_pre_ulimit(...)
+#define __sanitizer_syscall_pre_umount2(...)
+#define __sanitizer_syscall_pre_vm86(...)
+#define __sanitizer_syscall_pre_vm86old(...)
+#define __sanitizer_syscall_pre_vserver(...)
+
+#define __sanitizer_syscall_post_afs_syscall(res, ...)
+#define __sanitizer_syscall_post_arch_prctl(res, ...)
+#define __sanitizer_syscall_post_break(res, ...)
+#define __sanitizer_syscall_post_chown32(res, ...)
+#define __sanitizer_syscall_post_clone(res, ...)
+#define __sanitizer_syscall_post_create_module(res, ...)
+#define __sanitizer_syscall_post_epoll_ctl_old(res, ...)
+#define __sanitizer_syscall_post_epoll_wait_old(res, ...)
+#define __sanitizer_syscall_post_execve(res, ...)
+#define __sanitizer_syscall_post_fadvise64(res, ...)
+#define __sanitizer_syscall_post_fadvise64_64(res, ...)
+#define __sanitizer_syscall_post_fallocate(res, ...)
+#define __sanitizer_syscall_post_fanotify_init(res, ...)
+#define __sanitizer_syscall_post_fanotify_mark(res, ...)
+#define __sanitizer_syscall_post_fchown32(res, ...)
+#define __sanitizer_syscall_post_ftime(res, ...)
+#define __sanitizer_syscall_post_ftruncate64(res, ...)
+#define __sanitizer_syscall_post_futex(res, ...)
+#define __sanitizer_syscall_post_getegid32(res, ...)
+#define __sanitizer_syscall_post_geteuid32(res, ...)
+#define __sanitizer_syscall_post_getgid32(res, ...)
+#define __sanitizer_syscall_post_getgroups32(res, ...)
+#define __sanitizer_syscall_post_get_kernel_syms(res, ...)
+#define __sanitizer_syscall_post_getpmsg(res, ...)
+#define __sanitizer_syscall_post_getresgid32(res, ...)
+#define __sanitizer_syscall_post_getresuid32(res, ...)
+#define __sanitizer_syscall_post_get_thread_area(res, ...)
+#define __sanitizer_syscall_post_getuid32(res, ...)
+#define __sanitizer_syscall_post_gtty(res, ...)
+#define __sanitizer_syscall_post_idle(res, ...)
+#define __sanitizer_syscall_post_iopl(res, ...)
+#define __sanitizer_syscall_post_lchown32(res, ...)
+#define __sanitizer_syscall_post__llseek(res, ...)
+#define __sanitizer_syscall_post_lock(res, ...)
+#define __sanitizer_syscall_post_madvise1(res, ...)
+#define __sanitizer_syscall_post_mmap2(res, ...)
+#define __sanitizer_syscall_post_mmap(res, ...)
+#define __sanitizer_syscall_post_modify_ldt(res, ...)
+#define __sanitizer_syscall_post_mpx(res, ...)
+#define __sanitizer_syscall_post__newselect(res, ...)
+#define __sanitizer_syscall_post_nfsservctl(res, ...)
+#define __sanitizer_syscall_post_oldfstat(res, ...)
+#define __sanitizer_syscall_post_oldlstat(res, ...)
+#define __sanitizer_syscall_post_oldolduname(res, ...)
+#define __sanitizer_syscall_post_oldstat(res, ...)
+#define __sanitizer_syscall_post_prctl(res, ...)
+#define __sanitizer_syscall_post_profil(res, ...)
+#define __sanitizer_syscall_post_prof(res, ...)
+#define __sanitizer_syscall_post_putpmsg(res, ...)
+#define __sanitizer_syscall_post_query_module(res, ...)
+#define __sanitizer_syscall_post_readahead(res, ...)
+#define __sanitizer_syscall_post_readdir(res, ...)
+#define __sanitizer_syscall_post_rt_sigreturn(res, ...)
+#define __sanitizer_syscall_post_rt_sigsuspend(res, ...)
+#define __sanitizer_syscall_post_security(res, ...)
+#define __sanitizer_syscall_post_setfsgid32(res, ...)
+#define __sanitizer_syscall_post_setfsuid32(res, ...)
+#define __sanitizer_syscall_post_setgid32(res, ...)
+#define __sanitizer_syscall_post_setgroups32(res, ...)
+#define __sanitizer_syscall_post_setregid32(res, ...)
+#define __sanitizer_syscall_post_setresgid32(res, ...)
+#define __sanitizer_syscall_post_setresuid32(res, ...)
+#define __sanitizer_syscall_post_setreuid32(res, ...)
+#define __sanitizer_syscall_post_set_thread_area(res, ...)
+#define __sanitizer_syscall_post_setuid32(res, ...)
+#define __sanitizer_syscall_post_sigreturn(res, ...)
+#define __sanitizer_syscall_post_sigsuspend(res, ...)
+#define __sanitizer_syscall_post_stty(res, ...)
+#define __sanitizer_syscall_post_sync_file_range(res, ...)
+#define __sanitizer_syscall_post__sysctl(res, ...)
+#define __sanitizer_syscall_post_truncate64(res, ...)
+#define __sanitizer_syscall_post_tuxcall(res, ...)
+#define __sanitizer_syscall_post_ugetrlimit(res, ...)
+#define __sanitizer_syscall_post_ulimit(res, ...)
+#define __sanitizer_syscall_post_umount2(res, ...)
+#define __sanitizer_syscall_post_vm86old(res, ...)
+#define __sanitizer_syscall_post_vm86(res, ...)
+#define __sanitizer_syscall_post_vserver(res, ...)
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+// Private declarations. Do not call directly from user code. Use macros above.
+void __sanitizer_syscall_pre_impl_time(long tloc);
+void __sanitizer_syscall_post_impl_time(long res, long tloc);
+void __sanitizer_syscall_pre_impl_stime(long tptr);
+void __sanitizer_syscall_post_impl_stime(long res, long tptr);
+void __sanitizer_syscall_pre_impl_gettimeofday(long tv, long tz);
+void __sanitizer_syscall_post_impl_gettimeofday(long res, long tv, long tz);
+void __sanitizer_syscall_pre_impl_settimeofday(long tv, long tz);
+void __sanitizer_syscall_post_impl_settimeofday(long res, long tv, long tz);
+void __sanitizer_syscall_pre_impl_adjtimex(long txc_p);
+void __sanitizer_syscall_post_impl_adjtimex(long res, long txc_p);
+void __sanitizer_syscall_pre_impl_times(long tbuf);
+void __sanitizer_syscall_post_impl_times(long res, long tbuf);
+void __sanitizer_syscall_pre_impl_gettid();
+void __sanitizer_syscall_post_impl_gettid(long res);
+void __sanitizer_syscall_pre_impl_nanosleep(long rqtp, long rmtp);
+void __sanitizer_syscall_post_impl_nanosleep(long res, long rqtp, long rmtp);
+void __sanitizer_syscall_pre_impl_alarm(long seconds);
+void __sanitizer_syscall_post_impl_alarm(long res, long seconds);
+void __sanitizer_syscall_pre_impl_getpid();
+void __sanitizer_syscall_post_impl_getpid(long res);
+void __sanitizer_syscall_pre_impl_getppid();
+void __sanitizer_syscall_post_impl_getppid(long res);
+void __sanitizer_syscall_pre_impl_getuid();
+void __sanitizer_syscall_post_impl_getuid(long res);
+void __sanitizer_syscall_pre_impl_geteuid();
+void __sanitizer_syscall_post_impl_geteuid(long res);
+void __sanitizer_syscall_pre_impl_getgid();
+void __sanitizer_syscall_post_impl_getgid(long res);
+void __sanitizer_syscall_pre_impl_getegid();
+void __sanitizer_syscall_post_impl_getegid(long res);
+void __sanitizer_syscall_pre_impl_getresuid(long ruid, long euid, long suid);
+void __sanitizer_syscall_post_impl_getresuid(long res, long ruid, long euid,
+ long suid);
+void __sanitizer_syscall_pre_impl_getresgid(long rgid, long egid, long sgid);
+void __sanitizer_syscall_post_impl_getresgid(long res, long rgid, long egid,
+ long sgid);
+void __sanitizer_syscall_pre_impl_getpgid(long pid);
+void __sanitizer_syscall_post_impl_getpgid(long res, long pid);
+void __sanitizer_syscall_pre_impl_getpgrp();
+void __sanitizer_syscall_post_impl_getpgrp(long res);
+void __sanitizer_syscall_pre_impl_getsid(long pid);
+void __sanitizer_syscall_post_impl_getsid(long res, long pid);
+void __sanitizer_syscall_pre_impl_getgroups(long gidsetsize, long grouplist);
+void __sanitizer_syscall_post_impl_getgroups(long res, long gidsetsize,
+ long grouplist);
+void __sanitizer_syscall_pre_impl_setregid(long rgid, long egid);
+void __sanitizer_syscall_post_impl_setregid(long res, long rgid, long egid);
+void __sanitizer_syscall_pre_impl_setgid(long gid);
+void __sanitizer_syscall_post_impl_setgid(long res, long gid);
+void __sanitizer_syscall_pre_impl_setreuid(long ruid, long euid);
+void __sanitizer_syscall_post_impl_setreuid(long res, long ruid, long euid);
+void __sanitizer_syscall_pre_impl_setuid(long uid);
+void __sanitizer_syscall_post_impl_setuid(long res, long uid);
+void __sanitizer_syscall_pre_impl_setresuid(long ruid, long euid, long suid);
+void __sanitizer_syscall_post_impl_setresuid(long res, long ruid, long euid,
+ long suid);
+void __sanitizer_syscall_pre_impl_setresgid(long rgid, long egid, long sgid);
+void __sanitizer_syscall_post_impl_setresgid(long res, long rgid, long egid,
+ long sgid);
+void __sanitizer_syscall_pre_impl_setfsuid(long uid);
+void __sanitizer_syscall_post_impl_setfsuid(long res, long uid);
+void __sanitizer_syscall_pre_impl_setfsgid(long gid);
+void __sanitizer_syscall_post_impl_setfsgid(long res, long gid);
+void __sanitizer_syscall_pre_impl_setpgid(long pid, long pgid);
+void __sanitizer_syscall_post_impl_setpgid(long res, long pid, long pgid);
+void __sanitizer_syscall_pre_impl_setsid();
+void __sanitizer_syscall_post_impl_setsid(long res);
+void __sanitizer_syscall_pre_impl_setgroups(long gidsetsize, long grouplist);
+void __sanitizer_syscall_post_impl_setgroups(long res, long gidsetsize,
+ long grouplist);
+void __sanitizer_syscall_pre_impl_acct(long name);
+void __sanitizer_syscall_post_impl_acct(long res, long name);
+void __sanitizer_syscall_pre_impl_capget(long header, long dataptr);
+void __sanitizer_syscall_post_impl_capget(long res, long header, long dataptr);
+void __sanitizer_syscall_pre_impl_capset(long header, long data);
+void __sanitizer_syscall_post_impl_capset(long res, long header, long data);
+void __sanitizer_syscall_pre_impl_personality(long personality);
+void __sanitizer_syscall_post_impl_personality(long res, long personality);
+void __sanitizer_syscall_pre_impl_sigpending(long set);
+void __sanitizer_syscall_post_impl_sigpending(long res, long set);
+void __sanitizer_syscall_pre_impl_sigprocmask(long how, long set, long oset);
+void __sanitizer_syscall_post_impl_sigprocmask(long res, long how, long set,
+ long oset);
+void __sanitizer_syscall_pre_impl_getitimer(long which, long value);
+void __sanitizer_syscall_post_impl_getitimer(long res, long which, long value);
+void __sanitizer_syscall_pre_impl_setitimer(long which, long value,
+ long ovalue);
+void __sanitizer_syscall_post_impl_setitimer(long res, long which, long value,
+ long ovalue);
+void __sanitizer_syscall_pre_impl_timer_create(long which_clock,
+ long timer_event_spec,
+ long created_timer_id);
+void __sanitizer_syscall_post_impl_timer_create(long res, long which_clock,
+ long timer_event_spec,
+ long created_timer_id);
+void __sanitizer_syscall_pre_impl_timer_gettime(long timer_id, long setting);
+void __sanitizer_syscall_post_impl_timer_gettime(long res, long timer_id,
+ long setting);
+void __sanitizer_syscall_pre_impl_timer_getoverrun(long timer_id);
+void __sanitizer_syscall_post_impl_timer_getoverrun(long res, long timer_id);
+void __sanitizer_syscall_pre_impl_timer_settime(long timer_id, long flags,
+ long new_setting,
+ long old_setting);
+void __sanitizer_syscall_post_impl_timer_settime(long res, long timer_id,
+ long flags, long new_setting,
+ long old_setting);
+void __sanitizer_syscall_pre_impl_timer_delete(long timer_id);
+void __sanitizer_syscall_post_impl_timer_delete(long res, long timer_id);
+void __sanitizer_syscall_pre_impl_clock_settime(long which_clock, long tp);
+void __sanitizer_syscall_post_impl_clock_settime(long res, long which_clock,
+ long tp);
+void __sanitizer_syscall_pre_impl_clock_gettime(long which_clock, long tp);
+void __sanitizer_syscall_post_impl_clock_gettime(long res, long which_clock,
+ long tp);
+void __sanitizer_syscall_pre_impl_clock_adjtime(long which_clock, long tx);
+void __sanitizer_syscall_post_impl_clock_adjtime(long res, long which_clock,
+ long tx);
+void __sanitizer_syscall_pre_impl_clock_getres(long which_clock, long tp);
+void __sanitizer_syscall_post_impl_clock_getres(long res, long which_clock,
+ long tp);
+void __sanitizer_syscall_pre_impl_clock_nanosleep(long which_clock, long flags,
+ long rqtp, long rmtp);
+void __sanitizer_syscall_post_impl_clock_nanosleep(long res, long which_clock,
+ long flags, long rqtp,
+ long rmtp);
+void __sanitizer_syscall_pre_impl_nice(long increment);
+void __sanitizer_syscall_post_impl_nice(long res, long increment);
+void __sanitizer_syscall_pre_impl_sched_setscheduler(long pid, long policy,
+ long param);
+void __sanitizer_syscall_post_impl_sched_setscheduler(long res, long pid,
+ long policy, long param);
+void __sanitizer_syscall_pre_impl_sched_setparam(long pid, long param);
+void __sanitizer_syscall_post_impl_sched_setparam(long res, long pid,
+ long param);
+void __sanitizer_syscall_pre_impl_sched_getscheduler(long pid);
+void __sanitizer_syscall_post_impl_sched_getscheduler(long res, long pid);
+void __sanitizer_syscall_pre_impl_sched_getparam(long pid, long param);
+void __sanitizer_syscall_post_impl_sched_getparam(long res, long pid,
+ long param);
+void __sanitizer_syscall_pre_impl_sched_setaffinity(long pid, long len,
+ long user_mask_ptr);
+void __sanitizer_syscall_post_impl_sched_setaffinity(long res, long pid,
+ long len,
+ long user_mask_ptr);
+void __sanitizer_syscall_pre_impl_sched_getaffinity(long pid, long len,
+ long user_mask_ptr);
+void __sanitizer_syscall_post_impl_sched_getaffinity(long res, long pid,
+ long len,
+ long user_mask_ptr);
+void __sanitizer_syscall_pre_impl_sched_yield();
+void __sanitizer_syscall_post_impl_sched_yield(long res);
+void __sanitizer_syscall_pre_impl_sched_get_priority_max(long policy);
+void __sanitizer_syscall_post_impl_sched_get_priority_max(long res,
+ long policy);
+void __sanitizer_syscall_pre_impl_sched_get_priority_min(long policy);
+void __sanitizer_syscall_post_impl_sched_get_priority_min(long res,
+ long policy);
+void __sanitizer_syscall_pre_impl_sched_rr_get_interval(long pid,
+ long interval);
+void __sanitizer_syscall_post_impl_sched_rr_get_interval(long res, long pid,
+ long interval);
+void __sanitizer_syscall_pre_impl_setpriority(long which, long who,
+ long niceval);
+void __sanitizer_syscall_post_impl_setpriority(long res, long which, long who,
+ long niceval);
+void __sanitizer_syscall_pre_impl_getpriority(long which, long who);
+void __sanitizer_syscall_post_impl_getpriority(long res, long which, long who);
+void __sanitizer_syscall_pre_impl_shutdown(long arg0, long arg1);
+void __sanitizer_syscall_post_impl_shutdown(long res, long arg0, long arg1);
+void __sanitizer_syscall_pre_impl_reboot(long magic1, long magic2, long cmd,
+ long arg);
+void __sanitizer_syscall_post_impl_reboot(long res, long magic1, long magic2,
+ long cmd, long arg);
+void __sanitizer_syscall_pre_impl_restart_syscall();
+void __sanitizer_syscall_post_impl_restart_syscall(long res);
+void __sanitizer_syscall_pre_impl_kexec_load(long entry, long nr_segments,
+ long segments, long flags);
+void __sanitizer_syscall_post_impl_kexec_load(long res, long entry,
+ long nr_segments, long segments,
+ long flags);
+void __sanitizer_syscall_pre_impl_exit(long error_code);
+void __sanitizer_syscall_post_impl_exit(long res, long error_code);
+void __sanitizer_syscall_pre_impl_exit_group(long error_code);
+void __sanitizer_syscall_post_impl_exit_group(long res, long error_code);
+void __sanitizer_syscall_pre_impl_wait4(long pid, long stat_addr, long options,
+ long ru);
+void __sanitizer_syscall_post_impl_wait4(long res, long pid, long stat_addr,
+ long options, long ru);
+void __sanitizer_syscall_pre_impl_waitid(long which, long pid, long infop,
+ long options, long ru);
+void __sanitizer_syscall_post_impl_waitid(long res, long which, long pid,
+ long infop, long options, long ru);
+void __sanitizer_syscall_pre_impl_waitpid(long pid, long stat_addr,
+ long options);
+void __sanitizer_syscall_post_impl_waitpid(long res, long pid, long stat_addr,
+ long options);
+void __sanitizer_syscall_pre_impl_set_tid_address(long tidptr);
+void __sanitizer_syscall_post_impl_set_tid_address(long res, long tidptr);
+void __sanitizer_syscall_pre_impl_init_module(long umod, long len, long uargs);
+void __sanitizer_syscall_post_impl_init_module(long res, long umod, long len,
+ long uargs);
+void __sanitizer_syscall_pre_impl_delete_module(long name_user, long flags);
+void __sanitizer_syscall_post_impl_delete_module(long res, long name_user,
+ long flags);
+void __sanitizer_syscall_pre_impl_rt_sigprocmask(long how, long set, long oset,
+ long sigsetsize);
+void __sanitizer_syscall_post_impl_rt_sigprocmask(long res, long how, long set,
+ long oset, long sigsetsize);
+void __sanitizer_syscall_pre_impl_rt_sigpending(long set, long sigsetsize);
+void __sanitizer_syscall_post_impl_rt_sigpending(long res, long set,
+ long sigsetsize);
+void __sanitizer_syscall_pre_impl_rt_sigtimedwait(long uthese, long uinfo,
+ long uts, long sigsetsize);
+void __sanitizer_syscall_post_impl_rt_sigtimedwait(long res, long uthese,
+ long uinfo, long uts,
+ long sigsetsize);
+void __sanitizer_syscall_pre_impl_rt_tgsigqueueinfo(long tgid, long pid,
+ long sig, long uinfo);
+void __sanitizer_syscall_post_impl_rt_tgsigqueueinfo(long res, long tgid,
+ long pid, long sig,
+ long uinfo);
+void __sanitizer_syscall_pre_impl_kill(long pid, long sig);
+void __sanitizer_syscall_post_impl_kill(long res, long pid, long sig);
+void __sanitizer_syscall_pre_impl_tgkill(long tgid, long pid, long sig);
+void __sanitizer_syscall_post_impl_tgkill(long res, long tgid, long pid,
+ long sig);
+void __sanitizer_syscall_pre_impl_tkill(long pid, long sig);
+void __sanitizer_syscall_post_impl_tkill(long res, long pid, long sig);
+void __sanitizer_syscall_pre_impl_rt_sigqueueinfo(long pid, long sig,
+ long uinfo);
+void __sanitizer_syscall_post_impl_rt_sigqueueinfo(long res, long pid, long sig,
+ long uinfo);
+void __sanitizer_syscall_pre_impl_sgetmask();
+void __sanitizer_syscall_post_impl_sgetmask(long res);
+void __sanitizer_syscall_pre_impl_ssetmask(long newmask);
+void __sanitizer_syscall_post_impl_ssetmask(long res, long newmask);
+void __sanitizer_syscall_pre_impl_signal(long sig, long handler);
+void __sanitizer_syscall_post_impl_signal(long res, long sig, long handler);
+void __sanitizer_syscall_pre_impl_pause();
+void __sanitizer_syscall_post_impl_pause(long res);
+void __sanitizer_syscall_pre_impl_sync();
+void __sanitizer_syscall_post_impl_sync(long res);
+void __sanitizer_syscall_pre_impl_fsync(long fd);
+void __sanitizer_syscall_post_impl_fsync(long res, long fd);
+void __sanitizer_syscall_pre_impl_fdatasync(long fd);
+void __sanitizer_syscall_post_impl_fdatasync(long res, long fd);
+void __sanitizer_syscall_pre_impl_bdflush(long func, long data);
+void __sanitizer_syscall_post_impl_bdflush(long res, long func, long data);
+void __sanitizer_syscall_pre_impl_mount(long dev_name, long dir_name, long type,
+ long flags, long data);
+void __sanitizer_syscall_post_impl_mount(long res, long dev_name, long dir_name,
+ long type, long flags, long data);
+void __sanitizer_syscall_pre_impl_umount(long name, long flags);
+void __sanitizer_syscall_post_impl_umount(long res, long name, long flags);
+void __sanitizer_syscall_pre_impl_oldumount(long name);
+void __sanitizer_syscall_post_impl_oldumount(long res, long name);
+void __sanitizer_syscall_pre_impl_truncate(long path, long length);
+void __sanitizer_syscall_post_impl_truncate(long res, long path, long length);
+void __sanitizer_syscall_pre_impl_ftruncate(long fd, long length);
+void __sanitizer_syscall_post_impl_ftruncate(long res, long fd, long length);
+void __sanitizer_syscall_pre_impl_stat(long filename, long statbuf);
+void __sanitizer_syscall_post_impl_stat(long res, long filename, long statbuf);
+void __sanitizer_syscall_pre_impl_statfs(long path, long buf);
+void __sanitizer_syscall_post_impl_statfs(long res, long path, long buf);
+void __sanitizer_syscall_pre_impl_statfs64(long path, long sz, long buf);
+void __sanitizer_syscall_post_impl_statfs64(long res, long path, long sz,
+ long buf);
+void __sanitizer_syscall_pre_impl_fstatfs(long fd, long buf);
+void __sanitizer_syscall_post_impl_fstatfs(long res, long fd, long buf);
+void __sanitizer_syscall_pre_impl_fstatfs64(long fd, long sz, long buf);
+void __sanitizer_syscall_post_impl_fstatfs64(long res, long fd, long sz,
+ long buf);
+void __sanitizer_syscall_pre_impl_lstat(long filename, long statbuf);
+void __sanitizer_syscall_post_impl_lstat(long res, long filename, long statbuf);
+void __sanitizer_syscall_pre_impl_fstat(long fd, long statbuf);
+void __sanitizer_syscall_post_impl_fstat(long res, long fd, long statbuf);
+void __sanitizer_syscall_pre_impl_newstat(long filename, long statbuf);
+void __sanitizer_syscall_post_impl_newstat(long res, long filename,
+ long statbuf);
+void __sanitizer_syscall_pre_impl_newlstat(long filename, long statbuf);
+void __sanitizer_syscall_post_impl_newlstat(long res, long filename,
+ long statbuf);
+void __sanitizer_syscall_pre_impl_newfstat(long fd, long statbuf);
+void __sanitizer_syscall_post_impl_newfstat(long res, long fd, long statbuf);
+void __sanitizer_syscall_pre_impl_ustat(long dev, long ubuf);
+void __sanitizer_syscall_post_impl_ustat(long res, long dev, long ubuf);
+void __sanitizer_syscall_pre_impl_stat64(long filename, long statbuf);
+void __sanitizer_syscall_post_impl_stat64(long res, long filename,
+ long statbuf);
+void __sanitizer_syscall_pre_impl_fstat64(long fd, long statbuf);
+void __sanitizer_syscall_post_impl_fstat64(long res, long fd, long statbuf);
+void __sanitizer_syscall_pre_impl_lstat64(long filename, long statbuf);
+void __sanitizer_syscall_post_impl_lstat64(long res, long filename,
+ long statbuf);
+void __sanitizer_syscall_pre_impl_setxattr(long path, long name, long value,
+ long size, long flags);
+void __sanitizer_syscall_post_impl_setxattr(long res, long path, long name,
+ long value, long size, long flags);
+void __sanitizer_syscall_pre_impl_lsetxattr(long path, long name, long value,
+ long size, long flags);
+void __sanitizer_syscall_post_impl_lsetxattr(long res, long path, long name,
+ long value, long size, long flags);
+void __sanitizer_syscall_pre_impl_fsetxattr(long fd, long name, long value,
+ long size, long flags);
+void __sanitizer_syscall_post_impl_fsetxattr(long res, long fd, long name,
+ long value, long size, long flags);
+void __sanitizer_syscall_pre_impl_getxattr(long path, long name, long value,
+ long size);
+void __sanitizer_syscall_post_impl_getxattr(long res, long path, long name,
+ long value, long size);
+void __sanitizer_syscall_pre_impl_lgetxattr(long path, long name, long value,
+ long size);
+void __sanitizer_syscall_post_impl_lgetxattr(long res, long path, long name,
+ long value, long size);
+void __sanitizer_syscall_pre_impl_fgetxattr(long fd, long name, long value,
+ long size);
+void __sanitizer_syscall_post_impl_fgetxattr(long res, long fd, long name,
+ long value, long size);
+void __sanitizer_syscall_pre_impl_listxattr(long path, long list, long size);
+void __sanitizer_syscall_post_impl_listxattr(long res, long path, long list,
+ long size);
+void __sanitizer_syscall_pre_impl_llistxattr(long path, long list, long size);
+void __sanitizer_syscall_post_impl_llistxattr(long res, long path, long list,
+ long size);
+void __sanitizer_syscall_pre_impl_flistxattr(long fd, long list, long size);
+void __sanitizer_syscall_post_impl_flistxattr(long res, long fd, long list,
+ long size);
+void __sanitizer_syscall_pre_impl_removexattr(long path, long name);
+void __sanitizer_syscall_post_impl_removexattr(long res, long path, long name);
+void __sanitizer_syscall_pre_impl_lremovexattr(long path, long name);
+void __sanitizer_syscall_post_impl_lremovexattr(long res, long path, long name);
+void __sanitizer_syscall_pre_impl_fremovexattr(long fd, long name);
+void __sanitizer_syscall_post_impl_fremovexattr(long res, long fd, long name);
+void __sanitizer_syscall_pre_impl_brk(long brk);
+void __sanitizer_syscall_post_impl_brk(long res, long brk);
+void __sanitizer_syscall_pre_impl_mprotect(long start, long len, long prot);
+void __sanitizer_syscall_post_impl_mprotect(long res, long start, long len,
+ long prot);
+void __sanitizer_syscall_pre_impl_mremap(long addr, long old_len, long new_len,
+ long flags, long new_addr);
+void __sanitizer_syscall_post_impl_mremap(long res, long addr, long old_len,
+ long new_len, long flags,
+ long new_addr);
+void __sanitizer_syscall_pre_impl_remap_file_pages(long start, long size,
+ long prot, long pgoff,
+ long flags);
+void __sanitizer_syscall_post_impl_remap_file_pages(long res, long start,
+ long size, long prot,
+ long pgoff, long flags);
+void __sanitizer_syscall_pre_impl_msync(long start, long len, long flags);
+void __sanitizer_syscall_post_impl_msync(long res, long start, long len,
+ long flags);
+void __sanitizer_syscall_pre_impl_munmap(long addr, long len);
+void __sanitizer_syscall_post_impl_munmap(long res, long addr, long len);
+void __sanitizer_syscall_pre_impl_mlock(long start, long len);
+void __sanitizer_syscall_post_impl_mlock(long res, long start, long len);
+void __sanitizer_syscall_pre_impl_munlock(long start, long len);
+void __sanitizer_syscall_post_impl_munlock(long res, long start, long len);
+void __sanitizer_syscall_pre_impl_mlockall(long flags);
+void __sanitizer_syscall_post_impl_mlockall(long res, long flags);
+void __sanitizer_syscall_pre_impl_munlockall();
+void __sanitizer_syscall_post_impl_munlockall(long res);
+void __sanitizer_syscall_pre_impl_madvise(long start, long len, long behavior);
+void __sanitizer_syscall_post_impl_madvise(long res, long start, long len,
+ long behavior);
+void __sanitizer_syscall_pre_impl_mincore(long start, long len, long vec);
+void __sanitizer_syscall_post_impl_mincore(long res, long start, long len,
+ long vec);
+void __sanitizer_syscall_pre_impl_pivot_root(long new_root, long put_old);
+void __sanitizer_syscall_post_impl_pivot_root(long res, long new_root,
+ long put_old);
+void __sanitizer_syscall_pre_impl_chroot(long filename);
+void __sanitizer_syscall_post_impl_chroot(long res, long filename);
+void __sanitizer_syscall_pre_impl_mknod(long filename, long mode, long dev);
+void __sanitizer_syscall_post_impl_mknod(long res, long filename, long mode,
+ long dev);
+void __sanitizer_syscall_pre_impl_link(long oldname, long newname);
+void __sanitizer_syscall_post_impl_link(long res, long oldname, long newname);
+void __sanitizer_syscall_pre_impl_symlink(long old, long new_);
+void __sanitizer_syscall_post_impl_symlink(long res, long old, long new_);
+void __sanitizer_syscall_pre_impl_unlink(long pathname);
+void __sanitizer_syscall_post_impl_unlink(long res, long pathname);
+void __sanitizer_syscall_pre_impl_rename(long oldname, long newname);
+void __sanitizer_syscall_post_impl_rename(long res, long oldname, long newname);
+void __sanitizer_syscall_pre_impl_chmod(long filename, long mode);
+void __sanitizer_syscall_post_impl_chmod(long res, long filename, long mode);
+void __sanitizer_syscall_pre_impl_fchmod(long fd, long mode);
+void __sanitizer_syscall_post_impl_fchmod(long res, long fd, long mode);
+void __sanitizer_syscall_pre_impl_fcntl(long fd, long cmd, long arg);
+void __sanitizer_syscall_post_impl_fcntl(long res, long fd, long cmd, long arg);
+void __sanitizer_syscall_pre_impl_fcntl64(long fd, long cmd, long arg);
+void __sanitizer_syscall_post_impl_fcntl64(long res, long fd, long cmd,
+ long arg);
+void __sanitizer_syscall_pre_impl_pipe(long fildes);
+void __sanitizer_syscall_post_impl_pipe(long res, long fildes);
+void __sanitizer_syscall_pre_impl_pipe2(long fildes, long flags);
+void __sanitizer_syscall_post_impl_pipe2(long res, long fildes, long flags);
+void __sanitizer_syscall_pre_impl_dup(long fildes);
+void __sanitizer_syscall_post_impl_dup(long res, long fildes);
+void __sanitizer_syscall_pre_impl_dup2(long oldfd, long newfd);
+void __sanitizer_syscall_post_impl_dup2(long res, long oldfd, long newfd);
+void __sanitizer_syscall_pre_impl_dup3(long oldfd, long newfd, long flags);
+void __sanitizer_syscall_post_impl_dup3(long res, long oldfd, long newfd,
+ long flags);
+void __sanitizer_syscall_pre_impl_ioperm(long from, long num, long on);
+void __sanitizer_syscall_post_impl_ioperm(long res, long from, long num,
+ long on);
+void __sanitizer_syscall_pre_impl_ioctl(long fd, long cmd, long arg);
+void __sanitizer_syscall_post_impl_ioctl(long res, long fd, long cmd, long arg);
+void __sanitizer_syscall_pre_impl_flock(long fd, long cmd);
+void __sanitizer_syscall_post_impl_flock(long res, long fd, long cmd);
+void __sanitizer_syscall_pre_impl_io_setup(long nr_reqs, long ctx);
+void __sanitizer_syscall_post_impl_io_setup(long res, long nr_reqs, long ctx);
+void __sanitizer_syscall_pre_impl_io_destroy(long ctx);
+void __sanitizer_syscall_post_impl_io_destroy(long res, long ctx);
+void __sanitizer_syscall_pre_impl_io_getevents(long ctx_id, long min_nr,
+ long nr, long events,
+ long timeout);
+void __sanitizer_syscall_post_impl_io_getevents(long res, long ctx_id,
+ long min_nr, long nr,
+ long events, long timeout);
+void __sanitizer_syscall_pre_impl_io_submit(long ctx_id, long arg1, long arg2);
+void __sanitizer_syscall_post_impl_io_submit(long res, long ctx_id, long arg1,
+ long arg2);
+void __sanitizer_syscall_pre_impl_io_cancel(long ctx_id, long iocb,
+ long result);
+void __sanitizer_syscall_post_impl_io_cancel(long res, long ctx_id, long iocb,
+ long result);
+void __sanitizer_syscall_pre_impl_sendfile(long out_fd, long in_fd, long offset,
+ long count);
+void __sanitizer_syscall_post_impl_sendfile(long res, long out_fd, long in_fd,
+ long offset, long count);
+void __sanitizer_syscall_pre_impl_sendfile64(long out_fd, long in_fd,
+ long offset, long count);
+void __sanitizer_syscall_post_impl_sendfile64(long res, long out_fd, long in_fd,
+ long offset, long count);
+void __sanitizer_syscall_pre_impl_readlink(long path, long buf, long bufsiz);
+void __sanitizer_syscall_post_impl_readlink(long res, long path, long buf,
+ long bufsiz);
+void __sanitizer_syscall_pre_impl_creat(long pathname, long mode);
+void __sanitizer_syscall_post_impl_creat(long res, long pathname, long mode);
+void __sanitizer_syscall_pre_impl_open(long filename, long flags, long mode);
+void __sanitizer_syscall_post_impl_open(long res, long filename, long flags,
+ long mode);
+void __sanitizer_syscall_pre_impl_close(long fd);
+void __sanitizer_syscall_post_impl_close(long res, long fd);
+void __sanitizer_syscall_pre_impl_access(long filename, long mode);
+void __sanitizer_syscall_post_impl_access(long res, long filename, long mode);
+void __sanitizer_syscall_pre_impl_vhangup();
+void __sanitizer_syscall_post_impl_vhangup(long res);
+void __sanitizer_syscall_pre_impl_chown(long filename, long user, long group);
+void __sanitizer_syscall_post_impl_chown(long res, long filename, long user,
+ long group);
+void __sanitizer_syscall_pre_impl_lchown(long filename, long user, long group);
+void __sanitizer_syscall_post_impl_lchown(long res, long filename, long user,
+ long group);
+void __sanitizer_syscall_pre_impl_fchown(long fd, long user, long group);
+void __sanitizer_syscall_post_impl_fchown(long res, long fd, long user,
+ long group);
+void __sanitizer_syscall_pre_impl_chown16(long filename, long user, long group);
+void __sanitizer_syscall_post_impl_chown16(long res, long filename, long user,
+ long group);
+void __sanitizer_syscall_pre_impl_lchown16(long filename, long user,
+ long group);
+void __sanitizer_syscall_post_impl_lchown16(long res, long filename, long user,
+ long group);
+void __sanitizer_syscall_pre_impl_fchown16(long fd, long user, long group);
+void __sanitizer_syscall_post_impl_fchown16(long res, long fd, long user,
+ long group);
+void __sanitizer_syscall_pre_impl_setregid16(long rgid, long egid);
+void __sanitizer_syscall_post_impl_setregid16(long res, long rgid, long egid);
+void __sanitizer_syscall_pre_impl_setgid16(long gid);
+void __sanitizer_syscall_post_impl_setgid16(long res, long gid);
+void __sanitizer_syscall_pre_impl_setreuid16(long ruid, long euid);
+void __sanitizer_syscall_post_impl_setreuid16(long res, long ruid, long euid);
+void __sanitizer_syscall_pre_impl_setuid16(long uid);
+void __sanitizer_syscall_post_impl_setuid16(long res, long uid);
+void __sanitizer_syscall_pre_impl_setresuid16(long ruid, long euid, long suid);
+void __sanitizer_syscall_post_impl_setresuid16(long res, long ruid, long euid,
+ long suid);
+void __sanitizer_syscall_pre_impl_getresuid16(long ruid, long euid, long suid);
+void __sanitizer_syscall_post_impl_getresuid16(long res, long ruid, long euid,
+ long suid);
+void __sanitizer_syscall_pre_impl_setresgid16(long rgid, long egid, long sgid);
+void __sanitizer_syscall_post_impl_setresgid16(long res, long rgid, long egid,
+ long sgid);
+void __sanitizer_syscall_pre_impl_getresgid16(long rgid, long egid, long sgid);
+void __sanitizer_syscall_post_impl_getresgid16(long res, long rgid, long egid,
+ long sgid);
+void __sanitizer_syscall_pre_impl_setfsuid16(long uid);
+void __sanitizer_syscall_post_impl_setfsuid16(long res, long uid);
+void __sanitizer_syscall_pre_impl_setfsgid16(long gid);
+void __sanitizer_syscall_post_impl_setfsgid16(long res, long gid);
+void __sanitizer_syscall_pre_impl_getgroups16(long gidsetsize, long grouplist);
+void __sanitizer_syscall_post_impl_getgroups16(long res, long gidsetsize,
+ long grouplist);
+void __sanitizer_syscall_pre_impl_setgroups16(long gidsetsize, long grouplist);
+void __sanitizer_syscall_post_impl_setgroups16(long res, long gidsetsize,
+ long grouplist);
+void __sanitizer_syscall_pre_impl_getuid16();
+void __sanitizer_syscall_post_impl_getuid16(long res);
+void __sanitizer_syscall_pre_impl_geteuid16();
+void __sanitizer_syscall_post_impl_geteuid16(long res);
+void __sanitizer_syscall_pre_impl_getgid16();
+void __sanitizer_syscall_post_impl_getgid16(long res);
+void __sanitizer_syscall_pre_impl_getegid16();
+void __sanitizer_syscall_post_impl_getegid16(long res);
+void __sanitizer_syscall_pre_impl_utime(long filename, long times);
+void __sanitizer_syscall_post_impl_utime(long res, long filename, long times);
+void __sanitizer_syscall_pre_impl_utimes(long filename, long utimes);
+void __sanitizer_syscall_post_impl_utimes(long res, long filename, long utimes);
+void __sanitizer_syscall_pre_impl_lseek(long fd, long offset, long origin);
+void __sanitizer_syscall_post_impl_lseek(long res, long fd, long offset,
+ long origin);
+void __sanitizer_syscall_pre_impl_llseek(long fd, long offset_high,
+ long offset_low, long result,
+ long origin);
+void __sanitizer_syscall_post_impl_llseek(long res, long fd, long offset_high,
+ long offset_low, long result,
+ long origin);
+void __sanitizer_syscall_pre_impl_read(long fd, long buf, long count);
+void __sanitizer_syscall_post_impl_read(long res, long fd, long buf,
+ long count);
+void __sanitizer_syscall_pre_impl_readv(long fd, long vec, long vlen);
+void __sanitizer_syscall_post_impl_readv(long res, long fd, long vec,
+ long vlen);
+void __sanitizer_syscall_pre_impl_write(long fd, long buf, long count);
+void __sanitizer_syscall_post_impl_write(long res, long fd, long buf,
+ long count);
+void __sanitizer_syscall_pre_impl_writev(long fd, long vec, long vlen);
+void __sanitizer_syscall_post_impl_writev(long res, long fd, long vec,
+ long vlen);
+
+#ifdef _LP64
+void __sanitizer_syscall_pre_impl_pread64(long fd, long buf, long count,
+ long pos);
+void __sanitizer_syscall_post_impl_pread64(long res, long fd, long buf,
+ long count, long pos);
+void __sanitizer_syscall_pre_impl_pwrite64(long fd, long buf, long count,
+ long pos);
+void __sanitizer_syscall_post_impl_pwrite64(long res, long fd, long buf,
+ long count, long pos);
+#else
+void __sanitizer_syscall_pre_impl_pread64(long fd, long buf, long count,
+ long pos0, long pos1);
+void __sanitizer_syscall_post_impl_pread64(long res, long fd, long buf,
+ long count, long pos0, long pos1);
+void __sanitizer_syscall_pre_impl_pwrite64(long fd, long buf, long count,
+ long pos0, long pos1);
+void __sanitizer_syscall_post_impl_pwrite64(long res, long fd, long buf,
+ long count, long pos0, long pos1);
+#endif
+
+void __sanitizer_syscall_pre_impl_preadv(long fd, long vec, long vlen,
+ long pos_l, long pos_h);
+void __sanitizer_syscall_post_impl_preadv(long res, long fd, long vec,
+ long vlen, long pos_l, long pos_h);
+void __sanitizer_syscall_pre_impl_pwritev(long fd, long vec, long vlen,
+ long pos_l, long pos_h);
+void __sanitizer_syscall_post_impl_pwritev(long res, long fd, long vec,
+ long vlen, long pos_l, long pos_h);
+void __sanitizer_syscall_pre_impl_getcwd(long buf, long size);
+void __sanitizer_syscall_post_impl_getcwd(long res, long buf, long size);
+void __sanitizer_syscall_pre_impl_mkdir(long pathname, long mode);
+void __sanitizer_syscall_post_impl_mkdir(long res, long pathname, long mode);
+void __sanitizer_syscall_pre_impl_chdir(long filename);
+void __sanitizer_syscall_post_impl_chdir(long res, long filename);
+void __sanitizer_syscall_pre_impl_fchdir(long fd);
+void __sanitizer_syscall_post_impl_fchdir(long res, long fd);
+void __sanitizer_syscall_pre_impl_rmdir(long pathname);
+void __sanitizer_syscall_post_impl_rmdir(long res, long pathname);
+void __sanitizer_syscall_pre_impl_lookup_dcookie(long cookie64, long buf,
+ long len);
+void __sanitizer_syscall_post_impl_lookup_dcookie(long res, long cookie64,
+ long buf, long len);
+void __sanitizer_syscall_pre_impl_quotactl(long cmd, long special, long id,
+ long addr);
+void __sanitizer_syscall_post_impl_quotactl(long res, long cmd, long special,
+ long id, long addr);
+void __sanitizer_syscall_pre_impl_getdents(long fd, long dirent, long count);
+void __sanitizer_syscall_post_impl_getdents(long res, long fd, long dirent,
+ long count);
+void __sanitizer_syscall_pre_impl_getdents64(long fd, long dirent, long count);
+void __sanitizer_syscall_post_impl_getdents64(long res, long fd, long dirent,
+ long count);
+void __sanitizer_syscall_pre_impl_setsockopt(long fd, long level, long optname,
+ long optval, long optlen);
+void __sanitizer_syscall_post_impl_setsockopt(long res, long fd, long level,
+ long optname, long optval,
+ long optlen);
+void __sanitizer_syscall_pre_impl_getsockopt(long fd, long level, long optname,
+ long optval, long optlen);
+void __sanitizer_syscall_post_impl_getsockopt(long res, long fd, long level,
+ long optname, long optval,
+ long optlen);
+void __sanitizer_syscall_pre_impl_bind(long arg0, long arg1, long arg2);
+void __sanitizer_syscall_post_impl_bind(long res, long arg0, long arg1,
+ long arg2);
+void __sanitizer_syscall_pre_impl_connect(long arg0, long arg1, long arg2);
+void __sanitizer_syscall_post_impl_connect(long res, long arg0, long arg1,
+ long arg2);
+void __sanitizer_syscall_pre_impl_accept(long arg0, long arg1, long arg2);
+void __sanitizer_syscall_post_impl_accept(long res, long arg0, long arg1,
+ long arg2);
+void __sanitizer_syscall_pre_impl_accept4(long arg0, long arg1, long arg2,
+ long arg3);
+void __sanitizer_syscall_post_impl_accept4(long res, long arg0, long arg1,
+ long arg2, long arg3);
+void __sanitizer_syscall_pre_impl_getsockname(long arg0, long arg1, long arg2);
+void __sanitizer_syscall_post_impl_getsockname(long res, long arg0, long arg1,
+ long arg2);
+void __sanitizer_syscall_pre_impl_getpeername(long arg0, long arg1, long arg2);
+void __sanitizer_syscall_post_impl_getpeername(long res, long arg0, long arg1,
+ long arg2);
+void __sanitizer_syscall_pre_impl_send(long arg0, long arg1, long arg2,
+ long arg3);
+void __sanitizer_syscall_post_impl_send(long res, long arg0, long arg1,
+ long arg2, long arg3);
+void __sanitizer_syscall_pre_impl_sendto(long arg0, long arg1, long arg2,
+ long arg3, long arg4, long arg5);
+void __sanitizer_syscall_post_impl_sendto(long res, long arg0, long arg1,
+ long arg2, long arg3, long arg4,
+ long arg5);
+void __sanitizer_syscall_pre_impl_sendmsg(long fd, long msg, long flags);
+void __sanitizer_syscall_post_impl_sendmsg(long res, long fd, long msg,
+ long flags);
+void __sanitizer_syscall_pre_impl_sendmmsg(long fd, long msg, long vlen,
+ long flags);
+void __sanitizer_syscall_post_impl_sendmmsg(long res, long fd, long msg,
+ long vlen, long flags);
+void __sanitizer_syscall_pre_impl_recv(long arg0, long arg1, long arg2,
+ long arg3);
+void __sanitizer_syscall_post_impl_recv(long res, long arg0, long arg1,
+ long arg2, long arg3);
+void __sanitizer_syscall_pre_impl_recvfrom(long arg0, long arg1, long arg2,
+ long arg3, long arg4, long arg5);
+void __sanitizer_syscall_post_impl_recvfrom(long res, long arg0, long arg1,
+ long arg2, long arg3, long arg4,
+ long arg5);
+void __sanitizer_syscall_pre_impl_recvmsg(long fd, long msg, long flags);
+void __sanitizer_syscall_post_impl_recvmsg(long res, long fd, long msg,
+ long flags);
+void __sanitizer_syscall_pre_impl_recvmmsg(long fd, long msg, long vlen,
+ long flags, long timeout);
+void __sanitizer_syscall_post_impl_recvmmsg(long res, long fd, long msg,
+ long vlen, long flags,
+ long timeout);
+void __sanitizer_syscall_pre_impl_socket(long arg0, long arg1, long arg2);
+void __sanitizer_syscall_post_impl_socket(long res, long arg0, long arg1,
+ long arg2);
+void __sanitizer_syscall_pre_impl_socketpair(long arg0, long arg1, long arg2,
+ long arg3);
+void __sanitizer_syscall_post_impl_socketpair(long res, long arg0, long arg1,
+ long arg2, long arg3);
+void __sanitizer_syscall_pre_impl_socketcall(long call, long args);
+void __sanitizer_syscall_post_impl_socketcall(long res, long call, long args);
+void __sanitizer_syscall_pre_impl_listen(long arg0, long arg1);
+void __sanitizer_syscall_post_impl_listen(long res, long arg0, long arg1);
+void __sanitizer_syscall_pre_impl_poll(long ufds, long nfds, long timeout);
+void __sanitizer_syscall_post_impl_poll(long res, long ufds, long nfds,
+ long timeout);
+void __sanitizer_syscall_pre_impl_select(long n, long inp, long outp, long exp,
+ long tvp);
+void __sanitizer_syscall_post_impl_select(long res, long n, long inp, long outp,
+ long exp, long tvp);
+void __sanitizer_syscall_pre_impl_old_select(long arg);
+void __sanitizer_syscall_post_impl_old_select(long res, long arg);
+void __sanitizer_syscall_pre_impl_epoll_create(long size);
+void __sanitizer_syscall_post_impl_epoll_create(long res, long size);
+void __sanitizer_syscall_pre_impl_epoll_create1(long flags);
+void __sanitizer_syscall_post_impl_epoll_create1(long res, long flags);
+void __sanitizer_syscall_pre_impl_epoll_ctl(long epfd, long op, long fd,
+ long event);
+void __sanitizer_syscall_post_impl_epoll_ctl(long res, long epfd, long op,
+ long fd, long event);
+void __sanitizer_syscall_pre_impl_epoll_wait(long epfd, long events,
+ long maxevents, long timeout);
+void __sanitizer_syscall_post_impl_epoll_wait(long res, long epfd, long events,
+ long maxevents, long timeout);
+void __sanitizer_syscall_pre_impl_epoll_pwait(long epfd, long events,
+ long maxevents, long timeout,
+ long sigmask, long sigsetsize);
+void __sanitizer_syscall_post_impl_epoll_pwait(long res, long epfd, long events,
+ long maxevents, long timeout,
+ long sigmask, long sigsetsize);
+void __sanitizer_syscall_pre_impl_gethostname(long name, long len);
+void __sanitizer_syscall_post_impl_gethostname(long res, long name, long len);
+void __sanitizer_syscall_pre_impl_sethostname(long name, long len);
+void __sanitizer_syscall_post_impl_sethostname(long res, long name, long len);
+void __sanitizer_syscall_pre_impl_setdomainname(long name, long len);
+void __sanitizer_syscall_post_impl_setdomainname(long res, long name, long len);
+void __sanitizer_syscall_pre_impl_newuname(long name);
+void __sanitizer_syscall_post_impl_newuname(long res, long name);
+void __sanitizer_syscall_pre_impl_uname(long arg0);
+void __sanitizer_syscall_post_impl_uname(long res, long arg0);
+void __sanitizer_syscall_pre_impl_olduname(long arg0);
+void __sanitizer_syscall_post_impl_olduname(long res, long arg0);
+void __sanitizer_syscall_pre_impl_getrlimit(long resource, long rlim);
+void __sanitizer_syscall_post_impl_getrlimit(long res, long resource,
+ long rlim);
+void __sanitizer_syscall_pre_impl_old_getrlimit(long resource, long rlim);
+void __sanitizer_syscall_post_impl_old_getrlimit(long res, long resource,
+ long rlim);
+void __sanitizer_syscall_pre_impl_setrlimit(long resource, long rlim);
+void __sanitizer_syscall_post_impl_setrlimit(long res, long resource,
+ long rlim);
+void __sanitizer_syscall_pre_impl_prlimit64(long pid, long resource,
+ long new_rlim, long old_rlim);
+void __sanitizer_syscall_post_impl_prlimit64(long res, long pid, long resource,
+ long new_rlim, long old_rlim);
+void __sanitizer_syscall_pre_impl_getrusage(long who, long ru);
+void __sanitizer_syscall_post_impl_getrusage(long res, long who, long ru);
+void __sanitizer_syscall_pre_impl_umask(long mask);
+void __sanitizer_syscall_post_impl_umask(long res, long mask);
+void __sanitizer_syscall_pre_impl_msgget(long key, long msgflg);
+void __sanitizer_syscall_post_impl_msgget(long res, long key, long msgflg);
+void __sanitizer_syscall_pre_impl_msgsnd(long msqid, long msgp, long msgsz,
+ long msgflg);
+void __sanitizer_syscall_post_impl_msgsnd(long res, long msqid, long msgp,
+ long msgsz, long msgflg);
+void __sanitizer_syscall_pre_impl_msgrcv(long msqid, long msgp, long msgsz,
+ long msgtyp, long msgflg);
+void __sanitizer_syscall_post_impl_msgrcv(long res, long msqid, long msgp,
+ long msgsz, long msgtyp, long msgflg);
+void __sanitizer_syscall_pre_impl_msgctl(long msqid, long cmd, long buf);
+void __sanitizer_syscall_post_impl_msgctl(long res, long msqid, long cmd,
+ long buf);
+void __sanitizer_syscall_pre_impl_semget(long key, long nsems, long semflg);
+void __sanitizer_syscall_post_impl_semget(long res, long key, long nsems,
+ long semflg);
+void __sanitizer_syscall_pre_impl_semop(long semid, long sops, long nsops);
+void __sanitizer_syscall_post_impl_semop(long res, long semid, long sops,
+ long nsops);
+void __sanitizer_syscall_pre_impl_semctl(long semid, long semnum, long cmd,
+ long arg);
+void __sanitizer_syscall_post_impl_semctl(long res, long semid, long semnum,
+ long cmd, long arg);
+void __sanitizer_syscall_pre_impl_semtimedop(long semid, long sops, long nsops,
+ long timeout);
+void __sanitizer_syscall_post_impl_semtimedop(long res, long semid, long sops,
+ long nsops, long timeout);
+void __sanitizer_syscall_pre_impl_shmat(long shmid, long shmaddr, long shmflg);
+void __sanitizer_syscall_post_impl_shmat(long res, long shmid, long shmaddr,
+ long shmflg);
+void __sanitizer_syscall_pre_impl_shmget(long key, long size, long flag);
+void __sanitizer_syscall_post_impl_shmget(long res, long key, long size,
+ long flag);
+void __sanitizer_syscall_pre_impl_shmdt(long shmaddr);
+void __sanitizer_syscall_post_impl_shmdt(long res, long shmaddr);
+void __sanitizer_syscall_pre_impl_shmctl(long shmid, long cmd, long buf);
+void __sanitizer_syscall_post_impl_shmctl(long res, long shmid, long cmd,
+ long buf);
+void __sanitizer_syscall_pre_impl_ipc(long call, long first, long second,
+ long third, long ptr, long fifth);
+void __sanitizer_syscall_post_impl_ipc(long res, long call, long first,
+ long second, long third, long ptr,
+ long fifth);
+void __sanitizer_syscall_pre_impl_mq_open(long name, long oflag, long mode,
+ long attr);
+void __sanitizer_syscall_post_impl_mq_open(long res, long name, long oflag,
+ long mode, long attr);
+void __sanitizer_syscall_pre_impl_mq_unlink(long name);
+void __sanitizer_syscall_post_impl_mq_unlink(long res, long name);
+void __sanitizer_syscall_pre_impl_mq_timedsend(long mqdes, long msg_ptr,
+ long msg_len, long msg_prio,
+ long abs_timeout);
+void __sanitizer_syscall_post_impl_mq_timedsend(long res, long mqdes,
+ long msg_ptr, long msg_len,
+ long msg_prio,
+ long abs_timeout);
+void __sanitizer_syscall_pre_impl_mq_timedreceive(long mqdes, long msg_ptr,
+ long msg_len, long msg_prio,
+ long abs_timeout);
+void __sanitizer_syscall_post_impl_mq_timedreceive(long res, long mqdes,
+ long msg_ptr, long msg_len,
+ long msg_prio,
+ long abs_timeout);
+void __sanitizer_syscall_pre_impl_mq_notify(long mqdes, long notification);
+void __sanitizer_syscall_post_impl_mq_notify(long res, long mqdes,
+ long notification);
+void __sanitizer_syscall_pre_impl_mq_getsetattr(long mqdes, long mqstat,
+ long omqstat);
+void __sanitizer_syscall_post_impl_mq_getsetattr(long res, long mqdes,
+ long mqstat, long omqstat);
+void __sanitizer_syscall_pre_impl_pciconfig_iobase(long which, long bus,
+ long devfn);
+void __sanitizer_syscall_post_impl_pciconfig_iobase(long res, long which,
+ long bus, long devfn);
+void __sanitizer_syscall_pre_impl_pciconfig_read(long bus, long dfn, long off,
+ long len, long buf);
+void __sanitizer_syscall_post_impl_pciconfig_read(long res, long bus, long dfn,
+ long off, long len, long buf);
+void __sanitizer_syscall_pre_impl_pciconfig_write(long bus, long dfn, long off,
+ long len, long buf);
+void __sanitizer_syscall_post_impl_pciconfig_write(long res, long bus, long dfn,
+ long off, long len,
+ long buf);
+void __sanitizer_syscall_pre_impl_swapon(long specialfile, long swap_flags);
+void __sanitizer_syscall_post_impl_swapon(long res, long specialfile,
+ long swap_flags);
+void __sanitizer_syscall_pre_impl_swapoff(long specialfile);
+void __sanitizer_syscall_post_impl_swapoff(long res, long specialfile);
+void __sanitizer_syscall_pre_impl_sysctl(long args);
+void __sanitizer_syscall_post_impl_sysctl(long res, long args);
+void __sanitizer_syscall_pre_impl_sysinfo(long info);
+void __sanitizer_syscall_post_impl_sysinfo(long res, long info);
+void __sanitizer_syscall_pre_impl_sysfs(long option, long arg1, long arg2);
+void __sanitizer_syscall_post_impl_sysfs(long res, long option, long arg1,
+ long arg2);
+void __sanitizer_syscall_pre_impl_syslog(long type, long buf, long len);
+void __sanitizer_syscall_post_impl_syslog(long res, long type, long buf,
+ long len);
+void __sanitizer_syscall_pre_impl_uselib(long library);
+void __sanitizer_syscall_post_impl_uselib(long res, long library);
+void __sanitizer_syscall_pre_impl_ni_syscall();
+void __sanitizer_syscall_post_impl_ni_syscall(long res);
+void __sanitizer_syscall_pre_impl_ptrace(long request, long pid, long addr,
+ long data);
+void __sanitizer_syscall_post_impl_ptrace(long res, long request, long pid,
+ long addr, long data);
+void __sanitizer_syscall_pre_impl_add_key(long _type, long _description,
+ long _payload, long plen,
+ long destringid);
+void __sanitizer_syscall_post_impl_add_key(long res, long _type,
+ long _description, long _payload,
+ long plen, long destringid);
+void __sanitizer_syscall_pre_impl_request_key(long _type, long _description,
+ long _callout_info,
+ long destringid);
+void __sanitizer_syscall_post_impl_request_key(long res, long _type,
+ long _description,
+ long _callout_info,
+ long destringid);
+void __sanitizer_syscall_pre_impl_keyctl(long cmd, long arg2, long arg3,
+ long arg4, long arg5);
+void __sanitizer_syscall_post_impl_keyctl(long res, long cmd, long arg2,
+ long arg3, long arg4, long arg5);
+void __sanitizer_syscall_pre_impl_ioprio_set(long which, long who, long ioprio);
+void __sanitizer_syscall_post_impl_ioprio_set(long res, long which, long who,
+ long ioprio);
+void __sanitizer_syscall_pre_impl_ioprio_get(long which, long who);
+void __sanitizer_syscall_post_impl_ioprio_get(long res, long which, long who);
+void __sanitizer_syscall_pre_impl_set_mempolicy(long mode, long nmask,
+ long maxnode);
+void __sanitizer_syscall_post_impl_set_mempolicy(long res, long mode,
+ long nmask, long maxnode);
+void __sanitizer_syscall_pre_impl_migrate_pages(long pid, long maxnode,
+ long from, long to);
+void __sanitizer_syscall_post_impl_migrate_pages(long res, long pid,
+ long maxnode, long from,
+ long to);
+void __sanitizer_syscall_pre_impl_move_pages(long pid, long nr_pages,
+ long pages, long nodes,
+ long status, long flags);
+void __sanitizer_syscall_post_impl_move_pages(long res, long pid, long nr_pages,
+ long pages, long nodes,
+ long status, long flags);
+void __sanitizer_syscall_pre_impl_mbind(long start, long len, long mode,
+ long nmask, long maxnode, long flags);
+void __sanitizer_syscall_post_impl_mbind(long res, long start, long len,
+ long mode, long nmask, long maxnode,
+ long flags);
+void __sanitizer_syscall_pre_impl_get_mempolicy(long policy, long nmask,
+ long maxnode, long addr,
+ long flags);
+void __sanitizer_syscall_post_impl_get_mempolicy(long res, long policy,
+ long nmask, long maxnode,
+ long addr, long flags);
+void __sanitizer_syscall_pre_impl_inotify_init();
+void __sanitizer_syscall_post_impl_inotify_init(long res);
+void __sanitizer_syscall_pre_impl_inotify_init1(long flags);
+void __sanitizer_syscall_post_impl_inotify_init1(long res, long flags);
+void __sanitizer_syscall_pre_impl_inotify_add_watch(long fd, long path,
+ long mask);
+void __sanitizer_syscall_post_impl_inotify_add_watch(long res, long fd,
+ long path, long mask);
+void __sanitizer_syscall_pre_impl_inotify_rm_watch(long fd, long wd);
+void __sanitizer_syscall_post_impl_inotify_rm_watch(long res, long fd, long wd);
+void __sanitizer_syscall_pre_impl_spu_run(long fd, long unpc, long ustatus);
+void __sanitizer_syscall_post_impl_spu_run(long res, long fd, long unpc,
+ long ustatus);
+void __sanitizer_syscall_pre_impl_spu_create(long name, long flags, long mode,
+ long fd);
+void __sanitizer_syscall_post_impl_spu_create(long res, long name, long flags,
+ long mode, long fd);
+void __sanitizer_syscall_pre_impl_mknodat(long dfd, long filename, long mode,
+ long dev);
+void __sanitizer_syscall_post_impl_mknodat(long res, long dfd, long filename,
+ long mode, long dev);
+void __sanitizer_syscall_pre_impl_mkdirat(long dfd, long pathname, long mode);
+void __sanitizer_syscall_post_impl_mkdirat(long res, long dfd, long pathname,
+ long mode);
+void __sanitizer_syscall_pre_impl_unlinkat(long dfd, long pathname, long flag);
+void __sanitizer_syscall_post_impl_unlinkat(long res, long dfd, long pathname,
+ long flag);
+void __sanitizer_syscall_pre_impl_symlinkat(long oldname, long newdfd,
+ long newname);
+void __sanitizer_syscall_post_impl_symlinkat(long res, long oldname,
+ long newdfd, long newname);
+void __sanitizer_syscall_pre_impl_linkat(long olddfd, long oldname, long newdfd,
+ long newname, long flags);
+void __sanitizer_syscall_post_impl_linkat(long res, long olddfd, long oldname,
+ long newdfd, long newname,
+ long flags);
+void __sanitizer_syscall_pre_impl_renameat(long olddfd, long oldname,
+ long newdfd, long newname);
+void __sanitizer_syscall_post_impl_renameat(long res, long olddfd, long oldname,
+ long newdfd, long newname);
+void __sanitizer_syscall_pre_impl_futimesat(long dfd, long filename,
+ long utimes);
+void __sanitizer_syscall_post_impl_futimesat(long res, long dfd, long filename,
+ long utimes);
+void __sanitizer_syscall_pre_impl_faccessat(long dfd, long filename, long mode);
+void __sanitizer_syscall_post_impl_faccessat(long res, long dfd, long filename,
+ long mode);
+void __sanitizer_syscall_pre_impl_fchmodat(long dfd, long filename, long mode);
+void __sanitizer_syscall_post_impl_fchmodat(long res, long dfd, long filename,
+ long mode);
+void __sanitizer_syscall_pre_impl_fchownat(long dfd, long filename, long user,
+ long group, long flag);
+void __sanitizer_syscall_post_impl_fchownat(long res, long dfd, long filename,
+ long user, long group, long flag);
+void __sanitizer_syscall_pre_impl_openat(long dfd, long filename, long flags,
+ long mode);
+void __sanitizer_syscall_post_impl_openat(long res, long dfd, long filename,
+ long flags, long mode);
+void __sanitizer_syscall_pre_impl_newfstatat(long dfd, long filename,
+ long statbuf, long flag);
+void __sanitizer_syscall_post_impl_newfstatat(long res, long dfd, long filename,
+ long statbuf, long flag);
+void __sanitizer_syscall_pre_impl_fstatat64(long dfd, long filename,
+ long statbuf, long flag);
+void __sanitizer_syscall_post_impl_fstatat64(long res, long dfd, long filename,
+ long statbuf, long flag);
+void __sanitizer_syscall_pre_impl_readlinkat(long dfd, long path, long buf,
+ long bufsiz);
+void __sanitizer_syscall_post_impl_readlinkat(long res, long dfd, long path,
+ long buf, long bufsiz);
+void __sanitizer_syscall_pre_impl_utimensat(long dfd, long filename,
+ long utimes, long flags);
+void __sanitizer_syscall_post_impl_utimensat(long res, long dfd, long filename,
+ long utimes, long flags);
+void __sanitizer_syscall_pre_impl_unshare(long unshare_flags);
+void __sanitizer_syscall_post_impl_unshare(long res, long unshare_flags);
+void __sanitizer_syscall_pre_impl_splice(long fd_in, long off_in, long fd_out,
+ long off_out, long len, long flags);
+void __sanitizer_syscall_post_impl_splice(long res, long fd_in, long off_in,
+ long fd_out, long off_out, long len,
+ long flags);
+void __sanitizer_syscall_pre_impl_vmsplice(long fd, long iov, long nr_segs,
+ long flags);
+void __sanitizer_syscall_post_impl_vmsplice(long res, long fd, long iov,
+ long nr_segs, long flags);
+void __sanitizer_syscall_pre_impl_tee(long fdin, long fdout, long len,
+ long flags);
+void __sanitizer_syscall_post_impl_tee(long res, long fdin, long fdout,
+ long len, long flags);
+void __sanitizer_syscall_pre_impl_get_robust_list(long pid, long head_ptr,
+ long len_ptr);
+void __sanitizer_syscall_post_impl_get_robust_list(long res, long pid,
+ long head_ptr, long len_ptr);
+void __sanitizer_syscall_pre_impl_set_robust_list(long head, long len);
+void __sanitizer_syscall_post_impl_set_robust_list(long res, long head,
+ long len);
+void __sanitizer_syscall_pre_impl_getcpu(long cpu, long node, long cache);
+void __sanitizer_syscall_post_impl_getcpu(long res, long cpu, long node,
+ long cache);
+void __sanitizer_syscall_pre_impl_signalfd(long ufd, long user_mask,
+ long sizemask);
+void __sanitizer_syscall_post_impl_signalfd(long res, long ufd, long user_mask,
+ long sizemask);
+void __sanitizer_syscall_pre_impl_signalfd4(long ufd, long user_mask,
+ long sizemask, long flags);
+void __sanitizer_syscall_post_impl_signalfd4(long res, long ufd, long user_mask,
+ long sizemask, long flags);
+void __sanitizer_syscall_pre_impl_timerfd_create(long clockid, long flags);
+void __sanitizer_syscall_post_impl_timerfd_create(long res, long clockid,
+ long flags);
+void __sanitizer_syscall_pre_impl_timerfd_settime(long ufd, long flags,
+ long utmr, long otmr);
+void __sanitizer_syscall_post_impl_timerfd_settime(long res, long ufd,
+ long flags, long utmr,
+ long otmr);
+void __sanitizer_syscall_pre_impl_timerfd_gettime(long ufd, long otmr);
+void __sanitizer_syscall_post_impl_timerfd_gettime(long res, long ufd,
+ long otmr);
+void __sanitizer_syscall_pre_impl_eventfd(long count);
+void __sanitizer_syscall_post_impl_eventfd(long res, long count);
+void __sanitizer_syscall_pre_impl_eventfd2(long count, long flags);
+void __sanitizer_syscall_post_impl_eventfd2(long res, long count, long flags);
+void __sanitizer_syscall_pre_impl_old_readdir(long arg0, long arg1, long arg2);
+void __sanitizer_syscall_post_impl_old_readdir(long res, long arg0, long arg1,
+ long arg2);
+void __sanitizer_syscall_pre_impl_pselect6(long arg0, long arg1, long arg2,
+ long arg3, long arg4, long arg5);
+void __sanitizer_syscall_post_impl_pselect6(long res, long arg0, long arg1,
+ long arg2, long arg3, long arg4,
+ long arg5);
+void __sanitizer_syscall_pre_impl_ppoll(long arg0, long arg1, long arg2,
+ long arg3, long arg4);
+void __sanitizer_syscall_post_impl_ppoll(long res, long arg0, long arg1,
+ long arg2, long arg3, long arg4);
+void __sanitizer_syscall_pre_impl_fanotify_init(long flags, long event_f_flags);
+void __sanitizer_syscall_post_impl_fanotify_init(long res, long flags,
+ long event_f_flags);
+void __sanitizer_syscall_pre_impl_fanotify_mark(long fanotify_fd, long flags,
+ long mask, long fd,
+ long pathname);
+void __sanitizer_syscall_post_impl_fanotify_mark(long res, long fanotify_fd,
+ long flags, long mask, long fd,
+ long pathname);
+void __sanitizer_syscall_pre_impl_syncfs(long fd);
+void __sanitizer_syscall_post_impl_syncfs(long res, long fd);
+void __sanitizer_syscall_pre_impl_perf_event_open(long attr_uptr, long pid,
+ long cpu, long group_fd,
+ long flags);
+void __sanitizer_syscall_post_impl_perf_event_open(long res, long attr_uptr,
+ long pid, long cpu,
+ long group_fd, long flags);
+void __sanitizer_syscall_pre_impl_mmap_pgoff(long addr, long len, long prot,
+ long flags, long fd, long pgoff);
+void __sanitizer_syscall_post_impl_mmap_pgoff(long res, long addr, long len,
+ long prot, long flags, long fd,
+ long pgoff);
+void __sanitizer_syscall_pre_impl_old_mmap(long arg);
+void __sanitizer_syscall_post_impl_old_mmap(long res, long arg);
+void __sanitizer_syscall_pre_impl_name_to_handle_at(long dfd, long name,
+ long handle, long mnt_id,
+ long flag);
+void __sanitizer_syscall_post_impl_name_to_handle_at(long res, long dfd,
+ long name, long handle,
+ long mnt_id, long flag);
+void __sanitizer_syscall_pre_impl_open_by_handle_at(long mountdirfd,
+ long handle, long flags);
+void __sanitizer_syscall_post_impl_open_by_handle_at(long res, long mountdirfd,
+ long handle, long flags);
+void __sanitizer_syscall_pre_impl_setns(long fd, long nstype);
+void __sanitizer_syscall_post_impl_setns(long res, long fd, long nstype);
+void __sanitizer_syscall_pre_impl_process_vm_readv(long pid, long lvec,
+ long liovcnt, long rvec,
+ long riovcnt, long flags);
+void __sanitizer_syscall_post_impl_process_vm_readv(long res, long pid,
+ long lvec, long liovcnt,
+ long rvec, long riovcnt,
+ long flags);
+void __sanitizer_syscall_pre_impl_process_vm_writev(long pid, long lvec,
+ long liovcnt, long rvec,
+ long riovcnt, long flags);
+void __sanitizer_syscall_post_impl_process_vm_writev(long res, long pid,
+ long lvec, long liovcnt,
+ long rvec, long riovcnt,
+ long flags);
+void __sanitizer_syscall_pre_impl_fork();
+void __sanitizer_syscall_post_impl_fork(long res);
+void __sanitizer_syscall_pre_impl_vfork();
+void __sanitizer_syscall_post_impl_vfork(long res);
+void __sanitizer_syscall_pre_impl_sigaction(long signum, long act, long oldact);
+void __sanitizer_syscall_post_impl_sigaction(long res, long signum, long act,
+ long oldact);
+void __sanitizer_syscall_pre_impl_rt_sigaction(long signum, long act,
+ long oldact, long sz);
+void __sanitizer_syscall_post_impl_rt_sigaction(long res, long signum, long act,
+ long oldact, long sz);
+void __sanitizer_syscall_pre_impl_sigaltstack(long ss, long oss);
+void __sanitizer_syscall_post_impl_sigaltstack(long res, long ss, long oss);
+#ifdef __cplusplus
+} // extern "C"
+#endif
+
+#endif // SANITIZER_LINUX_SYSCALL_HOOKS_H
--- /dev/null
+//===-- sanitizer/lsan_interface.h ------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file is a part of LeakSanitizer.
+//
+// Public interface header.
+//===----------------------------------------------------------------------===//
+#ifndef SANITIZER_LSAN_INTERFACE_H
+#define SANITIZER_LSAN_INTERFACE_H
+
+#include <sanitizer/common_interface_defs.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+ // Allocations made between calls to __lsan_disable() and __lsan_enable() will
+ // be treated as non-leaks. Disable/enable pairs may be nested.
+ void __lsan_disable(void);
+ void __lsan_enable(void);
+
+ // The heap object into which p points will be treated as a non-leak.
+ void __lsan_ignore_object(const void *p);
+
+ // Memory regions registered through this interface will be treated as sources
+ // of live pointers during leak checking. Useful if you store pointers in
+ // mapped memory.
+ // Points of note:
+ // - __lsan_unregister_root_region() must be called with the same pointer and
+ // size that have earlier been passed to __lsan_register_root_region()
+ // - LSan will skip any inaccessible memory when scanning a root region. E.g.,
+ // if you map memory within a larger region that you have mprotect'ed, you can
+ // register the entire large region.
+ // - the implementation is not optimized for performance. This interface is
+ // intended to be used for a small number of relatively static regions.
+ void __lsan_register_root_region(const void *p, size_t size);
+ void __lsan_unregister_root_region(const void *p, size_t size);
+
+ // Check for leaks now. This function behaves identically to the default
+ // end-of-process leak check. In particular, it will terminate the process if
+ // leaks are found and the exitcode runtime flag is non-zero.
+ // Subsequent calls to this function will have no effect and end-of-process
+ // leak check will not run. Effectively, end-of-process leak check is moved to
+ // the time of first invocation of this function.
+ // By calling this function early during process shutdown, you can instruct
+ // LSan to ignore shutdown-only leaks which happen later on.
+ void __lsan_do_leak_check(void);
+
+ // Check for leaks now. Returns zero if no leaks have been found or if leak
+ // detection is disabled, non-zero otherwise.
+ // This function may be called repeatedly, e.g. to periodically check a
+ // long-running process. It prints a leak report if appropriate, but does not
+ // terminate the process. It does not affect the behavior of
+ // __lsan_do_leak_check() or the end-of-process leak check, and is not
+ // affected by them.
+ int __lsan_do_recoverable_leak_check(void);
+
+ // The user may optionally provide this function to disallow leak checking
+ // for the program it is linked into (if the return value is non-zero). This
+ // function must be defined as returning a constant value; any behavior beyond
+ // that is unsupported.
+ // To avoid dead stripping, you may need to define this function with
+ // __attribute__((used))
+ int __lsan_is_turned_off(void);
+
+ // This function may be optionally provided by user and should return
+ // a string containing LSan runtime options. See lsan_flags.inc for details.
+ const char *__lsan_default_options(void);
+
+ // This function may be optionally provided by the user and should return
+ // a string containing LSan suppressions.
+ const char *__lsan_default_suppressions(void);
+#ifdef __cplusplus
+} // extern "C"
+
+namespace __lsan {
+class ScopedDisabler {
+ public:
+ ScopedDisabler() { __lsan_disable(); }
+ ~ScopedDisabler() { __lsan_enable(); }
+};
+} // namespace __lsan
+#endif
+
+#endif // SANITIZER_LSAN_INTERFACE_H
--- /dev/null
+//===-- msan_interface.h --------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file is a part of MemorySanitizer.
+//
+// Public interface header.
+//===----------------------------------------------------------------------===//
+#ifndef MSAN_INTERFACE_H
+#define MSAN_INTERFACE_H
+
+#include <sanitizer/common_interface_defs.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+ /* Set raw origin for the memory range. */
+ void __msan_set_origin(const volatile void *a, size_t size, uint32_t origin);
+
+ /* Get raw origin for an address. */
+ uint32_t __msan_get_origin(const volatile void *a);
+
+ /* Test that this_id is a descendant of prev_id (or they are simply equal).
+ * "descendant" here means they are part of the same chain, created with
+ * __msan_chain_origin. */
+ int __msan_origin_is_descendant_or_same(uint32_t this_id, uint32_t prev_id);
+
+ /* Returns non-zero if tracking origins. */
+ int __msan_get_track_origins(void);
+
+ /* Returns the origin id of the latest UMR in the calling thread. */
+ uint32_t __msan_get_umr_origin(void);
+
+ /* Make memory region fully initialized (without changing its contents). */
+ void __msan_unpoison(const volatile void *a, size_t size);
+
+ /* Make a null-terminated string fully initialized (without changing its
+ contents). */
+ void __msan_unpoison_string(const volatile char *a);
+
+ /* Make first n parameters of the next function call fully initialized. */
+ void __msan_unpoison_param(size_t n);
+
+ /* Make memory region fully uninitialized (without changing its contents).
+ This is a legacy interface that does not update origin information. Use
+ __msan_allocated_memory() instead. */
+ void __msan_poison(const volatile void *a, size_t size);
+
+ /* Make memory region partially uninitialized (without changing its contents).
+ */
+ void __msan_partial_poison(const volatile void *data, void *shadow,
+ size_t size);
+
+ /* Returns the offset of the first (at least partially) poisoned byte in the
+ memory range, or -1 if the whole range is good. */
+ intptr_t __msan_test_shadow(const volatile void *x, size_t size);
+
+ /* Checks that memory range is fully initialized, and reports an error if it
+ * is not. */
+ void __msan_check_mem_is_initialized(const volatile void *x, size_t size);
+
+ /* For testing:
+ __msan_set_expect_umr(1);
+ ... some buggy code ...
+ __msan_set_expect_umr(0);
+ The last line will verify that a UMR happened. */
+ void __msan_set_expect_umr(int expect_umr);
+
+ /* Change the value of keep_going flag. Non-zero value means don't terminate
+ program execution when an error is detected. This will not affect error in
+ modules that were compiled without the corresponding compiler flag. */
+ void __msan_set_keep_going(int keep_going);
+
+ /* Print shadow and origin for the memory range to stderr in a human-readable
+ format. */
+ void __msan_print_shadow(const volatile void *x, size_t size);
+
+ /* Print shadow for the memory range to stderr in a minimalistic
+ human-readable format. */
+ void __msan_dump_shadow(const volatile void *x, size_t size);
+
+ /* Returns true if running under a dynamic tool (DynamoRio-based). */
+ int __msan_has_dynamic_component(void);
+
+ /* Tell MSan about newly allocated memory (ex.: custom allocator).
+ Memory will be marked uninitialized, with origin at the call site. */
+ void __msan_allocated_memory(const volatile void* data, size_t size);
+
+ /* Tell MSan about newly destroyed memory. Mark memory as uninitialized. */
+ void __sanitizer_dtor_callback(const volatile void* data, size_t size);
+
+ /* This function may be optionally provided by user and should return
+ a string containing Msan runtime options. See msan_flags.h for details. */
+ const char* __msan_default_options(void);
+
+ /* Deprecated. Call __sanitizer_set_death_callback instead. */
+ void __msan_set_death_callback(void (*callback)(void));
+
+ /* Update shadow for the application copy of size bytes from src to dst.
+ Src and dst are application addresses. This function does not copy the
+ actual application memory, it only updates shadow and origin for such
+ copy. Source and destination regions can overlap. */
+ void __msan_copy_shadow(const volatile void *dst, const volatile void *src,
+ size_t size);
+
+ /* Disables uninitialized memory checks in interceptors. */
+ void __msan_scoped_disable_interceptor_checks(void);
+
+ /* Re-enables uninitialized memory checks in interceptors after a previous
+ call to __msan_scoped_disable_interceptor_checks. */
+ void __msan_scoped_enable_interceptor_checks(void);
+
+#ifdef __cplusplus
+} // extern "C"
+#endif
+
+#endif
--- /dev/null
+//===-- netbsd_syscall_hooks.h --------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file is a part of public sanitizer interface.
+//
+// System call handlers.
+//
+// Interface methods declared in this header implement pre- and post- syscall
+// actions for the active sanitizer.
+// Usage:
+// __sanitizer_syscall_pre_getfoo(...args...);
+// long long res = syscall(SYS_getfoo, ...args...);
+// __sanitizer_syscall_post_getfoo(res, ...args...);
+//
+// DO NOT EDIT! THIS FILE HAS BEEN GENERATED!
+//
+// Generated with: generate_netbsd_syscalls.awk
+// Generated date: 2019-12-24
+// Generated from: syscalls.master,v 1.296 2019/09/22 22:59:39 christos Exp
+//
+//===----------------------------------------------------------------------===//
+#ifndef SANITIZER_NETBSD_SYSCALL_HOOKS_H
+#define SANITIZER_NETBSD_SYSCALL_HOOKS_H
+
+#define __sanitizer_syscall_pre_syscall(code, arg0, arg1, arg2, arg3, arg4, \
+ arg5, arg6, arg7) \
+ __sanitizer_syscall_pre_impl_syscall( \
+ (long long)(code), (long long)(arg0), (long long)(arg1), \
+ (long long)(arg2), (long long)(arg3), (long long)(arg4), \
+ (long long)(arg5), (long long)(arg6), (long long)(arg7))
+#define __sanitizer_syscall_post_syscall(res, code, arg0, arg1, arg2, arg3, \
+ arg4, arg5, arg6, arg7) \
+ __sanitizer_syscall_post_impl_syscall( \
+ res, (long long)(code), (long long)(arg0), (long long)(arg1), \
+ (long long)(arg2), (long long)(arg3), (long long)(arg4), \
+ (long long)(arg5), (long long)(arg6), (long long)(arg7))
+#define __sanitizer_syscall_pre_exit(rval) \
+ __sanitizer_syscall_pre_impl_exit((long long)(rval))
+#define __sanitizer_syscall_post_exit(res, rval) \
+ __sanitizer_syscall_post_impl_exit(res, (long long)(rval))
+#define __sanitizer_syscall_pre_fork() __sanitizer_syscall_pre_impl_fork()
+#define __sanitizer_syscall_post_fork(res) \
+ __sanitizer_syscall_post_impl_fork(res)
+#define __sanitizer_syscall_pre_read(fd, buf, nbyte) \
+ __sanitizer_syscall_pre_impl_read((long long)(fd), (long long)(buf), \
+ (long long)(nbyte))
+#define __sanitizer_syscall_post_read(res, fd, buf, nbyte) \
+ __sanitizer_syscall_post_impl_read(res, (long long)(fd), (long long)(buf), \
+ (long long)(nbyte))
+#define __sanitizer_syscall_pre_write(fd, buf, nbyte) \
+ __sanitizer_syscall_pre_impl_write((long long)(fd), (long long)(buf), \
+ (long long)(nbyte))
+#define __sanitizer_syscall_post_write(res, fd, buf, nbyte) \
+ __sanitizer_syscall_post_impl_write(res, (long long)(fd), (long long)(buf), \
+ (long long)(nbyte))
+#define __sanitizer_syscall_pre_open(path, flags, mode) \
+ __sanitizer_syscall_pre_impl_open((long long)(path), (long long)(flags), \
+ (long long)(mode))
+#define __sanitizer_syscall_post_open(res, path, flags, mode) \
+ __sanitizer_syscall_post_impl_open(res, (long long)(path), \
+ (long long)(flags), (long long)(mode))
+#define __sanitizer_syscall_pre_close(fd) \
+ __sanitizer_syscall_pre_impl_close((long long)(fd))
+#define __sanitizer_syscall_post_close(res, fd) \
+ __sanitizer_syscall_post_impl_close(res, (long long)(fd))
+#define __sanitizer_syscall_pre_compat_50_wait4(pid, status, options, rusage) \
+ __sanitizer_syscall_pre_impl_compat_50_wait4( \
+ (long long)(pid), (long long)(status), (long long)(options), \
+ (long long)(rusage))
+#define __sanitizer_syscall_post_compat_50_wait4(res, pid, status, options, \
+ rusage) \
+ __sanitizer_syscall_post_impl_compat_50_wait4( \
+ res, (long long)(pid), (long long)(status), (long long)(options), \
+ (long long)(rusage))
+#define __sanitizer_syscall_pre_compat_43_ocreat(path, mode) \
+ __sanitizer_syscall_pre_impl_compat_43_ocreat((long long)(path), \
+ (long long)(mode))
+#define __sanitizer_syscall_post_compat_43_ocreat(res, path, mode) \
+ __sanitizer_syscall_post_impl_compat_43_ocreat(res, (long long)(path), \
+ (long long)(mode))
+#define __sanitizer_syscall_pre_link(path, link) \
+ __sanitizer_syscall_pre_impl_link((long long)(path), (long long)(link))
+#define __sanitizer_syscall_post_link(res, path, link) \
+ __sanitizer_syscall_post_impl_link(res, (long long)(path), (long long)(link))
+#define __sanitizer_syscall_pre_unlink(path) \
+ __sanitizer_syscall_pre_impl_unlink((long long)(path))
+#define __sanitizer_syscall_post_unlink(res, path) \
+ __sanitizer_syscall_post_impl_unlink(res, (long long)(path))
+/* syscall 11 has been skipped */
+#define __sanitizer_syscall_pre_chdir(path) \
+ __sanitizer_syscall_pre_impl_chdir((long long)(path))
+#define __sanitizer_syscall_post_chdir(res, path) \
+ __sanitizer_syscall_post_impl_chdir(res, (long long)(path))
+#define __sanitizer_syscall_pre_fchdir(fd) \
+ __sanitizer_syscall_pre_impl_fchdir((long long)(fd))
+#define __sanitizer_syscall_post_fchdir(res, fd) \
+ __sanitizer_syscall_post_impl_fchdir(res, (long long)(fd))
+#define __sanitizer_syscall_pre_compat_50_mknod(path, mode, dev) \
+ __sanitizer_syscall_pre_impl_compat_50_mknod( \
+ (long long)(path), (long long)(mode), (long long)(dev))
+#define __sanitizer_syscall_post_compat_50_mknod(res, path, mode, dev) \
+ __sanitizer_syscall_post_impl_compat_50_mknod( \
+ res, (long long)(path), (long long)(mode), (long long)(dev))
+#define __sanitizer_syscall_pre_chmod(path, mode) \
+ __sanitizer_syscall_pre_impl_chmod((long long)(path), (long long)(mode))
+#define __sanitizer_syscall_post_chmod(res, path, mode) \
+ __sanitizer_syscall_post_impl_chmod(res, (long long)(path), (long long)(mode))
+#define __sanitizer_syscall_pre_chown(path, uid, gid) \
+ __sanitizer_syscall_pre_impl_chown((long long)(path), (long long)(uid), \
+ (long long)(gid))
+#define __sanitizer_syscall_post_chown(res, path, uid, gid) \
+ __sanitizer_syscall_post_impl_chown(res, (long long)(path), \
+ (long long)(uid), (long long)(gid))
+#define __sanitizer_syscall_pre_break(nsize) \
+ __sanitizer_syscall_pre_impl_break((long long)(nsize))
+#define __sanitizer_syscall_post_break(res, nsize) \
+ __sanitizer_syscall_post_impl_break(res, (long long)(nsize))
+#define __sanitizer_syscall_pre_compat_20_getfsstat(buf, bufsize, flags) \
+ __sanitizer_syscall_pre_impl_compat_20_getfsstat( \
+ (long long)(buf), (long long)(bufsize), (long long)(flags))
+#define __sanitizer_syscall_post_compat_20_getfsstat(res, buf, bufsize, flags) \
+ __sanitizer_syscall_post_impl_compat_20_getfsstat( \
+ res, (long long)(buf), (long long)(bufsize), (long long)(flags))
+#define __sanitizer_syscall_pre_compat_43_olseek(fd, offset, whence) \
+ __sanitizer_syscall_pre_impl_compat_43_olseek( \
+ (long long)(fd), (long long)(offset), (long long)(whence))
+#define __sanitizer_syscall_post_compat_43_olseek(res, fd, offset, whence) \
+ __sanitizer_syscall_post_impl_compat_43_olseek( \
+ res, (long long)(fd), (long long)(offset), (long long)(whence))
+#define __sanitizer_syscall_pre_getpid() __sanitizer_syscall_pre_impl_getpid()
+#define __sanitizer_syscall_post_getpid(res) \
+ __sanitizer_syscall_post_impl_getpid(res)
+#define __sanitizer_syscall_pre_compat_40_mount(type, path, flags, data) \
+ __sanitizer_syscall_pre_impl_compat_40_mount( \
+ (long long)(type), (long long)(path), (long long)(flags), \
+ (long long)(data))
+#define __sanitizer_syscall_post_compat_40_mount(res, type, path, flags, data) \
+ __sanitizer_syscall_post_impl_compat_40_mount( \
+ res, (long long)(type), (long long)(path), (long long)(flags), \
+ (long long)(data))
+#define __sanitizer_syscall_pre_unmount(path, flags) \
+ __sanitizer_syscall_pre_impl_unmount((long long)(path), (long long)(flags))
+#define __sanitizer_syscall_post_unmount(res, path, flags) \
+ __sanitizer_syscall_post_impl_unmount(res, (long long)(path), \
+ (long long)(flags))
+#define __sanitizer_syscall_pre_setuid(uid) \
+ __sanitizer_syscall_pre_impl_setuid((long long)(uid))
+#define __sanitizer_syscall_post_setuid(res, uid) \
+ __sanitizer_syscall_post_impl_setuid(res, (long long)(uid))
+#define __sanitizer_syscall_pre_getuid() __sanitizer_syscall_pre_impl_getuid()
+#define __sanitizer_syscall_post_getuid(res) \
+ __sanitizer_syscall_post_impl_getuid(res)
+#define __sanitizer_syscall_pre_geteuid() __sanitizer_syscall_pre_impl_geteuid()
+#define __sanitizer_syscall_post_geteuid(res) \
+ __sanitizer_syscall_post_impl_geteuid(res)
+#define __sanitizer_syscall_pre_ptrace(req, pid, addr, data) \
+ __sanitizer_syscall_pre_impl_ptrace((long long)(req), (long long)(pid), \
+ (long long)(addr), (long long)(data))
+#define __sanitizer_syscall_post_ptrace(res, req, pid, addr, data) \
+ __sanitizer_syscall_post_impl_ptrace(res, (long long)(req), \
+ (long long)(pid), (long long)(addr), \
+ (long long)(data))
+#define __sanitizer_syscall_pre_recvmsg(s, msg, flags) \
+ __sanitizer_syscall_pre_impl_recvmsg((long long)(s), (long long)(msg), \
+ (long long)(flags))
+#define __sanitizer_syscall_post_recvmsg(res, s, msg, flags) \
+ __sanitizer_syscall_post_impl_recvmsg(res, (long long)(s), (long long)(msg), \
+ (long long)(flags))
+#define __sanitizer_syscall_pre_sendmsg(s, msg, flags) \
+ __sanitizer_syscall_pre_impl_sendmsg((long long)(s), (long long)(msg), \
+ (long long)(flags))
+#define __sanitizer_syscall_post_sendmsg(res, s, msg, flags) \
+ __sanitizer_syscall_post_impl_sendmsg(res, (long long)(s), (long long)(msg), \
+ (long long)(flags))
+#define __sanitizer_syscall_pre_recvfrom(s, buf, len, flags, from, \
+ fromlenaddr) \
+ __sanitizer_syscall_pre_impl_recvfrom( \
+ (long long)(s), (long long)(buf), (long long)(len), (long long)(flags), \
+ (long long)(from), (long long)(fromlenaddr))
+#define __sanitizer_syscall_post_recvfrom(res, s, buf, len, flags, from, \
+ fromlenaddr) \
+ __sanitizer_syscall_post_impl_recvfrom( \
+ res, (long long)(s), (long long)(buf), (long long)(len), \
+ (long long)(flags), (long long)(from), (long long)(fromlenaddr))
+#define __sanitizer_syscall_pre_accept(s, name, anamelen) \
+ __sanitizer_syscall_pre_impl_accept((long long)(s), (long long)(name), \
+ (long long)(anamelen))
+#define __sanitizer_syscall_post_accept(res, s, name, anamelen) \
+ __sanitizer_syscall_post_impl_accept(res, (long long)(s), (long long)(name), \
+ (long long)(anamelen))
+#define __sanitizer_syscall_pre_getpeername(fdes, asa, alen) \
+ __sanitizer_syscall_pre_impl_getpeername( \
+ (long long)(fdes), (long long)(asa), (long long)(alen))
+#define __sanitizer_syscall_post_getpeername(res, fdes, asa, alen) \
+ __sanitizer_syscall_post_impl_getpeername( \
+ res, (long long)(fdes), (long long)(asa), (long long)(alen))
+#define __sanitizer_syscall_pre_getsockname(fdes, asa, alen) \
+ __sanitizer_syscall_pre_impl_getsockname( \
+ (long long)(fdes), (long long)(asa), (long long)(alen))
+#define __sanitizer_syscall_post_getsockname(res, fdes, asa, alen) \
+ __sanitizer_syscall_post_impl_getsockname( \
+ res, (long long)(fdes), (long long)(asa), (long long)(alen))
+#define __sanitizer_syscall_pre_access(path, flags) \
+ __sanitizer_syscall_pre_impl_access((long long)(path), (long long)(flags))
+#define __sanitizer_syscall_post_access(res, path, flags) \
+ __sanitizer_syscall_post_impl_access(res, (long long)(path), \
+ (long long)(flags))
+#define __sanitizer_syscall_pre_chflags(path, flags) \
+ __sanitizer_syscall_pre_impl_chflags((long long)(path), (long long)(flags))
+#define __sanitizer_syscall_post_chflags(res, path, flags) \
+ __sanitizer_syscall_post_impl_chflags(res, (long long)(path), \
+ (long long)(flags))
+#define __sanitizer_syscall_pre_fchflags(fd, flags) \
+ __sanitizer_syscall_pre_impl_fchflags((long long)(fd), (long long)(flags))
+#define __sanitizer_syscall_post_fchflags(res, fd, flags) \
+ __sanitizer_syscall_post_impl_fchflags(res, (long long)(fd), \
+ (long long)(flags))
+#define __sanitizer_syscall_pre_sync() __sanitizer_syscall_pre_impl_sync()
+#define __sanitizer_syscall_post_sync(res) \
+ __sanitizer_syscall_post_impl_sync(res)
+#define __sanitizer_syscall_pre_kill(pid, signum) \
+ __sanitizer_syscall_pre_impl_kill((long long)(pid), (long long)(signum))
+#define __sanitizer_syscall_post_kill(res, pid, signum) \
+ __sanitizer_syscall_post_impl_kill(res, (long long)(pid), (long long)(signum))
+#define __sanitizer_syscall_pre_compat_43_stat43(path, ub) \
+ __sanitizer_syscall_pre_impl_compat_43_stat43((long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_post_compat_43_stat43(res, path, ub) \
+ __sanitizer_syscall_post_impl_compat_43_stat43(res, (long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_pre_getppid() __sanitizer_syscall_pre_impl_getppid()
+#define __sanitizer_syscall_post_getppid(res) \
+ __sanitizer_syscall_post_impl_getppid(res)
+#define __sanitizer_syscall_pre_compat_43_lstat43(path, ub) \
+ __sanitizer_syscall_pre_impl_compat_43_lstat43((long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_post_compat_43_lstat43(res, path, ub) \
+ __sanitizer_syscall_post_impl_compat_43_lstat43(res, (long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_pre_dup(fd) \
+ __sanitizer_syscall_pre_impl_dup((long long)(fd))
+#define __sanitizer_syscall_post_dup(res, fd) \
+ __sanitizer_syscall_post_impl_dup(res, (long long)(fd))
+#define __sanitizer_syscall_pre_pipe() __sanitizer_syscall_pre_impl_pipe()
+#define __sanitizer_syscall_post_pipe(res) \
+ __sanitizer_syscall_post_impl_pipe(res)
+#define __sanitizer_syscall_pre_getegid() __sanitizer_syscall_pre_impl_getegid()
+#define __sanitizer_syscall_post_getegid(res) \
+ __sanitizer_syscall_post_impl_getegid(res)
+#define __sanitizer_syscall_pre_profil(samples, size, offset, scale) \
+ __sanitizer_syscall_pre_impl_profil((long long)(samples), (long long)(size), \
+ (long long)(offset), (long long)(scale))
+#define __sanitizer_syscall_post_profil(res, samples, size, offset, scale) \
+ __sanitizer_syscall_post_impl_profil(res, (long long)(samples), \
+ (long long)(size), (long long)(offset), \
+ (long long)(scale))
+#define __sanitizer_syscall_pre_ktrace(fname, ops, facs, pid) \
+ __sanitizer_syscall_pre_impl_ktrace((long long)(fname), (long long)(ops), \
+ (long long)(facs), (long long)(pid))
+#define __sanitizer_syscall_post_ktrace(res, fname, ops, facs, pid) \
+ __sanitizer_syscall_post_impl_ktrace(res, (long long)(fname), \
+ (long long)(ops), (long long)(facs), \
+ (long long)(pid))
+#define __sanitizer_syscall_pre_compat_13_sigaction13(signum, nsa, osa) \
+ __sanitizer_syscall_pre_impl_compat_13_sigaction13( \
+ (long long)(signum), (long long)(nsa), (long long)(osa))
+#define __sanitizer_syscall_post_compat_13_sigaction13(res, signum, nsa, osa) \
+ __sanitizer_syscall_post_impl_compat_13_sigaction13( \
+ res, (long long)(signum), (long long)(nsa), (long long)(osa))
+#define __sanitizer_syscall_pre_getgid() __sanitizer_syscall_pre_impl_getgid()
+#define __sanitizer_syscall_post_getgid(res) \
+ __sanitizer_syscall_post_impl_getgid(res)
+#define __sanitizer_syscall_pre_compat_13_sigprocmask13(how, mask) \
+ __sanitizer_syscall_pre_impl_compat_13_sigprocmask13((long long)(how), \
+ (long long)(mask))
+#define __sanitizer_syscall_post_compat_13_sigprocmask13(res, how, mask) \
+ __sanitizer_syscall_post_impl_compat_13_sigprocmask13(res, (long long)(how), \
+ (long long)(mask))
+#define __sanitizer_syscall_pre___getlogin(namebuf, namelen) \
+ __sanitizer_syscall_pre_impl___getlogin((long long)(namebuf), \
+ (long long)(namelen))
+#define __sanitizer_syscall_post___getlogin(res, namebuf, namelen) \
+ __sanitizer_syscall_post_impl___getlogin(res, (long long)(namebuf), \
+ (long long)(namelen))
+#define __sanitizer_syscall_pre___setlogin(namebuf) \
+ __sanitizer_syscall_pre_impl___setlogin((long long)(namebuf))
+#define __sanitizer_syscall_post___setlogin(res, namebuf) \
+ __sanitizer_syscall_post_impl___setlogin(res, (long long)(namebuf))
+#define __sanitizer_syscall_pre_acct(path) \
+ __sanitizer_syscall_pre_impl_acct((long long)(path))
+#define __sanitizer_syscall_post_acct(res, path) \
+ __sanitizer_syscall_post_impl_acct(res, (long long)(path))
+#define __sanitizer_syscall_pre_compat_13_sigpending13() \
+ __sanitizer_syscall_pre_impl_compat_13_sigpending13()
+#define __sanitizer_syscall_post_compat_13_sigpending13(res) \
+ __sanitizer_syscall_post_impl_compat_13_sigpending13(res)
+#define __sanitizer_syscall_pre_compat_13_sigaltstack13(nss, oss) \
+ __sanitizer_syscall_pre_impl_compat_13_sigaltstack13((long long)(nss), \
+ (long long)(oss))
+#define __sanitizer_syscall_post_compat_13_sigaltstack13(res, nss, oss) \
+ __sanitizer_syscall_post_impl_compat_13_sigaltstack13(res, (long long)(nss), \
+ (long long)(oss))
+#define __sanitizer_syscall_pre_ioctl(fd, com, data) \
+ __sanitizer_syscall_pre_impl_ioctl((long long)(fd), (long long)(com), \
+ (long long)(data))
+#define __sanitizer_syscall_post_ioctl(res, fd, com, data) \
+ __sanitizer_syscall_post_impl_ioctl(res, (long long)(fd), (long long)(com), \
+ (long long)(data))
+#define __sanitizer_syscall_pre_compat_12_oreboot(opt) \
+ __sanitizer_syscall_pre_impl_compat_12_oreboot((long long)(opt))
+#define __sanitizer_syscall_post_compat_12_oreboot(res, opt) \
+ __sanitizer_syscall_post_impl_compat_12_oreboot(res, (long long)(opt))
+#define __sanitizer_syscall_pre_revoke(path) \
+ __sanitizer_syscall_pre_impl_revoke((long long)(path))
+#define __sanitizer_syscall_post_revoke(res, path) \
+ __sanitizer_syscall_post_impl_revoke(res, (long long)(path))
+#define __sanitizer_syscall_pre_symlink(path, link) \
+ __sanitizer_syscall_pre_impl_symlink((long long)(path), (long long)(link))
+#define __sanitizer_syscall_post_symlink(res, path, link) \
+ __sanitizer_syscall_post_impl_symlink(res, (long long)(path), \
+ (long long)(link))
+#define __sanitizer_syscall_pre_readlink(path, buf, count) \
+ __sanitizer_syscall_pre_impl_readlink((long long)(path), (long long)(buf), \
+ (long long)(count))
+#define __sanitizer_syscall_post_readlink(res, path, buf, count) \
+ __sanitizer_syscall_post_impl_readlink(res, (long long)(path), \
+ (long long)(buf), (long long)(count))
+#define __sanitizer_syscall_pre_execve(path, argp, envp) \
+ __sanitizer_syscall_pre_impl_execve((long long)(path), (long long)(argp), \
+ (long long)(envp))
+#define __sanitizer_syscall_post_execve(res, path, argp, envp) \
+ __sanitizer_syscall_post_impl_execve(res, (long long)(path), \
+ (long long)(argp), (long long)(envp))
+#define __sanitizer_syscall_pre_umask(newmask) \
+ __sanitizer_syscall_pre_impl_umask((long long)(newmask))
+#define __sanitizer_syscall_post_umask(res, newmask) \
+ __sanitizer_syscall_post_impl_umask(res, (long long)(newmask))
+#define __sanitizer_syscall_pre_chroot(path) \
+ __sanitizer_syscall_pre_impl_chroot((long long)(path))
+#define __sanitizer_syscall_post_chroot(res, path) \
+ __sanitizer_syscall_post_impl_chroot(res, (long long)(path))
+#define __sanitizer_syscall_pre_compat_43_fstat43(fd, sb) \
+ __sanitizer_syscall_pre_impl_compat_43_fstat43((long long)(fd), \
+ (long long)(sb))
+#define __sanitizer_syscall_post_compat_43_fstat43(res, fd, sb) \
+ __sanitizer_syscall_post_impl_compat_43_fstat43(res, (long long)(fd), \
+ (long long)(sb))
+#define __sanitizer_syscall_pre_compat_43_ogetkerninfo(op, where, size, arg) \
+ __sanitizer_syscall_pre_impl_compat_43_ogetkerninfo( \
+ (long long)(op), (long long)(where), (long long)(size), \
+ (long long)(arg))
+#define __sanitizer_syscall_post_compat_43_ogetkerninfo(res, op, where, size, \
+ arg) \
+ __sanitizer_syscall_post_impl_compat_43_ogetkerninfo( \
+ res, (long long)(op), (long long)(where), (long long)(size), \
+ (long long)(arg))
+#define __sanitizer_syscall_pre_compat_43_ogetpagesize() \
+ __sanitizer_syscall_pre_impl_compat_43_ogetpagesize()
+#define __sanitizer_syscall_post_compat_43_ogetpagesize(res) \
+ __sanitizer_syscall_post_impl_compat_43_ogetpagesize(res)
+#define __sanitizer_syscall_pre_compat_12_msync(addr, len) \
+ __sanitizer_syscall_pre_impl_compat_12_msync((long long)(addr), \
+ (long long)(len))
+#define __sanitizer_syscall_post_compat_12_msync(res, addr, len) \
+ __sanitizer_syscall_post_impl_compat_12_msync(res, (long long)(addr), \
+ (long long)(len))
+#define __sanitizer_syscall_pre_vfork() __sanitizer_syscall_pre_impl_vfork()
+#define __sanitizer_syscall_post_vfork(res) \
+ __sanitizer_syscall_post_impl_vfork(res)
+/* syscall 67 has been skipped */
+/* syscall 68 has been skipped */
+/* syscall 69 has been skipped */
+/* syscall 70 has been skipped */
+#define __sanitizer_syscall_pre_compat_43_ommap(addr, len, prot, flags, fd, \
+ pos) \
+ __sanitizer_syscall_pre_impl_compat_43_ommap( \
+ (long long)(addr), (long long)(len), (long long)(prot), \
+ (long long)(flags), (long long)(fd), (long long)(pos))
+#define __sanitizer_syscall_post_compat_43_ommap(res, addr, len, prot, flags, \
+ fd, pos) \
+ __sanitizer_syscall_post_impl_compat_43_ommap( \
+ res, (long long)(addr), (long long)(len), (long long)(prot), \
+ (long long)(flags), (long long)(fd), (long long)(pos))
+#define __sanitizer_syscall_pre_vadvise(anom) \
+ __sanitizer_syscall_pre_impl_vadvise((long long)(anom))
+#define __sanitizer_syscall_post_vadvise(res, anom) \
+ __sanitizer_syscall_post_impl_vadvise(res, (long long)(anom))
+#define __sanitizer_syscall_pre_munmap(addr, len) \
+ __sanitizer_syscall_pre_impl_munmap((long long)(addr), (long long)(len))
+#define __sanitizer_syscall_post_munmap(res, addr, len) \
+ __sanitizer_syscall_post_impl_munmap(res, (long long)(addr), (long long)(len))
+#define __sanitizer_syscall_pre_mprotect(addr, len, prot) \
+ __sanitizer_syscall_pre_impl_mprotect((long long)(addr), (long long)(len), \
+ (long long)(prot))
+#define __sanitizer_syscall_post_mprotect(res, addr, len, prot) \
+ __sanitizer_syscall_post_impl_mprotect(res, (long long)(addr), \
+ (long long)(len), (long long)(prot))
+#define __sanitizer_syscall_pre_madvise(addr, len, behav) \
+ __sanitizer_syscall_pre_impl_madvise((long long)(addr), (long long)(len), \
+ (long long)(behav))
+#define __sanitizer_syscall_post_madvise(res, addr, len, behav) \
+ __sanitizer_syscall_post_impl_madvise(res, (long long)(addr), \
+ (long long)(len), (long long)(behav))
+/* syscall 76 has been skipped */
+/* syscall 77 has been skipped */
+#define __sanitizer_syscall_pre_mincore(addr, len, vec) \
+ __sanitizer_syscall_pre_impl_mincore((long long)(addr), (long long)(len), \
+ (long long)(vec))
+#define __sanitizer_syscall_post_mincore(res, addr, len, vec) \
+ __sanitizer_syscall_post_impl_mincore(res, (long long)(addr), \
+ (long long)(len), (long long)(vec))
+#define __sanitizer_syscall_pre_getgroups(gidsetsize, gidset) \
+ __sanitizer_syscall_pre_impl_getgroups((long long)(gidsetsize), \
+ (long long)(gidset))
+#define __sanitizer_syscall_post_getgroups(res, gidsetsize, gidset) \
+ __sanitizer_syscall_post_impl_getgroups(res, (long long)(gidsetsize), \
+ (long long)(gidset))
+#define __sanitizer_syscall_pre_setgroups(gidsetsize, gidset) \
+ __sanitizer_syscall_pre_impl_setgroups((long long)(gidsetsize), \
+ (long long)(gidset))
+#define __sanitizer_syscall_post_setgroups(res, gidsetsize, gidset) \
+ __sanitizer_syscall_post_impl_setgroups(res, (long long)(gidsetsize), \
+ (long long)(gidset))
+#define __sanitizer_syscall_pre_getpgrp() __sanitizer_syscall_pre_impl_getpgrp()
+#define __sanitizer_syscall_post_getpgrp(res) \
+ __sanitizer_syscall_post_impl_getpgrp(res)
+#define __sanitizer_syscall_pre_setpgid(pid, pgid) \
+ __sanitizer_syscall_pre_impl_setpgid((long long)(pid), (long long)(pgid))
+#define __sanitizer_syscall_post_setpgid(res, pid, pgid) \
+ __sanitizer_syscall_post_impl_setpgid(res, (long long)(pid), \
+ (long long)(pgid))
+#define __sanitizer_syscall_pre_compat_50_setitimer(which, itv, oitv) \
+ __sanitizer_syscall_pre_impl_compat_50_setitimer( \
+ (long long)(which), (long long)(itv), (long long)(oitv))
+#define __sanitizer_syscall_post_compat_50_setitimer(res, which, itv, oitv) \
+ __sanitizer_syscall_post_impl_compat_50_setitimer( \
+ res, (long long)(which), (long long)(itv), (long long)(oitv))
+#define __sanitizer_syscall_pre_compat_43_owait() \
+ __sanitizer_syscall_pre_impl_compat_43_owait()
+#define __sanitizer_syscall_post_compat_43_owait(res) \
+ __sanitizer_syscall_post_impl_compat_43_owait(res)
+#define __sanitizer_syscall_pre_compat_12_oswapon(name) \
+ __sanitizer_syscall_pre_impl_compat_12_oswapon((long long)(name))
+#define __sanitizer_syscall_post_compat_12_oswapon(res, name) \
+ __sanitizer_syscall_post_impl_compat_12_oswapon(res, (long long)(name))
+#define __sanitizer_syscall_pre_compat_50_getitimer(which, itv) \
+ __sanitizer_syscall_pre_impl_compat_50_getitimer((long long)(which), \
+ (long long)(itv))
+#define __sanitizer_syscall_post_compat_50_getitimer(res, which, itv) \
+ __sanitizer_syscall_post_impl_compat_50_getitimer(res, (long long)(which), \
+ (long long)(itv))
+#define __sanitizer_syscall_pre_compat_43_ogethostname(hostname, len) \
+ __sanitizer_syscall_pre_impl_compat_43_ogethostname((long long)(hostname), \
+ (long long)(len))
+#define __sanitizer_syscall_post_compat_43_ogethostname(res, hostname, len) \
+ __sanitizer_syscall_post_impl_compat_43_ogethostname( \
+ res, (long long)(hostname), (long long)(len))
+#define __sanitizer_syscall_pre_compat_43_osethostname(hostname, len) \
+ __sanitizer_syscall_pre_impl_compat_43_osethostname((long long)(hostname), \
+ (long long)(len))
+#define __sanitizer_syscall_post_compat_43_osethostname(res, hostname, len) \
+ __sanitizer_syscall_post_impl_compat_43_osethostname( \
+ res, (long long)(hostname), (long long)(len))
+#define __sanitizer_syscall_pre_compat_43_ogetdtablesize() \
+ __sanitizer_syscall_pre_impl_compat_43_ogetdtablesize()
+#define __sanitizer_syscall_post_compat_43_ogetdtablesize(res) \
+ __sanitizer_syscall_post_impl_compat_43_ogetdtablesize(res)
+#define __sanitizer_syscall_pre_dup2(from, to) \
+ __sanitizer_syscall_pre_impl_dup2((long long)(from), (long long)(to))
+#define __sanitizer_syscall_post_dup2(res, from, to) \
+ __sanitizer_syscall_post_impl_dup2(res, (long long)(from), (long long)(to))
+/* syscall 91 has been skipped */
+#define __sanitizer_syscall_pre_fcntl(fd, cmd, arg) \
+ __sanitizer_syscall_pre_impl_fcntl((long long)(fd), (long long)(cmd), \
+ (long long)(arg))
+#define __sanitizer_syscall_post_fcntl(res, fd, cmd, arg) \
+ __sanitizer_syscall_post_impl_fcntl(res, (long long)(fd), (long long)(cmd), \
+ (long long)(arg))
+#define __sanitizer_syscall_pre_compat_50_select(nd, in, ou, ex, tv) \
+ __sanitizer_syscall_pre_impl_compat_50_select( \
+ (long long)(nd), (long long)(in), (long long)(ou), (long long)(ex), \
+ (long long)(tv))
+#define __sanitizer_syscall_post_compat_50_select(res, nd, in, ou, ex, tv) \
+ __sanitizer_syscall_post_impl_compat_50_select( \
+ res, (long long)(nd), (long long)(in), (long long)(ou), (long long)(ex), \
+ (long long)(tv))
+/* syscall 94 has been skipped */
+#define __sanitizer_syscall_pre_fsync(fd) \
+ __sanitizer_syscall_pre_impl_fsync((long long)(fd))
+#define __sanitizer_syscall_post_fsync(res, fd) \
+ __sanitizer_syscall_post_impl_fsync(res, (long long)(fd))
+#define __sanitizer_syscall_pre_setpriority(which, who, prio) \
+ __sanitizer_syscall_pre_impl_setpriority( \
+ (long long)(which), (long long)(who), (long long)(prio))
+#define __sanitizer_syscall_post_setpriority(res, which, who, prio) \
+ __sanitizer_syscall_post_impl_setpriority( \
+ res, (long long)(which), (long long)(who), (long long)(prio))
+#define __sanitizer_syscall_pre_compat_30_socket(domain, type, protocol) \
+ __sanitizer_syscall_pre_impl_compat_30_socket( \
+ (long long)(domain), (long long)(type), (long long)(protocol))
+#define __sanitizer_syscall_post_compat_30_socket(res, domain, type, protocol) \
+ __sanitizer_syscall_post_impl_compat_30_socket( \
+ res, (long long)(domain), (long long)(type), (long long)(protocol))
+#define __sanitizer_syscall_pre_connect(s, name, namelen) \
+ __sanitizer_syscall_pre_impl_connect((long long)(s), (long long)(name), \
+ (long long)(namelen))
+#define __sanitizer_syscall_post_connect(res, s, name, namelen) \
+ __sanitizer_syscall_post_impl_connect( \
+ res, (long long)(s), (long long)(name), (long long)(namelen))
+#define __sanitizer_syscall_pre_compat_43_oaccept(s, name, anamelen) \
+ __sanitizer_syscall_pre_impl_compat_43_oaccept( \
+ (long long)(s), (long long)(name), (long long)(anamelen))
+#define __sanitizer_syscall_post_compat_43_oaccept(res, s, name, anamelen) \
+ __sanitizer_syscall_post_impl_compat_43_oaccept( \
+ res, (long long)(s), (long long)(name), (long long)(anamelen))
+#define __sanitizer_syscall_pre_getpriority(which, who) \
+ __sanitizer_syscall_pre_impl_getpriority((long long)(which), (long long)(who))
+#define __sanitizer_syscall_post_getpriority(res, which, who) \
+ __sanitizer_syscall_post_impl_getpriority(res, (long long)(which), \
+ (long long)(who))
+#define __sanitizer_syscall_pre_compat_43_osend(s, buf, len, flags) \
+ __sanitizer_syscall_pre_impl_compat_43_osend( \
+ (long long)(s), (long long)(buf), (long long)(len), (long long)(flags))
+#define __sanitizer_syscall_post_compat_43_osend(res, s, buf, len, flags) \
+ __sanitizer_syscall_post_impl_compat_43_osend( \
+ res, (long long)(s), (long long)(buf), (long long)(len), \
+ (long long)(flags))
+#define __sanitizer_syscall_pre_compat_43_orecv(s, buf, len, flags) \
+ __sanitizer_syscall_pre_impl_compat_43_orecv( \
+ (long long)(s), (long long)(buf), (long long)(len), (long long)(flags))
+#define __sanitizer_syscall_post_compat_43_orecv(res, s, buf, len, flags) \
+ __sanitizer_syscall_post_impl_compat_43_orecv( \
+ res, (long long)(s), (long long)(buf), (long long)(len), \
+ (long long)(flags))
+#define __sanitizer_syscall_pre_compat_13_sigreturn13(sigcntxp) \
+ __sanitizer_syscall_pre_impl_compat_13_sigreturn13((long long)(sigcntxp))
+#define __sanitizer_syscall_post_compat_13_sigreturn13(res, sigcntxp) \
+ __sanitizer_syscall_post_impl_compat_13_sigreturn13(res, \
+ (long long)(sigcntxp))
+#define __sanitizer_syscall_pre_bind(s, name, namelen) \
+ __sanitizer_syscall_pre_impl_bind((long long)(s), (long long)(name), \
+ (long long)(namelen))
+#define __sanitizer_syscall_post_bind(res, s, name, namelen) \
+ __sanitizer_syscall_post_impl_bind(res, (long long)(s), (long long)(name), \
+ (long long)(namelen))
+#define __sanitizer_syscall_pre_setsockopt(s, level, name, val, valsize) \
+ __sanitizer_syscall_pre_impl_setsockopt((long long)(s), (long long)(level), \
+ (long long)(name), (long long)(val), \
+ (long long)(valsize))
+#define __sanitizer_syscall_post_setsockopt(res, s, level, name, val, valsize) \
+ __sanitizer_syscall_post_impl_setsockopt( \
+ res, (long long)(s), (long long)(level), (long long)(name), \
+ (long long)(val), (long long)(valsize))
+#define __sanitizer_syscall_pre_listen(s, backlog) \
+ __sanitizer_syscall_pre_impl_listen((long long)(s), (long long)(backlog))
+#define __sanitizer_syscall_post_listen(res, s, backlog) \
+ __sanitizer_syscall_post_impl_listen(res, (long long)(s), \
+ (long long)(backlog))
+/* syscall 107 has been skipped */
+#define __sanitizer_syscall_pre_compat_43_osigvec(signum, nsv, osv) \
+ __sanitizer_syscall_pre_impl_compat_43_osigvec( \
+ (long long)(signum), (long long)(nsv), (long long)(osv))
+#define __sanitizer_syscall_post_compat_43_osigvec(res, signum, nsv, osv) \
+ __sanitizer_syscall_post_impl_compat_43_osigvec( \
+ res, (long long)(signum), (long long)(nsv), (long long)(osv))
+#define __sanitizer_syscall_pre_compat_43_osigblock(mask) \
+ __sanitizer_syscall_pre_impl_compat_43_osigblock((long long)(mask))
+#define __sanitizer_syscall_post_compat_43_osigblock(res, mask) \
+ __sanitizer_syscall_post_impl_compat_43_osigblock(res, (long long)(mask))
+#define __sanitizer_syscall_pre_compat_43_osigsetmask(mask) \
+ __sanitizer_syscall_pre_impl_compat_43_osigsetmask((long long)(mask))
+#define __sanitizer_syscall_post_compat_43_osigsetmask(res, mask) \
+ __sanitizer_syscall_post_impl_compat_43_osigsetmask(res, (long long)(mask))
+#define __sanitizer_syscall_pre_compat_13_sigsuspend13(mask) \
+ __sanitizer_syscall_pre_impl_compat_13_sigsuspend13((long long)(mask))
+#define __sanitizer_syscall_post_compat_13_sigsuspend13(res, mask) \
+ __sanitizer_syscall_post_impl_compat_13_sigsuspend13(res, (long long)(mask))
+#define __sanitizer_syscall_pre_compat_43_osigstack(nss, oss) \
+ __sanitizer_syscall_pre_impl_compat_43_osigstack((long long)(nss), \
+ (long long)(oss))
+#define __sanitizer_syscall_post_compat_43_osigstack(res, nss, oss) \
+ __sanitizer_syscall_post_impl_compat_43_osigstack(res, (long long)(nss), \
+ (long long)(oss))
+#define __sanitizer_syscall_pre_compat_43_orecvmsg(s, msg, flags) \
+ __sanitizer_syscall_pre_impl_compat_43_orecvmsg( \
+ (long long)(s), (long long)(msg), (long long)(flags))
+#define __sanitizer_syscall_post_compat_43_orecvmsg(res, s, msg, flags) \
+ __sanitizer_syscall_post_impl_compat_43_orecvmsg( \
+ res, (long long)(s), (long long)(msg), (long long)(flags))
+#define __sanitizer_syscall_pre_compat_43_osendmsg(s, msg, flags) \
+ __sanitizer_syscall_pre_impl_compat_43_osendmsg( \
+ (long long)(s), (long long)(msg), (long long)(flags))
+#define __sanitizer_syscall_post_compat_43_osendmsg(res, s, msg, flags) \
+ __sanitizer_syscall_post_impl_compat_43_osendmsg( \
+ res, (long long)(s), (long long)(msg), (long long)(flags))
+/* syscall 115 has been skipped */
+#define __sanitizer_syscall_pre_compat_50_gettimeofday(tp, tzp) \
+ __sanitizer_syscall_pre_impl_compat_50_gettimeofday((long long)(tp), \
+ (long long)(tzp))
+#define __sanitizer_syscall_post_compat_50_gettimeofday(res, tp, tzp) \
+ __sanitizer_syscall_post_impl_compat_50_gettimeofday(res, (long long)(tp), \
+ (long long)(tzp))
+#define __sanitizer_syscall_pre_compat_50_getrusage(who, rusage) \
+ __sanitizer_syscall_pre_impl_compat_50_getrusage((long long)(who), \
+ (long long)(rusage))
+#define __sanitizer_syscall_post_compat_50_getrusage(res, who, rusage) \
+ __sanitizer_syscall_post_impl_compat_50_getrusage(res, (long long)(who), \
+ (long long)(rusage))
+#define __sanitizer_syscall_pre_getsockopt(s, level, name, val, avalsize) \
+ __sanitizer_syscall_pre_impl_getsockopt((long long)(s), (long long)(level), \
+ (long long)(name), (long long)(val), \
+ (long long)(avalsize))
+#define __sanitizer_syscall_post_getsockopt(res, s, level, name, val, \
+ avalsize) \
+ __sanitizer_syscall_post_impl_getsockopt( \
+ res, (long long)(s), (long long)(level), (long long)(name), \
+ (long long)(val), (long long)(avalsize))
+/* syscall 119 has been skipped */
+#define __sanitizer_syscall_pre_readv(fd, iovp, iovcnt) \
+ __sanitizer_syscall_pre_impl_readv((long long)(fd), (long long)(iovp), \
+ (long long)(iovcnt))
+#define __sanitizer_syscall_post_readv(res, fd, iovp, iovcnt) \
+ __sanitizer_syscall_post_impl_readv(res, (long long)(fd), (long long)(iovp), \
+ (long long)(iovcnt))
+#define __sanitizer_syscall_pre_writev(fd, iovp, iovcnt) \
+ __sanitizer_syscall_pre_impl_writev((long long)(fd), (long long)(iovp), \
+ (long long)(iovcnt))
+#define __sanitizer_syscall_post_writev(res, fd, iovp, iovcnt) \
+ __sanitizer_syscall_post_impl_writev(res, (long long)(fd), \
+ (long long)(iovp), (long long)(iovcnt))
+#define __sanitizer_syscall_pre_compat_50_settimeofday(tv, tzp) \
+ __sanitizer_syscall_pre_impl_compat_50_settimeofday((long long)(tv), \
+ (long long)(tzp))
+#define __sanitizer_syscall_post_compat_50_settimeofday(res, tv, tzp) \
+ __sanitizer_syscall_post_impl_compat_50_settimeofday(res, (long long)(tv), \
+ (long long)(tzp))
+#define __sanitizer_syscall_pre_fchown(fd, uid, gid) \
+ __sanitizer_syscall_pre_impl_fchown((long long)(fd), (long long)(uid), \
+ (long long)(gid))
+#define __sanitizer_syscall_post_fchown(res, fd, uid, gid) \
+ __sanitizer_syscall_post_impl_fchown(res, (long long)(fd), (long long)(uid), \
+ (long long)(gid))
+#define __sanitizer_syscall_pre_fchmod(fd, mode) \
+ __sanitizer_syscall_pre_impl_fchmod((long long)(fd), (long long)(mode))
+#define __sanitizer_syscall_post_fchmod(res, fd, mode) \
+ __sanitizer_syscall_post_impl_fchmod(res, (long long)(fd), (long long)(mode))
+#define __sanitizer_syscall_pre_compat_43_orecvfrom(s, buf, len, flags, from, \
+ fromlenaddr) \
+ __sanitizer_syscall_pre_impl_compat_43_orecvfrom( \
+ (long long)(s), (long long)(buf), (long long)(len), (long long)(flags), \
+ (long long)(from), (long long)(fromlenaddr))
+#define __sanitizer_syscall_post_compat_43_orecvfrom(res, s, buf, len, flags, \
+ from, fromlenaddr) \
+ __sanitizer_syscall_post_impl_compat_43_orecvfrom( \
+ res, (long long)(s), (long long)(buf), (long long)(len), \
+ (long long)(flags), (long long)(from), (long long)(fromlenaddr))
+#define __sanitizer_syscall_pre_setreuid(ruid, euid) \
+ __sanitizer_syscall_pre_impl_setreuid((long long)(ruid), (long long)(euid))
+#define __sanitizer_syscall_post_setreuid(res, ruid, euid) \
+ __sanitizer_syscall_post_impl_setreuid(res, (long long)(ruid), \
+ (long long)(euid))
+#define __sanitizer_syscall_pre_setregid(rgid, egid) \
+ __sanitizer_syscall_pre_impl_setregid((long long)(rgid), (long long)(egid))
+#define __sanitizer_syscall_post_setregid(res, rgid, egid) \
+ __sanitizer_syscall_post_impl_setregid(res, (long long)(rgid), \
+ (long long)(egid))
+#define __sanitizer_syscall_pre_rename(from, to) \
+ __sanitizer_syscall_pre_impl_rename((long long)(from), (long long)(to))
+#define __sanitizer_syscall_post_rename(res, from, to) \
+ __sanitizer_syscall_post_impl_rename(res, (long long)(from), (long long)(to))
+#define __sanitizer_syscall_pre_compat_43_otruncate(path, length) \
+ __sanitizer_syscall_pre_impl_compat_43_otruncate((long long)(path), \
+ (long long)(length))
+#define __sanitizer_syscall_post_compat_43_otruncate(res, path, length) \
+ __sanitizer_syscall_post_impl_compat_43_otruncate(res, (long long)(path), \
+ (long long)(length))
+#define __sanitizer_syscall_pre_compat_43_oftruncate(fd, length) \
+ __sanitizer_syscall_pre_impl_compat_43_oftruncate((long long)(fd), \
+ (long long)(length))
+#define __sanitizer_syscall_post_compat_43_oftruncate(res, fd, length) \
+ __sanitizer_syscall_post_impl_compat_43_oftruncate(res, (long long)(fd), \
+ (long long)(length))
+#define __sanitizer_syscall_pre_flock(fd, how) \
+ __sanitizer_syscall_pre_impl_flock((long long)(fd), (long long)(how))
+#define __sanitizer_syscall_post_flock(res, fd, how) \
+ __sanitizer_syscall_post_impl_flock(res, (long long)(fd), (long long)(how))
+#define __sanitizer_syscall_pre_mkfifo(path, mode) \
+ __sanitizer_syscall_pre_impl_mkfifo((long long)(path), (long long)(mode))
+#define __sanitizer_syscall_post_mkfifo(res, path, mode) \
+ __sanitizer_syscall_post_impl_mkfifo(res, (long long)(path), \
+ (long long)(mode))
+#define __sanitizer_syscall_pre_sendto(s, buf, len, flags, to, tolen) \
+ __sanitizer_syscall_pre_impl_sendto((long long)(s), (long long)(buf), \
+ (long long)(len), (long long)(flags), \
+ (long long)(to), (long long)(tolen))
+#define __sanitizer_syscall_post_sendto(res, s, buf, len, flags, to, tolen) \
+ __sanitizer_syscall_post_impl_sendto(res, (long long)(s), (long long)(buf), \
+ (long long)(len), (long long)(flags), \
+ (long long)(to), (long long)(tolen))
+#define __sanitizer_syscall_pre_shutdown(s, how) \
+ __sanitizer_syscall_pre_impl_shutdown((long long)(s), (long long)(how))
+#define __sanitizer_syscall_post_shutdown(res, s, how) \
+ __sanitizer_syscall_post_impl_shutdown(res, (long long)(s), (long long)(how))
+#define __sanitizer_syscall_pre_socketpair(domain, type, protocol, rsv) \
+ __sanitizer_syscall_pre_impl_socketpair( \
+ (long long)(domain), (long long)(type), (long long)(protocol), \
+ (long long)(rsv))
+#define __sanitizer_syscall_post_socketpair(res, domain, type, protocol, rsv) \
+ __sanitizer_syscall_post_impl_socketpair( \
+ res, (long long)(domain), (long long)(type), (long long)(protocol), \
+ (long long)(rsv))
+#define __sanitizer_syscall_pre_mkdir(path, mode) \
+ __sanitizer_syscall_pre_impl_mkdir((long long)(path), (long long)(mode))
+#define __sanitizer_syscall_post_mkdir(res, path, mode) \
+ __sanitizer_syscall_post_impl_mkdir(res, (long long)(path), (long long)(mode))
+#define __sanitizer_syscall_pre_rmdir(path) \
+ __sanitizer_syscall_pre_impl_rmdir((long long)(path))
+#define __sanitizer_syscall_post_rmdir(res, path) \
+ __sanitizer_syscall_post_impl_rmdir(res, (long long)(path))
+#define __sanitizer_syscall_pre_compat_50_utimes(path, tptr) \
+ __sanitizer_syscall_pre_impl_compat_50_utimes((long long)(path), \
+ (long long)(tptr))
+#define __sanitizer_syscall_post_compat_50_utimes(res, path, tptr) \
+ __sanitizer_syscall_post_impl_compat_50_utimes(res, (long long)(path), \
+ (long long)(tptr))
+/* syscall 139 has been skipped */
+#define __sanitizer_syscall_pre_compat_50_adjtime(delta, olddelta) \
+ __sanitizer_syscall_pre_impl_compat_50_adjtime((long long)(delta), \
+ (long long)(olddelta))
+#define __sanitizer_syscall_post_compat_50_adjtime(res, delta, olddelta) \
+ __sanitizer_syscall_post_impl_compat_50_adjtime(res, (long long)(delta), \
+ (long long)(olddelta))
+#define __sanitizer_syscall_pre_compat_43_ogetpeername(fdes, asa, alen) \
+ __sanitizer_syscall_pre_impl_compat_43_ogetpeername( \
+ (long long)(fdes), (long long)(asa), (long long)(alen))
+#define __sanitizer_syscall_post_compat_43_ogetpeername(res, fdes, asa, alen) \
+ __sanitizer_syscall_post_impl_compat_43_ogetpeername( \
+ res, (long long)(fdes), (long long)(asa), (long long)(alen))
+#define __sanitizer_syscall_pre_compat_43_ogethostid() \
+ __sanitizer_syscall_pre_impl_compat_43_ogethostid()
+#define __sanitizer_syscall_post_compat_43_ogethostid(res) \
+ __sanitizer_syscall_post_impl_compat_43_ogethostid(res)
+#define __sanitizer_syscall_pre_compat_43_osethostid(hostid) \
+ __sanitizer_syscall_pre_impl_compat_43_osethostid((long long)(hostid))
+#define __sanitizer_syscall_post_compat_43_osethostid(res, hostid) \
+ __sanitizer_syscall_post_impl_compat_43_osethostid(res, (long long)(hostid))
+#define __sanitizer_syscall_pre_compat_43_ogetrlimit(which, rlp) \
+ __sanitizer_syscall_pre_impl_compat_43_ogetrlimit((long long)(which), \
+ (long long)(rlp))
+#define __sanitizer_syscall_post_compat_43_ogetrlimit(res, which, rlp) \
+ __sanitizer_syscall_post_impl_compat_43_ogetrlimit(res, (long long)(which), \
+ (long long)(rlp))
+#define __sanitizer_syscall_pre_compat_43_osetrlimit(which, rlp) \
+ __sanitizer_syscall_pre_impl_compat_43_osetrlimit((long long)(which), \
+ (long long)(rlp))
+#define __sanitizer_syscall_post_compat_43_osetrlimit(res, which, rlp) \
+ __sanitizer_syscall_post_impl_compat_43_osetrlimit(res, (long long)(which), \
+ (long long)(rlp))
+#define __sanitizer_syscall_pre_compat_43_okillpg(pgid, signum) \
+ __sanitizer_syscall_pre_impl_compat_43_okillpg((long long)(pgid), \
+ (long long)(signum))
+#define __sanitizer_syscall_post_compat_43_okillpg(res, pgid, signum) \
+ __sanitizer_syscall_post_impl_compat_43_okillpg(res, (long long)(pgid), \
+ (long long)(signum))
+#define __sanitizer_syscall_pre_setsid() __sanitizer_syscall_pre_impl_setsid()
+#define __sanitizer_syscall_post_setsid(res) \
+ __sanitizer_syscall_post_impl_setsid(res)
+#define __sanitizer_syscall_pre_compat_50_quotactl(path, cmd, uid, arg) \
+ __sanitizer_syscall_pre_impl_compat_50_quotactl( \
+ (long long)(path), (long long)(cmd), (long long)(uid), (long long)(arg))
+#define __sanitizer_syscall_post_compat_50_quotactl(res, path, cmd, uid, arg) \
+ __sanitizer_syscall_post_impl_compat_50_quotactl( \
+ res, (long long)(path), (long long)(cmd), (long long)(uid), \
+ (long long)(arg))
+#define __sanitizer_syscall_pre_compat_43_oquota() \
+ __sanitizer_syscall_pre_impl_compat_43_oquota()
+#define __sanitizer_syscall_post_compat_43_oquota(res) \
+ __sanitizer_syscall_post_impl_compat_43_oquota(res)
+#define __sanitizer_syscall_pre_compat_43_ogetsockname(fdec, asa, alen) \
+ __sanitizer_syscall_pre_impl_compat_43_ogetsockname( \
+ (long long)(fdec), (long long)(asa), (long long)(alen))
+#define __sanitizer_syscall_post_compat_43_ogetsockname(res, fdec, asa, alen) \
+ __sanitizer_syscall_post_impl_compat_43_ogetsockname( \
+ res, (long long)(fdec), (long long)(asa), (long long)(alen))
+/* syscall 151 has been skipped */
+/* syscall 152 has been skipped */
+/* syscall 153 has been skipped */
+/* syscall 154 has been skipped */
+#define __sanitizer_syscall_pre_nfssvc(flag, argp) \
+ __sanitizer_syscall_pre_impl_nfssvc((long long)(flag), (long long)(argp))
+#define __sanitizer_syscall_post_nfssvc(res, flag, argp) \
+ __sanitizer_syscall_post_impl_nfssvc(res, (long long)(flag), \
+ (long long)(argp))
+#define __sanitizer_syscall_pre_compat_43_ogetdirentries(fd, buf, count, \
+ basep) \
+ __sanitizer_syscall_pre_impl_compat_43_ogetdirentries( \
+ (long long)(fd), (long long)(buf), (long long)(count), \
+ (long long)(basep))
+#define __sanitizer_syscall_post_compat_43_ogetdirentries(res, fd, buf, count, \
+ basep) \
+ __sanitizer_syscall_post_impl_compat_43_ogetdirentries( \
+ res, (long long)(fd), (long long)(buf), (long long)(count), \
+ (long long)(basep))
+#define __sanitizer_syscall_pre_compat_20_statfs(path, buf) \
+ __sanitizer_syscall_pre_impl_compat_20_statfs((long long)(path), \
+ (long long)(buf))
+#define __sanitizer_syscall_post_compat_20_statfs(res, path, buf) \
+ __sanitizer_syscall_post_impl_compat_20_statfs(res, (long long)(path), \
+ (long long)(buf))
+#define __sanitizer_syscall_pre_compat_20_fstatfs(fd, buf) \
+ __sanitizer_syscall_pre_impl_compat_20_fstatfs((long long)(fd), \
+ (long long)(buf))
+#define __sanitizer_syscall_post_compat_20_fstatfs(res, fd, buf) \
+ __sanitizer_syscall_post_impl_compat_20_fstatfs(res, (long long)(fd), \
+ (long long)(buf))
+/* syscall 159 has been skipped */
+/* syscall 160 has been skipped */
+#define __sanitizer_syscall_pre_compat_30_getfh(fname, fhp) \
+ __sanitizer_syscall_pre_impl_compat_30_getfh((long long)(fname), \
+ (long long)(fhp))
+#define __sanitizer_syscall_post_compat_30_getfh(res, fname, fhp) \
+ __sanitizer_syscall_post_impl_compat_30_getfh(res, (long long)(fname), \
+ (long long)(fhp))
+#define __sanitizer_syscall_pre_compat_09_ogetdomainname(domainname, len) \
+ __sanitizer_syscall_pre_impl_compat_09_ogetdomainname( \
+ (long long)(domainname), (long long)(len))
+#define __sanitizer_syscall_post_compat_09_ogetdomainname(res, domainname, \
+ len) \
+ __sanitizer_syscall_post_impl_compat_09_ogetdomainname( \
+ res, (long long)(domainname), (long long)(len))
+#define __sanitizer_syscall_pre_compat_09_osetdomainname(domainname, len) \
+ __sanitizer_syscall_pre_impl_compat_09_osetdomainname( \
+ (long long)(domainname), (long long)(len))
+#define __sanitizer_syscall_post_compat_09_osetdomainname(res, domainname, \
+ len) \
+ __sanitizer_syscall_post_impl_compat_09_osetdomainname( \
+ res, (long long)(domainname), (long long)(len))
+#define __sanitizer_syscall_pre_compat_09_ouname(name) \
+ __sanitizer_syscall_pre_impl_compat_09_ouname((long long)(name))
+#define __sanitizer_syscall_post_compat_09_ouname(res, name) \
+ __sanitizer_syscall_post_impl_compat_09_ouname(res, (long long)(name))
+#define __sanitizer_syscall_pre_sysarch(op, parms) \
+ __sanitizer_syscall_pre_impl_sysarch((long long)(op), (long long)(parms))
+#define __sanitizer_syscall_post_sysarch(res, op, parms) \
+ __sanitizer_syscall_post_impl_sysarch(res, (long long)(op), \
+ (long long)(parms))
+/* syscall 166 has been skipped */
+/* syscall 167 has been skipped */
+/* syscall 168 has been skipped */
+#if !defined(_LP64)
+#define __sanitizer_syscall_pre_compat_10_osemsys(which, a2, a3, a4, a5) \
+ __sanitizer_syscall_pre_impl_compat_10_osemsys( \
+ (long long)(which), (long long)(a2), (long long)(a3), (long long)(a4), \
+ (long long)(a5))
+#define __sanitizer_syscall_post_compat_10_osemsys(res, which, a2, a3, a4, a5) \
+ __sanitizer_syscall_post_impl_compat_10_osemsys( \
+ res, (long long)(which), (long long)(a2), (long long)(a3), \
+ (long long)(a4), (long long)(a5))
+#else
+/* syscall 169 has been skipped */
+#endif
+#if !defined(_LP64)
+#define __sanitizer_syscall_pre_compat_10_omsgsys(which, a2, a3, a4, a5, a6) \
+ __sanitizer_syscall_pre_impl_compat_10_omsgsys( \
+ (long long)(which), (long long)(a2), (long long)(a3), (long long)(a4), \
+ (long long)(a5), (long long)(a6))
+#define __sanitizer_syscall_post_compat_10_omsgsys(res, which, a2, a3, a4, a5, \
+ a6) \
+ __sanitizer_syscall_post_impl_compat_10_omsgsys( \
+ res, (long long)(which), (long long)(a2), (long long)(a3), \
+ (long long)(a4), (long long)(a5), (long long)(a6))
+#else
+/* syscall 170 has been skipped */
+#endif
+#if !defined(_LP64)
+#define __sanitizer_syscall_pre_compat_10_oshmsys(which, a2, a3, a4) \
+ __sanitizer_syscall_pre_impl_compat_10_oshmsys( \
+ (long long)(which), (long long)(a2), (long long)(a3), (long long)(a4))
+#define __sanitizer_syscall_post_compat_10_oshmsys(res, which, a2, a3, a4) \
+ __sanitizer_syscall_post_impl_compat_10_oshmsys( \
+ res, (long long)(which), (long long)(a2), (long long)(a3), \
+ (long long)(a4))
+#else
+/* syscall 171 has been skipped */
+#endif
+/* syscall 172 has been skipped */
+#define __sanitizer_syscall_pre_pread(fd, buf, nbyte, PAD, offset) \
+ __sanitizer_syscall_pre_impl_pread((long long)(fd), (long long)(buf), \
+ (long long)(nbyte), (long long)(PAD), \
+ (long long)(offset))
+#define __sanitizer_syscall_post_pread(res, fd, buf, nbyte, PAD, offset) \
+ __sanitizer_syscall_post_impl_pread(res, (long long)(fd), (long long)(buf), \
+ (long long)(nbyte), (long long)(PAD), \
+ (long long)(offset))
+#define __sanitizer_syscall_pre_pwrite(fd, buf, nbyte, PAD, offset) \
+ __sanitizer_syscall_pre_impl_pwrite((long long)(fd), (long long)(buf), \
+ (long long)(nbyte), (long long)(PAD), \
+ (long long)(offset))
+#define __sanitizer_syscall_post_pwrite(res, fd, buf, nbyte, PAD, offset) \
+ __sanitizer_syscall_post_impl_pwrite(res, (long long)(fd), (long long)(buf), \
+ (long long)(nbyte), (long long)(PAD), \
+ (long long)(offset))
+#define __sanitizer_syscall_pre_compat_30_ntp_gettime(ntvp) \
+ __sanitizer_syscall_pre_impl_compat_30_ntp_gettime((long long)(ntvp))
+#define __sanitizer_syscall_post_compat_30_ntp_gettime(res, ntvp) \
+ __sanitizer_syscall_post_impl_compat_30_ntp_gettime(res, (long long)(ntvp))
+#if defined(NTP) || !defined(_KERNEL_OPT)
+#define __sanitizer_syscall_pre_ntp_adjtime(tp) \
+ __sanitizer_syscall_pre_impl_ntp_adjtime((long long)(tp))
+#define __sanitizer_syscall_post_ntp_adjtime(res, tp) \
+ __sanitizer_syscall_post_impl_ntp_adjtime(res, (long long)(tp))
+#else
+/* syscall 176 has been skipped */
+#endif
+/* syscall 177 has been skipped */
+/* syscall 178 has been skipped */
+/* syscall 179 has been skipped */
+/* syscall 180 has been skipped */
+#define __sanitizer_syscall_pre_setgid(gid) \
+ __sanitizer_syscall_pre_impl_setgid((long long)(gid))
+#define __sanitizer_syscall_post_setgid(res, gid) \
+ __sanitizer_syscall_post_impl_setgid(res, (long long)(gid))
+#define __sanitizer_syscall_pre_setegid(egid) \
+ __sanitizer_syscall_pre_impl_setegid((long long)(egid))
+#define __sanitizer_syscall_post_setegid(res, egid) \
+ __sanitizer_syscall_post_impl_setegid(res, (long long)(egid))
+#define __sanitizer_syscall_pre_seteuid(euid) \
+ __sanitizer_syscall_pre_impl_seteuid((long long)(euid))
+#define __sanitizer_syscall_post_seteuid(res, euid) \
+ __sanitizer_syscall_post_impl_seteuid(res, (long long)(euid))
+#define __sanitizer_syscall_pre_lfs_bmapv(fsidp, blkiov, blkcnt) \
+ __sanitizer_syscall_pre_impl_lfs_bmapv( \
+ (long long)(fsidp), (long long)(blkiov), (long long)(blkcnt))
+#define __sanitizer_syscall_post_lfs_bmapv(res, fsidp, blkiov, blkcnt) \
+ __sanitizer_syscall_post_impl_lfs_bmapv( \
+ res, (long long)(fsidp), (long long)(blkiov), (long long)(blkcnt))
+#define __sanitizer_syscall_pre_lfs_markv(fsidp, blkiov, blkcnt) \
+ __sanitizer_syscall_pre_impl_lfs_markv( \
+ (long long)(fsidp), (long long)(blkiov), (long long)(blkcnt))
+#define __sanitizer_syscall_post_lfs_markv(res, fsidp, blkiov, blkcnt) \
+ __sanitizer_syscall_post_impl_lfs_markv( \
+ res, (long long)(fsidp), (long long)(blkiov), (long long)(blkcnt))
+#define __sanitizer_syscall_pre_lfs_segclean(fsidp, segment) \
+ __sanitizer_syscall_pre_impl_lfs_segclean((long long)(fsidp), \
+ (long long)(segment))
+#define __sanitizer_syscall_post_lfs_segclean(res, fsidp, segment) \
+ __sanitizer_syscall_post_impl_lfs_segclean(res, (long long)(fsidp), \
+ (long long)(segment))
+#define __sanitizer_syscall_pre_compat_50_lfs_segwait(fsidp, tv) \
+ __sanitizer_syscall_pre_impl_compat_50_lfs_segwait((long long)(fsidp), \
+ (long long)(tv))
+#define __sanitizer_syscall_post_compat_50_lfs_segwait(res, fsidp, tv) \
+ __sanitizer_syscall_post_impl_compat_50_lfs_segwait(res, (long long)(fsidp), \
+ (long long)(tv))
+#define __sanitizer_syscall_pre_compat_12_stat12(path, ub) \
+ __sanitizer_syscall_pre_impl_compat_12_stat12((long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_post_compat_12_stat12(res, path, ub) \
+ __sanitizer_syscall_post_impl_compat_12_stat12(res, (long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_pre_compat_12_fstat12(fd, sb) \
+ __sanitizer_syscall_pre_impl_compat_12_fstat12((long long)(fd), \
+ (long long)(sb))
+#define __sanitizer_syscall_post_compat_12_fstat12(res, fd, sb) \
+ __sanitizer_syscall_post_impl_compat_12_fstat12(res, (long long)(fd), \
+ (long long)(sb))
+#define __sanitizer_syscall_pre_compat_12_lstat12(path, ub) \
+ __sanitizer_syscall_pre_impl_compat_12_lstat12((long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_post_compat_12_lstat12(res, path, ub) \
+ __sanitizer_syscall_post_impl_compat_12_lstat12(res, (long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_pre_pathconf(path, name) \
+ __sanitizer_syscall_pre_impl_pathconf((long long)(path), (long long)(name))
+#define __sanitizer_syscall_post_pathconf(res, path, name) \
+ __sanitizer_syscall_post_impl_pathconf(res, (long long)(path), \
+ (long long)(name))
+#define __sanitizer_syscall_pre_fpathconf(fd, name) \
+ __sanitizer_syscall_pre_impl_fpathconf((long long)(fd), (long long)(name))
+#define __sanitizer_syscall_post_fpathconf(res, fd, name) \
+ __sanitizer_syscall_post_impl_fpathconf(res, (long long)(fd), \
+ (long long)(name))
+#define __sanitizer_syscall_pre_getsockopt2(s, level, name, val, avalsize) \
+ __sanitizer_syscall_pre_impl_getsockopt2( \
+ (long long)(s), (long long)(level), (long long)(name), (long long)(val), \
+ (long long)(avalsize))
+#define __sanitizer_syscall_post_getsockopt2(res, s, level, name, val, \
+ avalsize) \
+ __sanitizer_syscall_post_impl_getsockopt2( \
+ res, (long long)(s), (long long)(level), (long long)(name), \
+ (long long)(val), (long long)(avalsize))
+#define __sanitizer_syscall_pre_getrlimit(which, rlp) \
+ __sanitizer_syscall_pre_impl_getrlimit((long long)(which), (long long)(rlp))
+#define __sanitizer_syscall_post_getrlimit(res, which, rlp) \
+ __sanitizer_syscall_post_impl_getrlimit(res, (long long)(which), \
+ (long long)(rlp))
+#define __sanitizer_syscall_pre_setrlimit(which, rlp) \
+ __sanitizer_syscall_pre_impl_setrlimit((long long)(which), (long long)(rlp))
+#define __sanitizer_syscall_post_setrlimit(res, which, rlp) \
+ __sanitizer_syscall_post_impl_setrlimit(res, (long long)(which), \
+ (long long)(rlp))
+#define __sanitizer_syscall_pre_compat_12_getdirentries(fd, buf, count, basep) \
+ __sanitizer_syscall_pre_impl_compat_12_getdirentries( \
+ (long long)(fd), (long long)(buf), (long long)(count), \
+ (long long)(basep))
+#define __sanitizer_syscall_post_compat_12_getdirentries(res, fd, buf, count, \
+ basep) \
+ __sanitizer_syscall_post_impl_compat_12_getdirentries( \
+ res, (long long)(fd), (long long)(buf), (long long)(count), \
+ (long long)(basep))
+#define __sanitizer_syscall_pre_mmap(addr, len, prot, flags, fd, PAD, pos) \
+ __sanitizer_syscall_pre_impl_mmap( \
+ (long long)(addr), (long long)(len), (long long)(prot), \
+ (long long)(flags), (long long)(fd), (long long)(PAD), (long long)(pos))
+#define __sanitizer_syscall_post_mmap(res, addr, len, prot, flags, fd, PAD, \
+ pos) \
+ __sanitizer_syscall_post_impl_mmap( \
+ res, (long long)(addr), (long long)(len), (long long)(prot), \
+ (long long)(flags), (long long)(fd), (long long)(PAD), (long long)(pos))
+#define __sanitizer_syscall_pre___syscall(code, arg0, arg1, arg2, arg3, arg4, \
+ arg5, arg6, arg7) \
+ __sanitizer_syscall_pre_impl___syscall( \
+ (long long)(code), (long long)(arg0), (long long)(arg1), \
+ (long long)(arg2), (long long)(arg3), (long long)(arg4), \
+ (long long)(arg5), (long long)(arg6), (long long)(arg7))
+#define __sanitizer_syscall_post___syscall(res, code, arg0, arg1, arg2, arg3, \
+ arg4, arg5, arg6, arg7) \
+ __sanitizer_syscall_post_impl___syscall( \
+ res, (long long)(code), (long long)(arg0), (long long)(arg1), \
+ (long long)(arg2), (long long)(arg3), (long long)(arg4), \
+ (long long)(arg5), (long long)(arg6), (long long)(arg7))
+#define __sanitizer_syscall_pre_lseek(fd, PAD, offset, whence) \
+ __sanitizer_syscall_pre_impl_lseek((long long)(fd), (long long)(PAD), \
+ (long long)(offset), (long long)(whence))
+#define __sanitizer_syscall_post_lseek(res, fd, PAD, offset, whence) \
+ __sanitizer_syscall_post_impl_lseek(res, (long long)(fd), (long long)(PAD), \
+ (long long)(offset), \
+ (long long)(whence))
+#define __sanitizer_syscall_pre_truncate(path, PAD, length) \
+ __sanitizer_syscall_pre_impl_truncate((long long)(path), (long long)(PAD), \
+ (long long)(length))
+#define __sanitizer_syscall_post_truncate(res, path, PAD, length) \
+ __sanitizer_syscall_post_impl_truncate( \
+ res, (long long)(path), (long long)(PAD), (long long)(length))
+#define __sanitizer_syscall_pre_ftruncate(fd, PAD, length) \
+ __sanitizer_syscall_pre_impl_ftruncate((long long)(fd), (long long)(PAD), \
+ (long long)(length))
+#define __sanitizer_syscall_post_ftruncate(res, fd, PAD, length) \
+ __sanitizer_syscall_post_impl_ftruncate( \
+ res, (long long)(fd), (long long)(PAD), (long long)(length))
+#define __sanitizer_syscall_pre___sysctl(name, namelen, oldv, oldlenp, newv, \
+ newlen) \
+ __sanitizer_syscall_pre_impl___sysctl( \
+ (long long)(name), (long long)(namelen), (long long)(oldv), \
+ (long long)(oldlenp), (long long)(newv), (long long)(newlen))
+#define __sanitizer_syscall_post___sysctl(res, name, namelen, oldv, oldlenp, \
+ newv, newlen) \
+ __sanitizer_syscall_post_impl___sysctl( \
+ res, (long long)(name), (long long)(namelen), (long long)(oldv), \
+ (long long)(oldlenp), (long long)(newv), (long long)(newlen))
+#define __sanitizer_syscall_pre_mlock(addr, len) \
+ __sanitizer_syscall_pre_impl_mlock((long long)(addr), (long long)(len))
+#define __sanitizer_syscall_post_mlock(res, addr, len) \
+ __sanitizer_syscall_post_impl_mlock(res, (long long)(addr), (long long)(len))
+#define __sanitizer_syscall_pre_munlock(addr, len) \
+ __sanitizer_syscall_pre_impl_munlock((long long)(addr), (long long)(len))
+#define __sanitizer_syscall_post_munlock(res, addr, len) \
+ __sanitizer_syscall_post_impl_munlock(res, (long long)(addr), \
+ (long long)(len))
+#define __sanitizer_syscall_pre_undelete(path) \
+ __sanitizer_syscall_pre_impl_undelete((long long)(path))
+#define __sanitizer_syscall_post_undelete(res, path) \
+ __sanitizer_syscall_post_impl_undelete(res, (long long)(path))
+#define __sanitizer_syscall_pre_compat_50_futimes(fd, tptr) \
+ __sanitizer_syscall_pre_impl_compat_50_futimes((long long)(fd), \
+ (long long)(tptr))
+#define __sanitizer_syscall_post_compat_50_futimes(res, fd, tptr) \
+ __sanitizer_syscall_post_impl_compat_50_futimes(res, (long long)(fd), \
+ (long long)(tptr))
+#define __sanitizer_syscall_pre_getpgid(pid) \
+ __sanitizer_syscall_pre_impl_getpgid((long long)(pid))
+#define __sanitizer_syscall_post_getpgid(res, pid) \
+ __sanitizer_syscall_post_impl_getpgid(res, (long long)(pid))
+#define __sanitizer_syscall_pre_reboot(opt, bootstr) \
+ __sanitizer_syscall_pre_impl_reboot((long long)(opt), (long long)(bootstr))
+#define __sanitizer_syscall_post_reboot(res, opt, bootstr) \
+ __sanitizer_syscall_post_impl_reboot(res, (long long)(opt), \
+ (long long)(bootstr))
+#define __sanitizer_syscall_pre_poll(fds, nfds, timeout) \
+ __sanitizer_syscall_pre_impl_poll((long long)(fds), (long long)(nfds), \
+ (long long)(timeout))
+#define __sanitizer_syscall_post_poll(res, fds, nfds, timeout) \
+ __sanitizer_syscall_post_impl_poll(res, (long long)(fds), (long long)(nfds), \
+ (long long)(timeout))
+#define __sanitizer_syscall_pre_afssys(id, a1, a2, a3, a4, a5, a6) \
+ __sanitizer_syscall_pre_impl_afssys( \
+ (long long)(id), (long long)(a1), (long long)(a2), (long long)(a3), \
+ (long long)(a4), (long long)(a5), (long long)(a6))
+#define __sanitizer_syscall_post_afssys(res, id, a1, a2, a3, a4, a5, a6) \
+ __sanitizer_syscall_post_impl_afssys( \
+ res, (long long)(id), (long long)(a1), (long long)(a2), (long long)(a3), \
+ (long long)(a4), (long long)(a5), (long long)(a6))
+/* syscall 211 has been skipped */
+/* syscall 212 has been skipped */
+/* syscall 213 has been skipped */
+/* syscall 214 has been skipped */
+/* syscall 215 has been skipped */
+/* syscall 216 has been skipped */
+/* syscall 217 has been skipped */
+/* syscall 218 has been skipped */
+/* syscall 219 has been skipped */
+#define __sanitizer_syscall_pre_compat_14___semctl(semid, semnum, cmd, arg) \
+ __sanitizer_syscall_pre_impl_compat_14___semctl( \
+ (long long)(semid), (long long)(semnum), (long long)(cmd), \
+ (long long)(arg))
+#define __sanitizer_syscall_post_compat_14___semctl(res, semid, semnum, cmd, \
+ arg) \
+ __sanitizer_syscall_post_impl_compat_14___semctl( \
+ res, (long long)(semid), (long long)(semnum), (long long)(cmd), \
+ (long long)(arg))
+#define __sanitizer_syscall_pre_semget(key, nsems, semflg) \
+ __sanitizer_syscall_pre_impl_semget((long long)(key), (long long)(nsems), \
+ (long long)(semflg))
+#define __sanitizer_syscall_post_semget(res, key, nsems, semflg) \
+ __sanitizer_syscall_post_impl_semget( \
+ res, (long long)(key), (long long)(nsems), (long long)(semflg))
+#define __sanitizer_syscall_pre_semop(semid, sops, nsops) \
+ __sanitizer_syscall_pre_impl_semop((long long)(semid), (long long)(sops), \
+ (long long)(nsops))
+#define __sanitizer_syscall_post_semop(res, semid, sops, nsops) \
+ __sanitizer_syscall_post_impl_semop(res, (long long)(semid), \
+ (long long)(sops), (long long)(nsops))
+#define __sanitizer_syscall_pre_semconfig(flag) \
+ __sanitizer_syscall_pre_impl_semconfig((long long)(flag))
+#define __sanitizer_syscall_post_semconfig(res, flag) \
+ __sanitizer_syscall_post_impl_semconfig(res, (long long)(flag))
+#define __sanitizer_syscall_pre_compat_14_msgctl(msqid, cmd, buf) \
+ __sanitizer_syscall_pre_impl_compat_14_msgctl( \
+ (long long)(msqid), (long long)(cmd), (long long)(buf))
+#define __sanitizer_syscall_post_compat_14_msgctl(res, msqid, cmd, buf) \
+ __sanitizer_syscall_post_impl_compat_14_msgctl( \
+ res, (long long)(msqid), (long long)(cmd), (long long)(buf))
+#define __sanitizer_syscall_pre_msgget(key, msgflg) \
+ __sanitizer_syscall_pre_impl_msgget((long long)(key), (long long)(msgflg))
+#define __sanitizer_syscall_post_msgget(res, key, msgflg) \
+ __sanitizer_syscall_post_impl_msgget(res, (long long)(key), \
+ (long long)(msgflg))
+#define __sanitizer_syscall_pre_msgsnd(msqid, msgp, msgsz, msgflg) \
+ __sanitizer_syscall_pre_impl_msgsnd((long long)(msqid), (long long)(msgp), \
+ (long long)(msgsz), (long long)(msgflg))
+#define __sanitizer_syscall_post_msgsnd(res, msqid, msgp, msgsz, msgflg) \
+ __sanitizer_syscall_post_impl_msgsnd(res, (long long)(msqid), \
+ (long long)(msgp), (long long)(msgsz), \
+ (long long)(msgflg))
+#define __sanitizer_syscall_pre_msgrcv(msqid, msgp, msgsz, msgtyp, msgflg) \
+ __sanitizer_syscall_pre_impl_msgrcv((long long)(msqid), (long long)(msgp), \
+ (long long)(msgsz), (long long)(msgtyp), \
+ (long long)(msgflg))
+#define __sanitizer_syscall_post_msgrcv(res, msqid, msgp, msgsz, msgtyp, \
+ msgflg) \
+ __sanitizer_syscall_post_impl_msgrcv( \
+ res, (long long)(msqid), (long long)(msgp), (long long)(msgsz), \
+ (long long)(msgtyp), (long long)(msgflg))
+#define __sanitizer_syscall_pre_shmat(shmid, shmaddr, shmflg) \
+ __sanitizer_syscall_pre_impl_shmat((long long)(shmid), (long long)(shmaddr), \
+ (long long)(shmflg))
+#define __sanitizer_syscall_post_shmat(res, shmid, shmaddr, shmflg) \
+ __sanitizer_syscall_post_impl_shmat( \
+ res, (long long)(shmid), (long long)(shmaddr), (long long)(shmflg))
+#define __sanitizer_syscall_pre_compat_14_shmctl(shmid, cmd, buf) \
+ __sanitizer_syscall_pre_impl_compat_14_shmctl( \
+ (long long)(shmid), (long long)(cmd), (long long)(buf))
+#define __sanitizer_syscall_post_compat_14_shmctl(res, shmid, cmd, buf) \
+ __sanitizer_syscall_post_impl_compat_14_shmctl( \
+ res, (long long)(shmid), (long long)(cmd), (long long)(buf))
+#define __sanitizer_syscall_pre_shmdt(shmaddr) \
+ __sanitizer_syscall_pre_impl_shmdt((long long)(shmaddr))
+#define __sanitizer_syscall_post_shmdt(res, shmaddr) \
+ __sanitizer_syscall_post_impl_shmdt(res, (long long)(shmaddr))
+#define __sanitizer_syscall_pre_shmget(key, size, shmflg) \
+ __sanitizer_syscall_pre_impl_shmget((long long)(key), (long long)(size), \
+ (long long)(shmflg))
+#define __sanitizer_syscall_post_shmget(res, key, size, shmflg) \
+ __sanitizer_syscall_post_impl_shmget(res, (long long)(key), \
+ (long long)(size), (long long)(shmflg))
+#define __sanitizer_syscall_pre_compat_50_clock_gettime(clock_id, tp) \
+ __sanitizer_syscall_pre_impl_compat_50_clock_gettime((long long)(clock_id), \
+ (long long)(tp))
+#define __sanitizer_syscall_post_compat_50_clock_gettime(res, clock_id, tp) \
+ __sanitizer_syscall_post_impl_compat_50_clock_gettime( \
+ res, (long long)(clock_id), (long long)(tp))
+#define __sanitizer_syscall_pre_compat_50_clock_settime(clock_id, tp) \
+ __sanitizer_syscall_pre_impl_compat_50_clock_settime((long long)(clock_id), \
+ (long long)(tp))
+#define __sanitizer_syscall_post_compat_50_clock_settime(res, clock_id, tp) \
+ __sanitizer_syscall_post_impl_compat_50_clock_settime( \
+ res, (long long)(clock_id), (long long)(tp))
+#define __sanitizer_syscall_pre_compat_50_clock_getres(clock_id, tp) \
+ __sanitizer_syscall_pre_impl_compat_50_clock_getres((long long)(clock_id), \
+ (long long)(tp))
+#define __sanitizer_syscall_post_compat_50_clock_getres(res, clock_id, tp) \
+ __sanitizer_syscall_post_impl_compat_50_clock_getres( \
+ res, (long long)(clock_id), (long long)(tp))
+#define __sanitizer_syscall_pre_timer_create(clock_id, evp, timerid) \
+ __sanitizer_syscall_pre_impl_timer_create( \
+ (long long)(clock_id), (long long)(evp), (long long)(timerid))
+#define __sanitizer_syscall_post_timer_create(res, clock_id, evp, timerid) \
+ __sanitizer_syscall_post_impl_timer_create( \
+ res, (long long)(clock_id), (long long)(evp), (long long)(timerid))
+#define __sanitizer_syscall_pre_timer_delete(timerid) \
+ __sanitizer_syscall_pre_impl_timer_delete((long long)(timerid))
+#define __sanitizer_syscall_post_timer_delete(res, timerid) \
+ __sanitizer_syscall_post_impl_timer_delete(res, (long long)(timerid))
+#define __sanitizer_syscall_pre_compat_50_timer_settime(timerid, flags, value, \
+ ovalue) \
+ __sanitizer_syscall_pre_impl_compat_50_timer_settime( \
+ (long long)(timerid), (long long)(flags), (long long)(value), \
+ (long long)(ovalue))
+#define __sanitizer_syscall_post_compat_50_timer_settime(res, timerid, flags, \
+ value, ovalue) \
+ __sanitizer_syscall_post_impl_compat_50_timer_settime( \
+ res, (long long)(timerid), (long long)(flags), (long long)(value), \
+ (long long)(ovalue))
+#define __sanitizer_syscall_pre_compat_50_timer_gettime(timerid, value) \
+ __sanitizer_syscall_pre_impl_compat_50_timer_gettime((long long)(timerid), \
+ (long long)(value))
+#define __sanitizer_syscall_post_compat_50_timer_gettime(res, timerid, value) \
+ __sanitizer_syscall_post_impl_compat_50_timer_gettime( \
+ res, (long long)(timerid), (long long)(value))
+#define __sanitizer_syscall_pre_timer_getoverrun(timerid) \
+ __sanitizer_syscall_pre_impl_timer_getoverrun((long long)(timerid))
+#define __sanitizer_syscall_post_timer_getoverrun(res, timerid) \
+ __sanitizer_syscall_post_impl_timer_getoverrun(res, (long long)(timerid))
+#define __sanitizer_syscall_pre_compat_50_nanosleep(rqtp, rmtp) \
+ __sanitizer_syscall_pre_impl_compat_50_nanosleep((long long)(rqtp), \
+ (long long)(rmtp))
+#define __sanitizer_syscall_post_compat_50_nanosleep(res, rqtp, rmtp) \
+ __sanitizer_syscall_post_impl_compat_50_nanosleep(res, (long long)(rqtp), \
+ (long long)(rmtp))
+#define __sanitizer_syscall_pre_fdatasync(fd) \
+ __sanitizer_syscall_pre_impl_fdatasync((long long)(fd))
+#define __sanitizer_syscall_post_fdatasync(res, fd) \
+ __sanitizer_syscall_post_impl_fdatasync(res, (long long)(fd))
+#define __sanitizer_syscall_pre_mlockall(flags) \
+ __sanitizer_syscall_pre_impl_mlockall((long long)(flags))
+#define __sanitizer_syscall_post_mlockall(res, flags) \
+ __sanitizer_syscall_post_impl_mlockall(res, (long long)(flags))
+#define __sanitizer_syscall_pre_munlockall() \
+ __sanitizer_syscall_pre_impl_munlockall()
+#define __sanitizer_syscall_post_munlockall(res) \
+ __sanitizer_syscall_post_impl_munlockall(res)
+#define __sanitizer_syscall_pre_compat_50___sigtimedwait(set, info, timeout) \
+ __sanitizer_syscall_pre_impl_compat_50___sigtimedwait( \
+ (long long)(set), (long long)(info), (long long)(timeout))
+#define __sanitizer_syscall_post_compat_50___sigtimedwait(res, set, info, \
+ timeout) \
+ __sanitizer_syscall_post_impl_compat_50___sigtimedwait( \
+ res, (long long)(set), (long long)(info), (long long)(timeout))
+#define __sanitizer_syscall_pre_sigqueueinfo(pid, info) \
+ __sanitizer_syscall_pre_impl_sigqueueinfo((long long)(pid), (long long)(info))
+#define __sanitizer_syscall_post_sigqueueinfo(res, pid, info) \
+ __sanitizer_syscall_post_impl_sigqueueinfo(res, (long long)(pid), \
+ (long long)(info))
+#define __sanitizer_syscall_pre_modctl(cmd, arg) \
+ __sanitizer_syscall_pre_impl_modctl((long long)(cmd), (long long)(arg))
+#define __sanitizer_syscall_post_modctl(res, cmd, arg) \
+ __sanitizer_syscall_post_impl_modctl(res, (long long)(cmd), (long long)(arg))
+#define __sanitizer_syscall_pre__ksem_init(value, idp) \
+ __sanitizer_syscall_pre_impl__ksem_init((long long)(value), (long long)(idp))
+#define __sanitizer_syscall_post__ksem_init(res, value, idp) \
+ __sanitizer_syscall_post_impl__ksem_init(res, (long long)(value), \
+ (long long)(idp))
+#define __sanitizer_syscall_pre__ksem_open(name, oflag, mode, value, idp) \
+ __sanitizer_syscall_pre_impl__ksem_open( \
+ (long long)(name), (long long)(oflag), (long long)(mode), \
+ (long long)(value), (long long)(idp))
+#define __sanitizer_syscall_post__ksem_open(res, name, oflag, mode, value, \
+ idp) \
+ __sanitizer_syscall_post_impl__ksem_open( \
+ res, (long long)(name), (long long)(oflag), (long long)(mode), \
+ (long long)(value), (long long)(idp))
+#define __sanitizer_syscall_pre__ksem_unlink(name) \
+ __sanitizer_syscall_pre_impl__ksem_unlink((long long)(name))
+#define __sanitizer_syscall_post__ksem_unlink(res, name) \
+ __sanitizer_syscall_post_impl__ksem_unlink(res, (long long)(name))
+#define __sanitizer_syscall_pre__ksem_close(id) \
+ __sanitizer_syscall_pre_impl__ksem_close((long long)(id))
+#define __sanitizer_syscall_post__ksem_close(res, id) \
+ __sanitizer_syscall_post_impl__ksem_close(res, (long long)(id))
+#define __sanitizer_syscall_pre__ksem_post(id) \
+ __sanitizer_syscall_pre_impl__ksem_post((long long)(id))
+#define __sanitizer_syscall_post__ksem_post(res, id) \
+ __sanitizer_syscall_post_impl__ksem_post(res, (long long)(id))
+#define __sanitizer_syscall_pre__ksem_wait(id) \
+ __sanitizer_syscall_pre_impl__ksem_wait((long long)(id))
+#define __sanitizer_syscall_post__ksem_wait(res, id) \
+ __sanitizer_syscall_post_impl__ksem_wait(res, (long long)(id))
+#define __sanitizer_syscall_pre__ksem_trywait(id) \
+ __sanitizer_syscall_pre_impl__ksem_trywait((long long)(id))
+#define __sanitizer_syscall_post__ksem_trywait(res, id) \
+ __sanitizer_syscall_post_impl__ksem_trywait(res, (long long)(id))
+#define __sanitizer_syscall_pre__ksem_getvalue(id, value) \
+ __sanitizer_syscall_pre_impl__ksem_getvalue((long long)(id), \
+ (long long)(value))
+#define __sanitizer_syscall_post__ksem_getvalue(res, id, value) \
+ __sanitizer_syscall_post_impl__ksem_getvalue(res, (long long)(id), \
+ (long long)(value))
+#define __sanitizer_syscall_pre__ksem_destroy(id) \
+ __sanitizer_syscall_pre_impl__ksem_destroy((long long)(id))
+#define __sanitizer_syscall_post__ksem_destroy(res, id) \
+ __sanitizer_syscall_post_impl__ksem_destroy(res, (long long)(id))
+#define __sanitizer_syscall_pre__ksem_timedwait(id, abstime) \
+ __sanitizer_syscall_pre_impl__ksem_timedwait((long long)(id), \
+ (long long)(abstime))
+#define __sanitizer_syscall_post__ksem_timedwait(res, id, abstime) \
+ __sanitizer_syscall_post_impl__ksem_timedwait(res, (long long)(id), \
+ (long long)(abstime))
+#define __sanitizer_syscall_pre_mq_open(name, oflag, mode, attr) \
+ __sanitizer_syscall_pre_impl_mq_open((long long)(name), (long long)(oflag), \
+ (long long)(mode), (long long)(attr))
+#define __sanitizer_syscall_post_mq_open(res, name, oflag, mode, attr) \
+ __sanitizer_syscall_post_impl_mq_open(res, (long long)(name), \
+ (long long)(oflag), (long long)(mode), \
+ (long long)(attr))
+#define __sanitizer_syscall_pre_mq_close(mqdes) \
+ __sanitizer_syscall_pre_impl_mq_close((long long)(mqdes))
+#define __sanitizer_syscall_post_mq_close(res, mqdes) \
+ __sanitizer_syscall_post_impl_mq_close(res, (long long)(mqdes))
+#define __sanitizer_syscall_pre_mq_unlink(name) \
+ __sanitizer_syscall_pre_impl_mq_unlink((long long)(name))
+#define __sanitizer_syscall_post_mq_unlink(res, name) \
+ __sanitizer_syscall_post_impl_mq_unlink(res, (long long)(name))
+#define __sanitizer_syscall_pre_mq_getattr(mqdes, mqstat) \
+ __sanitizer_syscall_pre_impl_mq_getattr((long long)(mqdes), \
+ (long long)(mqstat))
+#define __sanitizer_syscall_post_mq_getattr(res, mqdes, mqstat) \
+ __sanitizer_syscall_post_impl_mq_getattr(res, (long long)(mqdes), \
+ (long long)(mqstat))
+#define __sanitizer_syscall_pre_mq_setattr(mqdes, mqstat, omqstat) \
+ __sanitizer_syscall_pre_impl_mq_setattr( \
+ (long long)(mqdes), (long long)(mqstat), (long long)(omqstat))
+#define __sanitizer_syscall_post_mq_setattr(res, mqdes, mqstat, omqstat) \
+ __sanitizer_syscall_post_impl_mq_setattr( \
+ res, (long long)(mqdes), (long long)(mqstat), (long long)(omqstat))
+#define __sanitizer_syscall_pre_mq_notify(mqdes, notification) \
+ __sanitizer_syscall_pre_impl_mq_notify((long long)(mqdes), \
+ (long long)(notification))
+#define __sanitizer_syscall_post_mq_notify(res, mqdes, notification) \
+ __sanitizer_syscall_post_impl_mq_notify(res, (long long)(mqdes), \
+ (long long)(notification))
+#define __sanitizer_syscall_pre_mq_send(mqdes, msg_ptr, msg_len, msg_prio) \
+ __sanitizer_syscall_pre_impl_mq_send( \
+ (long long)(mqdes), (long long)(msg_ptr), (long long)(msg_len), \
+ (long long)(msg_prio))
+#define __sanitizer_syscall_post_mq_send(res, mqdes, msg_ptr, msg_len, \
+ msg_prio) \
+ __sanitizer_syscall_post_impl_mq_send( \
+ res, (long long)(mqdes), (long long)(msg_ptr), (long long)(msg_len), \
+ (long long)(msg_prio))
+#define __sanitizer_syscall_pre_mq_receive(mqdes, msg_ptr, msg_len, msg_prio) \
+ __sanitizer_syscall_pre_impl_mq_receive( \
+ (long long)(mqdes), (long long)(msg_ptr), (long long)(msg_len), \
+ (long long)(msg_prio))
+#define __sanitizer_syscall_post_mq_receive(res, mqdes, msg_ptr, msg_len, \
+ msg_prio) \
+ __sanitizer_syscall_post_impl_mq_receive( \
+ res, (long long)(mqdes), (long long)(msg_ptr), (long long)(msg_len), \
+ (long long)(msg_prio))
+#define __sanitizer_syscall_pre_compat_50_mq_timedsend( \
+ mqdes, msg_ptr, msg_len, msg_prio, abs_timeout) \
+ __sanitizer_syscall_pre_impl_compat_50_mq_timedsend( \
+ (long long)(mqdes), (long long)(msg_ptr), (long long)(msg_len), \
+ (long long)(msg_prio), (long long)(abs_timeout))
+#define __sanitizer_syscall_post_compat_50_mq_timedsend( \
+ res, mqdes, msg_ptr, msg_len, msg_prio, abs_timeout) \
+ __sanitizer_syscall_post_impl_compat_50_mq_timedsend( \
+ res, (long long)(mqdes), (long long)(msg_ptr), (long long)(msg_len), \
+ (long long)(msg_prio), (long long)(abs_timeout))
+#define __sanitizer_syscall_pre_compat_50_mq_timedreceive( \
+ mqdes, msg_ptr, msg_len, msg_prio, abs_timeout) \
+ __sanitizer_syscall_pre_impl_compat_50_mq_timedreceive( \
+ (long long)(mqdes), (long long)(msg_ptr), (long long)(msg_len), \
+ (long long)(msg_prio), (long long)(abs_timeout))
+#define __sanitizer_syscall_post_compat_50_mq_timedreceive( \
+ res, mqdes, msg_ptr, msg_len, msg_prio, abs_timeout) \
+ __sanitizer_syscall_post_impl_compat_50_mq_timedreceive( \
+ res, (long long)(mqdes), (long long)(msg_ptr), (long long)(msg_len), \
+ (long long)(msg_prio), (long long)(abs_timeout))
+/* syscall 267 has been skipped */
+/* syscall 268 has been skipped */
+/* syscall 269 has been skipped */
+#define __sanitizer_syscall_pre___posix_rename(from, to) \
+ __sanitizer_syscall_pre_impl___posix_rename((long long)(from), \
+ (long long)(to))
+#define __sanitizer_syscall_post___posix_rename(res, from, to) \
+ __sanitizer_syscall_post_impl___posix_rename(res, (long long)(from), \
+ (long long)(to))
+#define __sanitizer_syscall_pre_swapctl(cmd, arg, misc) \
+ __sanitizer_syscall_pre_impl_swapctl((long long)(cmd), (long long)(arg), \
+ (long long)(misc))
+#define __sanitizer_syscall_post_swapctl(res, cmd, arg, misc) \
+ __sanitizer_syscall_post_impl_swapctl(res, (long long)(cmd), \
+ (long long)(arg), (long long)(misc))
+#define __sanitizer_syscall_pre_compat_30_getdents(fd, buf, count) \
+ __sanitizer_syscall_pre_impl_compat_30_getdents( \
+ (long long)(fd), (long long)(buf), (long long)(count))
+#define __sanitizer_syscall_post_compat_30_getdents(res, fd, buf, count) \
+ __sanitizer_syscall_post_impl_compat_30_getdents( \
+ res, (long long)(fd), (long long)(buf), (long long)(count))
+#define __sanitizer_syscall_pre_minherit(addr, len, inherit) \
+ __sanitizer_syscall_pre_impl_minherit((long long)(addr), (long long)(len), \
+ (long long)(inherit))
+#define __sanitizer_syscall_post_minherit(res, addr, len, inherit) \
+ __sanitizer_syscall_post_impl_minherit( \
+ res, (long long)(addr), (long long)(len), (long long)(inherit))
+#define __sanitizer_syscall_pre_lchmod(path, mode) \
+ __sanitizer_syscall_pre_impl_lchmod((long long)(path), (long long)(mode))
+#define __sanitizer_syscall_post_lchmod(res, path, mode) \
+ __sanitizer_syscall_post_impl_lchmod(res, (long long)(path), \
+ (long long)(mode))
+#define __sanitizer_syscall_pre_lchown(path, uid, gid) \
+ __sanitizer_syscall_pre_impl_lchown((long long)(path), (long long)(uid), \
+ (long long)(gid))
+#define __sanitizer_syscall_post_lchown(res, path, uid, gid) \
+ __sanitizer_syscall_post_impl_lchown(res, (long long)(path), \
+ (long long)(uid), (long long)(gid))
+#define __sanitizer_syscall_pre_compat_50_lutimes(path, tptr) \
+ __sanitizer_syscall_pre_impl_compat_50_lutimes((long long)(path), \
+ (long long)(tptr))
+#define __sanitizer_syscall_post_compat_50_lutimes(res, path, tptr) \
+ __sanitizer_syscall_post_impl_compat_50_lutimes(res, (long long)(path), \
+ (long long)(tptr))
+#define __sanitizer_syscall_pre___msync13(addr, len, flags) \
+ __sanitizer_syscall_pre_impl___msync13((long long)(addr), (long long)(len), \
+ (long long)(flags))
+#define __sanitizer_syscall_post___msync13(res, addr, len, flags) \
+ __sanitizer_syscall_post_impl___msync13( \
+ res, (long long)(addr), (long long)(len), (long long)(flags))
+#define __sanitizer_syscall_pre_compat_30___stat13(path, ub) \
+ __sanitizer_syscall_pre_impl_compat_30___stat13((long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_post_compat_30___stat13(res, path, ub) \
+ __sanitizer_syscall_post_impl_compat_30___stat13(res, (long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_pre_compat_30___fstat13(fd, sb) \
+ __sanitizer_syscall_pre_impl_compat_30___fstat13((long long)(fd), \
+ (long long)(sb))
+#define __sanitizer_syscall_post_compat_30___fstat13(res, fd, sb) \
+ __sanitizer_syscall_post_impl_compat_30___fstat13(res, (long long)(fd), \
+ (long long)(sb))
+#define __sanitizer_syscall_pre_compat_30___lstat13(path, ub) \
+ __sanitizer_syscall_pre_impl_compat_30___lstat13((long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_post_compat_30___lstat13(res, path, ub) \
+ __sanitizer_syscall_post_impl_compat_30___lstat13(res, (long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_pre___sigaltstack14(nss, oss) \
+ __sanitizer_syscall_pre_impl___sigaltstack14((long long)(nss), \
+ (long long)(oss))
+#define __sanitizer_syscall_post___sigaltstack14(res, nss, oss) \
+ __sanitizer_syscall_post_impl___sigaltstack14(res, (long long)(nss), \
+ (long long)(oss))
+#define __sanitizer_syscall_pre___vfork14() \
+ __sanitizer_syscall_pre_impl___vfork14()
+#define __sanitizer_syscall_post___vfork14(res) \
+ __sanitizer_syscall_post_impl___vfork14(res)
+#define __sanitizer_syscall_pre___posix_chown(path, uid, gid) \
+ __sanitizer_syscall_pre_impl___posix_chown( \
+ (long long)(path), (long long)(uid), (long long)(gid))
+#define __sanitizer_syscall_post___posix_chown(res, path, uid, gid) \
+ __sanitizer_syscall_post_impl___posix_chown( \
+ res, (long long)(path), (long long)(uid), (long long)(gid))
+#define __sanitizer_syscall_pre___posix_fchown(fd, uid, gid) \
+ __sanitizer_syscall_pre_impl___posix_fchown( \
+ (long long)(fd), (long long)(uid), (long long)(gid))
+#define __sanitizer_syscall_post___posix_fchown(res, fd, uid, gid) \
+ __sanitizer_syscall_post_impl___posix_fchown( \
+ res, (long long)(fd), (long long)(uid), (long long)(gid))
+#define __sanitizer_syscall_pre___posix_lchown(path, uid, gid) \
+ __sanitizer_syscall_pre_impl___posix_lchown( \
+ (long long)(path), (long long)(uid), (long long)(gid))
+#define __sanitizer_syscall_post___posix_lchown(res, path, uid, gid) \
+ __sanitizer_syscall_post_impl___posix_lchown( \
+ res, (long long)(path), (long long)(uid), (long long)(gid))
+#define __sanitizer_syscall_pre_getsid(pid) \
+ __sanitizer_syscall_pre_impl_getsid((long long)(pid))
+#define __sanitizer_syscall_post_getsid(res, pid) \
+ __sanitizer_syscall_post_impl_getsid(res, (long long)(pid))
+#define __sanitizer_syscall_pre___clone(flags, stack) \
+ __sanitizer_syscall_pre_impl___clone((long long)(flags), (long long)(stack))
+#define __sanitizer_syscall_post___clone(res, flags, stack) \
+ __sanitizer_syscall_post_impl___clone(res, (long long)(flags), \
+ (long long)(stack))
+#define __sanitizer_syscall_pre_fktrace(fd, ops, facs, pid) \
+ __sanitizer_syscall_pre_impl_fktrace((long long)(fd), (long long)(ops), \
+ (long long)(facs), (long long)(pid))
+#define __sanitizer_syscall_post_fktrace(res, fd, ops, facs, pid) \
+ __sanitizer_syscall_post_impl_fktrace(res, (long long)(fd), \
+ (long long)(ops), (long long)(facs), \
+ (long long)(pid))
+#define __sanitizer_syscall_pre_preadv(fd, iovp, iovcnt, PAD, offset) \
+ __sanitizer_syscall_pre_impl_preadv((long long)(fd), (long long)(iovp), \
+ (long long)(iovcnt), (long long)(PAD), \
+ (long long)(offset))
+#define __sanitizer_syscall_post_preadv(res, fd, iovp, iovcnt, PAD, offset) \
+ __sanitizer_syscall_post_impl_preadv(res, (long long)(fd), \
+ (long long)(iovp), (long long)(iovcnt), \
+ (long long)(PAD), (long long)(offset))
+#define __sanitizer_syscall_pre_pwritev(fd, iovp, iovcnt, PAD, offset) \
+ __sanitizer_syscall_pre_impl_pwritev((long long)(fd), (long long)(iovp), \
+ (long long)(iovcnt), (long long)(PAD), \
+ (long long)(offset))
+#define __sanitizer_syscall_post_pwritev(res, fd, iovp, iovcnt, PAD, offset) \
+ __sanitizer_syscall_post_impl_pwritev( \
+ res, (long long)(fd), (long long)(iovp), (long long)(iovcnt), \
+ (long long)(PAD), (long long)(offset))
+#define __sanitizer_syscall_pre_compat_16___sigaction14(signum, nsa, osa) \
+ __sanitizer_syscall_pre_impl_compat_16___sigaction14( \
+ (long long)(signum), (long long)(nsa), (long long)(osa))
+#define __sanitizer_syscall_post_compat_16___sigaction14(res, signum, nsa, \
+ osa) \
+ __sanitizer_syscall_post_impl_compat_16___sigaction14( \
+ res, (long long)(signum), (long long)(nsa), (long long)(osa))
+#define __sanitizer_syscall_pre___sigpending14(set) \
+ __sanitizer_syscall_pre_impl___sigpending14((long long)(set))
+#define __sanitizer_syscall_post___sigpending14(res, set) \
+ __sanitizer_syscall_post_impl___sigpending14(res, (long long)(set))
+#define __sanitizer_syscall_pre___sigprocmask14(how, set, oset) \
+ __sanitizer_syscall_pre_impl___sigprocmask14( \
+ (long long)(how), (long long)(set), (long long)(oset))
+#define __sanitizer_syscall_post___sigprocmask14(res, how, set, oset) \
+ __sanitizer_syscall_post_impl___sigprocmask14( \
+ res, (long long)(how), (long long)(set), (long long)(oset))
+#define __sanitizer_syscall_pre___sigsuspend14(set) \
+ __sanitizer_syscall_pre_impl___sigsuspend14((long long)(set))
+#define __sanitizer_syscall_post___sigsuspend14(res, set) \
+ __sanitizer_syscall_post_impl___sigsuspend14(res, (long long)(set))
+#define __sanitizer_syscall_pre_compat_16___sigreturn14(sigcntxp) \
+ __sanitizer_syscall_pre_impl_compat_16___sigreturn14((long long)(sigcntxp))
+#define __sanitizer_syscall_post_compat_16___sigreturn14(res, sigcntxp) \
+ __sanitizer_syscall_post_impl_compat_16___sigreturn14(res, \
+ (long long)(sigcntxp))
+#define __sanitizer_syscall_pre___getcwd(bufp, length) \
+ __sanitizer_syscall_pre_impl___getcwd((long long)(bufp), (long long)(length))
+#define __sanitizer_syscall_post___getcwd(res, bufp, length) \
+ __sanitizer_syscall_post_impl___getcwd(res, (long long)(bufp), \
+ (long long)(length))
+#define __sanitizer_syscall_pre_fchroot(fd) \
+ __sanitizer_syscall_pre_impl_fchroot((long long)(fd))
+#define __sanitizer_syscall_post_fchroot(res, fd) \
+ __sanitizer_syscall_post_impl_fchroot(res, (long long)(fd))
+#define __sanitizer_syscall_pre_compat_30_fhopen(fhp, flags) \
+ __sanitizer_syscall_pre_impl_compat_30_fhopen((long long)(fhp), \
+ (long long)(flags))
+#define __sanitizer_syscall_post_compat_30_fhopen(res, fhp, flags) \
+ __sanitizer_syscall_post_impl_compat_30_fhopen(res, (long long)(fhp), \
+ (long long)(flags))
+#define __sanitizer_syscall_pre_compat_30_fhstat(fhp, sb) \
+ __sanitizer_syscall_pre_impl_compat_30_fhstat((long long)(fhp), \
+ (long long)(sb))
+#define __sanitizer_syscall_post_compat_30_fhstat(res, fhp, sb) \
+ __sanitizer_syscall_post_impl_compat_30_fhstat(res, (long long)(fhp), \
+ (long long)(sb))
+#define __sanitizer_syscall_pre_compat_20_fhstatfs(fhp, buf) \
+ __sanitizer_syscall_pre_impl_compat_20_fhstatfs((long long)(fhp), \
+ (long long)(buf))
+#define __sanitizer_syscall_post_compat_20_fhstatfs(res, fhp, buf) \
+ __sanitizer_syscall_post_impl_compat_20_fhstatfs(res, (long long)(fhp), \
+ (long long)(buf))
+#define __sanitizer_syscall_pre_compat_50_____semctl13(semid, semnum, cmd, \
+ arg) \
+ __sanitizer_syscall_pre_impl_compat_50_____semctl13( \
+ (long long)(semid), (long long)(semnum), (long long)(cmd), \
+ (long long)(arg))
+#define __sanitizer_syscall_post_compat_50_____semctl13(res, semid, semnum, \
+ cmd, arg) \
+ __sanitizer_syscall_post_impl_compat_50_____semctl13( \
+ res, (long long)(semid), (long long)(semnum), (long long)(cmd), \
+ (long long)(arg))
+#define __sanitizer_syscall_pre_compat_50___msgctl13(msqid, cmd, buf) \
+ __sanitizer_syscall_pre_impl_compat_50___msgctl13( \
+ (long long)(msqid), (long long)(cmd), (long long)(buf))
+#define __sanitizer_syscall_post_compat_50___msgctl13(res, msqid, cmd, buf) \
+ __sanitizer_syscall_post_impl_compat_50___msgctl13( \
+ res, (long long)(msqid), (long long)(cmd), (long long)(buf))
+#define __sanitizer_syscall_pre_compat_50___shmctl13(shmid, cmd, buf) \
+ __sanitizer_syscall_pre_impl_compat_50___shmctl13( \
+ (long long)(shmid), (long long)(cmd), (long long)(buf))
+#define __sanitizer_syscall_post_compat_50___shmctl13(res, shmid, cmd, buf) \
+ __sanitizer_syscall_post_impl_compat_50___shmctl13( \
+ res, (long long)(shmid), (long long)(cmd), (long long)(buf))
+#define __sanitizer_syscall_pre_lchflags(path, flags) \
+ __sanitizer_syscall_pre_impl_lchflags((long long)(path), (long long)(flags))
+#define __sanitizer_syscall_post_lchflags(res, path, flags) \
+ __sanitizer_syscall_post_impl_lchflags(res, (long long)(path), \
+ (long long)(flags))
+#define __sanitizer_syscall_pre_issetugid() \
+ __sanitizer_syscall_pre_impl_issetugid()
+#define __sanitizer_syscall_post_issetugid(res) \
+ __sanitizer_syscall_post_impl_issetugid(res)
+#define __sanitizer_syscall_pre_utrace(label, addr, len) \
+ __sanitizer_syscall_pre_impl_utrace((long long)(label), (long long)(addr), \
+ (long long)(len))
+#define __sanitizer_syscall_post_utrace(res, label, addr, len) \
+ __sanitizer_syscall_post_impl_utrace(res, (long long)(label), \
+ (long long)(addr), (long long)(len))
+#define __sanitizer_syscall_pre_getcontext(ucp) \
+ __sanitizer_syscall_pre_impl_getcontext((long long)(ucp))
+#define __sanitizer_syscall_post_getcontext(res, ucp) \
+ __sanitizer_syscall_post_impl_getcontext(res, (long long)(ucp))
+#define __sanitizer_syscall_pre_setcontext(ucp) \
+ __sanitizer_syscall_pre_impl_setcontext((long long)(ucp))
+#define __sanitizer_syscall_post_setcontext(res, ucp) \
+ __sanitizer_syscall_post_impl_setcontext(res, (long long)(ucp))
+#define __sanitizer_syscall_pre__lwp_create(ucp, flags, new_lwp) \
+ __sanitizer_syscall_pre_impl__lwp_create( \
+ (long long)(ucp), (long long)(flags), (long long)(new_lwp))
+#define __sanitizer_syscall_post__lwp_create(res, ucp, flags, new_lwp) \
+ __sanitizer_syscall_post_impl__lwp_create( \
+ res, (long long)(ucp), (long long)(flags), (long long)(new_lwp))
+#define __sanitizer_syscall_pre__lwp_exit() \
+ __sanitizer_syscall_pre_impl__lwp_exit()
+#define __sanitizer_syscall_post__lwp_exit(res) \
+ __sanitizer_syscall_post_impl__lwp_exit(res)
+#define __sanitizer_syscall_pre__lwp_self() \
+ __sanitizer_syscall_pre_impl__lwp_self()
+#define __sanitizer_syscall_post__lwp_self(res) \
+ __sanitizer_syscall_post_impl__lwp_self(res)
+#define __sanitizer_syscall_pre__lwp_wait(wait_for, departed) \
+ __sanitizer_syscall_pre_impl__lwp_wait((long long)(wait_for), \
+ (long long)(departed))
+#define __sanitizer_syscall_post__lwp_wait(res, wait_for, departed) \
+ __sanitizer_syscall_post_impl__lwp_wait(res, (long long)(wait_for), \
+ (long long)(departed))
+#define __sanitizer_syscall_pre__lwp_suspend(target) \
+ __sanitizer_syscall_pre_impl__lwp_suspend((long long)(target))
+#define __sanitizer_syscall_post__lwp_suspend(res, target) \
+ __sanitizer_syscall_post_impl__lwp_suspend(res, (long long)(target))
+#define __sanitizer_syscall_pre__lwp_continue(target) \
+ __sanitizer_syscall_pre_impl__lwp_continue((long long)(target))
+#define __sanitizer_syscall_post__lwp_continue(res, target) \
+ __sanitizer_syscall_post_impl__lwp_continue(res, (long long)(target))
+#define __sanitizer_syscall_pre__lwp_wakeup(target) \
+ __sanitizer_syscall_pre_impl__lwp_wakeup((long long)(target))
+#define __sanitizer_syscall_post__lwp_wakeup(res, target) \
+ __sanitizer_syscall_post_impl__lwp_wakeup(res, (long long)(target))
+#define __sanitizer_syscall_pre__lwp_getprivate() \
+ __sanitizer_syscall_pre_impl__lwp_getprivate()
+#define __sanitizer_syscall_post__lwp_getprivate(res) \
+ __sanitizer_syscall_post_impl__lwp_getprivate(res)
+#define __sanitizer_syscall_pre__lwp_setprivate(ptr) \
+ __sanitizer_syscall_pre_impl__lwp_setprivate((long long)(ptr))
+#define __sanitizer_syscall_post__lwp_setprivate(res, ptr) \
+ __sanitizer_syscall_post_impl__lwp_setprivate(res, (long long)(ptr))
+#define __sanitizer_syscall_pre__lwp_kill(target, signo) \
+ __sanitizer_syscall_pre_impl__lwp_kill((long long)(target), \
+ (long long)(signo))
+#define __sanitizer_syscall_post__lwp_kill(res, target, signo) \
+ __sanitizer_syscall_post_impl__lwp_kill(res, (long long)(target), \
+ (long long)(signo))
+#define __sanitizer_syscall_pre__lwp_detach(target) \
+ __sanitizer_syscall_pre_impl__lwp_detach((long long)(target))
+#define __sanitizer_syscall_post__lwp_detach(res, target) \
+ __sanitizer_syscall_post_impl__lwp_detach(res, (long long)(target))
+#define __sanitizer_syscall_pre_compat_50__lwp_park(ts, unpark, hint, \
+ unparkhint) \
+ __sanitizer_syscall_pre_impl_compat_50__lwp_park( \
+ (long long)(ts), (long long)(unpark), (long long)(hint), \
+ (long long)(unparkhint))
+#define __sanitizer_syscall_post_compat_50__lwp_park(res, ts, unpark, hint, \
+ unparkhint) \
+ __sanitizer_syscall_post_impl_compat_50__lwp_park( \
+ res, (long long)(ts), (long long)(unpark), (long long)(hint), \
+ (long long)(unparkhint))
+#define __sanitizer_syscall_pre__lwp_unpark(target, hint) \
+ __sanitizer_syscall_pre_impl__lwp_unpark((long long)(target), \
+ (long long)(hint))
+#define __sanitizer_syscall_post__lwp_unpark(res, target, hint) \
+ __sanitizer_syscall_post_impl__lwp_unpark(res, (long long)(target), \
+ (long long)(hint))
+#define __sanitizer_syscall_pre__lwp_unpark_all(targets, ntargets, hint) \
+ __sanitizer_syscall_pre_impl__lwp_unpark_all( \
+ (long long)(targets), (long long)(ntargets), (long long)(hint))
+#define __sanitizer_syscall_post__lwp_unpark_all(res, targets, ntargets, hint) \
+ __sanitizer_syscall_post_impl__lwp_unpark_all( \
+ res, (long long)(targets), (long long)(ntargets), (long long)(hint))
+#define __sanitizer_syscall_pre__lwp_setname(target, name) \
+ __sanitizer_syscall_pre_impl__lwp_setname((long long)(target), \
+ (long long)(name))
+#define __sanitizer_syscall_post__lwp_setname(res, target, name) \
+ __sanitizer_syscall_post_impl__lwp_setname(res, (long long)(target), \
+ (long long)(name))
+#define __sanitizer_syscall_pre__lwp_getname(target, name, len) \
+ __sanitizer_syscall_pre_impl__lwp_getname( \
+ (long long)(target), (long long)(name), (long long)(len))
+#define __sanitizer_syscall_post__lwp_getname(res, target, name, len) \
+ __sanitizer_syscall_post_impl__lwp_getname( \
+ res, (long long)(target), (long long)(name), (long long)(len))
+#define __sanitizer_syscall_pre__lwp_ctl(features, address) \
+ __sanitizer_syscall_pre_impl__lwp_ctl((long long)(features), \
+ (long long)(address))
+#define __sanitizer_syscall_post__lwp_ctl(res, features, address) \
+ __sanitizer_syscall_post_impl__lwp_ctl(res, (long long)(features), \
+ (long long)(address))
+/* syscall 326 has been skipped */
+/* syscall 327 has been skipped */
+/* syscall 328 has been skipped */
+/* syscall 329 has been skipped */
+#define __sanitizer_syscall_pre_compat_60_sa_register(newv, oldv, flags, \
+ stackinfo_offset) \
+ __sanitizer_syscall_pre_impl_compat_60_sa_register( \
+ (long long)(newv), (long long)(oldv), (long long)(flags), \
+ (long long)(stackinfo_offset))
+#define __sanitizer_syscall_post_compat_60_sa_register(res, newv, oldv, flags, \
+ stackinfo_offset) \
+ __sanitizer_syscall_post_impl_compat_60_sa_register( \
+ res, (long long)(newv), (long long)(oldv), (long long)(flags), \
+ (long long)(stackinfo_offset))
+#define __sanitizer_syscall_pre_compat_60_sa_stacks(num, stacks) \
+ __sanitizer_syscall_pre_impl_compat_60_sa_stacks((long long)(num), \
+ (long long)(stacks))
+#define __sanitizer_syscall_post_compat_60_sa_stacks(res, num, stacks) \
+ __sanitizer_syscall_post_impl_compat_60_sa_stacks(res, (long long)(num), \
+ (long long)(stacks))
+#define __sanitizer_syscall_pre_compat_60_sa_enable() \
+ __sanitizer_syscall_pre_impl_compat_60_sa_enable()
+#define __sanitizer_syscall_post_compat_60_sa_enable(res) \
+ __sanitizer_syscall_post_impl_compat_60_sa_enable(res)
+#define __sanitizer_syscall_pre_compat_60_sa_setconcurrency(concurrency) \
+ __sanitizer_syscall_pre_impl_compat_60_sa_setconcurrency( \
+ (long long)(concurrency))
+#define __sanitizer_syscall_post_compat_60_sa_setconcurrency(res, concurrency) \
+ __sanitizer_syscall_post_impl_compat_60_sa_setconcurrency( \
+ res, (long long)(concurrency))
+#define __sanitizer_syscall_pre_compat_60_sa_yield() \
+ __sanitizer_syscall_pre_impl_compat_60_sa_yield()
+#define __sanitizer_syscall_post_compat_60_sa_yield(res) \
+ __sanitizer_syscall_post_impl_compat_60_sa_yield(res)
+#define __sanitizer_syscall_pre_compat_60_sa_preempt(sa_id) \
+ __sanitizer_syscall_pre_impl_compat_60_sa_preempt((long long)(sa_id))
+#define __sanitizer_syscall_post_compat_60_sa_preempt(res, sa_id) \
+ __sanitizer_syscall_post_impl_compat_60_sa_preempt(res, (long long)(sa_id))
+/* syscall 336 has been skipped */
+/* syscall 337 has been skipped */
+/* syscall 338 has been skipped */
+/* syscall 339 has been skipped */
+#define __sanitizer_syscall_pre___sigaction_sigtramp(signum, nsa, osa, tramp, \
+ vers) \
+ __sanitizer_syscall_pre_impl___sigaction_sigtramp( \
+ (long long)(signum), (long long)(nsa), (long long)(osa), \
+ (long long)(tramp), (long long)(vers))
+#define __sanitizer_syscall_post___sigaction_sigtramp(res, signum, nsa, osa, \
+ tramp, vers) \
+ __sanitizer_syscall_post_impl___sigaction_sigtramp( \
+ res, (long long)(signum), (long long)(nsa), (long long)(osa), \
+ (long long)(tramp), (long long)(vers))
+/* syscall 341 has been skipped */
+/* syscall 342 has been skipped */
+#define __sanitizer_syscall_pre_rasctl(addr, len, op) \
+ __sanitizer_syscall_pre_impl_rasctl((long long)(addr), (long long)(len), \
+ (long long)(op))
+#define __sanitizer_syscall_post_rasctl(res, addr, len, op) \
+ __sanitizer_syscall_post_impl_rasctl(res, (long long)(addr), \
+ (long long)(len), (long long)(op))
+#define __sanitizer_syscall_pre_kqueue() __sanitizer_syscall_pre_impl_kqueue()
+#define __sanitizer_syscall_post_kqueue(res) \
+ __sanitizer_syscall_post_impl_kqueue(res)
+#define __sanitizer_syscall_pre_compat_50_kevent(fd, changelist, nchanges, \
+ eventlist, nevents, timeout) \
+ __sanitizer_syscall_pre_impl_compat_50_kevent( \
+ (long long)(fd), (long long)(changelist), (long long)(nchanges), \
+ (long long)(eventlist), (long long)(nevents), (long long)(timeout))
+#define __sanitizer_syscall_post_compat_50_kevent( \
+ res, fd, changelist, nchanges, eventlist, nevents, timeout) \
+ __sanitizer_syscall_post_impl_compat_50_kevent( \
+ res, (long long)(fd), (long long)(changelist), (long long)(nchanges), \
+ (long long)(eventlist), (long long)(nevents), (long long)(timeout))
+#define __sanitizer_syscall_pre__sched_setparam(pid, lid, policy, params) \
+ __sanitizer_syscall_pre_impl__sched_setparam( \
+ (long long)(pid), (long long)(lid), (long long)(policy), \
+ (long long)(params))
+#define __sanitizer_syscall_post__sched_setparam(res, pid, lid, policy, \
+ params) \
+ __sanitizer_syscall_post_impl__sched_setparam( \
+ res, (long long)(pid), (long long)(lid), (long long)(policy), \
+ (long long)(params))
+#define __sanitizer_syscall_pre__sched_getparam(pid, lid, policy, params) \
+ __sanitizer_syscall_pre_impl__sched_getparam( \
+ (long long)(pid), (long long)(lid), (long long)(policy), \
+ (long long)(params))
+#define __sanitizer_syscall_post__sched_getparam(res, pid, lid, policy, \
+ params) \
+ __sanitizer_syscall_post_impl__sched_getparam( \
+ res, (long long)(pid), (long long)(lid), (long long)(policy), \
+ (long long)(params))
+#define __sanitizer_syscall_pre__sched_setaffinity(pid, lid, size, cpuset) \
+ __sanitizer_syscall_pre_impl__sched_setaffinity( \
+ (long long)(pid), (long long)(lid), (long long)(size), \
+ (long long)(cpuset))
+#define __sanitizer_syscall_post__sched_setaffinity(res, pid, lid, size, \
+ cpuset) \
+ __sanitizer_syscall_post_impl__sched_setaffinity( \
+ res, (long long)(pid), (long long)(lid), (long long)(size), \
+ (long long)(cpuset))
+#define __sanitizer_syscall_pre__sched_getaffinity(pid, lid, size, cpuset) \
+ __sanitizer_syscall_pre_impl__sched_getaffinity( \
+ (long long)(pid), (long long)(lid), (long long)(size), \
+ (long long)(cpuset))
+#define __sanitizer_syscall_post__sched_getaffinity(res, pid, lid, size, \
+ cpuset) \
+ __sanitizer_syscall_post_impl__sched_getaffinity( \
+ res, (long long)(pid), (long long)(lid), (long long)(size), \
+ (long long)(cpuset))
+#define __sanitizer_syscall_pre_sched_yield() \
+ __sanitizer_syscall_pre_impl_sched_yield()
+#define __sanitizer_syscall_post_sched_yield(res) \
+ __sanitizer_syscall_post_impl_sched_yield(res)
+#define __sanitizer_syscall_pre__sched_protect(priority) \
+ __sanitizer_syscall_pre_impl__sched_protect((long long)(priority))
+#define __sanitizer_syscall_post__sched_protect(res, priority) \
+ __sanitizer_syscall_post_impl__sched_protect(res, (long long)(priority))
+/* syscall 352 has been skipped */
+/* syscall 353 has been skipped */
+#define __sanitizer_syscall_pre_fsync_range(fd, flags, start, length) \
+ __sanitizer_syscall_pre_impl_fsync_range( \
+ (long long)(fd), (long long)(flags), (long long)(start), \
+ (long long)(length))
+#define __sanitizer_syscall_post_fsync_range(res, fd, flags, start, length) \
+ __sanitizer_syscall_post_impl_fsync_range( \
+ res, (long long)(fd), (long long)(flags), (long long)(start), \
+ (long long)(length))
+#define __sanitizer_syscall_pre_uuidgen(store, count) \
+ __sanitizer_syscall_pre_impl_uuidgen((long long)(store), (long long)(count))
+#define __sanitizer_syscall_post_uuidgen(res, store, count) \
+ __sanitizer_syscall_post_impl_uuidgen(res, (long long)(store), \
+ (long long)(count))
+#define __sanitizer_syscall_pre_compat_90_getvfsstat(buf, bufsize, flags) \
+ __sanitizer_syscall_pre_impl_compat_90_getvfsstat( \
+ (long long)(buf), (long long)(bufsize), (long long)(flags))
+#define __sanitizer_syscall_post_compat_90_getvfsstat(res, buf, bufsize, \
+ flags) \
+ __sanitizer_syscall_post_impl_compat_90_getvfsstat( \
+ res, (long long)(buf), (long long)(bufsize), (long long)(flags))
+#define __sanitizer_syscall_pre_compat_90_statvfs1(path, buf, flags) \
+ __sanitizer_syscall_pre_impl_compat_90_statvfs1( \
+ (long long)(path), (long long)(buf), (long long)(flags))
+#define __sanitizer_syscall_post_compat_90_statvfs1(res, path, buf, flags) \
+ __sanitizer_syscall_post_impl_compat_90_statvfs1( \
+ res, (long long)(path), (long long)(buf), (long long)(flags))
+#define __sanitizer_syscall_pre_compat_90_fstatvfs1(fd, buf, flags) \
+ __sanitizer_syscall_pre_impl_compat_90_fstatvfs1( \
+ (long long)(fd), (long long)(buf), (long long)(flags))
+#define __sanitizer_syscall_post_compat_90_fstatvfs1(res, fd, buf, flags) \
+ __sanitizer_syscall_post_impl_compat_90_fstatvfs1( \
+ res, (long long)(fd), (long long)(buf), (long long)(flags))
+#define __sanitizer_syscall_pre_compat_30_fhstatvfs1(fhp, buf, flags) \
+ __sanitizer_syscall_pre_impl_compat_30_fhstatvfs1( \
+ (long long)(fhp), (long long)(buf), (long long)(flags))
+#define __sanitizer_syscall_post_compat_30_fhstatvfs1(res, fhp, buf, flags) \
+ __sanitizer_syscall_post_impl_compat_30_fhstatvfs1( \
+ res, (long long)(fhp), (long long)(buf), (long long)(flags))
+#define __sanitizer_syscall_pre_extattrctl(path, cmd, filename, attrnamespace, \
+ attrname) \
+ __sanitizer_syscall_pre_impl_extattrctl( \
+ (long long)(path), (long long)(cmd), (long long)(filename), \
+ (long long)(attrnamespace), (long long)(attrname))
+#define __sanitizer_syscall_post_extattrctl(res, path, cmd, filename, \
+ attrnamespace, attrname) \
+ __sanitizer_syscall_post_impl_extattrctl( \
+ res, (long long)(path), (long long)(cmd), (long long)(filename), \
+ (long long)(attrnamespace), (long long)(attrname))
+#define __sanitizer_syscall_pre_extattr_set_file(path, attrnamespace, \
+ attrname, data, nbytes) \
+ __sanitizer_syscall_pre_impl_extattr_set_file( \
+ (long long)(path), (long long)(attrnamespace), (long long)(attrname), \
+ (long long)(data), (long long)(nbytes))
+#define __sanitizer_syscall_post_extattr_set_file(res, path, attrnamespace, \
+ attrname, data, nbytes) \
+ __sanitizer_syscall_post_impl_extattr_set_file( \
+ res, (long long)(path), (long long)(attrnamespace), \
+ (long long)(attrname), (long long)(data), (long long)(nbytes))
+#define __sanitizer_syscall_pre_extattr_get_file(path, attrnamespace, \
+ attrname, data, nbytes) \
+ __sanitizer_syscall_pre_impl_extattr_get_file( \
+ (long long)(path), (long long)(attrnamespace), (long long)(attrname), \
+ (long long)(data), (long long)(nbytes))
+#define __sanitizer_syscall_post_extattr_get_file(res, path, attrnamespace, \
+ attrname, data, nbytes) \
+ __sanitizer_syscall_post_impl_extattr_get_file( \
+ res, (long long)(path), (long long)(attrnamespace), \
+ (long long)(attrname), (long long)(data), (long long)(nbytes))
+#define __sanitizer_syscall_pre_extattr_delete_file(path, attrnamespace, \
+ attrname) \
+ __sanitizer_syscall_pre_impl_extattr_delete_file( \
+ (long long)(path), (long long)(attrnamespace), (long long)(attrname))
+#define __sanitizer_syscall_post_extattr_delete_file(res, path, attrnamespace, \
+ attrname) \
+ __sanitizer_syscall_post_impl_extattr_delete_file( \
+ res, (long long)(path), (long long)(attrnamespace), \
+ (long long)(attrname))
+#define __sanitizer_syscall_pre_extattr_set_fd(fd, attrnamespace, attrname, \
+ data, nbytes) \
+ __sanitizer_syscall_pre_impl_extattr_set_fd( \
+ (long long)(fd), (long long)(attrnamespace), (long long)(attrname), \
+ (long long)(data), (long long)(nbytes))
+#define __sanitizer_syscall_post_extattr_set_fd(res, fd, attrnamespace, \
+ attrname, data, nbytes) \
+ __sanitizer_syscall_post_impl_extattr_set_fd( \
+ res, (long long)(fd), (long long)(attrnamespace), (long long)(attrname), \
+ (long long)(data), (long long)(nbytes))
+#define __sanitizer_syscall_pre_extattr_get_fd(fd, attrnamespace, attrname, \
+ data, nbytes) \
+ __sanitizer_syscall_pre_impl_extattr_get_fd( \
+ (long long)(fd), (long long)(attrnamespace), (long long)(attrname), \
+ (long long)(data), (long long)(nbytes))
+#define __sanitizer_syscall_post_extattr_get_fd(res, fd, attrnamespace, \
+ attrname, data, nbytes) \
+ __sanitizer_syscall_post_impl_extattr_get_fd( \
+ res, (long long)(fd), (long long)(attrnamespace), (long long)(attrname), \
+ (long long)(data), (long long)(nbytes))
+#define __sanitizer_syscall_pre_extattr_delete_fd(fd, attrnamespace, attrname) \
+ __sanitizer_syscall_pre_impl_extattr_delete_fd( \
+ (long long)(fd), (long long)(attrnamespace), (long long)(attrname))
+#define __sanitizer_syscall_post_extattr_delete_fd(res, fd, attrnamespace, \
+ attrname) \
+ __sanitizer_syscall_post_impl_extattr_delete_fd( \
+ res, (long long)(fd), (long long)(attrnamespace), (long long)(attrname))
+#define __sanitizer_syscall_pre_extattr_set_link(path, attrnamespace, \
+ attrname, data, nbytes) \
+ __sanitizer_syscall_pre_impl_extattr_set_link( \
+ (long long)(path), (long long)(attrnamespace), (long long)(attrname), \
+ (long long)(data), (long long)(nbytes))
+#define __sanitizer_syscall_post_extattr_set_link(res, path, attrnamespace, \
+ attrname, data, nbytes) \
+ __sanitizer_syscall_post_impl_extattr_set_link( \
+ res, (long long)(path), (long long)(attrnamespace), \
+ (long long)(attrname), (long long)(data), (long long)(nbytes))
+#define __sanitizer_syscall_pre_extattr_get_link(path, attrnamespace, \
+ attrname, data, nbytes) \
+ __sanitizer_syscall_pre_impl_extattr_get_link( \
+ (long long)(path), (long long)(attrnamespace), (long long)(attrname), \
+ (long long)(data), (long long)(nbytes))
+#define __sanitizer_syscall_post_extattr_get_link(res, path, attrnamespace, \
+ attrname, data, nbytes) \
+ __sanitizer_syscall_post_impl_extattr_get_link( \
+ res, (long long)(path), (long long)(attrnamespace), \
+ (long long)(attrname), (long long)(data), (long long)(nbytes))
+#define __sanitizer_syscall_pre_extattr_delete_link(path, attrnamespace, \
+ attrname) \
+ __sanitizer_syscall_pre_impl_extattr_delete_link( \
+ (long long)(path), (long long)(attrnamespace), (long long)(attrname))
+#define __sanitizer_syscall_post_extattr_delete_link(res, path, attrnamespace, \
+ attrname) \
+ __sanitizer_syscall_post_impl_extattr_delete_link( \
+ res, (long long)(path), (long long)(attrnamespace), \
+ (long long)(attrname))
+#define __sanitizer_syscall_pre_extattr_list_fd(fd, attrnamespace, data, \
+ nbytes) \
+ __sanitizer_syscall_pre_impl_extattr_list_fd( \
+ (long long)(fd), (long long)(attrnamespace), (long long)(data), \
+ (long long)(nbytes))
+#define __sanitizer_syscall_post_extattr_list_fd(res, fd, attrnamespace, data, \
+ nbytes) \
+ __sanitizer_syscall_post_impl_extattr_list_fd( \
+ res, (long long)(fd), (long long)(attrnamespace), (long long)(data), \
+ (long long)(nbytes))
+#define __sanitizer_syscall_pre_extattr_list_file(path, attrnamespace, data, \
+ nbytes) \
+ __sanitizer_syscall_pre_impl_extattr_list_file( \
+ (long long)(path), (long long)(attrnamespace), (long long)(data), \
+ (long long)(nbytes))
+#define __sanitizer_syscall_post_extattr_list_file(res, path, attrnamespace, \
+ data, nbytes) \
+ __sanitizer_syscall_post_impl_extattr_list_file( \
+ res, (long long)(path), (long long)(attrnamespace), (long long)(data), \
+ (long long)(nbytes))
+#define __sanitizer_syscall_pre_extattr_list_link(path, attrnamespace, data, \
+ nbytes) \
+ __sanitizer_syscall_pre_impl_extattr_list_link( \
+ (long long)(path), (long long)(attrnamespace), (long long)(data), \
+ (long long)(nbytes))
+#define __sanitizer_syscall_post_extattr_list_link(res, path, attrnamespace, \
+ data, nbytes) \
+ __sanitizer_syscall_post_impl_extattr_list_link( \
+ res, (long long)(path), (long long)(attrnamespace), (long long)(data), \
+ (long long)(nbytes))
+#define __sanitizer_syscall_pre_compat_50_pselect(nd, in, ou, ex, ts, mask) \
+ __sanitizer_syscall_pre_impl_compat_50_pselect( \
+ (long long)(nd), (long long)(in), (long long)(ou), (long long)(ex), \
+ (long long)(ts), (long long)(mask))
+#define __sanitizer_syscall_post_compat_50_pselect(res, nd, in, ou, ex, ts, \
+ mask) \
+ __sanitizer_syscall_post_impl_compat_50_pselect( \
+ res, (long long)(nd), (long long)(in), (long long)(ou), (long long)(ex), \
+ (long long)(ts), (long long)(mask))
+#define __sanitizer_syscall_pre_compat_50_pollts(fds, nfds, ts, mask) \
+ __sanitizer_syscall_pre_impl_compat_50_pollts( \
+ (long long)(fds), (long long)(nfds), (long long)(ts), (long long)(mask))
+#define __sanitizer_syscall_post_compat_50_pollts(res, fds, nfds, ts, mask) \
+ __sanitizer_syscall_post_impl_compat_50_pollts( \
+ res, (long long)(fds), (long long)(nfds), (long long)(ts), \
+ (long long)(mask))
+#define __sanitizer_syscall_pre_setxattr(path, name, value, size, flags) \
+ __sanitizer_syscall_pre_impl_setxattr((long long)(path), (long long)(name), \
+ (long long)(value), (long long)(size), \
+ (long long)(flags))
+#define __sanitizer_syscall_post_setxattr(res, path, name, value, size, flags) \
+ __sanitizer_syscall_post_impl_setxattr( \
+ res, (long long)(path), (long long)(name), (long long)(value), \
+ (long long)(size), (long long)(flags))
+#define __sanitizer_syscall_pre_lsetxattr(path, name, value, size, flags) \
+ __sanitizer_syscall_pre_impl_lsetxattr( \
+ (long long)(path), (long long)(name), (long long)(value), \
+ (long long)(size), (long long)(flags))
+#define __sanitizer_syscall_post_lsetxattr(res, path, name, value, size, \
+ flags) \
+ __sanitizer_syscall_post_impl_lsetxattr( \
+ res, (long long)(path), (long long)(name), (long long)(value), \
+ (long long)(size), (long long)(flags))
+#define __sanitizer_syscall_pre_fsetxattr(fd, name, value, size, flags) \
+ __sanitizer_syscall_pre_impl_fsetxattr( \
+ (long long)(fd), (long long)(name), (long long)(value), \
+ (long long)(size), (long long)(flags))
+#define __sanitizer_syscall_post_fsetxattr(res, fd, name, value, size, flags) \
+ __sanitizer_syscall_post_impl_fsetxattr( \
+ res, (long long)(fd), (long long)(name), (long long)(value), \
+ (long long)(size), (long long)(flags))
+#define __sanitizer_syscall_pre_getxattr(path, name, value, size) \
+ __sanitizer_syscall_pre_impl_getxattr((long long)(path), (long long)(name), \
+ (long long)(value), (long long)(size))
+#define __sanitizer_syscall_post_getxattr(res, path, name, value, size) \
+ __sanitizer_syscall_post_impl_getxattr( \
+ res, (long long)(path), (long long)(name), (long long)(value), \
+ (long long)(size))
+#define __sanitizer_syscall_pre_lgetxattr(path, name, value, size) \
+ __sanitizer_syscall_pre_impl_lgetxattr((long long)(path), (long long)(name), \
+ (long long)(value), \
+ (long long)(size))
+#define __sanitizer_syscall_post_lgetxattr(res, path, name, value, size) \
+ __sanitizer_syscall_post_impl_lgetxattr( \
+ res, (long long)(path), (long long)(name), (long long)(value), \
+ (long long)(size))
+#define __sanitizer_syscall_pre_fgetxattr(fd, name, value, size) \
+ __sanitizer_syscall_pre_impl_fgetxattr((long long)(fd), (long long)(name), \
+ (long long)(value), \
+ (long long)(size))
+#define __sanitizer_syscall_post_fgetxattr(res, fd, name, value, size) \
+ __sanitizer_syscall_post_impl_fgetxattr( \
+ res, (long long)(fd), (long long)(name), (long long)(value), \
+ (long long)(size))
+#define __sanitizer_syscall_pre_listxattr(path, list, size) \
+ __sanitizer_syscall_pre_impl_listxattr((long long)(path), (long long)(list), \
+ (long long)(size))
+#define __sanitizer_syscall_post_listxattr(res, path, list, size) \
+ __sanitizer_syscall_post_impl_listxattr( \
+ res, (long long)(path), (long long)(list), (long long)(size))
+#define __sanitizer_syscall_pre_llistxattr(path, list, size) \
+ __sanitizer_syscall_pre_impl_llistxattr( \
+ (long long)(path), (long long)(list), (long long)(size))
+#define __sanitizer_syscall_post_llistxattr(res, path, list, size) \
+ __sanitizer_syscall_post_impl_llistxattr( \
+ res, (long long)(path), (long long)(list), (long long)(size))
+#define __sanitizer_syscall_pre_flistxattr(fd, list, size) \
+ __sanitizer_syscall_pre_impl_flistxattr((long long)(fd), (long long)(list), \
+ (long long)(size))
+#define __sanitizer_syscall_post_flistxattr(res, fd, list, size) \
+ __sanitizer_syscall_post_impl_flistxattr( \
+ res, (long long)(fd), (long long)(list), (long long)(size))
+#define __sanitizer_syscall_pre_removexattr(path, name) \
+ __sanitizer_syscall_pre_impl_removexattr((long long)(path), (long long)(name))
+#define __sanitizer_syscall_post_removexattr(res, path, name) \
+ __sanitizer_syscall_post_impl_removexattr(res, (long long)(path), \
+ (long long)(name))
+#define __sanitizer_syscall_pre_lremovexattr(path, name) \
+ __sanitizer_syscall_pre_impl_lremovexattr((long long)(path), \
+ (long long)(name))
+#define __sanitizer_syscall_post_lremovexattr(res, path, name) \
+ __sanitizer_syscall_post_impl_lremovexattr(res, (long long)(path), \
+ (long long)(name))
+#define __sanitizer_syscall_pre_fremovexattr(fd, name) \
+ __sanitizer_syscall_pre_impl_fremovexattr((long long)(fd), (long long)(name))
+#define __sanitizer_syscall_post_fremovexattr(res, fd, name) \
+ __sanitizer_syscall_post_impl_fremovexattr(res, (long long)(fd), \
+ (long long)(name))
+#define __sanitizer_syscall_pre_compat_50___stat30(path, ub) \
+ __sanitizer_syscall_pre_impl_compat_50___stat30((long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_post_compat_50___stat30(res, path, ub) \
+ __sanitizer_syscall_post_impl_compat_50___stat30(res, (long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_pre_compat_50___fstat30(fd, sb) \
+ __sanitizer_syscall_pre_impl_compat_50___fstat30((long long)(fd), \
+ (long long)(sb))
+#define __sanitizer_syscall_post_compat_50___fstat30(res, fd, sb) \
+ __sanitizer_syscall_post_impl_compat_50___fstat30(res, (long long)(fd), \
+ (long long)(sb))
+#define __sanitizer_syscall_pre_compat_50___lstat30(path, ub) \
+ __sanitizer_syscall_pre_impl_compat_50___lstat30((long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_post_compat_50___lstat30(res, path, ub) \
+ __sanitizer_syscall_post_impl_compat_50___lstat30(res, (long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_pre___getdents30(fd, buf, count) \
+ __sanitizer_syscall_pre_impl___getdents30((long long)(fd), (long long)(buf), \
+ (long long)(count))
+#define __sanitizer_syscall_post___getdents30(res, fd, buf, count) \
+ __sanitizer_syscall_post_impl___getdents30( \
+ res, (long long)(fd), (long long)(buf), (long long)(count))
+#define __sanitizer_syscall_pre_posix_fadvise() \
+ __sanitizer_syscall_pre_impl_posix_fadvise((long long)())
+#define __sanitizer_syscall_post_posix_fadvise(res) \
+ __sanitizer_syscall_post_impl_posix_fadvise(res, (long long)())
+#define __sanitizer_syscall_pre_compat_30___fhstat30(fhp, sb) \
+ __sanitizer_syscall_pre_impl_compat_30___fhstat30((long long)(fhp), \
+ (long long)(sb))
+#define __sanitizer_syscall_post_compat_30___fhstat30(res, fhp, sb) \
+ __sanitizer_syscall_post_impl_compat_30___fhstat30(res, (long long)(fhp), \
+ (long long)(sb))
+#define __sanitizer_syscall_pre_compat_50___ntp_gettime30(ntvp) \
+ __sanitizer_syscall_pre_impl_compat_50___ntp_gettime30((long long)(ntvp))
+#define __sanitizer_syscall_post_compat_50___ntp_gettime30(res, ntvp) \
+ __sanitizer_syscall_post_impl_compat_50___ntp_gettime30(res, \
+ (long long)(ntvp))
+#define __sanitizer_syscall_pre___socket30(domain, type, protocol) \
+ __sanitizer_syscall_pre_impl___socket30( \
+ (long long)(domain), (long long)(type), (long long)(protocol))
+#define __sanitizer_syscall_post___socket30(res, domain, type, protocol) \
+ __sanitizer_syscall_post_impl___socket30( \
+ res, (long long)(domain), (long long)(type), (long long)(protocol))
+#define __sanitizer_syscall_pre___getfh30(fname, fhp, fh_size) \
+ __sanitizer_syscall_pre_impl___getfh30((long long)(fname), (long long)(fhp), \
+ (long long)(fh_size))
+#define __sanitizer_syscall_post___getfh30(res, fname, fhp, fh_size) \
+ __sanitizer_syscall_post_impl___getfh30( \
+ res, (long long)(fname), (long long)(fhp), (long long)(fh_size))
+#define __sanitizer_syscall_pre___fhopen40(fhp, fh_size, flags) \
+ __sanitizer_syscall_pre_impl___fhopen40( \
+ (long long)(fhp), (long long)(fh_size), (long long)(flags))
+#define __sanitizer_syscall_post___fhopen40(res, fhp, fh_size, flags) \
+ __sanitizer_syscall_post_impl___fhopen40( \
+ res, (long long)(fhp), (long long)(fh_size), (long long)(flags))
+#define __sanitizer_syscall_pre_compat_90_fhstatvfs1(fhp, fh_size, buf, flags) \
+ __sanitizer_syscall_pre_impl_compat_90_fhstatvfs1( \
+ (long long)(fhp), (long long)(fh_size), (long long)(buf), \
+ (long long)(flags))
+#define __sanitizer_syscall_post_compat_90_fhstatvfs1(res, fhp, fh_size, buf, \
+ flags) \
+ __sanitizer_syscall_post_impl_compat_90_fhstatvfs1( \
+ res, (long long)(fhp), (long long)(fh_size), (long long)(buf), \
+ (long long)(flags))
+#define __sanitizer_syscall_pre_compat_50___fhstat40(fhp, fh_size, sb) \
+ __sanitizer_syscall_pre_impl_compat_50___fhstat40( \
+ (long long)(fhp), (long long)(fh_size), (long long)(sb))
+#define __sanitizer_syscall_post_compat_50___fhstat40(res, fhp, fh_size, sb) \
+ __sanitizer_syscall_post_impl_compat_50___fhstat40( \
+ res, (long long)(fhp), (long long)(fh_size), (long long)(sb))
+#define __sanitizer_syscall_pre_aio_cancel(fildes, aiocbp) \
+ __sanitizer_syscall_pre_impl_aio_cancel((long long)(fildes), \
+ (long long)(aiocbp))
+#define __sanitizer_syscall_post_aio_cancel(res, fildes, aiocbp) \
+ __sanitizer_syscall_post_impl_aio_cancel(res, (long long)(fildes), \
+ (long long)(aiocbp))
+#define __sanitizer_syscall_pre_aio_error(aiocbp) \
+ __sanitizer_syscall_pre_impl_aio_error((long long)(aiocbp))
+#define __sanitizer_syscall_post_aio_error(res, aiocbp) \
+ __sanitizer_syscall_post_impl_aio_error(res, (long long)(aiocbp))
+#define __sanitizer_syscall_pre_aio_fsync(op, aiocbp) \
+ __sanitizer_syscall_pre_impl_aio_fsync((long long)(op), (long long)(aiocbp))
+#define __sanitizer_syscall_post_aio_fsync(res, op, aiocbp) \
+ __sanitizer_syscall_post_impl_aio_fsync(res, (long long)(op), \
+ (long long)(aiocbp))
+#define __sanitizer_syscall_pre_aio_read(aiocbp) \
+ __sanitizer_syscall_pre_impl_aio_read((long long)(aiocbp))
+#define __sanitizer_syscall_post_aio_read(res, aiocbp) \
+ __sanitizer_syscall_post_impl_aio_read(res, (long long)(aiocbp))
+#define __sanitizer_syscall_pre_aio_return(aiocbp) \
+ __sanitizer_syscall_pre_impl_aio_return((long long)(aiocbp))
+#define __sanitizer_syscall_post_aio_return(res, aiocbp) \
+ __sanitizer_syscall_post_impl_aio_return(res, (long long)(aiocbp))
+#define __sanitizer_syscall_pre_compat_50_aio_suspend(list, nent, timeout) \
+ __sanitizer_syscall_pre_impl_compat_50_aio_suspend( \
+ (long long)(list), (long long)(nent), (long long)(timeout))
+#define __sanitizer_syscall_post_compat_50_aio_suspend(res, list, nent, \
+ timeout) \
+ __sanitizer_syscall_post_impl_compat_50_aio_suspend( \
+ res, (long long)(list), (long long)(nent), (long long)(timeout))
+#define __sanitizer_syscall_pre_aio_write(aiocbp) \
+ __sanitizer_syscall_pre_impl_aio_write((long long)(aiocbp))
+#define __sanitizer_syscall_post_aio_write(res, aiocbp) \
+ __sanitizer_syscall_post_impl_aio_write(res, (long long)(aiocbp))
+#define __sanitizer_syscall_pre_lio_listio(mode, list, nent, sig) \
+ __sanitizer_syscall_pre_impl_lio_listio((long long)(mode), \
+ (long long)(list), \
+ (long long)(nent), (long long)(sig))
+#define __sanitizer_syscall_post_lio_listio(res, mode, list, nent, sig) \
+ __sanitizer_syscall_post_impl_lio_listio( \
+ res, (long long)(mode), (long long)(list), (long long)(nent), \
+ (long long)(sig))
+/* syscall 407 has been skipped */
+/* syscall 408 has been skipped */
+/* syscall 409 has been skipped */
+#define __sanitizer_syscall_pre___mount50(type, path, flags, data, data_len) \
+ __sanitizer_syscall_pre_impl___mount50( \
+ (long long)(type), (long long)(path), (long long)(flags), \
+ (long long)(data), (long long)(data_len))
+#define __sanitizer_syscall_post___mount50(res, type, path, flags, data, \
+ data_len) \
+ __sanitizer_syscall_post_impl___mount50( \
+ res, (long long)(type), (long long)(path), (long long)(flags), \
+ (long long)(data), (long long)(data_len))
+#define __sanitizer_syscall_pre_mremap(old_address, old_size, new_address, \
+ new_size, flags) \
+ __sanitizer_syscall_pre_impl_mremap( \
+ (long long)(old_address), (long long)(old_size), \
+ (long long)(new_address), (long long)(new_size), (long long)(flags))
+#define __sanitizer_syscall_post_mremap(res, old_address, old_size, \
+ new_address, new_size, flags) \
+ __sanitizer_syscall_post_impl_mremap( \
+ res, (long long)(old_address), (long long)(old_size), \
+ (long long)(new_address), (long long)(new_size), (long long)(flags))
+#define __sanitizer_syscall_pre_pset_create(psid) \
+ __sanitizer_syscall_pre_impl_pset_create((long long)(psid))
+#define __sanitizer_syscall_post_pset_create(res, psid) \
+ __sanitizer_syscall_post_impl_pset_create(res, (long long)(psid))
+#define __sanitizer_syscall_pre_pset_destroy(psid) \
+ __sanitizer_syscall_pre_impl_pset_destroy((long long)(psid))
+#define __sanitizer_syscall_post_pset_destroy(res, psid) \
+ __sanitizer_syscall_post_impl_pset_destroy(res, (long long)(psid))
+#define __sanitizer_syscall_pre_pset_assign(psid, cpuid, opsid) \
+ __sanitizer_syscall_pre_impl_pset_assign( \
+ (long long)(psid), (long long)(cpuid), (long long)(opsid))
+#define __sanitizer_syscall_post_pset_assign(res, psid, cpuid, opsid) \
+ __sanitizer_syscall_post_impl_pset_assign( \
+ res, (long long)(psid), (long long)(cpuid), (long long)(opsid))
+#define __sanitizer_syscall_pre__pset_bind(idtype, first_id, second_id, psid, \
+ opsid) \
+ __sanitizer_syscall_pre_impl__pset_bind( \
+ (long long)(idtype), (long long)(first_id), (long long)(second_id), \
+ (long long)(psid), (long long)(opsid))
+#define __sanitizer_syscall_post__pset_bind(res, idtype, first_id, second_id, \
+ psid, opsid) \
+ __sanitizer_syscall_post_impl__pset_bind( \
+ res, (long long)(idtype), (long long)(first_id), (long long)(second_id), \
+ (long long)(psid), (long long)(opsid))
+#define __sanitizer_syscall_pre___posix_fadvise50(fd, PAD, offset, len, \
+ advice) \
+ __sanitizer_syscall_pre_impl___posix_fadvise50( \
+ (long long)(fd), (long long)(PAD), (long long)(offset), \
+ (long long)(len), (long long)(advice))
+#define __sanitizer_syscall_post___posix_fadvise50(res, fd, PAD, offset, len, \
+ advice) \
+ __sanitizer_syscall_post_impl___posix_fadvise50( \
+ res, (long long)(fd), (long long)(PAD), (long long)(offset), \
+ (long long)(len), (long long)(advice))
+#define __sanitizer_syscall_pre___select50(nd, in, ou, ex, tv) \
+ __sanitizer_syscall_pre_impl___select50((long long)(nd), (long long)(in), \
+ (long long)(ou), (long long)(ex), \
+ (long long)(tv))
+#define __sanitizer_syscall_post___select50(res, nd, in, ou, ex, tv) \
+ __sanitizer_syscall_post_impl___select50(res, (long long)(nd), \
+ (long long)(in), (long long)(ou), \
+ (long long)(ex), (long long)(tv))
+#define __sanitizer_syscall_pre___gettimeofday50(tp, tzp) \
+ __sanitizer_syscall_pre_impl___gettimeofday50((long long)(tp), \
+ (long long)(tzp))
+#define __sanitizer_syscall_post___gettimeofday50(res, tp, tzp) \
+ __sanitizer_syscall_post_impl___gettimeofday50(res, (long long)(tp), \
+ (long long)(tzp))
+#define __sanitizer_syscall_pre___settimeofday50(tv, tzp) \
+ __sanitizer_syscall_pre_impl___settimeofday50((long long)(tv), \
+ (long long)(tzp))
+#define __sanitizer_syscall_post___settimeofday50(res, tv, tzp) \
+ __sanitizer_syscall_post_impl___settimeofday50(res, (long long)(tv), \
+ (long long)(tzp))
+#define __sanitizer_syscall_pre___utimes50(path, tptr) \
+ __sanitizer_syscall_pre_impl___utimes50((long long)(path), (long long)(tptr))
+#define __sanitizer_syscall_post___utimes50(res, path, tptr) \
+ __sanitizer_syscall_post_impl___utimes50(res, (long long)(path), \
+ (long long)(tptr))
+#define __sanitizer_syscall_pre___adjtime50(delta, olddelta) \
+ __sanitizer_syscall_pre_impl___adjtime50((long long)(delta), \
+ (long long)(olddelta))
+#define __sanitizer_syscall_post___adjtime50(res, delta, olddelta) \
+ __sanitizer_syscall_post_impl___adjtime50(res, (long long)(delta), \
+ (long long)(olddelta))
+#define __sanitizer_syscall_pre___lfs_segwait50(fsidp, tv) \
+ __sanitizer_syscall_pre_impl___lfs_segwait50((long long)(fsidp), \
+ (long long)(tv))
+#define __sanitizer_syscall_post___lfs_segwait50(res, fsidp, tv) \
+ __sanitizer_syscall_post_impl___lfs_segwait50(res, (long long)(fsidp), \
+ (long long)(tv))
+#define __sanitizer_syscall_pre___futimes50(fd, tptr) \
+ __sanitizer_syscall_pre_impl___futimes50((long long)(fd), (long long)(tptr))
+#define __sanitizer_syscall_post___futimes50(res, fd, tptr) \
+ __sanitizer_syscall_post_impl___futimes50(res, (long long)(fd), \
+ (long long)(tptr))
+#define __sanitizer_syscall_pre___lutimes50(path, tptr) \
+ __sanitizer_syscall_pre_impl___lutimes50((long long)(path), (long long)(tptr))
+#define __sanitizer_syscall_post___lutimes50(res, path, tptr) \
+ __sanitizer_syscall_post_impl___lutimes50(res, (long long)(path), \
+ (long long)(tptr))
+#define __sanitizer_syscall_pre___setitimer50(which, itv, oitv) \
+ __sanitizer_syscall_pre_impl___setitimer50( \
+ (long long)(which), (long long)(itv), (long long)(oitv))
+#define __sanitizer_syscall_post___setitimer50(res, which, itv, oitv) \
+ __sanitizer_syscall_post_impl___setitimer50( \
+ res, (long long)(which), (long long)(itv), (long long)(oitv))
+#define __sanitizer_syscall_pre___getitimer50(which, itv) \
+ __sanitizer_syscall_pre_impl___getitimer50((long long)(which), \
+ (long long)(itv))
+#define __sanitizer_syscall_post___getitimer50(res, which, itv) \
+ __sanitizer_syscall_post_impl___getitimer50(res, (long long)(which), \
+ (long long)(itv))
+#define __sanitizer_syscall_pre___clock_gettime50(clock_id, tp) \
+ __sanitizer_syscall_pre_impl___clock_gettime50((long long)(clock_id), \
+ (long long)(tp))
+#define __sanitizer_syscall_post___clock_gettime50(res, clock_id, tp) \
+ __sanitizer_syscall_post_impl___clock_gettime50(res, (long long)(clock_id), \
+ (long long)(tp))
+#define __sanitizer_syscall_pre___clock_settime50(clock_id, tp) \
+ __sanitizer_syscall_pre_impl___clock_settime50((long long)(clock_id), \
+ (long long)(tp))
+#define __sanitizer_syscall_post___clock_settime50(res, clock_id, tp) \
+ __sanitizer_syscall_post_impl___clock_settime50(res, (long long)(clock_id), \
+ (long long)(tp))
+#define __sanitizer_syscall_pre___clock_getres50(clock_id, tp) \
+ __sanitizer_syscall_pre_impl___clock_getres50((long long)(clock_id), \
+ (long long)(tp))
+#define __sanitizer_syscall_post___clock_getres50(res, clock_id, tp) \
+ __sanitizer_syscall_post_impl___clock_getres50(res, (long long)(clock_id), \
+ (long long)(tp))
+#define __sanitizer_syscall_pre___nanosleep50(rqtp, rmtp) \
+ __sanitizer_syscall_pre_impl___nanosleep50((long long)(rqtp), \
+ (long long)(rmtp))
+#define __sanitizer_syscall_post___nanosleep50(res, rqtp, rmtp) \
+ __sanitizer_syscall_post_impl___nanosleep50(res, (long long)(rqtp), \
+ (long long)(rmtp))
+#define __sanitizer_syscall_pre_____sigtimedwait50(set, info, timeout) \
+ __sanitizer_syscall_pre_impl_____sigtimedwait50( \
+ (long long)(set), (long long)(info), (long long)(timeout))
+#define __sanitizer_syscall_post_____sigtimedwait50(res, set, info, timeout) \
+ __sanitizer_syscall_post_impl_____sigtimedwait50( \
+ res, (long long)(set), (long long)(info), (long long)(timeout))
+#define __sanitizer_syscall_pre___mq_timedsend50(mqdes, msg_ptr, msg_len, \
+ msg_prio, abs_timeout) \
+ __sanitizer_syscall_pre_impl___mq_timedsend50( \
+ (long long)(mqdes), (long long)(msg_ptr), (long long)(msg_len), \
+ (long long)(msg_prio), (long long)(abs_timeout))
+#define __sanitizer_syscall_post___mq_timedsend50( \
+ res, mqdes, msg_ptr, msg_len, msg_prio, abs_timeout) \
+ __sanitizer_syscall_post_impl___mq_timedsend50( \
+ res, (long long)(mqdes), (long long)(msg_ptr), (long long)(msg_len), \
+ (long long)(msg_prio), (long long)(abs_timeout))
+#define __sanitizer_syscall_pre___mq_timedreceive50(mqdes, msg_ptr, msg_len, \
+ msg_prio, abs_timeout) \
+ __sanitizer_syscall_pre_impl___mq_timedreceive50( \
+ (long long)(mqdes), (long long)(msg_ptr), (long long)(msg_len), \
+ (long long)(msg_prio), (long long)(abs_timeout))
+#define __sanitizer_syscall_post___mq_timedreceive50( \
+ res, mqdes, msg_ptr, msg_len, msg_prio, abs_timeout) \
+ __sanitizer_syscall_post_impl___mq_timedreceive50( \
+ res, (long long)(mqdes), (long long)(msg_ptr), (long long)(msg_len), \
+ (long long)(msg_prio), (long long)(abs_timeout))
+#define __sanitizer_syscall_pre_compat_60__lwp_park(ts, unpark, hint, \
+ unparkhint) \
+ __sanitizer_syscall_pre_impl_compat_60__lwp_park( \
+ (long long)(ts), (long long)(unpark), (long long)(hint), \
+ (long long)(unparkhint))
+#define __sanitizer_syscall_post_compat_60__lwp_park(res, ts, unpark, hint, \
+ unparkhint) \
+ __sanitizer_syscall_post_impl_compat_60__lwp_park( \
+ res, (long long)(ts), (long long)(unpark), (long long)(hint), \
+ (long long)(unparkhint))
+#define __sanitizer_syscall_pre___kevent50(fd, changelist, nchanges, \
+ eventlist, nevents, timeout) \
+ __sanitizer_syscall_pre_impl___kevent50( \
+ (long long)(fd), (long long)(changelist), (long long)(nchanges), \
+ (long long)(eventlist), (long long)(nevents), (long long)(timeout))
+#define __sanitizer_syscall_post___kevent50(res, fd, changelist, nchanges, \
+ eventlist, nevents, timeout) \
+ __sanitizer_syscall_post_impl___kevent50( \
+ res, (long long)(fd), (long long)(changelist), (long long)(nchanges), \
+ (long long)(eventlist), (long long)(nevents), (long long)(timeout))
+#define __sanitizer_syscall_pre___pselect50(nd, in, ou, ex, ts, mask) \
+ __sanitizer_syscall_pre_impl___pselect50((long long)(nd), (long long)(in), \
+ (long long)(ou), (long long)(ex), \
+ (long long)(ts), (long long)(mask))
+#define __sanitizer_syscall_post___pselect50(res, nd, in, ou, ex, ts, mask) \
+ __sanitizer_syscall_post_impl___pselect50( \
+ res, (long long)(nd), (long long)(in), (long long)(ou), (long long)(ex), \
+ (long long)(ts), (long long)(mask))
+#define __sanitizer_syscall_pre___pollts50(fds, nfds, ts, mask) \
+ __sanitizer_syscall_pre_impl___pollts50((long long)(fds), (long long)(nfds), \
+ (long long)(ts), (long long)(mask))
+#define __sanitizer_syscall_post___pollts50(res, fds, nfds, ts, mask) \
+ __sanitizer_syscall_post_impl___pollts50(res, (long long)(fds), \
+ (long long)(nfds), (long long)(ts), \
+ (long long)(mask))
+#define __sanitizer_syscall_pre___aio_suspend50(list, nent, timeout) \
+ __sanitizer_syscall_pre_impl___aio_suspend50( \
+ (long long)(list), (long long)(nent), (long long)(timeout))
+#define __sanitizer_syscall_post___aio_suspend50(res, list, nent, timeout) \
+ __sanitizer_syscall_post_impl___aio_suspend50( \
+ res, (long long)(list), (long long)(nent), (long long)(timeout))
+#define __sanitizer_syscall_pre___stat50(path, ub) \
+ __sanitizer_syscall_pre_impl___stat50((long long)(path), (long long)(ub))
+#define __sanitizer_syscall_post___stat50(res, path, ub) \
+ __sanitizer_syscall_post_impl___stat50(res, (long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_pre___fstat50(fd, sb) \
+ __sanitizer_syscall_pre_impl___fstat50((long long)(fd), (long long)(sb))
+#define __sanitizer_syscall_post___fstat50(res, fd, sb) \
+ __sanitizer_syscall_post_impl___fstat50(res, (long long)(fd), (long long)(sb))
+#define __sanitizer_syscall_pre___lstat50(path, ub) \
+ __sanitizer_syscall_pre_impl___lstat50((long long)(path), (long long)(ub))
+#define __sanitizer_syscall_post___lstat50(res, path, ub) \
+ __sanitizer_syscall_post_impl___lstat50(res, (long long)(path), \
+ (long long)(ub))
+#define __sanitizer_syscall_pre_____semctl50(semid, semnum, cmd, arg) \
+ __sanitizer_syscall_pre_impl_____semctl50( \
+ (long long)(semid), (long long)(semnum), (long long)(cmd), \
+ (long long)(arg))
+#define __sanitizer_syscall_post_____semctl50(res, semid, semnum, cmd, arg) \
+ __sanitizer_syscall_post_impl_____semctl50( \
+ res, (long long)(semid), (long long)(semnum), (long long)(cmd), \
+ (long long)(arg))
+#define __sanitizer_syscall_pre___shmctl50(shmid, cmd, buf) \
+ __sanitizer_syscall_pre_impl___shmctl50((long long)(shmid), \
+ (long long)(cmd), (long long)(buf))
+#define __sanitizer_syscall_post___shmctl50(res, shmid, cmd, buf) \
+ __sanitizer_syscall_post_impl___shmctl50(res, (long long)(shmid), \
+ (long long)(cmd), (long long)(buf))
+#define __sanitizer_syscall_pre___msgctl50(msqid, cmd, buf) \
+ __sanitizer_syscall_pre_impl___msgctl50((long long)(msqid), \
+ (long long)(cmd), (long long)(buf))
+#define __sanitizer_syscall_post___msgctl50(res, msqid, cmd, buf) \
+ __sanitizer_syscall_post_impl___msgctl50(res, (long long)(msqid), \
+ (long long)(cmd), (long long)(buf))
+#define __sanitizer_syscall_pre___getrusage50(who, rusage) \
+ __sanitizer_syscall_pre_impl___getrusage50((long long)(who), \
+ (long long)(rusage))
+#define __sanitizer_syscall_post___getrusage50(res, who, rusage) \
+ __sanitizer_syscall_post_impl___getrusage50(res, (long long)(who), \
+ (long long)(rusage))
+#define __sanitizer_syscall_pre___timer_settime50(timerid, flags, value, \
+ ovalue) \
+ __sanitizer_syscall_pre_impl___timer_settime50( \
+ (long long)(timerid), (long long)(flags), (long long)(value), \
+ (long long)(ovalue))
+#define __sanitizer_syscall_post___timer_settime50(res, timerid, flags, value, \
+ ovalue) \
+ __sanitizer_syscall_post_impl___timer_settime50( \
+ res, (long long)(timerid), (long long)(flags), (long long)(value), \
+ (long long)(ovalue))
+#define __sanitizer_syscall_pre___timer_gettime50(timerid, value) \
+ __sanitizer_syscall_pre_impl___timer_gettime50((long long)(timerid), \
+ (long long)(value))
+#define __sanitizer_syscall_post___timer_gettime50(res, timerid, value) \
+ __sanitizer_syscall_post_impl___timer_gettime50(res, (long long)(timerid), \
+ (long long)(value))
+#if defined(NTP) || !defined(_KERNEL_OPT)
+#define __sanitizer_syscall_pre___ntp_gettime50(ntvp) \
+ __sanitizer_syscall_pre_impl___ntp_gettime50((long long)(ntvp))
+#define __sanitizer_syscall_post___ntp_gettime50(res, ntvp) \
+ __sanitizer_syscall_post_impl___ntp_gettime50(res, (long long)(ntvp))
+#else
+/* syscall 448 has been skipped */
+#endif
+#define __sanitizer_syscall_pre___wait450(pid, status, options, rusage) \
+ __sanitizer_syscall_pre_impl___wait450( \
+ (long long)(pid), (long long)(status), (long long)(options), \
+ (long long)(rusage))
+#define __sanitizer_syscall_post___wait450(res, pid, status, options, rusage) \
+ __sanitizer_syscall_post_impl___wait450( \
+ res, (long long)(pid), (long long)(status), (long long)(options), \
+ (long long)(rusage))
+#define __sanitizer_syscall_pre___mknod50(path, mode, dev) \
+ __sanitizer_syscall_pre_impl___mknod50((long long)(path), (long long)(mode), \
+ (long long)(dev))
+#define __sanitizer_syscall_post___mknod50(res, path, mode, dev) \
+ __sanitizer_syscall_post_impl___mknod50(res, (long long)(path), \
+ (long long)(mode), (long long)(dev))
+#define __sanitizer_syscall_pre___fhstat50(fhp, fh_size, sb) \
+ __sanitizer_syscall_pre_impl___fhstat50( \
+ (long long)(fhp), (long long)(fh_size), (long long)(sb))
+#define __sanitizer_syscall_post___fhstat50(res, fhp, fh_size, sb) \
+ __sanitizer_syscall_post_impl___fhstat50( \
+ res, (long long)(fhp), (long long)(fh_size), (long long)(sb))
+/* syscall 452 has been skipped */
+#define __sanitizer_syscall_pre_pipe2(fildes, flags) \
+ __sanitizer_syscall_pre_impl_pipe2((long long)(fildes), (long long)(flags))
+#define __sanitizer_syscall_post_pipe2(res, fildes, flags) \
+ __sanitizer_syscall_post_impl_pipe2(res, (long long)(fildes), \
+ (long long)(flags))
+#define __sanitizer_syscall_pre_dup3(from, to, flags) \
+ __sanitizer_syscall_pre_impl_dup3((long long)(from), (long long)(to), \
+ (long long)(flags))
+#define __sanitizer_syscall_post_dup3(res, from, to, flags) \
+ __sanitizer_syscall_post_impl_dup3(res, (long long)(from), (long long)(to), \
+ (long long)(flags))
+#define __sanitizer_syscall_pre_kqueue1(flags) \
+ __sanitizer_syscall_pre_impl_kqueue1((long long)(flags))
+#define __sanitizer_syscall_post_kqueue1(res, flags) \
+ __sanitizer_syscall_post_impl_kqueue1(res, (long long)(flags))
+#define __sanitizer_syscall_pre_paccept(s, name, anamelen, mask, flags) \
+ __sanitizer_syscall_pre_impl_paccept((long long)(s), (long long)(name), \
+ (long long)(anamelen), \
+ (long long)(mask), (long long)(flags))
+#define __sanitizer_syscall_post_paccept(res, s, name, anamelen, mask, flags) \
+ __sanitizer_syscall_post_impl_paccept( \
+ res, (long long)(s), (long long)(name), (long long)(anamelen), \
+ (long long)(mask), (long long)(flags))
+#define __sanitizer_syscall_pre_linkat(fd1, name1, fd2, name2, flags) \
+ __sanitizer_syscall_pre_impl_linkat((long long)(fd1), (long long)(name1), \
+ (long long)(fd2), (long long)(name2), \
+ (long long)(flags))
+#define __sanitizer_syscall_post_linkat(res, fd1, name1, fd2, name2, flags) \
+ __sanitizer_syscall_post_impl_linkat(res, (long long)(fd1), \
+ (long long)(name1), (long long)(fd2), \
+ (long long)(name2), (long long)(flags))
+#define __sanitizer_syscall_pre_renameat(fromfd, from, tofd, to) \
+ __sanitizer_syscall_pre_impl_renameat((long long)(fromfd), \
+ (long long)(from), (long long)(tofd), \
+ (long long)(to))
+#define __sanitizer_syscall_post_renameat(res, fromfd, from, tofd, to) \
+ __sanitizer_syscall_post_impl_renameat(res, (long long)(fromfd), \
+ (long long)(from), (long long)(tofd), \
+ (long long)(to))
+#define __sanitizer_syscall_pre_mkfifoat(fd, path, mode) \
+ __sanitizer_syscall_pre_impl_mkfifoat((long long)(fd), (long long)(path), \
+ (long long)(mode))
+#define __sanitizer_syscall_post_mkfifoat(res, fd, path, mode) \
+ __sanitizer_syscall_post_impl_mkfifoat(res, (long long)(fd), \
+ (long long)(path), (long long)(mode))
+#define __sanitizer_syscall_pre_mknodat(fd, path, mode, PAD, dev) \
+ __sanitizer_syscall_pre_impl_mknodat((long long)(fd), (long long)(path), \
+ (long long)(mode), (long long)(PAD), \
+ (long long)(dev))
+#define __sanitizer_syscall_post_mknodat(res, fd, path, mode, PAD, dev) \
+ __sanitizer_syscall_post_impl_mknodat(res, (long long)(fd), \
+ (long long)(path), (long long)(mode), \
+ (long long)(PAD), (long long)(dev))
+#define __sanitizer_syscall_pre_mkdirat(fd, path, mode) \
+ __sanitizer_syscall_pre_impl_mkdirat((long long)(fd), (long long)(path), \
+ (long long)(mode))
+#define __sanitizer_syscall_post_mkdirat(res, fd, path, mode) \
+ __sanitizer_syscall_post_impl_mkdirat(res, (long long)(fd), \
+ (long long)(path), (long long)(mode))
+#define __sanitizer_syscall_pre_faccessat(fd, path, amode, flag) \
+ __sanitizer_syscall_pre_impl_faccessat((long long)(fd), (long long)(path), \
+ (long long)(amode), \
+ (long long)(flag))
+#define __sanitizer_syscall_post_faccessat(res, fd, path, amode, flag) \
+ __sanitizer_syscall_post_impl_faccessat( \
+ res, (long long)(fd), (long long)(path), (long long)(amode), \
+ (long long)(flag))
+#define __sanitizer_syscall_pre_fchmodat(fd, path, mode, flag) \
+ __sanitizer_syscall_pre_impl_fchmodat((long long)(fd), (long long)(path), \
+ (long long)(mode), (long long)(flag))
+#define __sanitizer_syscall_post_fchmodat(res, fd, path, mode, flag) \
+ __sanitizer_syscall_post_impl_fchmodat(res, (long long)(fd), \
+ (long long)(path), (long long)(mode), \
+ (long long)(flag))
+#define __sanitizer_syscall_pre_fchownat(fd, path, owner, group, flag) \
+ __sanitizer_syscall_pre_impl_fchownat((long long)(fd), (long long)(path), \
+ (long long)(owner), \
+ (long long)(group), (long long)(flag))
+#define __sanitizer_syscall_post_fchownat(res, fd, path, owner, group, flag) \
+ __sanitizer_syscall_post_impl_fchownat( \
+ res, (long long)(fd), (long long)(path), (long long)(owner), \
+ (long long)(group), (long long)(flag))
+#define __sanitizer_syscall_pre_fexecve(fd, argp, envp) \
+ __sanitizer_syscall_pre_impl_fexecve((long long)(fd), (long long)(argp), \
+ (long long)(envp))
+#define __sanitizer_syscall_post_fexecve(res, fd, argp, envp) \
+ __sanitizer_syscall_post_impl_fexecve(res, (long long)(fd), \
+ (long long)(argp), (long long)(envp))
+#define __sanitizer_syscall_pre_fstatat(fd, path, buf, flag) \
+ __sanitizer_syscall_pre_impl_fstatat((long long)(fd), (long long)(path), \
+ (long long)(buf), (long long)(flag))
+#define __sanitizer_syscall_post_fstatat(res, fd, path, buf, flag) \
+ __sanitizer_syscall_post_impl_fstatat(res, (long long)(fd), \
+ (long long)(path), (long long)(buf), \
+ (long long)(flag))
+#define __sanitizer_syscall_pre_utimensat(fd, path, tptr, flag) \
+ __sanitizer_syscall_pre_impl_utimensat((long long)(fd), (long long)(path), \
+ (long long)(tptr), (long long)(flag))
+#define __sanitizer_syscall_post_utimensat(res, fd, path, tptr, flag) \
+ __sanitizer_syscall_post_impl_utimensat( \
+ res, (long long)(fd), (long long)(path), (long long)(tptr), \
+ (long long)(flag))
+#define __sanitizer_syscall_pre_openat(fd, path, oflags, mode) \
+ __sanitizer_syscall_pre_impl_openat((long long)(fd), (long long)(path), \
+ (long long)(oflags), (long long)(mode))
+#define __sanitizer_syscall_post_openat(res, fd, path, oflags, mode) \
+ __sanitizer_syscall_post_impl_openat(res, (long long)(fd), \
+ (long long)(path), (long long)(oflags), \
+ (long long)(mode))
+#define __sanitizer_syscall_pre_readlinkat(fd, path, buf, bufsize) \
+ __sanitizer_syscall_pre_impl_readlinkat((long long)(fd), (long long)(path), \
+ (long long)(buf), \
+ (long long)(bufsize))
+#define __sanitizer_syscall_post_readlinkat(res, fd, path, buf, bufsize) \
+ __sanitizer_syscall_post_impl_readlinkat( \
+ res, (long long)(fd), (long long)(path), (long long)(buf), \
+ (long long)(bufsize))
+#define __sanitizer_syscall_pre_symlinkat(path1, fd, path2) \
+ __sanitizer_syscall_pre_impl_symlinkat((long long)(path1), (long long)(fd), \
+ (long long)(path2))
+#define __sanitizer_syscall_post_symlinkat(res, path1, fd, path2) \
+ __sanitizer_syscall_post_impl_symlinkat(res, (long long)(path1), \
+ (long long)(fd), (long long)(path2))
+#define __sanitizer_syscall_pre_unlinkat(fd, path, flag) \
+ __sanitizer_syscall_pre_impl_unlinkat((long long)(fd), (long long)(path), \
+ (long long)(flag))
+#define __sanitizer_syscall_post_unlinkat(res, fd, path, flag) \
+ __sanitizer_syscall_post_impl_unlinkat(res, (long long)(fd), \
+ (long long)(path), (long long)(flag))
+#define __sanitizer_syscall_pre_futimens(fd, tptr) \
+ __sanitizer_syscall_pre_impl_futimens((long long)(fd), (long long)(tptr))
+#define __sanitizer_syscall_post_futimens(res, fd, tptr) \
+ __sanitizer_syscall_post_impl_futimens(res, (long long)(fd), \
+ (long long)(tptr))
+#define __sanitizer_syscall_pre___quotactl(path, args) \
+ __sanitizer_syscall_pre_impl___quotactl((long long)(path), (long long)(args))
+#define __sanitizer_syscall_post___quotactl(res, path, args) \
+ __sanitizer_syscall_post_impl___quotactl(res, (long long)(path), \
+ (long long)(args))
+#define __sanitizer_syscall_pre_posix_spawn(pid, path, file_actions, attrp, \
+ argv, envp) \
+ __sanitizer_syscall_pre_impl_posix_spawn( \
+ (long long)(pid), (long long)(path), (long long)(file_actions), \
+ (long long)(attrp), (long long)(argv), (long long)(envp))
+#define __sanitizer_syscall_post_posix_spawn(res, pid, path, file_actions, \
+ attrp, argv, envp) \
+ __sanitizer_syscall_post_impl_posix_spawn( \
+ res, (long long)(pid), (long long)(path), (long long)(file_actions), \
+ (long long)(attrp), (long long)(argv), (long long)(envp))
+#define __sanitizer_syscall_pre_recvmmsg(s, mmsg, vlen, flags, timeout) \
+ __sanitizer_syscall_pre_impl_recvmmsg((long long)(s), (long long)(mmsg), \
+ (long long)(vlen), (long long)(flags), \
+ (long long)(timeout))
+#define __sanitizer_syscall_post_recvmmsg(res, s, mmsg, vlen, flags, timeout) \
+ __sanitizer_syscall_post_impl_recvmmsg( \
+ res, (long long)(s), (long long)(mmsg), (long long)(vlen), \
+ (long long)(flags), (long long)(timeout))
+#define __sanitizer_syscall_pre_sendmmsg(s, mmsg, vlen, flags) \
+ __sanitizer_syscall_pre_impl_sendmmsg((long long)(s), (long long)(mmsg), \
+ (long long)(vlen), (long long)(flags))
+#define __sanitizer_syscall_post_sendmmsg(res, s, mmsg, vlen, flags) \
+ __sanitizer_syscall_post_impl_sendmmsg(res, (long long)(s), \
+ (long long)(mmsg), (long long)(vlen), \
+ (long long)(flags))
+#define __sanitizer_syscall_pre_clock_nanosleep(clock_id, flags, rqtp, rmtp) \
+ __sanitizer_syscall_pre_impl_clock_nanosleep( \
+ (long long)(clock_id), (long long)(flags), (long long)(rqtp), \
+ (long long)(rmtp))
+#define __sanitizer_syscall_post_clock_nanosleep(res, clock_id, flags, rqtp, \
+ rmtp) \
+ __sanitizer_syscall_post_impl_clock_nanosleep( \
+ res, (long long)(clock_id), (long long)(flags), (long long)(rqtp), \
+ (long long)(rmtp))
+#define __sanitizer_syscall_pre____lwp_park60(clock_id, flags, ts, unpark, \
+ hint, unparkhint) \
+ __sanitizer_syscall_pre_impl____lwp_park60( \
+ (long long)(clock_id), (long long)(flags), (long long)(ts), \
+ (long long)(unpark), (long long)(hint), (long long)(unparkhint))
+#define __sanitizer_syscall_post____lwp_park60(res, clock_id, flags, ts, \
+ unpark, hint, unparkhint) \
+ __sanitizer_syscall_post_impl____lwp_park60( \
+ res, (long long)(clock_id), (long long)(flags), (long long)(ts), \
+ (long long)(unpark), (long long)(hint), (long long)(unparkhint))
+#define __sanitizer_syscall_pre_posix_fallocate(fd, PAD, pos, len) \
+ __sanitizer_syscall_pre_impl_posix_fallocate( \
+ (long long)(fd), (long long)(PAD), (long long)(pos), (long long)(len))
+#define __sanitizer_syscall_post_posix_fallocate(res, fd, PAD, pos, len) \
+ __sanitizer_syscall_post_impl_posix_fallocate( \
+ res, (long long)(fd), (long long)(PAD), (long long)(pos), \
+ (long long)(len))
+#define __sanitizer_syscall_pre_fdiscard(fd, PAD, pos, len) \
+ __sanitizer_syscall_pre_impl_fdiscard((long long)(fd), (long long)(PAD), \
+ (long long)(pos), (long long)(len))
+#define __sanitizer_syscall_post_fdiscard(res, fd, PAD, pos, len) \
+ __sanitizer_syscall_post_impl_fdiscard(res, (long long)(fd), \
+ (long long)(PAD), (long long)(pos), \
+ (long long)(len))
+#define __sanitizer_syscall_pre_wait6(idtype, id, status, options, wru, info) \
+ __sanitizer_syscall_pre_impl_wait6( \
+ (long long)(idtype), (long long)(id), (long long)(status), \
+ (long long)(options), (long long)(wru), (long long)(info))
+#define __sanitizer_syscall_post_wait6(res, idtype, id, status, options, wru, \
+ info) \
+ __sanitizer_syscall_post_impl_wait6( \
+ res, (long long)(idtype), (long long)(id), (long long)(status), \
+ (long long)(options), (long long)(wru), (long long)(info))
+#define __sanitizer_syscall_pre_clock_getcpuclockid2(idtype, id, clock_id) \
+ __sanitizer_syscall_pre_impl_clock_getcpuclockid2( \
+ (long long)(idtype), (long long)(id), (long long)(clock_id))
+#define __sanitizer_syscall_post_clock_getcpuclockid2(res, idtype, id, \
+ clock_id) \
+ __sanitizer_syscall_post_impl_clock_getcpuclockid2( \
+ res, (long long)(idtype), (long long)(id), (long long)(clock_id))
+#define __sanitizer_syscall_pre___getvfsstat90(buf, bufsize, flags) \
+ __sanitizer_syscall_pre_impl___getvfsstat90( \
+ (long long)(buf), (long long)(bufsize), (long long)(flags))
+#define __sanitizer_syscall_post___getvfsstat90(res, buf, bufsize, flags) \
+ __sanitizer_syscall_post_impl___getvfsstat90( \
+ res, (long long)(buf), (long long)(bufsize), (long long)(flags))
+#define __sanitizer_syscall_pre___statvfs190(path, buf, flags) \
+ __sanitizer_syscall_pre_impl___statvfs190( \
+ (long long)(path), (long long)(buf), (long long)(flags))
+#define __sanitizer_syscall_post___statvfs190(res, path, buf, flags) \
+ __sanitizer_syscall_post_impl___statvfs190( \
+ res, (long long)(path), (long long)(buf), (long long)(flags))
+#define __sanitizer_syscall_pre___fstatvfs190(fd, buf, flags) \
+ __sanitizer_syscall_pre_impl___fstatvfs190( \
+ (long long)(fd), (long long)(buf), (long long)(flags))
+#define __sanitizer_syscall_post___fstatvfs190(res, fd, buf, flags) \
+ __sanitizer_syscall_post_impl___fstatvfs190( \
+ res, (long long)(fd), (long long)(buf), (long long)(flags))
+#define __sanitizer_syscall_pre___fhstatvfs190(fhp, fh_size, buf, flags) \
+ __sanitizer_syscall_pre_impl___fhstatvfs190( \
+ (long long)(fhp), (long long)(fh_size), (long long)(buf), \
+ (long long)(flags))
+#define __sanitizer_syscall_post___fhstatvfs190(res, fhp, fh_size, buf, flags) \
+ __sanitizer_syscall_post_impl___fhstatvfs190( \
+ res, (long long)(fhp), (long long)(fh_size), (long long)(buf), \
+ (long long)(flags))
+
+/* Compat with older releases */
+#define __sanitizer_syscall_pre_getvfsstat \
+ __sanitizer_syscall_pre_compat_90_getvfsstat
+#define __sanitizer_syscall_post_getvfsstat \
+ __sanitizer_syscall_post_compat_90_getvfsstat
+
+#define __sanitizer_syscall_pre_statvfs1 \
+ __sanitizer_syscall_pre_compat_90_statvfs1
+#define __sanitizer_syscall_post_statvfs1 \
+ __sanitizer_syscall_post_compat_90_statvfs1
+
+#define __sanitizer_syscall_pre_fstatvfs1 \
+ __sanitizer_syscall_pre_compat_90_fstatvfs1
+#define __sanitizer_syscall_post_fstatvfs1 \
+ __sanitizer_syscall_post_compat_90_fstatvfs1
+
+#define __sanitizer_syscall_pre___fhstatvfs140 \
+ __sanitizer_syscall_pre_compat_90_fhstatvfs1
+#define __sanitizer_syscall_post___fhstatvfs140 \
+ __sanitizer_syscall_post_compat_90_fhstatvfs1
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+// Private declarations. Do not call directly from user code. Use macros above.
+
+// DO NOT EDIT! THIS FILE HAS BEEN GENERATED!
+
+void __sanitizer_syscall_pre_impl_syscall(long long code, long long arg0,
+ long long arg1, long long arg2,
+ long long arg3, long long arg4,
+ long long arg5, long long arg6,
+ long long arg7);
+void __sanitizer_syscall_post_impl_syscall(long long res, long long code,
+ long long arg0, long long arg1,
+ long long arg2, long long arg3,
+ long long arg4, long long arg5,
+ long long arg6, long long arg7);
+void __sanitizer_syscall_pre_impl_exit(long long rval);
+void __sanitizer_syscall_post_impl_exit(long long res, long long rval);
+void __sanitizer_syscall_pre_impl_fork(void);
+void __sanitizer_syscall_post_impl_fork(long long res);
+void __sanitizer_syscall_pre_impl_read(long long fd, long long buf,
+ long long nbyte);
+void __sanitizer_syscall_post_impl_read(long long res, long long fd,
+ long long buf, long long nbyte);
+void __sanitizer_syscall_pre_impl_write(long long fd, long long buf,
+ long long nbyte);
+void __sanitizer_syscall_post_impl_write(long long res, long long fd,
+ long long buf, long long nbyte);
+void __sanitizer_syscall_pre_impl_open(long long path, long long flags,
+ long long mode);
+void __sanitizer_syscall_post_impl_open(long long res, long long path,
+ long long flags, long long mode);
+void __sanitizer_syscall_pre_impl_close(long long fd);
+void __sanitizer_syscall_post_impl_close(long long res, long long fd);
+void __sanitizer_syscall_pre_impl_compat_50_wait4(long long pid,
+ long long status,
+ long long options,
+ long long rusage);
+void __sanitizer_syscall_post_impl_compat_50_wait4(long long res, long long pid,
+ long long status,
+ long long options,
+ long long rusage);
+void __sanitizer_syscall_pre_impl_compat_43_ocreat(long long path,
+ long long mode);
+void __sanitizer_syscall_post_impl_compat_43_ocreat(long long res,
+ long long path,
+ long long mode);
+void __sanitizer_syscall_pre_impl_link(long long path, long long link);
+void __sanitizer_syscall_post_impl_link(long long res, long long path,
+ long long link);
+void __sanitizer_syscall_pre_impl_unlink(long long path);
+void __sanitizer_syscall_post_impl_unlink(long long res, long long path);
+/* syscall 11 has been skipped */
+void __sanitizer_syscall_pre_impl_chdir(long long path);
+void __sanitizer_syscall_post_impl_chdir(long long res, long long path);
+void __sanitizer_syscall_pre_impl_fchdir(long long fd);
+void __sanitizer_syscall_post_impl_fchdir(long long res, long long fd);
+void __sanitizer_syscall_pre_impl_compat_50_mknod(long long path,
+ long long mode,
+ long long dev);
+void __sanitizer_syscall_post_impl_compat_50_mknod(long long res,
+ long long path,
+ long long mode,
+ long long dev);
+void __sanitizer_syscall_pre_impl_chmod(long long path, long long mode);
+void __sanitizer_syscall_post_impl_chmod(long long res, long long path,
+ long long mode);
+void __sanitizer_syscall_pre_impl_chown(long long path, long long uid,
+ long long gid);
+void __sanitizer_syscall_post_impl_chown(long long res, long long path,
+ long long uid, long long gid);
+void __sanitizer_syscall_pre_impl_break(long long nsize);
+void __sanitizer_syscall_post_impl_break(long long res, long long nsize);
+void __sanitizer_syscall_pre_impl_compat_20_getfsstat(long long buf,
+ long long bufsize,
+ long long flags);
+void __sanitizer_syscall_post_impl_compat_20_getfsstat(long long res,
+ long long buf,
+ long long bufsize,
+ long long flags);
+void __sanitizer_syscall_pre_impl_compat_43_olseek(long long fd,
+ long long offset,
+ long long whence);
+void __sanitizer_syscall_post_impl_compat_43_olseek(long long res, long long fd,
+ long long offset,
+ long long whence);
+void __sanitizer_syscall_pre_impl_getpid(void);
+void __sanitizer_syscall_post_impl_getpid(long long res);
+void __sanitizer_syscall_pre_impl_compat_40_mount(long long type,
+ long long path,
+ long long flags,
+ long long data);
+void __sanitizer_syscall_post_impl_compat_40_mount(long long res,
+ long long type,
+ long long path,
+ long long flags,
+ long long data);
+void __sanitizer_syscall_pre_impl_unmount(long long path, long long flags);
+void __sanitizer_syscall_post_impl_unmount(long long res, long long path,
+ long long flags);
+void __sanitizer_syscall_pre_impl_setuid(long long uid);
+void __sanitizer_syscall_post_impl_setuid(long long res, long long uid);
+void __sanitizer_syscall_pre_impl_getuid(void);
+void __sanitizer_syscall_post_impl_getuid(long long res);
+void __sanitizer_syscall_pre_impl_geteuid(void);
+void __sanitizer_syscall_post_impl_geteuid(long long res);
+void __sanitizer_syscall_pre_impl_ptrace(long long req, long long pid,
+ long long addr, long long data);
+void __sanitizer_syscall_post_impl_ptrace(long long res, long long req,
+ long long pid, long long addr,
+ long long data);
+void __sanitizer_syscall_pre_impl_recvmsg(long long s, long long msg,
+ long long flags);
+void __sanitizer_syscall_post_impl_recvmsg(long long res, long long s,
+ long long msg, long long flags);
+void __sanitizer_syscall_pre_impl_sendmsg(long long s, long long msg,
+ long long flags);
+void __sanitizer_syscall_post_impl_sendmsg(long long res, long long s,
+ long long msg, long long flags);
+void __sanitizer_syscall_pre_impl_recvfrom(long long s, long long buf,
+ long long len, long long flags,
+ long long from,
+ long long fromlenaddr);
+void __sanitizer_syscall_post_impl_recvfrom(long long res, long long s,
+ long long buf, long long len,
+ long long flags, long long from,
+ long long fromlenaddr);
+void __sanitizer_syscall_pre_impl_accept(long long s, long long name,
+ long long anamelen);
+void __sanitizer_syscall_post_impl_accept(long long res, long long s,
+ long long name, long long anamelen);
+void __sanitizer_syscall_pre_impl_getpeername(long long fdes, long long asa,
+ long long alen);
+void __sanitizer_syscall_post_impl_getpeername(long long res, long long fdes,
+ long long asa, long long alen);
+void __sanitizer_syscall_pre_impl_getsockname(long long fdes, long long asa,
+ long long alen);
+void __sanitizer_syscall_post_impl_getsockname(long long res, long long fdes,
+ long long asa, long long alen);
+void __sanitizer_syscall_pre_impl_access(long long path, long long flags);
+void __sanitizer_syscall_post_impl_access(long long res, long long path,
+ long long flags);
+void __sanitizer_syscall_pre_impl_chflags(long long path, long long flags);
+void __sanitizer_syscall_post_impl_chflags(long long res, long long path,
+ long long flags);
+void __sanitizer_syscall_pre_impl_fchflags(long long fd, long long flags);
+void __sanitizer_syscall_post_impl_fchflags(long long res, long long fd,
+ long long flags);
+void __sanitizer_syscall_pre_impl_sync(void);
+void __sanitizer_syscall_post_impl_sync(long long res);
+void __sanitizer_syscall_pre_impl_kill(long long pid, long long signum);
+void __sanitizer_syscall_post_impl_kill(long long res, long long pid,
+ long long signum);
+void __sanitizer_syscall_pre_impl_compat_43_stat43(long long path,
+ long long ub);
+void __sanitizer_syscall_post_impl_compat_43_stat43(long long res,
+ long long path,
+ long long ub);
+void __sanitizer_syscall_pre_impl_getppid(void);
+void __sanitizer_syscall_post_impl_getppid(long long res);
+void __sanitizer_syscall_pre_impl_compat_43_lstat43(long long path,
+ long long ub);
+void __sanitizer_syscall_post_impl_compat_43_lstat43(long long res,
+ long long path,
+ long long ub);
+void __sanitizer_syscall_pre_impl_dup(long long fd);
+void __sanitizer_syscall_post_impl_dup(long long res, long long fd);
+void __sanitizer_syscall_pre_impl_pipe(void);
+void __sanitizer_syscall_post_impl_pipe(long long res);
+void __sanitizer_syscall_pre_impl_getegid(void);
+void __sanitizer_syscall_post_impl_getegid(long long res);
+void __sanitizer_syscall_pre_impl_profil(long long samples, long long size,
+ long long offset, long long scale);
+void __sanitizer_syscall_post_impl_profil(long long res, long long samples,
+ long long size, long long offset,
+ long long scale);
+void __sanitizer_syscall_pre_impl_ktrace(long long fname, long long ops,
+ long long facs, long long pid);
+void __sanitizer_syscall_post_impl_ktrace(long long res, long long fname,
+ long long ops, long long facs,
+ long long pid);
+void __sanitizer_syscall_pre_impl_compat_13_sigaction13(long long signum,
+ long long nsa,
+ long long osa);
+void __sanitizer_syscall_post_impl_compat_13_sigaction13(long long res,
+ long long signum,
+ long long nsa,
+ long long osa);
+void __sanitizer_syscall_pre_impl_getgid(void);
+void __sanitizer_syscall_post_impl_getgid(long long res);
+void __sanitizer_syscall_pre_impl_compat_13_sigprocmask13(long long how,
+ long long mask);
+void __sanitizer_syscall_post_impl_compat_13_sigprocmask13(long long res,
+ long long how,
+ long long mask);
+void __sanitizer_syscall_pre_impl___getlogin(long long namebuf,
+ long long namelen);
+void __sanitizer_syscall_post_impl___getlogin(long long res, long long namebuf,
+ long long namelen);
+void __sanitizer_syscall_pre_impl___setlogin(long long namebuf);
+void __sanitizer_syscall_post_impl___setlogin(long long res, long long namebuf);
+void __sanitizer_syscall_pre_impl_acct(long long path);
+void __sanitizer_syscall_post_impl_acct(long long res, long long path);
+void __sanitizer_syscall_pre_impl_compat_13_sigpending13(void);
+void __sanitizer_syscall_post_impl_compat_13_sigpending13(long long res);
+void __sanitizer_syscall_pre_impl_compat_13_sigaltstack13(long long nss,
+ long long oss);
+void __sanitizer_syscall_post_impl_compat_13_sigaltstack13(long long res,
+ long long nss,
+ long long oss);
+void __sanitizer_syscall_pre_impl_ioctl(long long fd, long long com,
+ long long data);
+void __sanitizer_syscall_post_impl_ioctl(long long res, long long fd,
+ long long com, long long data);
+void __sanitizer_syscall_pre_impl_compat_12_oreboot(long long opt);
+void __sanitizer_syscall_post_impl_compat_12_oreboot(long long res,
+ long long opt);
+void __sanitizer_syscall_pre_impl_revoke(long long path);
+void __sanitizer_syscall_post_impl_revoke(long long res, long long path);
+void __sanitizer_syscall_pre_impl_symlink(long long path, long long link);
+void __sanitizer_syscall_post_impl_symlink(long long res, long long path,
+ long long link);
+void __sanitizer_syscall_pre_impl_readlink(long long path, long long buf,
+ long long count);
+void __sanitizer_syscall_post_impl_readlink(long long res, long long path,
+ long long buf, long long count);
+void __sanitizer_syscall_pre_impl_execve(long long path, long long argp,
+ long long envp);
+void __sanitizer_syscall_post_impl_execve(long long res, long long path,
+ long long argp, long long envp);
+void __sanitizer_syscall_pre_impl_umask(long long newmask);
+void __sanitizer_syscall_post_impl_umask(long long res, long long newmask);
+void __sanitizer_syscall_pre_impl_chroot(long long path);
+void __sanitizer_syscall_post_impl_chroot(long long res, long long path);
+void __sanitizer_syscall_pre_impl_compat_43_fstat43(long long fd, long long sb);
+void __sanitizer_syscall_post_impl_compat_43_fstat43(long long res,
+ long long fd,
+ long long sb);
+void __sanitizer_syscall_pre_impl_compat_43_ogetkerninfo(long long op,
+ long long where,
+ long long size,
+ long long arg);
+void __sanitizer_syscall_post_impl_compat_43_ogetkerninfo(long long res,
+ long long op,
+ long long where,
+ long long size,
+ long long arg);
+void __sanitizer_syscall_pre_impl_compat_43_ogetpagesize(void);
+void __sanitizer_syscall_post_impl_compat_43_ogetpagesize(long long res);
+void __sanitizer_syscall_pre_impl_compat_12_msync(long long addr,
+ long long len);
+void __sanitizer_syscall_post_impl_compat_12_msync(long long res,
+ long long addr,
+ long long len);
+void __sanitizer_syscall_pre_impl_vfork(void);
+void __sanitizer_syscall_post_impl_vfork(long long res);
+/* syscall 67 has been skipped */
+/* syscall 68 has been skipped */
+/* syscall 69 has been skipped */
+/* syscall 70 has been skipped */
+void __sanitizer_syscall_pre_impl_compat_43_ommap(long long addr, long long len,
+ long long prot,
+ long long flags, long long fd,
+ long long pos);
+void __sanitizer_syscall_post_impl_compat_43_ommap(
+ long long res, long long addr, long long len, long long prot,
+ long long flags, long long fd, long long pos);
+void __sanitizer_syscall_pre_impl_vadvise(long long anom);
+void __sanitizer_syscall_post_impl_vadvise(long long res, long long anom);
+void __sanitizer_syscall_pre_impl_munmap(long long addr, long long len);
+void __sanitizer_syscall_post_impl_munmap(long long res, long long addr,
+ long long len);
+void __sanitizer_syscall_pre_impl_mprotect(long long addr, long long len,
+ long long prot);
+void __sanitizer_syscall_post_impl_mprotect(long long res, long long addr,
+ long long len, long long prot);
+void __sanitizer_syscall_pre_impl_madvise(long long addr, long long len,
+ long long behav);
+void __sanitizer_syscall_post_impl_madvise(long long res, long long addr,
+ long long len, long long behav);
+/* syscall 76 has been skipped */
+/* syscall 77 has been skipped */
+void __sanitizer_syscall_pre_impl_mincore(long long addr, long long len,
+ long long vec);
+void __sanitizer_syscall_post_impl_mincore(long long res, long long addr,
+ long long len, long long vec);
+void __sanitizer_syscall_pre_impl_getgroups(long long gidsetsize,
+ long long gidset);
+void __sanitizer_syscall_post_impl_getgroups(long long res,
+ long long gidsetsize,
+ long long gidset);
+void __sanitizer_syscall_pre_impl_setgroups(long long gidsetsize,
+ long long gidset);
+void __sanitizer_syscall_post_impl_setgroups(long long res,
+ long long gidsetsize,
+ long long gidset);
+void __sanitizer_syscall_pre_impl_getpgrp(void);
+void __sanitizer_syscall_post_impl_getpgrp(long long res);
+void __sanitizer_syscall_pre_impl_setpgid(long long pid, long long pgid);
+void __sanitizer_syscall_post_impl_setpgid(long long res, long long pid,
+ long long pgid);
+void __sanitizer_syscall_pre_impl_compat_50_setitimer(long long which,
+ long long itv,
+ long long oitv);
+void __sanitizer_syscall_post_impl_compat_50_setitimer(long long res,
+ long long which,
+ long long itv,
+ long long oitv);
+void __sanitizer_syscall_pre_impl_compat_43_owait(void);
+void __sanitizer_syscall_post_impl_compat_43_owait(long long res);
+void __sanitizer_syscall_pre_impl_compat_12_oswapon(long long name);
+void __sanitizer_syscall_post_impl_compat_12_oswapon(long long res,
+ long long name);
+void __sanitizer_syscall_pre_impl_compat_50_getitimer(long long which,
+ long long itv);
+void __sanitizer_syscall_post_impl_compat_50_getitimer(long long res,
+ long long which,
+ long long itv);
+void __sanitizer_syscall_pre_impl_compat_43_ogethostname(long long hostname,
+ long long len);
+void __sanitizer_syscall_post_impl_compat_43_ogethostname(long long res,
+ long long hostname,
+ long long len);
+void __sanitizer_syscall_pre_impl_compat_43_osethostname(long long hostname,
+ long long len);
+void __sanitizer_syscall_post_impl_compat_43_osethostname(long long res,
+ long long hostname,
+ long long len);
+void __sanitizer_syscall_pre_impl_compat_43_ogetdtablesize(void);
+void __sanitizer_syscall_post_impl_compat_43_ogetdtablesize(long long res);
+void __sanitizer_syscall_pre_impl_dup2(long long from, long long to);
+void __sanitizer_syscall_post_impl_dup2(long long res, long long from,
+ long long to);
+/* syscall 91 has been skipped */
+void __sanitizer_syscall_pre_impl_fcntl(long long fd, long long cmd,
+ long long arg);
+void __sanitizer_syscall_post_impl_fcntl(long long res, long long fd,
+ long long cmd, long long arg);
+void __sanitizer_syscall_pre_impl_compat_50_select(long long nd, long long in,
+ long long ou, long long ex,
+ long long tv);
+void __sanitizer_syscall_post_impl_compat_50_select(long long res, long long nd,
+ long long in, long long ou,
+ long long ex, long long tv);
+/* syscall 94 has been skipped */
+void __sanitizer_syscall_pre_impl_fsync(long long fd);
+void __sanitizer_syscall_post_impl_fsync(long long res, long long fd);
+void __sanitizer_syscall_pre_impl_setpriority(long long which, long long who,
+ long long prio);
+void __sanitizer_syscall_post_impl_setpriority(long long res, long long which,
+ long long who, long long prio);
+void __sanitizer_syscall_pre_impl_compat_30_socket(long long domain,
+ long long type,
+ long long protocol);
+void __sanitizer_syscall_post_impl_compat_30_socket(long long res,
+ long long domain,
+ long long type,
+ long long protocol);
+void __sanitizer_syscall_pre_impl_connect(long long s, long long name,
+ long long namelen);
+void __sanitizer_syscall_post_impl_connect(long long res, long long s,
+ long long name, long long namelen);
+void __sanitizer_syscall_pre_impl_compat_43_oaccept(long long s, long long name,
+ long long anamelen);
+void __sanitizer_syscall_post_impl_compat_43_oaccept(long long res, long long s,
+ long long name,
+ long long anamelen);
+void __sanitizer_syscall_pre_impl_getpriority(long long which, long long who);
+void __sanitizer_syscall_post_impl_getpriority(long long res, long long which,
+ long long who);
+void __sanitizer_syscall_pre_impl_compat_43_osend(long long s, long long buf,
+ long long len,
+ long long flags);
+void __sanitizer_syscall_post_impl_compat_43_osend(long long res, long long s,
+ long long buf, long long len,
+ long long flags);
+void __sanitizer_syscall_pre_impl_compat_43_orecv(long long s, long long buf,
+ long long len,
+ long long flags);
+void __sanitizer_syscall_post_impl_compat_43_orecv(long long res, long long s,
+ long long buf, long long len,
+ long long flags);
+void __sanitizer_syscall_pre_impl_compat_13_sigreturn13(long long sigcntxp);
+void __sanitizer_syscall_post_impl_compat_13_sigreturn13(long long res,
+ long long sigcntxp);
+void __sanitizer_syscall_pre_impl_bind(long long s, long long name,
+ long long namelen);
+void __sanitizer_syscall_post_impl_bind(long long res, long long s,
+ long long name, long long namelen);
+void __sanitizer_syscall_pre_impl_setsockopt(long long s, long long level,
+ long long name, long long val,
+ long long valsize);
+void __sanitizer_syscall_post_impl_setsockopt(long long res, long long s,
+ long long level, long long name,
+ long long val, long long valsize);
+void __sanitizer_syscall_pre_impl_listen(long long s, long long backlog);
+void __sanitizer_syscall_post_impl_listen(long long res, long long s,
+ long long backlog);
+/* syscall 107 has been skipped */
+void __sanitizer_syscall_pre_impl_compat_43_osigvec(long long signum,
+ long long nsv,
+ long long osv);
+void __sanitizer_syscall_post_impl_compat_43_osigvec(long long res,
+ long long signum,
+ long long nsv,
+ long long osv);
+void __sanitizer_syscall_pre_impl_compat_43_osigblock(long long mask);
+void __sanitizer_syscall_post_impl_compat_43_osigblock(long long res,
+ long long mask);
+void __sanitizer_syscall_pre_impl_compat_43_osigsetmask(long long mask);
+void __sanitizer_syscall_post_impl_compat_43_osigsetmask(long long res,
+ long long mask);
+void __sanitizer_syscall_pre_impl_compat_13_sigsuspend13(long long mask);
+void __sanitizer_syscall_post_impl_compat_13_sigsuspend13(long long res,
+ long long mask);
+void __sanitizer_syscall_pre_impl_compat_43_osigstack(long long nss,
+ long long oss);
+void __sanitizer_syscall_post_impl_compat_43_osigstack(long long res,
+ long long nss,
+ long long oss);
+void __sanitizer_syscall_pre_impl_compat_43_orecvmsg(long long s, long long msg,
+ long long flags);
+void __sanitizer_syscall_post_impl_compat_43_orecvmsg(long long res,
+ long long s,
+ long long msg,
+ long long flags);
+void __sanitizer_syscall_pre_impl_compat_43_osendmsg(long long s, long long msg,
+ long long flags);
+void __sanitizer_syscall_post_impl_compat_43_osendmsg(long long res,
+ long long s,
+ long long msg,
+ long long flags);
+/* syscall 115 has been skipped */
+void __sanitizer_syscall_pre_impl_compat_50_gettimeofday(long long tp,
+ long long tzp);
+void __sanitizer_syscall_post_impl_compat_50_gettimeofday(long long res,
+ long long tp,
+ long long tzp);
+void __sanitizer_syscall_pre_impl_compat_50_getrusage(long long who,
+ long long rusage);
+void __sanitizer_syscall_post_impl_compat_50_getrusage(long long res,
+ long long who,
+ long long rusage);
+void __sanitizer_syscall_pre_impl_getsockopt(long long s, long long level,
+ long long name, long long val,
+ long long avalsize);
+void __sanitizer_syscall_post_impl_getsockopt(long long res, long long s,
+ long long level, long long name,
+ long long val,
+ long long avalsize);
+/* syscall 119 has been skipped */
+void __sanitizer_syscall_pre_impl_readv(long long fd, long long iovp,
+ long long iovcnt);
+void __sanitizer_syscall_post_impl_readv(long long res, long long fd,
+ long long iovp, long long iovcnt);
+void __sanitizer_syscall_pre_impl_writev(long long fd, long long iovp,
+ long long iovcnt);
+void __sanitizer_syscall_post_impl_writev(long long res, long long fd,
+ long long iovp, long long iovcnt);
+void __sanitizer_syscall_pre_impl_compat_50_settimeofday(long long tv,
+ long long tzp);
+void __sanitizer_syscall_post_impl_compat_50_settimeofday(long long res,
+ long long tv,
+ long long tzp);
+void __sanitizer_syscall_pre_impl_fchown(long long fd, long long uid,
+ long long gid);
+void __sanitizer_syscall_post_impl_fchown(long long res, long long fd,
+ long long uid, long long gid);
+void __sanitizer_syscall_pre_impl_fchmod(long long fd, long long mode);
+void __sanitizer_syscall_post_impl_fchmod(long long res, long long fd,
+ long long mode);
+void __sanitizer_syscall_pre_impl_compat_43_orecvfrom(
+ long long s, long long buf, long long len, long long flags, long long from,
+ long long fromlenaddr);
+void __sanitizer_syscall_post_impl_compat_43_orecvfrom(
+ long long res, long long s, long long buf, long long len, long long flags,
+ long long from, long long fromlenaddr);
+void __sanitizer_syscall_pre_impl_setreuid(long long ruid, long long euid);
+void __sanitizer_syscall_post_impl_setreuid(long long res, long long ruid,
+ long long euid);
+void __sanitizer_syscall_pre_impl_setregid(long long rgid, long long egid);
+void __sanitizer_syscall_post_impl_setregid(long long res, long long rgid,
+ long long egid);
+void __sanitizer_syscall_pre_impl_rename(long long from, long long to);
+void __sanitizer_syscall_post_impl_rename(long long res, long long from,
+ long long to);
+void __sanitizer_syscall_pre_impl_compat_43_otruncate(long long path,
+ long long length);
+void __sanitizer_syscall_post_impl_compat_43_otruncate(long long res,
+ long long path,
+ long long length);
+void __sanitizer_syscall_pre_impl_compat_43_oftruncate(long long fd,
+ long long length);
+void __sanitizer_syscall_post_impl_compat_43_oftruncate(long long res,
+ long long fd,
+ long long length);
+void __sanitizer_syscall_pre_impl_flock(long long fd, long long how);
+void __sanitizer_syscall_post_impl_flock(long long res, long long fd,
+ long long how);
+void __sanitizer_syscall_pre_impl_mkfifo(long long path, long long mode);
+void __sanitizer_syscall_post_impl_mkfifo(long long res, long long path,
+ long long mode);
+void __sanitizer_syscall_pre_impl_sendto(long long s, long long buf,
+ long long len, long long flags,
+ long long to, long long tolen);
+void __sanitizer_syscall_post_impl_sendto(long long res, long long s,
+ long long buf, long long len,
+ long long flags, long long to,
+ long long tolen);
+void __sanitizer_syscall_pre_impl_shutdown(long long s, long long how);
+void __sanitizer_syscall_post_impl_shutdown(long long res, long long s,
+ long long how);
+void __sanitizer_syscall_pre_impl_socketpair(long long domain, long long type,
+ long long protocol, long long rsv);
+void __sanitizer_syscall_post_impl_socketpair(long long res, long long domain,
+ long long type,
+ long long protocol,
+ long long rsv);
+void __sanitizer_syscall_pre_impl_mkdir(long long path, long long mode);
+void __sanitizer_syscall_post_impl_mkdir(long long res, long long path,
+ long long mode);
+void __sanitizer_syscall_pre_impl_rmdir(long long path);
+void __sanitizer_syscall_post_impl_rmdir(long long res, long long path);
+void __sanitizer_syscall_pre_impl_compat_50_utimes(long long path,
+ long long tptr);
+void __sanitizer_syscall_post_impl_compat_50_utimes(long long res,
+ long long path,
+ long long tptr);
+/* syscall 139 has been skipped */
+void __sanitizer_syscall_pre_impl_compat_50_adjtime(long long delta,
+ long long olddelta);
+void __sanitizer_syscall_post_impl_compat_50_adjtime(long long res,
+ long long delta,
+ long long olddelta);
+void __sanitizer_syscall_pre_impl_compat_43_ogetpeername(long long fdes,
+ long long asa,
+ long long alen);
+void __sanitizer_syscall_post_impl_compat_43_ogetpeername(long long res,
+ long long fdes,
+ long long asa,
+ long long alen);
+void __sanitizer_syscall_pre_impl_compat_43_ogethostid(void);
+void __sanitizer_syscall_post_impl_compat_43_ogethostid(long long res);
+void __sanitizer_syscall_pre_impl_compat_43_osethostid(long long hostid);
+void __sanitizer_syscall_post_impl_compat_43_osethostid(long long res,
+ long long hostid);
+void __sanitizer_syscall_pre_impl_compat_43_ogetrlimit(long long which,
+ long long rlp);
+void __sanitizer_syscall_post_impl_compat_43_ogetrlimit(long long res,
+ long long which,
+ long long rlp);
+void __sanitizer_syscall_pre_impl_compat_43_osetrlimit(long long which,
+ long long rlp);
+void __sanitizer_syscall_post_impl_compat_43_osetrlimit(long long res,
+ long long which,
+ long long rlp);
+void __sanitizer_syscall_pre_impl_compat_43_okillpg(long long pgid,
+ long long signum);
+void __sanitizer_syscall_post_impl_compat_43_okillpg(long long res,
+ long long pgid,
+ long long signum);
+void __sanitizer_syscall_pre_impl_setsid(void);
+void __sanitizer_syscall_post_impl_setsid(long long res);
+void __sanitizer_syscall_pre_impl_compat_50_quotactl(long long path,
+ long long cmd,
+ long long uid,
+ long long arg);
+void __sanitizer_syscall_post_impl_compat_50_quotactl(
+ long long res, long long path, long long cmd, long long uid, long long arg);
+void __sanitizer_syscall_pre_impl_compat_43_oquota(void);
+void __sanitizer_syscall_post_impl_compat_43_oquota(long long res);
+void __sanitizer_syscall_pre_impl_compat_43_ogetsockname(long long fdec,
+ long long asa,
+ long long alen);
+void __sanitizer_syscall_post_impl_compat_43_ogetsockname(long long res,
+ long long fdec,
+ long long asa,
+ long long alen);
+/* syscall 151 has been skipped */
+/* syscall 152 has been skipped */
+/* syscall 153 has been skipped */
+/* syscall 154 has been skipped */
+void __sanitizer_syscall_pre_impl_nfssvc(long long flag, long long argp);
+void __sanitizer_syscall_post_impl_nfssvc(long long res, long long flag,
+ long long argp);
+void __sanitizer_syscall_pre_impl_compat_43_ogetdirentries(long long fd,
+ long long buf,
+ long long count,
+ long long basep);
+void __sanitizer_syscall_post_impl_compat_43_ogetdirentries(long long res,
+ long long fd,
+ long long buf,
+ long long count,
+ long long basep);
+void __sanitizer_syscall_pre_impl_compat_20_statfs(long long path,
+ long long buf);
+void __sanitizer_syscall_post_impl_compat_20_statfs(long long res,
+ long long path,
+ long long buf);
+void __sanitizer_syscall_pre_impl_compat_20_fstatfs(long long fd,
+ long long buf);
+void __sanitizer_syscall_post_impl_compat_20_fstatfs(long long res,
+ long long fd,
+ long long buf);
+/* syscall 159 has been skipped */
+/* syscall 160 has been skipped */
+void __sanitizer_syscall_pre_impl_compat_30_getfh(long long fname,
+ long long fhp);
+void __sanitizer_syscall_post_impl_compat_30_getfh(long long res,
+ long long fname,
+ long long fhp);
+void __sanitizer_syscall_pre_impl_compat_09_ogetdomainname(long long domainname,
+ long long len);
+void __sanitizer_syscall_post_impl_compat_09_ogetdomainname(
+ long long res, long long domainname, long long len);
+void __sanitizer_syscall_pre_impl_compat_09_osetdomainname(long long domainname,
+ long long len);
+void __sanitizer_syscall_post_impl_compat_09_osetdomainname(
+ long long res, long long domainname, long long len);
+void __sanitizer_syscall_pre_impl_compat_09_ouname(long long name);
+void __sanitizer_syscall_post_impl_compat_09_ouname(long long res,
+ long long name);
+void __sanitizer_syscall_pre_impl_sysarch(long long op, long long parms);
+void __sanitizer_syscall_post_impl_sysarch(long long res, long long op,
+ long long parms);
+/* syscall 166 has been skipped */
+/* syscall 167 has been skipped */
+/* syscall 168 has been skipped */
+#if !defined(_LP64)
+void __sanitizer_syscall_pre_impl_compat_10_osemsys(long long which,
+ long long a2, long long a3,
+ long long a4, long long a5);
+void __sanitizer_syscall_post_impl_compat_10_osemsys(long long res,
+ long long which,
+ long long a2, long long a3,
+ long long a4,
+ long long a5);
+#else
+/* syscall 169 has been skipped */
+#endif
+#if !defined(_LP64)
+void __sanitizer_syscall_pre_impl_compat_10_omsgsys(long long which,
+ long long a2, long long a3,
+ long long a4, long long a5,
+ long long a6);
+void __sanitizer_syscall_post_impl_compat_10_omsgsys(long long res,
+ long long which,
+ long long a2, long long a3,
+ long long a4, long long a5,
+ long long a6);
+#else
+/* syscall 170 has been skipped */
+#endif
+#if !defined(_LP64)
+void __sanitizer_syscall_pre_impl_compat_10_oshmsys(long long which,
+ long long a2, long long a3,
+ long long a4);
+void __sanitizer_syscall_post_impl_compat_10_oshmsys(long long res,
+ long long which,
+ long long a2, long long a3,
+ long long a4);
+#else
+/* syscall 171 has been skipped */
+#endif
+/* syscall 172 has been skipped */
+void __sanitizer_syscall_pre_impl_pread(long long fd, long long buf,
+ long long nbyte, long long PAD,
+ long long offset);
+void __sanitizer_syscall_post_impl_pread(long long res, long long fd,
+ long long buf, long long nbyte,
+ long long PAD, long long offset);
+void __sanitizer_syscall_pre_impl_pwrite(long long fd, long long buf,
+ long long nbyte, long long PAD,
+ long long offset);
+void __sanitizer_syscall_post_impl_pwrite(long long res, long long fd,
+ long long buf, long long nbyte,
+ long long PAD, long long offset);
+void __sanitizer_syscall_pre_impl_compat_30_ntp_gettime(long long ntvp);
+void __sanitizer_syscall_post_impl_compat_30_ntp_gettime(long long res,
+ long long ntvp);
+#if defined(NTP) || !defined(_KERNEL_OPT)
+void __sanitizer_syscall_pre_impl_ntp_adjtime(long long tp);
+void __sanitizer_syscall_post_impl_ntp_adjtime(long long res, long long tp);
+#else
+/* syscall 176 has been skipped */
+#endif
+/* syscall 177 has been skipped */
+/* syscall 178 has been skipped */
+/* syscall 179 has been skipped */
+/* syscall 180 has been skipped */
+void __sanitizer_syscall_pre_impl_setgid(long long gid);
+void __sanitizer_syscall_post_impl_setgid(long long res, long long gid);
+void __sanitizer_syscall_pre_impl_setegid(long long egid);
+void __sanitizer_syscall_post_impl_setegid(long long res, long long egid);
+void __sanitizer_syscall_pre_impl_seteuid(long long euid);
+void __sanitizer_syscall_post_impl_seteuid(long long res, long long euid);
+void __sanitizer_syscall_pre_impl_lfs_bmapv(long long fsidp, long long blkiov,
+ long long blkcnt);
+void __sanitizer_syscall_post_impl_lfs_bmapv(long long res, long long fsidp,
+ long long blkiov,
+ long long blkcnt);
+void __sanitizer_syscall_pre_impl_lfs_markv(long long fsidp, long long blkiov,
+ long long blkcnt);
+void __sanitizer_syscall_post_impl_lfs_markv(long long res, long long fsidp,
+ long long blkiov,
+ long long blkcnt);
+void __sanitizer_syscall_pre_impl_lfs_segclean(long long fsidp,
+ long long segment);
+void __sanitizer_syscall_post_impl_lfs_segclean(long long res, long long fsidp,
+ long long segment);
+void __sanitizer_syscall_pre_impl_compat_50_lfs_segwait(long long fsidp,
+ long long tv);
+void __sanitizer_syscall_post_impl_compat_50_lfs_segwait(long long res,
+ long long fsidp,
+ long long tv);
+void __sanitizer_syscall_pre_impl_compat_12_stat12(long long path,
+ long long ub);
+void __sanitizer_syscall_post_impl_compat_12_stat12(long long res,
+ long long path,
+ long long ub);
+void __sanitizer_syscall_pre_impl_compat_12_fstat12(long long fd, long long sb);
+void __sanitizer_syscall_post_impl_compat_12_fstat12(long long res,
+ long long fd,
+ long long sb);
+void __sanitizer_syscall_pre_impl_compat_12_lstat12(long long path,
+ long long ub);
+void __sanitizer_syscall_post_impl_compat_12_lstat12(long long res,
+ long long path,
+ long long ub);
+void __sanitizer_syscall_pre_impl_pathconf(long long path, long long name);
+void __sanitizer_syscall_post_impl_pathconf(long long res, long long path,
+ long long name);
+void __sanitizer_syscall_pre_impl_fpathconf(long long fd, long long name);
+void __sanitizer_syscall_post_impl_fpathconf(long long res, long long fd,
+ long long name);
+void __sanitizer_syscall_pre_impl_getsockopt2(long long s, long long level,
+ long long name, long long val,
+ long long avalsize);
+void __sanitizer_syscall_post_impl_getsockopt2(long long res, long long s,
+ long long level, long long name,
+ long long val,
+ long long avalsize);
+void __sanitizer_syscall_pre_impl_getrlimit(long long which, long long rlp);
+void __sanitizer_syscall_post_impl_getrlimit(long long res, long long which,
+ long long rlp);
+void __sanitizer_syscall_pre_impl_setrlimit(long long which, long long rlp);
+void __sanitizer_syscall_post_impl_setrlimit(long long res, long long which,
+ long long rlp);
+void __sanitizer_syscall_pre_impl_compat_12_getdirentries(long long fd,
+ long long buf,
+ long long count,
+ long long basep);
+void __sanitizer_syscall_post_impl_compat_12_getdirentries(long long res,
+ long long fd,
+ long long buf,
+ long long count,
+ long long basep);
+void __sanitizer_syscall_pre_impl_mmap(long long addr, long long len,
+ long long prot, long long flags,
+ long long fd, long long PAD,
+ long long pos);
+void __sanitizer_syscall_post_impl_mmap(long long res, long long addr,
+ long long len, long long prot,
+ long long flags, long long fd,
+ long long PAD, long long pos);
+void __sanitizer_syscall_pre_impl___syscall(long long code, long long arg0,
+ long long arg1, long long arg2,
+ long long arg3, long long arg4,
+ long long arg5, long long arg6,
+ long long arg7);
+void __sanitizer_syscall_post_impl___syscall(long long res, long long code,
+ long long arg0, long long arg1,
+ long long arg2, long long arg3,
+ long long arg4, long long arg5,
+ long long arg6, long long arg7);
+void __sanitizer_syscall_pre_impl_lseek(long long fd, long long PAD,
+ long long offset, long long whence);
+void __sanitizer_syscall_post_impl_lseek(long long res, long long fd,
+ long long PAD, long long offset,
+ long long whence);
+void __sanitizer_syscall_pre_impl_truncate(long long path, long long PAD,
+ long long length);
+void __sanitizer_syscall_post_impl_truncate(long long res, long long path,
+ long long PAD, long long length);
+void __sanitizer_syscall_pre_impl_ftruncate(long long fd, long long PAD,
+ long long length);
+void __sanitizer_syscall_post_impl_ftruncate(long long res, long long fd,
+ long long PAD, long long length);
+void __sanitizer_syscall_pre_impl___sysctl(long long name, long long namelen,
+ long long oldv, long long oldlenp,
+ long long newv, long long newlen);
+void __sanitizer_syscall_post_impl___sysctl(long long res, long long name,
+ long long namelen, long long oldv,
+ long long oldlenp, long long newv,
+ long long newlen);
+void __sanitizer_syscall_pre_impl_mlock(long long addr, long long len);
+void __sanitizer_syscall_post_impl_mlock(long long res, long long addr,
+ long long len);
+void __sanitizer_syscall_pre_impl_munlock(long long addr, long long len);
+void __sanitizer_syscall_post_impl_munlock(long long res, long long addr,
+ long long len);
+void __sanitizer_syscall_pre_impl_undelete(long long path);
+void __sanitizer_syscall_post_impl_undelete(long long res, long long path);
+void __sanitizer_syscall_pre_impl_compat_50_futimes(long long fd,
+ long long tptr);
+void __sanitizer_syscall_post_impl_compat_50_futimes(long long res,
+ long long fd,
+ long long tptr);
+void __sanitizer_syscall_pre_impl_getpgid(long long pid);
+void __sanitizer_syscall_post_impl_getpgid(long long res, long long pid);
+void __sanitizer_syscall_pre_impl_reboot(long long opt, long long bootstr);
+void __sanitizer_syscall_post_impl_reboot(long long res, long long opt,
+ long long bootstr);
+void __sanitizer_syscall_pre_impl_poll(long long fds, long long nfds,
+ long long timeout);
+void __sanitizer_syscall_post_impl_poll(long long res, long long fds,
+ long long nfds, long long timeout);
+void __sanitizer_syscall_pre_impl_afssys(long long id, long long a1,
+ long long a2, long long a3,
+ long long a4, long long a5,
+ long long a6);
+void __sanitizer_syscall_post_impl_afssys(long long res, long long id,
+ long long a1, long long a2,
+ long long a3, long long a4,
+ long long a5, long long a6);
+/* syscall 211 has been skipped */
+/* syscall 212 has been skipped */
+/* syscall 213 has been skipped */
+/* syscall 214 has been skipped */
+/* syscall 215 has been skipped */
+/* syscall 216 has been skipped */
+/* syscall 217 has been skipped */
+/* syscall 218 has been skipped */
+/* syscall 219 has been skipped */
+void __sanitizer_syscall_pre_impl_compat_14___semctl(long long semid,
+ long long semnum,
+ long long cmd,
+ long long arg);
+void __sanitizer_syscall_post_impl_compat_14___semctl(long long res,
+ long long semid,
+ long long semnum,
+ long long cmd,
+ long long arg);
+void __sanitizer_syscall_pre_impl_semget(long long key, long long nsems,
+ long long semflg);
+void __sanitizer_syscall_post_impl_semget(long long res, long long key,
+ long long nsems, long long semflg);
+void __sanitizer_syscall_pre_impl_semop(long long semid, long long sops,
+ long long nsops);
+void __sanitizer_syscall_post_impl_semop(long long res, long long semid,
+ long long sops, long long nsops);
+void __sanitizer_syscall_pre_impl_semconfig(long long flag);
+void __sanitizer_syscall_post_impl_semconfig(long long res, long long flag);
+void __sanitizer_syscall_pre_impl_compat_14_msgctl(long long msqid,
+ long long cmd,
+ long long buf);
+void __sanitizer_syscall_post_impl_compat_14_msgctl(long long res,
+ long long msqid,
+ long long cmd,
+ long long buf);
+void __sanitizer_syscall_pre_impl_msgget(long long key, long long msgflg);
+void __sanitizer_syscall_post_impl_msgget(long long res, long long key,
+ long long msgflg);
+void __sanitizer_syscall_pre_impl_msgsnd(long long msqid, long long msgp,
+ long long msgsz, long long msgflg);
+void __sanitizer_syscall_post_impl_msgsnd(long long res, long long msqid,
+ long long msgp, long long msgsz,
+ long long msgflg);
+void __sanitizer_syscall_pre_impl_msgrcv(long long msqid, long long msgp,
+ long long msgsz, long long msgtyp,
+ long long msgflg);
+void __sanitizer_syscall_post_impl_msgrcv(long long res, long long msqid,
+ long long msgp, long long msgsz,
+ long long msgtyp, long long msgflg);
+void __sanitizer_syscall_pre_impl_shmat(long long shmid, long long shmaddr,
+ long long shmflg);
+void __sanitizer_syscall_post_impl_shmat(long long res, long long shmid,
+ long long shmaddr, long long shmflg);
+void __sanitizer_syscall_pre_impl_compat_14_shmctl(long long shmid,
+ long long cmd,
+ long long buf);
+void __sanitizer_syscall_post_impl_compat_14_shmctl(long long res,
+ long long shmid,
+ long long cmd,
+ long long buf);
+void __sanitizer_syscall_pre_impl_shmdt(long long shmaddr);
+void __sanitizer_syscall_post_impl_shmdt(long long res, long long shmaddr);
+void __sanitizer_syscall_pre_impl_shmget(long long key, long long size,
+ long long shmflg);
+void __sanitizer_syscall_post_impl_shmget(long long res, long long key,
+ long long size, long long shmflg);
+void __sanitizer_syscall_pre_impl_compat_50_clock_gettime(long long clock_id,
+ long long tp);
+void __sanitizer_syscall_post_impl_compat_50_clock_gettime(long long res,
+ long long clock_id,
+ long long tp);
+void __sanitizer_syscall_pre_impl_compat_50_clock_settime(long long clock_id,
+ long long tp);
+void __sanitizer_syscall_post_impl_compat_50_clock_settime(long long res,
+ long long clock_id,
+ long long tp);
+void __sanitizer_syscall_pre_impl_compat_50_clock_getres(long long clock_id,
+ long long tp);
+void __sanitizer_syscall_post_impl_compat_50_clock_getres(long long res,
+ long long clock_id,
+ long long tp);
+void __sanitizer_syscall_pre_impl_timer_create(long long clock_id,
+ long long evp,
+ long long timerid);
+void __sanitizer_syscall_post_impl_timer_create(long long res,
+ long long clock_id,
+ long long evp,
+ long long timerid);
+void __sanitizer_syscall_pre_impl_timer_delete(long long timerid);
+void __sanitizer_syscall_post_impl_timer_delete(long long res,
+ long long timerid);
+void __sanitizer_syscall_pre_impl_compat_50_timer_settime(long long timerid,
+ long long flags,
+ long long value,
+ long long ovalue);
+void __sanitizer_syscall_post_impl_compat_50_timer_settime(long long res,
+ long long timerid,
+ long long flags,
+ long long value,
+ long long ovalue);
+void __sanitizer_syscall_pre_impl_compat_50_timer_gettime(long long timerid,
+ long long value);
+void __sanitizer_syscall_post_impl_compat_50_timer_gettime(long long res,
+ long long timerid,
+ long long value);
+void __sanitizer_syscall_pre_impl_timer_getoverrun(long long timerid);
+void __sanitizer_syscall_post_impl_timer_getoverrun(long long res,
+ long long timerid);
+void __sanitizer_syscall_pre_impl_compat_50_nanosleep(long long rqtp,
+ long long rmtp);
+void __sanitizer_syscall_post_impl_compat_50_nanosleep(long long res,
+ long long rqtp,
+ long long rmtp);
+void __sanitizer_syscall_pre_impl_fdatasync(long long fd);
+void __sanitizer_syscall_post_impl_fdatasync(long long res, long long fd);
+void __sanitizer_syscall_pre_impl_mlockall(long long flags);
+void __sanitizer_syscall_post_impl_mlockall(long long res, long long flags);
+void __sanitizer_syscall_pre_impl_munlockall(void);
+void __sanitizer_syscall_post_impl_munlockall(long long res);
+void __sanitizer_syscall_pre_impl_compat_50___sigtimedwait(long long set,
+ long long info,
+ long long timeout);
+void __sanitizer_syscall_post_impl_compat_50___sigtimedwait(long long res,
+ long long set,
+ long long info,
+ long long timeout);
+void __sanitizer_syscall_pre_impl_sigqueueinfo(long long pid, long long info);
+void __sanitizer_syscall_post_impl_sigqueueinfo(long long res, long long pid,
+ long long info);
+void __sanitizer_syscall_pre_impl_modctl(long long cmd, long long arg);
+void __sanitizer_syscall_post_impl_modctl(long long res, long long cmd,
+ long long arg);
+void __sanitizer_syscall_pre_impl__ksem_init(long long value, long long idp);
+void __sanitizer_syscall_post_impl__ksem_init(long long res, long long value,
+ long long idp);
+void __sanitizer_syscall_pre_impl__ksem_open(long long name, long long oflag,
+ long long mode, long long value,
+ long long idp);
+void __sanitizer_syscall_post_impl__ksem_open(long long res, long long name,
+ long long oflag, long long mode,
+ long long value, long long idp);
+void __sanitizer_syscall_pre_impl__ksem_unlink(long long name);
+void __sanitizer_syscall_post_impl__ksem_unlink(long long res, long long name);
+void __sanitizer_syscall_pre_impl__ksem_close(long long id);
+void __sanitizer_syscall_post_impl__ksem_close(long long res, long long id);
+void __sanitizer_syscall_pre_impl__ksem_post(long long id);
+void __sanitizer_syscall_post_impl__ksem_post(long long res, long long id);
+void __sanitizer_syscall_pre_impl__ksem_wait(long long id);
+void __sanitizer_syscall_post_impl__ksem_wait(long long res, long long id);
+void __sanitizer_syscall_pre_impl__ksem_trywait(long long id);
+void __sanitizer_syscall_post_impl__ksem_trywait(long long res, long long id);
+void __sanitizer_syscall_pre_impl__ksem_getvalue(long long id, long long value);
+void __sanitizer_syscall_post_impl__ksem_getvalue(long long res, long long id,
+ long long value);
+void __sanitizer_syscall_pre_impl__ksem_destroy(long long id);
+void __sanitizer_syscall_post_impl__ksem_destroy(long long res, long long id);
+void __sanitizer_syscall_pre_impl__ksem_timedwait(long long id,
+ long long abstime);
+void __sanitizer_syscall_post_impl__ksem_timedwait(long long res, long long id,
+ long long abstime);
+void __sanitizer_syscall_pre_impl_mq_open(long long name, long long oflag,
+ long long mode, long long attr);
+void __sanitizer_syscall_post_impl_mq_open(long long res, long long name,
+ long long oflag, long long mode,
+ long long attr);
+void __sanitizer_syscall_pre_impl_mq_close(long long mqdes);
+void __sanitizer_syscall_post_impl_mq_close(long long res, long long mqdes);
+void __sanitizer_syscall_pre_impl_mq_unlink(long long name);
+void __sanitizer_syscall_post_impl_mq_unlink(long long res, long long name);
+void __sanitizer_syscall_pre_impl_mq_getattr(long long mqdes, long long mqstat);
+void __sanitizer_syscall_post_impl_mq_getattr(long long res, long long mqdes,
+ long long mqstat);
+void __sanitizer_syscall_pre_impl_mq_setattr(long long mqdes, long long mqstat,
+ long long omqstat);
+void __sanitizer_syscall_post_impl_mq_setattr(long long res, long long mqdes,
+ long long mqstat,
+ long long omqstat);
+void __sanitizer_syscall_pre_impl_mq_notify(long long mqdes,
+ long long notification);
+void __sanitizer_syscall_post_impl_mq_notify(long long res, long long mqdes,
+ long long notification);
+void __sanitizer_syscall_pre_impl_mq_send(long long mqdes, long long msg_ptr,
+ long long msg_len,
+ long long msg_prio);
+void __sanitizer_syscall_post_impl_mq_send(long long res, long long mqdes,
+ long long msg_ptr, long long msg_len,
+ long long msg_prio);
+void __sanitizer_syscall_pre_impl_mq_receive(long long mqdes, long long msg_ptr,
+ long long msg_len,
+ long long msg_prio);
+void __sanitizer_syscall_post_impl_mq_receive(long long res, long long mqdes,
+ long long msg_ptr,
+ long long msg_len,
+ long long msg_prio);
+void __sanitizer_syscall_pre_impl_compat_50_mq_timedsend(long long mqdes,
+ long long msg_ptr,
+ long long msg_len,
+ long long msg_prio,
+ long long abs_timeout);
+void __sanitizer_syscall_post_impl_compat_50_mq_timedsend(
+ long long res, long long mqdes, long long msg_ptr, long long msg_len,
+ long long msg_prio, long long abs_timeout);
+void __sanitizer_syscall_pre_impl_compat_50_mq_timedreceive(
+ long long mqdes, long long msg_ptr, long long msg_len, long long msg_prio,
+ long long abs_timeout);
+void __sanitizer_syscall_post_impl_compat_50_mq_timedreceive(
+ long long res, long long mqdes, long long msg_ptr, long long msg_len,
+ long long msg_prio, long long abs_timeout);
+/* syscall 267 has been skipped */
+/* syscall 268 has been skipped */
+/* syscall 269 has been skipped */
+void __sanitizer_syscall_pre_impl___posix_rename(long long from, long long to);
+void __sanitizer_syscall_post_impl___posix_rename(long long res, long long from,
+ long long to);
+void __sanitizer_syscall_pre_impl_swapctl(long long cmd, long long arg,
+ long long misc);
+void __sanitizer_syscall_post_impl_swapctl(long long res, long long cmd,
+ long long arg, long long misc);
+void __sanitizer_syscall_pre_impl_compat_30_getdents(long long fd,
+ long long buf,
+ long long count);
+void __sanitizer_syscall_post_impl_compat_30_getdents(long long res,
+ long long fd,
+ long long buf,
+ long long count);
+void __sanitizer_syscall_pre_impl_minherit(long long addr, long long len,
+ long long inherit);
+void __sanitizer_syscall_post_impl_minherit(long long res, long long addr,
+ long long len, long long inherit);
+void __sanitizer_syscall_pre_impl_lchmod(long long path, long long mode);
+void __sanitizer_syscall_post_impl_lchmod(long long res, long long path,
+ long long mode);
+void __sanitizer_syscall_pre_impl_lchown(long long path, long long uid,
+ long long gid);
+void __sanitizer_syscall_post_impl_lchown(long long res, long long path,
+ long long uid, long long gid);
+void __sanitizer_syscall_pre_impl_compat_50_lutimes(long long path,
+ long long tptr);
+void __sanitizer_syscall_post_impl_compat_50_lutimes(long long res,
+ long long path,
+ long long tptr);
+void __sanitizer_syscall_pre_impl___msync13(long long addr, long long len,
+ long long flags);
+void __sanitizer_syscall_post_impl___msync13(long long res, long long addr,
+ long long len, long long flags);
+void __sanitizer_syscall_pre_impl_compat_30___stat13(long long path,
+ long long ub);
+void __sanitizer_syscall_post_impl_compat_30___stat13(long long res,
+ long long path,
+ long long ub);
+void __sanitizer_syscall_pre_impl_compat_30___fstat13(long long fd,
+ long long sb);
+void __sanitizer_syscall_post_impl_compat_30___fstat13(long long res,
+ long long fd,
+ long long sb);
+void __sanitizer_syscall_pre_impl_compat_30___lstat13(long long path,
+ long long ub);
+void __sanitizer_syscall_post_impl_compat_30___lstat13(long long res,
+ long long path,
+ long long ub);
+void __sanitizer_syscall_pre_impl___sigaltstack14(long long nss, long long oss);
+void __sanitizer_syscall_post_impl___sigaltstack14(long long res, long long nss,
+ long long oss);
+void __sanitizer_syscall_pre_impl___vfork14(void);
+void __sanitizer_syscall_post_impl___vfork14(long long res);
+void __sanitizer_syscall_pre_impl___posix_chown(long long path, long long uid,
+ long long gid);
+void __sanitizer_syscall_post_impl___posix_chown(long long res, long long path,
+ long long uid, long long gid);
+void __sanitizer_syscall_pre_impl___posix_fchown(long long fd, long long uid,
+ long long gid);
+void __sanitizer_syscall_post_impl___posix_fchown(long long res, long long fd,
+ long long uid, long long gid);
+void __sanitizer_syscall_pre_impl___posix_lchown(long long path, long long uid,
+ long long gid);
+void __sanitizer_syscall_post_impl___posix_lchown(long long res, long long path,
+ long long uid, long long gid);
+void __sanitizer_syscall_pre_impl_getsid(long long pid);
+void __sanitizer_syscall_post_impl_getsid(long long res, long long pid);
+void __sanitizer_syscall_pre_impl___clone(long long flags, long long stack);
+void __sanitizer_syscall_post_impl___clone(long long res, long long flags,
+ long long stack);
+void __sanitizer_syscall_pre_impl_fktrace(long long fd, long long ops,
+ long long facs, long long pid);
+void __sanitizer_syscall_post_impl_fktrace(long long res, long long fd,
+ long long ops, long long facs,
+ long long pid);
+void __sanitizer_syscall_pre_impl_preadv(long long fd, long long iovp,
+ long long iovcnt, long long PAD,
+ long long offset);
+void __sanitizer_syscall_post_impl_preadv(long long res, long long fd,
+ long long iovp, long long iovcnt,
+ long long PAD, long long offset);
+void __sanitizer_syscall_pre_impl_pwritev(long long fd, long long iovp,
+ long long iovcnt, long long PAD,
+ long long offset);
+void __sanitizer_syscall_post_impl_pwritev(long long res, long long fd,
+ long long iovp, long long iovcnt,
+ long long PAD, long long offset);
+void __sanitizer_syscall_pre_impl_compat_16___sigaction14(long long signum,
+ long long nsa,
+ long long osa);
+void __sanitizer_syscall_post_impl_compat_16___sigaction14(long long res,
+ long long signum,
+ long long nsa,
+ long long osa);
+void __sanitizer_syscall_pre_impl___sigpending14(long long set);
+void __sanitizer_syscall_post_impl___sigpending14(long long res, long long set);
+void __sanitizer_syscall_pre_impl___sigprocmask14(long long how, long long set,
+ long long oset);
+void __sanitizer_syscall_post_impl___sigprocmask14(long long res, long long how,
+ long long set,
+ long long oset);
+void __sanitizer_syscall_pre_impl___sigsuspend14(long long set);
+void __sanitizer_syscall_post_impl___sigsuspend14(long long res, long long set);
+void __sanitizer_syscall_pre_impl_compat_16___sigreturn14(long long sigcntxp);
+void __sanitizer_syscall_post_impl_compat_16___sigreturn14(long long res,
+ long long sigcntxp);
+void __sanitizer_syscall_pre_impl___getcwd(long long bufp, long long length);
+void __sanitizer_syscall_post_impl___getcwd(long long res, long long bufp,
+ long long length);
+void __sanitizer_syscall_pre_impl_fchroot(long long fd);
+void __sanitizer_syscall_post_impl_fchroot(long long res, long long fd);
+void __sanitizer_syscall_pre_impl_compat_30_fhopen(long long fhp,
+ long long flags);
+void __sanitizer_syscall_post_impl_compat_30_fhopen(long long res,
+ long long fhp,
+ long long flags);
+void __sanitizer_syscall_pre_impl_compat_30_fhstat(long long fhp, long long sb);
+void __sanitizer_syscall_post_impl_compat_30_fhstat(long long res,
+ long long fhp,
+ long long sb);
+void __sanitizer_syscall_pre_impl_compat_20_fhstatfs(long long fhp,
+ long long buf);
+void __sanitizer_syscall_post_impl_compat_20_fhstatfs(long long res,
+ long long fhp,
+ long long buf);
+void __sanitizer_syscall_pre_impl_compat_50_____semctl13(long long semid,
+ long long semnum,
+ long long cmd,
+ long long arg);
+void __sanitizer_syscall_post_impl_compat_50_____semctl13(long long res,
+ long long semid,
+ long long semnum,
+ long long cmd,
+ long long arg);
+void __sanitizer_syscall_pre_impl_compat_50___msgctl13(long long msqid,
+ long long cmd,
+ long long buf);
+void __sanitizer_syscall_post_impl_compat_50___msgctl13(long long res,
+ long long msqid,
+ long long cmd,
+ long long buf);
+void __sanitizer_syscall_pre_impl_compat_50___shmctl13(long long shmid,
+ long long cmd,
+ long long buf);
+void __sanitizer_syscall_post_impl_compat_50___shmctl13(long long res,
+ long long shmid,
+ long long cmd,
+ long long buf);
+void __sanitizer_syscall_pre_impl_lchflags(long long path, long long flags);
+void __sanitizer_syscall_post_impl_lchflags(long long res, long long path,
+ long long flags);
+void __sanitizer_syscall_pre_impl_issetugid(void);
+void __sanitizer_syscall_post_impl_issetugid(long long res);
+void __sanitizer_syscall_pre_impl_utrace(long long label, long long addr,
+ long long len);
+void __sanitizer_syscall_post_impl_utrace(long long res, long long label,
+ long long addr, long long len);
+void __sanitizer_syscall_pre_impl_getcontext(long long ucp);
+void __sanitizer_syscall_post_impl_getcontext(long long res, long long ucp);
+void __sanitizer_syscall_pre_impl_setcontext(long long ucp);
+void __sanitizer_syscall_post_impl_setcontext(long long res, long long ucp);
+void __sanitizer_syscall_pre_impl__lwp_create(long long ucp, long long flags,
+ long long new_lwp);
+void __sanitizer_syscall_post_impl__lwp_create(long long res, long long ucp,
+ long long flags,
+ long long new_lwp);
+void __sanitizer_syscall_pre_impl__lwp_exit(void);
+void __sanitizer_syscall_post_impl__lwp_exit(long long res);
+void __sanitizer_syscall_pre_impl__lwp_self(void);
+void __sanitizer_syscall_post_impl__lwp_self(long long res);
+void __sanitizer_syscall_pre_impl__lwp_wait(long long wait_for,
+ long long departed);
+void __sanitizer_syscall_post_impl__lwp_wait(long long res, long long wait_for,
+ long long departed);
+void __sanitizer_syscall_pre_impl__lwp_suspend(long long target);
+void __sanitizer_syscall_post_impl__lwp_suspend(long long res,
+ long long target);
+void __sanitizer_syscall_pre_impl__lwp_continue(long long target);
+void __sanitizer_syscall_post_impl__lwp_continue(long long res,
+ long long target);
+void __sanitizer_syscall_pre_impl__lwp_wakeup(long long target);
+void __sanitizer_syscall_post_impl__lwp_wakeup(long long res, long long target);
+void __sanitizer_syscall_pre_impl__lwp_getprivate(void);
+void __sanitizer_syscall_post_impl__lwp_getprivate(long long res);
+void __sanitizer_syscall_pre_impl__lwp_setprivate(long long ptr);
+void __sanitizer_syscall_post_impl__lwp_setprivate(long long res,
+ long long ptr);
+void __sanitizer_syscall_pre_impl__lwp_kill(long long target, long long signo);
+void __sanitizer_syscall_post_impl__lwp_kill(long long res, long long target,
+ long long signo);
+void __sanitizer_syscall_pre_impl__lwp_detach(long long target);
+void __sanitizer_syscall_post_impl__lwp_detach(long long res, long long target);
+void __sanitizer_syscall_pre_impl_compat_50__lwp_park(long long ts,
+ long long unpark,
+ long long hint,
+ long long unparkhint);
+void __sanitizer_syscall_post_impl_compat_50__lwp_park(long long res,
+ long long ts,
+ long long unpark,
+ long long hint,
+ long long unparkhint);
+void __sanitizer_syscall_pre_impl__lwp_unpark(long long target, long long hint);
+void __sanitizer_syscall_post_impl__lwp_unpark(long long res, long long target,
+ long long hint);
+void __sanitizer_syscall_pre_impl__lwp_unpark_all(long long targets,
+ long long ntargets,
+ long long hint);
+void __sanitizer_syscall_post_impl__lwp_unpark_all(long long res,
+ long long targets,
+ long long ntargets,
+ long long hint);
+void __sanitizer_syscall_pre_impl__lwp_setname(long long target,
+ long long name);
+void __sanitizer_syscall_post_impl__lwp_setname(long long res, long long target,
+ long long name);
+void __sanitizer_syscall_pre_impl__lwp_getname(long long target, long long name,
+ long long len);
+void __sanitizer_syscall_post_impl__lwp_getname(long long res, long long target,
+ long long name, long long len);
+void __sanitizer_syscall_pre_impl__lwp_ctl(long long features,
+ long long address);
+void __sanitizer_syscall_post_impl__lwp_ctl(long long res, long long features,
+ long long address);
+/* syscall 326 has been skipped */
+/* syscall 327 has been skipped */
+/* syscall 328 has been skipped */
+/* syscall 329 has been skipped */
+void __sanitizer_syscall_pre_impl_compat_60_sa_register(
+ long long newv, long long oldv, long long flags,
+ long long stackinfo_offset);
+void __sanitizer_syscall_post_impl_compat_60_sa_register(
+ long long res, long long newv, long long oldv, long long flags,
+ long long stackinfo_offset);
+void __sanitizer_syscall_pre_impl_compat_60_sa_stacks(long long num,
+ long long stacks);
+void __sanitizer_syscall_post_impl_compat_60_sa_stacks(long long res,
+ long long num,
+ long long stacks);
+void __sanitizer_syscall_pre_impl_compat_60_sa_enable(void);
+void __sanitizer_syscall_post_impl_compat_60_sa_enable(long long res);
+void __sanitizer_syscall_pre_impl_compat_60_sa_setconcurrency(
+ long long concurrency);
+void __sanitizer_syscall_post_impl_compat_60_sa_setconcurrency(
+ long long res, long long concurrency);
+void __sanitizer_syscall_pre_impl_compat_60_sa_yield(void);
+void __sanitizer_syscall_post_impl_compat_60_sa_yield(long long res);
+void __sanitizer_syscall_pre_impl_compat_60_sa_preempt(long long sa_id);
+void __sanitizer_syscall_post_impl_compat_60_sa_preempt(long long res,
+ long long sa_id);
+/* syscall 336 has been skipped */
+/* syscall 337 has been skipped */
+/* syscall 338 has been skipped */
+/* syscall 339 has been skipped */
+void __sanitizer_syscall_pre_impl___sigaction_sigtramp(long long signum,
+ long long nsa,
+ long long osa,
+ long long tramp,
+ long long vers);
+void __sanitizer_syscall_post_impl___sigaction_sigtramp(
+ long long res, long long signum, long long nsa, long long osa,
+ long long tramp, long long vers);
+/* syscall 341 has been skipped */
+/* syscall 342 has been skipped */
+void __sanitizer_syscall_pre_impl_rasctl(long long addr, long long len,
+ long long op);
+void __sanitizer_syscall_post_impl_rasctl(long long res, long long addr,
+ long long len, long long op);
+void __sanitizer_syscall_pre_impl_kqueue(void);
+void __sanitizer_syscall_post_impl_kqueue(long long res);
+void __sanitizer_syscall_pre_impl_compat_50_kevent(
+ long long fd, long long changelist, long long nchanges, long long eventlist,
+ long long nevents, long long timeout);
+void __sanitizer_syscall_post_impl_compat_50_kevent(
+ long long res, long long fd, long long changelist, long long nchanges,
+ long long eventlist, long long nevents, long long timeout);
+void __sanitizer_syscall_pre_impl__sched_setparam(long long pid, long long lid,
+ long long policy,
+ long long params);
+void __sanitizer_syscall_post_impl__sched_setparam(long long res, long long pid,
+ long long lid,
+ long long policy,
+ long long params);
+void __sanitizer_syscall_pre_impl__sched_getparam(long long pid, long long lid,
+ long long policy,
+ long long params);
+void __sanitizer_syscall_post_impl__sched_getparam(long long res, long long pid,
+ long long lid,
+ long long policy,
+ long long params);
+void __sanitizer_syscall_pre_impl__sched_setaffinity(long long pid,
+ long long lid,
+ long long size,
+ long long cpuset);
+void __sanitizer_syscall_post_impl__sched_setaffinity(long long res,
+ long long pid,
+ long long lid,
+ long long size,
+ long long cpuset);
+void __sanitizer_syscall_pre_impl__sched_getaffinity(long long pid,
+ long long lid,
+ long long size,
+ long long cpuset);
+void __sanitizer_syscall_post_impl__sched_getaffinity(long long res,
+ long long pid,
+ long long lid,
+ long long size,
+ long long cpuset);
+void __sanitizer_syscall_pre_impl_sched_yield(void);
+void __sanitizer_syscall_post_impl_sched_yield(long long res);
+void __sanitizer_syscall_pre_impl__sched_protect(long long priority);
+void __sanitizer_syscall_post_impl__sched_protect(long long res,
+ long long priority);
+/* syscall 352 has been skipped */
+/* syscall 353 has been skipped */
+void __sanitizer_syscall_pre_impl_fsync_range(long long fd, long long flags,
+ long long start,
+ long long length);
+void __sanitizer_syscall_post_impl_fsync_range(long long res, long long fd,
+ long long flags, long long start,
+ long long length);
+void __sanitizer_syscall_pre_impl_uuidgen(long long store, long long count);
+void __sanitizer_syscall_post_impl_uuidgen(long long res, long long store,
+ long long count);
+void __sanitizer_syscall_pre_impl_compat_90_getvfsstat(long long buf,
+ long long bufsize,
+ long long flags);
+void __sanitizer_syscall_post_impl_compat_90_getvfsstat(long long res,
+ long long buf,
+ long long bufsize,
+ long long flags);
+void __sanitizer_syscall_pre_impl_compat_90_statvfs1(long long path,
+ long long buf,
+ long long flags);
+void __sanitizer_syscall_post_impl_compat_90_statvfs1(long long res,
+ long long path,
+ long long buf,
+ long long flags);
+void __sanitizer_syscall_pre_impl_compat_90_fstatvfs1(long long fd,
+ long long buf,
+ long long flags);
+void __sanitizer_syscall_post_impl_compat_90_fstatvfs1(long long res,
+ long long fd,
+ long long buf,
+ long long flags);
+void __sanitizer_syscall_pre_impl_compat_30_fhstatvfs1(long long fhp,
+ long long buf,
+ long long flags);
+void __sanitizer_syscall_post_impl_compat_30_fhstatvfs1(long long res,
+ long long fhp,
+ long long buf,
+ long long flags);
+void __sanitizer_syscall_pre_impl_extattrctl(long long path, long long cmd,
+ long long filename,
+ long long attrnamespace,
+ long long attrname);
+void __sanitizer_syscall_post_impl_extattrctl(long long res, long long path,
+ long long cmd, long long filename,
+ long long attrnamespace,
+ long long attrname);
+void __sanitizer_syscall_pre_impl_extattr_set_file(long long path,
+ long long attrnamespace,
+ long long attrname,
+ long long data,
+ long long nbytes);
+void __sanitizer_syscall_post_impl_extattr_set_file(
+ long long res, long long path, long long attrnamespace, long long attrname,
+ long long data, long long nbytes);
+void __sanitizer_syscall_pre_impl_extattr_get_file(long long path,
+ long long attrnamespace,
+ long long attrname,
+ long long data,
+ long long nbytes);
+void __sanitizer_syscall_post_impl_extattr_get_file(
+ long long res, long long path, long long attrnamespace, long long attrname,
+ long long data, long long nbytes);
+void __sanitizer_syscall_pre_impl_extattr_delete_file(long long path,
+ long long attrnamespace,
+ long long attrname);
+void __sanitizer_syscall_post_impl_extattr_delete_file(long long res,
+ long long path,
+ long long attrnamespace,
+ long long attrname);
+void __sanitizer_syscall_pre_impl_extattr_set_fd(long long fd,
+ long long attrnamespace,
+ long long attrname,
+ long long data,
+ long long nbytes);
+void __sanitizer_syscall_post_impl_extattr_set_fd(long long res, long long fd,
+ long long attrnamespace,
+ long long attrname,
+ long long data,
+ long long nbytes);
+void __sanitizer_syscall_pre_impl_extattr_get_fd(long long fd,
+ long long attrnamespace,
+ long long attrname,
+ long long data,
+ long long nbytes);
+void __sanitizer_syscall_post_impl_extattr_get_fd(long long res, long long fd,
+ long long attrnamespace,
+ long long attrname,
+ long long data,
+ long long nbytes);
+void __sanitizer_syscall_pre_impl_extattr_delete_fd(long long fd,
+ long long attrnamespace,
+ long long attrname);
+void __sanitizer_syscall_post_impl_extattr_delete_fd(long long res,
+ long long fd,
+ long long attrnamespace,
+ long long attrname);
+void __sanitizer_syscall_pre_impl_extattr_set_link(long long path,
+ long long attrnamespace,
+ long long attrname,
+ long long data,
+ long long nbytes);
+void __sanitizer_syscall_post_impl_extattr_set_link(
+ long long res, long long path, long long attrnamespace, long long attrname,
+ long long data, long long nbytes);
+void __sanitizer_syscall_pre_impl_extattr_get_link(long long path,
+ long long attrnamespace,
+ long long attrname,
+ long long data,
+ long long nbytes);
+void __sanitizer_syscall_post_impl_extattr_get_link(
+ long long res, long long path, long long attrnamespace, long long attrname,
+ long long data, long long nbytes);
+void __sanitizer_syscall_pre_impl_extattr_delete_link(long long path,
+ long long attrnamespace,
+ long long attrname);
+void __sanitizer_syscall_post_impl_extattr_delete_link(long long res,
+ long long path,
+ long long attrnamespace,
+ long long attrname);
+void __sanitizer_syscall_pre_impl_extattr_list_fd(long long fd,
+ long long attrnamespace,
+ long long data,
+ long long nbytes);
+void __sanitizer_syscall_post_impl_extattr_list_fd(long long res, long long fd,
+ long long attrnamespace,
+ long long data,
+ long long nbytes);
+void __sanitizer_syscall_pre_impl_extattr_list_file(long long path,
+ long long attrnamespace,
+ long long data,
+ long long nbytes);
+void __sanitizer_syscall_post_impl_extattr_list_file(long long res,
+ long long path,
+ long long attrnamespace,
+ long long data,
+ long long nbytes);
+void __sanitizer_syscall_pre_impl_extattr_list_link(long long path,
+ long long attrnamespace,
+ long long data,
+ long long nbytes);
+void __sanitizer_syscall_post_impl_extattr_list_link(long long res,
+ long long path,
+ long long attrnamespace,
+ long long data,
+ long long nbytes);
+void __sanitizer_syscall_pre_impl_compat_50_pselect(long long nd, long long in,
+ long long ou, long long ex,
+ long long ts,
+ long long mask);
+void __sanitizer_syscall_post_impl_compat_50_pselect(long long res,
+ long long nd, long long in,
+ long long ou, long long ex,
+ long long ts,
+ long long mask);
+void __sanitizer_syscall_pre_impl_compat_50_pollts(long long fds,
+ long long nfds, long long ts,
+ long long mask);
+void __sanitizer_syscall_post_impl_compat_50_pollts(
+ long long res, long long fds, long long nfds, long long ts, long long mask);
+void __sanitizer_syscall_pre_impl_setxattr(long long path, long long name,
+ long long value, long long size,
+ long long flags);
+void __sanitizer_syscall_post_impl_setxattr(long long res, long long path,
+ long long name, long long value,
+ long long size, long long flags);
+void __sanitizer_syscall_pre_impl_lsetxattr(long long path, long long name,
+ long long value, long long size,
+ long long flags);
+void __sanitizer_syscall_post_impl_lsetxattr(long long res, long long path,
+ long long name, long long value,
+ long long size, long long flags);
+void __sanitizer_syscall_pre_impl_fsetxattr(long long fd, long long name,
+ long long value, long long size,
+ long long flags);
+void __sanitizer_syscall_post_impl_fsetxattr(long long res, long long fd,
+ long long name, long long value,
+ long long size, long long flags);
+void __sanitizer_syscall_pre_impl_getxattr(long long path, long long name,
+ long long value, long long size);
+void __sanitizer_syscall_post_impl_getxattr(long long res, long long path,
+ long long name, long long value,
+ long long size);
+void __sanitizer_syscall_pre_impl_lgetxattr(long long path, long long name,
+ long long value, long long size);
+void __sanitizer_syscall_post_impl_lgetxattr(long long res, long long path,
+ long long name, long long value,
+ long long size);
+void __sanitizer_syscall_pre_impl_fgetxattr(long long fd, long long name,
+ long long value, long long size);
+void __sanitizer_syscall_post_impl_fgetxattr(long long res, long long fd,
+ long long name, long long value,
+ long long size);
+void __sanitizer_syscall_pre_impl_listxattr(long long path, long long list,
+ long long size);
+void __sanitizer_syscall_post_impl_listxattr(long long res, long long path,
+ long long list, long long size);
+void __sanitizer_syscall_pre_impl_llistxattr(long long path, long long list,
+ long long size);
+void __sanitizer_syscall_post_impl_llistxattr(long long res, long long path,
+ long long list, long long size);
+void __sanitizer_syscall_pre_impl_flistxattr(long long fd, long long list,
+ long long size);
+void __sanitizer_syscall_post_impl_flistxattr(long long res, long long fd,
+ long long list, long long size);
+void __sanitizer_syscall_pre_impl_removexattr(long long path, long long name);
+void __sanitizer_syscall_post_impl_removexattr(long long res, long long path,
+ long long name);
+void __sanitizer_syscall_pre_impl_lremovexattr(long long path, long long name);
+void __sanitizer_syscall_post_impl_lremovexattr(long long res, long long path,
+ long long name);
+void __sanitizer_syscall_pre_impl_fremovexattr(long long fd, long long name);
+void __sanitizer_syscall_post_impl_fremovexattr(long long res, long long fd,
+ long long name);
+void __sanitizer_syscall_pre_impl_compat_50___stat30(long long path,
+ long long ub);
+void __sanitizer_syscall_post_impl_compat_50___stat30(long long res,
+ long long path,
+ long long ub);
+void __sanitizer_syscall_pre_impl_compat_50___fstat30(long long fd,
+ long long sb);
+void __sanitizer_syscall_post_impl_compat_50___fstat30(long long res,
+ long long fd,
+ long long sb);
+void __sanitizer_syscall_pre_impl_compat_50___lstat30(long long path,
+ long long ub);
+void __sanitizer_syscall_post_impl_compat_50___lstat30(long long res,
+ long long path,
+ long long ub);
+void __sanitizer_syscall_pre_impl___getdents30(long long fd, long long buf,
+ long long count);
+void __sanitizer_syscall_post_impl___getdents30(long long res, long long fd,
+ long long buf, long long count);
+void __sanitizer_syscall_pre_impl_posix_fadvise(long long);
+void __sanitizer_syscall_post_impl_posix_fadvise(long long res, long long);
+void __sanitizer_syscall_pre_impl_compat_30___fhstat30(long long fhp,
+ long long sb);
+void __sanitizer_syscall_post_impl_compat_30___fhstat30(long long res,
+ long long fhp,
+ long long sb);
+void __sanitizer_syscall_pre_impl_compat_50___ntp_gettime30(long long ntvp);
+void __sanitizer_syscall_post_impl_compat_50___ntp_gettime30(long long res,
+ long long ntvp);
+void __sanitizer_syscall_pre_impl___socket30(long long domain, long long type,
+ long long protocol);
+void __sanitizer_syscall_post_impl___socket30(long long res, long long domain,
+ long long type,
+ long long protocol);
+void __sanitizer_syscall_pre_impl___getfh30(long long fname, long long fhp,
+ long long fh_size);
+void __sanitizer_syscall_post_impl___getfh30(long long res, long long fname,
+ long long fhp, long long fh_size);
+void __sanitizer_syscall_pre_impl___fhopen40(long long fhp, long long fh_size,
+ long long flags);
+void __sanitizer_syscall_post_impl___fhopen40(long long res, long long fhp,
+ long long fh_size,
+ long long flags);
+void __sanitizer_syscall_pre_impl_compat_90_fhstatvfs1(long long fhp,
+ long long fh_size,
+ long long buf,
+ long long flags);
+void __sanitizer_syscall_post_impl_compat_90_fhstatvfs1(long long res,
+ long long fhp,
+ long long fh_size,
+ long long buf,
+ long long flags);
+void __sanitizer_syscall_pre_impl_compat_50___fhstat40(long long fhp,
+ long long fh_size,
+ long long sb);
+void __sanitizer_syscall_post_impl_compat_50___fhstat40(long long res,
+ long long fhp,
+ long long fh_size,
+ long long sb);
+void __sanitizer_syscall_pre_impl_aio_cancel(long long fildes,
+ long long aiocbp);
+void __sanitizer_syscall_post_impl_aio_cancel(long long res, long long fildes,
+ long long aiocbp);
+void __sanitizer_syscall_pre_impl_aio_error(long long aiocbp);
+void __sanitizer_syscall_post_impl_aio_error(long long res, long long aiocbp);
+void __sanitizer_syscall_pre_impl_aio_fsync(long long op, long long aiocbp);
+void __sanitizer_syscall_post_impl_aio_fsync(long long res, long long op,
+ long long aiocbp);
+void __sanitizer_syscall_pre_impl_aio_read(long long aiocbp);
+void __sanitizer_syscall_post_impl_aio_read(long long res, long long aiocbp);
+void __sanitizer_syscall_pre_impl_aio_return(long long aiocbp);
+void __sanitizer_syscall_post_impl_aio_return(long long res, long long aiocbp);
+void __sanitizer_syscall_pre_impl_compat_50_aio_suspend(long long list,
+ long long nent,
+ long long timeout);
+void __sanitizer_syscall_post_impl_compat_50_aio_suspend(long long res,
+ long long list,
+ long long nent,
+ long long timeout);
+void __sanitizer_syscall_pre_impl_aio_write(long long aiocbp);
+void __sanitizer_syscall_post_impl_aio_write(long long res, long long aiocbp);
+void __sanitizer_syscall_pre_impl_lio_listio(long long mode, long long list,
+ long long nent, long long sig);
+void __sanitizer_syscall_post_impl_lio_listio(long long res, long long mode,
+ long long list, long long nent,
+ long long sig);
+/* syscall 407 has been skipped */
+/* syscall 408 has been skipped */
+/* syscall 409 has been skipped */
+void __sanitizer_syscall_pre_impl___mount50(long long type, long long path,
+ long long flags, long long data,
+ long long data_len);
+void __sanitizer_syscall_post_impl___mount50(long long res, long long type,
+ long long path, long long flags,
+ long long data,
+ long long data_len);
+void __sanitizer_syscall_pre_impl_mremap(long long old_address,
+ long long old_size,
+ long long new_address,
+ long long new_size, long long flags);
+void __sanitizer_syscall_post_impl_mremap(long long res, long long old_address,
+ long long old_size,
+ long long new_address,
+ long long new_size, long long flags);
+void __sanitizer_syscall_pre_impl_pset_create(long long psid);
+void __sanitizer_syscall_post_impl_pset_create(long long res, long long psid);
+void __sanitizer_syscall_pre_impl_pset_destroy(long long psid);
+void __sanitizer_syscall_post_impl_pset_destroy(long long res, long long psid);
+void __sanitizer_syscall_pre_impl_pset_assign(long long psid, long long cpuid,
+ long long opsid);
+void __sanitizer_syscall_post_impl_pset_assign(long long res, long long psid,
+ long long cpuid,
+ long long opsid);
+void __sanitizer_syscall_pre_impl__pset_bind(long long idtype,
+ long long first_id,
+ long long second_id,
+ long long psid, long long opsid);
+void __sanitizer_syscall_post_impl__pset_bind(long long res, long long idtype,
+ long long first_id,
+ long long second_id,
+ long long psid, long long opsid);
+void __sanitizer_syscall_pre_impl___posix_fadvise50(long long fd, long long PAD,
+ long long offset,
+ long long len,
+ long long advice);
+void __sanitizer_syscall_post_impl___posix_fadvise50(
+ long long res, long long fd, long long PAD, long long offset, long long len,
+ long long advice);
+void __sanitizer_syscall_pre_impl___select50(long long nd, long long in,
+ long long ou, long long ex,
+ long long tv);
+void __sanitizer_syscall_post_impl___select50(long long res, long long nd,
+ long long in, long long ou,
+ long long ex, long long tv);
+void __sanitizer_syscall_pre_impl___gettimeofday50(long long tp, long long tzp);
+void __sanitizer_syscall_post_impl___gettimeofday50(long long res, long long tp,
+ long long tzp);
+void __sanitizer_syscall_pre_impl___settimeofday50(long long tv, long long tzp);
+void __sanitizer_syscall_post_impl___settimeofday50(long long res, long long tv,
+ long long tzp);
+void __sanitizer_syscall_pre_impl___utimes50(long long path, long long tptr);
+void __sanitizer_syscall_post_impl___utimes50(long long res, long long path,
+ long long tptr);
+void __sanitizer_syscall_pre_impl___adjtime50(long long delta,
+ long long olddelta);
+void __sanitizer_syscall_post_impl___adjtime50(long long res, long long delta,
+ long long olddelta);
+void __sanitizer_syscall_pre_impl___lfs_segwait50(long long fsidp,
+ long long tv);
+void __sanitizer_syscall_post_impl___lfs_segwait50(long long res,
+ long long fsidp,
+ long long tv);
+void __sanitizer_syscall_pre_impl___futimes50(long long fd, long long tptr);
+void __sanitizer_syscall_post_impl___futimes50(long long res, long long fd,
+ long long tptr);
+void __sanitizer_syscall_pre_impl___lutimes50(long long path, long long tptr);
+void __sanitizer_syscall_post_impl___lutimes50(long long res, long long path,
+ long long tptr);
+void __sanitizer_syscall_pre_impl___setitimer50(long long which, long long itv,
+ long long oitv);
+void __sanitizer_syscall_post_impl___setitimer50(long long res, long long which,
+ long long itv, long long oitv);
+void __sanitizer_syscall_pre_impl___getitimer50(long long which, long long itv);
+void __sanitizer_syscall_post_impl___getitimer50(long long res, long long which,
+ long long itv);
+void __sanitizer_syscall_pre_impl___clock_gettime50(long long clock_id,
+ long long tp);
+void __sanitizer_syscall_post_impl___clock_gettime50(long long res,
+ long long clock_id,
+ long long tp);
+void __sanitizer_syscall_pre_impl___clock_settime50(long long clock_id,
+ long long tp);
+void __sanitizer_syscall_post_impl___clock_settime50(long long res,
+ long long clock_id,
+ long long tp);
+void __sanitizer_syscall_pre_impl___clock_getres50(long long clock_id,
+ long long tp);
+void __sanitizer_syscall_post_impl___clock_getres50(long long res,
+ long long clock_id,
+ long long tp);
+void __sanitizer_syscall_pre_impl___nanosleep50(long long rqtp, long long rmtp);
+void __sanitizer_syscall_post_impl___nanosleep50(long long res, long long rqtp,
+ long long rmtp);
+void __sanitizer_syscall_pre_impl_____sigtimedwait50(long long set,
+ long long info,
+ long long timeout);
+void __sanitizer_syscall_post_impl_____sigtimedwait50(long long res,
+ long long set,
+ long long info,
+ long long timeout);
+void __sanitizer_syscall_pre_impl___mq_timedsend50(long long mqdes,
+ long long msg_ptr,
+ long long msg_len,
+ long long msg_prio,
+ long long abs_timeout);
+void __sanitizer_syscall_post_impl___mq_timedsend50(
+ long long res, long long mqdes, long long msg_ptr, long long msg_len,
+ long long msg_prio, long long abs_timeout);
+void __sanitizer_syscall_pre_impl___mq_timedreceive50(long long mqdes,
+ long long msg_ptr,
+ long long msg_len,
+ long long msg_prio,
+ long long abs_timeout);
+void __sanitizer_syscall_post_impl___mq_timedreceive50(
+ long long res, long long mqdes, long long msg_ptr, long long msg_len,
+ long long msg_prio, long long abs_timeout);
+void __sanitizer_syscall_pre_impl_compat_60__lwp_park(long long ts,
+ long long unpark,
+ long long hint,
+ long long unparkhint);
+void __sanitizer_syscall_post_impl_compat_60__lwp_park(long long res,
+ long long ts,
+ long long unpark,
+ long long hint,
+ long long unparkhint);
+void __sanitizer_syscall_pre_impl___kevent50(long long fd, long long changelist,
+ long long nchanges,
+ long long eventlist,
+ long long nevents,
+ long long timeout);
+void __sanitizer_syscall_post_impl___kevent50(
+ long long res, long long fd, long long changelist, long long nchanges,
+ long long eventlist, long long nevents, long long timeout);
+void __sanitizer_syscall_pre_impl___pselect50(long long nd, long long in,
+ long long ou, long long ex,
+ long long ts, long long mask);
+void __sanitizer_syscall_post_impl___pselect50(long long res, long long nd,
+ long long in, long long ou,
+ long long ex, long long ts,
+ long long mask);
+void __sanitizer_syscall_pre_impl___pollts50(long long fds, long long nfds,
+ long long ts, long long mask);
+void __sanitizer_syscall_post_impl___pollts50(long long res, long long fds,
+ long long nfds, long long ts,
+ long long mask);
+void __sanitizer_syscall_pre_impl___aio_suspend50(long long list,
+ long long nent,
+ long long timeout);
+void __sanitizer_syscall_post_impl___aio_suspend50(long long res,
+ long long list,
+ long long nent,
+ long long timeout);
+void __sanitizer_syscall_pre_impl___stat50(long long path, long long ub);
+void __sanitizer_syscall_post_impl___stat50(long long res, long long path,
+ long long ub);
+void __sanitizer_syscall_pre_impl___fstat50(long long fd, long long sb);
+void __sanitizer_syscall_post_impl___fstat50(long long res, long long fd,
+ long long sb);
+void __sanitizer_syscall_pre_impl___lstat50(long long path, long long ub);
+void __sanitizer_syscall_post_impl___lstat50(long long res, long long path,
+ long long ub);
+void __sanitizer_syscall_pre_impl_____semctl50(long long semid,
+ long long semnum, long long cmd,
+ long long arg);
+void __sanitizer_syscall_post_impl_____semctl50(long long res, long long semid,
+ long long semnum, long long cmd,
+ long long arg);
+void __sanitizer_syscall_pre_impl___shmctl50(long long shmid, long long cmd,
+ long long buf);
+void __sanitizer_syscall_post_impl___shmctl50(long long res, long long shmid,
+ long long cmd, long long buf);
+void __sanitizer_syscall_pre_impl___msgctl50(long long msqid, long long cmd,
+ long long buf);
+void __sanitizer_syscall_post_impl___msgctl50(long long res, long long msqid,
+ long long cmd, long long buf);
+void __sanitizer_syscall_pre_impl___getrusage50(long long who,
+ long long rusage);
+void __sanitizer_syscall_post_impl___getrusage50(long long res, long long who,
+ long long rusage);
+void __sanitizer_syscall_pre_impl___timer_settime50(long long timerid,
+ long long flags,
+ long long value,
+ long long ovalue);
+void __sanitizer_syscall_post_impl___timer_settime50(long long res,
+ long long timerid,
+ long long flags,
+ long long value,
+ long long ovalue);
+void __sanitizer_syscall_pre_impl___timer_gettime50(long long timerid,
+ long long value);
+void __sanitizer_syscall_post_impl___timer_gettime50(long long res,
+ long long timerid,
+ long long value);
+#if defined(NTP) || !defined(_KERNEL_OPT)
+void __sanitizer_syscall_pre_impl___ntp_gettime50(long long ntvp);
+void __sanitizer_syscall_post_impl___ntp_gettime50(long long res,
+ long long ntvp);
+#else
+/* syscall 448 has been skipped */
+#endif
+void __sanitizer_syscall_pre_impl___wait450(long long pid, long long status,
+ long long options,
+ long long rusage);
+void __sanitizer_syscall_post_impl___wait450(long long res, long long pid,
+ long long status,
+ long long options,
+ long long rusage);
+void __sanitizer_syscall_pre_impl___mknod50(long long path, long long mode,
+ long long dev);
+void __sanitizer_syscall_post_impl___mknod50(long long res, long long path,
+ long long mode, long long dev);
+void __sanitizer_syscall_pre_impl___fhstat50(long long fhp, long long fh_size,
+ long long sb);
+void __sanitizer_syscall_post_impl___fhstat50(long long res, long long fhp,
+ long long fh_size, long long sb);
+/* syscall 452 has been skipped */
+void __sanitizer_syscall_pre_impl_pipe2(long long fildes, long long flags);
+void __sanitizer_syscall_post_impl_pipe2(long long res, long long fildes,
+ long long flags);
+void __sanitizer_syscall_pre_impl_dup3(long long from, long long to,
+ long long flags);
+void __sanitizer_syscall_post_impl_dup3(long long res, long long from,
+ long long to, long long flags);
+void __sanitizer_syscall_pre_impl_kqueue1(long long flags);
+void __sanitizer_syscall_post_impl_kqueue1(long long res, long long flags);
+void __sanitizer_syscall_pre_impl_paccept(long long s, long long name,
+ long long anamelen, long long mask,
+ long long flags);
+void __sanitizer_syscall_post_impl_paccept(long long res, long long s,
+ long long name, long long anamelen,
+ long long mask, long long flags);
+void __sanitizer_syscall_pre_impl_linkat(long long fd1, long long name1,
+ long long fd2, long long name2,
+ long long flags);
+void __sanitizer_syscall_post_impl_linkat(long long res, long long fd1,
+ long long name1, long long fd2,
+ long long name2, long long flags);
+void __sanitizer_syscall_pre_impl_renameat(long long fromfd, long long from,
+ long long tofd, long long to);
+void __sanitizer_syscall_post_impl_renameat(long long res, long long fromfd,
+ long long from, long long tofd,
+ long long to);
+void __sanitizer_syscall_pre_impl_mkfifoat(long long fd, long long path,
+ long long mode);
+void __sanitizer_syscall_post_impl_mkfifoat(long long res, long long fd,
+ long long path, long long mode);
+void __sanitizer_syscall_pre_impl_mknodat(long long fd, long long path,
+ long long mode, long long PAD,
+ long long dev);
+void __sanitizer_syscall_post_impl_mknodat(long long res, long long fd,
+ long long path, long long mode,
+ long long PAD, long long dev);
+void __sanitizer_syscall_pre_impl_mkdirat(long long fd, long long path,
+ long long mode);
+void __sanitizer_syscall_post_impl_mkdirat(long long res, long long fd,
+ long long path, long long mode);
+void __sanitizer_syscall_pre_impl_faccessat(long long fd, long long path,
+ long long amode, long long flag);
+void __sanitizer_syscall_post_impl_faccessat(long long res, long long fd,
+ long long path, long long amode,
+ long long flag);
+void __sanitizer_syscall_pre_impl_fchmodat(long long fd, long long path,
+ long long mode, long long flag);
+void __sanitizer_syscall_post_impl_fchmodat(long long res, long long fd,
+ long long path, long long mode,
+ long long flag);
+void __sanitizer_syscall_pre_impl_fchownat(long long fd, long long path,
+ long long owner, long long group,
+ long long flag);
+void __sanitizer_syscall_post_impl_fchownat(long long res, long long fd,
+ long long path, long long owner,
+ long long group, long long flag);
+void __sanitizer_syscall_pre_impl_fexecve(long long fd, long long argp,
+ long long envp);
+void __sanitizer_syscall_post_impl_fexecve(long long res, long long fd,
+ long long argp, long long envp);
+void __sanitizer_syscall_pre_impl_fstatat(long long fd, long long path,
+ long long buf, long long flag);
+void __sanitizer_syscall_post_impl_fstatat(long long res, long long fd,
+ long long path, long long buf,
+ long long flag);
+void __sanitizer_syscall_pre_impl_utimensat(long long fd, long long path,
+ long long tptr, long long flag);
+void __sanitizer_syscall_post_impl_utimensat(long long res, long long fd,
+ long long path, long long tptr,
+ long long flag);
+void __sanitizer_syscall_pre_impl_openat(long long fd, long long path,
+ long long oflags, long long mode);
+void __sanitizer_syscall_post_impl_openat(long long res, long long fd,
+ long long path, long long oflags,
+ long long mode);
+void __sanitizer_syscall_pre_impl_readlinkat(long long fd, long long path,
+ long long buf, long long bufsize);
+void __sanitizer_syscall_post_impl_readlinkat(long long res, long long fd,
+ long long path, long long buf,
+ long long bufsize);
+void __sanitizer_syscall_pre_impl_symlinkat(long long path1, long long fd,
+ long long path2);
+void __sanitizer_syscall_post_impl_symlinkat(long long res, long long path1,
+ long long fd, long long path2);
+void __sanitizer_syscall_pre_impl_unlinkat(long long fd, long long path,
+ long long flag);
+void __sanitizer_syscall_post_impl_unlinkat(long long res, long long fd,
+ long long path, long long flag);
+void __sanitizer_syscall_pre_impl_futimens(long long fd, long long tptr);
+void __sanitizer_syscall_post_impl_futimens(long long res, long long fd,
+ long long tptr);
+void __sanitizer_syscall_pre_impl___quotactl(long long path, long long args);
+void __sanitizer_syscall_post_impl___quotactl(long long res, long long path,
+ long long args);
+void __sanitizer_syscall_pre_impl_posix_spawn(long long pid, long long path,
+ long long file_actions,
+ long long attrp, long long argv,
+ long long envp);
+void __sanitizer_syscall_post_impl_posix_spawn(long long res, long long pid,
+ long long path,
+ long long file_actions,
+ long long attrp, long long argv,
+ long long envp);
+void __sanitizer_syscall_pre_impl_recvmmsg(long long s, long long mmsg,
+ long long vlen, long long flags,
+ long long timeout);
+void __sanitizer_syscall_post_impl_recvmmsg(long long res, long long s,
+ long long mmsg, long long vlen,
+ long long flags, long long timeout);
+void __sanitizer_syscall_pre_impl_sendmmsg(long long s, long long mmsg,
+ long long vlen, long long flags);
+void __sanitizer_syscall_post_impl_sendmmsg(long long res, long long s,
+ long long mmsg, long long vlen,
+ long long flags);
+void __sanitizer_syscall_pre_impl_clock_nanosleep(long long clock_id,
+ long long flags,
+ long long rqtp,
+ long long rmtp);
+void __sanitizer_syscall_post_impl_clock_nanosleep(long long res,
+ long long clock_id,
+ long long flags,
+ long long rqtp,
+ long long rmtp);
+void __sanitizer_syscall_pre_impl____lwp_park60(long long clock_id,
+ long long flags, long long ts,
+ long long unpark,
+ long long hint,
+ long long unparkhint);
+void __sanitizer_syscall_post_impl____lwp_park60(
+ long long res, long long clock_id, long long flags, long long ts,
+ long long unpark, long long hint, long long unparkhint);
+void __sanitizer_syscall_pre_impl_posix_fallocate(long long fd, long long PAD,
+ long long pos, long long len);
+void __sanitizer_syscall_post_impl_posix_fallocate(long long res, long long fd,
+ long long PAD, long long pos,
+ long long len);
+void __sanitizer_syscall_pre_impl_fdiscard(long long fd, long long PAD,
+ long long pos, long long len);
+void __sanitizer_syscall_post_impl_fdiscard(long long res, long long fd,
+ long long PAD, long long pos,
+ long long len);
+void __sanitizer_syscall_pre_impl_wait6(long long idtype, long long id,
+ long long status, long long options,
+ long long wru, long long info);
+void __sanitizer_syscall_post_impl_wait6(long long res, long long idtype,
+ long long id, long long status,
+ long long options, long long wru,
+ long long info);
+void __sanitizer_syscall_pre_impl_clock_getcpuclockid2(long long idtype,
+ long long id,
+ long long clock_id);
+void __sanitizer_syscall_post_impl_clock_getcpuclockid2(long long res,
+ long long idtype,
+ long long id,
+ long long clock_id);
+void __sanitizer_syscall_pre_impl___getvfsstat90(long long buf,
+ long long bufsize,
+ long long flags);
+void __sanitizer_syscall_post_impl___getvfsstat90(long long res, long long buf,
+ long long bufsize,
+ long long flags);
+void __sanitizer_syscall_pre_impl___statvfs190(long long path, long long buf,
+ long long flags);
+void __sanitizer_syscall_post_impl___statvfs190(long long res, long long path,
+ long long buf, long long flags);
+void __sanitizer_syscall_pre_impl___fstatvfs190(long long fd, long long buf,
+ long long flags);
+void __sanitizer_syscall_post_impl___fstatvfs190(long long res, long long fd,
+ long long buf,
+ long long flags);
+void __sanitizer_syscall_pre_impl___fhstatvfs190(long long fhp,
+ long long fh_size,
+ long long buf,
+ long long flags);
+void __sanitizer_syscall_post_impl___fhstatvfs190(long long res, long long fhp,
+ long long fh_size,
+ long long buf,
+ long long flags);
+
+#ifdef __cplusplus
+} // extern "C"
+#endif
+
+// DO NOT EDIT! THIS FILE HAS BEEN GENERATED!
+
+#endif // SANITIZER_NETBSD_SYSCALL_HOOKS_H
--- /dev/null
+//===-- sanitizer/scudo_interface.h -----------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+/// Public Scudo interface header.
+//
+//===----------------------------------------------------------------------===//
+#ifndef SANITIZER_SCUDO_INTERFACE_H_
+#define SANITIZER_SCUDO_INTERFACE_H_
+
+#include <sanitizer/common_interface_defs.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+ // This function may be optionally provided by a user and should return
+ // a string containing Scudo runtime options. See scudo_flags.h for details.
+ const char* __scudo_default_options(void);
+
+ // This function allows to set the RSS limit at runtime. This can be either
+ // the hard limit (HardLimit=1) or the soft limit (HardLimit=0). The limit
+ // can be removed by setting LimitMb to 0. This function's parameters should
+ // be fully trusted to avoid security mishaps.
+ void __scudo_set_rss_limit(size_t LimitMb, int HardLimit);
+
+ // This function outputs various allocator statistics for both the Primary
+ // and Secondary allocators, including memory usage, number of allocations
+ // and deallocations.
+ void __scudo_print_stats(void);
+#ifdef __cplusplus
+} // extern "C"
+#endif
+
+#endif // SANITIZER_SCUDO_INTERFACE_H_
--- /dev/null
+//===-- tsan_interface.h ----------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file is a part of ThreadSanitizer (TSan), a race detector.
+//
+// Public interface header for TSan.
+//===----------------------------------------------------------------------===//
+#ifndef SANITIZER_TSAN_INTERFACE_H
+#define SANITIZER_TSAN_INTERFACE_H
+
+#include <sanitizer/common_interface_defs.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+// __tsan_release establishes a happens-before relation with a preceding
+// __tsan_acquire on the same address.
+void __tsan_acquire(void *addr);
+void __tsan_release(void *addr);
+
+// Annotations for custom mutexes.
+// The annotations allow to get better reports (with sets of locked mutexes),
+// detect more types of bugs (e.g. mutex misuses, races between lock/unlock and
+// destruction and potential deadlocks) and improve precision and performance
+// (by ignoring individual atomic operations in mutex code). However, the
+// downside is that annotated mutex code itself is not checked for correctness.
+
+// Mutex creation flags are passed to __tsan_mutex_create annotation.
+// If mutex has no constructor and __tsan_mutex_create is not called,
+// the flags may be passed to __tsan_mutex_pre_lock/__tsan_mutex_post_lock
+// annotations.
+
+// Mutex has static storage duration and no-op constructor and destructor.
+// This effectively makes tsan ignore destroy annotation.
+static const unsigned __tsan_mutex_linker_init = 1 << 0;
+// Mutex is write reentrant.
+static const unsigned __tsan_mutex_write_reentrant = 1 << 1;
+// Mutex is read reentrant.
+static const unsigned __tsan_mutex_read_reentrant = 1 << 2;
+// Mutex does not have static storage duration, and must not be used after
+// its destructor runs. The opposite of __tsan_mutex_linker_init.
+// If this flag is passed to __tsan_mutex_destroy, then the destruction
+// is ignored unless this flag was previously set on the mutex.
+static const unsigned __tsan_mutex_not_static = 1 << 8;
+
+// Mutex operation flags:
+
+// Denotes read lock operation.
+static const unsigned __tsan_mutex_read_lock = 1 << 3;
+// Denotes try lock operation.
+static const unsigned __tsan_mutex_try_lock = 1 << 4;
+// Denotes that a try lock operation has failed to acquire the mutex.
+static const unsigned __tsan_mutex_try_lock_failed = 1 << 5;
+// Denotes that the lock operation acquires multiple recursion levels.
+// Number of levels is passed in recursion parameter.
+// This is useful for annotation of e.g. Java builtin monitors,
+// for which wait operation releases all recursive acquisitions of the mutex.
+static const unsigned __tsan_mutex_recursive_lock = 1 << 6;
+// Denotes that the unlock operation releases all recursion levels.
+// Number of released levels is returned and later must be passed to
+// the corresponding __tsan_mutex_post_lock annotation.
+static const unsigned __tsan_mutex_recursive_unlock = 1 << 7;
+
+// Annotate creation of a mutex.
+// Supported flags: mutex creation flags.
+void __tsan_mutex_create(void *addr, unsigned flags);
+
+// Annotate destruction of a mutex.
+// Supported flags:
+// - __tsan_mutex_linker_init
+// - __tsan_mutex_not_static
+void __tsan_mutex_destroy(void *addr, unsigned flags);
+
+// Annotate start of lock operation.
+// Supported flags:
+// - __tsan_mutex_read_lock
+// - __tsan_mutex_try_lock
+// - all mutex creation flags
+void __tsan_mutex_pre_lock(void *addr, unsigned flags);
+
+// Annotate end of lock operation.
+// Supported flags:
+// - __tsan_mutex_read_lock (must match __tsan_mutex_pre_lock)
+// - __tsan_mutex_try_lock (must match __tsan_mutex_pre_lock)
+// - __tsan_mutex_try_lock_failed
+// - __tsan_mutex_recursive_lock
+// - all mutex creation flags
+void __tsan_mutex_post_lock(void *addr, unsigned flags, int recursion);
+
+// Annotate start of unlock operation.
+// Supported flags:
+// - __tsan_mutex_read_lock
+// - __tsan_mutex_recursive_unlock
+int __tsan_mutex_pre_unlock(void *addr, unsigned flags);
+
+// Annotate end of unlock operation.
+// Supported flags:
+// - __tsan_mutex_read_lock (must match __tsan_mutex_pre_unlock)
+void __tsan_mutex_post_unlock(void *addr, unsigned flags);
+
+// Annotate start/end of notify/signal/broadcast operation.
+// Supported flags: none.
+void __tsan_mutex_pre_signal(void *addr, unsigned flags);
+void __tsan_mutex_post_signal(void *addr, unsigned flags);
+
+// Annotate start/end of a region of code where lock/unlock/signal operation
+// diverts to do something else unrelated to the mutex. This can be used to
+// annotate, for example, calls into cooperative scheduler or contention
+// profiling code.
+// These annotations must be called only from within
+// __tsan_mutex_pre/post_lock, __tsan_mutex_pre/post_unlock,
+// __tsan_mutex_pre/post_signal regions.
+// Supported flags: none.
+void __tsan_mutex_pre_divert(void *addr, unsigned flags);
+void __tsan_mutex_post_divert(void *addr, unsigned flags);
+
+// External race detection API.
+// Can be used by non-instrumented libraries to detect when their objects are
+// being used in an unsafe manner.
+// - __tsan_external_read/__tsan_external_write annotates the logical reads
+// and writes of the object at the specified address. 'caller_pc' should
+// be the PC of the library user, which the library can obtain with e.g.
+// `__builtin_return_address(0)`.
+// - __tsan_external_register_tag registers a 'tag' with the specified name,
+// which is later used in read/write annotations to denote the object type
+// - __tsan_external_assign_tag can optionally mark a heap object with a tag
+void *__tsan_external_register_tag(const char *object_type);
+void __tsan_external_register_header(void *tag, const char *header);
+void __tsan_external_assign_tag(void *addr, void *tag);
+void __tsan_external_read(void *addr, void *caller_pc, void *tag);
+void __tsan_external_write(void *addr, void *caller_pc, void *tag);
+
+// Fiber switching API.
+// - TSAN context for fiber can be created by __tsan_create_fiber
+// and freed by __tsan_destroy_fiber.
+// - TSAN context of current fiber or thread can be obtained
+// by calling __tsan_get_current_fiber.
+// - __tsan_switch_to_fiber should be called immediatly before switch
+// to fiber, such as call of swapcontext.
+// - Fiber name can be set by __tsan_set_fiber_name.
+void *__tsan_get_current_fiber(void);
+void *__tsan_create_fiber(unsigned flags);
+void __tsan_destroy_fiber(void *fiber);
+void __tsan_switch_to_fiber(void *fiber, unsigned flags);
+void __tsan_set_fiber_name(void *fiber, const char *name);
+
+// Flags for __tsan_switch_to_fiber:
+// Do not establish a happens-before relation between fibers
+static const unsigned __tsan_switch_to_fiber_no_sync = 1 << 0;
+
+#ifdef __cplusplus
+} // extern "C"
+#endif
+
+#endif // SANITIZER_TSAN_INTERFACE_H
--- /dev/null
+//===-- tsan_interface_atomic.h ---------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file is a part of ThreadSanitizer (TSan), a race detector.
+//
+// Public interface header for TSan atomics.
+//===----------------------------------------------------------------------===//
+#ifndef TSAN_INTERFACE_ATOMIC_H
+#define TSAN_INTERFACE_ATOMIC_H
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+typedef char __tsan_atomic8;
+typedef short __tsan_atomic16;
+typedef int __tsan_atomic32;
+typedef long __tsan_atomic64;
+#if defined(__SIZEOF_INT128__) \
+ || (__clang_major__ * 100 + __clang_minor__ >= 302)
+__extension__ typedef __int128 __tsan_atomic128;
+# define __TSAN_HAS_INT128 1
+#else
+# define __TSAN_HAS_INT128 0
+#endif
+
+// Part of ABI, do not change.
+// https://github.com/llvm/llvm-project/blob/master/libcxx/include/atomic
+typedef enum {
+ __tsan_memory_order_relaxed,
+ __tsan_memory_order_consume,
+ __tsan_memory_order_acquire,
+ __tsan_memory_order_release,
+ __tsan_memory_order_acq_rel,
+ __tsan_memory_order_seq_cst
+} __tsan_memory_order;
+
+__tsan_atomic8 __tsan_atomic8_load(const volatile __tsan_atomic8 *a,
+ __tsan_memory_order mo);
+__tsan_atomic16 __tsan_atomic16_load(const volatile __tsan_atomic16 *a,
+ __tsan_memory_order mo);
+__tsan_atomic32 __tsan_atomic32_load(const volatile __tsan_atomic32 *a,
+ __tsan_memory_order mo);
+__tsan_atomic64 __tsan_atomic64_load(const volatile __tsan_atomic64 *a,
+ __tsan_memory_order mo);
+#if __TSAN_HAS_INT128
+__tsan_atomic128 __tsan_atomic128_load(const volatile __tsan_atomic128 *a,
+ __tsan_memory_order mo);
+#endif
+
+void __tsan_atomic8_store(volatile __tsan_atomic8 *a, __tsan_atomic8 v,
+ __tsan_memory_order mo);
+void __tsan_atomic16_store(volatile __tsan_atomic16 *a, __tsan_atomic16 v,
+ __tsan_memory_order mo);
+void __tsan_atomic32_store(volatile __tsan_atomic32 *a, __tsan_atomic32 v,
+ __tsan_memory_order mo);
+void __tsan_atomic64_store(volatile __tsan_atomic64 *a, __tsan_atomic64 v,
+ __tsan_memory_order mo);
+#if __TSAN_HAS_INT128
+void __tsan_atomic128_store(volatile __tsan_atomic128 *a, __tsan_atomic128 v,
+ __tsan_memory_order mo);
+#endif
+
+__tsan_atomic8 __tsan_atomic8_exchange(volatile __tsan_atomic8 *a,
+ __tsan_atomic8 v, __tsan_memory_order mo);
+__tsan_atomic16 __tsan_atomic16_exchange(volatile __tsan_atomic16 *a,
+ __tsan_atomic16 v, __tsan_memory_order mo);
+__tsan_atomic32 __tsan_atomic32_exchange(volatile __tsan_atomic32 *a,
+ __tsan_atomic32 v, __tsan_memory_order mo);
+__tsan_atomic64 __tsan_atomic64_exchange(volatile __tsan_atomic64 *a,
+ __tsan_atomic64 v, __tsan_memory_order mo);
+#if __TSAN_HAS_INT128
+__tsan_atomic128 __tsan_atomic128_exchange(volatile __tsan_atomic128 *a,
+ __tsan_atomic128 v, __tsan_memory_order mo);
+#endif
+
+__tsan_atomic8 __tsan_atomic8_fetch_add(volatile __tsan_atomic8 *a,
+ __tsan_atomic8 v, __tsan_memory_order mo);
+__tsan_atomic16 __tsan_atomic16_fetch_add(volatile __tsan_atomic16 *a,
+ __tsan_atomic16 v, __tsan_memory_order mo);
+__tsan_atomic32 __tsan_atomic32_fetch_add(volatile __tsan_atomic32 *a,
+ __tsan_atomic32 v, __tsan_memory_order mo);
+__tsan_atomic64 __tsan_atomic64_fetch_add(volatile __tsan_atomic64 *a,
+ __tsan_atomic64 v, __tsan_memory_order mo);
+#if __TSAN_HAS_INT128
+__tsan_atomic128 __tsan_atomic128_fetch_add(volatile __tsan_atomic128 *a,
+ __tsan_atomic128 v, __tsan_memory_order mo);
+#endif
+
+__tsan_atomic8 __tsan_atomic8_fetch_sub(volatile __tsan_atomic8 *a,
+ __tsan_atomic8 v, __tsan_memory_order mo);
+__tsan_atomic16 __tsan_atomic16_fetch_sub(volatile __tsan_atomic16 *a,
+ __tsan_atomic16 v, __tsan_memory_order mo);
+__tsan_atomic32 __tsan_atomic32_fetch_sub(volatile __tsan_atomic32 *a,
+ __tsan_atomic32 v, __tsan_memory_order mo);
+__tsan_atomic64 __tsan_atomic64_fetch_sub(volatile __tsan_atomic64 *a,
+ __tsan_atomic64 v, __tsan_memory_order mo);
+#if __TSAN_HAS_INT128
+__tsan_atomic128 __tsan_atomic128_fetch_sub(volatile __tsan_atomic128 *a,
+ __tsan_atomic128 v, __tsan_memory_order mo);
+#endif
+
+__tsan_atomic8 __tsan_atomic8_fetch_and(volatile __tsan_atomic8 *a,
+ __tsan_atomic8 v, __tsan_memory_order mo);
+__tsan_atomic16 __tsan_atomic16_fetch_and(volatile __tsan_atomic16 *a,
+ __tsan_atomic16 v, __tsan_memory_order mo);
+__tsan_atomic32 __tsan_atomic32_fetch_and(volatile __tsan_atomic32 *a,
+ __tsan_atomic32 v, __tsan_memory_order mo);
+__tsan_atomic64 __tsan_atomic64_fetch_and(volatile __tsan_atomic64 *a,
+ __tsan_atomic64 v, __tsan_memory_order mo);
+#if __TSAN_HAS_INT128
+__tsan_atomic128 __tsan_atomic128_fetch_and(volatile __tsan_atomic128 *a,
+ __tsan_atomic128 v, __tsan_memory_order mo);
+#endif
+
+__tsan_atomic8 __tsan_atomic8_fetch_or(volatile __tsan_atomic8 *a,
+ __tsan_atomic8 v, __tsan_memory_order mo);
+__tsan_atomic16 __tsan_atomic16_fetch_or(volatile __tsan_atomic16 *a,
+ __tsan_atomic16 v, __tsan_memory_order mo);
+__tsan_atomic32 __tsan_atomic32_fetch_or(volatile __tsan_atomic32 *a,
+ __tsan_atomic32 v, __tsan_memory_order mo);
+__tsan_atomic64 __tsan_atomic64_fetch_or(volatile __tsan_atomic64 *a,
+ __tsan_atomic64 v, __tsan_memory_order mo);
+#if __TSAN_HAS_INT128
+__tsan_atomic128 __tsan_atomic128_fetch_or(volatile __tsan_atomic128 *a,
+ __tsan_atomic128 v, __tsan_memory_order mo);
+#endif
+
+__tsan_atomic8 __tsan_atomic8_fetch_xor(volatile __tsan_atomic8 *a,
+ __tsan_atomic8 v, __tsan_memory_order mo);
+__tsan_atomic16 __tsan_atomic16_fetch_xor(volatile __tsan_atomic16 *a,
+ __tsan_atomic16 v, __tsan_memory_order mo);
+__tsan_atomic32 __tsan_atomic32_fetch_xor(volatile __tsan_atomic32 *a,
+ __tsan_atomic32 v, __tsan_memory_order mo);
+__tsan_atomic64 __tsan_atomic64_fetch_xor(volatile __tsan_atomic64 *a,
+ __tsan_atomic64 v, __tsan_memory_order mo);
+#if __TSAN_HAS_INT128
+__tsan_atomic128 __tsan_atomic128_fetch_xor(volatile __tsan_atomic128 *a,
+ __tsan_atomic128 v, __tsan_memory_order mo);
+#endif
+
+__tsan_atomic8 __tsan_atomic8_fetch_nand(volatile __tsan_atomic8 *a,
+ __tsan_atomic8 v, __tsan_memory_order mo);
+__tsan_atomic16 __tsan_atomic16_fetch_nand(volatile __tsan_atomic16 *a,
+ __tsan_atomic16 v, __tsan_memory_order mo);
+__tsan_atomic32 __tsan_atomic32_fetch_nand(volatile __tsan_atomic32 *a,
+ __tsan_atomic32 v, __tsan_memory_order mo);
+__tsan_atomic64 __tsan_atomic64_fetch_nand(volatile __tsan_atomic64 *a,
+ __tsan_atomic64 v, __tsan_memory_order mo);
+#if __TSAN_HAS_INT128
+__tsan_atomic128 __tsan_atomic128_fetch_nand(volatile __tsan_atomic128 *a,
+ __tsan_atomic128 v, __tsan_memory_order mo);
+#endif
+
+int __tsan_atomic8_compare_exchange_weak(volatile __tsan_atomic8 *a,
+ __tsan_atomic8 *c, __tsan_atomic8 v, __tsan_memory_order mo,
+ __tsan_memory_order fail_mo);
+int __tsan_atomic16_compare_exchange_weak(volatile __tsan_atomic16 *a,
+ __tsan_atomic16 *c, __tsan_atomic16 v, __tsan_memory_order mo,
+ __tsan_memory_order fail_mo);
+int __tsan_atomic32_compare_exchange_weak(volatile __tsan_atomic32 *a,
+ __tsan_atomic32 *c, __tsan_atomic32 v, __tsan_memory_order mo,
+ __tsan_memory_order fail_mo);
+int __tsan_atomic64_compare_exchange_weak(volatile __tsan_atomic64 *a,
+ __tsan_atomic64 *c, __tsan_atomic64 v, __tsan_memory_order mo,
+ __tsan_memory_order fail_mo);
+#if __TSAN_HAS_INT128
+int __tsan_atomic128_compare_exchange_weak(volatile __tsan_atomic128 *a,
+ __tsan_atomic128 *c, __tsan_atomic128 v, __tsan_memory_order mo,
+ __tsan_memory_order fail_mo);
+#endif
+
+int __tsan_atomic8_compare_exchange_strong(volatile __tsan_atomic8 *a,
+ __tsan_atomic8 *c, __tsan_atomic8 v, __tsan_memory_order mo,
+ __tsan_memory_order fail_mo);
+int __tsan_atomic16_compare_exchange_strong(volatile __tsan_atomic16 *a,
+ __tsan_atomic16 *c, __tsan_atomic16 v, __tsan_memory_order mo,
+ __tsan_memory_order fail_mo);
+int __tsan_atomic32_compare_exchange_strong(volatile __tsan_atomic32 *a,
+ __tsan_atomic32 *c, __tsan_atomic32 v, __tsan_memory_order mo,
+ __tsan_memory_order fail_mo);
+int __tsan_atomic64_compare_exchange_strong(volatile __tsan_atomic64 *a,
+ __tsan_atomic64 *c, __tsan_atomic64 v, __tsan_memory_order mo,
+ __tsan_memory_order fail_mo);
+#if __TSAN_HAS_INT128
+int __tsan_atomic128_compare_exchange_strong(volatile __tsan_atomic128 *a,
+ __tsan_atomic128 *c, __tsan_atomic128 v, __tsan_memory_order mo,
+ __tsan_memory_order fail_mo);
+#endif
+
+__tsan_atomic8 __tsan_atomic8_compare_exchange_val(
+ volatile __tsan_atomic8 *a, __tsan_atomic8 c, __tsan_atomic8 v,
+ __tsan_memory_order mo, __tsan_memory_order fail_mo);
+__tsan_atomic16 __tsan_atomic16_compare_exchange_val(
+ volatile __tsan_atomic16 *a, __tsan_atomic16 c, __tsan_atomic16 v,
+ __tsan_memory_order mo, __tsan_memory_order fail_mo);
+__tsan_atomic32 __tsan_atomic32_compare_exchange_val(
+ volatile __tsan_atomic32 *a, __tsan_atomic32 c, __tsan_atomic32 v,
+ __tsan_memory_order mo, __tsan_memory_order fail_mo);
+__tsan_atomic64 __tsan_atomic64_compare_exchange_val(
+ volatile __tsan_atomic64 *a, __tsan_atomic64 c, __tsan_atomic64 v,
+ __tsan_memory_order mo, __tsan_memory_order fail_mo);
+#if __TSAN_HAS_INT128
+__tsan_atomic128 __tsan_atomic128_compare_exchange_val(
+ volatile __tsan_atomic128 *a, __tsan_atomic128 c, __tsan_atomic128 v,
+ __tsan_memory_order mo, __tsan_memory_order fail_mo);
+#endif
+
+void __tsan_atomic_thread_fence(__tsan_memory_order mo);
+void __tsan_atomic_signal_fence(__tsan_memory_order mo);
+
+#ifdef __cplusplus
+} // extern "C"
+#endif
+
+#endif // TSAN_INTERFACE_ATOMIC_H
--- /dev/null
+//===-- sanitizer/ubsan_interface.h -----------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file is a part of UBSanitizer (UBSan).
+//
+// Public interface header.
+//===----------------------------------------------------------------------===//
+#ifndef SANITIZER_UBSAN_INTERFACE_H
+#define SANITIZER_UBSAN_INTERFACE_H
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+/// User-provided default option settings.
+///
+/// You can provide your own implementation of this function to return a string
+/// containing UBSan runtime options (for example,
+/// <c>verbosity=1:halt_on_error=0</c>).
+///
+/// \returns Default options string.
+const char* __ubsan_default_options(void);
+
+#ifdef __cplusplus
+} // extern "C"
+#endif
+
+#endif // SANITIZER_UBSAN_INTERFACE_H
--- /dev/null
+/*===--------------- serializeintrin.h - serialize intrinsics --------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <serializeintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __SERIALIZEINTRIN_H
+#define __SERIALIZEINTRIN_H
+
+/// Serialize instruction fetch and execution.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the <c> SERIALIZE </c> instruction.
+///
+static __inline__ void
+__attribute__((__always_inline__, __nodebug__, __target__("serialize")))
+_serialize (void)
+{
+ __builtin_ia32_serialize ();
+}
+
+#endif /* __SERIALIZEINTRIN_H */
--- /dev/null
+/*===---- sgxintrin.h - X86 SGX intrinsics configuration -------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#if !defined __X86INTRIN_H && !defined __IMMINTRIN_H
+#error "Never use <sgxintrin.h> directly; include <x86intrin.h> instead."
+#endif
+
+#ifndef __SGXINTRIN_H
+#define __SGXINTRIN_H
+
+#if __has_extension(gnu_asm)
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("sgx")))
+
+static __inline unsigned int __DEFAULT_FN_ATTRS
+_enclu_u32(unsigned int __leaf, __SIZE_TYPE__ __d[])
+{
+ unsigned int __result;
+ __asm__ ("enclu"
+ : "=a" (__result), "=b" (__d[0]), "=c" (__d[1]), "=d" (__d[2])
+ : "a" (__leaf), "b" (__d[0]), "c" (__d[1]), "d" (__d[2])
+ : "cc");
+ return __result;
+}
+
+static __inline unsigned int __DEFAULT_FN_ATTRS
+_encls_u32(unsigned int __leaf, __SIZE_TYPE__ __d[])
+{
+ unsigned int __result;
+ __asm__ ("encls"
+ : "=a" (__result), "=b" (__d[0]), "=c" (__d[1]), "=d" (__d[2])
+ : "a" (__leaf), "b" (__d[0]), "c" (__d[1]), "d" (__d[2])
+ : "cc");
+ return __result;
+}
+
+static __inline unsigned int __DEFAULT_FN_ATTRS
+_enclv_u32(unsigned int __leaf, __SIZE_TYPE__ __d[])
+{
+ unsigned int __result;
+ __asm__ ("enclv"
+ : "=a" (__result), "=b" (__d[0]), "=c" (__d[1]), "=d" (__d[2])
+ : "a" (__leaf), "b" (__d[0]), "c" (__d[1]), "d" (__d[2])
+ : "cc");
+ return __result;
+}
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif /* __has_extension(gnu_asm) */
+
+#endif
/*===---- shaintrin.h - SHA intrinsics -------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#define __SHAINTRIN_H
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("sha")))
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("sha"), __min_vector_width__(128)))
-#define _mm_sha1rnds4_epu32(V1, V2, M) __extension__ ({ \
- __builtin_ia32_sha1rnds4((__v4si)(__m128i)(V1), (__v4si)(__m128i)(V2), (M)); })
+#define _mm_sha1rnds4_epu32(V1, V2, M) \
+ __builtin_ia32_sha1rnds4((__v4si)(__m128i)(V1), (__v4si)(__m128i)(V2), (M))
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_sha1nexte_epu32(__m128i __X, __m128i __Y)
/*===---- smmintrin.h - SSE4 intrinsics ------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
-#ifndef _SMMINTRIN_H
-#define _SMMINTRIN_H
+#ifndef __SMMINTRIN_H
+#define __SMMINTRIN_H
#include <tmmintrin.h>
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("sse4.1")))
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("sse4.1"), __min_vector_width__(128)))
/* SSE4 Rounding macros. */
#define _MM_FROUND_TO_NEAREST_INT 0x00
#define _MM_FROUND_RINT (_MM_FROUND_RAISE_EXC | _MM_FROUND_CUR_DIRECTION)
#define _MM_FROUND_NEARBYINT (_MM_FROUND_NO_EXC | _MM_FROUND_CUR_DIRECTION)
-/// \brief Rounds up each element of the 128-bit vector of [4 x float] to an
+/// Rounds up each element of the 128-bit vector of [4 x float] to an
/// integer and returns the rounded values in a 128-bit vector of
/// [4 x float].
///
/// \returns A 128-bit vector of [4 x float] containing the rounded values.
#define _mm_ceil_ps(X) _mm_round_ps((X), _MM_FROUND_CEIL)
-/// \brief Rounds up each element of the 128-bit vector of [2 x double] to an
+/// Rounds up each element of the 128-bit vector of [2 x double] to an
/// integer and returns the rounded values in a 128-bit vector of
/// [2 x double].
///
/// \returns A 128-bit vector of [2 x double] containing the rounded values.
#define _mm_ceil_pd(X) _mm_round_pd((X), _MM_FROUND_CEIL)
-/// \brief Copies three upper elements of the first 128-bit vector operand to
+/// Copies three upper elements of the first 128-bit vector operand to
/// the corresponding three upper elements of the 128-bit result vector of
/// [4 x float]. Rounds up the lowest element of the second 128-bit vector
/// operand to an integer and copies it to the lowest element of the 128-bit
/// values.
#define _mm_ceil_ss(X, Y) _mm_round_ss((X), (Y), _MM_FROUND_CEIL)
-/// \brief Copies the upper element of the first 128-bit vector operand to the
+/// Copies the upper element of the first 128-bit vector operand to the
/// corresponding upper element of the 128-bit result vector of [2 x double].
/// Rounds up the lower element of the second 128-bit vector operand to an
/// integer and copies it to the lower element of the 128-bit result vector
/// values.
#define _mm_ceil_sd(X, Y) _mm_round_sd((X), (Y), _MM_FROUND_CEIL)
-/// \brief Rounds down each element of the 128-bit vector of [4 x float] to an
+/// Rounds down each element of the 128-bit vector of [4 x float] to an
/// an integer and returns the rounded values in a 128-bit vector of
/// [4 x float].
///
/// \returns A 128-bit vector of [4 x float] containing the rounded values.
#define _mm_floor_ps(X) _mm_round_ps((X), _MM_FROUND_FLOOR)
-/// \brief Rounds down each element of the 128-bit vector of [2 x double] to an
+/// Rounds down each element of the 128-bit vector of [2 x double] to an
/// integer and returns the rounded values in a 128-bit vector of
/// [2 x double].
///
/// \returns A 128-bit vector of [2 x double] containing the rounded values.
#define _mm_floor_pd(X) _mm_round_pd((X), _MM_FROUND_FLOOR)
-/// \brief Copies three upper elements of the first 128-bit vector operand to
+/// Copies three upper elements of the first 128-bit vector operand to
/// the corresponding three upper elements of the 128-bit result vector of
/// [4 x float]. Rounds down the lowest element of the second 128-bit vector
/// operand to an integer and copies it to the lowest element of the 128-bit
/// values.
#define _mm_floor_ss(X, Y) _mm_round_ss((X), (Y), _MM_FROUND_FLOOR)
-/// \brief Copies the upper element of the first 128-bit vector operand to the
+/// Copies the upper element of the first 128-bit vector operand to the
/// corresponding upper element of the 128-bit result vector of [2 x double].
/// Rounds down the lower element of the second 128-bit vector operand to an
/// integer and copies it to the lower element of the 128-bit result vector
/// values.
#define _mm_floor_sd(X, Y) _mm_round_sd((X), (Y), _MM_FROUND_FLOOR)
-/// \brief Rounds each element of the 128-bit vector of [4 x float] to an
+/// Rounds each element of the 128-bit vector of [4 x float] to an
/// integer value according to the rounding control specified by the second
/// argument and returns the rounded values in a 128-bit vector of
/// [4 x float].
/// 10: Upward (toward positive infinity) \n
/// 11: Truncated
/// \returns A 128-bit vector of [4 x float] containing the rounded values.
-#define _mm_round_ps(X, M) __extension__ ({ \
- (__m128)__builtin_ia32_roundps((__v4sf)(__m128)(X), (M)); })
+#define _mm_round_ps(X, M) \
+ (__m128)__builtin_ia32_roundps((__v4sf)(__m128)(X), (M))
-/// \brief Copies three upper elements of the first 128-bit vector operand to
+/// Copies three upper elements of the first 128-bit vector operand to
/// the corresponding three upper elements of the 128-bit result vector of
/// [4 x float]. Rounds the lowest element of the second 128-bit vector
/// operand to an integer value according to the rounding control specified
/// 11: Truncated
/// \returns A 128-bit vector of [4 x float] containing the copied and rounded
/// values.
-#define _mm_round_ss(X, Y, M) __extension__ ({ \
+#define _mm_round_ss(X, Y, M) \
(__m128)__builtin_ia32_roundss((__v4sf)(__m128)(X), \
- (__v4sf)(__m128)(Y), (M)); })
+ (__v4sf)(__m128)(Y), (M))
-/// \brief Rounds each element of the 128-bit vector of [2 x double] to an
+/// Rounds each element of the 128-bit vector of [2 x double] to an
/// integer value according to the rounding control specified by the second
/// argument and returns the rounded values in a 128-bit vector of
/// [2 x double].
/// 10: Upward (toward positive infinity) \n
/// 11: Truncated
/// \returns A 128-bit vector of [2 x double] containing the rounded values.
-#define _mm_round_pd(X, M) __extension__ ({ \
- (__m128d)__builtin_ia32_roundpd((__v2df)(__m128d)(X), (M)); })
+#define _mm_round_pd(X, M) \
+ (__m128d)__builtin_ia32_roundpd((__v2df)(__m128d)(X), (M))
-/// \brief Copies the upper element of the first 128-bit vector operand to the
+/// Copies the upper element of the first 128-bit vector operand to the
/// corresponding upper element of the 128-bit result vector of [2 x double].
/// Rounds the lower element of the second 128-bit vector operand to an
/// integer value according to the rounding control specified by the third
/// 11: Truncated
/// \returns A 128-bit vector of [2 x double] containing the copied and rounded
/// values.
-#define _mm_round_sd(X, Y, M) __extension__ ({ \
+#define _mm_round_sd(X, Y, M) \
(__m128d)__builtin_ia32_roundsd((__v2df)(__m128d)(X), \
- (__v2df)(__m128d)(Y), (M)); })
+ (__v2df)(__m128d)(Y), (M))
/* SSE4 Packed Blending Intrinsics. */
-/// \brief Returns a 128-bit vector of [2 x double] where the values are
+/// Returns a 128-bit vector of [2 x double] where the values are
/// selected from either the first or second operand as specified by the
/// third operand, the control mask.
///
/// When a mask bit is 1, the corresponding 64-bit element in operand \a V2
/// is copied to the same position in the result.
/// \returns A 128-bit vector of [2 x double] containing the copied values.
-#define _mm_blend_pd(V1, V2, M) __extension__ ({ \
- (__m128d)__builtin_shufflevector((__v2df)(__m128d)(V1), \
- (__v2df)(__m128d)(V2), \
- (((M) & 0x01) ? 2 : 0), \
- (((M) & 0x02) ? 3 : 1)); })
+#define _mm_blend_pd(V1, V2, M) \
+ (__m128d) __builtin_ia32_blendpd ((__v2df)(__m128d)(V1), \
+ (__v2df)(__m128d)(V2), (int)(M))
-/// \brief Returns a 128-bit vector of [4 x float] where the values are selected
+/// Returns a 128-bit vector of [4 x float] where the values are selected
/// from either the first or second operand as specified by the third
/// operand, the control mask.
///
/// When a mask bit is 1, the corresponding 32-bit element in operand \a V2
/// is copied to the same position in the result.
/// \returns A 128-bit vector of [4 x float] containing the copied values.
-#define _mm_blend_ps(V1, V2, M) __extension__ ({ \
- (__m128)__builtin_shufflevector((__v4sf)(__m128)(V1), (__v4sf)(__m128)(V2), \
- (((M) & 0x01) ? 4 : 0), \
- (((M) & 0x02) ? 5 : 1), \
- (((M) & 0x04) ? 6 : 2), \
- (((M) & 0x08) ? 7 : 3)); })
-
-/// \brief Returns a 128-bit vector of [2 x double] where the values are
+#define _mm_blend_ps(V1, V2, M) \
+ (__m128) __builtin_ia32_blendps ((__v4sf)(__m128)(V1), \
+ (__v4sf)(__m128)(V2), (int)(M))
+
+/// Returns a 128-bit vector of [2 x double] where the values are
/// selected from either the first or second operand as specified by the
/// third operand, the control mask.
///
(__v2df)__M);
}
-/// \brief Returns a 128-bit vector of [4 x float] where the values are
+/// Returns a 128-bit vector of [4 x float] where the values are
/// selected from either the first or second operand as specified by the
/// third operand, the control mask.
///
(__v4sf)__M);
}
-/// \brief Returns a 128-bit vector of [16 x i8] where the values are selected
+/// Returns a 128-bit vector of [16 x i8] where the values are selected
/// from either of the first or second operand as specified by the third
/// operand, the control mask.
///
/// \param __V2
/// A 128-bit vector of [16 x i8].
/// \param __M
-/// A 128-bit vector operand, with mask bits 127, 119, 111 ... 7 specifying
+/// A 128-bit vector operand, with mask bits 127, 119, 111...7 specifying
/// how the values are to be copied. The position of the mask bit corresponds
/// to the most significant bit of a copied value. When a mask bit is 0, the
/// corresponding 8-bit element in operand \a __V1 is copied to the same
(__v16qi)__M);
}
-/// \brief Returns a 128-bit vector of [8 x i16] where the values are selected
+/// Returns a 128-bit vector of [8 x i16] where the values are selected
/// from either of the first or second operand as specified by the third
/// operand, the control mask.
///
/// When a mask bit is 1, the corresponding 16-bit element in operand \a V2
/// is copied to the same position in the result.
/// \returns A 128-bit vector of [8 x i16] containing the copied values.
-#define _mm_blend_epi16(V1, V2, M) __extension__ ({ \
- (__m128i)__builtin_shufflevector((__v8hi)(__m128i)(V1), \
- (__v8hi)(__m128i)(V2), \
- (((M) & 0x01) ? 8 : 0), \
- (((M) & 0x02) ? 9 : 1), \
- (((M) & 0x04) ? 10 : 2), \
- (((M) & 0x08) ? 11 : 3), \
- (((M) & 0x10) ? 12 : 4), \
- (((M) & 0x20) ? 13 : 5), \
- (((M) & 0x40) ? 14 : 6), \
- (((M) & 0x80) ? 15 : 7)); })
+#define _mm_blend_epi16(V1, V2, M) \
+ (__m128i) __builtin_ia32_pblendw128 ((__v8hi)(__m128i)(V1), \
+ (__v8hi)(__m128i)(V2), (int)(M))
/* SSE4 Dword Multiply Instructions. */
-/// \brief Multiples corresponding elements of two 128-bit vectors of [4 x i32]
+/// Multiples corresponding elements of two 128-bit vectors of [4 x i32]
/// and returns the lower 32 bits of the each product in a 128-bit vector of
/// [4 x i32].
///
return (__m128i) ((__v4su)__V1 * (__v4su)__V2);
}
-/// \brief Multiplies corresponding even-indexed elements of two 128-bit
+/// Multiplies corresponding even-indexed elements of two 128-bit
/// vectors of [4 x i32] and returns a 128-bit vector of [2 x i64]
/// containing the products.
///
}
/* SSE4 Floating Point Dot Product Instructions. */
-/// \brief Computes the dot product of the two 128-bit vectors of [4 x float]
+/// Computes the dot product of the two 128-bit vectors of [4 x float]
/// and returns it in the elements of the 128-bit result vector of
/// [4 x float].
///
/// each [4 x float] subvector. If a bit is set, the dot product is returned
/// in the corresponding element; otherwise that element is set to zero.
/// \returns A 128-bit vector of [4 x float] containing the dot product.
-#define _mm_dp_ps(X, Y, M) __extension__ ({ \
+#define _mm_dp_ps(X, Y, M) \
(__m128) __builtin_ia32_dpps((__v4sf)(__m128)(X), \
- (__v4sf)(__m128)(Y), (M)); })
+ (__v4sf)(__m128)(Y), (M))
-/// \brief Computes the dot product of the two 128-bit vectors of [2 x double]
+/// Computes the dot product of the two 128-bit vectors of [2 x double]
/// and returns it in the elements of the 128-bit result vector of
/// [2 x double].
///
/// input vectors are used as an input for dot product; otherwise that input
/// is treated as zero. Bits [1:0] determine which elements of the result
/// will receive a copy of the final dot product, with bit [0] corresponding
-/// to the lowest element and bit [3] corresponding to the highest element of
+/// to the lowest element and bit [1] corresponding to the highest element of
/// each [2 x double] vector. If a bit is set, the dot product is returned in
/// the corresponding element; otherwise that element is set to zero.
-#define _mm_dp_pd(X, Y, M) __extension__ ({\
+#define _mm_dp_pd(X, Y, M) \
(__m128d) __builtin_ia32_dppd((__v2df)(__m128d)(X), \
- (__v2df)(__m128d)(Y), (M)); })
+ (__v2df)(__m128d)(Y), (M))
/* SSE4 Streaming Load Hint Instruction. */
-/// \brief Loads integer values from a 128-bit aligned memory location to a
+/// Loads integer values from a 128-bit aligned memory location to a
/// 128-bit integer vector.
///
/// \headerfile <x86intrin.h>
}
/* SSE4 Packed Integer Min/Max Instructions. */
-/// \brief Compares the corresponding elements of two 128-bit vectors of
+/// Compares the corresponding elements of two 128-bit vectors of
/// [16 x i8] and returns a 128-bit vector of [16 x i8] containing the lesser
/// of the two values.
///
return (__m128i) __builtin_ia32_pminsb128 ((__v16qi) __V1, (__v16qi) __V2);
}
-/// \brief Compares the corresponding elements of two 128-bit vectors of
+/// Compares the corresponding elements of two 128-bit vectors of
/// [16 x i8] and returns a 128-bit vector of [16 x i8] containing the
/// greater value of the two.
///
return (__m128i) __builtin_ia32_pmaxsb128 ((__v16qi) __V1, (__v16qi) __V2);
}
-/// \brief Compares the corresponding elements of two 128-bit vectors of
+/// Compares the corresponding elements of two 128-bit vectors of
/// [8 x u16] and returns a 128-bit vector of [8 x u16] containing the lesser
/// value of the two.
///
return (__m128i) __builtin_ia32_pminuw128 ((__v8hi) __V1, (__v8hi) __V2);
}
-/// \brief Compares the corresponding elements of two 128-bit vectors of
+/// Compares the corresponding elements of two 128-bit vectors of
/// [8 x u16] and returns a 128-bit vector of [8 x u16] containing the
/// greater value of the two.
///
return (__m128i) __builtin_ia32_pmaxuw128 ((__v8hi) __V1, (__v8hi) __V2);
}
-/// \brief Compares the corresponding elements of two 128-bit vectors of
+/// Compares the corresponding elements of two 128-bit vectors of
/// [4 x i32] and returns a 128-bit vector of [4 x i32] containing the lesser
/// value of the two.
///
return (__m128i) __builtin_ia32_pminsd128 ((__v4si) __V1, (__v4si) __V2);
}
-/// \brief Compares the corresponding elements of two 128-bit vectors of
+/// Compares the corresponding elements of two 128-bit vectors of
/// [4 x i32] and returns a 128-bit vector of [4 x i32] containing the
/// greater value of the two.
///
return (__m128i) __builtin_ia32_pmaxsd128 ((__v4si) __V1, (__v4si) __V2);
}
-/// \brief Compares the corresponding elements of two 128-bit vectors of
+/// Compares the corresponding elements of two 128-bit vectors of
/// [4 x u32] and returns a 128-bit vector of [4 x u32] containing the lesser
/// value of the two.
///
return (__m128i) __builtin_ia32_pminud128((__v4si) __V1, (__v4si) __V2);
}
-/// \brief Compares the corresponding elements of two 128-bit vectors of
+/// Compares the corresponding elements of two 128-bit vectors of
/// [4 x u32] and returns a 128-bit vector of [4 x u32] containing the
/// greater value of the two.
///
}
/* SSE4 Insertion and Extraction from XMM Register Instructions. */
-/// \brief Takes the first argument \a X and inserts an element from the second
+/// Takes the first argument \a X and inserts an element from the second
/// argument \a Y as selected by the third argument \a N. That result then
/// has elements zeroed out also as selected by the third argument \a N. The
/// resulting 128-bit vector of [4 x float] is then returned.
/// 11: Copies the selected bits from \a Y to result bits [127:96]. \n
/// Bits[3:0]: If any of these bits are set, the corresponding result
/// element is cleared.
-/// \returns A 128-bit vector of [4 x float] containing the copied single-
-/// precision floating point elements from the operands.
+/// \returns A 128-bit vector of [4 x float] containing the copied
+/// single-precision floating point elements from the operands.
#define _mm_insert_ps(X, Y, N) __builtin_ia32_insertps128((X), (Y), (N))
-/// \brief Extracts a 32-bit integer from a 128-bit vector of [4 x float] and
+/// Extracts a 32-bit integer from a 128-bit vector of [4 x float] and
/// returns it, using the immediate value parameter \a N as a selector.
///
/// \headerfile <x86intrin.h>
/// 11: Bits [127:96] of parameter \a X are returned.
/// \returns A 32-bit integer containing the extracted 32 bits of float data.
#define _mm_extract_ps(X, N) (__extension__ \
- ({ union { int __i; float __f; } __t; \
- __v4sf __a = (__v4sf)(__m128)(X); \
- __t.__f = __a[(N) & 3]; \
- __t.__i;}))
+ ({ union { int __i; float __f; } __t; \
+ __t.__f = __builtin_ia32_vec_ext_v4sf((__v4sf)(__m128)(X), (int)(N)); \
+ __t.__i;}))
/* Miscellaneous insert and extract macros. */
/* Extract a single-precision float from X at index N into D. */
-#define _MM_EXTRACT_FLOAT(D, X, N) (__extension__ ({ __v4sf __a = (__v4sf)(X); \
- (D) = __a[N]; }))
+#define _MM_EXTRACT_FLOAT(D, X, N) \
+ { (D) = __builtin_ia32_vec_ext_v4sf((__v4sf)(__m128)(X), (int)(N)); }
/* Or together 2 sets of indexes (X and Y) with the zeroing bits (Z) to create
an index suitable for _mm_insert_ps. */
_MM_MK_INSERTPS_NDX((N), 0, 0x0e))
/* Insert int into packed integer array at index. */
-/// \brief Constructs a 128-bit vector of [16 x i8] by first making a copy of
+/// Constructs a 128-bit vector of [16 x i8] by first making a copy of
/// the 128-bit integer vector parameter, and then inserting the lower 8 bits
/// of an integer parameter \a I into an offset specified by the immediate
/// value parameter \a N.
/// 1110: Bits [119:112] of the result are used for insertion. \n
/// 1111: Bits [127:120] of the result are used for insertion.
/// \returns A 128-bit integer vector containing the constructed values.
-#define _mm_insert_epi8(X, I, N) (__extension__ \
- ({ __v16qi __a = (__v16qi)(__m128i)(X); \
- __a[(N) & 15] = (I); \
- (__m128i)__a;}))
+#define _mm_insert_epi8(X, I, N) \
+ (__m128i)__builtin_ia32_vec_set_v16qi((__v16qi)(__m128i)(X), \
+ (int)(I), (int)(N))
-/// \brief Constructs a 128-bit vector of [4 x i32] by first making a copy of
+/// Constructs a 128-bit vector of [4 x i32] by first making a copy of
/// the 128-bit integer vector parameter, and then inserting the 32-bit
/// integer parameter \a I at the offset specified by the immediate value
/// parameter \a N.
/// 10: Bits [95:64] of the result are used for insertion. \n
/// 11: Bits [127:96] of the result are used for insertion.
/// \returns A 128-bit integer vector containing the constructed values.
-#define _mm_insert_epi32(X, I, N) (__extension__ \
- ({ __v4si __a = (__v4si)(__m128i)(X); \
- __a[(N) & 3] = (I); \
- (__m128i)__a;}))
+#define _mm_insert_epi32(X, I, N) \
+ (__m128i)__builtin_ia32_vec_set_v4si((__v4si)(__m128i)(X), \
+ (int)(I), (int)(N))
#ifdef __x86_64__
-/// \brief Constructs a 128-bit vector of [2 x i64] by first making a copy of
+/// Constructs a 128-bit vector of [2 x i64] by first making a copy of
/// the 128-bit integer vector parameter, and then inserting the 64-bit
/// integer parameter \a I, using the immediate value parameter \a N as an
/// insertion location selector.
/// 0: Bits [63:0] of the result are used for insertion. \n
/// 1: Bits [127:64] of the result are used for insertion. \n
/// \returns A 128-bit integer vector containing the constructed values.
-#define _mm_insert_epi64(X, I, N) (__extension__ \
- ({ __v2di __a = (__v2di)(__m128i)(X); \
- __a[(N) & 1] = (I); \
- (__m128i)__a;}))
+#define _mm_insert_epi64(X, I, N) \
+ (__m128i)__builtin_ia32_vec_set_v2di((__v2di)(__m128i)(X), \
+ (long long)(I), (int)(N))
#endif /* __x86_64__ */
/* Extract int from packed integer array at index. This returns the element
* as a zero extended value, so it is unsigned.
*/
-/// \brief Extracts an 8-bit element from the 128-bit integer vector of
+/// Extracts an 8-bit element from the 128-bit integer vector of
/// [16 x i8], using the immediate value parameter \a N as a selector.
///
/// \headerfile <x86intrin.h>
/// \returns An unsigned integer, whose lower 8 bits are selected from the
/// 128-bit integer vector parameter and the remaining bits are assigned
/// zeros.
-#define _mm_extract_epi8(X, N) (__extension__ \
- ({ __v16qi __a = (__v16qi)(__m128i)(X); \
- (int)(unsigned char) __a[(N) & 15];}))
+#define _mm_extract_epi8(X, N) \
+ (int)(unsigned char)__builtin_ia32_vec_ext_v16qi((__v16qi)(__m128i)(X), \
+ (int)(N))
-/// \brief Extracts a 32-bit element from the 128-bit integer vector of
+/// Extracts a 32-bit element from the 128-bit integer vector of
/// [4 x i32], using the immediate value parameter \a N as a selector.
///
/// \headerfile <x86intrin.h>
/// 11: Bits [127:96] of the parameter \a X are exracted.
/// \returns An integer, whose lower 32 bits are selected from the 128-bit
/// integer vector parameter and the remaining bits are assigned zeros.
-#define _mm_extract_epi32(X, N) (__extension__ \
- ({ __v4si __a = (__v4si)(__m128i)(X); \
- (int)__a[(N) & 3];}))
+#define _mm_extract_epi32(X, N) \
+ (int)__builtin_ia32_vec_ext_v4si((__v4si)(__m128i)(X), (int)(N))
#ifdef __x86_64__
-/// \brief Extracts a 64-bit element from the 128-bit integer vector of
+/// Extracts a 64-bit element from the 128-bit integer vector of
/// [2 x i64], using the immediate value parameter \a N as a selector.
///
/// \headerfile <x86intrin.h>
/// 0: Bits [63:0] are returned. \n
/// 1: Bits [127:64] are returned. \n
/// \returns A 64-bit integer.
-#define _mm_extract_epi64(X, N) (__extension__ \
- ({ __v2di __a = (__v2di)(__m128i)(X); \
- (long long)__a[(N) & 1];}))
+#define _mm_extract_epi64(X, N) \
+ (long long)__builtin_ia32_vec_ext_v2di((__v2di)(__m128i)(X), (int)(N))
#endif /* __x86_64 */
/* SSE4 128-bit Packed Integer Comparisons. */
-/// \brief Tests whether the specified bits in a 128-bit integer vector are all
+/// Tests whether the specified bits in a 128-bit integer vector are all
/// zeros.
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_ptestz128((__v2di)__M, (__v2di)__V);
}
-/// \brief Tests whether the specified bits in a 128-bit integer vector are all
+/// Tests whether the specified bits in a 128-bit integer vector are all
/// ones.
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_ptestc128((__v2di)__M, (__v2di)__V);
}
-/// \brief Tests whether the specified bits in a 128-bit integer vector are
+/// Tests whether the specified bits in a 128-bit integer vector are
/// neither all zeros nor all ones.
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_ptestnzc128((__v2di)__M, (__v2di)__V);
}
-/// \brief Tests whether the specified bits in a 128-bit integer vector are all
+/// Tests whether the specified bits in a 128-bit integer vector are all
/// ones.
///
/// \headerfile <x86intrin.h>
/// otherwise.
#define _mm_test_all_ones(V) _mm_testc_si128((V), _mm_cmpeq_epi32((V), (V)))
-/// \brief Tests whether the specified bits in a 128-bit integer vector are
+/// Tests whether the specified bits in a 128-bit integer vector are
/// neither all zeros nor all ones.
///
/// \headerfile <x86intrin.h>
/// FALSE otherwise.
#define _mm_test_mix_ones_zeros(M, V) _mm_testnzc_si128((M), (V))
-/// \brief Tests whether the specified bits in a 128-bit integer vector are all
+/// Tests whether the specified bits in a 128-bit integer vector are all
/// zeros.
///
/// \headerfile <x86intrin.h>
#define _mm_test_all_zeros(M, V) _mm_testz_si128 ((M), (V))
/* SSE4 64-bit Packed Integer Comparisons. */
-/// \brief Compares each of the corresponding 64-bit values of the 128-bit
+/// Compares each of the corresponding 64-bit values of the 128-bit
/// integer vectors for equality.
///
/// \headerfile <x86intrin.h>
}
/* SSE4 Packed Integer Sign-Extension. */
-/// \brief Sign-extends each of the lower eight 8-bit integer elements of a
+/// Sign-extends each of the lower eight 8-bit integer elements of a
/// 128-bit vector of [16 x i8] to 16-bit values and returns them in a
/// 128-bit vector of [8 x i16]. The upper eight elements of the input vector
/// are unused.
return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v16qs)__V, (__v16qs)__V, 0, 1, 2, 3, 4, 5, 6, 7), __v8hi);
}
-/// \brief Sign-extends each of the lower four 8-bit integer elements of a
+/// Sign-extends each of the lower four 8-bit integer elements of a
/// 128-bit vector of [16 x i8] to 32-bit values and returns them in a
/// 128-bit vector of [4 x i32]. The upper twelve elements of the input
/// vector are unused.
/// This intrinsic corresponds to the <c> VPMOVSXBD / PMOVSXBD </c> instruction.
///
/// \param __V
-/// A 128-bit vector of [16 x i8]. The lower four 8-bit elements are sign-
-/// extended to 32-bit values.
+/// A 128-bit vector of [16 x i8]. The lower four 8-bit elements are
+/// sign-extended to 32-bit values.
/// \returns A 128-bit vector of [4 x i32] containing the sign-extended values.
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_cvtepi8_epi32(__m128i __V)
return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v16qs)__V, (__v16qs)__V, 0, 1, 2, 3), __v4si);
}
-/// \brief Sign-extends each of the lower two 8-bit integer elements of a
+/// Sign-extends each of the lower two 8-bit integer elements of a
/// 128-bit integer vector of [16 x i8] to 64-bit values and returns them in
/// a 128-bit vector of [2 x i64]. The upper fourteen elements of the input
/// vector are unused.
/// This intrinsic corresponds to the <c> VPMOVSXBQ / PMOVSXBQ </c> instruction.
///
/// \param __V
-/// A 128-bit vector of [16 x i8]. The lower two 8-bit elements are sign-
-/// extended to 64-bit values.
+/// A 128-bit vector of [16 x i8]. The lower two 8-bit elements are
+/// sign-extended to 64-bit values.
/// \returns A 128-bit vector of [2 x i64] containing the sign-extended values.
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_cvtepi8_epi64(__m128i __V)
return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v16qs)__V, (__v16qs)__V, 0, 1), __v2di);
}
-/// \brief Sign-extends each of the lower four 16-bit integer elements of a
+/// Sign-extends each of the lower four 16-bit integer elements of a
/// 128-bit integer vector of [8 x i16] to 32-bit values and returns them in
/// a 128-bit vector of [4 x i32]. The upper four elements of the input
/// vector are unused.
/// This intrinsic corresponds to the <c> VPMOVSXWD / PMOVSXWD </c> instruction.
///
/// \param __V
-/// A 128-bit vector of [8 x i16]. The lower four 16-bit elements are sign-
-/// extended to 32-bit values.
+/// A 128-bit vector of [8 x i16]. The lower four 16-bit elements are
+/// sign-extended to 32-bit values.
/// \returns A 128-bit vector of [4 x i32] containing the sign-extended values.
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_cvtepi16_epi32(__m128i __V)
return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v8hi)__V, (__v8hi)__V, 0, 1, 2, 3), __v4si);
}
-/// \brief Sign-extends each of the lower two 16-bit integer elements of a
+/// Sign-extends each of the lower two 16-bit integer elements of a
/// 128-bit integer vector of [8 x i16] to 64-bit values and returns them in
/// a 128-bit vector of [2 x i64]. The upper six elements of the input
/// vector are unused.
/// This intrinsic corresponds to the <c> VPMOVSXWQ / PMOVSXWQ </c> instruction.
///
/// \param __V
-/// A 128-bit vector of [8 x i16]. The lower two 16-bit elements are sign-
-/// extended to 64-bit values.
+/// A 128-bit vector of [8 x i16]. The lower two 16-bit elements are
+/// sign-extended to 64-bit values.
/// \returns A 128-bit vector of [2 x i64] containing the sign-extended values.
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_cvtepi16_epi64(__m128i __V)
return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v8hi)__V, (__v8hi)__V, 0, 1), __v2di);
}
-/// \brief Sign-extends each of the lower two 32-bit integer elements of a
+/// Sign-extends each of the lower two 32-bit integer elements of a
/// 128-bit integer vector of [4 x i32] to 64-bit values and returns them in
/// a 128-bit vector of [2 x i64]. The upper two elements of the input vector
/// are unused.
/// This intrinsic corresponds to the <c> VPMOVSXDQ / PMOVSXDQ </c> instruction.
///
/// \param __V
-/// A 128-bit vector of [4 x i32]. The lower two 32-bit elements are sign-
-/// extended to 64-bit values.
+/// A 128-bit vector of [4 x i32]. The lower two 32-bit elements are
+/// sign-extended to 64-bit values.
/// \returns A 128-bit vector of [2 x i64] containing the sign-extended values.
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_cvtepi32_epi64(__m128i __V)
}
/* SSE4 Packed Integer Zero-Extension. */
-/// \brief Zero-extends each of the lower eight 8-bit integer elements of a
+/// Zero-extends each of the lower eight 8-bit integer elements of a
/// 128-bit vector of [16 x i8] to 16-bit values and returns them in a
/// 128-bit vector of [8 x i16]. The upper eight elements of the input vector
/// are unused.
/// This intrinsic corresponds to the <c> VPMOVZXBW / PMOVZXBW </c> instruction.
///
/// \param __V
-/// A 128-bit vector of [16 x i8]. The lower eight 8-bit elements are zero-
-/// extended to 16-bit values.
+/// A 128-bit vector of [16 x i8]. The lower eight 8-bit elements are
+/// zero-extended to 16-bit values.
/// \returns A 128-bit vector of [8 x i16] containing the zero-extended values.
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_cvtepu8_epi16(__m128i __V)
return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v16qu)__V, (__v16qu)__V, 0, 1, 2, 3, 4, 5, 6, 7), __v8hi);
}
-/// \brief Zero-extends each of the lower four 8-bit integer elements of a
+/// Zero-extends each of the lower four 8-bit integer elements of a
/// 128-bit vector of [16 x i8] to 32-bit values and returns them in a
/// 128-bit vector of [4 x i32]. The upper twelve elements of the input
/// vector are unused.
/// This intrinsic corresponds to the <c> VPMOVZXBD / PMOVZXBD </c> instruction.
///
/// \param __V
-/// A 128-bit vector of [16 x i8]. The lower four 8-bit elements are zero-
-/// extended to 32-bit values.
+/// A 128-bit vector of [16 x i8]. The lower four 8-bit elements are
+/// zero-extended to 32-bit values.
/// \returns A 128-bit vector of [4 x i32] containing the zero-extended values.
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_cvtepu8_epi32(__m128i __V)
return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v16qu)__V, (__v16qu)__V, 0, 1, 2, 3), __v4si);
}
-/// \brief Zero-extends each of the lower two 8-bit integer elements of a
+/// Zero-extends each of the lower two 8-bit integer elements of a
/// 128-bit integer vector of [16 x i8] to 64-bit values and returns them in
/// a 128-bit vector of [2 x i64]. The upper fourteen elements of the input
/// vector are unused.
/// This intrinsic corresponds to the <c> VPMOVZXBQ / PMOVZXBQ </c> instruction.
///
/// \param __V
-/// A 128-bit vector of [16 x i8]. The lower two 8-bit elements are zero-
-/// extended to 64-bit values.
+/// A 128-bit vector of [16 x i8]. The lower two 8-bit elements are
+/// zero-extended to 64-bit values.
/// \returns A 128-bit vector of [2 x i64] containing the zero-extended values.
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_cvtepu8_epi64(__m128i __V)
return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v16qu)__V, (__v16qu)__V, 0, 1), __v2di);
}
-/// \brief Zero-extends each of the lower four 16-bit integer elements of a
+/// Zero-extends each of the lower four 16-bit integer elements of a
/// 128-bit integer vector of [8 x i16] to 32-bit values and returns them in
/// a 128-bit vector of [4 x i32]. The upper four elements of the input
/// vector are unused.
/// This intrinsic corresponds to the <c> VPMOVZXWD / PMOVZXWD </c> instruction.
///
/// \param __V
-/// A 128-bit vector of [8 x i16]. The lower four 16-bit elements are zero-
-/// extended to 32-bit values.
+/// A 128-bit vector of [8 x i16]. The lower four 16-bit elements are
+/// zero-extended to 32-bit values.
/// \returns A 128-bit vector of [4 x i32] containing the zero-extended values.
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_cvtepu16_epi32(__m128i __V)
return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v8hu)__V, (__v8hu)__V, 0, 1, 2, 3), __v4si);
}
-/// \brief Zero-extends each of the lower two 16-bit integer elements of a
+/// Zero-extends each of the lower two 16-bit integer elements of a
/// 128-bit integer vector of [8 x i16] to 64-bit values and returns them in
/// a 128-bit vector of [2 x i64]. The upper six elements of the input vector
/// are unused.
/// This intrinsic corresponds to the <c> VPMOVZXWQ / PMOVZXWQ </c> instruction.
///
/// \param __V
-/// A 128-bit vector of [8 x i16]. The lower two 16-bit elements are zero-
-/// extended to 64-bit values.
+/// A 128-bit vector of [8 x i16]. The lower two 16-bit elements are
+/// zero-extended to 64-bit values.
/// \returns A 128-bit vector of [2 x i64] containing the zero-extended values.
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_cvtepu16_epi64(__m128i __V)
return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v8hu)__V, (__v8hu)__V, 0, 1), __v2di);
}
-/// \brief Zero-extends each of the lower two 32-bit integer elements of a
+/// Zero-extends each of the lower two 32-bit integer elements of a
/// 128-bit integer vector of [4 x i32] to 64-bit values and returns them in
/// a 128-bit vector of [2 x i64]. The upper two elements of the input vector
/// are unused.
/// This intrinsic corresponds to the <c> VPMOVZXDQ / PMOVZXDQ </c> instruction.
///
/// \param __V
-/// A 128-bit vector of [4 x i32]. The lower two 32-bit elements are zero-
-/// extended to 64-bit values.
+/// A 128-bit vector of [4 x i32]. The lower two 32-bit elements are
+/// zero-extended to 64-bit values.
/// \returns A 128-bit vector of [2 x i64] containing the zero-extended values.
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_cvtepu32_epi64(__m128i __V)
}
/* SSE4 Pack with Unsigned Saturation. */
-/// \brief Converts 32-bit signed integers from both 128-bit integer vector
+/// Converts 32-bit signed integers from both 128-bit integer vector
/// operands into 16-bit unsigned integers, and returns the packed result.
/// Values greater than 0xFFFF are saturated to 0xFFFF. Values less than
/// 0x0000 are saturated to 0x0000.
}
/* SSE4 Multiple Packed Sums of Absolute Difference. */
-/// \brief Subtracts 8-bit unsigned integer values and computes the absolute
+/// Subtracts 8-bit unsigned integer values and computes the absolute
/// values of the differences to the corresponding bits in the destination.
/// Then sums of the absolute differences are returned according to the bit
/// fields in the immediate operand.
/// \code
/// // M2 represents bit 2 of the immediate operand
/// // M10 represents bits [1:0] of the immediate operand
-/// i = M2 * 4
-/// j = M10 * 4
+/// i = M2 * 4;
+/// j = M10 * 4;
/// for (k = 0; k < 8; k = k + 1) {
-/// d0 = abs(X[i + k + 0] - Y[j + 0])
-/// d1 = abs(X[i + k + 1] - Y[j + 1])
-/// d2 = abs(X[i + k + 2] - Y[j + 2])
-/// d3 = abs(X[i + k + 3] - Y[j + 3])
-/// r[k] = d0 + d1 + d2 + d3
+/// d0 = abs(X[i + k + 0] - Y[j + 0]);
+/// d1 = abs(X[i + k + 1] - Y[j + 1]);
+/// d2 = abs(X[i + k + 2] - Y[j + 2]);
+/// d3 = abs(X[i + k + 3] - Y[j + 3]);
+/// r[k] = d0 + d1 + d2 + d3;
/// }
/// \endcode
/// \returns A 128-bit integer vector containing the sums of the sets of
/// absolute differences between both operands.
-#define _mm_mpsadbw_epu8(X, Y, M) __extension__ ({ \
+#define _mm_mpsadbw_epu8(X, Y, M) \
(__m128i) __builtin_ia32_mpsadbw128((__v16qi)(__m128i)(X), \
- (__v16qi)(__m128i)(Y), (M)); })
+ (__v16qi)(__m128i)(Y), (M))
-/// \brief Finds the minimum unsigned 16-bit element in the input 128-bit
+/// Finds the minimum unsigned 16-bit element in the input 128-bit
/// vector of [8 x u16] and returns it and along with its index.
///
/// \headerfile <x86intrin.h>
#define _SIDD_UNIT_MASK 0x40
/* SSE4.2 Packed Comparison Intrinsics. */
-/// \brief Uses the immediate operand \a M to perform a comparison of string
+/// Uses the immediate operand \a M to perform a comparison of string
/// data with implicitly defined lengths that is contained in source operands
/// \a A and \a B. Returns a 128-bit integer vector representing the result
/// mask of the comparison.
(__m128i)__builtin_ia32_pcmpistrm128((__v16qi)(__m128i)(A), \
(__v16qi)(__m128i)(B), (int)(M))
-/// \brief Uses the immediate operand \a M to perform a comparison of string
+/// Uses the immediate operand \a M to perform a comparison of string
/// data with implicitly defined lengths that is contained in source operands
/// \a A and \a B. Returns an integer representing the result index of the
/// comparison.
(int)__builtin_ia32_pcmpistri128((__v16qi)(__m128i)(A), \
(__v16qi)(__m128i)(B), (int)(M))
-/// \brief Uses the immediate operand \a M to perform a comparison of string
+/// Uses the immediate operand \a M to perform a comparison of string
/// data with explicitly defined lengths that is contained in source operands
/// \a A and \a B. Returns a 128-bit integer vector representing the result
/// mask of the comparison.
(__v16qi)(__m128i)(B), (int)(LB), \
(int)(M))
-/// \brief Uses the immediate operand \a M to perform a comparison of string
+/// Uses the immediate operand \a M to perform a comparison of string
/// data with explicitly defined lengths that is contained in source operands
/// \a A and \a B. Returns an integer representing the result index of the
/// comparison.
(int)(M))
/* SSE4.2 Packed Comparison Intrinsics and EFlag Reading. */
-/// \brief Uses the immediate operand \a M to perform a comparison of string
+/// Uses the immediate operand \a M to perform a comparison of string
/// data with implicitly defined lengths that is contained in source operands
/// \a A and \a B. Returns 1 if the bit mask is zero and the length of the
/// string in \a B is the maximum, otherwise, returns 0.
(int)__builtin_ia32_pcmpistria128((__v16qi)(__m128i)(A), \
(__v16qi)(__m128i)(B), (int)(M))
-/// \brief Uses the immediate operand \a M to perform a comparison of string
+/// Uses the immediate operand \a M to perform a comparison of string
/// data with implicitly defined lengths that is contained in source operands
/// \a A and \a B. Returns 1 if the bit mask is non-zero, otherwise, returns
/// 0.
(int)__builtin_ia32_pcmpistric128((__v16qi)(__m128i)(A), \
(__v16qi)(__m128i)(B), (int)(M))
-/// \brief Uses the immediate operand \a M to perform a comparison of string
+/// Uses the immediate operand \a M to perform a comparison of string
/// data with implicitly defined lengths that is contained in source operands
/// \a A and \a B. Returns bit 0 of the resulting bit mask.
///
(int)__builtin_ia32_pcmpistrio128((__v16qi)(__m128i)(A), \
(__v16qi)(__m128i)(B), (int)(M))
-/// \brief Uses the immediate operand \a M to perform a comparison of string
+/// Uses the immediate operand \a M to perform a comparison of string
/// data with implicitly defined lengths that is contained in source operands
/// \a A and \a B. Returns 1 if the length of the string in \a A is less than
/// the maximum, otherwise, returns 0.
(int)__builtin_ia32_pcmpistris128((__v16qi)(__m128i)(A), \
(__v16qi)(__m128i)(B), (int)(M))
-/// \brief Uses the immediate operand \a M to perform a comparison of string
+/// Uses the immediate operand \a M to perform a comparison of string
/// data with implicitly defined lengths that is contained in source operands
/// \a A and \a B. Returns 1 if the length of the string in \a B is less than
/// the maximum, otherwise, returns 0.
(int)__builtin_ia32_pcmpistriz128((__v16qi)(__m128i)(A), \
(__v16qi)(__m128i)(B), (int)(M))
-/// \brief Uses the immediate operand \a M to perform a comparison of string
+/// Uses the immediate operand \a M to perform a comparison of string
/// data with explicitly defined lengths that is contained in source operands
/// \a A and \a B. Returns 1 if the bit mask is zero and the length of the
/// string in \a B is the maximum, otherwise, returns 0.
(__v16qi)(__m128i)(B), (int)(LB), \
(int)(M))
-/// \brief Uses the immediate operand \a M to perform a comparison of string
+/// Uses the immediate operand \a M to perform a comparison of string
/// data with explicitly defined lengths that is contained in source operands
/// \a A and \a B. Returns 1 if the resulting mask is non-zero, otherwise,
/// returns 0.
(__v16qi)(__m128i)(B), (int)(LB), \
(int)(M))
-/// \brief Uses the immediate operand \a M to perform a comparison of string
+/// Uses the immediate operand \a M to perform a comparison of string
/// data with explicitly defined lengths that is contained in source operands
/// \a A and \a B. Returns bit 0 of the resulting bit mask.
///
(__v16qi)(__m128i)(B), (int)(LB), \
(int)(M))
-/// \brief Uses the immediate operand \a M to perform a comparison of string
+/// Uses the immediate operand \a M to perform a comparison of string
/// data with explicitly defined lengths that is contained in source operands
/// \a A and \a B. Returns 1 if the length of the string in \a A is less than
/// the maximum, otherwise, returns 0.
(__v16qi)(__m128i)(B), (int)(LB), \
(int)(M))
-/// \brief Uses the immediate operand \a M to perform a comparison of string
+/// Uses the immediate operand \a M to perform a comparison of string
/// data with explicitly defined lengths that is contained in source operands
/// \a A and \a B. Returns 1 if the length of the string in \a B is less than
/// the maximum, otherwise, returns 0.
(int)(M))
/* SSE4.2 Compare Packed Data -- Greater Than. */
-/// \brief Compares each of the corresponding 64-bit values of the 128-bit
+/// Compares each of the corresponding 64-bit values of the 128-bit
/// integer vectors to determine if the values in the first operand are
/// greater than those in the second operand.
///
}
/* SSE4.2 Accumulate CRC32. */
-/// \brief Adds the unsigned integer operand to the CRC-32C checksum of the
+/// Adds the unsigned integer operand to the CRC-32C checksum of the
/// unsigned char operand.
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_crc32qi(__C, __D);
}
-/// \brief Adds the unsigned integer operand to the CRC-32C checksum of the
+/// Adds the unsigned integer operand to the CRC-32C checksum of the
/// unsigned short operand.
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_crc32hi(__C, __D);
}
-/// \brief Adds the first unsigned integer operand to the CRC-32C checksum of
+/// Adds the first unsigned integer operand to the CRC-32C checksum of
/// the second unsigned integer operand.
///
/// \headerfile <x86intrin.h>
}
#ifdef __x86_64__
-/// \brief Adds the unsigned integer operand to the CRC-32C checksum of the
+/// Adds the unsigned integer operand to the CRC-32C checksum of the
/// unsigned 64-bit integer operand.
///
/// \headerfile <x86intrin.h>
#undef __DEFAULT_FN_ATTRS
-#ifdef __POPCNT__
#include <popcntintrin.h>
-#endif
-#endif /* _SMMINTRIN_H */
+#endif /* __SMMINTRIN_H */
--- /dev/null
+/*===---- stdalign.h - Standard header for alignment ------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __STDALIGN_H
+#define __STDALIGN_H
+
+#ifndef __cplusplus
+#define alignas _Alignas
+#define alignof _Alignof
+#endif
+
+#define __alignas_is_defined 1
+#define __alignof_is_defined 1
+
+#endif /* __STDALIGN_H */
--- /dev/null
+/*===---- stdarg.h - Variable argument handling ----------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __STDARG_H
+#define __STDARG_H
+
+#ifndef _VA_LIST
+typedef __builtin_va_list va_list;
+#define _VA_LIST
+#endif
+#define va_start(ap, param) __builtin_va_start(ap, param)
+#define va_end(ap) __builtin_va_end(ap)
+#define va_arg(ap, type) __builtin_va_arg(ap, type)
+
+/* GCC always defines __va_copy, but does not define va_copy unless in c99 mode
+ * or -ansi is not specified, since it was not part of C90.
+ */
+#define __va_copy(d,s) __builtin_va_copy(d,s)
+
+#if __STDC_VERSION__ >= 199901L || __cplusplus >= 201103L || !defined(__STRICT_ANSI__)
+#define va_copy(dest, src) __builtin_va_copy(dest, src)
+#endif
+
+#ifndef __GNUC_VA_LIST
+#define __GNUC_VA_LIST 1
+typedef __builtin_va_list __gnuc_va_list;
+#endif
+
+#endif /* __STDARG_H */
/*===---- stdatomic.h - Standard header for atomic types and operations -----===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
--- /dev/null
+/*===---- stdbool.h - Standard header for booleans -------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __STDBOOL_H
+#define __STDBOOL_H
+
+/* Don't define bool, true, and false in C++, except as a GNU extension. */
+#ifndef __cplusplus
+#define bool _Bool
+#define true 1
+#define false 0
+#elif defined(__GNUC__) && !defined(__STRICT_ANSI__)
+/* Define _Bool as a GNU extension. */
+#define _Bool bool
+#if __cplusplus < 201103L
+/* For C++98, define bool, false, true as a GNU extension. */
+#define bool bool
+#define false false
+#define true true
+#endif
+#endif
+
+#define __bool_true_false_are_defined 1
+
+#endif /* __STDBOOL_H */
/*===---- stddef.h - Basic type definitions --------------------------------===
*
- * Copyright (c) 2008 Eli Friedman
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
/*===---- stdint.h - Standard header for sized integer types --------------===*\
*
- * Copyright (c) 2009 Chris Lattner
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
\*===----------------------------------------------------------------------===*/
#ifndef __CLANG_STDINT_H
+// AIX system headers need stdint.h to be re-enterable while _STD_TYPES_T
+// is defined until an inclusion of it without _STD_TYPES_T occurs, in which
+// case the header guard macro is defined.
+#if !defined(_AIX) || !defined(_STD_TYPES_T) || !defined(__STDC_HOSTED__)
#define __CLANG_STDINT_H
+#endif
/* If we're hosted, fall back to the system's stdint.h, which might have
* additional definitions.
*
* To accommodate targets that are missing types that are exactly 8, 16, 32, or
* 64 bits wide, this implementation takes an approach of cascading
- * redefintions, redefining __int_leastN_t to successively smaller exact-width
+ * redefinitions, redefining __int_leastN_t to successively smaller exact-width
* types. It is therefore important that the types are defined in order of
* descending widths.
*
* As in the type definitions, this section takes an approach of
* successive-shrinking to determine which limits to use for the standard (8,
* 16, 32, 64) bit widths when they don't have exact representations. It is
- * therefore important that the defintions be kept in order of decending
+ * therefore important that the definitions be kept in order of decending
* widths.
*
* Note that C++ should not check __STDC_LIMIT_MACROS here, contrary to the
--- /dev/null
+/*===---- stdnoreturn.h - Standard header for noreturn macro ---------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __STDNORETURN_H
+#define __STDNORETURN_H
+
+#define noreturn _Noreturn
+#define __noreturn_is_defined 1
+
+#endif /* __STDNORETURN_H */
/*===---- tbmintrin.h - TBM intrinsics -------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
/*===---- tgmath.h - Standard header for type generic math ----------------===*\
*
- * Copyright (c) 2009 Howard Hinnant
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
\*===----------------------------------------------------------------------===*/
/*===---- tmmintrin.h - SSSE3 intrinsics -----------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#include <pmmintrin.h>
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("ssse3")))
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("ssse3"), __min_vector_width__(64)))
+#define __DEFAULT_FN_ATTRS_MMX __attribute__((__always_inline__, __nodebug__, __target__("mmx,ssse3"), __min_vector_width__(64)))
-/// \brief Computes the absolute value of each of the packed 8-bit signed
+/// Computes the absolute value of each of the packed 8-bit signed
/// integers in the source operand and stores the 8-bit unsigned integer
/// results in the destination.
///
/// A 64-bit vector of [8 x i8].
/// \returns A 64-bit integer vector containing the absolute values of the
/// elements in the operand.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_abs_pi8(__m64 __a)
{
return (__m64)__builtin_ia32_pabsb((__v8qi)__a);
}
-/// \brief Computes the absolute value of each of the packed 8-bit signed
+/// Computes the absolute value of each of the packed 8-bit signed
/// integers in the source operand and stores the 8-bit unsigned integer
/// results in the destination.
///
return (__m128i)__builtin_ia32_pabsb128((__v16qi)__a);
}
-/// \brief Computes the absolute value of each of the packed 16-bit signed
+/// Computes the absolute value of each of the packed 16-bit signed
/// integers in the source operand and stores the 16-bit unsigned integer
/// results in the destination.
///
/// A 64-bit vector of [4 x i16].
/// \returns A 64-bit integer vector containing the absolute values of the
/// elements in the operand.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_abs_pi16(__m64 __a)
{
return (__m64)__builtin_ia32_pabsw((__v4hi)__a);
}
-/// \brief Computes the absolute value of each of the packed 16-bit signed
+/// Computes the absolute value of each of the packed 16-bit signed
/// integers in the source operand and stores the 16-bit unsigned integer
/// results in the destination.
///
return (__m128i)__builtin_ia32_pabsw128((__v8hi)__a);
}
-/// \brief Computes the absolute value of each of the packed 32-bit signed
+/// Computes the absolute value of each of the packed 32-bit signed
/// integers in the source operand and stores the 32-bit unsigned integer
/// results in the destination.
///
/// A 64-bit vector of [2 x i32].
/// \returns A 64-bit integer vector containing the absolute values of the
/// elements in the operand.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_abs_pi32(__m64 __a)
{
return (__m64)__builtin_ia32_pabsd((__v2si)__a);
}
-/// \brief Computes the absolute value of each of the packed 32-bit signed
+/// Computes the absolute value of each of the packed 32-bit signed
/// integers in the source operand and stores the 32-bit unsigned integer
/// results in the destination.
///
return (__m128i)__builtin_ia32_pabsd128((__v4si)__a);
}
-/// \brief Concatenates the two 128-bit integer vector operands, and
+/// Concatenates the two 128-bit integer vector operands, and
/// right-shifts the result by the number of bytes specified in the immediate
/// operand.
///
/// An immediate operand specifying how many bytes to right-shift the result.
/// \returns A 128-bit integer vector containing the concatenated right-shifted
/// value.
-#define _mm_alignr_epi8(a, b, n) __extension__ ({ \
+#define _mm_alignr_epi8(a, b, n) \
(__m128i)__builtin_ia32_palignr128((__v16qi)(__m128i)(a), \
- (__v16qi)(__m128i)(b), (n)); })
+ (__v16qi)(__m128i)(b), (n))
-/// \brief Concatenates the two 64-bit integer vector operands, and right-shifts
+/// Concatenates the two 64-bit integer vector operands, and right-shifts
/// the result by the number of bytes specified in the immediate operand.
///
/// \headerfile <x86intrin.h>
/// An immediate operand specifying how many bytes to right-shift the result.
/// \returns A 64-bit integer vector containing the concatenated right-shifted
/// value.
-#define _mm_alignr_pi8(a, b, n) __extension__ ({ \
- (__m64)__builtin_ia32_palignr((__v8qi)(__m64)(a), (__v8qi)(__m64)(b), (n)); })
+#define _mm_alignr_pi8(a, b, n) \
+ (__m64)__builtin_ia32_palignr((__v8qi)(__m64)(a), (__v8qi)(__m64)(b), (n))
-/// \brief Horizontally adds the adjacent pairs of values contained in 2 packed
+/// Horizontally adds the adjacent pairs of values contained in 2 packed
/// 128-bit vectors of [8 x i16].
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_ia32_phaddw128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Horizontally adds the adjacent pairs of values contained in 2 packed
+/// Horizontally adds the adjacent pairs of values contained in 2 packed
/// 128-bit vectors of [4 x i32].
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_ia32_phaddd128((__v4si)__a, (__v4si)__b);
}
-/// \brief Horizontally adds the adjacent pairs of values contained in 2 packed
+/// Horizontally adds the adjacent pairs of values contained in 2 packed
/// 64-bit vectors of [4 x i16].
///
/// \headerfile <x86intrin.h>
/// destination.
/// \returns A 64-bit vector of [4 x i16] containing the horizontal sums of both
/// operands.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_hadd_pi16(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_phaddw((__v4hi)__a, (__v4hi)__b);
}
-/// \brief Horizontally adds the adjacent pairs of values contained in 2 packed
+/// Horizontally adds the adjacent pairs of values contained in 2 packed
/// 64-bit vectors of [2 x i32].
///
/// \headerfile <x86intrin.h>
/// destination.
/// \returns A 64-bit vector of [2 x i32] containing the horizontal sums of both
/// operands.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_hadd_pi32(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_phaddd((__v2si)__a, (__v2si)__b);
}
-/// \brief Horizontally adds the adjacent pairs of values contained in 2 packed
-/// 128-bit vectors of [8 x i16]. Positive sums greater than 7FFFh are
-/// saturated to 7FFFh. Negative sums less than 8000h are saturated to 8000h.
+/// Horizontally adds the adjacent pairs of values contained in 2 packed
+/// 128-bit vectors of [8 x i16]. Positive sums greater than 0x7FFF are
+/// saturated to 0x7FFF. Negative sums less than 0x8000 are saturated to
+/// 0x8000.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)__builtin_ia32_phaddsw128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Horizontally adds the adjacent pairs of values contained in 2 packed
-/// 64-bit vectors of [4 x i16]. Positive sums greater than 7FFFh are
-/// saturated to 7FFFh. Negative sums less than 8000h are saturated to 8000h.
+/// Horizontally adds the adjacent pairs of values contained in 2 packed
+/// 64-bit vectors of [4 x i16]. Positive sums greater than 0x7FFF are
+/// saturated to 0x7FFF. Negative sums less than 0x8000 are saturated to
+/// 0x8000.
///
/// \headerfile <x86intrin.h>
///
/// destination.
/// \returns A 64-bit vector of [4 x i16] containing the horizontal saturated
/// sums of both operands.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_hadds_pi16(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_phaddsw((__v4hi)__a, (__v4hi)__b);
}
-/// \brief Horizontally subtracts the adjacent pairs of values contained in 2
+/// Horizontally subtracts the adjacent pairs of values contained in 2
/// packed 128-bit vectors of [8 x i16].
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_ia32_phsubw128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Horizontally subtracts the adjacent pairs of values contained in 2
+/// Horizontally subtracts the adjacent pairs of values contained in 2
/// packed 128-bit vectors of [4 x i32].
///
/// \headerfile <x86intrin.h>
return (__m128i)__builtin_ia32_phsubd128((__v4si)__a, (__v4si)__b);
}
-/// \brief Horizontally subtracts the adjacent pairs of values contained in 2
+/// Horizontally subtracts the adjacent pairs of values contained in 2
/// packed 64-bit vectors of [4 x i16].
///
/// \headerfile <x86intrin.h>
/// the destination.
/// \returns A 64-bit vector of [4 x i16] containing the horizontal differences
/// of both operands.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_hsub_pi16(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_phsubw((__v4hi)__a, (__v4hi)__b);
}
-/// \brief Horizontally subtracts the adjacent pairs of values contained in 2
+/// Horizontally subtracts the adjacent pairs of values contained in 2
/// packed 64-bit vectors of [2 x i32].
///
/// \headerfile <x86intrin.h>
/// the destination.
/// \returns A 64-bit vector of [2 x i32] containing the horizontal differences
/// of both operands.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_hsub_pi32(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_phsubd((__v2si)__a, (__v2si)__b);
}
-/// \brief Horizontally subtracts the adjacent pairs of values contained in 2
+/// Horizontally subtracts the adjacent pairs of values contained in 2
/// packed 128-bit vectors of [8 x i16]. Positive differences greater than
-/// 7FFFh are saturated to 7FFFh. Negative differences less than 8000h are
-/// saturated to 8000h.
+/// 0x7FFF are saturated to 0x7FFF. Negative differences less than 0x8000 are
+/// saturated to 0x8000.
///
/// \headerfile <x86intrin.h>
///
return (__m128i)__builtin_ia32_phsubsw128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Horizontally subtracts the adjacent pairs of values contained in 2
+/// Horizontally subtracts the adjacent pairs of values contained in 2
/// packed 64-bit vectors of [4 x i16]. Positive differences greater than
-/// 7FFFh are saturated to 7FFFh. Negative differences less than 8000h are
-/// saturated to 8000h.
+/// 0x7FFF are saturated to 0x7FFF. Negative differences less than 0x8000 are
+/// saturated to 0x8000.
///
/// \headerfile <x86intrin.h>
///
/// the destination.
/// \returns A 64-bit vector of [4 x i16] containing the horizontal saturated
/// differences of both operands.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_hsubs_pi16(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_phsubsw((__v4hi)__a, (__v4hi)__b);
}
-/// \brief Multiplies corresponding pairs of packed 8-bit unsigned integer
+/// Multiplies corresponding pairs of packed 8-bit unsigned integer
/// values contained in the first source operand and packed 8-bit signed
/// integer values contained in the second source operand, adds pairs of
/// contiguous products with signed saturation, and writes the 16-bit sums to
return (__m128i)__builtin_ia32_pmaddubsw128((__v16qi)__a, (__v16qi)__b);
}
-/// \brief Multiplies corresponding pairs of packed 8-bit unsigned integer
+/// Multiplies corresponding pairs of packed 8-bit unsigned integer
/// values contained in the first source operand and packed 8-bit signed
/// integer values contained in the second source operand, adds pairs of
/// contiguous products with signed saturation, and writes the 16-bit sums to
/// \a R1 := (\a __a2 * \a __b2) + (\a __a3 * \a __b3) \n
/// \a R2 := (\a __a4 * \a __b4) + (\a __a5 * \a __b5) \n
/// \a R3 := (\a __a6 * \a __b6) + (\a __a7 * \a __b7)
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_maddubs_pi16(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_pmaddubsw((__v8qi)__a, (__v8qi)__b);
}
-/// \brief Multiplies packed 16-bit signed integer values, truncates the 32-bit
+/// Multiplies packed 16-bit signed integer values, truncates the 32-bit
/// products to the 18 most significant bits by right-shifting, rounds the
/// truncated value by adding 1, and writes bits [16:1] to the destination.
///
return (__m128i)__builtin_ia32_pmulhrsw128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief Multiplies packed 16-bit signed integer values, truncates the 32-bit
+/// Multiplies packed 16-bit signed integer values, truncates the 32-bit
/// products to the 18 most significant bits by right-shifting, rounds the
/// truncated value by adding 1, and writes bits [16:1] to the destination.
///
/// A 64-bit vector of [4 x i16] containing one of the source operands.
/// \returns A 64-bit vector of [4 x i16] containing the rounded and scaled
/// products of both operands.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_mulhrs_pi16(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_pmulhrsw((__v4hi)__a, (__v4hi)__b);
}
-/// \brief Copies the 8-bit integers from a 128-bit integer vector to the
+/// Copies the 8-bit integers from a 128-bit integer vector to the
/// destination or clears 8-bit values in the destination, as specified by
/// the second source operand.
///
return (__m128i)__builtin_ia32_pshufb128((__v16qi)__a, (__v16qi)__b);
}
-/// \brief Copies the 8-bit integers from a 64-bit integer vector to the
+/// Copies the 8-bit integers from a 64-bit integer vector to the
/// destination or clears 8-bit values in the destination, as specified by
/// the second source operand.
///
/// destination. \n
/// Bits [3:0] select the source byte to be copied.
/// \returns A 64-bit integer vector containing the copied or cleared values.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_shuffle_pi8(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_pshufb((__v8qi)__a, (__v8qi)__b);
}
-/// \brief For each 8-bit integer in the first source operand, perform one of
+/// For each 8-bit integer in the first source operand, perform one of
/// the following actions as specified by the second source operand.
///
/// If the byte in the second source is negative, calculate the two's
return (__m128i)__builtin_ia32_psignb128((__v16qi)__a, (__v16qi)__b);
}
-/// \brief For each 16-bit integer in the first source operand, perform one of
+/// For each 16-bit integer in the first source operand, perform one of
/// the following actions as specified by the second source operand.
///
/// If the word in the second source is negative, calculate the two's
return (__m128i)__builtin_ia32_psignw128((__v8hi)__a, (__v8hi)__b);
}
-/// \brief For each 32-bit integer in the first source operand, perform one of
+/// For each 32-bit integer in the first source operand, perform one of
/// the following actions as specified by the second source operand.
///
/// If the doubleword in the second source is negative, calculate the two's
return (__m128i)__builtin_ia32_psignd128((__v4si)__a, (__v4si)__b);
}
-/// \brief For each 8-bit integer in the first source operand, perform one of
+/// For each 8-bit integer in the first source operand, perform one of
/// the following actions as specified by the second source operand.
///
/// If the byte in the second source is negative, calculate the two's
/// A 64-bit integer vector containing control bytes corresponding to
/// positions in the destination.
/// \returns A 64-bit integer vector containing the resultant values.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_sign_pi8(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_psignb((__v8qi)__a, (__v8qi)__b);
}
-/// \brief For each 16-bit integer in the first source operand, perform one of
+/// For each 16-bit integer in the first source operand, perform one of
/// the following actions as specified by the second source operand.
///
/// If the word in the second source is negative, calculate the two's
/// A 64-bit integer vector containing control words corresponding to
/// positions in the destination.
/// \returns A 64-bit integer vector containing the resultant values.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_sign_pi16(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_psignw((__v4hi)__a, (__v4hi)__b);
}
-/// \brief For each 32-bit integer in the first source operand, perform one of
+/// For each 32-bit integer in the first source operand, perform one of
/// the following actions as specified by the second source operand.
///
/// If the doubleword in the second source is negative, calculate the two's
/// A 64-bit integer vector containing two control doublewords corresponding
/// to positions in the destination.
/// \returns A 64-bit integer vector containing the resultant values.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_sign_pi32(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_psignd((__v2si)__a, (__v2si)__b);
}
#undef __DEFAULT_FN_ATTRS
+#undef __DEFAULT_FN_ATTRS_MMX
#endif /* __TMMINTRIN_H */
--- /dev/null
+/*===------------- tsxldtrkintrin.h - tsxldtrk intrinsics ------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <tsxldtrkintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __TSXLDTRKINTRIN_H
+#define __TSXLDTRKINTRIN_H
+
+/* Define the default attributes for the functions in this file */
+#define _DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("tsxldtrk")))
+
+/// Marks the start of an TSX (RTM) suspend load address tracking region. If
+/// this intrinsic is used inside a transactional region, subsequent loads
+/// are not added to the read set of the transaction. If it's used inside a
+/// suspend load address tracking region it will cause transaction abort.
+/// If it's used outside of a transactional region it behaves like a NOP.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the \c XSUSLDTRK instruction.
+///
+static __inline__ void _DEFAULT_FN_ATTRS
+_xsusldtrk (void)
+{
+ __builtin_ia32_xsusldtrk();
+}
+
+/// Marks the end of an TSX (RTM) suspend load address tracking region. If this
+/// intrinsic is used inside a suspend load address tracking region it will
+/// end the suspend region and all following load addresses will be added to
+/// the transaction read set. If it's used inside an active transaction but
+/// not in a suspend region it will cause transaction abort. If it's used
+/// outside of a transactional region it behaves like a NOP.
+///
+/// \headerfile <x86intrin.h>
+///
+/// This intrinsic corresponds to the \c XRESLDTRK instruction.
+///
+static __inline__ void _DEFAULT_FN_ATTRS
+_xresldtrk (void)
+{
+ __builtin_ia32_xresldtrk();
+}
+
+#undef _DEFAULT_FN_ATTRS
+
+#endif /* __TSXLDTRKINTRIN_H */
/*===---- unwind.h - Stack unwinding ----------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#pragma GCC visibility push(default)
#endif
-typedef uintptr_t _Unwind_Word;
-typedef intptr_t _Unwind_Sword;
+typedef uintptr_t _Unwind_Word __attribute__((__mode__(__unwind_word__)));
+typedef intptr_t _Unwind_Sword __attribute__((__mode__(__unwind_word__)));
typedef uintptr_t _Unwind_Ptr;
typedef uintptr_t _Unwind_Internal_Ptr;
typedef uint64_t _Unwind_Exception_Class;
typedef uintptr_t _uleb128_t;
struct _Unwind_Context;
+#if defined(__arm__) && !(defined(__USING_SJLJ_EXCEPTIONS__) || defined(__ARM_DWARF_EH__))
+struct _Unwind_Control_Block;
+typedef struct _Unwind_Control_Block _Unwind_Exception; /* Alias */
+#else
struct _Unwind_Exception;
+typedef struct _Unwind_Exception _Unwind_Exception;
+#endif
typedef enum {
_URC_NO_REASON = 0,
#if defined(__arm__) && !defined(__USING_SJLJ_EXCEPTIONS__) && \
} _Unwind_Action;
typedef void (*_Unwind_Exception_Cleanup_Fn)(_Unwind_Reason_Code,
- struct _Unwind_Exception *);
-
+ _Unwind_Exception *);
+
+#if defined(__arm__) && !(defined(__USING_SJLJ_EXCEPTIONS__) || defined(__ARM_DWARF_EH__))
+typedef struct _Unwind_Control_Block _Unwind_Control_Block;
+typedef uint32_t _Unwind_EHT_Header;
+
+struct _Unwind_Control_Block {
+ uint64_t exception_class;
+ void (*exception_cleanup)(_Unwind_Reason_Code, _Unwind_Control_Block *);
+ /* unwinder cache (private fields for the unwinder's use) */
+ struct {
+ uint32_t reserved1; /* forced unwind stop function, 0 if not forced */
+ uint32_t reserved2; /* personality routine */
+ uint32_t reserved3; /* callsite */
+ uint32_t reserved4; /* forced unwind stop argument */
+ uint32_t reserved5;
+ } unwinder_cache;
+ /* propagation barrier cache (valid after phase 1) */
+ struct {
+ uint32_t sp;
+ uint32_t bitpattern[5];
+ } barrier_cache;
+ /* cleanup cache (preserved over cleanup) */
+ struct {
+ uint32_t bitpattern[4];
+ } cleanup_cache;
+ /* personality cache (for personality's benefit) */
+ struct {
+ uint32_t fnstart; /* function start address */
+ _Unwind_EHT_Header *ehtp; /* pointer to EHT entry header word */
+ uint32_t additional; /* additional data */
+ uint32_t reserved1;
+ } pr_cache;
+ long long int : 0; /* force alignment of next item to 8-byte boundary */
+} __attribute__((__aligned__(8)));
+#else
struct _Unwind_Exception {
_Unwind_Exception_Class exception_class;
_Unwind_Exception_Cleanup_Fn exception_cleanup;
+#if !defined (__USING_SJLJ_EXCEPTIONS__) && defined (__SEH__)
+ _Unwind_Word private_[6];
+#else
_Unwind_Word private_1;
_Unwind_Word private_2;
+#endif
/* The Itanium ABI requires that _Unwind_Exception objects are "double-word
* aligned". GCC has interpreted this to mean "use the maximum useful
* alignment for the target"; so do we. */
} __attribute__((__aligned__));
+#endif
typedef _Unwind_Reason_Code (*_Unwind_Stop_Fn)(int, _Unwind_Action,
_Unwind_Exception_Class,
- struct _Unwind_Exception *,
+ _Unwind_Exception *,
struct _Unwind_Context *,
void *);
-typedef _Unwind_Reason_Code (*_Unwind_Personality_Fn)(
- int, _Unwind_Action, _Unwind_Exception_Class, struct _Unwind_Exception *,
- struct _Unwind_Context *);
+typedef _Unwind_Reason_Code (*_Unwind_Personality_Fn)(int, _Unwind_Action,
+ _Unwind_Exception_Class,
+ _Unwind_Exception *,
+ struct _Unwind_Context *);
typedef _Unwind_Personality_Fn __personality_routine;
typedef _Unwind_Reason_Code (*_Unwind_Trace_Fn)(struct _Unwind_Context *,
void *);
-#if defined(__arm__) && !defined(__APPLE__)
-
+#if defined(__arm__) && !(defined(__USING_SJLJ_EXCEPTIONS__) || defined(__ARM_DWARF_EH__))
typedef enum {
_UVRSC_CORE = 0, /* integer register */
_UVRSC_VFP = 1, /* vfp */
_UVRSR_FAILED = 2
} _Unwind_VRS_Result;
-#if !defined(__USING_SJLJ_EXCEPTIONS__) && !defined(__ARM_DWARF_EH__)
typedef uint32_t _Unwind_State;
#define _US_VIRTUAL_UNWIND_FRAME ((_Unwind_State)0)
#define _US_UNWIND_FRAME_STARTING ((_Unwind_State)1)
#define _US_UNWIND_FRAME_RESUME ((_Unwind_State)2)
#define _US_ACTION_MASK ((_Unwind_State)3)
#define _US_FORCE_UNWIND ((_Unwind_State)8)
-#endif
_Unwind_VRS_Result _Unwind_VRS_Get(struct _Unwind_Context *__context,
_Unwind_VRS_RegClass __regclass,
/* DWARF EH functions; currently not available on Darwin/ARM */
#if !defined(__APPLE__) || !defined(__arm__)
-
-_Unwind_Reason_Code _Unwind_RaiseException(struct _Unwind_Exception *);
-_Unwind_Reason_Code _Unwind_ForcedUnwind(struct _Unwind_Exception *,
- _Unwind_Stop_Fn, void *);
-void _Unwind_DeleteException(struct _Unwind_Exception *);
-void _Unwind_Resume(struct _Unwind_Exception *);
-_Unwind_Reason_Code _Unwind_Resume_or_Rethrow(struct _Unwind_Exception *);
+_Unwind_Reason_Code _Unwind_RaiseException(_Unwind_Exception *);
+_Unwind_Reason_Code _Unwind_ForcedUnwind(_Unwind_Exception *, _Unwind_Stop_Fn,
+ void *);
+void _Unwind_DeleteException(_Unwind_Exception *);
+void _Unwind_Resume(_Unwind_Exception *);
+_Unwind_Reason_Code _Unwind_Resume_or_Rethrow(_Unwind_Exception *);
#endif
void _Unwind_SjLj_Register(_Unwind_FunctionContext_t);
void _Unwind_SjLj_Unregister(_Unwind_FunctionContext_t);
-_Unwind_Reason_Code _Unwind_SjLj_RaiseException(struct _Unwind_Exception *);
-_Unwind_Reason_Code _Unwind_SjLj_ForcedUnwind(struct _Unwind_Exception *,
+_Unwind_Reason_Code _Unwind_SjLj_RaiseException(_Unwind_Exception *);
+_Unwind_Reason_Code _Unwind_SjLj_ForcedUnwind(_Unwind_Exception *,
_Unwind_Stop_Fn, void *);
-void _Unwind_SjLj_Resume(struct _Unwind_Exception *);
-_Unwind_Reason_Code _Unwind_SjLj_Resume_or_Rethrow(struct _Unwind_Exception *);
+void _Unwind_SjLj_Resume(_Unwind_Exception *);
+_Unwind_Reason_Code _Unwind_SjLj_Resume_or_Rethrow(_Unwind_Exception *);
void *_Unwind_FindEnclosingFunction(void *);
/* ===-------- vadefs.h ---------------------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
--- /dev/null
+/*===------------------ vaesintrin.h - VAES intrinsics ---------------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <vaesintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __VAESINTRIN_H
+#define __VAESINTRIN_H
+
+/* Default attributes for YMM forms. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("vaes"), __min_vector_width__(256)))
+
+/* Default attributes for ZMM forms. */
+#define __DEFAULT_FN_ATTRS_F __attribute__((__always_inline__, __nodebug__, __target__("avx512f,vaes"), __min_vector_width__(512)))
+
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS
+ _mm256_aesenc_epi128(__m256i __A, __m256i __B)
+{
+ return (__m256i) __builtin_ia32_aesenc256((__v4di) __A,
+ (__v4di) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS_F
+ _mm512_aesenc_epi128(__m512i __A, __m512i __B)
+{
+ return (__m512i) __builtin_ia32_aesenc512((__v8di) __A,
+ (__v8di) __B);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS
+ _mm256_aesdec_epi128(__m256i __A, __m256i __B)
+{
+ return (__m256i) __builtin_ia32_aesdec256((__v4di) __A,
+ (__v4di) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS_F
+ _mm512_aesdec_epi128(__m512i __A, __m512i __B)
+{
+ return (__m512i) __builtin_ia32_aesdec512((__v8di) __A,
+ (__v8di) __B);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS
+ _mm256_aesenclast_epi128(__m256i __A, __m256i __B)
+{
+ return (__m256i) __builtin_ia32_aesenclast256((__v4di) __A,
+ (__v4di) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS_F
+ _mm512_aesenclast_epi128(__m512i __A, __m512i __B)
+{
+ return (__m512i) __builtin_ia32_aesenclast512((__v8di) __A,
+ (__v8di) __B);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS
+ _mm256_aesdeclast_epi128(__m256i __A, __m256i __B)
+{
+ return (__m256i) __builtin_ia32_aesdeclast256((__v4di) __A,
+ (__v4di) __B);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS_F
+ _mm512_aesdeclast_epi128(__m512i __A, __m512i __B)
+{
+ return (__m512i) __builtin_ia32_aesdeclast512((__v8di) __A,
+ (__v8di) __B);
+}
+
+
+#undef __DEFAULT_FN_ATTRS
+#undef __DEFAULT_FN_ATTRS_F
+
+#endif
--- /dev/null
+/*===---- varargs.h - Variable argument handling -------------------------------------===
+*
+* Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+* See https://llvm.org/LICENSE.txt for license information.
+* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+*
+*===-----------------------------------------------------------------------===
+*/
+#ifndef __VARARGS_H
+#define __VARARGS_H
+ #error "Please use <stdarg.h> instead of <varargs.h>"
+#endif
--- /dev/null
+/*===---- vecintrin.h - Vector intrinsics ----------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#if defined(__s390x__) && defined(__VEC__)
+
+#define __ATTRS_ai __attribute__((__always_inline__))
+#define __ATTRS_o __attribute__((__overloadable__))
+#define __ATTRS_o_ai __attribute__((__overloadable__, __always_inline__))
+
+#define __constant(PARM) \
+ __attribute__((__enable_if__ ((PARM) == (PARM), \
+ "argument must be a constant integer")))
+#define __constant_range(PARM, LOW, HIGH) \
+ __attribute__((__enable_if__ ((PARM) >= (LOW) && (PARM) <= (HIGH), \
+ "argument must be a constant integer from " #LOW " to " #HIGH)))
+#define __constant_pow2_range(PARM, LOW, HIGH) \
+ __attribute__((__enable_if__ ((PARM) >= (LOW) && (PARM) <= (HIGH) && \
+ ((PARM) & ((PARM) - 1)) == 0, \
+ "argument must be a constant power of 2 from " #LOW " to " #HIGH)))
+
+/*-- __lcbb -----------------------------------------------------------------*/
+
+extern __ATTRS_o unsigned int
+__lcbb(const void *__ptr, unsigned short __len)
+ __constant_pow2_range(__len, 64, 4096);
+
+#define __lcbb(X, Y) ((__typeof__((__lcbb)((X), (Y)))) \
+ __builtin_s390_lcbb((X), __builtin_constant_p((Y))? \
+ ((Y) == 64 ? 0 : \
+ (Y) == 128 ? 1 : \
+ (Y) == 256 ? 2 : \
+ (Y) == 512 ? 3 : \
+ (Y) == 1024 ? 4 : \
+ (Y) == 2048 ? 5 : \
+ (Y) == 4096 ? 6 : 0) : 0))
+
+/*-- vec_extract ------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai signed char
+vec_extract(__vector signed char __vec, int __index) {
+ return __vec[__index & 15];
+}
+
+static inline __ATTRS_o_ai unsigned char
+vec_extract(__vector __bool char __vec, int __index) {
+ return __vec[__index & 15];
+}
+
+static inline __ATTRS_o_ai unsigned char
+vec_extract(__vector unsigned char __vec, int __index) {
+ return __vec[__index & 15];
+}
+
+static inline __ATTRS_o_ai signed short
+vec_extract(__vector signed short __vec, int __index) {
+ return __vec[__index & 7];
+}
+
+static inline __ATTRS_o_ai unsigned short
+vec_extract(__vector __bool short __vec, int __index) {
+ return __vec[__index & 7];
+}
+
+static inline __ATTRS_o_ai unsigned short
+vec_extract(__vector unsigned short __vec, int __index) {
+ return __vec[__index & 7];
+}
+
+static inline __ATTRS_o_ai signed int
+vec_extract(__vector signed int __vec, int __index) {
+ return __vec[__index & 3];
+}
+
+static inline __ATTRS_o_ai unsigned int
+vec_extract(__vector __bool int __vec, int __index) {
+ return __vec[__index & 3];
+}
+
+static inline __ATTRS_o_ai unsigned int
+vec_extract(__vector unsigned int __vec, int __index) {
+ return __vec[__index & 3];
+}
+
+static inline __ATTRS_o_ai signed long long
+vec_extract(__vector signed long long __vec, int __index) {
+ return __vec[__index & 1];
+}
+
+static inline __ATTRS_o_ai unsigned long long
+vec_extract(__vector __bool long long __vec, int __index) {
+ return __vec[__index & 1];
+}
+
+static inline __ATTRS_o_ai unsigned long long
+vec_extract(__vector unsigned long long __vec, int __index) {
+ return __vec[__index & 1];
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai float
+vec_extract(__vector float __vec, int __index) {
+ return __vec[__index & 3];
+}
+#endif
+
+static inline __ATTRS_o_ai double
+vec_extract(__vector double __vec, int __index) {
+ return __vec[__index & 1];
+}
+
+/*-- vec_insert -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_insert(signed char __scalar, __vector signed char __vec, int __index) {
+ __vec[__index & 15] = __scalar;
+ return __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_insert(unsigned char __scalar, __vector __bool char __vec, int __index) {
+ __vector unsigned char __newvec = (__vector unsigned char)__vec;
+ __newvec[__index & 15] = (unsigned char)__scalar;
+ return __newvec;
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_insert(unsigned char __scalar, __vector unsigned char __vec, int __index) {
+ __vec[__index & 15] = __scalar;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_insert(signed short __scalar, __vector signed short __vec, int __index) {
+ __vec[__index & 7] = __scalar;
+ return __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_insert(unsigned short __scalar, __vector __bool short __vec,
+ int __index) {
+ __vector unsigned short __newvec = (__vector unsigned short)__vec;
+ __newvec[__index & 7] = (unsigned short)__scalar;
+ return __newvec;
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_insert(unsigned short __scalar, __vector unsigned short __vec,
+ int __index) {
+ __vec[__index & 7] = __scalar;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_insert(signed int __scalar, __vector signed int __vec, int __index) {
+ __vec[__index & 3] = __scalar;
+ return __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_insert(unsigned int __scalar, __vector __bool int __vec, int __index) {
+ __vector unsigned int __newvec = (__vector unsigned int)__vec;
+ __newvec[__index & 3] = __scalar;
+ return __newvec;
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_insert(unsigned int __scalar, __vector unsigned int __vec, int __index) {
+ __vec[__index & 3] = __scalar;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_insert(signed long long __scalar, __vector signed long long __vec,
+ int __index) {
+ __vec[__index & 1] = __scalar;
+ return __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_insert(unsigned long long __scalar, __vector __bool long long __vec,
+ int __index) {
+ __vector unsigned long long __newvec = (__vector unsigned long long)__vec;
+ __newvec[__index & 1] = __scalar;
+ return __newvec;
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_insert(unsigned long long __scalar, __vector unsigned long long __vec,
+ int __index) {
+ __vec[__index & 1] = __scalar;
+ return __vec;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_insert(float __scalar, __vector float __vec, int __index) {
+ __vec[__index & 1] = __scalar;
+ return __vec;
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_insert(double __scalar, __vector double __vec, int __index) {
+ __vec[__index & 1] = __scalar;
+ return __vec;
+}
+
+/*-- vec_promote ------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_promote(signed char __scalar, int __index) {
+ const __vector signed char __zero = (__vector signed char)0;
+ __vector signed char __vec = __builtin_shufflevector(__zero, __zero,
+ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1);
+ __vec[__index & 15] = __scalar;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_promote(unsigned char __scalar, int __index) {
+ const __vector unsigned char __zero = (__vector unsigned char)0;
+ __vector unsigned char __vec = __builtin_shufflevector(__zero, __zero,
+ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1);
+ __vec[__index & 15] = __scalar;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_promote(signed short __scalar, int __index) {
+ const __vector signed short __zero = (__vector signed short)0;
+ __vector signed short __vec = __builtin_shufflevector(__zero, __zero,
+ -1, -1, -1, -1, -1, -1, -1, -1);
+ __vec[__index & 7] = __scalar;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_promote(unsigned short __scalar, int __index) {
+ const __vector unsigned short __zero = (__vector unsigned short)0;
+ __vector unsigned short __vec = __builtin_shufflevector(__zero, __zero,
+ -1, -1, -1, -1, -1, -1, -1, -1);
+ __vec[__index & 7] = __scalar;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_promote(signed int __scalar, int __index) {
+ const __vector signed int __zero = (__vector signed int)0;
+ __vector signed int __vec = __builtin_shufflevector(__zero, __zero,
+ -1, -1, -1, -1);
+ __vec[__index & 3] = __scalar;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_promote(unsigned int __scalar, int __index) {
+ const __vector unsigned int __zero = (__vector unsigned int)0;
+ __vector unsigned int __vec = __builtin_shufflevector(__zero, __zero,
+ -1, -1, -1, -1);
+ __vec[__index & 3] = __scalar;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_promote(signed long long __scalar, int __index) {
+ const __vector signed long long __zero = (__vector signed long long)0;
+ __vector signed long long __vec = __builtin_shufflevector(__zero, __zero,
+ -1, -1);
+ __vec[__index & 1] = __scalar;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_promote(unsigned long long __scalar, int __index) {
+ const __vector unsigned long long __zero = (__vector unsigned long long)0;
+ __vector unsigned long long __vec = __builtin_shufflevector(__zero, __zero,
+ -1, -1);
+ __vec[__index & 1] = __scalar;
+ return __vec;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_promote(float __scalar, int __index) {
+ const __vector float __zero = (__vector float)0.0f;
+ __vector float __vec = __builtin_shufflevector(__zero, __zero,
+ -1, -1, -1, -1);
+ __vec[__index & 3] = __scalar;
+ return __vec;
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_promote(double __scalar, int __index) {
+ const __vector double __zero = (__vector double)0.0;
+ __vector double __vec = __builtin_shufflevector(__zero, __zero, -1, -1);
+ __vec[__index & 1] = __scalar;
+ return __vec;
+}
+
+/*-- vec_insert_and_zero ----------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_insert_and_zero(const signed char *__ptr) {
+ __vector signed char __vec = (__vector signed char)0;
+ __vec[7] = *__ptr;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_insert_and_zero(const unsigned char *__ptr) {
+ __vector unsigned char __vec = (__vector unsigned char)0;
+ __vec[7] = *__ptr;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_insert_and_zero(const signed short *__ptr) {
+ __vector signed short __vec = (__vector signed short)0;
+ __vec[3] = *__ptr;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_insert_and_zero(const unsigned short *__ptr) {
+ __vector unsigned short __vec = (__vector unsigned short)0;
+ __vec[3] = *__ptr;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_insert_and_zero(const signed int *__ptr) {
+ __vector signed int __vec = (__vector signed int)0;
+ __vec[1] = *__ptr;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_insert_and_zero(const unsigned int *__ptr) {
+ __vector unsigned int __vec = (__vector unsigned int)0;
+ __vec[1] = *__ptr;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_insert_and_zero(const signed long long *__ptr) {
+ __vector signed long long __vec = (__vector signed long long)0;
+ __vec[0] = *__ptr;
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_insert_and_zero(const unsigned long long *__ptr) {
+ __vector unsigned long long __vec = (__vector unsigned long long)0;
+ __vec[0] = *__ptr;
+ return __vec;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_insert_and_zero(const float *__ptr) {
+ __vector float __vec = (__vector float)0.0f;
+ __vec[1] = *__ptr;
+ return __vec;
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_insert_and_zero(const double *__ptr) {
+ __vector double __vec = (__vector double)0.0;
+ __vec[0] = *__ptr;
+ return __vec;
+}
+
+/*-- vec_perm ---------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_perm(__vector signed char __a, __vector signed char __b,
+ __vector unsigned char __c) {
+ return (__vector signed char)__builtin_s390_vperm(
+ (__vector unsigned char)__a, (__vector unsigned char)__b, __c);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_perm(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c) {
+ return (__vector unsigned char)__builtin_s390_vperm(
+ (__vector unsigned char)__a, (__vector unsigned char)__b, __c);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_perm(__vector __bool char __a, __vector __bool char __b,
+ __vector unsigned char __c) {
+ return (__vector __bool char)__builtin_s390_vperm(
+ (__vector unsigned char)__a, (__vector unsigned char)__b, __c);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_perm(__vector signed short __a, __vector signed short __b,
+ __vector unsigned char __c) {
+ return (__vector signed short)__builtin_s390_vperm(
+ (__vector unsigned char)__a, (__vector unsigned char)__b, __c);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_perm(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned char __c) {
+ return (__vector unsigned short)__builtin_s390_vperm(
+ (__vector unsigned char)__a, (__vector unsigned char)__b, __c);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_perm(__vector __bool short __a, __vector __bool short __b,
+ __vector unsigned char __c) {
+ return (__vector __bool short)__builtin_s390_vperm(
+ (__vector unsigned char)__a, (__vector unsigned char)__b, __c);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_perm(__vector signed int __a, __vector signed int __b,
+ __vector unsigned char __c) {
+ return (__vector signed int)__builtin_s390_vperm(
+ (__vector unsigned char)__a, (__vector unsigned char)__b, __c);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_perm(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned char __c) {
+ return (__vector unsigned int)__builtin_s390_vperm(
+ (__vector unsigned char)__a, (__vector unsigned char)__b, __c);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_perm(__vector __bool int __a, __vector __bool int __b,
+ __vector unsigned char __c) {
+ return (__vector __bool int)__builtin_s390_vperm(
+ (__vector unsigned char)__a, (__vector unsigned char)__b, __c);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_perm(__vector signed long long __a, __vector signed long long __b,
+ __vector unsigned char __c) {
+ return (__vector signed long long)__builtin_s390_vperm(
+ (__vector unsigned char)__a, (__vector unsigned char)__b, __c);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_perm(__vector unsigned long long __a, __vector unsigned long long __b,
+ __vector unsigned char __c) {
+ return (__vector unsigned long long)__builtin_s390_vperm(
+ (__vector unsigned char)__a, (__vector unsigned char)__b, __c);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_perm(__vector __bool long long __a, __vector __bool long long __b,
+ __vector unsigned char __c) {
+ return (__vector __bool long long)__builtin_s390_vperm(
+ (__vector unsigned char)__a, (__vector unsigned char)__b, __c);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_perm(__vector float __a, __vector float __b,
+ __vector unsigned char __c) {
+ return (__vector float)__builtin_s390_vperm(
+ (__vector unsigned char)__a, (__vector unsigned char)__b, __c);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_perm(__vector double __a, __vector double __b,
+ __vector unsigned char __c) {
+ return (__vector double)__builtin_s390_vperm(
+ (__vector unsigned char)__a, (__vector unsigned char)__b, __c);
+}
+
+/*-- vec_permi --------------------------------------------------------------*/
+
+// This prototype is deprecated.
+extern __ATTRS_o __vector signed long long
+vec_permi(__vector signed long long __a, __vector signed long long __b,
+ int __c)
+ __constant_range(__c, 0, 3);
+
+// This prototype is deprecated.
+extern __ATTRS_o __vector unsigned long long
+vec_permi(__vector unsigned long long __a, __vector unsigned long long __b,
+ int __c)
+ __constant_range(__c, 0, 3);
+
+// This prototype is deprecated.
+extern __ATTRS_o __vector __bool long long
+vec_permi(__vector __bool long long __a, __vector __bool long long __b,
+ int __c)
+ __constant_range(__c, 0, 3);
+
+// This prototype is deprecated.
+extern __ATTRS_o __vector double
+vec_permi(__vector double __a, __vector double __b, int __c)
+ __constant_range(__c, 0, 3);
+
+#define vec_permi(X, Y, Z) ((__typeof__((vec_permi)((X), (Y), (Z)))) \
+ __builtin_s390_vpdi((__vector unsigned long long)(X), \
+ (__vector unsigned long long)(Y), \
+ (((Z) & 2) << 1) | ((Z) & 1)))
+
+/*-- vec_bperm_u128 ---------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_ai __vector unsigned long long
+vec_bperm_u128(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vbperm(__a, __b);
+}
+#endif
+
+/*-- vec_revb ---------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed short
+vec_revb(__vector signed short __vec) {
+ return (__vector signed short)
+ __builtin_s390_vlbrh((__vector unsigned short)__vec);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_revb(__vector unsigned short __vec) {
+ return __builtin_s390_vlbrh(__vec);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_revb(__vector signed int __vec) {
+ return (__vector signed int)
+ __builtin_s390_vlbrf((__vector unsigned int)__vec);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_revb(__vector unsigned int __vec) {
+ return __builtin_s390_vlbrf(__vec);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_revb(__vector signed long long __vec) {
+ return (__vector signed long long)
+ __builtin_s390_vlbrg((__vector unsigned long long)__vec);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_revb(__vector unsigned long long __vec) {
+ return __builtin_s390_vlbrg(__vec);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_revb(__vector float __vec) {
+ return (__vector float)
+ __builtin_s390_vlbrf((__vector unsigned int)__vec);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_revb(__vector double __vec) {
+ return (__vector double)
+ __builtin_s390_vlbrg((__vector unsigned long long)__vec);
+}
+
+/*-- vec_reve ---------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_reve(__vector signed char __vec) {
+ return (__vector signed char) { __vec[15], __vec[14], __vec[13], __vec[12],
+ __vec[11], __vec[10], __vec[9], __vec[8],
+ __vec[7], __vec[6], __vec[5], __vec[4],
+ __vec[3], __vec[2], __vec[1], __vec[0] };
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_reve(__vector unsigned char __vec) {
+ return (__vector unsigned char) { __vec[15], __vec[14], __vec[13], __vec[12],
+ __vec[11], __vec[10], __vec[9], __vec[8],
+ __vec[7], __vec[6], __vec[5], __vec[4],
+ __vec[3], __vec[2], __vec[1], __vec[0] };
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_reve(__vector __bool char __vec) {
+ return (__vector __bool char) { __vec[15], __vec[14], __vec[13], __vec[12],
+ __vec[11], __vec[10], __vec[9], __vec[8],
+ __vec[7], __vec[6], __vec[5], __vec[4],
+ __vec[3], __vec[2], __vec[1], __vec[0] };
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_reve(__vector signed short __vec) {
+ return (__vector signed short) { __vec[7], __vec[6], __vec[5], __vec[4],
+ __vec[3], __vec[2], __vec[1], __vec[0] };
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_reve(__vector unsigned short __vec) {
+ return (__vector unsigned short) { __vec[7], __vec[6], __vec[5], __vec[4],
+ __vec[3], __vec[2], __vec[1], __vec[0] };
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_reve(__vector __bool short __vec) {
+ return (__vector __bool short) { __vec[7], __vec[6], __vec[5], __vec[4],
+ __vec[3], __vec[2], __vec[1], __vec[0] };
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_reve(__vector signed int __vec) {
+ return (__vector signed int) { __vec[3], __vec[2], __vec[1], __vec[0] };
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_reve(__vector unsigned int __vec) {
+ return (__vector unsigned int) { __vec[3], __vec[2], __vec[1], __vec[0] };
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_reve(__vector __bool int __vec) {
+ return (__vector __bool int) { __vec[3], __vec[2], __vec[1], __vec[0] };
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_reve(__vector signed long long __vec) {
+ return (__vector signed long long) { __vec[1], __vec[0] };
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_reve(__vector unsigned long long __vec) {
+ return (__vector unsigned long long) { __vec[1], __vec[0] };
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_reve(__vector __bool long long __vec) {
+ return (__vector __bool long long) { __vec[1], __vec[0] };
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_reve(__vector float __vec) {
+ return (__vector float) { __vec[3], __vec[2], __vec[1], __vec[0] };
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_reve(__vector double __vec) {
+ return (__vector double) { __vec[1], __vec[0] };
+}
+
+/*-- vec_sel ----------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_sel(__vector signed char __a, __vector signed char __b,
+ __vector unsigned char __c) {
+ return (((__vector signed char)__c & __b) |
+ (~(__vector signed char)__c & __a));
+}
+
+static inline __ATTRS_o_ai __vector signed char
+vec_sel(__vector signed char __a, __vector signed char __b,
+ __vector __bool char __c) {
+ return (((__vector signed char)__c & __b) |
+ (~(__vector signed char)__c & __a));
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_sel(__vector __bool char __a, __vector __bool char __b,
+ __vector unsigned char __c) {
+ return (((__vector __bool char)__c & __b) |
+ (~(__vector __bool char)__c & __a));
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_sel(__vector __bool char __a, __vector __bool char __b,
+ __vector __bool char __c) {
+ return (__c & __b) | (~__c & __a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_sel(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c) {
+ return (__c & __b) | (~__c & __a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_sel(__vector unsigned char __a, __vector unsigned char __b,
+ __vector __bool char __c) {
+ return (((__vector unsigned char)__c & __b) |
+ (~(__vector unsigned char)__c & __a));
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_sel(__vector signed short __a, __vector signed short __b,
+ __vector unsigned short __c) {
+ return (((__vector signed short)__c & __b) |
+ (~(__vector signed short)__c & __a));
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_sel(__vector signed short __a, __vector signed short __b,
+ __vector __bool short __c) {
+ return (((__vector signed short)__c & __b) |
+ (~(__vector signed short)__c & __a));
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_sel(__vector __bool short __a, __vector __bool short __b,
+ __vector unsigned short __c) {
+ return (((__vector __bool short)__c & __b) |
+ (~(__vector __bool short)__c & __a));
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_sel(__vector __bool short __a, __vector __bool short __b,
+ __vector __bool short __c) {
+ return (__c & __b) | (~__c & __a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_sel(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned short __c) {
+ return (__c & __b) | (~__c & __a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_sel(__vector unsigned short __a, __vector unsigned short __b,
+ __vector __bool short __c) {
+ return (((__vector unsigned short)__c & __b) |
+ (~(__vector unsigned short)__c & __a));
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_sel(__vector signed int __a, __vector signed int __b,
+ __vector unsigned int __c) {
+ return (((__vector signed int)__c & __b) |
+ (~(__vector signed int)__c & __a));
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_sel(__vector signed int __a, __vector signed int __b,
+ __vector __bool int __c) {
+ return (((__vector signed int)__c & __b) |
+ (~(__vector signed int)__c & __a));
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_sel(__vector __bool int __a, __vector __bool int __b,
+ __vector unsigned int __c) {
+ return (((__vector __bool int)__c & __b) |
+ (~(__vector __bool int)__c & __a));
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_sel(__vector __bool int __a, __vector __bool int __b,
+ __vector __bool int __c) {
+ return (__c & __b) | (~__c & __a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_sel(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned int __c) {
+ return (__c & __b) | (~__c & __a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_sel(__vector unsigned int __a, __vector unsigned int __b,
+ __vector __bool int __c) {
+ return (((__vector unsigned int)__c & __b) |
+ (~(__vector unsigned int)__c & __a));
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_sel(__vector signed long long __a, __vector signed long long __b,
+ __vector unsigned long long __c) {
+ return (((__vector signed long long)__c & __b) |
+ (~(__vector signed long long)__c & __a));
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_sel(__vector signed long long __a, __vector signed long long __b,
+ __vector __bool long long __c) {
+ return (((__vector signed long long)__c & __b) |
+ (~(__vector signed long long)__c & __a));
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_sel(__vector __bool long long __a, __vector __bool long long __b,
+ __vector unsigned long long __c) {
+ return (((__vector __bool long long)__c & __b) |
+ (~(__vector __bool long long)__c & __a));
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_sel(__vector __bool long long __a, __vector __bool long long __b,
+ __vector __bool long long __c) {
+ return (__c & __b) | (~__c & __a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_sel(__vector unsigned long long __a, __vector unsigned long long __b,
+ __vector unsigned long long __c) {
+ return (__c & __b) | (~__c & __a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_sel(__vector unsigned long long __a, __vector unsigned long long __b,
+ __vector __bool long long __c) {
+ return (((__vector unsigned long long)__c & __b) |
+ (~(__vector unsigned long long)__c & __a));
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_sel(__vector float __a, __vector float __b, __vector unsigned int __c) {
+ return (__vector float)((__c & (__vector unsigned int)__b) |
+ (~__c & (__vector unsigned int)__a));
+}
+
+static inline __ATTRS_o_ai __vector float
+vec_sel(__vector float __a, __vector float __b, __vector __bool int __c) {
+ __vector unsigned int __ac = (__vector unsigned int)__a;
+ __vector unsigned int __bc = (__vector unsigned int)__b;
+ __vector unsigned int __cc = (__vector unsigned int)__c;
+ return (__vector float)((__cc & __bc) | (~__cc & __ac));
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_sel(__vector double __a, __vector double __b,
+ __vector unsigned long long __c) {
+ return (__vector double)((__c & (__vector unsigned long long)__b) |
+ (~__c & (__vector unsigned long long)__a));
+}
+
+static inline __ATTRS_o_ai __vector double
+vec_sel(__vector double __a, __vector double __b,
+ __vector __bool long long __c) {
+ __vector unsigned long long __ac = (__vector unsigned long long)__a;
+ __vector unsigned long long __bc = (__vector unsigned long long)__b;
+ __vector unsigned long long __cc = (__vector unsigned long long)__c;
+ return (__vector double)((__cc & __bc) | (~__cc & __ac));
+}
+
+/*-- vec_gather_element -----------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed int
+vec_gather_element(__vector signed int __vec,
+ __vector unsigned int __offset,
+ const signed int *__ptr, int __index)
+ __constant_range(__index, 0, 3) {
+ __vec[__index] = *(const signed int *)(
+ (const char *)__ptr + __offset[__index]);
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_gather_element(__vector __bool int __vec,
+ __vector unsigned int __offset,
+ const unsigned int *__ptr, int __index)
+ __constant_range(__index, 0, 3) {
+ __vec[__index] = *(const unsigned int *)(
+ (const char *)__ptr + __offset[__index]);
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_gather_element(__vector unsigned int __vec,
+ __vector unsigned int __offset,
+ const unsigned int *__ptr, int __index)
+ __constant_range(__index, 0, 3) {
+ __vec[__index] = *(const unsigned int *)(
+ (const char *)__ptr + __offset[__index]);
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_gather_element(__vector signed long long __vec,
+ __vector unsigned long long __offset,
+ const signed long long *__ptr, int __index)
+ __constant_range(__index, 0, 1) {
+ __vec[__index] = *(const signed long long *)(
+ (const char *)__ptr + __offset[__index]);
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_gather_element(__vector __bool long long __vec,
+ __vector unsigned long long __offset,
+ const unsigned long long *__ptr, int __index)
+ __constant_range(__index, 0, 1) {
+ __vec[__index] = *(const unsigned long long *)(
+ (const char *)__ptr + __offset[__index]);
+ return __vec;
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_gather_element(__vector unsigned long long __vec,
+ __vector unsigned long long __offset,
+ const unsigned long long *__ptr, int __index)
+ __constant_range(__index, 0, 1) {
+ __vec[__index] = *(const unsigned long long *)(
+ (const char *)__ptr + __offset[__index]);
+ return __vec;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_gather_element(__vector float __vec,
+ __vector unsigned int __offset,
+ const float *__ptr, int __index)
+ __constant_range(__index, 0, 3) {
+ __vec[__index] = *(const float *)(
+ (const char *)__ptr + __offset[__index]);
+ return __vec;
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_gather_element(__vector double __vec,
+ __vector unsigned long long __offset,
+ const double *__ptr, int __index)
+ __constant_range(__index, 0, 1) {
+ __vec[__index] = *(const double *)(
+ (const char *)__ptr + __offset[__index]);
+ return __vec;
+}
+
+/*-- vec_scatter_element ----------------------------------------------------*/
+
+static inline __ATTRS_o_ai void
+vec_scatter_element(__vector signed int __vec,
+ __vector unsigned int __offset,
+ signed int *__ptr, int __index)
+ __constant_range(__index, 0, 3) {
+ *(signed int *)((char *)__ptr + __offset[__index]) =
+ __vec[__index];
+}
+
+static inline __ATTRS_o_ai void
+vec_scatter_element(__vector __bool int __vec,
+ __vector unsigned int __offset,
+ unsigned int *__ptr, int __index)
+ __constant_range(__index, 0, 3) {
+ *(unsigned int *)((char *)__ptr + __offset[__index]) =
+ __vec[__index];
+}
+
+static inline __ATTRS_o_ai void
+vec_scatter_element(__vector unsigned int __vec,
+ __vector unsigned int __offset,
+ unsigned int *__ptr, int __index)
+ __constant_range(__index, 0, 3) {
+ *(unsigned int *)((char *)__ptr + __offset[__index]) =
+ __vec[__index];
+}
+
+static inline __ATTRS_o_ai void
+vec_scatter_element(__vector signed long long __vec,
+ __vector unsigned long long __offset,
+ signed long long *__ptr, int __index)
+ __constant_range(__index, 0, 1) {
+ *(signed long long *)((char *)__ptr + __offset[__index]) =
+ __vec[__index];
+}
+
+static inline __ATTRS_o_ai void
+vec_scatter_element(__vector __bool long long __vec,
+ __vector unsigned long long __offset,
+ unsigned long long *__ptr, int __index)
+ __constant_range(__index, 0, 1) {
+ *(unsigned long long *)((char *)__ptr + __offset[__index]) =
+ __vec[__index];
+}
+
+static inline __ATTRS_o_ai void
+vec_scatter_element(__vector unsigned long long __vec,
+ __vector unsigned long long __offset,
+ unsigned long long *__ptr, int __index)
+ __constant_range(__index, 0, 1) {
+ *(unsigned long long *)((char *)__ptr + __offset[__index]) =
+ __vec[__index];
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai void
+vec_scatter_element(__vector float __vec,
+ __vector unsigned int __offset,
+ float *__ptr, int __index)
+ __constant_range(__index, 0, 3) {
+ *(float *)((char *)__ptr + __offset[__index]) =
+ __vec[__index];
+}
+#endif
+
+static inline __ATTRS_o_ai void
+vec_scatter_element(__vector double __vec,
+ __vector unsigned long long __offset,
+ double *__ptr, int __index)
+ __constant_range(__index, 0, 1) {
+ *(double *)((char *)__ptr + __offset[__index]) =
+ __vec[__index];
+}
+
+/*-- vec_xl -----------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_xl(long __offset, const signed char *__ptr) {
+ return *(const __vector signed char *)
+ ((const char *)__ptr + __offset);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_xl(long __offset, const unsigned char *__ptr) {
+ return *(const __vector unsigned char *)
+ ((const char *)__ptr + __offset);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_xl(long __offset, const signed short *__ptr) {
+ return *(const __vector signed short *)
+ ((const char *)__ptr + __offset);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_xl(long __offset, const unsigned short *__ptr) {
+ return *(const __vector unsigned short *)
+ ((const char *)__ptr + __offset);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_xl(long __offset, const signed int *__ptr) {
+ return *(const __vector signed int *)
+ ((const char *)__ptr + __offset);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_xl(long __offset, const unsigned int *__ptr) {
+ return *(const __vector unsigned int *)
+ ((const char *)__ptr + __offset);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_xl(long __offset, const signed long long *__ptr) {
+ return *(const __vector signed long long *)
+ ((const char *)__ptr + __offset);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_xl(long __offset, const unsigned long long *__ptr) {
+ return *(const __vector unsigned long long *)
+ ((const char *)__ptr + __offset);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_xl(long __offset, const float *__ptr) {
+ return *(const __vector float *)
+ ((const char *)__ptr + __offset);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_xl(long __offset, const double *__ptr) {
+ return *(const __vector double *)
+ ((const char *)__ptr + __offset);
+}
+
+/*-- vec_xld2 ---------------------------------------------------------------*/
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_xld2(long __offset, const signed char *__ptr) {
+ return *(const __vector signed char *)
+ ((const char *)__ptr + __offset);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_xld2(long __offset, const unsigned char *__ptr) {
+ return *(const __vector unsigned char *)
+ ((const char *)__ptr + __offset);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_xld2(long __offset, const signed short *__ptr) {
+ return *(const __vector signed short *)
+ ((const char *)__ptr + __offset);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_xld2(long __offset, const unsigned short *__ptr) {
+ return *(const __vector unsigned short *)
+ ((const char *)__ptr + __offset);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_xld2(long __offset, const signed int *__ptr) {
+ return *(const __vector signed int *)
+ ((const char *)__ptr + __offset);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_xld2(long __offset, const unsigned int *__ptr) {
+ return *(const __vector unsigned int *)
+ ((const char *)__ptr + __offset);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_xld2(long __offset, const signed long long *__ptr) {
+ return *(const __vector signed long long *)
+ ((const char *)__ptr + __offset);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_xld2(long __offset, const unsigned long long *__ptr) {
+ return *(const __vector unsigned long long *)
+ ((const char *)__ptr + __offset);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector double
+vec_xld2(long __offset, const double *__ptr) {
+ return *(const __vector double *)
+ ((const char *)__ptr + __offset);
+}
+
+/*-- vec_xlw4 ---------------------------------------------------------------*/
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_xlw4(long __offset, const signed char *__ptr) {
+ return *(const __vector signed char *)
+ ((const char *)__ptr + __offset);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_xlw4(long __offset, const unsigned char *__ptr) {
+ return *(const __vector unsigned char *)
+ ((const char *)__ptr + __offset);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_xlw4(long __offset, const signed short *__ptr) {
+ return *(const __vector signed short *)
+ ((const char *)__ptr + __offset);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_xlw4(long __offset, const unsigned short *__ptr) {
+ return *(const __vector unsigned short *)
+ ((const char *)__ptr + __offset);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_xlw4(long __offset, const signed int *__ptr) {
+ return *(const __vector signed int *)
+ ((const char *)__ptr + __offset);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_xlw4(long __offset, const unsigned int *__ptr) {
+ return *(const __vector unsigned int *)
+ ((const char *)__ptr + __offset);
+}
+
+/*-- vec_xst ----------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai void
+vec_xst(__vector signed char __vec, long __offset, signed char *__ptr) {
+ *(__vector signed char *)((char *)__ptr + __offset) = __vec;
+}
+
+static inline __ATTRS_o_ai void
+vec_xst(__vector unsigned char __vec, long __offset, unsigned char *__ptr) {
+ *(__vector unsigned char *)((char *)__ptr + __offset) = __vec;
+}
+
+static inline __ATTRS_o_ai void
+vec_xst(__vector signed short __vec, long __offset, signed short *__ptr) {
+ *(__vector signed short *)((char *)__ptr + __offset) = __vec;
+}
+
+static inline __ATTRS_o_ai void
+vec_xst(__vector unsigned short __vec, long __offset, unsigned short *__ptr) {
+ *(__vector unsigned short *)((char *)__ptr + __offset) = __vec;
+}
+
+static inline __ATTRS_o_ai void
+vec_xst(__vector signed int __vec, long __offset, signed int *__ptr) {
+ *(__vector signed int *)((char *)__ptr + __offset) = __vec;
+}
+
+static inline __ATTRS_o_ai void
+vec_xst(__vector unsigned int __vec, long __offset, unsigned int *__ptr) {
+ *(__vector unsigned int *)((char *)__ptr + __offset) = __vec;
+}
+
+static inline __ATTRS_o_ai void
+vec_xst(__vector signed long long __vec, long __offset,
+ signed long long *__ptr) {
+ *(__vector signed long long *)((char *)__ptr + __offset) = __vec;
+}
+
+static inline __ATTRS_o_ai void
+vec_xst(__vector unsigned long long __vec, long __offset,
+ unsigned long long *__ptr) {
+ *(__vector unsigned long long *)((char *)__ptr + __offset) = __vec;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai void
+vec_xst(__vector float __vec, long __offset, float *__ptr) {
+ *(__vector float *)((char *)__ptr + __offset) = __vec;
+}
+#endif
+
+static inline __ATTRS_o_ai void
+vec_xst(__vector double __vec, long __offset, double *__ptr) {
+ *(__vector double *)((char *)__ptr + __offset) = __vec;
+}
+
+/*-- vec_xstd2 --------------------------------------------------------------*/
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai void
+vec_xstd2(__vector signed char __vec, long __offset, signed char *__ptr) {
+ *(__vector signed char *)((char *)__ptr + __offset) = __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai void
+vec_xstd2(__vector unsigned char __vec, long __offset, unsigned char *__ptr) {
+ *(__vector unsigned char *)((char *)__ptr + __offset) = __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai void
+vec_xstd2(__vector signed short __vec, long __offset, signed short *__ptr) {
+ *(__vector signed short *)((char *)__ptr + __offset) = __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai void
+vec_xstd2(__vector unsigned short __vec, long __offset, unsigned short *__ptr) {
+ *(__vector unsigned short *)((char *)__ptr + __offset) = __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai void
+vec_xstd2(__vector signed int __vec, long __offset, signed int *__ptr) {
+ *(__vector signed int *)((char *)__ptr + __offset) = __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai void
+vec_xstd2(__vector unsigned int __vec, long __offset, unsigned int *__ptr) {
+ *(__vector unsigned int *)((char *)__ptr + __offset) = __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai void
+vec_xstd2(__vector signed long long __vec, long __offset,
+ signed long long *__ptr) {
+ *(__vector signed long long *)((char *)__ptr + __offset) = __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai void
+vec_xstd2(__vector unsigned long long __vec, long __offset,
+ unsigned long long *__ptr) {
+ *(__vector unsigned long long *)((char *)__ptr + __offset) = __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai void
+vec_xstd2(__vector double __vec, long __offset, double *__ptr) {
+ *(__vector double *)((char *)__ptr + __offset) = __vec;
+}
+
+/*-- vec_xstw4 --------------------------------------------------------------*/
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai void
+vec_xstw4(__vector signed char __vec, long __offset, signed char *__ptr) {
+ *(__vector signed char *)((char *)__ptr + __offset) = __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai void
+vec_xstw4(__vector unsigned char __vec, long __offset, unsigned char *__ptr) {
+ *(__vector unsigned char *)((char *)__ptr + __offset) = __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai void
+vec_xstw4(__vector signed short __vec, long __offset, signed short *__ptr) {
+ *(__vector signed short *)((char *)__ptr + __offset) = __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai void
+vec_xstw4(__vector unsigned short __vec, long __offset, unsigned short *__ptr) {
+ *(__vector unsigned short *)((char *)__ptr + __offset) = __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai void
+vec_xstw4(__vector signed int __vec, long __offset, signed int *__ptr) {
+ *(__vector signed int *)((char *)__ptr + __offset) = __vec;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai void
+vec_xstw4(__vector unsigned int __vec, long __offset, unsigned int *__ptr) {
+ *(__vector unsigned int *)((char *)__ptr + __offset) = __vec;
+}
+
+/*-- vec_load_bndry ---------------------------------------------------------*/
+
+extern __ATTRS_o __vector signed char
+vec_load_bndry(const signed char *__ptr, unsigned short __len)
+ __constant_pow2_range(__len, 64, 4096);
+
+extern __ATTRS_o __vector unsigned char
+vec_load_bndry(const unsigned char *__ptr, unsigned short __len)
+ __constant_pow2_range(__len, 64, 4096);
+
+extern __ATTRS_o __vector signed short
+vec_load_bndry(const signed short *__ptr, unsigned short __len)
+ __constant_pow2_range(__len, 64, 4096);
+
+extern __ATTRS_o __vector unsigned short
+vec_load_bndry(const unsigned short *__ptr, unsigned short __len)
+ __constant_pow2_range(__len, 64, 4096);
+
+extern __ATTRS_o __vector signed int
+vec_load_bndry(const signed int *__ptr, unsigned short __len)
+ __constant_pow2_range(__len, 64, 4096);
+
+extern __ATTRS_o __vector unsigned int
+vec_load_bndry(const unsigned int *__ptr, unsigned short __len)
+ __constant_pow2_range(__len, 64, 4096);
+
+extern __ATTRS_o __vector signed long long
+vec_load_bndry(const signed long long *__ptr, unsigned short __len)
+ __constant_pow2_range(__len, 64, 4096);
+
+extern __ATTRS_o __vector unsigned long long
+vec_load_bndry(const unsigned long long *__ptr, unsigned short __len)
+ __constant_pow2_range(__len, 64, 4096);
+
+#if __ARCH__ >= 12
+extern __ATTRS_o __vector float
+vec_load_bndry(const float *__ptr, unsigned short __len)
+ __constant_pow2_range(__len, 64, 4096);
+#endif
+
+extern __ATTRS_o __vector double
+vec_load_bndry(const double *__ptr, unsigned short __len)
+ __constant_pow2_range(__len, 64, 4096);
+
+#define vec_load_bndry(X, Y) ((__typeof__((vec_load_bndry)((X), (Y)))) \
+ __builtin_s390_vlbb((X), ((Y) == 64 ? 0 : \
+ (Y) == 128 ? 1 : \
+ (Y) == 256 ? 2 : \
+ (Y) == 512 ? 3 : \
+ (Y) == 1024 ? 4 : \
+ (Y) == 2048 ? 5 : \
+ (Y) == 4096 ? 6 : -1)))
+
+/*-- vec_load_len -----------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_load_len(const signed char *__ptr, unsigned int __len) {
+ return (__vector signed char)__builtin_s390_vll(__len, __ptr);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_load_len(const unsigned char *__ptr, unsigned int __len) {
+ return (__vector unsigned char)__builtin_s390_vll(__len, __ptr);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_load_len(const signed short *__ptr, unsigned int __len) {
+ return (__vector signed short)__builtin_s390_vll(__len, __ptr);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_load_len(const unsigned short *__ptr, unsigned int __len) {
+ return (__vector unsigned short)__builtin_s390_vll(__len, __ptr);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_load_len(const signed int *__ptr, unsigned int __len) {
+ return (__vector signed int)__builtin_s390_vll(__len, __ptr);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_load_len(const unsigned int *__ptr, unsigned int __len) {
+ return (__vector unsigned int)__builtin_s390_vll(__len, __ptr);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_load_len(const signed long long *__ptr, unsigned int __len) {
+ return (__vector signed long long)__builtin_s390_vll(__len, __ptr);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_load_len(const unsigned long long *__ptr, unsigned int __len) {
+ return (__vector unsigned long long)__builtin_s390_vll(__len, __ptr);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_load_len(const float *__ptr, unsigned int __len) {
+ return (__vector float)__builtin_s390_vll(__len, __ptr);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_load_len(const double *__ptr, unsigned int __len) {
+ return (__vector double)__builtin_s390_vll(__len, __ptr);
+}
+
+/*-- vec_load_len_r ---------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_ai __vector unsigned char
+vec_load_len_r(const unsigned char *__ptr, unsigned int __len) {
+ return (__vector unsigned char)__builtin_s390_vlrl(__len, __ptr);
+}
+#endif
+
+/*-- vec_store_len ----------------------------------------------------------*/
+
+static inline __ATTRS_o_ai void
+vec_store_len(__vector signed char __vec, signed char *__ptr,
+ unsigned int __len) {
+ __builtin_s390_vstl((__vector signed char)__vec, __len, __ptr);
+}
+
+static inline __ATTRS_o_ai void
+vec_store_len(__vector unsigned char __vec, unsigned char *__ptr,
+ unsigned int __len) {
+ __builtin_s390_vstl((__vector signed char)__vec, __len, __ptr);
+}
+
+static inline __ATTRS_o_ai void
+vec_store_len(__vector signed short __vec, signed short *__ptr,
+ unsigned int __len) {
+ __builtin_s390_vstl((__vector signed char)__vec, __len, __ptr);
+}
+
+static inline __ATTRS_o_ai void
+vec_store_len(__vector unsigned short __vec, unsigned short *__ptr,
+ unsigned int __len) {
+ __builtin_s390_vstl((__vector signed char)__vec, __len, __ptr);
+}
+
+static inline __ATTRS_o_ai void
+vec_store_len(__vector signed int __vec, signed int *__ptr,
+ unsigned int __len) {
+ __builtin_s390_vstl((__vector signed char)__vec, __len, __ptr);
+}
+
+static inline __ATTRS_o_ai void
+vec_store_len(__vector unsigned int __vec, unsigned int *__ptr,
+ unsigned int __len) {
+ __builtin_s390_vstl((__vector signed char)__vec, __len, __ptr);
+}
+
+static inline __ATTRS_o_ai void
+vec_store_len(__vector signed long long __vec, signed long long *__ptr,
+ unsigned int __len) {
+ __builtin_s390_vstl((__vector signed char)__vec, __len, __ptr);
+}
+
+static inline __ATTRS_o_ai void
+vec_store_len(__vector unsigned long long __vec, unsigned long long *__ptr,
+ unsigned int __len) {
+ __builtin_s390_vstl((__vector signed char)__vec, __len, __ptr);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai void
+vec_store_len(__vector float __vec, float *__ptr,
+ unsigned int __len) {
+ __builtin_s390_vstl((__vector signed char)__vec, __len, __ptr);
+}
+#endif
+
+static inline __ATTRS_o_ai void
+vec_store_len(__vector double __vec, double *__ptr,
+ unsigned int __len) {
+ __builtin_s390_vstl((__vector signed char)__vec, __len, __ptr);
+}
+
+/*-- vec_store_len_r --------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_ai void
+vec_store_len_r(__vector unsigned char __vec, unsigned char *__ptr,
+ unsigned int __len) {
+ __builtin_s390_vstrl((__vector signed char)__vec, __len, __ptr);
+}
+#endif
+
+/*-- vec_load_pair ----------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_load_pair(signed long long __a, signed long long __b) {
+ return (__vector signed long long)(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_load_pair(unsigned long long __a, unsigned long long __b) {
+ return (__vector unsigned long long)(__a, __b);
+}
+
+/*-- vec_genmask ------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_genmask(unsigned short __mask)
+ __constant(__mask) {
+ return (__vector unsigned char)(
+ __mask & 0x8000 ? 0xff : 0,
+ __mask & 0x4000 ? 0xff : 0,
+ __mask & 0x2000 ? 0xff : 0,
+ __mask & 0x1000 ? 0xff : 0,
+ __mask & 0x0800 ? 0xff : 0,
+ __mask & 0x0400 ? 0xff : 0,
+ __mask & 0x0200 ? 0xff : 0,
+ __mask & 0x0100 ? 0xff : 0,
+ __mask & 0x0080 ? 0xff : 0,
+ __mask & 0x0040 ? 0xff : 0,
+ __mask & 0x0020 ? 0xff : 0,
+ __mask & 0x0010 ? 0xff : 0,
+ __mask & 0x0008 ? 0xff : 0,
+ __mask & 0x0004 ? 0xff : 0,
+ __mask & 0x0002 ? 0xff : 0,
+ __mask & 0x0001 ? 0xff : 0);
+}
+
+/*-- vec_genmasks_* ---------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_genmasks_8(unsigned char __first, unsigned char __last)
+ __constant(__first) __constant(__last) {
+ unsigned char __bit1 = __first & 7;
+ unsigned char __bit2 = __last & 7;
+ unsigned char __mask1 = (unsigned char)(1U << (7 - __bit1) << 1) - 1;
+ unsigned char __mask2 = (unsigned char)(1U << (7 - __bit2)) - 1;
+ unsigned char __value = (__bit1 <= __bit2 ?
+ __mask1 & ~__mask2 :
+ __mask1 | ~__mask2);
+ return (__vector unsigned char)__value;
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_genmasks_16(unsigned char __first, unsigned char __last)
+ __constant(__first) __constant(__last) {
+ unsigned char __bit1 = __first & 15;
+ unsigned char __bit2 = __last & 15;
+ unsigned short __mask1 = (unsigned short)(1U << (15 - __bit1) << 1) - 1;
+ unsigned short __mask2 = (unsigned short)(1U << (15 - __bit2)) - 1;
+ unsigned short __value = (__bit1 <= __bit2 ?
+ __mask1 & ~__mask2 :
+ __mask1 | ~__mask2);
+ return (__vector unsigned short)__value;
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_genmasks_32(unsigned char __first, unsigned char __last)
+ __constant(__first) __constant(__last) {
+ unsigned char __bit1 = __first & 31;
+ unsigned char __bit2 = __last & 31;
+ unsigned int __mask1 = (1U << (31 - __bit1) << 1) - 1;
+ unsigned int __mask2 = (1U << (31 - __bit2)) - 1;
+ unsigned int __value = (__bit1 <= __bit2 ?
+ __mask1 & ~__mask2 :
+ __mask1 | ~__mask2);
+ return (__vector unsigned int)__value;
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_genmasks_64(unsigned char __first, unsigned char __last)
+ __constant(__first) __constant(__last) {
+ unsigned char __bit1 = __first & 63;
+ unsigned char __bit2 = __last & 63;
+ unsigned long long __mask1 = (1ULL << (63 - __bit1) << 1) - 1;
+ unsigned long long __mask2 = (1ULL << (63 - __bit2)) - 1;
+ unsigned long long __value = (__bit1 <= __bit2 ?
+ __mask1 & ~__mask2 :
+ __mask1 | ~__mask2);
+ return (__vector unsigned long long)__value;
+}
+
+/*-- vec_splat --------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_splat(__vector signed char __vec, int __index)
+ __constant_range(__index, 0, 15) {
+ return (__vector signed char)__vec[__index];
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_splat(__vector __bool char __vec, int __index)
+ __constant_range(__index, 0, 15) {
+ return (__vector __bool char)(__vector unsigned char)__vec[__index];
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_splat(__vector unsigned char __vec, int __index)
+ __constant_range(__index, 0, 15) {
+ return (__vector unsigned char)__vec[__index];
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_splat(__vector signed short __vec, int __index)
+ __constant_range(__index, 0, 7) {
+ return (__vector signed short)__vec[__index];
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_splat(__vector __bool short __vec, int __index)
+ __constant_range(__index, 0, 7) {
+ return (__vector __bool short)(__vector unsigned short)__vec[__index];
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_splat(__vector unsigned short __vec, int __index)
+ __constant_range(__index, 0, 7) {
+ return (__vector unsigned short)__vec[__index];
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_splat(__vector signed int __vec, int __index)
+ __constant_range(__index, 0, 3) {
+ return (__vector signed int)__vec[__index];
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_splat(__vector __bool int __vec, int __index)
+ __constant_range(__index, 0, 3) {
+ return (__vector __bool int)(__vector unsigned int)__vec[__index];
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_splat(__vector unsigned int __vec, int __index)
+ __constant_range(__index, 0, 3) {
+ return (__vector unsigned int)__vec[__index];
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_splat(__vector signed long long __vec, int __index)
+ __constant_range(__index, 0, 1) {
+ return (__vector signed long long)__vec[__index];
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_splat(__vector __bool long long __vec, int __index)
+ __constant_range(__index, 0, 1) {
+ return ((__vector __bool long long)
+ (__vector unsigned long long)__vec[__index]);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_splat(__vector unsigned long long __vec, int __index)
+ __constant_range(__index, 0, 1) {
+ return (__vector unsigned long long)__vec[__index];
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_splat(__vector float __vec, int __index)
+ __constant_range(__index, 0, 3) {
+ return (__vector float)__vec[__index];
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_splat(__vector double __vec, int __index)
+ __constant_range(__index, 0, 1) {
+ return (__vector double)__vec[__index];
+}
+
+/*-- vec_splat_s* -----------------------------------------------------------*/
+
+static inline __ATTRS_ai __vector signed char
+vec_splat_s8(signed char __scalar)
+ __constant(__scalar) {
+ return (__vector signed char)__scalar;
+}
+
+static inline __ATTRS_ai __vector signed short
+vec_splat_s16(signed short __scalar)
+ __constant(__scalar) {
+ return (__vector signed short)__scalar;
+}
+
+static inline __ATTRS_ai __vector signed int
+vec_splat_s32(signed short __scalar)
+ __constant(__scalar) {
+ return (__vector signed int)(signed int)__scalar;
+}
+
+static inline __ATTRS_ai __vector signed long long
+vec_splat_s64(signed short __scalar)
+ __constant(__scalar) {
+ return (__vector signed long long)(signed long)__scalar;
+}
+
+/*-- vec_splat_u* -----------------------------------------------------------*/
+
+static inline __ATTRS_ai __vector unsigned char
+vec_splat_u8(unsigned char __scalar)
+ __constant(__scalar) {
+ return (__vector unsigned char)__scalar;
+}
+
+static inline __ATTRS_ai __vector unsigned short
+vec_splat_u16(unsigned short __scalar)
+ __constant(__scalar) {
+ return (__vector unsigned short)__scalar;
+}
+
+static inline __ATTRS_ai __vector unsigned int
+vec_splat_u32(signed short __scalar)
+ __constant(__scalar) {
+ return (__vector unsigned int)(signed int)__scalar;
+}
+
+static inline __ATTRS_ai __vector unsigned long long
+vec_splat_u64(signed short __scalar)
+ __constant(__scalar) {
+ return (__vector unsigned long long)(signed long long)__scalar;
+}
+
+/*-- vec_splats -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_splats(signed char __scalar) {
+ return (__vector signed char)__scalar;
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_splats(unsigned char __scalar) {
+ return (__vector unsigned char)__scalar;
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_splats(signed short __scalar) {
+ return (__vector signed short)__scalar;
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_splats(unsigned short __scalar) {
+ return (__vector unsigned short)__scalar;
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_splats(signed int __scalar) {
+ return (__vector signed int)__scalar;
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_splats(unsigned int __scalar) {
+ return (__vector unsigned int)__scalar;
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_splats(signed long long __scalar) {
+ return (__vector signed long long)__scalar;
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_splats(unsigned long long __scalar) {
+ return (__vector unsigned long long)__scalar;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_splats(float __scalar) {
+ return (__vector float)__scalar;
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_splats(double __scalar) {
+ return (__vector double)__scalar;
+}
+
+/*-- vec_extend_s64 ---------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_extend_s64(__vector signed char __a) {
+ return (__vector signed long long)(__a[7], __a[15]);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_extend_s64(__vector signed short __a) {
+ return (__vector signed long long)(__a[3], __a[7]);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_extend_s64(__vector signed int __a) {
+ return (__vector signed long long)(__a[1], __a[3]);
+}
+
+/*-- vec_mergeh -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_mergeh(__vector signed char __a, __vector signed char __b) {
+ return (__vector signed char)(
+ __a[0], __b[0], __a[1], __b[1], __a[2], __b[2], __a[3], __b[3],
+ __a[4], __b[4], __a[5], __b[5], __a[6], __b[6], __a[7], __b[7]);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_mergeh(__vector __bool char __a, __vector __bool char __b) {
+ return (__vector __bool char)(
+ __a[0], __b[0], __a[1], __b[1], __a[2], __b[2], __a[3], __b[3],
+ __a[4], __b[4], __a[5], __b[5], __a[6], __b[6], __a[7], __b[7]);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_mergeh(__vector unsigned char __a, __vector unsigned char __b) {
+ return (__vector unsigned char)(
+ __a[0], __b[0], __a[1], __b[1], __a[2], __b[2], __a[3], __b[3],
+ __a[4], __b[4], __a[5], __b[5], __a[6], __b[6], __a[7], __b[7]);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_mergeh(__vector signed short __a, __vector signed short __b) {
+ return (__vector signed short)(
+ __a[0], __b[0], __a[1], __b[1], __a[2], __b[2], __a[3], __b[3]);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_mergeh(__vector __bool short __a, __vector __bool short __b) {
+ return (__vector __bool short)(
+ __a[0], __b[0], __a[1], __b[1], __a[2], __b[2], __a[3], __b[3]);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_mergeh(__vector unsigned short __a, __vector unsigned short __b) {
+ return (__vector unsigned short)(
+ __a[0], __b[0], __a[1], __b[1], __a[2], __b[2], __a[3], __b[3]);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_mergeh(__vector signed int __a, __vector signed int __b) {
+ return (__vector signed int)(__a[0], __b[0], __a[1], __b[1]);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_mergeh(__vector __bool int __a, __vector __bool int __b) {
+ return (__vector __bool int)(__a[0], __b[0], __a[1], __b[1]);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_mergeh(__vector unsigned int __a, __vector unsigned int __b) {
+ return (__vector unsigned int)(__a[0], __b[0], __a[1], __b[1]);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_mergeh(__vector signed long long __a, __vector signed long long __b) {
+ return (__vector signed long long)(__a[0], __b[0]);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_mergeh(__vector __bool long long __a, __vector __bool long long __b) {
+ return (__vector __bool long long)(__a[0], __b[0]);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_mergeh(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return (__vector unsigned long long)(__a[0], __b[0]);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_mergeh(__vector float __a, __vector float __b) {
+ return (__vector float)(__a[0], __b[0], __a[1], __b[1]);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_mergeh(__vector double __a, __vector double __b) {
+ return (__vector double)(__a[0], __b[0]);
+}
+
+/*-- vec_mergel -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_mergel(__vector signed char __a, __vector signed char __b) {
+ return (__vector signed char)(
+ __a[8], __b[8], __a[9], __b[9], __a[10], __b[10], __a[11], __b[11],
+ __a[12], __b[12], __a[13], __b[13], __a[14], __b[14], __a[15], __b[15]);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_mergel(__vector __bool char __a, __vector __bool char __b) {
+ return (__vector __bool char)(
+ __a[8], __b[8], __a[9], __b[9], __a[10], __b[10], __a[11], __b[11],
+ __a[12], __b[12], __a[13], __b[13], __a[14], __b[14], __a[15], __b[15]);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_mergel(__vector unsigned char __a, __vector unsigned char __b) {
+ return (__vector unsigned char)(
+ __a[8], __b[8], __a[9], __b[9], __a[10], __b[10], __a[11], __b[11],
+ __a[12], __b[12], __a[13], __b[13], __a[14], __b[14], __a[15], __b[15]);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_mergel(__vector signed short __a, __vector signed short __b) {
+ return (__vector signed short)(
+ __a[4], __b[4], __a[5], __b[5], __a[6], __b[6], __a[7], __b[7]);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_mergel(__vector __bool short __a, __vector __bool short __b) {
+ return (__vector __bool short)(
+ __a[4], __b[4], __a[5], __b[5], __a[6], __b[6], __a[7], __b[7]);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_mergel(__vector unsigned short __a, __vector unsigned short __b) {
+ return (__vector unsigned short)(
+ __a[4], __b[4], __a[5], __b[5], __a[6], __b[6], __a[7], __b[7]);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_mergel(__vector signed int __a, __vector signed int __b) {
+ return (__vector signed int)(__a[2], __b[2], __a[3], __b[3]);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_mergel(__vector __bool int __a, __vector __bool int __b) {
+ return (__vector __bool int)(__a[2], __b[2], __a[3], __b[3]);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_mergel(__vector unsigned int __a, __vector unsigned int __b) {
+ return (__vector unsigned int)(__a[2], __b[2], __a[3], __b[3]);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_mergel(__vector signed long long __a, __vector signed long long __b) {
+ return (__vector signed long long)(__a[1], __b[1]);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_mergel(__vector __bool long long __a, __vector __bool long long __b) {
+ return (__vector __bool long long)(__a[1], __b[1]);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_mergel(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return (__vector unsigned long long)(__a[1], __b[1]);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_mergel(__vector float __a, __vector float __b) {
+ return (__vector float)(__a[2], __b[2], __a[3], __b[3]);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_mergel(__vector double __a, __vector double __b) {
+ return (__vector double)(__a[1], __b[1]);
+}
+
+/*-- vec_pack ---------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_pack(__vector signed short __a, __vector signed short __b) {
+ __vector signed char __ac = (__vector signed char)__a;
+ __vector signed char __bc = (__vector signed char)__b;
+ return (__vector signed char)(
+ __ac[1], __ac[3], __ac[5], __ac[7], __ac[9], __ac[11], __ac[13], __ac[15],
+ __bc[1], __bc[3], __bc[5], __bc[7], __bc[9], __bc[11], __bc[13], __bc[15]);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_pack(__vector __bool short __a, __vector __bool short __b) {
+ __vector __bool char __ac = (__vector __bool char)__a;
+ __vector __bool char __bc = (__vector __bool char)__b;
+ return (__vector __bool char)(
+ __ac[1], __ac[3], __ac[5], __ac[7], __ac[9], __ac[11], __ac[13], __ac[15],
+ __bc[1], __bc[3], __bc[5], __bc[7], __bc[9], __bc[11], __bc[13], __bc[15]);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_pack(__vector unsigned short __a, __vector unsigned short __b) {
+ __vector unsigned char __ac = (__vector unsigned char)__a;
+ __vector unsigned char __bc = (__vector unsigned char)__b;
+ return (__vector unsigned char)(
+ __ac[1], __ac[3], __ac[5], __ac[7], __ac[9], __ac[11], __ac[13], __ac[15],
+ __bc[1], __bc[3], __bc[5], __bc[7], __bc[9], __bc[11], __bc[13], __bc[15]);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_pack(__vector signed int __a, __vector signed int __b) {
+ __vector signed short __ac = (__vector signed short)__a;
+ __vector signed short __bc = (__vector signed short)__b;
+ return (__vector signed short)(
+ __ac[1], __ac[3], __ac[5], __ac[7],
+ __bc[1], __bc[3], __bc[5], __bc[7]);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_pack(__vector __bool int __a, __vector __bool int __b) {
+ __vector __bool short __ac = (__vector __bool short)__a;
+ __vector __bool short __bc = (__vector __bool short)__b;
+ return (__vector __bool short)(
+ __ac[1], __ac[3], __ac[5], __ac[7],
+ __bc[1], __bc[3], __bc[5], __bc[7]);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_pack(__vector unsigned int __a, __vector unsigned int __b) {
+ __vector unsigned short __ac = (__vector unsigned short)__a;
+ __vector unsigned short __bc = (__vector unsigned short)__b;
+ return (__vector unsigned short)(
+ __ac[1], __ac[3], __ac[5], __ac[7],
+ __bc[1], __bc[3], __bc[5], __bc[7]);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_pack(__vector signed long long __a, __vector signed long long __b) {
+ __vector signed int __ac = (__vector signed int)__a;
+ __vector signed int __bc = (__vector signed int)__b;
+ return (__vector signed int)(__ac[1], __ac[3], __bc[1], __bc[3]);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_pack(__vector __bool long long __a, __vector __bool long long __b) {
+ __vector __bool int __ac = (__vector __bool int)__a;
+ __vector __bool int __bc = (__vector __bool int)__b;
+ return (__vector __bool int)(__ac[1], __ac[3], __bc[1], __bc[3]);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_pack(__vector unsigned long long __a, __vector unsigned long long __b) {
+ __vector unsigned int __ac = (__vector unsigned int)__a;
+ __vector unsigned int __bc = (__vector unsigned int)__b;
+ return (__vector unsigned int)(__ac[1], __ac[3], __bc[1], __bc[3]);
+}
+
+/*-- vec_packs --------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_packs(__vector signed short __a, __vector signed short __b) {
+ return __builtin_s390_vpksh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_packs(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vpklsh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_packs(__vector signed int __a, __vector signed int __b) {
+ return __builtin_s390_vpksf(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_packs(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vpklsf(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_packs(__vector signed long long __a, __vector signed long long __b) {
+ return __builtin_s390_vpksg(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_packs(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return __builtin_s390_vpklsg(__a, __b);
+}
+
+/*-- vec_packs_cc -----------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_packs_cc(__vector signed short __a, __vector signed short __b, int *__cc) {
+ return __builtin_s390_vpkshs(__a, __b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_packs_cc(__vector unsigned short __a, __vector unsigned short __b,
+ int *__cc) {
+ return __builtin_s390_vpklshs(__a, __b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_packs_cc(__vector signed int __a, __vector signed int __b, int *__cc) {
+ return __builtin_s390_vpksfs(__a, __b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_packs_cc(__vector unsigned int __a, __vector unsigned int __b, int *__cc) {
+ return __builtin_s390_vpklsfs(__a, __b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_packs_cc(__vector signed long long __a, __vector signed long long __b,
+ int *__cc) {
+ return __builtin_s390_vpksgs(__a, __b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_packs_cc(__vector unsigned long long __a, __vector unsigned long long __b,
+ int *__cc) {
+ return __builtin_s390_vpklsgs(__a, __b, __cc);
+}
+
+/*-- vec_packsu -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_packsu(__vector signed short __a, __vector signed short __b) {
+ const __vector signed short __zero = (__vector signed short)0;
+ return __builtin_s390_vpklsh(
+ (__vector unsigned short)(__a >= __zero) & (__vector unsigned short)__a,
+ (__vector unsigned short)(__b >= __zero) & (__vector unsigned short)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_packsu(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vpklsh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_packsu(__vector signed int __a, __vector signed int __b) {
+ const __vector signed int __zero = (__vector signed int)0;
+ return __builtin_s390_vpklsf(
+ (__vector unsigned int)(__a >= __zero) & (__vector unsigned int)__a,
+ (__vector unsigned int)(__b >= __zero) & (__vector unsigned int)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_packsu(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vpklsf(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_packsu(__vector signed long long __a, __vector signed long long __b) {
+ const __vector signed long long __zero = (__vector signed long long)0;
+ return __builtin_s390_vpklsg(
+ (__vector unsigned long long)(__a >= __zero) &
+ (__vector unsigned long long)__a,
+ (__vector unsigned long long)(__b >= __zero) &
+ (__vector unsigned long long)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_packsu(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return __builtin_s390_vpklsg(__a, __b);
+}
+
+/*-- vec_packsu_cc ----------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_packsu_cc(__vector unsigned short __a, __vector unsigned short __b,
+ int *__cc) {
+ return __builtin_s390_vpklshs(__a, __b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_packsu_cc(__vector unsigned int __a, __vector unsigned int __b, int *__cc) {
+ return __builtin_s390_vpklsfs(__a, __b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_packsu_cc(__vector unsigned long long __a, __vector unsigned long long __b,
+ int *__cc) {
+ return __builtin_s390_vpklsgs(__a, __b, __cc);
+}
+
+/*-- vec_unpackh ------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed short
+vec_unpackh(__vector signed char __a) {
+ return __builtin_s390_vuphb(__a);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_unpackh(__vector __bool char __a) {
+ return ((__vector __bool short)
+ __builtin_s390_vuphb((__vector signed char)__a));
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_unpackh(__vector unsigned char __a) {
+ return __builtin_s390_vuplhb(__a);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_unpackh(__vector signed short __a) {
+ return __builtin_s390_vuphh(__a);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_unpackh(__vector __bool short __a) {
+ return (__vector __bool int)__builtin_s390_vuphh((__vector signed short)__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_unpackh(__vector unsigned short __a) {
+ return __builtin_s390_vuplhh(__a);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_unpackh(__vector signed int __a) {
+ return __builtin_s390_vuphf(__a);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_unpackh(__vector __bool int __a) {
+ return ((__vector __bool long long)
+ __builtin_s390_vuphf((__vector signed int)__a));
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_unpackh(__vector unsigned int __a) {
+ return __builtin_s390_vuplhf(__a);
+}
+
+/*-- vec_unpackl ------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed short
+vec_unpackl(__vector signed char __a) {
+ return __builtin_s390_vuplb(__a);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_unpackl(__vector __bool char __a) {
+ return ((__vector __bool short)
+ __builtin_s390_vuplb((__vector signed char)__a));
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_unpackl(__vector unsigned char __a) {
+ return __builtin_s390_vupllb(__a);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_unpackl(__vector signed short __a) {
+ return __builtin_s390_vuplhw(__a);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_unpackl(__vector __bool short __a) {
+ return ((__vector __bool int)
+ __builtin_s390_vuplhw((__vector signed short)__a));
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_unpackl(__vector unsigned short __a) {
+ return __builtin_s390_vupllh(__a);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_unpackl(__vector signed int __a) {
+ return __builtin_s390_vuplf(__a);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_unpackl(__vector __bool int __a) {
+ return ((__vector __bool long long)
+ __builtin_s390_vuplf((__vector signed int)__a));
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_unpackl(__vector unsigned int __a) {
+ return __builtin_s390_vupllf(__a);
+}
+
+/*-- vec_cmpeq --------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cmpeq(__vector __bool char __a, __vector __bool char __b) {
+ return (__vector __bool char)(__a == __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cmpeq(__vector signed char __a, __vector signed char __b) {
+ return (__vector __bool char)(__a == __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cmpeq(__vector unsigned char __a, __vector unsigned char __b) {
+ return (__vector __bool char)(__a == __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cmpeq(__vector __bool short __a, __vector __bool short __b) {
+ return (__vector __bool short)(__a == __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cmpeq(__vector signed short __a, __vector signed short __b) {
+ return (__vector __bool short)(__a == __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cmpeq(__vector unsigned short __a, __vector unsigned short __b) {
+ return (__vector __bool short)(__a == __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmpeq(__vector __bool int __a, __vector __bool int __b) {
+ return (__vector __bool int)(__a == __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmpeq(__vector signed int __a, __vector signed int __b) {
+ return (__vector __bool int)(__a == __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmpeq(__vector unsigned int __a, __vector unsigned int __b) {
+ return (__vector __bool int)(__a == __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmpeq(__vector __bool long long __a, __vector __bool long long __b) {
+ return (__vector __bool long long)(__a == __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmpeq(__vector signed long long __a, __vector signed long long __b) {
+ return (__vector __bool long long)(__a == __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmpeq(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return (__vector __bool long long)(__a == __b);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmpeq(__vector float __a, __vector float __b) {
+ return (__vector __bool int)(__a == __b);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmpeq(__vector double __a, __vector double __b) {
+ return (__vector __bool long long)(__a == __b);
+}
+
+/*-- vec_cmpge --------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cmpge(__vector signed char __a, __vector signed char __b) {
+ return (__vector __bool char)(__a >= __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cmpge(__vector unsigned char __a, __vector unsigned char __b) {
+ return (__vector __bool char)(__a >= __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cmpge(__vector signed short __a, __vector signed short __b) {
+ return (__vector __bool short)(__a >= __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cmpge(__vector unsigned short __a, __vector unsigned short __b) {
+ return (__vector __bool short)(__a >= __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmpge(__vector signed int __a, __vector signed int __b) {
+ return (__vector __bool int)(__a >= __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmpge(__vector unsigned int __a, __vector unsigned int __b) {
+ return (__vector __bool int)(__a >= __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmpge(__vector signed long long __a, __vector signed long long __b) {
+ return (__vector __bool long long)(__a >= __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmpge(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return (__vector __bool long long)(__a >= __b);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmpge(__vector float __a, __vector float __b) {
+ return (__vector __bool int)(__a >= __b);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmpge(__vector double __a, __vector double __b) {
+ return (__vector __bool long long)(__a >= __b);
+}
+
+/*-- vec_cmpgt --------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cmpgt(__vector signed char __a, __vector signed char __b) {
+ return (__vector __bool char)(__a > __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cmpgt(__vector unsigned char __a, __vector unsigned char __b) {
+ return (__vector __bool char)(__a > __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cmpgt(__vector signed short __a, __vector signed short __b) {
+ return (__vector __bool short)(__a > __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cmpgt(__vector unsigned short __a, __vector unsigned short __b) {
+ return (__vector __bool short)(__a > __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmpgt(__vector signed int __a, __vector signed int __b) {
+ return (__vector __bool int)(__a > __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmpgt(__vector unsigned int __a, __vector unsigned int __b) {
+ return (__vector __bool int)(__a > __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmpgt(__vector signed long long __a, __vector signed long long __b) {
+ return (__vector __bool long long)(__a > __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmpgt(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return (__vector __bool long long)(__a > __b);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmpgt(__vector float __a, __vector float __b) {
+ return (__vector __bool int)(__a > __b);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmpgt(__vector double __a, __vector double __b) {
+ return (__vector __bool long long)(__a > __b);
+}
+
+/*-- vec_cmple --------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cmple(__vector signed char __a, __vector signed char __b) {
+ return (__vector __bool char)(__a <= __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cmple(__vector unsigned char __a, __vector unsigned char __b) {
+ return (__vector __bool char)(__a <= __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cmple(__vector signed short __a, __vector signed short __b) {
+ return (__vector __bool short)(__a <= __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cmple(__vector unsigned short __a, __vector unsigned short __b) {
+ return (__vector __bool short)(__a <= __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmple(__vector signed int __a, __vector signed int __b) {
+ return (__vector __bool int)(__a <= __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmple(__vector unsigned int __a, __vector unsigned int __b) {
+ return (__vector __bool int)(__a <= __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmple(__vector signed long long __a, __vector signed long long __b) {
+ return (__vector __bool long long)(__a <= __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmple(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return (__vector __bool long long)(__a <= __b);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmple(__vector float __a, __vector float __b) {
+ return (__vector __bool int)(__a <= __b);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmple(__vector double __a, __vector double __b) {
+ return (__vector __bool long long)(__a <= __b);
+}
+
+/*-- vec_cmplt --------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cmplt(__vector signed char __a, __vector signed char __b) {
+ return (__vector __bool char)(__a < __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cmplt(__vector unsigned char __a, __vector unsigned char __b) {
+ return (__vector __bool char)(__a < __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cmplt(__vector signed short __a, __vector signed short __b) {
+ return (__vector __bool short)(__a < __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cmplt(__vector unsigned short __a, __vector unsigned short __b) {
+ return (__vector __bool short)(__a < __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmplt(__vector signed int __a, __vector signed int __b) {
+ return (__vector __bool int)(__a < __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmplt(__vector unsigned int __a, __vector unsigned int __b) {
+ return (__vector __bool int)(__a < __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmplt(__vector signed long long __a, __vector signed long long __b) {
+ return (__vector __bool long long)(__a < __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmplt(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return (__vector __bool long long)(__a < __b);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmplt(__vector float __a, __vector float __b) {
+ return (__vector __bool int)(__a < __b);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_cmplt(__vector double __a, __vector double __b) {
+ return (__vector __bool long long)(__a < __b);
+}
+
+/*-- vec_all_eq -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector signed char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vceqbs(__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector signed char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vceqbs(__a, (__vector signed char)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector __bool char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a, __b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector unsigned char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector unsigned char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector __bool char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector __bool char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector signed short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vceqhs(__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector signed short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vceqhs(__a, (__vector signed short)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector __bool short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a, __b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector unsigned short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector unsigned short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector __bool short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector __bool short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector signed int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vceqfs(__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector signed int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vceqfs(__a, (__vector signed int)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector __bool int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a, __b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector unsigned int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector unsigned int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector __bool int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector __bool int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector signed long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs(__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector signed long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs(__a, (__vector signed long long)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector __bool long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a, __b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector unsigned long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector unsigned long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector __bool long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector __bool long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc == 0;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfcesbs(__a, __b, &__cc);
+ return __cc == 0;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_all_eq(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfcedbs(__a, __b, &__cc);
+ return __cc == 0;
+}
+
+/*-- vec_all_ne -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector signed char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vceqbs(__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector signed char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vceqbs(__a, (__vector signed char)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector __bool char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a, __b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector unsigned char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector unsigned char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector __bool char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector __bool char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector signed short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vceqhs(__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector signed short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vceqhs(__a, (__vector signed short)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector __bool short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a, __b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector unsigned short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector unsigned short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector __bool short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector __bool short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector signed int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vceqfs(__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector signed int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vceqfs(__a, (__vector signed int)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector __bool int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a, __b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector unsigned int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector unsigned int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector __bool int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector __bool int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector signed long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs(__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector signed long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs(__a, (__vector signed long long)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector __bool long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a, __b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector unsigned long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector unsigned long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector __bool long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector __bool long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc == 3;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfcesbs(__a, __b, &__cc);
+ return __cc == 3;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_all_ne(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfcedbs(__a, __b, &__cc);
+ return __cc == 3;
+}
+
+/*-- vec_all_ge -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector signed char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector signed char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchbs((__vector signed char)__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector __bool char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__b, (__vector signed char)__a, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector unsigned char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector unsigned char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector __bool char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__b, (__vector unsigned char)__a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector __bool char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__b,
+ (__vector unsigned char)__a, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector signed short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector signed short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchhs((__vector signed short)__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector __bool short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__b, (__vector signed short)__a, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector unsigned short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector unsigned short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector __bool short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__b, (__vector unsigned short)__a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector __bool short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__b,
+ (__vector unsigned short)__a, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector signed int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector signed int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchfs((__vector signed int)__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector __bool int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__b, (__vector signed int)__a, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector unsigned int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector unsigned int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector __bool int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__b, (__vector unsigned int)__a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector __bool int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__b,
+ (__vector unsigned int)__a, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector signed long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector signed long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchgs((__vector signed long long)__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector __bool long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__b, (__vector signed long long)__a, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector unsigned long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector unsigned long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__b, __a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector __bool long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__b, (__vector unsigned long long)__a, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector __bool long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__b,
+ (__vector unsigned long long)__a, &__cc);
+ return __cc == 3;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchesbs(__a, __b, &__cc);
+ return __cc == 0;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_all_ge(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchedbs(__a, __b, &__cc);
+ return __cc == 0;
+}
+
+/*-- vec_all_gt -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector signed char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector signed char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__a, (__vector signed char)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector __bool char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs((__vector signed char)__a, __b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector unsigned char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector unsigned char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__a, (__vector unsigned char)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector __bool char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector __bool char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector signed short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector signed short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__a, (__vector signed short)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector __bool short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs((__vector signed short)__a, __b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector unsigned short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector unsigned short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__a, (__vector unsigned short)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector __bool short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector __bool short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector signed int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector signed int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__a, (__vector signed int)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector __bool int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs((__vector signed int)__a, __b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector unsigned int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector unsigned int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__a, (__vector unsigned int)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector __bool int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector __bool int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector signed long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector signed long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__a, (__vector signed long long)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector __bool long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs((__vector signed long long)__a, __b, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector unsigned long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector unsigned long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__a, (__vector unsigned long long)__b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector __bool long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__a, __b, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector __bool long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__a,
+ (__vector unsigned long long)__b, &__cc);
+ return __cc == 0;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchsbs(__a, __b, &__cc);
+ return __cc == 0;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_all_gt(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchdbs(__a, __b, &__cc);
+ return __cc == 0;
+}
+
+/*-- vec_all_le -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai int
+vec_all_le(__vector signed char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector signed char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__a, (__vector signed char)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector __bool char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs((__vector signed char)__a, __b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_le(__vector unsigned char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector unsigned char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__a, (__vector unsigned char)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector __bool char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector __bool char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_le(__vector signed short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector signed short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__a, (__vector signed short)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector __bool short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs((__vector signed short)__a, __b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_le(__vector unsigned short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector unsigned short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__a, (__vector unsigned short)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector __bool short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector __bool short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_le(__vector signed int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector signed int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__a, (__vector signed int)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector __bool int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs((__vector signed int)__a, __b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_le(__vector unsigned int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector unsigned int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__a, (__vector unsigned int)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector __bool int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector __bool int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_le(__vector signed long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector signed long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__a, (__vector signed long long)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector __bool long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs((__vector signed long long)__a, __b, &__cc);
+ return __cc == 3;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_le(__vector unsigned long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector unsigned long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__a, (__vector unsigned long long)__b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector __bool long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__a, __b, &__cc);
+ return __cc == 3;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_le(__vector __bool long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__a,
+ (__vector unsigned long long)__b, &__cc);
+ return __cc == 3;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_all_le(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchesbs(__b, __a, &__cc);
+ return __cc == 0;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_all_le(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchedbs(__b, __a, &__cc);
+ return __cc == 0;
+}
+
+/*-- vec_all_lt -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector signed char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector signed char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchbs((__vector signed char)__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector __bool char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__b, (__vector signed char)__a, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector unsigned char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector unsigned char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector __bool char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__b, (__vector unsigned char)__a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector __bool char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__b,
+ (__vector unsigned char)__a, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector signed short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector signed short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchhs((__vector signed short)__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector __bool short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__b, (__vector signed short)__a, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector unsigned short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector unsigned short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector __bool short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__b, (__vector unsigned short)__a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector __bool short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__b,
+ (__vector unsigned short)__a, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector signed int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector signed int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchfs((__vector signed int)__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector __bool int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__b, (__vector signed int)__a, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector unsigned int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector unsigned int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector __bool int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__b, (__vector unsigned int)__a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector __bool int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__b,
+ (__vector unsigned int)__a, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector signed long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector signed long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchgs((__vector signed long long)__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector __bool long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__b, (__vector signed long long)__a, &__cc);
+ return __cc == 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector unsigned long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector unsigned long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__b, __a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector __bool long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__b, (__vector unsigned long long)__a, &__cc);
+ return __cc == 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector __bool long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__b,
+ (__vector unsigned long long)__a, &__cc);
+ return __cc == 0;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchsbs(__b, __a, &__cc);
+ return __cc == 0;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_all_lt(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchdbs(__b, __a, &__cc);
+ return __cc == 0;
+}
+
+/*-- vec_all_nge ------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_all_nge(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchesbs(__a, __b, &__cc);
+ return __cc == 3;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_all_nge(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchedbs(__a, __b, &__cc);
+ return __cc == 3;
+}
+
+/*-- vec_all_ngt ------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_all_ngt(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchsbs(__a, __b, &__cc);
+ return __cc == 3;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_all_ngt(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchdbs(__a, __b, &__cc);
+ return __cc == 3;
+}
+
+/*-- vec_all_nle ------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_all_nle(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchesbs(__b, __a, &__cc);
+ return __cc == 3;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_all_nle(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchedbs(__b, __a, &__cc);
+ return __cc == 3;
+}
+
+/*-- vec_all_nlt ------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_all_nlt(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchsbs(__b, __a, &__cc);
+ return __cc == 3;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_all_nlt(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchdbs(__b, __a, &__cc);
+ return __cc == 3;
+}
+
+/*-- vec_all_nan ------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_all_nan(__vector float __a) {
+ int __cc;
+ __builtin_s390_vftcisb(__a, 15, &__cc);
+ return __cc == 0;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_all_nan(__vector double __a) {
+ int __cc;
+ __builtin_s390_vftcidb(__a, 15, &__cc);
+ return __cc == 0;
+}
+
+/*-- vec_all_numeric --------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_all_numeric(__vector float __a) {
+ int __cc;
+ __builtin_s390_vftcisb(__a, 15, &__cc);
+ return __cc == 3;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_all_numeric(__vector double __a) {
+ int __cc;
+ __builtin_s390_vftcidb(__a, 15, &__cc);
+ return __cc == 3;
+}
+
+/*-- vec_any_eq -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector signed char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vceqbs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector signed char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vceqbs(__a, (__vector signed char)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector __bool char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector unsigned char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector unsigned char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector __bool char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector __bool char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector signed short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vceqhs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector signed short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vceqhs(__a, (__vector signed short)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector __bool short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector unsigned short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector unsigned short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector __bool short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector __bool short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector signed int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vceqfs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector signed int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vceqfs(__a, (__vector signed int)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector __bool int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector unsigned int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector unsigned int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector __bool int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector __bool int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector signed long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector signed long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs(__a, (__vector signed long long)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector __bool long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector unsigned long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector unsigned long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector __bool long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector __bool long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc <= 1;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfcesbs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_any_eq(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfcedbs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+/*-- vec_any_ne -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector signed char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vceqbs(__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector signed char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vceqbs(__a, (__vector signed char)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector __bool char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a, __b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector unsigned char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector unsigned char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector __bool char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector __bool char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vceqbs((__vector signed char)__a,
+ (__vector signed char)__b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector signed short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vceqhs(__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector signed short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vceqhs(__a, (__vector signed short)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector __bool short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a, __b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector unsigned short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector unsigned short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector __bool short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector __bool short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vceqhs((__vector signed short)__a,
+ (__vector signed short)__b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector signed int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vceqfs(__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector signed int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vceqfs(__a, (__vector signed int)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector __bool int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a, __b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector unsigned int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector unsigned int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector __bool int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector __bool int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vceqfs((__vector signed int)__a,
+ (__vector signed int)__b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector signed long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs(__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector signed long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs(__a, (__vector signed long long)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector __bool long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a, __b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector unsigned long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector unsigned long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector __bool long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector __bool long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vceqgs((__vector signed long long)__a,
+ (__vector signed long long)__b, &__cc);
+ return __cc != 0;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfcesbs(__a, __b, &__cc);
+ return __cc != 0;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_any_ne(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfcedbs(__a, __b, &__cc);
+ return __cc != 0;
+}
+
+/*-- vec_any_ge -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector signed char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector signed char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchbs((__vector signed char)__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector __bool char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__b, (__vector signed char)__a, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector unsigned char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector unsigned char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector __bool char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__b, (__vector unsigned char)__a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector __bool char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__b,
+ (__vector unsigned char)__a, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector signed short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector signed short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchhs((__vector signed short)__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector __bool short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__b, (__vector signed short)__a, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector unsigned short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector unsigned short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector __bool short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__b, (__vector unsigned short)__a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector __bool short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__b,
+ (__vector unsigned short)__a, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector signed int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector signed int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchfs((__vector signed int)__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector __bool int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__b, (__vector signed int)__a, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector unsigned int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector unsigned int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector __bool int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__b, (__vector unsigned int)__a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector __bool int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__b,
+ (__vector unsigned int)__a, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector signed long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector signed long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchgs((__vector signed long long)__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector __bool long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__b, (__vector signed long long)__a, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector unsigned long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector unsigned long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__b, __a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector __bool long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__b, (__vector unsigned long long)__a, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector __bool long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__b,
+ (__vector unsigned long long)__a, &__cc);
+ return __cc != 0;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchesbs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_any_ge(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchedbs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+/*-- vec_any_gt -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector signed char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector signed char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__a, (__vector signed char)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector __bool char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs((__vector signed char)__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector unsigned char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector unsigned char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__a, (__vector unsigned char)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector __bool char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector __bool char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector signed short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector signed short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__a, (__vector signed short)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector __bool short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs((__vector signed short)__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector unsigned short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector unsigned short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__a, (__vector unsigned short)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector __bool short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector __bool short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector signed int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector signed int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__a, (__vector signed int)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector __bool int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs((__vector signed int)__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector unsigned int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector unsigned int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__a, (__vector unsigned int)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector __bool int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector __bool int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector signed long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector signed long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__a, (__vector signed long long)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector __bool long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs((__vector signed long long)__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector unsigned long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector unsigned long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__a, (__vector unsigned long long)__b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector __bool long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector __bool long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__a,
+ (__vector unsigned long long)__b, &__cc);
+ return __cc <= 1;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchsbs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_any_gt(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchdbs(__a, __b, &__cc);
+ return __cc <= 1;
+}
+
+/*-- vec_any_le -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai int
+vec_any_le(__vector signed char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector signed char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__a, (__vector signed char)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector __bool char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs((__vector signed char)__a, __b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_le(__vector unsigned char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector unsigned char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__a, (__vector unsigned char)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector __bool char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector __bool char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_le(__vector signed short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector signed short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__a, (__vector signed short)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector __bool short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs((__vector signed short)__a, __b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_le(__vector unsigned short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector unsigned short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__a, (__vector unsigned short)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector __bool short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector __bool short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_le(__vector signed int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector signed int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__a, (__vector signed int)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector __bool int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs((__vector signed int)__a, __b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_le(__vector unsigned int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector unsigned int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__a, (__vector unsigned int)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector __bool int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector __bool int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_le(__vector signed long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector signed long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__a, (__vector signed long long)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector __bool long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs((__vector signed long long)__a, __b, &__cc);
+ return __cc != 0;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_le(__vector unsigned long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector unsigned long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__a, (__vector unsigned long long)__b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector __bool long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__a, __b, &__cc);
+ return __cc != 0;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_le(__vector __bool long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__a,
+ (__vector unsigned long long)__b, &__cc);
+ return __cc != 0;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_any_le(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchesbs(__b, __a, &__cc);
+ return __cc <= 1;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_any_le(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchedbs(__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+/*-- vec_any_lt -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector signed char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector signed char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchbs((__vector signed char)__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector __bool char __a, __vector signed char __b) {
+ int __cc;
+ __builtin_s390_vchbs(__b, (__vector signed char)__a, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector unsigned char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector unsigned char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector __bool char __a, __vector unsigned char __b) {
+ int __cc;
+ __builtin_s390_vchlbs(__b, (__vector unsigned char)__a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector __bool char __a, __vector __bool char __b) {
+ int __cc;
+ __builtin_s390_vchlbs((__vector unsigned char)__b,
+ (__vector unsigned char)__a, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector signed short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector signed short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchhs((__vector signed short)__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector __bool short __a, __vector signed short __b) {
+ int __cc;
+ __builtin_s390_vchhs(__b, (__vector signed short)__a, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector unsigned short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector unsigned short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector __bool short __a, __vector unsigned short __b) {
+ int __cc;
+ __builtin_s390_vchlhs(__b, (__vector unsigned short)__a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector __bool short __a, __vector __bool short __b) {
+ int __cc;
+ __builtin_s390_vchlhs((__vector unsigned short)__b,
+ (__vector unsigned short)__a, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector signed int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector signed int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchfs((__vector signed int)__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector __bool int __a, __vector signed int __b) {
+ int __cc;
+ __builtin_s390_vchfs(__b, (__vector signed int)__a, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector unsigned int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector unsigned int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector __bool int __a, __vector unsigned int __b) {
+ int __cc;
+ __builtin_s390_vchlfs(__b, (__vector unsigned int)__a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector __bool int __a, __vector __bool int __b) {
+ int __cc;
+ __builtin_s390_vchlfs((__vector unsigned int)__b,
+ (__vector unsigned int)__a, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector signed long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector signed long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchgs((__vector signed long long)__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector __bool long long __a, __vector signed long long __b) {
+ int __cc;
+ __builtin_s390_vchgs(__b, (__vector signed long long)__a, &__cc);
+ return __cc <= 1;
+}
+
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector unsigned long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector unsigned long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector __bool long long __a, __vector unsigned long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs(__b, (__vector unsigned long long)__a, &__cc);
+ return __cc <= 1;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector __bool long long __a, __vector __bool long long __b) {
+ int __cc;
+ __builtin_s390_vchlgs((__vector unsigned long long)__b,
+ (__vector unsigned long long)__a, &__cc);
+ return __cc <= 1;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchsbs(__b, __a, &__cc);
+ return __cc <= 1;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_any_lt(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchdbs(__b, __a, &__cc);
+ return __cc <= 1;
+}
+
+/*-- vec_any_nge ------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_any_nge(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchesbs(__a, __b, &__cc);
+ return __cc != 0;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_any_nge(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchedbs(__a, __b, &__cc);
+ return __cc != 0;
+}
+
+/*-- vec_any_ngt ------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_any_ngt(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchsbs(__a, __b, &__cc);
+ return __cc != 0;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_any_ngt(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchdbs(__a, __b, &__cc);
+ return __cc != 0;
+}
+
+/*-- vec_any_nle ------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_any_nle(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchesbs(__b, __a, &__cc);
+ return __cc != 0;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_any_nle(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchedbs(__b, __a, &__cc);
+ return __cc != 0;
+}
+
+/*-- vec_any_nlt ------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_any_nlt(__vector float __a, __vector float __b) {
+ int __cc;
+ __builtin_s390_vfchsbs(__b, __a, &__cc);
+ return __cc != 0;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_any_nlt(__vector double __a, __vector double __b) {
+ int __cc;
+ __builtin_s390_vfchdbs(__b, __a, &__cc);
+ return __cc != 0;
+}
+
+/*-- vec_any_nan ------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_any_nan(__vector float __a) {
+ int __cc;
+ __builtin_s390_vftcisb(__a, 15, &__cc);
+ return __cc != 3;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_any_nan(__vector double __a) {
+ int __cc;
+ __builtin_s390_vftcidb(__a, 15, &__cc);
+ return __cc != 3;
+}
+
+/*-- vec_any_numeric --------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_any_numeric(__vector float __a) {
+ int __cc;
+ __builtin_s390_vftcisb(__a, 15, &__cc);
+ return __cc != 0;
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_any_numeric(__vector double __a) {
+ int __cc;
+ __builtin_s390_vftcidb(__a, 15, &__cc);
+ return __cc != 0;
+}
+
+/*-- vec_andc ---------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_andc(__vector __bool char __a, __vector __bool char __b) {
+ return __a & ~__b;
+}
+
+static inline __ATTRS_o_ai __vector signed char
+vec_andc(__vector signed char __a, __vector signed char __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_andc(__vector __bool char __a, __vector signed char __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_andc(__vector signed char __a, __vector __bool char __b) {
+ return __a & ~__b;
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_andc(__vector unsigned char __a, __vector unsigned char __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_andc(__vector __bool char __a, __vector unsigned char __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_andc(__vector unsigned char __a, __vector __bool char __b) {
+ return __a & ~__b;
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_andc(__vector __bool short __a, __vector __bool short __b) {
+ return __a & ~__b;
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_andc(__vector signed short __a, __vector signed short __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_andc(__vector __bool short __a, __vector signed short __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_andc(__vector signed short __a, __vector __bool short __b) {
+ return __a & ~__b;
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_andc(__vector unsigned short __a, __vector unsigned short __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_andc(__vector __bool short __a, __vector unsigned short __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_andc(__vector unsigned short __a, __vector __bool short __b) {
+ return __a & ~__b;
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_andc(__vector __bool int __a, __vector __bool int __b) {
+ return __a & ~__b;
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_andc(__vector signed int __a, __vector signed int __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_andc(__vector __bool int __a, __vector signed int __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_andc(__vector signed int __a, __vector __bool int __b) {
+ return __a & ~__b;
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_andc(__vector unsigned int __a, __vector unsigned int __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_andc(__vector __bool int __a, __vector unsigned int __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_andc(__vector unsigned int __a, __vector __bool int __b) {
+ return __a & ~__b;
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_andc(__vector __bool long long __a, __vector __bool long long __b) {
+ return __a & ~__b;
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_andc(__vector signed long long __a, __vector signed long long __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_andc(__vector __bool long long __a, __vector signed long long __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_andc(__vector signed long long __a, __vector __bool long long __b) {
+ return __a & ~__b;
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_andc(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_andc(__vector __bool long long __a, __vector unsigned long long __b) {
+ return __a & ~__b;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_andc(__vector unsigned long long __a, __vector __bool long long __b) {
+ return __a & ~__b;
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_andc(__vector float __a, __vector float __b) {
+ return (__vector float)((__vector unsigned int)__a &
+ ~(__vector unsigned int)__b);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_andc(__vector double __a, __vector double __b) {
+ return (__vector double)((__vector unsigned long long)__a &
+ ~(__vector unsigned long long)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector double
+vec_andc(__vector __bool long long __a, __vector double __b) {
+ return (__vector double)((__vector unsigned long long)__a &
+ ~(__vector unsigned long long)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector double
+vec_andc(__vector double __a, __vector __bool long long __b) {
+ return (__vector double)((__vector unsigned long long)__a &
+ ~(__vector unsigned long long)__b);
+}
+
+/*-- vec_nor ----------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_nor(__vector __bool char __a, __vector __bool char __b) {
+ return ~(__a | __b);
+}
+
+static inline __ATTRS_o_ai __vector signed char
+vec_nor(__vector signed char __a, __vector signed char __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_nor(__vector __bool char __a, __vector signed char __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_nor(__vector signed char __a, __vector __bool char __b) {
+ return ~(__a | __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_nor(__vector unsigned char __a, __vector unsigned char __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_nor(__vector __bool char __a, __vector unsigned char __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_nor(__vector unsigned char __a, __vector __bool char __b) {
+ return ~(__a | __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_nor(__vector __bool short __a, __vector __bool short __b) {
+ return ~(__a | __b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_nor(__vector signed short __a, __vector signed short __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_nor(__vector __bool short __a, __vector signed short __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_nor(__vector signed short __a, __vector __bool short __b) {
+ return ~(__a | __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_nor(__vector unsigned short __a, __vector unsigned short __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_nor(__vector __bool short __a, __vector unsigned short __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_nor(__vector unsigned short __a, __vector __bool short __b) {
+ return ~(__a | __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_nor(__vector __bool int __a, __vector __bool int __b) {
+ return ~(__a | __b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_nor(__vector signed int __a, __vector signed int __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_nor(__vector __bool int __a, __vector signed int __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_nor(__vector signed int __a, __vector __bool int __b) {
+ return ~(__a | __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_nor(__vector unsigned int __a, __vector unsigned int __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_nor(__vector __bool int __a, __vector unsigned int __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_nor(__vector unsigned int __a, __vector __bool int __b) {
+ return ~(__a | __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_nor(__vector __bool long long __a, __vector __bool long long __b) {
+ return ~(__a | __b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_nor(__vector signed long long __a, __vector signed long long __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_nor(__vector __bool long long __a, __vector signed long long __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_nor(__vector signed long long __a, __vector __bool long long __b) {
+ return ~(__a | __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_nor(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_nor(__vector __bool long long __a, __vector unsigned long long __b) {
+ return ~(__a | __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_nor(__vector unsigned long long __a, __vector __bool long long __b) {
+ return ~(__a | __b);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_nor(__vector float __a, __vector float __b) {
+ return (__vector float)~((__vector unsigned int)__a |
+ (__vector unsigned int)__b);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_nor(__vector double __a, __vector double __b) {
+ return (__vector double)~((__vector unsigned long long)__a |
+ (__vector unsigned long long)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector double
+vec_nor(__vector __bool long long __a, __vector double __b) {
+ return (__vector double)~((__vector unsigned long long)__a |
+ (__vector unsigned long long)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector double
+vec_nor(__vector double __a, __vector __bool long long __b) {
+ return (__vector double)~((__vector unsigned long long)__a |
+ (__vector unsigned long long)__b);
+}
+
+/*-- vec_orc ----------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector __bool char
+vec_orc(__vector __bool char __a, __vector __bool char __b) {
+ return __a | ~__b;
+}
+
+static inline __ATTRS_o_ai __vector signed char
+vec_orc(__vector signed char __a, __vector signed char __b) {
+ return __a | ~__b;
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_orc(__vector unsigned char __a, __vector unsigned char __b) {
+ return __a | ~__b;
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_orc(__vector __bool short __a, __vector __bool short __b) {
+ return __a | ~__b;
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_orc(__vector signed short __a, __vector signed short __b) {
+ return __a | ~__b;
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_orc(__vector unsigned short __a, __vector unsigned short __b) {
+ return __a | ~__b;
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_orc(__vector __bool int __a, __vector __bool int __b) {
+ return __a | ~__b;
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_orc(__vector signed int __a, __vector signed int __b) {
+ return __a | ~__b;
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_orc(__vector unsigned int __a, __vector unsigned int __b) {
+ return __a | ~__b;
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_orc(__vector __bool long long __a, __vector __bool long long __b) {
+ return __a | ~__b;
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_orc(__vector signed long long __a, __vector signed long long __b) {
+ return __a | ~__b;
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_orc(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return __a | ~__b;
+}
+
+static inline __ATTRS_o_ai __vector float
+vec_orc(__vector float __a, __vector float __b) {
+ return (__vector float)((__vector unsigned int)__a |
+ ~(__vector unsigned int)__b);
+}
+
+static inline __ATTRS_o_ai __vector double
+vec_orc(__vector double __a, __vector double __b) {
+ return (__vector double)((__vector unsigned long long)__a |
+ ~(__vector unsigned long long)__b);
+}
+#endif
+
+/*-- vec_nand ---------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector __bool char
+vec_nand(__vector __bool char __a, __vector __bool char __b) {
+ return ~(__a & __b);
+}
+
+static inline __ATTRS_o_ai __vector signed char
+vec_nand(__vector signed char __a, __vector signed char __b) {
+ return ~(__a & __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_nand(__vector unsigned char __a, __vector unsigned char __b) {
+ return ~(__a & __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_nand(__vector __bool short __a, __vector __bool short __b) {
+ return ~(__a & __b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_nand(__vector signed short __a, __vector signed short __b) {
+ return ~(__a & __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_nand(__vector unsigned short __a, __vector unsigned short __b) {
+ return ~(__a & __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_nand(__vector __bool int __a, __vector __bool int __b) {
+ return ~(__a & __b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_nand(__vector signed int __a, __vector signed int __b) {
+ return ~(__a & __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_nand(__vector unsigned int __a, __vector unsigned int __b) {
+ return ~(__a & __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_nand(__vector __bool long long __a, __vector __bool long long __b) {
+ return ~(__a & __b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_nand(__vector signed long long __a, __vector signed long long __b) {
+ return ~(__a & __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_nand(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return ~(__a & __b);
+}
+
+static inline __ATTRS_o_ai __vector float
+vec_nand(__vector float __a, __vector float __b) {
+ return (__vector float)~((__vector unsigned int)__a &
+ (__vector unsigned int)__b);
+}
+
+static inline __ATTRS_o_ai __vector double
+vec_nand(__vector double __a, __vector double __b) {
+ return (__vector double)~((__vector unsigned long long)__a &
+ (__vector unsigned long long)__b);
+}
+#endif
+
+/*-- vec_eqv ----------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector __bool char
+vec_eqv(__vector __bool char __a, __vector __bool char __b) {
+ return ~(__a ^ __b);
+}
+
+static inline __ATTRS_o_ai __vector signed char
+vec_eqv(__vector signed char __a, __vector signed char __b) {
+ return ~(__a ^ __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_eqv(__vector unsigned char __a, __vector unsigned char __b) {
+ return ~(__a ^ __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_eqv(__vector __bool short __a, __vector __bool short __b) {
+ return ~(__a ^ __b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_eqv(__vector signed short __a, __vector signed short __b) {
+ return ~(__a ^ __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_eqv(__vector unsigned short __a, __vector unsigned short __b) {
+ return ~(__a ^ __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_eqv(__vector __bool int __a, __vector __bool int __b) {
+ return ~(__a ^ __b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_eqv(__vector signed int __a, __vector signed int __b) {
+ return ~(__a ^ __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_eqv(__vector unsigned int __a, __vector unsigned int __b) {
+ return ~(__a ^ __b);
+}
+
+static inline __ATTRS_o_ai __vector __bool long long
+vec_eqv(__vector __bool long long __a, __vector __bool long long __b) {
+ return ~(__a ^ __b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_eqv(__vector signed long long __a, __vector signed long long __b) {
+ return ~(__a ^ __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_eqv(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return ~(__a ^ __b);
+}
+
+static inline __ATTRS_o_ai __vector float
+vec_eqv(__vector float __a, __vector float __b) {
+ return (__vector float)~((__vector unsigned int)__a ^
+ (__vector unsigned int)__b);
+}
+
+static inline __ATTRS_o_ai __vector double
+vec_eqv(__vector double __a, __vector double __b) {
+ return (__vector double)~((__vector unsigned long long)__a ^
+ (__vector unsigned long long)__b);
+}
+#endif
+
+/*-- vec_cntlz --------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cntlz(__vector signed char __a) {
+ return __builtin_s390_vclzb((__vector unsigned char)__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cntlz(__vector unsigned char __a) {
+ return __builtin_s390_vclzb(__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cntlz(__vector signed short __a) {
+ return __builtin_s390_vclzh((__vector unsigned short)__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cntlz(__vector unsigned short __a) {
+ return __builtin_s390_vclzh(__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cntlz(__vector signed int __a) {
+ return __builtin_s390_vclzf((__vector unsigned int)__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cntlz(__vector unsigned int __a) {
+ return __builtin_s390_vclzf(__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_cntlz(__vector signed long long __a) {
+ return __builtin_s390_vclzg((__vector unsigned long long)__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_cntlz(__vector unsigned long long __a) {
+ return __builtin_s390_vclzg(__a);
+}
+
+/*-- vec_cnttz --------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cnttz(__vector signed char __a) {
+ return __builtin_s390_vctzb((__vector unsigned char)__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cnttz(__vector unsigned char __a) {
+ return __builtin_s390_vctzb(__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cnttz(__vector signed short __a) {
+ return __builtin_s390_vctzh((__vector unsigned short)__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cnttz(__vector unsigned short __a) {
+ return __builtin_s390_vctzh(__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cnttz(__vector signed int __a) {
+ return __builtin_s390_vctzf((__vector unsigned int)__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cnttz(__vector unsigned int __a) {
+ return __builtin_s390_vctzf(__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_cnttz(__vector signed long long __a) {
+ return __builtin_s390_vctzg((__vector unsigned long long)__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_cnttz(__vector unsigned long long __a) {
+ return __builtin_s390_vctzg(__a);
+}
+
+/*-- vec_popcnt -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_popcnt(__vector signed char __a) {
+ return __builtin_s390_vpopctb((__vector unsigned char)__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_popcnt(__vector unsigned char __a) {
+ return __builtin_s390_vpopctb(__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_popcnt(__vector signed short __a) {
+ return __builtin_s390_vpopcth((__vector unsigned short)__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_popcnt(__vector unsigned short __a) {
+ return __builtin_s390_vpopcth(__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_popcnt(__vector signed int __a) {
+ return __builtin_s390_vpopctf((__vector unsigned int)__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_popcnt(__vector unsigned int __a) {
+ return __builtin_s390_vpopctf(__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_popcnt(__vector signed long long __a) {
+ return __builtin_s390_vpopctg((__vector unsigned long long)__a);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_popcnt(__vector unsigned long long __a) {
+ return __builtin_s390_vpopctg(__a);
+}
+
+/*-- vec_rl -----------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_rl(__vector signed char __a, __vector unsigned char __b) {
+ return (__vector signed char)__builtin_s390_verllvb(
+ (__vector unsigned char)__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_rl(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_verllvb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_rl(__vector signed short __a, __vector unsigned short __b) {
+ return (__vector signed short)__builtin_s390_verllvh(
+ (__vector unsigned short)__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_rl(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_verllvh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_rl(__vector signed int __a, __vector unsigned int __b) {
+ return (__vector signed int)__builtin_s390_verllvf(
+ (__vector unsigned int)__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_rl(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_verllvf(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_rl(__vector signed long long __a, __vector unsigned long long __b) {
+ return (__vector signed long long)__builtin_s390_verllvg(
+ (__vector unsigned long long)__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_rl(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return __builtin_s390_verllvg(__a, __b);
+}
+
+/*-- vec_rli ----------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_rli(__vector signed char __a, unsigned long __b) {
+ return (__vector signed char)__builtin_s390_verllb(
+ (__vector unsigned char)__a, (int)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_rli(__vector unsigned char __a, unsigned long __b) {
+ return __builtin_s390_verllb(__a, (int)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_rli(__vector signed short __a, unsigned long __b) {
+ return (__vector signed short)__builtin_s390_verllh(
+ (__vector unsigned short)__a, (int)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_rli(__vector unsigned short __a, unsigned long __b) {
+ return __builtin_s390_verllh(__a, (int)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_rli(__vector signed int __a, unsigned long __b) {
+ return (__vector signed int)__builtin_s390_verllf(
+ (__vector unsigned int)__a, (int)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_rli(__vector unsigned int __a, unsigned long __b) {
+ return __builtin_s390_verllf(__a, (int)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_rli(__vector signed long long __a, unsigned long __b) {
+ return (__vector signed long long)__builtin_s390_verllg(
+ (__vector unsigned long long)__a, (int)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_rli(__vector unsigned long long __a, unsigned long __b) {
+ return __builtin_s390_verllg(__a, (int)__b);
+}
+
+/*-- vec_rl_mask ------------------------------------------------------------*/
+
+extern __ATTRS_o __vector signed char
+vec_rl_mask(__vector signed char __a, __vector unsigned char __b,
+ unsigned char __c) __constant(__c);
+
+extern __ATTRS_o __vector unsigned char
+vec_rl_mask(__vector unsigned char __a, __vector unsigned char __b,
+ unsigned char __c) __constant(__c);
+
+extern __ATTRS_o __vector signed short
+vec_rl_mask(__vector signed short __a, __vector unsigned short __b,
+ unsigned char __c) __constant(__c);
+
+extern __ATTRS_o __vector unsigned short
+vec_rl_mask(__vector unsigned short __a, __vector unsigned short __b,
+ unsigned char __c) __constant(__c);
+
+extern __ATTRS_o __vector signed int
+vec_rl_mask(__vector signed int __a, __vector unsigned int __b,
+ unsigned char __c) __constant(__c);
+
+extern __ATTRS_o __vector unsigned int
+vec_rl_mask(__vector unsigned int __a, __vector unsigned int __b,
+ unsigned char __c) __constant(__c);
+
+extern __ATTRS_o __vector signed long long
+vec_rl_mask(__vector signed long long __a, __vector unsigned long long __b,
+ unsigned char __c) __constant(__c);
+
+extern __ATTRS_o __vector unsigned long long
+vec_rl_mask(__vector unsigned long long __a, __vector unsigned long long __b,
+ unsigned char __c) __constant(__c);
+
+#define vec_rl_mask(X, Y, Z) ((__typeof__((vec_rl_mask)((X), (Y), (Z)))) \
+ __extension__ ({ \
+ __vector unsigned char __res; \
+ __vector unsigned char __x = (__vector unsigned char)(X); \
+ __vector unsigned char __y = (__vector unsigned char)(Y); \
+ switch (sizeof ((X)[0])) { \
+ case 1: __res = (__vector unsigned char) __builtin_s390_verimb( \
+ (__vector unsigned char)__x, (__vector unsigned char)__x, \
+ (__vector unsigned char)__y, (Z)); break; \
+ case 2: __res = (__vector unsigned char) __builtin_s390_verimh( \
+ (__vector unsigned short)__x, (__vector unsigned short)__x, \
+ (__vector unsigned short)__y, (Z)); break; \
+ case 4: __res = (__vector unsigned char) __builtin_s390_verimf( \
+ (__vector unsigned int)__x, (__vector unsigned int)__x, \
+ (__vector unsigned int)__y, (Z)); break; \
+ default: __res = (__vector unsigned char) __builtin_s390_verimg( \
+ (__vector unsigned long long)__x, (__vector unsigned long long)__x, \
+ (__vector unsigned long long)__y, (Z)); break; \
+ } __res; }))
+
+/*-- vec_sll ----------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_sll(__vector signed char __a, __vector unsigned char __b) {
+ return (__vector signed char)__builtin_s390_vsl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_sll(__vector signed char __a, __vector unsigned short __b) {
+ return (__vector signed char)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_sll(__vector signed char __a, __vector unsigned int __b) {
+ return (__vector signed char)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool char
+vec_sll(__vector __bool char __a, __vector unsigned char __b) {
+ return (__vector __bool char)__builtin_s390_vsl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool char
+vec_sll(__vector __bool char __a, __vector unsigned short __b) {
+ return (__vector __bool char)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool char
+vec_sll(__vector __bool char __a, __vector unsigned int __b) {
+ return (__vector __bool char)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_sll(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vsl(__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_sll(__vector unsigned char __a, __vector unsigned short __b) {
+ return __builtin_s390_vsl(__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_sll(__vector unsigned char __a, __vector unsigned int __b) {
+ return __builtin_s390_vsl(__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_sll(__vector signed short __a, __vector unsigned char __b) {
+ return (__vector signed short)__builtin_s390_vsl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_sll(__vector signed short __a, __vector unsigned short __b) {
+ return (__vector signed short)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_sll(__vector signed short __a, __vector unsigned int __b) {
+ return (__vector signed short)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool short
+vec_sll(__vector __bool short __a, __vector unsigned char __b) {
+ return (__vector __bool short)__builtin_s390_vsl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool short
+vec_sll(__vector __bool short __a, __vector unsigned short __b) {
+ return (__vector __bool short)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool short
+vec_sll(__vector __bool short __a, __vector unsigned int __b) {
+ return (__vector __bool short)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_sll(__vector unsigned short __a, __vector unsigned char __b) {
+ return (__vector unsigned short)__builtin_s390_vsl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_sll(__vector unsigned short __a, __vector unsigned short __b) {
+ return (__vector unsigned short)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_sll(__vector unsigned short __a, __vector unsigned int __b) {
+ return (__vector unsigned short)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_sll(__vector signed int __a, __vector unsigned char __b) {
+ return (__vector signed int)__builtin_s390_vsl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_sll(__vector signed int __a, __vector unsigned short __b) {
+ return (__vector signed int)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_sll(__vector signed int __a, __vector unsigned int __b) {
+ return (__vector signed int)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool int
+vec_sll(__vector __bool int __a, __vector unsigned char __b) {
+ return (__vector __bool int)__builtin_s390_vsl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool int
+vec_sll(__vector __bool int __a, __vector unsigned short __b) {
+ return (__vector __bool int)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool int
+vec_sll(__vector __bool int __a, __vector unsigned int __b) {
+ return (__vector __bool int)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_sll(__vector unsigned int __a, __vector unsigned char __b) {
+ return (__vector unsigned int)__builtin_s390_vsl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_sll(__vector unsigned int __a, __vector unsigned short __b) {
+ return (__vector unsigned int)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_sll(__vector unsigned int __a, __vector unsigned int __b) {
+ return (__vector unsigned int)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_sll(__vector signed long long __a, __vector unsigned char __b) {
+ return (__vector signed long long)__builtin_s390_vsl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_sll(__vector signed long long __a, __vector unsigned short __b) {
+ return (__vector signed long long)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_sll(__vector signed long long __a, __vector unsigned int __b) {
+ return (__vector signed long long)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool long long
+vec_sll(__vector __bool long long __a, __vector unsigned char __b) {
+ return (__vector __bool long long)__builtin_s390_vsl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool long long
+vec_sll(__vector __bool long long __a, __vector unsigned short __b) {
+ return (__vector __bool long long)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool long long
+vec_sll(__vector __bool long long __a, __vector unsigned int __b) {
+ return (__vector __bool long long)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_sll(__vector unsigned long long __a, __vector unsigned char __b) {
+ return (__vector unsigned long long)__builtin_s390_vsl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_sll(__vector unsigned long long __a, __vector unsigned short __b) {
+ return (__vector unsigned long long)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_sll(__vector unsigned long long __a, __vector unsigned int __b) {
+ return (__vector unsigned long long)__builtin_s390_vsl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+/*-- vec_slb ----------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_slb(__vector signed char __a, __vector signed char __b) {
+ return (__vector signed char)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed char
+vec_slb(__vector signed char __a, __vector unsigned char __b) {
+ return (__vector signed char)__builtin_s390_vslb(
+ (__vector unsigned char)__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_slb(__vector unsigned char __a, __vector signed char __b) {
+ return __builtin_s390_vslb(__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_slb(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vslb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_slb(__vector signed short __a, __vector signed short __b) {
+ return (__vector signed short)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_slb(__vector signed short __a, __vector unsigned short __b) {
+ return (__vector signed short)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_slb(__vector unsigned short __a, __vector signed short __b) {
+ return (__vector unsigned short)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_slb(__vector unsigned short __a, __vector unsigned short __b) {
+ return (__vector unsigned short)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_slb(__vector signed int __a, __vector signed int __b) {
+ return (__vector signed int)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_slb(__vector signed int __a, __vector unsigned int __b) {
+ return (__vector signed int)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_slb(__vector unsigned int __a, __vector signed int __b) {
+ return (__vector unsigned int)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_slb(__vector unsigned int __a, __vector unsigned int __b) {
+ return (__vector unsigned int)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_slb(__vector signed long long __a, __vector signed long long __b) {
+ return (__vector signed long long)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_slb(__vector signed long long __a, __vector unsigned long long __b) {
+ return (__vector signed long long)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_slb(__vector unsigned long long __a, __vector signed long long __b) {
+ return (__vector unsigned long long)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_slb(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return (__vector unsigned long long)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_slb(__vector float __a, __vector signed int __b) {
+ return (__vector float)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector float
+vec_slb(__vector float __a, __vector unsigned int __b) {
+ return (__vector float)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_slb(__vector double __a, __vector signed long long __b) {
+ return (__vector double)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector double
+vec_slb(__vector double __a, __vector unsigned long long __b) {
+ return (__vector double)__builtin_s390_vslb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+/*-- vec_sld ----------------------------------------------------------------*/
+
+extern __ATTRS_o __vector signed char
+vec_sld(__vector signed char __a, __vector signed char __b, int __c)
+ __constant_range(__c, 0, 15);
+
+extern __ATTRS_o __vector __bool char
+vec_sld(__vector __bool char __a, __vector __bool char __b, int __c)
+ __constant_range(__c, 0, 15);
+
+extern __ATTRS_o __vector unsigned char
+vec_sld(__vector unsigned char __a, __vector unsigned char __b, int __c)
+ __constant_range(__c, 0, 15);
+
+extern __ATTRS_o __vector signed short
+vec_sld(__vector signed short __a, __vector signed short __b, int __c)
+ __constant_range(__c, 0, 15);
+
+extern __ATTRS_o __vector __bool short
+vec_sld(__vector __bool short __a, __vector __bool short __b, int __c)
+ __constant_range(__c, 0, 15);
+
+extern __ATTRS_o __vector unsigned short
+vec_sld(__vector unsigned short __a, __vector unsigned short __b, int __c)
+ __constant_range(__c, 0, 15);
+
+extern __ATTRS_o __vector signed int
+vec_sld(__vector signed int __a, __vector signed int __b, int __c)
+ __constant_range(__c, 0, 15);
+
+extern __ATTRS_o __vector __bool int
+vec_sld(__vector __bool int __a, __vector __bool int __b, int __c)
+ __constant_range(__c, 0, 15);
+
+extern __ATTRS_o __vector unsigned int
+vec_sld(__vector unsigned int __a, __vector unsigned int __b, int __c)
+ __constant_range(__c, 0, 15);
+
+extern __ATTRS_o __vector signed long long
+vec_sld(__vector signed long long __a, __vector signed long long __b, int __c)
+ __constant_range(__c, 0, 15);
+
+extern __ATTRS_o __vector __bool long long
+vec_sld(__vector __bool long long __a, __vector __bool long long __b, int __c)
+ __constant_range(__c, 0, 15);
+
+extern __ATTRS_o __vector unsigned long long
+vec_sld(__vector unsigned long long __a, __vector unsigned long long __b,
+ int __c)
+ __constant_range(__c, 0, 15);
+
+#if __ARCH__ >= 12
+extern __ATTRS_o __vector float
+vec_sld(__vector float __a, __vector float __b, int __c)
+ __constant_range(__c, 0, 15);
+#endif
+
+extern __ATTRS_o __vector double
+vec_sld(__vector double __a, __vector double __b, int __c)
+ __constant_range(__c, 0, 15);
+
+#define vec_sld(X, Y, Z) ((__typeof__((vec_sld)((X), (Y), (Z)))) \
+ __builtin_s390_vsldb((__vector unsigned char)(X), \
+ (__vector unsigned char)(Y), (Z)))
+
+/*-- vec_sldw ---------------------------------------------------------------*/
+
+extern __ATTRS_o __vector signed char
+vec_sldw(__vector signed char __a, __vector signed char __b, int __c)
+ __constant_range(__c, 0, 3);
+
+extern __ATTRS_o __vector unsigned char
+vec_sldw(__vector unsigned char __a, __vector unsigned char __b, int __c)
+ __constant_range(__c, 0, 3);
+
+extern __ATTRS_o __vector signed short
+vec_sldw(__vector signed short __a, __vector signed short __b, int __c)
+ __constant_range(__c, 0, 3);
+
+extern __ATTRS_o __vector unsigned short
+vec_sldw(__vector unsigned short __a, __vector unsigned short __b, int __c)
+ __constant_range(__c, 0, 3);
+
+extern __ATTRS_o __vector signed int
+vec_sldw(__vector signed int __a, __vector signed int __b, int __c)
+ __constant_range(__c, 0, 3);
+
+extern __ATTRS_o __vector unsigned int
+vec_sldw(__vector unsigned int __a, __vector unsigned int __b, int __c)
+ __constant_range(__c, 0, 3);
+
+extern __ATTRS_o __vector signed long long
+vec_sldw(__vector signed long long __a, __vector signed long long __b, int __c)
+ __constant_range(__c, 0, 3);
+
+extern __ATTRS_o __vector unsigned long long
+vec_sldw(__vector unsigned long long __a, __vector unsigned long long __b,
+ int __c)
+ __constant_range(__c, 0, 3);
+
+// This prototype is deprecated.
+extern __ATTRS_o __vector double
+vec_sldw(__vector double __a, __vector double __b, int __c)
+ __constant_range(__c, 0, 3);
+
+#define vec_sldw(X, Y, Z) ((__typeof__((vec_sldw)((X), (Y), (Z)))) \
+ __builtin_s390_vsldb((__vector unsigned char)(X), \
+ (__vector unsigned char)(Y), (Z) * 4))
+
+/*-- vec_sldb ---------------------------------------------------------------*/
+
+#if __ARCH__ >= 13
+
+extern __ATTRS_o __vector signed char
+vec_sldb(__vector signed char __a, __vector signed char __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector unsigned char
+vec_sldb(__vector unsigned char __a, __vector unsigned char __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector signed short
+vec_sldb(__vector signed short __a, __vector signed short __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector unsigned short
+vec_sldb(__vector unsigned short __a, __vector unsigned short __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector signed int
+vec_sldb(__vector signed int __a, __vector signed int __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector unsigned int
+vec_sldb(__vector unsigned int __a, __vector unsigned int __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector signed long long
+vec_sldb(__vector signed long long __a, __vector signed long long __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector unsigned long long
+vec_sldb(__vector unsigned long long __a, __vector unsigned long long __b,
+ int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector float
+vec_sldb(__vector float __a, __vector float __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector double
+vec_sldb(__vector double __a, __vector double __b, int __c)
+ __constant_range(__c, 0, 7);
+
+#define vec_sldb(X, Y, Z) ((__typeof__((vec_sldb)((X), (Y), (Z)))) \
+ __builtin_s390_vsld((__vector unsigned char)(X), \
+ (__vector unsigned char)(Y), (Z)))
+
+#endif
+
+/*-- vec_sral ---------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_sral(__vector signed char __a, __vector unsigned char __b) {
+ return (__vector signed char)__builtin_s390_vsra(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_sral(__vector signed char __a, __vector unsigned short __b) {
+ return (__vector signed char)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_sral(__vector signed char __a, __vector unsigned int __b) {
+ return (__vector signed char)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool char
+vec_sral(__vector __bool char __a, __vector unsigned char __b) {
+ return (__vector __bool char)__builtin_s390_vsra(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool char
+vec_sral(__vector __bool char __a, __vector unsigned short __b) {
+ return (__vector __bool char)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool char
+vec_sral(__vector __bool char __a, __vector unsigned int __b) {
+ return (__vector __bool char)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_sral(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vsra(__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_sral(__vector unsigned char __a, __vector unsigned short __b) {
+ return __builtin_s390_vsra(__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_sral(__vector unsigned char __a, __vector unsigned int __b) {
+ return __builtin_s390_vsra(__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_sral(__vector signed short __a, __vector unsigned char __b) {
+ return (__vector signed short)__builtin_s390_vsra(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_sral(__vector signed short __a, __vector unsigned short __b) {
+ return (__vector signed short)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_sral(__vector signed short __a, __vector unsigned int __b) {
+ return (__vector signed short)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool short
+vec_sral(__vector __bool short __a, __vector unsigned char __b) {
+ return (__vector __bool short)__builtin_s390_vsra(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool short
+vec_sral(__vector __bool short __a, __vector unsigned short __b) {
+ return (__vector __bool short)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool short
+vec_sral(__vector __bool short __a, __vector unsigned int __b) {
+ return (__vector __bool short)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_sral(__vector unsigned short __a, __vector unsigned char __b) {
+ return (__vector unsigned short)__builtin_s390_vsra(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_sral(__vector unsigned short __a, __vector unsigned short __b) {
+ return (__vector unsigned short)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_sral(__vector unsigned short __a, __vector unsigned int __b) {
+ return (__vector unsigned short)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_sral(__vector signed int __a, __vector unsigned char __b) {
+ return (__vector signed int)__builtin_s390_vsra(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_sral(__vector signed int __a, __vector unsigned short __b) {
+ return (__vector signed int)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_sral(__vector signed int __a, __vector unsigned int __b) {
+ return (__vector signed int)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool int
+vec_sral(__vector __bool int __a, __vector unsigned char __b) {
+ return (__vector __bool int)__builtin_s390_vsra(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool int
+vec_sral(__vector __bool int __a, __vector unsigned short __b) {
+ return (__vector __bool int)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool int
+vec_sral(__vector __bool int __a, __vector unsigned int __b) {
+ return (__vector __bool int)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_sral(__vector unsigned int __a, __vector unsigned char __b) {
+ return (__vector unsigned int)__builtin_s390_vsra(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_sral(__vector unsigned int __a, __vector unsigned short __b) {
+ return (__vector unsigned int)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_sral(__vector unsigned int __a, __vector unsigned int __b) {
+ return (__vector unsigned int)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_sral(__vector signed long long __a, __vector unsigned char __b) {
+ return (__vector signed long long)__builtin_s390_vsra(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_sral(__vector signed long long __a, __vector unsigned short __b) {
+ return (__vector signed long long)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_sral(__vector signed long long __a, __vector unsigned int __b) {
+ return (__vector signed long long)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool long long
+vec_sral(__vector __bool long long __a, __vector unsigned char __b) {
+ return (__vector __bool long long)__builtin_s390_vsra(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool long long
+vec_sral(__vector __bool long long __a, __vector unsigned short __b) {
+ return (__vector __bool long long)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool long long
+vec_sral(__vector __bool long long __a, __vector unsigned int __b) {
+ return (__vector __bool long long)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_sral(__vector unsigned long long __a, __vector unsigned char __b) {
+ return (__vector unsigned long long)__builtin_s390_vsra(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_sral(__vector unsigned long long __a, __vector unsigned short __b) {
+ return (__vector unsigned long long)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_sral(__vector unsigned long long __a, __vector unsigned int __b) {
+ return (__vector unsigned long long)__builtin_s390_vsra(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+/*-- vec_srab ---------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_srab(__vector signed char __a, __vector signed char __b) {
+ return (__vector signed char)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed char
+vec_srab(__vector signed char __a, __vector unsigned char __b) {
+ return (__vector signed char)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_srab(__vector unsigned char __a, __vector signed char __b) {
+ return __builtin_s390_vsrab(__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_srab(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vsrab(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_srab(__vector signed short __a, __vector signed short __b) {
+ return (__vector signed short)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_srab(__vector signed short __a, __vector unsigned short __b) {
+ return (__vector signed short)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_srab(__vector unsigned short __a, __vector signed short __b) {
+ return (__vector unsigned short)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_srab(__vector unsigned short __a, __vector unsigned short __b) {
+ return (__vector unsigned short)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_srab(__vector signed int __a, __vector signed int __b) {
+ return (__vector signed int)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_srab(__vector signed int __a, __vector unsigned int __b) {
+ return (__vector signed int)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_srab(__vector unsigned int __a, __vector signed int __b) {
+ return (__vector unsigned int)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_srab(__vector unsigned int __a, __vector unsigned int __b) {
+ return (__vector unsigned int)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_srab(__vector signed long long __a, __vector signed long long __b) {
+ return (__vector signed long long)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_srab(__vector signed long long __a, __vector unsigned long long __b) {
+ return (__vector signed long long)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_srab(__vector unsigned long long __a, __vector signed long long __b) {
+ return (__vector unsigned long long)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_srab(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return (__vector unsigned long long)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_srab(__vector float __a, __vector signed int __b) {
+ return (__vector float)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector float
+vec_srab(__vector float __a, __vector unsigned int __b) {
+ return (__vector float)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_srab(__vector double __a, __vector signed long long __b) {
+ return (__vector double)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector double
+vec_srab(__vector double __a, __vector unsigned long long __b) {
+ return (__vector double)__builtin_s390_vsrab(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+/*-- vec_srl ----------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_srl(__vector signed char __a, __vector unsigned char __b) {
+ return (__vector signed char)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_srl(__vector signed char __a, __vector unsigned short __b) {
+ return (__vector signed char)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_srl(__vector signed char __a, __vector unsigned int __b) {
+ return (__vector signed char)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool char
+vec_srl(__vector __bool char __a, __vector unsigned char __b) {
+ return (__vector __bool char)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool char
+vec_srl(__vector __bool char __a, __vector unsigned short __b) {
+ return (__vector __bool char)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool char
+vec_srl(__vector __bool char __a, __vector unsigned int __b) {
+ return (__vector __bool char)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_srl(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vsrl(__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_srl(__vector unsigned char __a, __vector unsigned short __b) {
+ return __builtin_s390_vsrl(__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_srl(__vector unsigned char __a, __vector unsigned int __b) {
+ return __builtin_s390_vsrl(__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_srl(__vector signed short __a, __vector unsigned char __b) {
+ return (__vector signed short)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_srl(__vector signed short __a, __vector unsigned short __b) {
+ return (__vector signed short)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_srl(__vector signed short __a, __vector unsigned int __b) {
+ return (__vector signed short)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool short
+vec_srl(__vector __bool short __a, __vector unsigned char __b) {
+ return (__vector __bool short)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool short
+vec_srl(__vector __bool short __a, __vector unsigned short __b) {
+ return (__vector __bool short)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool short
+vec_srl(__vector __bool short __a, __vector unsigned int __b) {
+ return (__vector __bool short)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_srl(__vector unsigned short __a, __vector unsigned char __b) {
+ return (__vector unsigned short)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_srl(__vector unsigned short __a, __vector unsigned short __b) {
+ return (__vector unsigned short)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_srl(__vector unsigned short __a, __vector unsigned int __b) {
+ return (__vector unsigned short)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_srl(__vector signed int __a, __vector unsigned char __b) {
+ return (__vector signed int)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_srl(__vector signed int __a, __vector unsigned short __b) {
+ return (__vector signed int)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_srl(__vector signed int __a, __vector unsigned int __b) {
+ return (__vector signed int)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool int
+vec_srl(__vector __bool int __a, __vector unsigned char __b) {
+ return (__vector __bool int)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool int
+vec_srl(__vector __bool int __a, __vector unsigned short __b) {
+ return (__vector __bool int)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool int
+vec_srl(__vector __bool int __a, __vector unsigned int __b) {
+ return (__vector __bool int)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_srl(__vector unsigned int __a, __vector unsigned char __b) {
+ return (__vector unsigned int)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_srl(__vector unsigned int __a, __vector unsigned short __b) {
+ return (__vector unsigned int)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_srl(__vector unsigned int __a, __vector unsigned int __b) {
+ return (__vector unsigned int)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_srl(__vector signed long long __a, __vector unsigned char __b) {
+ return (__vector signed long long)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_srl(__vector signed long long __a, __vector unsigned short __b) {
+ return (__vector signed long long)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_srl(__vector signed long long __a, __vector unsigned int __b) {
+ return (__vector signed long long)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool long long
+vec_srl(__vector __bool long long __a, __vector unsigned char __b) {
+ return (__vector __bool long long)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool long long
+vec_srl(__vector __bool long long __a, __vector unsigned short __b) {
+ return (__vector __bool long long)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector __bool long long
+vec_srl(__vector __bool long long __a, __vector unsigned int __b) {
+ return (__vector __bool long long)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_srl(__vector unsigned long long __a, __vector unsigned char __b) {
+ return (__vector unsigned long long)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, __b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_srl(__vector unsigned long long __a, __vector unsigned short __b) {
+ return (__vector unsigned long long)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_srl(__vector unsigned long long __a, __vector unsigned int __b) {
+ return (__vector unsigned long long)__builtin_s390_vsrl(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+/*-- vec_srb ----------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_srb(__vector signed char __a, __vector signed char __b) {
+ return (__vector signed char)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed char
+vec_srb(__vector signed char __a, __vector unsigned char __b) {
+ return (__vector signed char)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_srb(__vector unsigned char __a, __vector signed char __b) {
+ return __builtin_s390_vsrlb(__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_srb(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vsrlb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_srb(__vector signed short __a, __vector signed short __b) {
+ return (__vector signed short)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_srb(__vector signed short __a, __vector unsigned short __b) {
+ return (__vector signed short)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_srb(__vector unsigned short __a, __vector signed short __b) {
+ return (__vector unsigned short)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_srb(__vector unsigned short __a, __vector unsigned short __b) {
+ return (__vector unsigned short)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_srb(__vector signed int __a, __vector signed int __b) {
+ return (__vector signed int)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_srb(__vector signed int __a, __vector unsigned int __b) {
+ return (__vector signed int)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_srb(__vector unsigned int __a, __vector signed int __b) {
+ return (__vector unsigned int)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_srb(__vector unsigned int __a, __vector unsigned int __b) {
+ return (__vector unsigned int)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_srb(__vector signed long long __a, __vector signed long long __b) {
+ return (__vector signed long long)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_srb(__vector signed long long __a, __vector unsigned long long __b) {
+ return (__vector signed long long)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_srb(__vector unsigned long long __a, __vector signed long long __b) {
+ return (__vector unsigned long long)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_srb(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return (__vector unsigned long long)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_srb(__vector float __a, __vector signed int __b) {
+ return (__vector float)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector float
+vec_srb(__vector float __a, __vector unsigned int __b) {
+ return (__vector float)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_srb(__vector double __a, __vector signed long long __b) {
+ return (__vector double)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector double
+vec_srb(__vector double __a, __vector unsigned long long __b) {
+ return (__vector double)__builtin_s390_vsrlb(
+ (__vector unsigned char)__a, (__vector unsigned char)__b);
+}
+
+/*-- vec_srdb ---------------------------------------------------------------*/
+
+#if __ARCH__ >= 13
+
+extern __ATTRS_o __vector signed char
+vec_srdb(__vector signed char __a, __vector signed char __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector unsigned char
+vec_srdb(__vector unsigned char __a, __vector unsigned char __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector signed short
+vec_srdb(__vector signed short __a, __vector signed short __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector unsigned short
+vec_srdb(__vector unsigned short __a, __vector unsigned short __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector signed int
+vec_srdb(__vector signed int __a, __vector signed int __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector unsigned int
+vec_srdb(__vector unsigned int __a, __vector unsigned int __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector signed long long
+vec_srdb(__vector signed long long __a, __vector signed long long __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector unsigned long long
+vec_srdb(__vector unsigned long long __a, __vector unsigned long long __b,
+ int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector float
+vec_srdb(__vector float __a, __vector float __b, int __c)
+ __constant_range(__c, 0, 7);
+
+extern __ATTRS_o __vector double
+vec_srdb(__vector double __a, __vector double __b, int __c)
+ __constant_range(__c, 0, 7);
+
+#define vec_srdb(X, Y, Z) ((__typeof__((vec_srdb)((X), (Y), (Z)))) \
+ __builtin_s390_vsrd((__vector unsigned char)(X), \
+ (__vector unsigned char)(Y), (Z)))
+
+#endif
+
+/*-- vec_abs ----------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_abs(__vector signed char __a) {
+ return vec_sel(__a, -__a, vec_cmplt(__a, (__vector signed char)0));
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_abs(__vector signed short __a) {
+ return vec_sel(__a, -__a, vec_cmplt(__a, (__vector signed short)0));
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_abs(__vector signed int __a) {
+ return vec_sel(__a, -__a, vec_cmplt(__a, (__vector signed int)0));
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_abs(__vector signed long long __a) {
+ return vec_sel(__a, -__a, vec_cmplt(__a, (__vector signed long long)0));
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_abs(__vector float __a) {
+ return __builtin_s390_vflpsb(__a);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_abs(__vector double __a) {
+ return __builtin_s390_vflpdb(__a);
+}
+
+/*-- vec_nabs ---------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_nabs(__vector float __a) {
+ return __builtin_s390_vflnsb(__a);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_nabs(__vector double __a) {
+ return __builtin_s390_vflndb(__a);
+}
+
+/*-- vec_max ----------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_max(__vector signed char __a, __vector signed char __b) {
+ return vec_sel(__b, __a, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_max(__vector signed char __a, __vector __bool char __b) {
+ __vector signed char __bc = (__vector signed char)__b;
+ return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_max(__vector __bool char __a, __vector signed char __b) {
+ __vector signed char __ac = (__vector signed char)__a;
+ return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_max(__vector unsigned char __a, __vector unsigned char __b) {
+ return vec_sel(__b, __a, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_max(__vector unsigned char __a, __vector __bool char __b) {
+ __vector unsigned char __bc = (__vector unsigned char)__b;
+ return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_max(__vector __bool char __a, __vector unsigned char __b) {
+ __vector unsigned char __ac = (__vector unsigned char)__a;
+ return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_max(__vector signed short __a, __vector signed short __b) {
+ return vec_sel(__b, __a, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_max(__vector signed short __a, __vector __bool short __b) {
+ __vector signed short __bc = (__vector signed short)__b;
+ return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_max(__vector __bool short __a, __vector signed short __b) {
+ __vector signed short __ac = (__vector signed short)__a;
+ return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_max(__vector unsigned short __a, __vector unsigned short __b) {
+ return vec_sel(__b, __a, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_max(__vector unsigned short __a, __vector __bool short __b) {
+ __vector unsigned short __bc = (__vector unsigned short)__b;
+ return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_max(__vector __bool short __a, __vector unsigned short __b) {
+ __vector unsigned short __ac = (__vector unsigned short)__a;
+ return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_max(__vector signed int __a, __vector signed int __b) {
+ return vec_sel(__b, __a, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_max(__vector signed int __a, __vector __bool int __b) {
+ __vector signed int __bc = (__vector signed int)__b;
+ return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_max(__vector __bool int __a, __vector signed int __b) {
+ __vector signed int __ac = (__vector signed int)__a;
+ return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_max(__vector unsigned int __a, __vector unsigned int __b) {
+ return vec_sel(__b, __a, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_max(__vector unsigned int __a, __vector __bool int __b) {
+ __vector unsigned int __bc = (__vector unsigned int)__b;
+ return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_max(__vector __bool int __a, __vector unsigned int __b) {
+ __vector unsigned int __ac = (__vector unsigned int)__a;
+ return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_max(__vector signed long long __a, __vector signed long long __b) {
+ return vec_sel(__b, __a, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_max(__vector signed long long __a, __vector __bool long long __b) {
+ __vector signed long long __bc = (__vector signed long long)__b;
+ return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_max(__vector __bool long long __a, __vector signed long long __b) {
+ __vector signed long long __ac = (__vector signed long long)__a;
+ return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_max(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return vec_sel(__b, __a, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_max(__vector unsigned long long __a, __vector __bool long long __b) {
+ __vector unsigned long long __bc = (__vector unsigned long long)__b;
+ return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_max(__vector __bool long long __a, __vector unsigned long long __b) {
+ __vector unsigned long long __ac = (__vector unsigned long long)__a;
+ return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_max(__vector float __a, __vector float __b) {
+ return __builtin_s390_vfmaxsb(__a, __b, 0);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_max(__vector double __a, __vector double __b) {
+#if __ARCH__ >= 12
+ return __builtin_s390_vfmaxdb(__a, __b, 0);
+#else
+ return vec_sel(__b, __a, vec_cmpgt(__a, __b));
+#endif
+}
+
+/*-- vec_min ----------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_min(__vector signed char __a, __vector signed char __b) {
+ return vec_sel(__a, __b, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_min(__vector signed char __a, __vector __bool char __b) {
+ __vector signed char __bc = (__vector signed char)__b;
+ return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed char
+vec_min(__vector __bool char __a, __vector signed char __b) {
+ __vector signed char __ac = (__vector signed char)__a;
+ return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_min(__vector unsigned char __a, __vector unsigned char __b) {
+ return vec_sel(__a, __b, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_min(__vector unsigned char __a, __vector __bool char __b) {
+ __vector unsigned char __bc = (__vector unsigned char)__b;
+ return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned char
+vec_min(__vector __bool char __a, __vector unsigned char __b) {
+ __vector unsigned char __ac = (__vector unsigned char)__a;
+ return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_min(__vector signed short __a, __vector signed short __b) {
+ return vec_sel(__a, __b, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_min(__vector signed short __a, __vector __bool short __b) {
+ __vector signed short __bc = (__vector signed short)__b;
+ return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed short
+vec_min(__vector __bool short __a, __vector signed short __b) {
+ __vector signed short __ac = (__vector signed short)__a;
+ return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_min(__vector unsigned short __a, __vector unsigned short __b) {
+ return vec_sel(__a, __b, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_min(__vector unsigned short __a, __vector __bool short __b) {
+ __vector unsigned short __bc = (__vector unsigned short)__b;
+ return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned short
+vec_min(__vector __bool short __a, __vector unsigned short __b) {
+ __vector unsigned short __ac = (__vector unsigned short)__a;
+ return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_min(__vector signed int __a, __vector signed int __b) {
+ return vec_sel(__a, __b, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_min(__vector signed int __a, __vector __bool int __b) {
+ __vector signed int __bc = (__vector signed int)__b;
+ return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed int
+vec_min(__vector __bool int __a, __vector signed int __b) {
+ __vector signed int __ac = (__vector signed int)__a;
+ return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_min(__vector unsigned int __a, __vector unsigned int __b) {
+ return vec_sel(__a, __b, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_min(__vector unsigned int __a, __vector __bool int __b) {
+ __vector unsigned int __bc = (__vector unsigned int)__b;
+ return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned int
+vec_min(__vector __bool int __a, __vector unsigned int __b) {
+ __vector unsigned int __ac = (__vector unsigned int)__a;
+ return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_min(__vector signed long long __a, __vector signed long long __b) {
+ return vec_sel(__a, __b, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_min(__vector signed long long __a, __vector __bool long long __b) {
+ __vector signed long long __bc = (__vector signed long long)__b;
+ return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_min(__vector __bool long long __a, __vector signed long long __b) {
+ __vector signed long long __ac = (__vector signed long long)__a;
+ return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_min(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return vec_sel(__a, __b, vec_cmpgt(__a, __b));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_min(__vector unsigned long long __a, __vector __bool long long __b) {
+ __vector unsigned long long __bc = (__vector unsigned long long)__b;
+ return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_min(__vector __bool long long __a, __vector unsigned long long __b) {
+ __vector unsigned long long __ac = (__vector unsigned long long)__a;
+ return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_min(__vector float __a, __vector float __b) {
+ return __builtin_s390_vfminsb(__a, __b, 0);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_min(__vector double __a, __vector double __b) {
+#if __ARCH__ >= 12
+ return __builtin_s390_vfmindb(__a, __b, 0);
+#else
+ return vec_sel(__a, __b, vec_cmpgt(__a, __b));
+#endif
+}
+
+/*-- vec_add_u128 -----------------------------------------------------------*/
+
+static inline __ATTRS_ai __vector unsigned char
+vec_add_u128(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vaq(__a, __b);
+}
+
+/*-- vec_addc ---------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_addc(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vaccb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_addc(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vacch(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_addc(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vaccf(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_addc(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return __builtin_s390_vaccg(__a, __b);
+}
+
+/*-- vec_addc_u128 ----------------------------------------------------------*/
+
+static inline __ATTRS_ai __vector unsigned char
+vec_addc_u128(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vaccq(__a, __b);
+}
+
+/*-- vec_adde_u128 ----------------------------------------------------------*/
+
+static inline __ATTRS_ai __vector unsigned char
+vec_adde_u128(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c) {
+ return __builtin_s390_vacq(__a, __b, __c);
+}
+
+/*-- vec_addec_u128 ---------------------------------------------------------*/
+
+static inline __ATTRS_ai __vector unsigned char
+vec_addec_u128(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c) {
+ return __builtin_s390_vacccq(__a, __b, __c);
+}
+
+/*-- vec_avg ----------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_avg(__vector signed char __a, __vector signed char __b) {
+ return __builtin_s390_vavgb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_avg(__vector signed short __a, __vector signed short __b) {
+ return __builtin_s390_vavgh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_avg(__vector signed int __a, __vector signed int __b) {
+ return __builtin_s390_vavgf(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_avg(__vector signed long long __a, __vector signed long long __b) {
+ return __builtin_s390_vavgg(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_avg(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vavglb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_avg(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vavglh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_avg(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vavglf(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_avg(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return __builtin_s390_vavglg(__a, __b);
+}
+
+/*-- vec_checksum -----------------------------------------------------------*/
+
+static inline __ATTRS_ai __vector unsigned int
+vec_checksum(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vcksm(__a, __b);
+}
+
+/*-- vec_gfmsum -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_gfmsum(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vgfmb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_gfmsum(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vgfmh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_gfmsum(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vgfmf(__a, __b);
+}
+
+/*-- vec_gfmsum_128 ---------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_gfmsum_128(__vector unsigned long long __a,
+ __vector unsigned long long __b) {
+ return __builtin_s390_vgfmg(__a, __b);
+}
+
+/*-- vec_gfmsum_accum -------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_gfmsum_accum(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned short __c) {
+ return __builtin_s390_vgfmab(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_gfmsum_accum(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned int __c) {
+ return __builtin_s390_vgfmah(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_gfmsum_accum(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned long long __c) {
+ return __builtin_s390_vgfmaf(__a, __b, __c);
+}
+
+/*-- vec_gfmsum_accum_128 ---------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_gfmsum_accum_128(__vector unsigned long long __a,
+ __vector unsigned long long __b,
+ __vector unsigned char __c) {
+ return __builtin_s390_vgfmag(__a, __b, __c);
+}
+
+/*-- vec_mladd --------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_mladd(__vector signed char __a, __vector signed char __b,
+ __vector signed char __c) {
+ return __a * __b + __c;
+}
+
+static inline __ATTRS_o_ai __vector signed char
+vec_mladd(__vector unsigned char __a, __vector signed char __b,
+ __vector signed char __c) {
+ return (__vector signed char)__a * __b + __c;
+}
+
+static inline __ATTRS_o_ai __vector signed char
+vec_mladd(__vector signed char __a, __vector unsigned char __b,
+ __vector unsigned char __c) {
+ return __a * (__vector signed char)__b + (__vector signed char)__c;
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_mladd(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c) {
+ return __a * __b + __c;
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_mladd(__vector signed short __a, __vector signed short __b,
+ __vector signed short __c) {
+ return __a * __b + __c;
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_mladd(__vector unsigned short __a, __vector signed short __b,
+ __vector signed short __c) {
+ return (__vector signed short)__a * __b + __c;
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_mladd(__vector signed short __a, __vector unsigned short __b,
+ __vector unsigned short __c) {
+ return __a * (__vector signed short)__b + (__vector signed short)__c;
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_mladd(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned short __c) {
+ return __a * __b + __c;
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_mladd(__vector signed int __a, __vector signed int __b,
+ __vector signed int __c) {
+ return __a * __b + __c;
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_mladd(__vector unsigned int __a, __vector signed int __b,
+ __vector signed int __c) {
+ return (__vector signed int)__a * __b + __c;
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_mladd(__vector signed int __a, __vector unsigned int __b,
+ __vector unsigned int __c) {
+ return __a * (__vector signed int)__b + (__vector signed int)__c;
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_mladd(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned int __c) {
+ return __a * __b + __c;
+}
+
+/*-- vec_mhadd --------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_mhadd(__vector signed char __a, __vector signed char __b,
+ __vector signed char __c) {
+ return __builtin_s390_vmahb(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_mhadd(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c) {
+ return __builtin_s390_vmalhb(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_mhadd(__vector signed short __a, __vector signed short __b,
+ __vector signed short __c) {
+ return __builtin_s390_vmahh(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_mhadd(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned short __c) {
+ return __builtin_s390_vmalhh(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_mhadd(__vector signed int __a, __vector signed int __b,
+ __vector signed int __c) {
+ return __builtin_s390_vmahf(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_mhadd(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned int __c) {
+ return __builtin_s390_vmalhf(__a, __b, __c);
+}
+
+/*-- vec_meadd --------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed short
+vec_meadd(__vector signed char __a, __vector signed char __b,
+ __vector signed short __c) {
+ return __builtin_s390_vmaeb(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_meadd(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned short __c) {
+ return __builtin_s390_vmaleb(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_meadd(__vector signed short __a, __vector signed short __b,
+ __vector signed int __c) {
+ return __builtin_s390_vmaeh(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_meadd(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned int __c) {
+ return __builtin_s390_vmaleh(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_meadd(__vector signed int __a, __vector signed int __b,
+ __vector signed long long __c) {
+ return __builtin_s390_vmaef(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_meadd(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned long long __c) {
+ return __builtin_s390_vmalef(__a, __b, __c);
+}
+
+/*-- vec_moadd --------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed short
+vec_moadd(__vector signed char __a, __vector signed char __b,
+ __vector signed short __c) {
+ return __builtin_s390_vmaob(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_moadd(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned short __c) {
+ return __builtin_s390_vmalob(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_moadd(__vector signed short __a, __vector signed short __b,
+ __vector signed int __c) {
+ return __builtin_s390_vmaoh(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_moadd(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned int __c) {
+ return __builtin_s390_vmaloh(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_moadd(__vector signed int __a, __vector signed int __b,
+ __vector signed long long __c) {
+ return __builtin_s390_vmaof(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_moadd(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned long long __c) {
+ return __builtin_s390_vmalof(__a, __b, __c);
+}
+
+/*-- vec_mulh ---------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_mulh(__vector signed char __a, __vector signed char __b) {
+ return __builtin_s390_vmhb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_mulh(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vmlhb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_mulh(__vector signed short __a, __vector signed short __b) {
+ return __builtin_s390_vmhh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_mulh(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vmlhh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_mulh(__vector signed int __a, __vector signed int __b) {
+ return __builtin_s390_vmhf(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_mulh(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vmlhf(__a, __b);
+}
+
+/*-- vec_mule ---------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed short
+vec_mule(__vector signed char __a, __vector signed char __b) {
+ return __builtin_s390_vmeb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_mule(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vmleb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_mule(__vector signed short __a, __vector signed short __b) {
+ return __builtin_s390_vmeh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_mule(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vmleh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_mule(__vector signed int __a, __vector signed int __b) {
+ return __builtin_s390_vmef(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_mule(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vmlef(__a, __b);
+}
+
+/*-- vec_mulo ---------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed short
+vec_mulo(__vector signed char __a, __vector signed char __b) {
+ return __builtin_s390_vmob(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_mulo(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vmlob(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_mulo(__vector signed short __a, __vector signed short __b) {
+ return __builtin_s390_vmoh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_mulo(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vmloh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_mulo(__vector signed int __a, __vector signed int __b) {
+ return __builtin_s390_vmof(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_mulo(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vmlof(__a, __b);
+}
+
+/*-- vec_msum_u128 ----------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+#define vec_msum_u128(X, Y, Z, W) \
+ ((__vector unsigned char)__builtin_s390_vmslg((X), (Y), (Z), (W)));
+#endif
+
+/*-- vec_sub_u128 -----------------------------------------------------------*/
+
+static inline __ATTRS_ai __vector unsigned char
+vec_sub_u128(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vsq(__a, __b);
+}
+
+/*-- vec_subc ---------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_subc(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vscbib(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_subc(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vscbih(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_subc(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vscbif(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_subc(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return __builtin_s390_vscbig(__a, __b);
+}
+
+/*-- vec_subc_u128 ----------------------------------------------------------*/
+
+static inline __ATTRS_ai __vector unsigned char
+vec_subc_u128(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vscbiq(__a, __b);
+}
+
+/*-- vec_sube_u128 ----------------------------------------------------------*/
+
+static inline __ATTRS_ai __vector unsigned char
+vec_sube_u128(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c) {
+ return __builtin_s390_vsbiq(__a, __b, __c);
+}
+
+/*-- vec_subec_u128 ---------------------------------------------------------*/
+
+static inline __ATTRS_ai __vector unsigned char
+vec_subec_u128(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c) {
+ return __builtin_s390_vsbcbiq(__a, __b, __c);
+}
+
+/*-- vec_sum2 ---------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_sum2(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vsumgh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_sum2(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vsumgf(__a, __b);
+}
+
+/*-- vec_sum_u128 -----------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_sum_u128(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vsumqf(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_sum_u128(__vector unsigned long long __a, __vector unsigned long long __b) {
+ return __builtin_s390_vsumqg(__a, __b);
+}
+
+/*-- vec_sum4 ---------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_sum4(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vsumb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_sum4(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vsumh(__a, __b);
+}
+
+/*-- vec_test_mask ----------------------------------------------------------*/
+
+static inline __ATTRS_o_ai int
+vec_test_mask(__vector signed char __a, __vector unsigned char __b) {
+ return __builtin_s390_vtm((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai int
+vec_test_mask(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vtm(__a, __b);
+}
+
+static inline __ATTRS_o_ai int
+vec_test_mask(__vector signed short __a, __vector unsigned short __b) {
+ return __builtin_s390_vtm((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai int
+vec_test_mask(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vtm((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai int
+vec_test_mask(__vector signed int __a, __vector unsigned int __b) {
+ return __builtin_s390_vtm((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai int
+vec_test_mask(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vtm((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai int
+vec_test_mask(__vector signed long long __a, __vector unsigned long long __b) {
+ return __builtin_s390_vtm((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai int
+vec_test_mask(__vector unsigned long long __a,
+ __vector unsigned long long __b) {
+ return __builtin_s390_vtm((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai int
+vec_test_mask(__vector float __a, __vector unsigned int __b) {
+ return __builtin_s390_vtm((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+#endif
+
+static inline __ATTRS_o_ai int
+vec_test_mask(__vector double __a, __vector unsigned long long __b) {
+ return __builtin_s390_vtm((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+/*-- vec_madd ---------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_madd(__vector float __a, __vector float __b, __vector float __c) {
+ return __builtin_s390_vfmasb(__a, __b, __c);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_madd(__vector double __a, __vector double __b, __vector double __c) {
+ return __builtin_s390_vfmadb(__a, __b, __c);
+}
+
+/*-- vec_msub ---------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_msub(__vector float __a, __vector float __b, __vector float __c) {
+ return __builtin_s390_vfmssb(__a, __b, __c);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_msub(__vector double __a, __vector double __b, __vector double __c) {
+ return __builtin_s390_vfmsdb(__a, __b, __c);
+}
+
+/*-- vec_nmadd ---------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_nmadd(__vector float __a, __vector float __b, __vector float __c) {
+ return __builtin_s390_vfnmasb(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector double
+vec_nmadd(__vector double __a, __vector double __b, __vector double __c) {
+ return __builtin_s390_vfnmadb(__a, __b, __c);
+}
+#endif
+
+/*-- vec_nmsub ---------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_nmsub(__vector float __a, __vector float __b, __vector float __c) {
+ return __builtin_s390_vfnmssb(__a, __b, __c);
+}
+
+static inline __ATTRS_o_ai __vector double
+vec_nmsub(__vector double __a, __vector double __b, __vector double __c) {
+ return __builtin_s390_vfnmsdb(__a, __b, __c);
+}
+#endif
+
+/*-- vec_sqrt ---------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_sqrt(__vector float __a) {
+ return __builtin_s390_vfsqsb(__a);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_sqrt(__vector double __a) {
+ return __builtin_s390_vfsqdb(__a);
+}
+
+/*-- vec_ld2f ---------------------------------------------------------------*/
+
+// This prototype is deprecated.
+static inline __ATTRS_ai __vector double
+vec_ld2f(const float *__ptr) {
+ typedef float __v2f32 __attribute__((__vector_size__(8)));
+ return __builtin_convertvector(*(const __v2f32 *)__ptr, __vector double);
+}
+
+/*-- vec_st2f ---------------------------------------------------------------*/
+
+// This prototype is deprecated.
+static inline __ATTRS_ai void
+vec_st2f(__vector double __a, float *__ptr) {
+ typedef float __v2f32 __attribute__((__vector_size__(8)));
+ *(__v2f32 *)__ptr = __builtin_convertvector(__a, __v2f32);
+}
+
+/*-- vec_ctd ----------------------------------------------------------------*/
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector double
+vec_ctd(__vector signed long long __a, int __b)
+ __constant_range(__b, 0, 31) {
+ __vector double __conv = __builtin_convertvector(__a, __vector double);
+ __conv *= ((__vector double)(__vector unsigned long long)
+ ((0x3ffULL - __b) << 52));
+ return __conv;
+}
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector double
+vec_ctd(__vector unsigned long long __a, int __b)
+ __constant_range(__b, 0, 31) {
+ __vector double __conv = __builtin_convertvector(__a, __vector double);
+ __conv *= ((__vector double)(__vector unsigned long long)
+ ((0x3ffULL - __b) << 52));
+ return __conv;
+}
+
+/*-- vec_ctsl ---------------------------------------------------------------*/
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector signed long long
+vec_ctsl(__vector double __a, int __b)
+ __constant_range(__b, 0, 31) {
+ __a *= ((__vector double)(__vector unsigned long long)
+ ((0x3ffULL + __b) << 52));
+ return __builtin_convertvector(__a, __vector signed long long);
+}
+
+/*-- vec_ctul ---------------------------------------------------------------*/
+
+// This prototype is deprecated.
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_ctul(__vector double __a, int __b)
+ __constant_range(__b, 0, 31) {
+ __a *= ((__vector double)(__vector unsigned long long)
+ ((0x3ffULL + __b) << 52));
+ return __builtin_convertvector(__a, __vector unsigned long long);
+}
+
+/*-- vec_doublee ------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_ai __vector double
+vec_doublee(__vector float __a) {
+ typedef float __v2f32 __attribute__((__vector_size__(8)));
+ __v2f32 __pack = __builtin_shufflevector(__a, __a, 0, 2);
+ return __builtin_convertvector(__pack, __vector double);
+}
+#endif
+
+/*-- vec_floate -------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_ai __vector float
+vec_floate(__vector double __a) {
+ typedef float __v2f32 __attribute__((__vector_size__(8)));
+ __v2f32 __pack = __builtin_convertvector(__a, __v2f32);
+ return __builtin_shufflevector(__pack, __pack, 0, -1, 1, -1);
+}
+#endif
+
+/*-- vec_double -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector double
+vec_double(__vector signed long long __a) {
+ return __builtin_convertvector(__a, __vector double);
+}
+
+static inline __ATTRS_o_ai __vector double
+vec_double(__vector unsigned long long __a) {
+ return __builtin_convertvector(__a, __vector double);
+}
+
+/*-- vec_float --------------------------------------------------------------*/
+
+#if __ARCH__ >= 13
+
+static inline __ATTRS_o_ai __vector float
+vec_float(__vector signed int __a) {
+ return __builtin_convertvector(__a, __vector float);
+}
+
+static inline __ATTRS_o_ai __vector float
+vec_float(__vector unsigned int __a) {
+ return __builtin_convertvector(__a, __vector float);
+}
+
+#endif
+
+/*-- vec_signed -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed long long
+vec_signed(__vector double __a) {
+ return __builtin_convertvector(__a, __vector signed long long);
+}
+
+#if __ARCH__ >= 13
+static inline __ATTRS_o_ai __vector signed int
+vec_signed(__vector float __a) {
+ return __builtin_convertvector(__a, __vector signed int);
+}
+#endif
+
+/*-- vec_unsigned -----------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned long long
+vec_unsigned(__vector double __a) {
+ return __builtin_convertvector(__a, __vector unsigned long long);
+}
+
+#if __ARCH__ >= 13
+static inline __ATTRS_o_ai __vector unsigned int
+vec_unsigned(__vector float __a) {
+ return __builtin_convertvector(__a, __vector unsigned int);
+}
+#endif
+
+/*-- vec_roundp -------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_roundp(__vector float __a) {
+ return __builtin_s390_vfisb(__a, 4, 6);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_roundp(__vector double __a) {
+ return __builtin_s390_vfidb(__a, 4, 6);
+}
+
+/*-- vec_ceil ---------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_ceil(__vector float __a) {
+ // On this platform, vec_ceil never triggers the IEEE-inexact exception.
+ return __builtin_s390_vfisb(__a, 4, 6);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_ceil(__vector double __a) {
+ // On this platform, vec_ceil never triggers the IEEE-inexact exception.
+ return __builtin_s390_vfidb(__a, 4, 6);
+}
+
+/*-- vec_roundm -------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_roundm(__vector float __a) {
+ return __builtin_s390_vfisb(__a, 4, 7);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_roundm(__vector double __a) {
+ return __builtin_s390_vfidb(__a, 4, 7);
+}
+
+/*-- vec_floor --------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_floor(__vector float __a) {
+ // On this platform, vec_floor never triggers the IEEE-inexact exception.
+ return __builtin_s390_vfisb(__a, 4, 7);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_floor(__vector double __a) {
+ // On this platform, vec_floor never triggers the IEEE-inexact exception.
+ return __builtin_s390_vfidb(__a, 4, 7);
+}
+
+/*-- vec_roundz -------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_roundz(__vector float __a) {
+ return __builtin_s390_vfisb(__a, 4, 5);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_roundz(__vector double __a) {
+ return __builtin_s390_vfidb(__a, 4, 5);
+}
+
+/*-- vec_trunc --------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_trunc(__vector float __a) {
+ // On this platform, vec_trunc never triggers the IEEE-inexact exception.
+ return __builtin_s390_vfisb(__a, 4, 5);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_trunc(__vector double __a) {
+ // On this platform, vec_trunc never triggers the IEEE-inexact exception.
+ return __builtin_s390_vfidb(__a, 4, 5);
+}
+
+/*-- vec_roundc -------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_roundc(__vector float __a) {
+ return __builtin_s390_vfisb(__a, 4, 0);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_roundc(__vector double __a) {
+ return __builtin_s390_vfidb(__a, 4, 0);
+}
+
+/*-- vec_rint ---------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_rint(__vector float __a) {
+ // vec_rint may trigger the IEEE-inexact exception.
+ return __builtin_s390_vfisb(__a, 0, 0);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_rint(__vector double __a) {
+ // vec_rint may trigger the IEEE-inexact exception.
+ return __builtin_s390_vfidb(__a, 0, 0);
+}
+
+/*-- vec_round --------------------------------------------------------------*/
+
+#if __ARCH__ >= 12
+static inline __ATTRS_o_ai __vector float
+vec_round(__vector float __a) {
+ return __builtin_s390_vfisb(__a, 4, 4);
+}
+#endif
+
+static inline __ATTRS_o_ai __vector double
+vec_round(__vector double __a) {
+ return __builtin_s390_vfidb(__a, 4, 4);
+}
+
+/*-- vec_fp_test_data_class -------------------------------------------------*/
+
+#if __ARCH__ >= 12
+extern __ATTRS_o __vector __bool int
+vec_fp_test_data_class(__vector float __a, int __b, int *__c)
+ __constant_range(__b, 0, 4095);
+
+extern __ATTRS_o __vector __bool long long
+vec_fp_test_data_class(__vector double __a, int __b, int *__c)
+ __constant_range(__b, 0, 4095);
+
+#define vec_fp_test_data_class(X, Y, Z) \
+ ((__typeof__((vec_fp_test_data_class)((X), (Y), (Z)))) \
+ __extension__ ({ \
+ __vector unsigned char __res; \
+ __vector unsigned char __x = (__vector unsigned char)(X); \
+ int *__z = (Z); \
+ switch (sizeof ((X)[0])) { \
+ case 4: __res = (__vector unsigned char) \
+ __builtin_s390_vftcisb((__vector float)__x, (Y), __z); \
+ break; \
+ default: __res = (__vector unsigned char) \
+ __builtin_s390_vftcidb((__vector double)__x, (Y), __z); \
+ break; \
+ } __res; }))
+#else
+#define vec_fp_test_data_class(X, Y, Z) \
+ ((__vector __bool long long)__builtin_s390_vftcidb((X), (Y), (Z)))
+#endif
+
+#define __VEC_CLASS_FP_ZERO_P (1 << 11)
+#define __VEC_CLASS_FP_ZERO_N (1 << 10)
+#define __VEC_CLASS_FP_ZERO (__VEC_CLASS_FP_ZERO_P | __VEC_CLASS_FP_ZERO_N)
+#define __VEC_CLASS_FP_NORMAL_P (1 << 9)
+#define __VEC_CLASS_FP_NORMAL_N (1 << 8)
+#define __VEC_CLASS_FP_NORMAL (__VEC_CLASS_FP_NORMAL_P | \
+ __VEC_CLASS_FP_NORMAL_N)
+#define __VEC_CLASS_FP_SUBNORMAL_P (1 << 7)
+#define __VEC_CLASS_FP_SUBNORMAL_N (1 << 6)
+#define __VEC_CLASS_FP_SUBNORMAL (__VEC_CLASS_FP_SUBNORMAL_P | \
+ __VEC_CLASS_FP_SUBNORMAL_N)
+#define __VEC_CLASS_FP_INFINITY_P (1 << 5)
+#define __VEC_CLASS_FP_INFINITY_N (1 << 4)
+#define __VEC_CLASS_FP_INFINITY (__VEC_CLASS_FP_INFINITY_P | \
+ __VEC_CLASS_FP_INFINITY_N)
+#define __VEC_CLASS_FP_QNAN_P (1 << 3)
+#define __VEC_CLASS_FP_QNAN_N (1 << 2)
+#define __VEC_CLASS_FP_QNAN (__VEC_CLASS_FP_QNAN_P | __VEC_CLASS_FP_QNAN_N)
+#define __VEC_CLASS_FP_SNAN_P (1 << 1)
+#define __VEC_CLASS_FP_SNAN_N (1 << 0)
+#define __VEC_CLASS_FP_SNAN (__VEC_CLASS_FP_SNAN_P | __VEC_CLASS_FP_SNAN_N)
+#define __VEC_CLASS_FP_NAN (__VEC_CLASS_FP_QNAN | __VEC_CLASS_FP_SNAN)
+#define __VEC_CLASS_FP_NOT_NORMAL (__VEC_CLASS_FP_NAN | \
+ __VEC_CLASS_FP_SUBNORMAL | \
+ __VEC_CLASS_FP_ZERO | \
+ __VEC_CLASS_FP_INFINITY)
+
+/*-- vec_cp_until_zero ------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_cp_until_zero(__vector signed char __a) {
+ return ((__vector signed char)
+ __builtin_s390_vistrb((__vector unsigned char)__a));
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cp_until_zero(__vector __bool char __a) {
+ return ((__vector __bool char)
+ __builtin_s390_vistrb((__vector unsigned char)__a));
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cp_until_zero(__vector unsigned char __a) {
+ return __builtin_s390_vistrb(__a);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_cp_until_zero(__vector signed short __a) {
+ return ((__vector signed short)
+ __builtin_s390_vistrh((__vector unsigned short)__a));
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cp_until_zero(__vector __bool short __a) {
+ return ((__vector __bool short)
+ __builtin_s390_vistrh((__vector unsigned short)__a));
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cp_until_zero(__vector unsigned short __a) {
+ return __builtin_s390_vistrh(__a);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_cp_until_zero(__vector signed int __a) {
+ return ((__vector signed int)
+ __builtin_s390_vistrf((__vector unsigned int)__a));
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cp_until_zero(__vector __bool int __a) {
+ return ((__vector __bool int)
+ __builtin_s390_vistrf((__vector unsigned int)__a));
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cp_until_zero(__vector unsigned int __a) {
+ return __builtin_s390_vistrf(__a);
+}
+
+/*-- vec_cp_until_zero_cc ---------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_cp_until_zero_cc(__vector signed char __a, int *__cc) {
+ return (__vector signed char)
+ __builtin_s390_vistrbs((__vector unsigned char)__a, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cp_until_zero_cc(__vector __bool char __a, int *__cc) {
+ return (__vector __bool char)
+ __builtin_s390_vistrbs((__vector unsigned char)__a, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cp_until_zero_cc(__vector unsigned char __a, int *__cc) {
+ return __builtin_s390_vistrbs(__a, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_cp_until_zero_cc(__vector signed short __a, int *__cc) {
+ return (__vector signed short)
+ __builtin_s390_vistrhs((__vector unsigned short)__a, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cp_until_zero_cc(__vector __bool short __a, int *__cc) {
+ return (__vector __bool short)
+ __builtin_s390_vistrhs((__vector unsigned short)__a, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cp_until_zero_cc(__vector unsigned short __a, int *__cc) {
+ return __builtin_s390_vistrhs(__a, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_cp_until_zero_cc(__vector signed int __a, int *__cc) {
+ return (__vector signed int)
+ __builtin_s390_vistrfs((__vector unsigned int)__a, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cp_until_zero_cc(__vector __bool int __a, int *__cc) {
+ return (__vector __bool int)
+ __builtin_s390_vistrfs((__vector unsigned int)__a, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cp_until_zero_cc(__vector unsigned int __a, int *__cc) {
+ return __builtin_s390_vistrfs(__a, __cc);
+}
+
+/*-- vec_cmpeq_idx ----------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_cmpeq_idx(__vector signed char __a, __vector signed char __b) {
+ return (__vector signed char)
+ __builtin_s390_vfeeb((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpeq_idx(__vector __bool char __a, __vector __bool char __b) {
+ return __builtin_s390_vfeeb((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpeq_idx(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vfeeb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_cmpeq_idx(__vector signed short __a, __vector signed short __b) {
+ return (__vector signed short)
+ __builtin_s390_vfeeh((__vector unsigned short)__a,
+ (__vector unsigned short)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpeq_idx(__vector __bool short __a, __vector __bool short __b) {
+ return __builtin_s390_vfeeh((__vector unsigned short)__a,
+ (__vector unsigned short)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpeq_idx(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vfeeh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_cmpeq_idx(__vector signed int __a, __vector signed int __b) {
+ return (__vector signed int)
+ __builtin_s390_vfeef((__vector unsigned int)__a,
+ (__vector unsigned int)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpeq_idx(__vector __bool int __a, __vector __bool int __b) {
+ return __builtin_s390_vfeef((__vector unsigned int)__a,
+ (__vector unsigned int)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpeq_idx(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vfeef(__a, __b);
+}
+
+/*-- vec_cmpeq_idx_cc -------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_cmpeq_idx_cc(__vector signed char __a, __vector signed char __b, int *__cc) {
+ return (__vector signed char)
+ __builtin_s390_vfeebs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpeq_idx_cc(__vector __bool char __a, __vector __bool char __b, int *__cc) {
+ return __builtin_s390_vfeebs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpeq_idx_cc(__vector unsigned char __a, __vector unsigned char __b,
+ int *__cc) {
+ return __builtin_s390_vfeebs(__a, __b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_cmpeq_idx_cc(__vector signed short __a, __vector signed short __b,
+ int *__cc) {
+ return (__vector signed short)
+ __builtin_s390_vfeehs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpeq_idx_cc(__vector __bool short __a, __vector __bool short __b, int *__cc) {
+ return __builtin_s390_vfeehs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpeq_idx_cc(__vector unsigned short __a, __vector unsigned short __b,
+ int *__cc) {
+ return __builtin_s390_vfeehs(__a, __b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_cmpeq_idx_cc(__vector signed int __a, __vector signed int __b, int *__cc) {
+ return (__vector signed int)
+ __builtin_s390_vfeefs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpeq_idx_cc(__vector __bool int __a, __vector __bool int __b, int *__cc) {
+ return __builtin_s390_vfeefs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpeq_idx_cc(__vector unsigned int __a, __vector unsigned int __b,
+ int *__cc) {
+ return __builtin_s390_vfeefs(__a, __b, __cc);
+}
+
+/*-- vec_cmpeq_or_0_idx -----------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_cmpeq_or_0_idx(__vector signed char __a, __vector signed char __b) {
+ return (__vector signed char)
+ __builtin_s390_vfeezb((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpeq_or_0_idx(__vector __bool char __a, __vector __bool char __b) {
+ return __builtin_s390_vfeezb((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpeq_or_0_idx(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vfeezb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_cmpeq_or_0_idx(__vector signed short __a, __vector signed short __b) {
+ return (__vector signed short)
+ __builtin_s390_vfeezh((__vector unsigned short)__a,
+ (__vector unsigned short)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpeq_or_0_idx(__vector __bool short __a, __vector __bool short __b) {
+ return __builtin_s390_vfeezh((__vector unsigned short)__a,
+ (__vector unsigned short)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpeq_or_0_idx(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vfeezh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_cmpeq_or_0_idx(__vector signed int __a, __vector signed int __b) {
+ return (__vector signed int)
+ __builtin_s390_vfeezf((__vector unsigned int)__a,
+ (__vector unsigned int)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpeq_or_0_idx(__vector __bool int __a, __vector __bool int __b) {
+ return __builtin_s390_vfeezf((__vector unsigned int)__a,
+ (__vector unsigned int)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpeq_or_0_idx(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vfeezf(__a, __b);
+}
+
+/*-- vec_cmpeq_or_0_idx_cc --------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_cmpeq_or_0_idx_cc(__vector signed char __a, __vector signed char __b,
+ int *__cc) {
+ return (__vector signed char)
+ __builtin_s390_vfeezbs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpeq_or_0_idx_cc(__vector __bool char __a, __vector __bool char __b,
+ int *__cc) {
+ return __builtin_s390_vfeezbs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpeq_or_0_idx_cc(__vector unsigned char __a, __vector unsigned char __b,
+ int *__cc) {
+ return __builtin_s390_vfeezbs(__a, __b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_cmpeq_or_0_idx_cc(__vector signed short __a, __vector signed short __b,
+ int *__cc) {
+ return (__vector signed short)
+ __builtin_s390_vfeezhs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpeq_or_0_idx_cc(__vector __bool short __a, __vector __bool short __b,
+ int *__cc) {
+ return __builtin_s390_vfeezhs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpeq_or_0_idx_cc(__vector unsigned short __a, __vector unsigned short __b,
+ int *__cc) {
+ return __builtin_s390_vfeezhs(__a, __b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_cmpeq_or_0_idx_cc(__vector signed int __a, __vector signed int __b,
+ int *__cc) {
+ return (__vector signed int)
+ __builtin_s390_vfeezfs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpeq_or_0_idx_cc(__vector __bool int __a, __vector __bool int __b,
+ int *__cc) {
+ return __builtin_s390_vfeezfs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpeq_or_0_idx_cc(__vector unsigned int __a, __vector unsigned int __b,
+ int *__cc) {
+ return __builtin_s390_vfeezfs(__a, __b, __cc);
+}
+
+/*-- vec_cmpne_idx ----------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_cmpne_idx(__vector signed char __a, __vector signed char __b) {
+ return (__vector signed char)
+ __builtin_s390_vfeneb((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpne_idx(__vector __bool char __a, __vector __bool char __b) {
+ return __builtin_s390_vfeneb((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpne_idx(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vfeneb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_cmpne_idx(__vector signed short __a, __vector signed short __b) {
+ return (__vector signed short)
+ __builtin_s390_vfeneh((__vector unsigned short)__a,
+ (__vector unsigned short)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpne_idx(__vector __bool short __a, __vector __bool short __b) {
+ return __builtin_s390_vfeneh((__vector unsigned short)__a,
+ (__vector unsigned short)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpne_idx(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vfeneh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_cmpne_idx(__vector signed int __a, __vector signed int __b) {
+ return (__vector signed int)
+ __builtin_s390_vfenef((__vector unsigned int)__a,
+ (__vector unsigned int)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpne_idx(__vector __bool int __a, __vector __bool int __b) {
+ return __builtin_s390_vfenef((__vector unsigned int)__a,
+ (__vector unsigned int)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpne_idx(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vfenef(__a, __b);
+}
+
+/*-- vec_cmpne_idx_cc -------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_cmpne_idx_cc(__vector signed char __a, __vector signed char __b, int *__cc) {
+ return (__vector signed char)
+ __builtin_s390_vfenebs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpne_idx_cc(__vector __bool char __a, __vector __bool char __b, int *__cc) {
+ return __builtin_s390_vfenebs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpne_idx_cc(__vector unsigned char __a, __vector unsigned char __b,
+ int *__cc) {
+ return __builtin_s390_vfenebs(__a, __b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_cmpne_idx_cc(__vector signed short __a, __vector signed short __b,
+ int *__cc) {
+ return (__vector signed short)
+ __builtin_s390_vfenehs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpne_idx_cc(__vector __bool short __a, __vector __bool short __b,
+ int *__cc) {
+ return __builtin_s390_vfenehs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpne_idx_cc(__vector unsigned short __a, __vector unsigned short __b,
+ int *__cc) {
+ return __builtin_s390_vfenehs(__a, __b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_cmpne_idx_cc(__vector signed int __a, __vector signed int __b, int *__cc) {
+ return (__vector signed int)
+ __builtin_s390_vfenefs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpne_idx_cc(__vector __bool int __a, __vector __bool int __b, int *__cc) {
+ return __builtin_s390_vfenefs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpne_idx_cc(__vector unsigned int __a, __vector unsigned int __b,
+ int *__cc) {
+ return __builtin_s390_vfenefs(__a, __b, __cc);
+}
+
+/*-- vec_cmpne_or_0_idx -----------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_cmpne_or_0_idx(__vector signed char __a, __vector signed char __b) {
+ return (__vector signed char)
+ __builtin_s390_vfenezb((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpne_or_0_idx(__vector __bool char __a, __vector __bool char __b) {
+ return __builtin_s390_vfenezb((__vector unsigned char)__a,
+ (__vector unsigned char)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpne_or_0_idx(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vfenezb(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_cmpne_or_0_idx(__vector signed short __a, __vector signed short __b) {
+ return (__vector signed short)
+ __builtin_s390_vfenezh((__vector unsigned short)__a,
+ (__vector unsigned short)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpne_or_0_idx(__vector __bool short __a, __vector __bool short __b) {
+ return __builtin_s390_vfenezh((__vector unsigned short)__a,
+ (__vector unsigned short)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpne_or_0_idx(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vfenezh(__a, __b);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_cmpne_or_0_idx(__vector signed int __a, __vector signed int __b) {
+ return (__vector signed int)
+ __builtin_s390_vfenezf((__vector unsigned int)__a,
+ (__vector unsigned int)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpne_or_0_idx(__vector __bool int __a, __vector __bool int __b) {
+ return __builtin_s390_vfenezf((__vector unsigned int)__a,
+ (__vector unsigned int)__b);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpne_or_0_idx(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vfenezf(__a, __b);
+}
+
+/*-- vec_cmpne_or_0_idx_cc --------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_cmpne_or_0_idx_cc(__vector signed char __a, __vector signed char __b,
+ int *__cc) {
+ return (__vector signed char)
+ __builtin_s390_vfenezbs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpne_or_0_idx_cc(__vector __bool char __a, __vector __bool char __b,
+ int *__cc) {
+ return __builtin_s390_vfenezbs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpne_or_0_idx_cc(__vector unsigned char __a, __vector unsigned char __b,
+ int *__cc) {
+ return __builtin_s390_vfenezbs(__a, __b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_cmpne_or_0_idx_cc(__vector signed short __a, __vector signed short __b,
+ int *__cc) {
+ return (__vector signed short)
+ __builtin_s390_vfenezhs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpne_or_0_idx_cc(__vector __bool short __a, __vector __bool short __b,
+ int *__cc) {
+ return __builtin_s390_vfenezhs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpne_or_0_idx_cc(__vector unsigned short __a, __vector unsigned short __b,
+ int *__cc) {
+ return __builtin_s390_vfenezhs(__a, __b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_cmpne_or_0_idx_cc(__vector signed int __a, __vector signed int __b,
+ int *__cc) {
+ return (__vector signed int)
+ __builtin_s390_vfenezfs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpne_or_0_idx_cc(__vector __bool int __a, __vector __bool int __b,
+ int *__cc) {
+ return __builtin_s390_vfenezfs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpne_or_0_idx_cc(__vector unsigned int __a, __vector unsigned int __b,
+ int *__cc) {
+ return __builtin_s390_vfenezfs(__a, __b, __cc);
+}
+
+/*-- vec_cmprg --------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cmprg(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c) {
+ return (__vector __bool char)__builtin_s390_vstrcb(__a, __b, __c, 4);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cmprg(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned short __c) {
+ return (__vector __bool short)__builtin_s390_vstrch(__a, __b, __c, 4);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmprg(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned int __c) {
+ return (__vector __bool int)__builtin_s390_vstrcf(__a, __b, __c, 4);
+}
+
+/*-- vec_cmprg_cc -----------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cmprg_cc(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c, int *__cc) {
+ return (__vector __bool char)__builtin_s390_vstrcbs(__a, __b, __c, 4, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cmprg_cc(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned short __c, int *__cc) {
+ return (__vector __bool short)__builtin_s390_vstrchs(__a, __b, __c, 4, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmprg_cc(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned int __c, int *__cc) {
+ return (__vector __bool int)__builtin_s390_vstrcfs(__a, __b, __c, 4, __cc);
+}
+
+/*-- vec_cmprg_idx ----------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmprg_idx(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c) {
+ return __builtin_s390_vstrcb(__a, __b, __c, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmprg_idx(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned short __c) {
+ return __builtin_s390_vstrch(__a, __b, __c, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmprg_idx(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned int __c) {
+ return __builtin_s390_vstrcf(__a, __b, __c, 0);
+}
+
+/*-- vec_cmprg_idx_cc -------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmprg_idx_cc(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrcbs(__a, __b, __c, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmprg_idx_cc(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned short __c, int *__cc) {
+ return __builtin_s390_vstrchs(__a, __b, __c, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmprg_idx_cc(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned int __c, int *__cc) {
+ return __builtin_s390_vstrcfs(__a, __b, __c, 0, __cc);
+}
+
+/*-- vec_cmprg_or_0_idx -----------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmprg_or_0_idx(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c) {
+ return __builtin_s390_vstrczb(__a, __b, __c, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmprg_or_0_idx(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned short __c) {
+ return __builtin_s390_vstrczh(__a, __b, __c, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmprg_or_0_idx(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned int __c) {
+ return __builtin_s390_vstrczf(__a, __b, __c, 0);
+}
+
+/*-- vec_cmprg_or_0_idx_cc --------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmprg_or_0_idx_cc(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrczbs(__a, __b, __c, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmprg_or_0_idx_cc(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned short __c, int *__cc) {
+ return __builtin_s390_vstrczhs(__a, __b, __c, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmprg_or_0_idx_cc(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned int __c, int *__cc) {
+ return __builtin_s390_vstrczfs(__a, __b, __c, 0, __cc);
+}
+
+/*-- vec_cmpnrg -------------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cmpnrg(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c) {
+ return (__vector __bool char)__builtin_s390_vstrcb(__a, __b, __c, 12);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cmpnrg(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned short __c) {
+ return (__vector __bool short)__builtin_s390_vstrch(__a, __b, __c, 12);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmpnrg(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned int __c) {
+ return (__vector __bool int)__builtin_s390_vstrcf(__a, __b, __c, 12);
+}
+
+/*-- vec_cmpnrg_cc ----------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_cmpnrg_cc(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c, int *__cc) {
+ return (__vector __bool char)
+ __builtin_s390_vstrcbs(__a, __b, __c, 12, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_cmpnrg_cc(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned short __c, int *__cc) {
+ return (__vector __bool short)
+ __builtin_s390_vstrchs(__a, __b, __c, 12, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_cmpnrg_cc(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned int __c, int *__cc) {
+ return (__vector __bool int)
+ __builtin_s390_vstrcfs(__a, __b, __c, 12, __cc);
+}
+
+/*-- vec_cmpnrg_idx ---------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpnrg_idx(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c) {
+ return __builtin_s390_vstrcb(__a, __b, __c, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpnrg_idx(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned short __c) {
+ return __builtin_s390_vstrch(__a, __b, __c, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpnrg_idx(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned int __c) {
+ return __builtin_s390_vstrcf(__a, __b, __c, 8);
+}
+
+/*-- vec_cmpnrg_idx_cc ------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpnrg_idx_cc(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrcbs(__a, __b, __c, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpnrg_idx_cc(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned short __c, int *__cc) {
+ return __builtin_s390_vstrchs(__a, __b, __c, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpnrg_idx_cc(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned int __c, int *__cc) {
+ return __builtin_s390_vstrcfs(__a, __b, __c, 8, __cc);
+}
+
+/*-- vec_cmpnrg_or_0_idx ----------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpnrg_or_0_idx(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c) {
+ return __builtin_s390_vstrczb(__a, __b, __c, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpnrg_or_0_idx(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned short __c) {
+ return __builtin_s390_vstrczh(__a, __b, __c, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpnrg_or_0_idx(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned int __c) {
+ return __builtin_s390_vstrczf(__a, __b, __c, 8);
+}
+
+/*-- vec_cmpnrg_or_0_idx_cc -------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_cmpnrg_or_0_idx_cc(__vector unsigned char __a,
+ __vector unsigned char __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrczbs(__a, __b, __c, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_cmpnrg_or_0_idx_cc(__vector unsigned short __a,
+ __vector unsigned short __b,
+ __vector unsigned short __c, int *__cc) {
+ return __builtin_s390_vstrczhs(__a, __b, __c, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_cmpnrg_or_0_idx_cc(__vector unsigned int __a,
+ __vector unsigned int __b,
+ __vector unsigned int __c, int *__cc) {
+ return __builtin_s390_vstrczfs(__a, __b, __c, 8, __cc);
+}
+
+/*-- vec_find_any_eq --------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_find_any_eq(__vector signed char __a, __vector signed char __b) {
+ return (__vector __bool char)
+ __builtin_s390_vfaeb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 4);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_find_any_eq(__vector __bool char __a, __vector __bool char __b) {
+ return (__vector __bool char)
+ __builtin_s390_vfaeb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 4);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_find_any_eq(__vector unsigned char __a, __vector unsigned char __b) {
+ return (__vector __bool char)__builtin_s390_vfaeb(__a, __b, 4);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_find_any_eq(__vector signed short __a, __vector signed short __b) {
+ return (__vector __bool short)
+ __builtin_s390_vfaeh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 4);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_find_any_eq(__vector __bool short __a, __vector __bool short __b) {
+ return (__vector __bool short)
+ __builtin_s390_vfaeh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 4);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_find_any_eq(__vector unsigned short __a, __vector unsigned short __b) {
+ return (__vector __bool short)__builtin_s390_vfaeh(__a, __b, 4);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_find_any_eq(__vector signed int __a, __vector signed int __b) {
+ return (__vector __bool int)
+ __builtin_s390_vfaef((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 4);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_find_any_eq(__vector __bool int __a, __vector __bool int __b) {
+ return (__vector __bool int)
+ __builtin_s390_vfaef((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 4);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_find_any_eq(__vector unsigned int __a, __vector unsigned int __b) {
+ return (__vector __bool int)__builtin_s390_vfaef(__a, __b, 4);
+}
+
+/*-- vec_find_any_eq_cc -----------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_find_any_eq_cc(__vector signed char __a, __vector signed char __b,
+ int *__cc) {
+ return (__vector __bool char)
+ __builtin_s390_vfaebs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 4, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_find_any_eq_cc(__vector __bool char __a, __vector __bool char __b,
+ int *__cc) {
+ return (__vector __bool char)
+ __builtin_s390_vfaebs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 4, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_find_any_eq_cc(__vector unsigned char __a, __vector unsigned char __b,
+ int *__cc) {
+ return (__vector __bool char)__builtin_s390_vfaebs(__a, __b, 4, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_find_any_eq_cc(__vector signed short __a, __vector signed short __b,
+ int *__cc) {
+ return (__vector __bool short)
+ __builtin_s390_vfaehs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 4, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_find_any_eq_cc(__vector __bool short __a, __vector __bool short __b,
+ int *__cc) {
+ return (__vector __bool short)
+ __builtin_s390_vfaehs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 4, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_find_any_eq_cc(__vector unsigned short __a, __vector unsigned short __b,
+ int *__cc) {
+ return (__vector __bool short)__builtin_s390_vfaehs(__a, __b, 4, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_find_any_eq_cc(__vector signed int __a, __vector signed int __b,
+ int *__cc) {
+ return (__vector __bool int)
+ __builtin_s390_vfaefs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 4, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_find_any_eq_cc(__vector __bool int __a, __vector __bool int __b,
+ int *__cc) {
+ return (__vector __bool int)
+ __builtin_s390_vfaefs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 4, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_find_any_eq_cc(__vector unsigned int __a, __vector unsigned int __b,
+ int *__cc) {
+ return (__vector __bool int)__builtin_s390_vfaefs(__a, __b, 4, __cc);
+}
+
+/*-- vec_find_any_eq_idx ----------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_find_any_eq_idx(__vector signed char __a, __vector signed char __b) {
+ return (__vector signed char)
+ __builtin_s390_vfaeb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_eq_idx(__vector __bool char __a, __vector __bool char __b) {
+ return __builtin_s390_vfaeb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_eq_idx(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vfaeb(__a, __b, 0);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_find_any_eq_idx(__vector signed short __a, __vector signed short __b) {
+ return (__vector signed short)
+ __builtin_s390_vfaeh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_eq_idx(__vector __bool short __a, __vector __bool short __b) {
+ return __builtin_s390_vfaeh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_eq_idx(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vfaeh(__a, __b, 0);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_find_any_eq_idx(__vector signed int __a, __vector signed int __b) {
+ return (__vector signed int)
+ __builtin_s390_vfaef((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_eq_idx(__vector __bool int __a, __vector __bool int __b) {
+ return __builtin_s390_vfaef((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_eq_idx(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vfaef(__a, __b, 0);
+}
+
+/*-- vec_find_any_eq_idx_cc -------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_find_any_eq_idx_cc(__vector signed char __a,
+ __vector signed char __b, int *__cc) {
+ return (__vector signed char)
+ __builtin_s390_vfaebs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_eq_idx_cc(__vector __bool char __a,
+ __vector __bool char __b, int *__cc) {
+ return __builtin_s390_vfaebs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_eq_idx_cc(__vector unsigned char __a,
+ __vector unsigned char __b, int *__cc) {
+ return __builtin_s390_vfaebs(__a, __b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_find_any_eq_idx_cc(__vector signed short __a,
+ __vector signed short __b, int *__cc) {
+ return (__vector signed short)
+ __builtin_s390_vfaehs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_eq_idx_cc(__vector __bool short __a,
+ __vector __bool short __b, int *__cc) {
+ return __builtin_s390_vfaehs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_eq_idx_cc(__vector unsigned short __a,
+ __vector unsigned short __b, int *__cc) {
+ return __builtin_s390_vfaehs(__a, __b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_find_any_eq_idx_cc(__vector signed int __a,
+ __vector signed int __b, int *__cc) {
+ return (__vector signed int)
+ __builtin_s390_vfaefs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_eq_idx_cc(__vector __bool int __a,
+ __vector __bool int __b, int *__cc) {
+ return __builtin_s390_vfaefs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_eq_idx_cc(__vector unsigned int __a,
+ __vector unsigned int __b, int *__cc) {
+ return __builtin_s390_vfaefs(__a, __b, 0, __cc);
+}
+
+/*-- vec_find_any_eq_or_0_idx -----------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_find_any_eq_or_0_idx(__vector signed char __a,
+ __vector signed char __b) {
+ return (__vector signed char)
+ __builtin_s390_vfaezb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_eq_or_0_idx(__vector __bool char __a,
+ __vector __bool char __b) {
+ return __builtin_s390_vfaezb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_eq_or_0_idx(__vector unsigned char __a,
+ __vector unsigned char __b) {
+ return __builtin_s390_vfaezb(__a, __b, 0);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_find_any_eq_or_0_idx(__vector signed short __a,
+ __vector signed short __b) {
+ return (__vector signed short)
+ __builtin_s390_vfaezh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_eq_or_0_idx(__vector __bool short __a,
+ __vector __bool short __b) {
+ return __builtin_s390_vfaezh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_eq_or_0_idx(__vector unsigned short __a,
+ __vector unsigned short __b) {
+ return __builtin_s390_vfaezh(__a, __b, 0);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_find_any_eq_or_0_idx(__vector signed int __a,
+ __vector signed int __b) {
+ return (__vector signed int)
+ __builtin_s390_vfaezf((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_eq_or_0_idx(__vector __bool int __a,
+ __vector __bool int __b) {
+ return __builtin_s390_vfaezf((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 0);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_eq_or_0_idx(__vector unsigned int __a,
+ __vector unsigned int __b) {
+ return __builtin_s390_vfaezf(__a, __b, 0);
+}
+
+/*-- vec_find_any_eq_or_0_idx_cc --------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_find_any_eq_or_0_idx_cc(__vector signed char __a,
+ __vector signed char __b, int *__cc) {
+ return (__vector signed char)
+ __builtin_s390_vfaezbs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_eq_or_0_idx_cc(__vector __bool char __a,
+ __vector __bool char __b, int *__cc) {
+ return __builtin_s390_vfaezbs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_eq_or_0_idx_cc(__vector unsigned char __a,
+ __vector unsigned char __b, int *__cc) {
+ return __builtin_s390_vfaezbs(__a, __b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_find_any_eq_or_0_idx_cc(__vector signed short __a,
+ __vector signed short __b, int *__cc) {
+ return (__vector signed short)
+ __builtin_s390_vfaezhs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_eq_or_0_idx_cc(__vector __bool short __a,
+ __vector __bool short __b, int *__cc) {
+ return __builtin_s390_vfaezhs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_eq_or_0_idx_cc(__vector unsigned short __a,
+ __vector unsigned short __b, int *__cc) {
+ return __builtin_s390_vfaezhs(__a, __b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_find_any_eq_or_0_idx_cc(__vector signed int __a,
+ __vector signed int __b, int *__cc) {
+ return (__vector signed int)
+ __builtin_s390_vfaezfs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_eq_or_0_idx_cc(__vector __bool int __a,
+ __vector __bool int __b, int *__cc) {
+ return __builtin_s390_vfaezfs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 0, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_eq_or_0_idx_cc(__vector unsigned int __a,
+ __vector unsigned int __b, int *__cc) {
+ return __builtin_s390_vfaezfs(__a, __b, 0, __cc);
+}
+
+/*-- vec_find_any_ne --------------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_find_any_ne(__vector signed char __a, __vector signed char __b) {
+ return (__vector __bool char)
+ __builtin_s390_vfaeb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 12);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_find_any_ne(__vector __bool char __a, __vector __bool char __b) {
+ return (__vector __bool char)
+ __builtin_s390_vfaeb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 12);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_find_any_ne(__vector unsigned char __a, __vector unsigned char __b) {
+ return (__vector __bool char)__builtin_s390_vfaeb(__a, __b, 12);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_find_any_ne(__vector signed short __a, __vector signed short __b) {
+ return (__vector __bool short)
+ __builtin_s390_vfaeh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 12);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_find_any_ne(__vector __bool short __a, __vector __bool short __b) {
+ return (__vector __bool short)
+ __builtin_s390_vfaeh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 12);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_find_any_ne(__vector unsigned short __a, __vector unsigned short __b) {
+ return (__vector __bool short)__builtin_s390_vfaeh(__a, __b, 12);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_find_any_ne(__vector signed int __a, __vector signed int __b) {
+ return (__vector __bool int)
+ __builtin_s390_vfaef((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 12);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_find_any_ne(__vector __bool int __a, __vector __bool int __b) {
+ return (__vector __bool int)
+ __builtin_s390_vfaef((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 12);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_find_any_ne(__vector unsigned int __a, __vector unsigned int __b) {
+ return (__vector __bool int)__builtin_s390_vfaef(__a, __b, 12);
+}
+
+/*-- vec_find_any_ne_cc -----------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_find_any_ne_cc(__vector signed char __a,
+ __vector signed char __b, int *__cc) {
+ return (__vector __bool char)
+ __builtin_s390_vfaebs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 12, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_find_any_ne_cc(__vector __bool char __a,
+ __vector __bool char __b, int *__cc) {
+ return (__vector __bool char)
+ __builtin_s390_vfaebs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 12, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool char
+vec_find_any_ne_cc(__vector unsigned char __a,
+ __vector unsigned char __b, int *__cc) {
+ return (__vector __bool char)__builtin_s390_vfaebs(__a, __b, 12, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_find_any_ne_cc(__vector signed short __a,
+ __vector signed short __b, int *__cc) {
+ return (__vector __bool short)
+ __builtin_s390_vfaehs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 12, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_find_any_ne_cc(__vector __bool short __a,
+ __vector __bool short __b, int *__cc) {
+ return (__vector __bool short)
+ __builtin_s390_vfaehs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 12, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool short
+vec_find_any_ne_cc(__vector unsigned short __a,
+ __vector unsigned short __b, int *__cc) {
+ return (__vector __bool short)__builtin_s390_vfaehs(__a, __b, 12, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_find_any_ne_cc(__vector signed int __a,
+ __vector signed int __b, int *__cc) {
+ return (__vector __bool int)
+ __builtin_s390_vfaefs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 12, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_find_any_ne_cc(__vector __bool int __a,
+ __vector __bool int __b, int *__cc) {
+ return (__vector __bool int)
+ __builtin_s390_vfaefs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 12, __cc);
+}
+
+static inline __ATTRS_o_ai __vector __bool int
+vec_find_any_ne_cc(__vector unsigned int __a,
+ __vector unsigned int __b, int *__cc) {
+ return (__vector __bool int)__builtin_s390_vfaefs(__a, __b, 12, __cc);
+}
+
+/*-- vec_find_any_ne_idx ----------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_find_any_ne_idx(__vector signed char __a, __vector signed char __b) {
+ return (__vector signed char)
+ __builtin_s390_vfaeb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_ne_idx(__vector __bool char __a, __vector __bool char __b) {
+ return __builtin_s390_vfaeb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_ne_idx(__vector unsigned char __a, __vector unsigned char __b) {
+ return __builtin_s390_vfaeb(__a, __b, 8);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_find_any_ne_idx(__vector signed short __a, __vector signed short __b) {
+ return (__vector signed short)
+ __builtin_s390_vfaeh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_ne_idx(__vector __bool short __a, __vector __bool short __b) {
+ return __builtin_s390_vfaeh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_ne_idx(__vector unsigned short __a, __vector unsigned short __b) {
+ return __builtin_s390_vfaeh(__a, __b, 8);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_find_any_ne_idx(__vector signed int __a, __vector signed int __b) {
+ return (__vector signed int)
+ __builtin_s390_vfaef((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_ne_idx(__vector __bool int __a, __vector __bool int __b) {
+ return __builtin_s390_vfaef((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_ne_idx(__vector unsigned int __a, __vector unsigned int __b) {
+ return __builtin_s390_vfaef(__a, __b, 8);
+}
+
+/*-- vec_find_any_ne_idx_cc -------------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_find_any_ne_idx_cc(__vector signed char __a,
+ __vector signed char __b, int *__cc) {
+ return (__vector signed char)
+ __builtin_s390_vfaebs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_ne_idx_cc(__vector __bool char __a,
+ __vector __bool char __b, int *__cc) {
+ return __builtin_s390_vfaebs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_ne_idx_cc(__vector unsigned char __a,
+ __vector unsigned char __b,
+ int *__cc) {
+ return __builtin_s390_vfaebs(__a, __b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_find_any_ne_idx_cc(__vector signed short __a,
+ __vector signed short __b, int *__cc) {
+ return (__vector signed short)
+ __builtin_s390_vfaehs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_ne_idx_cc(__vector __bool short __a,
+ __vector __bool short __b, int *__cc) {
+ return __builtin_s390_vfaehs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_ne_idx_cc(__vector unsigned short __a,
+ __vector unsigned short __b, int *__cc) {
+ return __builtin_s390_vfaehs(__a, __b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_find_any_ne_idx_cc(__vector signed int __a,
+ __vector signed int __b, int *__cc) {
+ return (__vector signed int)
+ __builtin_s390_vfaefs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_ne_idx_cc(__vector __bool int __a,
+ __vector __bool int __b, int *__cc) {
+ return __builtin_s390_vfaefs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_ne_idx_cc(__vector unsigned int __a,
+ __vector unsigned int __b, int *__cc) {
+ return __builtin_s390_vfaefs(__a, __b, 8, __cc);
+}
+
+/*-- vec_find_any_ne_or_0_idx -----------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_find_any_ne_or_0_idx(__vector signed char __a,
+ __vector signed char __b) {
+ return (__vector signed char)
+ __builtin_s390_vfaezb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_ne_or_0_idx(__vector __bool char __a,
+ __vector __bool char __b) {
+ return __builtin_s390_vfaezb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_ne_or_0_idx(__vector unsigned char __a,
+ __vector unsigned char __b) {
+ return __builtin_s390_vfaezb(__a, __b, 8);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_find_any_ne_or_0_idx(__vector signed short __a,
+ __vector signed short __b) {
+ return (__vector signed short)
+ __builtin_s390_vfaezh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_ne_or_0_idx(__vector __bool short __a,
+ __vector __bool short __b) {
+ return __builtin_s390_vfaezh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_ne_or_0_idx(__vector unsigned short __a,
+ __vector unsigned short __b) {
+ return __builtin_s390_vfaezh(__a, __b, 8);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_find_any_ne_or_0_idx(__vector signed int __a,
+ __vector signed int __b) {
+ return (__vector signed int)
+ __builtin_s390_vfaezf((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_ne_or_0_idx(__vector __bool int __a,
+ __vector __bool int __b) {
+ return __builtin_s390_vfaezf((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 8);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_ne_or_0_idx(__vector unsigned int __a,
+ __vector unsigned int __b) {
+ return __builtin_s390_vfaezf(__a, __b, 8);
+}
+
+/*-- vec_find_any_ne_or_0_idx_cc --------------------------------------------*/
+
+static inline __ATTRS_o_ai __vector signed char
+vec_find_any_ne_or_0_idx_cc(__vector signed char __a,
+ __vector signed char __b, int *__cc) {
+ return (__vector signed char)
+ __builtin_s390_vfaezbs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_ne_or_0_idx_cc(__vector __bool char __a,
+ __vector __bool char __b, int *__cc) {
+ return __builtin_s390_vfaezbs((__vector unsigned char)__a,
+ (__vector unsigned char)__b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_find_any_ne_or_0_idx_cc(__vector unsigned char __a,
+ __vector unsigned char __b, int *__cc) {
+ return __builtin_s390_vfaezbs(__a, __b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed short
+vec_find_any_ne_or_0_idx_cc(__vector signed short __a,
+ __vector signed short __b, int *__cc) {
+ return (__vector signed short)
+ __builtin_s390_vfaezhs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_ne_or_0_idx_cc(__vector __bool short __a,
+ __vector __bool short __b, int *__cc) {
+ return __builtin_s390_vfaezhs((__vector unsigned short)__a,
+ (__vector unsigned short)__b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned short
+vec_find_any_ne_or_0_idx_cc(__vector unsigned short __a,
+ __vector unsigned short __b, int *__cc) {
+ return __builtin_s390_vfaezhs(__a, __b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector signed int
+vec_find_any_ne_or_0_idx_cc(__vector signed int __a,
+ __vector signed int __b, int *__cc) {
+ return (__vector signed int)
+ __builtin_s390_vfaezfs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_ne_or_0_idx_cc(__vector __bool int __a,
+ __vector __bool int __b, int *__cc) {
+ return __builtin_s390_vfaezfs((__vector unsigned int)__a,
+ (__vector unsigned int)__b, 8, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned int
+vec_find_any_ne_or_0_idx_cc(__vector unsigned int __a,
+ __vector unsigned int __b, int *__cc) {
+ return __builtin_s390_vfaezfs(__a, __b, 8, __cc);
+}
+
+/*-- vec_search_string_cc ---------------------------------------------------*/
+
+#if __ARCH__ >= 13
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_cc(__vector signed char __a, __vector signed char __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrsb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_cc(__vector __bool char __a, __vector __bool char __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrsb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_cc(__vector unsigned char __a, __vector unsigned char __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrsb(__a, __b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_cc(__vector signed short __a, __vector signed short __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrsh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_cc(__vector __bool short __a, __vector __bool short __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrsh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_cc(__vector unsigned short __a, __vector unsigned short __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrsh(__a, __b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_cc(__vector signed int __a, __vector signed int __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrsf((__vector unsigned int)__a,
+ (__vector unsigned int)__b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_cc(__vector __bool int __a, __vector __bool int __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrsf((__vector unsigned int)__a,
+ (__vector unsigned int)__b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_cc(__vector unsigned int __a, __vector unsigned int __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrsf(__a, __b, __c, __cc);
+}
+
+#endif
+
+/*-- vec_search_string_until_zero_cc ----------------------------------------*/
+
+#if __ARCH__ >= 13
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_until_zero_cc(__vector signed char __a,
+ __vector signed char __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrszb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_until_zero_cc(__vector __bool char __a,
+ __vector __bool char __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrszb((__vector unsigned char)__a,
+ (__vector unsigned char)__b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_until_zero_cc(__vector unsigned char __a,
+ __vector unsigned char __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrszb(__a, __b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_until_zero_cc(__vector signed short __a,
+ __vector signed short __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrszh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_until_zero_cc(__vector __bool short __a,
+ __vector __bool short __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrszh((__vector unsigned short)__a,
+ (__vector unsigned short)__b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_until_zero_cc(__vector unsigned short __a,
+ __vector unsigned short __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrszh(__a, __b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_until_zero_cc(__vector signed int __a,
+ __vector signed int __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrszf((__vector unsigned int)__a,
+ (__vector unsigned int)__b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_until_zero_cc(__vector __bool int __a,
+ __vector __bool int __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrszf((__vector unsigned int)__a,
+ (__vector unsigned int)__b, __c, __cc);
+}
+
+static inline __ATTRS_o_ai __vector unsigned char
+vec_search_string_until_zero_cc(__vector unsigned int __a,
+ __vector unsigned int __b,
+ __vector unsigned char __c, int *__cc) {
+ return __builtin_s390_vstrszf(__a, __b, __c, __cc);
+}
+
+#endif
+
+#undef __constant_pow2_range
+#undef __constant_range
+#undef __constant
+#undef __ATTRS_o
+#undef __ATTRS_o_ai
+#undef __ATTRS_ai
+
+#else
+
+#error "Use -fzvector to enable vector extensions"
+
+#endif
--- /dev/null
+/*===------------ vpclmulqdqintrin.h - VPCLMULQDQ intrinsics ---------------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error "Never use <vpclmulqdqintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __VPCLMULQDQINTRIN_H
+#define __VPCLMULQDQINTRIN_H
+
+#define _mm256_clmulepi64_epi128(A, B, I) \
+ (__m256i)__builtin_ia32_pclmulqdq256((__v4di)(__m256i)(A), \
+ (__v4di)(__m256i)(B), \
+ (char)(I))
+
+#define _mm512_clmulepi64_epi128(A, B, I) \
+ (__m512i)__builtin_ia32_pclmulqdq512((__v8di)(__m512i)(A), \
+ (__v8di)(__m512i)(B), \
+ (char)(I))
+
+#endif /* __VPCLMULQDQINTRIN_H */
+
--- /dev/null
+/*===----------------------- waitpkgintrin.h - WAITPKG --------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#if !defined __X86INTRIN_H && !defined __IMMINTRIN_H
+#error "Never use <waitpkgintrin.h> directly; include <x86intrin.h> instead."
+#endif
+
+#ifndef __WAITPKGINTRIN_H
+#define __WAITPKGINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("waitpkg")))
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_umonitor (void * __address)
+{
+ __builtin_ia32_umonitor (__address);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_umwait (unsigned int __control, unsigned long long __counter)
+{
+ return __builtin_ia32_umwait (__control,
+ (unsigned int)(__counter >> 32), (unsigned int)__counter);
+}
+
+static __inline__ unsigned char __DEFAULT_FN_ATTRS
+_tpause (unsigned int __control, unsigned long long __counter)
+{
+ return __builtin_ia32_tpause (__control,
+ (unsigned int)(__counter >> 32), (unsigned int)__counter);
+}
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif /* __WAITPKGINTRIN_H */
--- /dev/null
+/*===---- wasm_simd128.h - WebAssembly portable SIMD intrinsics ------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __WASM_SIMD128_H
+#define __WASM_SIMD128_H
+
+#include <stdbool.h>
+#include <stdint.h>
+
+// User-facing type
+typedef int32_t v128_t __attribute__((__vector_size__(16), __aligned__(16)));
+
+// Internal types determined by clang builtin definitions
+typedef int32_t __v128_u __attribute__((__vector_size__(16), __aligned__(1)));
+typedef char __i8x16 __attribute__((__vector_size__(16), __aligned__(16)));
+typedef signed char __s8x16
+ __attribute__((__vector_size__(16), __aligned__(16)));
+typedef unsigned char __u8x16
+ __attribute__((__vector_size__(16), __aligned__(16)));
+typedef short __i16x8 __attribute__((__vector_size__(16), __aligned__(16)));
+typedef unsigned short __u16x8
+ __attribute__((__vector_size__(16), __aligned__(16)));
+typedef int __i32x4 __attribute__((__vector_size__(16), __aligned__(16)));
+typedef unsigned int __u32x4
+ __attribute__((__vector_size__(16), __aligned__(16)));
+typedef long long __i64x2 __attribute__((__vector_size__(16), __aligned__(16)));
+typedef unsigned long long __u64x2
+ __attribute__((__vector_size__(16), __aligned__(16)));
+typedef float __f32x4 __attribute__((__vector_size__(16), __aligned__(16)));
+typedef double __f64x2 __attribute__((__vector_size__(16), __aligned__(16)));
+
+#define __DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("simd128"), \
+ __min_vector_width__(128)))
+
+#define __REQUIRE_CONSTANT(e) \
+ _Static_assert(__builtin_constant_p(e), "Expected constant")
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_v128_load(const void *__mem) {
+ // UB-free unaligned access copied from xmmintrin.h
+ struct __wasm_v128_load_struct {
+ __v128_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ return ((const struct __wasm_v128_load_struct *)__mem)->__v;
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_v8x16_load_splat(const void *__mem) {
+ struct __wasm_v8x16_load_splat_struct {
+ uint8_t __v;
+ } __attribute__((__packed__, __may_alias__));
+ uint8_t __v = ((const struct __wasm_v8x16_load_splat_struct *)__mem)->__v;
+ return (v128_t)(__u8x16){__v, __v, __v, __v, __v, __v, __v, __v,
+ __v, __v, __v, __v, __v, __v, __v, __v};
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_v16x8_load_splat(const void *__mem) {
+ struct __wasm_v16x8_load_splat_struct {
+ uint16_t __v;
+ } __attribute__((__packed__, __may_alias__));
+ uint16_t __v = ((const struct __wasm_v16x8_load_splat_struct *)__mem)->__v;
+ return (v128_t)(__u16x8){__v, __v, __v, __v, __v, __v, __v, __v};
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_v32x4_load_splat(const void *__mem) {
+ struct __wasm_v32x4_load_splat_struct {
+ uint32_t __v;
+ } __attribute__((__packed__, __may_alias__));
+ uint32_t __v = ((const struct __wasm_v32x4_load_splat_struct *)__mem)->__v;
+ return (v128_t)(__u32x4){__v, __v, __v, __v};
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_v64x2_load_splat(const void *__mem) {
+ struct __wasm_v64x2_load_splat_struct {
+ uint64_t __v;
+ } __attribute__((__packed__, __may_alias__));
+ uint64_t __v = ((const struct __wasm_v64x2_load_splat_struct *)__mem)->__v;
+ return (v128_t)(__u64x2){__v, __v};
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i16x8_load_8x8(const void *__mem) {
+ typedef int8_t __i8x8 __attribute__((__vector_size__(8), __aligned__(8)));
+ struct __wasm_i16x8_load_8x8_struct {
+ __i8x8 __v;
+ } __attribute__((__packed__, __may_alias__));
+ __i8x8 __v = ((const struct __wasm_i16x8_load_8x8_struct *)__mem)->__v;
+ return (v128_t) __builtin_convertvector(__v, __i16x8);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_u16x8_load_8x8(const void *__mem) {
+ typedef uint8_t __u8x8 __attribute__((__vector_size__(8), __aligned__(8)));
+ struct __wasm_u16x8_load_8x8_struct {
+ __u8x8 __v;
+ } __attribute__((__packed__, __may_alias__));
+ __u8x8 __v = ((const struct __wasm_u16x8_load_8x8_struct *)__mem)->__v;
+ return (v128_t) __builtin_convertvector(__v, __u16x8);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i32x4_load_16x4(const void *__mem) {
+ typedef int16_t __i16x4 __attribute__((__vector_size__(8), __aligned__(8)));
+ struct __wasm_i32x4_load_16x4_struct {
+ __i16x4 __v;
+ } __attribute__((__packed__, __may_alias__));
+ __i16x4 __v = ((const struct __wasm_i32x4_load_16x4_struct *)__mem)->__v;
+ return (v128_t) __builtin_convertvector(__v, __i32x4);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_u32x4_load_16x4(const void *__mem) {
+ typedef uint16_t __u16x4 __attribute__((__vector_size__(8), __aligned__(8)));
+ struct __wasm_u32x4_load_16x4_struct {
+ __u16x4 __v;
+ } __attribute__((__packed__, __may_alias__));
+ __u16x4 __v = ((const struct __wasm_u32x4_load_16x4_struct *)__mem)->__v;
+ return (v128_t) __builtin_convertvector(__v, __u32x4);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i64x2_load_32x2(const void *__mem) {
+ typedef int32_t __i32x2 __attribute__((__vector_size__(8), __aligned__(8)));
+ struct __wasm_i64x2_load_32x2_struct {
+ __i32x2 __v;
+ } __attribute__((__packed__, __may_alias__));
+ __i32x2 __v = ((const struct __wasm_i64x2_load_32x2_struct *)__mem)->__v;
+ return (v128_t) __builtin_convertvector(__v, __i64x2);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_u64x2_load_32x2(const void *__mem) {
+ typedef uint32_t __u32x2 __attribute__((__vector_size__(8), __aligned__(8)));
+ struct __wasm_u64x2_load_32x2_struct {
+ __u32x2 __v;
+ } __attribute__((__packed__, __may_alias__));
+ __u32x2 __v = ((const struct __wasm_u64x2_load_32x2_struct *)__mem)->__v;
+ return (v128_t) __builtin_convertvector(__v, __u64x2);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS wasm_v128_store(void *__mem,
+ v128_t __a) {
+ // UB-free unaligned access copied from xmmintrin.h
+ struct __wasm_v128_store_struct {
+ __v128_u __v;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __wasm_v128_store_struct *)__mem)->__v = __a;
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i8x16_make(int8_t __c0, int8_t __c1, int8_t __c2, int8_t __c3, int8_t __c4,
+ int8_t __c5, int8_t __c6, int8_t __c7, int8_t __c8, int8_t __c9,
+ int8_t __c10, int8_t __c11, int8_t __c12, int8_t __c13,
+ int8_t __c14, int8_t __c15) {
+ return (v128_t)(__i8x16){__c0, __c1, __c2, __c3, __c4, __c5,
+ __c6, __c7, __c8, __c9, __c10, __c11,
+ __c12, __c13, __c14, __c15};
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i16x8_make(int16_t __c0, int16_t __c1, int16_t __c2, int16_t __c3,
+ int16_t __c4, int16_t __c5, int16_t __c6, int16_t __c7) {
+ return (v128_t)(__i16x8){__c0, __c1, __c2, __c3, __c4, __c5, __c6, __c7};
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_make(int32_t __c0,
+ int32_t __c1,
+ int32_t __c2,
+ int32_t __c3) {
+ return (v128_t)(__i32x4){__c0, __c1, __c2, __c3};
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_make(float __c0,
+ float __c1,
+ float __c2,
+ float __c3) {
+ return (v128_t)(__f32x4){__c0, __c1, __c2, __c3};
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i64x2_make(int64_t __c0,
+ int64_t __c1) {
+ return (v128_t)(__i64x2){__c0, __c1};
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_make(double __c0,
+ double __c1) {
+ return (v128_t)(__f64x2){__c0, __c1};
+}
+
+#define wasm_i8x16_const(__c0, __c1, __c2, __c3, __c4, __c5, __c6, __c7, __c8, \
+ __c9, __c10, __c11, __c12, __c13, __c14, __c15) \
+ __extension__({ \
+ __REQUIRE_CONSTANT(__c0); \
+ __REQUIRE_CONSTANT(__c1); \
+ __REQUIRE_CONSTANT(__c2); \
+ __REQUIRE_CONSTANT(__c3); \
+ __REQUIRE_CONSTANT(__c4); \
+ __REQUIRE_CONSTANT(__c5); \
+ __REQUIRE_CONSTANT(__c6); \
+ __REQUIRE_CONSTANT(__c7); \
+ __REQUIRE_CONSTANT(__c8); \
+ __REQUIRE_CONSTANT(__c9); \
+ __REQUIRE_CONSTANT(__c10); \
+ __REQUIRE_CONSTANT(__c11); \
+ __REQUIRE_CONSTANT(__c12); \
+ __REQUIRE_CONSTANT(__c13); \
+ __REQUIRE_CONSTANT(__c14); \
+ __REQUIRE_CONSTANT(__c15); \
+ (v128_t)(__i8x16){__c0, __c1, __c2, __c3, __c4, __c5, __c6, __c7, \
+ __c8, __c9, __c10, __c11, __c12, __c13, __c14, __c15}; \
+ })
+
+#define wasm_i16x8_const(__c0, __c1, __c2, __c3, __c4, __c5, __c6, __c7) \
+ __extension__({ \
+ __REQUIRE_CONSTANT(__c0); \
+ __REQUIRE_CONSTANT(__c1); \
+ __REQUIRE_CONSTANT(__c2); \
+ __REQUIRE_CONSTANT(__c3); \
+ __REQUIRE_CONSTANT(__c4); \
+ __REQUIRE_CONSTANT(__c5); \
+ __REQUIRE_CONSTANT(__c6); \
+ __REQUIRE_CONSTANT(__c7); \
+ (v128_t)(__i16x8){__c0, __c1, __c2, __c3, __c4, __c5, __c6, __c7}; \
+ })
+
+#define wasm_i32x4_const(__c0, __c1, __c2, __c3) \
+ __extension__({ \
+ __REQUIRE_CONSTANT(__c0); \
+ __REQUIRE_CONSTANT(__c1); \
+ __REQUIRE_CONSTANT(__c2); \
+ __REQUIRE_CONSTANT(__c3); \
+ (v128_t)(__i32x4){__c0, __c1, __c2, __c3}; \
+ })
+
+#define wasm_f32x4_const(__c0, __c1, __c2, __c3) \
+ __extension__({ \
+ __REQUIRE_CONSTANT(__c0); \
+ __REQUIRE_CONSTANT(__c1); \
+ __REQUIRE_CONSTANT(__c2); \
+ __REQUIRE_CONSTANT(__c3); \
+ (v128_t)(__f32x4){__c0, __c1, __c2, __c3}; \
+ })
+
+#define wasm_i64x2_const(__c0, __c1) \
+ __extension__({ \
+ __REQUIRE_CONSTANT(__c0); \
+ __REQUIRE_CONSTANT(__c1); \
+ (v128_t)(__i64x2){__c0, __c1}; \
+ })
+
+#define wasm_f64x2_const(__c0, __c1) \
+ __extension__({ \
+ __REQUIRE_CONSTANT(__c0); \
+ __REQUIRE_CONSTANT(__c1); \
+ (v128_t)(__f64x2){__c0, __c1}; \
+ })
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i8x16_splat(int8_t __a) {
+ return (v128_t)(__i8x16){__a, __a, __a, __a, __a, __a, __a, __a,
+ __a, __a, __a, __a, __a, __a, __a, __a};
+}
+
+#define wasm_i8x16_extract_lane(__a, __i) \
+ (__builtin_wasm_extract_lane_s_i8x16((__i8x16)(__a), __i))
+
+#define wasm_u8x16_extract_lane(__a, __i) \
+ (__builtin_wasm_extract_lane_u_i8x16((__i8x16)(__a), __i))
+
+#define wasm_i8x16_replace_lane(__a, __i, __b) \
+ ((v128_t)__builtin_wasm_replace_lane_i8x16((__i8x16)(__a), __i, __b))
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_splat(int16_t __a) {
+ return (v128_t)(__i16x8){__a, __a, __a, __a, __a, __a, __a, __a};
+}
+
+#define wasm_i16x8_extract_lane(__a, __i) \
+ (__builtin_wasm_extract_lane_s_i16x8((__i16x8)(__a), __i))
+
+#define wasm_u16x8_extract_lane(__a, __i) \
+ (__builtin_wasm_extract_lane_u_i16x8((__i16x8)(__a), __i))
+
+#define wasm_i16x8_replace_lane(__a, __i, __b) \
+ ((v128_t)__builtin_wasm_replace_lane_i16x8((__i16x8)(__a), __i, __b))
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_splat(int32_t __a) {
+ return (v128_t)(__i32x4){__a, __a, __a, __a};
+}
+
+#define wasm_i32x4_extract_lane(__a, __i) \
+ (__builtin_wasm_extract_lane_i32x4((__i32x4)(__a), __i))
+
+#define wasm_i32x4_replace_lane(__a, __i, __b) \
+ ((v128_t)__builtin_wasm_replace_lane_i32x4((__i32x4)(__a), __i, __b))
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i64x2_splat(int64_t __a) {
+ return (v128_t)(__i64x2){__a, __a};
+}
+
+#define wasm_i64x2_extract_lane(__a, __i) \
+ (__builtin_wasm_extract_lane_i64x2((__i64x2)(__a), __i))
+
+#define wasm_i64x2_replace_lane(__a, __i, __b) \
+ ((v128_t)__builtin_wasm_replace_lane_i64x2((__i64x2)(__a), __i, __b))
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_splat(float __a) {
+ return (v128_t)(__f32x4){__a, __a, __a, __a};
+}
+
+#define wasm_f32x4_extract_lane(__a, __i) \
+ (__builtin_wasm_extract_lane_f32x4((__f32x4)(__a), __i))
+
+#define wasm_f32x4_replace_lane(__a, __i, __b) \
+ ((v128_t)__builtin_wasm_replace_lane_f32x4((__f32x4)(__a), __i, __b))
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_splat(double __a) {
+ return (v128_t)(__f64x2){__a, __a};
+}
+
+#define wasm_f64x2_extract_lane(__a, __i) \
+ (__builtin_wasm_extract_lane_f64x2((__f64x2)(__a), __i))
+
+#define wasm_f64x2_replace_lane(__a, __i, __b) \
+ ((v128_t)__builtin_wasm_replace_lane_f64x2((__f64x2)(__a), __i, __b))
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i8x16_eq(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__s8x16)__a == (__s8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i8x16_ne(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__s8x16)__a != (__s8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i8x16_lt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__s8x16)__a < (__s8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u8x16_lt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u8x16)__a < (__u8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i8x16_gt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__s8x16)__a > (__s8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u8x16_gt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u8x16)__a > (__u8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i8x16_le(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__s8x16)__a <= (__s8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u8x16_le(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u8x16)__a <= (__u8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i8x16_ge(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__s8x16)__a >= (__s8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u8x16_ge(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u8x16)__a >= (__u8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_eq(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__i16x8)__a == (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_ne(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u16x8)__a != (__u16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_lt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__i16x8)__a < (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u16x8_lt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u16x8)__a < (__u16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_gt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__i16x8)__a > (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u16x8_gt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u16x8)__a > (__u16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_le(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__i16x8)__a <= (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u16x8_le(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u16x8)__a <= (__u16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_ge(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__i16x8)__a >= (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u16x8_ge(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u16x8)__a >= (__u16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_eq(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__i32x4)__a == (__i32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_ne(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__i32x4)__a != (__i32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_lt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__i32x4)__a < (__i32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u32x4_lt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u32x4)__a < (__u32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_gt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__i32x4)__a > (__i32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u32x4_gt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u32x4)__a > (__u32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_le(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__i32x4)__a <= (__i32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u32x4_le(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u32x4)__a <= (__u32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_ge(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__i32x4)__a >= (__i32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u32x4_ge(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u32x4)__a >= (__u32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_eq(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f32x4)__a == (__f32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_ne(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f32x4)__a != (__f32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_lt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f32x4)__a < (__f32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_gt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f32x4)__a > (__f32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_le(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f32x4)__a <= (__f32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_ge(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f32x4)__a >= (__f32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_eq(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f64x2)__a == (__f64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_ne(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f64x2)__a != (__f64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_lt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f64x2)__a < (__f64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_gt(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f64x2)__a > (__f64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_le(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f64x2)__a <= (__f64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_ge(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f64x2)__a >= (__f64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_v128_not(v128_t __a) {
+ return ~__a;
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_v128_and(v128_t __a,
+ v128_t __b) {
+ return __a & __b;
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_v128_or(v128_t __a,
+ v128_t __b) {
+ return __a | __b;
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_v128_xor(v128_t __a,
+ v128_t __b) {
+ return __a ^ __b;
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_v128_andnot(v128_t __a,
+ v128_t __b) {
+ return __a & ~__b;
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_v128_bitselect(v128_t __a,
+ v128_t __b,
+ v128_t __mask) {
+ return (v128_t)__builtin_wasm_bitselect((__i32x4)__a, (__i32x4)__b,
+ (__i32x4)__mask);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i8x16_abs(v128_t __a) {
+ return (v128_t)__builtin_wasm_abs_i8x16((__i8x16)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i8x16_neg(v128_t __a) {
+ return (v128_t)(-(__u8x16)__a);
+}
+
+static __inline__ bool __DEFAULT_FN_ATTRS wasm_i8x16_any_true(v128_t __a) {
+ return __builtin_wasm_any_true_i8x16((__i8x16)__a);
+}
+
+static __inline__ bool __DEFAULT_FN_ATTRS wasm_i8x16_all_true(v128_t __a) {
+ return __builtin_wasm_all_true_i8x16((__i8x16)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i8x16_shl(v128_t __a,
+ int32_t __b) {
+ return (v128_t)((__i8x16)__a << __b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i8x16_shr(v128_t __a,
+ int32_t __b) {
+ return (v128_t)((__s8x16)__a >> __b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u8x16_shr(v128_t __a,
+ int32_t __b) {
+ return (v128_t)((__u8x16)__a >> __b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i8x16_add(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u8x16)__a + (__u8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i8x16_add_saturate(v128_t __a, v128_t __b) {
+ return (v128_t)__builtin_wasm_add_saturate_s_i8x16((__i8x16)__a,
+ (__i8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_u8x16_add_saturate(v128_t __a, v128_t __b) {
+ return (v128_t)__builtin_wasm_add_saturate_u_i8x16((__i8x16)__a,
+ (__i8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i8x16_sub(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u8x16)__a - (__u8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i8x16_sub_saturate(v128_t __a, v128_t __b) {
+ return (v128_t)__builtin_wasm_sub_saturate_s_i8x16((__i8x16)__a,
+ (__i8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_u8x16_sub_saturate(v128_t __a, v128_t __b) {
+ return (v128_t)__builtin_wasm_sub_saturate_u_i8x16((__i8x16)__a,
+ (__i8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i8x16_min(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_min_s_i8x16((__i8x16)__a, (__i8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u8x16_min(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_min_u_i8x16((__i8x16)__a, (__i8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i8x16_max(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_max_s_i8x16((__i8x16)__a, (__i8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u8x16_max(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_max_u_i8x16((__i8x16)__a, (__i8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u8x16_avgr(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_avgr_u_i8x16((__i8x16)__a, (__i8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_abs(v128_t __a) {
+ return (v128_t)__builtin_wasm_abs_i16x8((__i16x8)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_neg(v128_t __a) {
+ return (v128_t)(-(__u16x8)__a);
+}
+
+static __inline__ bool __DEFAULT_FN_ATTRS wasm_i16x8_any_true(v128_t __a) {
+ return __builtin_wasm_any_true_i16x8((__i16x8)__a);
+}
+
+static __inline__ bool __DEFAULT_FN_ATTRS wasm_i16x8_all_true(v128_t __a) {
+ return __builtin_wasm_all_true_i16x8((__i16x8)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_shl(v128_t __a,
+ int32_t __b) {
+ return (v128_t)((__i16x8)__a << __b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_shr(v128_t __a,
+ int32_t __b) {
+ return (v128_t)((__i16x8)__a >> __b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u16x8_shr(v128_t __a,
+ int32_t __b) {
+ return (v128_t)((__u16x8)__a >> __b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_add(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u16x8)__a + (__u16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i16x8_add_saturate(v128_t __a, v128_t __b) {
+ return (v128_t)__builtin_wasm_add_saturate_s_i16x8((__i16x8)__a,
+ (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_u16x8_add_saturate(v128_t __a, v128_t __b) {
+ return (v128_t)__builtin_wasm_add_saturate_u_i16x8((__i16x8)__a,
+ (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_sub(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__i16x8)__a - (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i16x8_sub_saturate(v128_t __a, v128_t __b) {
+ return (v128_t)__builtin_wasm_sub_saturate_s_i16x8((__i16x8)__a,
+ (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_u16x8_sub_saturate(v128_t __a, v128_t __b) {
+ return (v128_t)__builtin_wasm_sub_saturate_u_i16x8((__i16x8)__a,
+ (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_mul(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u16x8)__a * (__u16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_min(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_min_s_i16x8((__i16x8)__a, (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u16x8_min(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_min_u_i16x8((__i16x8)__a, (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i16x8_max(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_max_s_i16x8((__i16x8)__a, (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u16x8_max(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_max_u_i16x8((__i16x8)__a, (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u16x8_avgr(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_avgr_u_i16x8((__i16x8)__a, (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_abs(v128_t __a) {
+ return (v128_t)__builtin_wasm_abs_i32x4((__i32x4)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_neg(v128_t __a) {
+ return (v128_t)(-(__u32x4)__a);
+}
+
+static __inline__ bool __DEFAULT_FN_ATTRS wasm_i32x4_any_true(v128_t __a) {
+ return __builtin_wasm_any_true_i32x4((__i32x4)__a);
+}
+
+static __inline__ bool __DEFAULT_FN_ATTRS wasm_i32x4_all_true(v128_t __a) {
+ return __builtin_wasm_all_true_i32x4((__i32x4)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_shl(v128_t __a,
+ int32_t __b) {
+ return (v128_t)((__i32x4)__a << __b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_shr(v128_t __a,
+ int32_t __b) {
+ return (v128_t)((__i32x4)__a >> __b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u32x4_shr(v128_t __a,
+ int32_t __b) {
+ return (v128_t)((__u32x4)__a >> __b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_add(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u32x4)__a + (__u32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_sub(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u32x4)__a - (__u32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_mul(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u32x4)__a * (__u32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_min(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_min_s_i32x4((__i32x4)__a, (__i32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u32x4_min(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_min_u_i32x4((__i32x4)__a, (__i32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i32x4_max(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_max_s_i32x4((__i32x4)__a, (__i32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u32x4_max(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_max_u_i32x4((__i32x4)__a, (__i32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i64x2_neg(v128_t __a) {
+ return (v128_t)(-(__u64x2)__a);
+}
+
+#ifdef __wasm_unimplemented_simd128__
+
+static __inline__ bool __DEFAULT_FN_ATTRS wasm_i64x2_any_true(v128_t __a) {
+ return __builtin_wasm_any_true_i64x2((__i64x2)__a);
+}
+
+static __inline__ bool __DEFAULT_FN_ATTRS wasm_i64x2_all_true(v128_t __a) {
+ return __builtin_wasm_all_true_i64x2((__i64x2)__a);
+}
+
+#endif // __wasm_unimplemented_simd128__
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i64x2_shl(v128_t __a,
+ int32_t __b) {
+ return (v128_t)((__i64x2)__a << (int64_t)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i64x2_shr(v128_t __a,
+ int32_t __b) {
+ return (v128_t)((__i64x2)__a >> (int64_t)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_u64x2_shr(v128_t __a,
+ int32_t __b) {
+ return (v128_t)((__u64x2)__a >> (int64_t)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i64x2_add(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u64x2)__a + (__u64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i64x2_sub(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u64x2)__a - (__u64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_i64x2_mul(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__u64x2)__a * (__u64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_abs(v128_t __a) {
+ return (v128_t)__builtin_wasm_abs_f32x4((__f32x4)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_neg(v128_t __a) {
+ return (v128_t)(-(__f32x4)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_sqrt(v128_t __a) {
+ return (v128_t)__builtin_wasm_sqrt_f32x4((__f32x4)__a);
+}
+
+#ifdef __wasm_unimplemented_simd128__
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_qfma(v128_t __a,
+ v128_t __b,
+ v128_t __c) {
+ return (v128_t)__builtin_wasm_qfma_f32x4((__f32x4)__a, (__f32x4)__b,
+ (__f32x4)__c);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_qfms(v128_t __a,
+ v128_t __b,
+ v128_t __c) {
+ return (v128_t)__builtin_wasm_qfms_f32x4((__f32x4)__a, (__f32x4)__b,
+ (__f32x4)__c);
+}
+
+#endif // __wasm_unimplemented_simd128__
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_add(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f32x4)__a + (__f32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_sub(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f32x4)__a - (__f32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_mul(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f32x4)__a * (__f32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_div(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f32x4)__a / (__f32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_min(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_min_f32x4((__f32x4)__a, (__f32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_max(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_max_f32x4((__f32x4)__a, (__f32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_pmin(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_pmin_f32x4((__f32x4)__a, (__f32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f32x4_pmax(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_pmax_f32x4((__f32x4)__a, (__f32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_abs(v128_t __a) {
+ return (v128_t)__builtin_wasm_abs_f64x2((__f64x2)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_neg(v128_t __a) {
+ return (v128_t)(-(__f64x2)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_sqrt(v128_t __a) {
+ return (v128_t)__builtin_wasm_sqrt_f64x2((__f64x2)__a);
+}
+
+#ifdef __wasm_unimplemented_simd128__
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_qfma(v128_t __a,
+ v128_t __b,
+ v128_t __c) {
+ return (v128_t)__builtin_wasm_qfma_f64x2((__f64x2)__a, (__f64x2)__b,
+ (__f64x2)__c);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_qfms(v128_t __a,
+ v128_t __b,
+ v128_t __c) {
+ return (v128_t)__builtin_wasm_qfms_f64x2((__f64x2)__a, (__f64x2)__b,
+ (__f64x2)__c);
+}
+
+#endif // __wasm_unimplemented_simd128__
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_add(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f64x2)__a + (__f64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_sub(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f64x2)__a - (__f64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_mul(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f64x2)__a * (__f64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_div(v128_t __a,
+ v128_t __b) {
+ return (v128_t)((__f64x2)__a / (__f64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_min(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_min_f64x2((__f64x2)__a, (__f64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_max(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_max_f64x2((__f64x2)__a, (__f64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_pmin(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_pmin_f64x2((__f64x2)__a, (__f64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_f64x2_pmax(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_pmax_f64x2((__f64x2)__a, (__f64x2)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i32x4_trunc_saturate_f32x4(v128_t __a) {
+ return (v128_t)__builtin_wasm_trunc_saturate_s_i32x4_f32x4((__f32x4)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_u32x4_trunc_saturate_f32x4(v128_t __a) {
+ return (v128_t)__builtin_wasm_trunc_saturate_u_i32x4_f32x4((__f32x4)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_f32x4_convert_i32x4(v128_t __a) {
+ return (v128_t) __builtin_convertvector((__i32x4)__a, __f32x4);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_f32x4_convert_u32x4(v128_t __a) {
+ return (v128_t) __builtin_convertvector((__u32x4)__a, __f32x4);
+}
+
+#define wasm_v8x16_shuffle(__a, __b, __c0, __c1, __c2, __c3, __c4, __c5, __c6, \
+ __c7, __c8, __c9, __c10, __c11, __c12, __c13, \
+ __c14, __c15) \
+ ((v128_t)__builtin_wasm_shuffle_v8x16( \
+ (__i8x16)(__a), (__i8x16)(__b), __c0, __c1, __c2, __c3, __c4, __c5, \
+ __c6, __c7, __c8, __c9, __c10, __c11, __c12, __c13, __c14, __c15))
+
+#define wasm_v16x8_shuffle(__a, __b, __c0, __c1, __c2, __c3, __c4, __c5, __c6, \
+ __c7) \
+ ((v128_t)__builtin_wasm_shuffle_v8x16( \
+ (__i8x16)(__a), (__i8x16)(__b), (__c0)*2, (__c0)*2 + 1, (__c1)*2, \
+ (__c1)*2 + 1, (__c2)*2, (__c2)*2 + 1, (__c3)*2, (__c3)*2 + 1, (__c4)*2, \
+ (__c4)*2 + 1, (__c5)*2, (__c5)*2 + 1, (__c6)*2, (__c6)*2 + 1, (__c7)*2, \
+ (__c7)*2 + 1))
+
+#define wasm_v32x4_shuffle(__a, __b, __c0, __c1, __c2, __c3) \
+ ((v128_t)__builtin_wasm_shuffle_v8x16( \
+ (__i8x16)(__a), (__i8x16)(__b), (__c0)*4, (__c0)*4 + 1, (__c0)*4 + 2, \
+ (__c0)*4 + 3, (__c1)*4, (__c1)*4 + 1, (__c1)*4 + 2, (__c1)*4 + 3, \
+ (__c2)*4, (__c2)*4 + 1, (__c2)*4 + 2, (__c2)*4 + 3, (__c3)*4, \
+ (__c3)*4 + 1, (__c3)*4 + 2, (__c3)*4 + 3))
+
+#define wasm_v64x2_shuffle(__a, __b, __c0, __c1) \
+ ((v128_t)__builtin_wasm_shuffle_v8x16( \
+ (__i8x16)(__a), (__i8x16)(__b), (__c0)*8, (__c0)*8 + 1, (__c0)*8 + 2, \
+ (__c0)*8 + 3, (__c0)*8 + 4, (__c0)*8 + 5, (__c0)*8 + 6, (__c0)*8 + 7, \
+ (__c1)*8, (__c1)*8 + 1, (__c1)*8 + 2, (__c1)*8 + 3, (__c1)*8 + 4, \
+ (__c1)*8 + 5, (__c1)*8 + 6, (__c1)*8 + 7))
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS wasm_v8x16_swizzle(v128_t __a,
+ v128_t __b) {
+ return (v128_t)__builtin_wasm_swizzle_v8x16((__i8x16)__a, (__i8x16)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i8x16_narrow_i16x8(v128_t __a, v128_t __b) {
+ return (v128_t)__builtin_wasm_narrow_s_i8x16_i16x8((__i16x8)__a,
+ (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_u8x16_narrow_i16x8(v128_t __a, v128_t __b) {
+ return (v128_t)__builtin_wasm_narrow_u_i8x16_i16x8((__i16x8)__a,
+ (__i16x8)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i16x8_narrow_i32x4(v128_t __a, v128_t __b) {
+ return (v128_t)__builtin_wasm_narrow_s_i16x8_i32x4((__i32x4)__a,
+ (__i32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_u16x8_narrow_i32x4(v128_t __a, v128_t __b) {
+ return (v128_t)__builtin_wasm_narrow_u_i16x8_i32x4((__i32x4)__a,
+ (__i32x4)__b);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i16x8_widen_low_i8x16(v128_t __a) {
+ return (v128_t)__builtin_wasm_widen_low_s_i16x8_i8x16((__i8x16)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i16x8_widen_high_i8x16(v128_t __a) {
+ return (v128_t)__builtin_wasm_widen_high_s_i16x8_i8x16((__i8x16)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i16x8_widen_low_u8x16(v128_t __a) {
+ return (v128_t)__builtin_wasm_widen_low_u_i16x8_i8x16((__i8x16)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i16x8_widen_high_u8x16(v128_t __a) {
+ return (v128_t)__builtin_wasm_widen_high_u_i16x8_i8x16((__i8x16)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i32x4_widen_low_i16x8(v128_t __a) {
+ return (v128_t)__builtin_wasm_widen_low_s_i32x4_i16x8((__i16x8)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i32x4_widen_high_i16x8(v128_t __a) {
+ return (v128_t)__builtin_wasm_widen_high_s_i32x4_i16x8((__i16x8)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i32x4_widen_low_u16x8(v128_t __a) {
+ return (v128_t)__builtin_wasm_widen_low_u_i32x4_i16x8((__i16x8)__a);
+}
+
+static __inline__ v128_t __DEFAULT_FN_ATTRS
+wasm_i32x4_widen_high_u16x8(v128_t __a) {
+ return (v128_t)__builtin_wasm_widen_high_u_i32x4_i16x8((__i16x8)__a);
+}
+
+// Undefine helper macros
+#undef __DEFAULT_FN_ATTRS
+
+#endif // __WASM_SIMD128_H
--- /dev/null
+/*===-------------- wbnoinvdintrin.h - wbnoinvd intrinsic-------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#if !defined __X86INTRIN_H && !defined __IMMINTRIN_H
+#error "Never use <wbnoinvdintrin.h> directly; include <x86intrin.h> instead."
+#endif
+
+#ifndef __WBNOINVDINTRIN_H
+#define __WBNOINVDINTRIN_H
+
+static __inline__ void
+ __attribute__((__always_inline__, __nodebug__, __target__("wbnoinvd")))
+_wbnoinvd (void)
+{
+ __builtin_ia32_wbnoinvd ();
+}
+
+#endif /* __WBNOINVDINTRIN_H */
--- /dev/null
+/*===---- wmmintrin.h - AES intrinsics ------------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __WMMINTRIN_H
+#define __WMMINTRIN_H
+
+#include <emmintrin.h>
+
+#include <__wmmintrin_aes.h>
+
+#include <__wmmintrin_pclmul.h>
+
+#endif /* __WMMINTRIN_H */
--- /dev/null
+/*===---- x86intrin.h - X86 intrinsics -------------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __X86INTRIN_H
+#define __X86INTRIN_H
+
+#include <ia32intrin.h>
+
+#include <immintrin.h>
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__3dNOW__)
+#include <mm3dnow.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__PRFCHW__)
+#include <prfchwintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__SSE4A__)
+#include <ammintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__FMA4__)
+#include <fma4intrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__XOP__)
+#include <xopintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__TBM__)
+#include <tbmintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__LWP__)
+#include <lwpintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__MWAITX__)
+#include <mwaitxintrin.h>
+#endif
+
+#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \
+ defined(__CLZERO__)
+#include <clzerointrin.h>
+#endif
+
+
+#endif /* __X86INTRIN_H */
/*===---- xmmintrin.h - SSE intrinsics -------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
typedef int __v4si __attribute__((__vector_size__(16)));
typedef float __v4sf __attribute__((__vector_size__(16)));
-typedef float __m128 __attribute__((__vector_size__(16)));
+typedef float __m128 __attribute__((__vector_size__(16), __aligned__(16)));
+
+typedef float __m128_u __attribute__((__vector_size__(16), __aligned__(1)));
/* Unsigned types */
typedef unsigned int __v4su __attribute__((__vector_size__(16)));
#endif
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("sse")))
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("sse"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS_MMX __attribute__((__always_inline__, __nodebug__, __target__("mmx,sse"), __min_vector_width__(64)))
-/// \brief Adds the 32-bit float values in the low-order bits of the operands.
+/// Adds the 32-bit float values in the low-order bits of the operands.
///
/// \headerfile <x86intrin.h>
///
return __a;
}
-/// \brief Adds two 128-bit vectors of [4 x float], and returns the results of
+/// Adds two 128-bit vectors of [4 x float], and returns the results of
/// the addition.
///
/// \headerfile <x86intrin.h>
return (__m128)((__v4sf)__a + (__v4sf)__b);
}
-/// \brief Subtracts the 32-bit float value in the low-order bits of the second
+/// Subtracts the 32-bit float value in the low-order bits of the second
/// operand from the corresponding value in the first operand.
///
/// \headerfile <x86intrin.h>
return __a;
}
-/// \brief Subtracts each of the values of the second operand from the first
+/// Subtracts each of the values of the second operand from the first
/// operand, both of which are 128-bit vectors of [4 x float] and returns
/// the results of the subtraction.
///
return (__m128)((__v4sf)__a - (__v4sf)__b);
}
-/// \brief Multiplies two 32-bit float values in the low-order bits of the
+/// Multiplies two 32-bit float values in the low-order bits of the
/// operands.
///
/// \headerfile <x86intrin.h>
return __a;
}
-/// \brief Multiplies two 128-bit vectors of [4 x float] and returns the
+/// Multiplies two 128-bit vectors of [4 x float] and returns the
/// results of the multiplication.
///
/// \headerfile <x86intrin.h>
return (__m128)((__v4sf)__a * (__v4sf)__b);
}
-/// \brief Divides the value in the low-order 32 bits of the first operand by
+/// Divides the value in the low-order 32 bits of the first operand by
/// the corresponding value in the second operand.
///
/// \headerfile <x86intrin.h>
return __a;
}
-/// \brief Divides two 128-bit vectors of [4 x float].
+/// Divides two 128-bit vectors of [4 x float].
///
/// \headerfile <x86intrin.h>
///
return (__m128)((__v4sf)__a / (__v4sf)__b);
}
-/// \brief Calculates the square root of the value stored in the low-order bits
+/// Calculates the square root of the value stored in the low-order bits
/// of a 128-bit vector of [4 x float].
///
/// \headerfile <x86intrin.h>
static __inline__ __m128 __DEFAULT_FN_ATTRS
_mm_sqrt_ss(__m128 __a)
{
- __m128 __c = __builtin_ia32_sqrtss((__v4sf)__a);
- return (__m128) { __c[0], __a[1], __a[2], __a[3] };
+ return (__m128)__builtin_ia32_sqrtss((__v4sf)__a);
}
-/// \brief Calculates the square roots of the values stored in a 128-bit vector
+/// Calculates the square roots of the values stored in a 128-bit vector
/// of [4 x float].
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_sqrtps((__v4sf)__a);
}
-/// \brief Calculates the approximate reciprocal of the value stored in the
+/// Calculates the approximate reciprocal of the value stored in the
/// low-order bits of a 128-bit vector of [4 x float].
///
/// \headerfile <x86intrin.h>
static __inline__ __m128 __DEFAULT_FN_ATTRS
_mm_rcp_ss(__m128 __a)
{
- __m128 __c = __builtin_ia32_rcpss((__v4sf)__a);
- return (__m128) { __c[0], __a[1], __a[2], __a[3] };
+ return (__m128)__builtin_ia32_rcpss((__v4sf)__a);
}
-/// \brief Calculates the approximate reciprocals of the values stored in a
+/// Calculates the approximate reciprocals of the values stored in a
/// 128-bit vector of [4 x float].
///
/// \headerfile <x86intrin.h>
static __inline__ __m128 __DEFAULT_FN_ATTRS
_mm_rcp_ps(__m128 __a)
{
- return __builtin_ia32_rcpps((__v4sf)__a);
+ return (__m128)__builtin_ia32_rcpps((__v4sf)__a);
}
-/// \brief Calculates the approximate reciprocal of the square root of the value
+/// Calculates the approximate reciprocal of the square root of the value
/// stored in the low-order bits of a 128-bit vector of [4 x float].
///
/// \headerfile <x86intrin.h>
static __inline__ __m128 __DEFAULT_FN_ATTRS
_mm_rsqrt_ss(__m128 __a)
{
- __m128 __c = __builtin_ia32_rsqrtss((__v4sf)__a);
- return (__m128) { __c[0], __a[1], __a[2], __a[3] };
+ return __builtin_ia32_rsqrtss((__v4sf)__a);
}
-/// \brief Calculates the approximate reciprocals of the square roots of the
+/// Calculates the approximate reciprocals of the square roots of the
/// values stored in a 128-bit vector of [4 x float].
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_rsqrtps((__v4sf)__a);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands and returns the lesser value in the low-order bits of the
/// vector of [4 x float].
///
return __builtin_ia32_minss((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 128-bit vectors of [4 x float] and returns the lesser
+/// Compares two 128-bit vectors of [4 x float] and returns the lesser
/// of each pair of values.
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_minps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands and returns the greater value in the low-order bits of a 128-bit
/// vector of [4 x float].
///
return __builtin_ia32_maxss((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 128-bit vectors of [4 x float] and returns the greater
+/// Compares two 128-bit vectors of [4 x float] and returns the greater
/// of each pair of values.
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_maxps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Performs a bitwise AND of two 128-bit vectors of [4 x float].
+/// Performs a bitwise AND of two 128-bit vectors of [4 x float].
///
/// \headerfile <x86intrin.h>
///
return (__m128)((__v4su)__a & (__v4su)__b);
}
-/// \brief Performs a bitwise AND of two 128-bit vectors of [4 x float], using
+/// Performs a bitwise AND of two 128-bit vectors of [4 x float], using
/// the one's complement of the values contained in the first source
/// operand.
///
return (__m128)(~(__v4su)__a & (__v4su)__b);
}
-/// \brief Performs a bitwise OR of two 128-bit vectors of [4 x float].
+/// Performs a bitwise OR of two 128-bit vectors of [4 x float].
///
/// \headerfile <x86intrin.h>
///
return (__m128)((__v4su)__a | (__v4su)__b);
}
-/// \brief Performs a bitwise exclusive OR of two 128-bit vectors of
+/// Performs a bitwise exclusive OR of two 128-bit vectors of
/// [4 x float].
///
/// \headerfile <x86intrin.h>
return (__m128)((__v4su)__a ^ (__v4su)__b);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands for equality and returns the result of the comparison in the
/// low-order bits of a vector [4 x float].
///
return (__m128)__builtin_ia32_cmpeqss((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares each of the corresponding 32-bit float values of the
+/// Compares each of the corresponding 32-bit float values of the
/// 128-bit vectors of [4 x float] for equality.
///
/// \headerfile <x86intrin.h>
return (__m128)__builtin_ia32_cmpeqps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands to determine if the value in the first operand is less than the
/// corresponding value in the second operand and returns the result of the
/// comparison in the low-order bits of a vector of [4 x float].
return (__m128)__builtin_ia32_cmpltss((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares each of the corresponding 32-bit float values of the
+/// Compares each of the corresponding 32-bit float values of the
/// 128-bit vectors of [4 x float] to determine if the values in the first
/// operand are less than those in the second operand.
///
return (__m128)__builtin_ia32_cmpltps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands to determine if the value in the first operand is less than or
/// equal to the corresponding value in the second operand and returns the
/// result of the comparison in the low-order bits of a vector of
return (__m128)__builtin_ia32_cmpless((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares each of the corresponding 32-bit float values of the
+/// Compares each of the corresponding 32-bit float values of the
/// 128-bit vectors of [4 x float] to determine if the values in the first
/// operand are less than or equal to those in the second operand.
///
return (__m128)__builtin_ia32_cmpleps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands to determine if the value in the first operand is greater than
/// the corresponding value in the second operand and returns the result of
/// the comparison in the low-order bits of a vector of [4 x float].
4, 1, 2, 3);
}
-/// \brief Compares each of the corresponding 32-bit float values of the
+/// Compares each of the corresponding 32-bit float values of the
/// 128-bit vectors of [4 x float] to determine if the values in the first
/// operand are greater than those in the second operand.
///
return (__m128)__builtin_ia32_cmpltps((__v4sf)__b, (__v4sf)__a);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands to determine if the value in the first operand is greater than
/// or equal to the corresponding value in the second operand and returns
/// the result of the comparison in the low-order bits of a vector of
4, 1, 2, 3);
}
-/// \brief Compares each of the corresponding 32-bit float values of the
+/// Compares each of the corresponding 32-bit float values of the
/// 128-bit vectors of [4 x float] to determine if the values in the first
/// operand are greater than or equal to those in the second operand.
///
return (__m128)__builtin_ia32_cmpleps((__v4sf)__b, (__v4sf)__a);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands for inequality and returns the result of the comparison in the
/// low-order bits of a vector of [4 x float].
///
return (__m128)__builtin_ia32_cmpneqss((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares each of the corresponding 32-bit float values of the
+/// Compares each of the corresponding 32-bit float values of the
/// 128-bit vectors of [4 x float] for inequality.
///
/// \headerfile <x86intrin.h>
return (__m128)__builtin_ia32_cmpneqps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands to determine if the value in the first operand is not less than
/// the corresponding value in the second operand and returns the result of
/// the comparison in the low-order bits of a vector of [4 x float].
return (__m128)__builtin_ia32_cmpnltss((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares each of the corresponding 32-bit float values of the
+/// Compares each of the corresponding 32-bit float values of the
/// 128-bit vectors of [4 x float] to determine if the values in the first
/// operand are not less than those in the second operand.
///
return (__m128)__builtin_ia32_cmpnltps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands to determine if the value in the first operand is not less than
/// or equal to the corresponding value in the second operand and returns
/// the result of the comparison in the low-order bits of a vector of
return (__m128)__builtin_ia32_cmpnless((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares each of the corresponding 32-bit float values of the
+/// Compares each of the corresponding 32-bit float values of the
/// 128-bit vectors of [4 x float] to determine if the values in the first
/// operand are not less than or equal to those in the second operand.
///
return (__m128)__builtin_ia32_cmpnleps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands to determine if the value in the first operand is not greater
/// than the corresponding value in the second operand and returns the
/// result of the comparison in the low-order bits of a vector of
4, 1, 2, 3);
}
-/// \brief Compares each of the corresponding 32-bit float values of the
+/// Compares each of the corresponding 32-bit float values of the
/// 128-bit vectors of [4 x float] to determine if the values in the first
/// operand are not greater than those in the second operand.
///
return (__m128)__builtin_ia32_cmpnltps((__v4sf)__b, (__v4sf)__a);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands to determine if the value in the first operand is not greater
/// than or equal to the corresponding value in the second operand and
/// returns the result of the comparison in the low-order bits of a vector
4, 1, 2, 3);
}
-/// \brief Compares each of the corresponding 32-bit float values of the
+/// Compares each of the corresponding 32-bit float values of the
/// 128-bit vectors of [4 x float] to determine if the values in the first
/// operand are not greater than or equal to those in the second operand.
///
return (__m128)__builtin_ia32_cmpnleps((__v4sf)__b, (__v4sf)__a);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands to determine if the value in the first operand is ordered with
/// respect to the corresponding value in the second operand and returns the
/// result of the comparison in the low-order bits of a vector of
return (__m128)__builtin_ia32_cmpordss((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares each of the corresponding 32-bit float values of the
+/// Compares each of the corresponding 32-bit float values of the
/// 128-bit vectors of [4 x float] to determine if the values in the first
/// operand are ordered with respect to those in the second operand.
///
return (__m128)__builtin_ia32_cmpordps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands to determine if the value in the first operand is unordered
/// with respect to the corresponding value in the second operand and
/// returns the result of the comparison in the low-order bits of a vector
return (__m128)__builtin_ia32_cmpunordss((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares each of the corresponding 32-bit float values of the
+/// Compares each of the corresponding 32-bit float values of the
/// 128-bit vectors of [4 x float] to determine if the values in the first
/// operand are unordered with respect to those in the second operand.
///
return (__m128)__builtin_ia32_cmpunordps((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands for equality and returns the result of the comparison.
///
+/// If either of the two lower 32-bit values is NaN, 0 is returned.
+///
/// \headerfile <x86intrin.h>
///
/// This intrinsic corresponds to the <c> VCOMISS / COMISS </c>
/// \param __b
/// A 128-bit vector of [4 x float]. The lower 32 bits of this operand are
/// used in the comparison.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the
+/// two lower 32-bit values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_comieq_ss(__m128 __a, __m128 __b)
{
return __builtin_ia32_comieq((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands to determine if the first operand is less than the second
/// operand and returns the result of the comparison.
///
+/// If either of the two lower 32-bit values is NaN, 0 is returned.
+///
/// \headerfile <x86intrin.h>
///
/// This intrinsic corresponds to the <c> VCOMISS / COMISS </c>
/// \param __b
/// A 128-bit vector of [4 x float]. The lower 32 bits of this operand are
/// used in the comparison.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the two
+/// lower 32-bit values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_comilt_ss(__m128 __a, __m128 __b)
{
return __builtin_ia32_comilt((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands to determine if the first operand is less than or equal to the
/// second operand and returns the result of the comparison.
///
+/// If either of the two lower 32-bit values is NaN, 0 is returned.
+///
/// \headerfile <x86intrin.h>
///
/// This intrinsic corresponds to the <c> VCOMISS / COMISS </c> instructions.
/// \param __b
/// A 128-bit vector of [4 x float]. The lower 32 bits of this operand are
/// used in the comparison.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the two
+/// lower 32-bit values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_comile_ss(__m128 __a, __m128 __b)
{
return __builtin_ia32_comile((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands to determine if the first operand is greater than the second
/// operand and returns the result of the comparison.
///
+/// If either of the two lower 32-bit values is NaN, 0 is returned.
+///
/// \headerfile <x86intrin.h>
///
/// This intrinsic corresponds to the <c> VCOMISS / COMISS </c> instructions.
/// \param __b
/// A 128-bit vector of [4 x float]. The lower 32 bits of this operand are
/// used in the comparison.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the
+/// two lower 32-bit values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_comigt_ss(__m128 __a, __m128 __b)
{
return __builtin_ia32_comigt((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands to determine if the first operand is greater than or equal to
/// the second operand and returns the result of the comparison.
///
+/// If either of the two lower 32-bit values is NaN, 0 is returned.
+///
/// \headerfile <x86intrin.h>
///
/// This intrinsic corresponds to the <c> VCOMISS / COMISS </c> instructions.
/// \param __b
/// A 128-bit vector of [4 x float]. The lower 32 bits of this operand are
/// used in the comparison.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the two
+/// lower 32-bit values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_comige_ss(__m128 __a, __m128 __b)
{
return __builtin_ia32_comige((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Compares two 32-bit float values in the low-order bits of both
+/// Compares two 32-bit float values in the low-order bits of both
/// operands to determine if the first operand is not equal to the second
/// operand and returns the result of the comparison.
///
+/// If either of the two lower 32-bit values is NaN, 1 is returned.
+///
/// \headerfile <x86intrin.h>
///
/// This intrinsic corresponds to the <c> VCOMISS / COMISS </c> instructions.
/// \param __b
/// A 128-bit vector of [4 x float]. The lower 32 bits of this operand are
/// used in the comparison.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the
+/// two lower 32-bit values is NaN, 1 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_comineq_ss(__m128 __a, __m128 __b)
{
return __builtin_ia32_comineq((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Performs an unordered comparison of two 32-bit float values using
+/// Performs an unordered comparison of two 32-bit float values using
/// the low-order bits of both operands to determine equality and returns
/// the result of the comparison.
///
+/// If either of the two lower 32-bit values is NaN, 0 is returned.
+///
/// \headerfile <x86intrin.h>
///
/// This intrinsic corresponds to the <c> VUCOMISS / UCOMISS </c> instructions.
/// \param __b
/// A 128-bit vector of [4 x float]. The lower 32 bits of this operand are
/// used in the comparison.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the two
+/// lower 32-bit values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_ucomieq_ss(__m128 __a, __m128 __b)
{
return __builtin_ia32_ucomieq((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Performs an unordered comparison of two 32-bit float values using
+/// Performs an unordered comparison of two 32-bit float values using
/// the low-order bits of both operands to determine if the first operand is
/// less than the second operand and returns the result of the comparison.
///
+/// If either of the two lower 32-bit values is NaN, 0 is returned.
+///
/// \headerfile <x86intrin.h>
///
/// This intrinsic corresponds to the <c> VUCOMISS / UCOMISS </c> instructions.
/// \param __b
/// A 128-bit vector of [4 x float]. The lower 32 bits of this operand are
/// used in the comparison.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the two
+/// lower 32-bit values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_ucomilt_ss(__m128 __a, __m128 __b)
{
return __builtin_ia32_ucomilt((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Performs an unordered comparison of two 32-bit float values using
+/// Performs an unordered comparison of two 32-bit float values using
/// the low-order bits of both operands to determine if the first operand is
/// less than or equal to the second operand and returns the result of the
/// comparison.
///
+/// If either of the two lower 32-bit values is NaN, 0 is returned.
+///
/// \headerfile <x86intrin.h>
///
/// This intrinsic corresponds to the <c> VUCOMISS / UCOMISS </c> instructions.
/// \param __b
/// A 128-bit vector of [4 x float]. The lower 32 bits of this operand are
/// used in the comparison.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the two
+/// lower 32-bit values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_ucomile_ss(__m128 __a, __m128 __b)
{
return __builtin_ia32_ucomile((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Performs an unordered comparison of two 32-bit float values using
+/// Performs an unordered comparison of two 32-bit float values using
/// the low-order bits of both operands to determine if the first operand is
/// greater than the second operand and returns the result of the
/// comparison.
///
+/// If either of the two lower 32-bit values is NaN, 0 is returned.
+///
/// \headerfile <x86intrin.h>
///
/// This intrinsic corresponds to the <c> VUCOMISS / UCOMISS </c> instructions.
/// \param __b
/// A 128-bit vector of [4 x float]. The lower 32 bits of this operand are
/// used in the comparison.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the two
+/// lower 32-bit values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_ucomigt_ss(__m128 __a, __m128 __b)
{
return __builtin_ia32_ucomigt((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Performs an unordered comparison of two 32-bit float values using
+/// Performs an unordered comparison of two 32-bit float values using
/// the low-order bits of both operands to determine if the first operand is
/// greater than or equal to the second operand and returns the result of
/// the comparison.
///
+/// If either of the two lower 32-bit values is NaN, 0 is returned.
+///
/// \headerfile <x86intrin.h>
///
/// This intrinsic corresponds to the <c> VUCOMISS / UCOMISS </c> instructions.
/// \param __b
/// A 128-bit vector of [4 x float]. The lower 32 bits of this operand are
/// used in the comparison.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the two
+/// lower 32-bit values is NaN, 0 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_ucomige_ss(__m128 __a, __m128 __b)
{
return __builtin_ia32_ucomige((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Performs an unordered comparison of two 32-bit float values using
+/// Performs an unordered comparison of two 32-bit float values using
/// the low-order bits of both operands to determine inequality and returns
/// the result of the comparison.
///
+/// If either of the two lower 32-bit values is NaN, 1 is returned.
+///
/// \headerfile <x86intrin.h>
///
/// This intrinsic corresponds to the <c> VUCOMISS / UCOMISS </c> instructions.
/// \param __b
/// A 128-bit vector of [4 x float]. The lower 32 bits of this operand are
/// used in the comparison.
-/// \returns An integer containing the comparison results.
+/// \returns An integer containing the comparison results. If either of the two
+/// lower 32-bit values is NaN, 1 is returned.
static __inline__ int __DEFAULT_FN_ATTRS
_mm_ucomineq_ss(__m128 __a, __m128 __b)
{
return __builtin_ia32_ucomineq((__v4sf)__a, (__v4sf)__b);
}
-/// \brief Converts a float value contained in the lower 32 bits of a vector of
+/// Converts a float value contained in the lower 32 bits of a vector of
/// [4 x float] into a 32-bit integer.
///
/// \headerfile <x86intrin.h>
return __builtin_ia32_cvtss2si((__v4sf)__a);
}
-/// \brief Converts a float value contained in the lower 32 bits of a vector of
+/// Converts a float value contained in the lower 32 bits of a vector of
/// [4 x float] into a 32-bit integer.
///
/// \headerfile <x86intrin.h>
#ifdef __x86_64__
-/// \brief Converts a float value contained in the lower 32 bits of a vector of
+/// Converts a float value contained in the lower 32 bits of a vector of
/// [4 x float] into a 64-bit integer.
///
/// \headerfile <x86intrin.h>
#endif
-/// \brief Converts two low-order float values in a 128-bit vector of
+/// Converts two low-order float values in a 128-bit vector of
/// [4 x float] into a 64-bit vector of [2 x i32].
///
/// \headerfile <x86intrin.h>
/// \param __a
/// A 128-bit vector of [4 x float].
/// \returns A 64-bit integer vector containing the converted values.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_cvtps_pi32(__m128 __a)
{
return (__m64)__builtin_ia32_cvtps2pi((__v4sf)__a);
}
-/// \brief Converts two low-order float values in a 128-bit vector of
+/// Converts two low-order float values in a 128-bit vector of
/// [4 x float] into a 64-bit vector of [2 x i32].
///
/// \headerfile <x86intrin.h>
/// \param __a
/// A 128-bit vector of [4 x float].
/// \returns A 64-bit integer vector containing the converted values.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_cvt_ps2pi(__m128 __a)
{
return _mm_cvtps_pi32(__a);
}
-/// \brief Converts a float value contained in the lower 32 bits of a vector of
+/// Converts a float value contained in the lower 32 bits of a vector of
/// [4 x float] into a 32-bit integer, truncating the result when it is
/// inexact.
///
return __builtin_ia32_cvttss2si((__v4sf)__a);
}
-/// \brief Converts a float value contained in the lower 32 bits of a vector of
+/// Converts a float value contained in the lower 32 bits of a vector of
/// [4 x float] into a 32-bit integer, truncating the result when it is
/// inexact.
///
}
#ifdef __x86_64__
-/// \brief Converts a float value contained in the lower 32 bits of a vector of
+/// Converts a float value contained in the lower 32 bits of a vector of
/// [4 x float] into a 64-bit integer, truncating the result when it is
/// inexact.
///
}
#endif
-/// \brief Converts two low-order float values in a 128-bit vector of
+/// Converts two low-order float values in a 128-bit vector of
/// [4 x float] into a 64-bit vector of [2 x i32], truncating the result
/// when it is inexact.
///
/// \param __a
/// A 128-bit vector of [4 x float].
/// \returns A 64-bit integer vector containing the converted values.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_cvttps_pi32(__m128 __a)
{
return (__m64)__builtin_ia32_cvttps2pi((__v4sf)__a);
}
-/// \brief Converts two low-order float values in a 128-bit vector of [4 x
+/// Converts two low-order float values in a 128-bit vector of [4 x
/// float] into a 64-bit vector of [2 x i32], truncating the result when it
/// is inexact.
///
/// \param __a
/// A 128-bit vector of [4 x float].
/// \returns A 64-bit integer vector containing the converted values.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_cvtt_ps2pi(__m128 __a)
{
return _mm_cvttps_pi32(__a);
}
-/// \brief Converts a 32-bit signed integer value into a floating point value
+/// Converts a 32-bit signed integer value into a floating point value
/// and writes it to the lower 32 bits of the destination. The remaining
/// higher order elements of the destination vector are copied from the
/// corresponding elements in the first operand.
return __a;
}
-/// \brief Converts a 32-bit signed integer value into a floating point value
+/// Converts a 32-bit signed integer value into a floating point value
/// and writes it to the lower 32 bits of the destination. The remaining
/// higher order elements of the destination are copied from the
/// corresponding elements in the first operand.
#ifdef __x86_64__
-/// \brief Converts a 64-bit signed integer value into a floating point value
+/// Converts a 64-bit signed integer value into a floating point value
/// and writes it to the lower 32 bits of the destination. The remaining
/// higher order elements of the destination are copied from the
/// corresponding elements in the first operand.
#endif
-/// \brief Converts two elements of a 64-bit vector of [2 x i32] into two
+/// Converts two elements of a 64-bit vector of [2 x i32] into two
/// floating point values and writes them to the lower 64-bits of the
/// destination. The remaining higher order elements of the destination are
/// copied from the corresponding elements in the first operand.
/// \returns A 128-bit vector of [4 x float] whose lower 64 bits contain the
/// converted value of the second operand. The upper 64 bits are copied from
/// the upper 64 bits of the first operand.
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS_MMX
_mm_cvtpi32_ps(__m128 __a, __m64 __b)
{
return __builtin_ia32_cvtpi2ps((__v4sf)__a, (__v2si)__b);
}
-/// \brief Converts two elements of a 64-bit vector of [2 x i32] into two
+/// Converts two elements of a 64-bit vector of [2 x i32] into two
/// floating point values and writes them to the lower 64-bits of the
/// destination. The remaining higher order elements of the destination are
/// copied from the corresponding elements in the first operand.
/// \returns A 128-bit vector of [4 x float] whose lower 64 bits contain the
/// converted value from the second operand. The upper 64 bits are copied
/// from the upper 64 bits of the first operand.
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS_MMX
_mm_cvt_pi2ps(__m128 __a, __m64 __b)
{
return _mm_cvtpi32_ps(__a, __b);
}
-/// \brief Extracts a float value contained in the lower 32 bits of a vector of
+/// Extracts a float value contained in the lower 32 bits of a vector of
/// [4 x float].
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c> VMOVSS / MOVSS </c> instruction.
+/// This intrinsic has no corresponding instruction.
///
/// \param __a
/// A 128-bit vector of [4 x float]. The lower 32 bits of this operand are
return __a[0];
}
-/// \brief Loads two packed float values from the address \a __p into the
+/// Loads two packed float values from the address \a __p into the
/// high-order bits of a 128-bit vector of [4 x float]. The low-order bits
/// are copied from the low-order bits of the first operand.
///
struct __mm_loadh_pi_struct {
__mm_loadh_pi_v2f32 __u;
} __attribute__((__packed__, __may_alias__));
- __mm_loadh_pi_v2f32 __b = ((struct __mm_loadh_pi_struct*)__p)->__u;
+ __mm_loadh_pi_v2f32 __b = ((const struct __mm_loadh_pi_struct*)__p)->__u;
__m128 __bb = __builtin_shufflevector(__b, __b, 0, 1, 0, 1);
return __builtin_shufflevector(__a, __bb, 0, 1, 4, 5);
}
-/// \brief Loads two packed float values from the address \a __p into the
+/// Loads two packed float values from the address \a __p into the
/// low-order bits of a 128-bit vector of [4 x float]. The high-order bits
/// are copied from the high-order bits of the first operand.
///
struct __mm_loadl_pi_struct {
__mm_loadl_pi_v2f32 __u;
} __attribute__((__packed__, __may_alias__));
- __mm_loadl_pi_v2f32 __b = ((struct __mm_loadl_pi_struct*)__p)->__u;
+ __mm_loadl_pi_v2f32 __b = ((const struct __mm_loadl_pi_struct*)__p)->__u;
__m128 __bb = __builtin_shufflevector(__b, __b, 0, 1, 0, 1);
return __builtin_shufflevector(__a, __bb, 4, 5, 2, 3);
}
-/// \brief Constructs a 128-bit floating-point vector of [4 x float]. The lower
+/// Constructs a 128-bit floating-point vector of [4 x float]. The lower
/// 32 bits of the vector are initialized with the single-precision
/// floating-point value loaded from a specified memory location. The upper
/// 96 bits are set to zero.
struct __mm_load_ss_struct {
float __u;
} __attribute__((__packed__, __may_alias__));
- float __u = ((struct __mm_load_ss_struct*)__p)->__u;
- return (__m128){ __u, 0, 0, 0 };
+ float __u = ((const struct __mm_load_ss_struct*)__p)->__u;
+ return __extension__ (__m128){ __u, 0, 0, 0 };
}
-/// \brief Loads a 32-bit float value and duplicates it to all four vector
+/// Loads a 32-bit float value and duplicates it to all four vector
/// elements of a 128-bit vector of [4 x float].
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c> VMOVSS / MOVSS + shuffling </c>
+/// This intrinsic corresponds to the <c> VBROADCASTSS / MOVSS + shuffling </c>
/// instruction.
///
/// \param __p
struct __mm_load1_ps_struct {
float __u;
} __attribute__((__packed__, __may_alias__));
- float __u = ((struct __mm_load1_ps_struct*)__p)->__u;
- return (__m128){ __u, __u, __u, __u };
+ float __u = ((const struct __mm_load1_ps_struct*)__p)->__u;
+ return __extension__ (__m128){ __u, __u, __u, __u };
}
#define _mm_load_ps1(p) _mm_load1_ps(p)
-/// \brief Loads a 128-bit floating-point vector of [4 x float] from an aligned
+/// Loads a 128-bit floating-point vector of [4 x float] from an aligned
/// memory location.
///
/// \headerfile <x86intrin.h>
/// \param __p
/// A pointer to a 128-bit memory location. The address of the memory
/// location has to be 128-bit aligned.
-/// \returns A 128-bit vector of [4 x float] containing the loaded valus.
+/// \returns A 128-bit vector of [4 x float] containing the loaded values.
static __inline__ __m128 __DEFAULT_FN_ATTRS
_mm_load_ps(const float *__p)
{
- return *(__m128*)__p;
+ return *(const __m128*)__p;
}
-/// \brief Loads a 128-bit floating-point vector of [4 x float] from an
+/// Loads a 128-bit floating-point vector of [4 x float] from an
/// unaligned memory location.
///
/// \headerfile <x86intrin.h>
_mm_loadu_ps(const float *__p)
{
struct __loadu_ps {
- __m128 __v;
+ __m128_u __v;
} __attribute__((__packed__, __may_alias__));
- return ((struct __loadu_ps*)__p)->__v;
+ return ((const struct __loadu_ps*)__p)->__v;
}
-/// \brief Loads four packed float values, in reverse order, from an aligned
+/// Loads four packed float values, in reverse order, from an aligned
/// memory location to 32-bit elements in a 128-bit vector of [4 x float].
///
/// \headerfile <x86intrin.h>
return __builtin_shufflevector((__v4sf)__a, (__v4sf)__a, 3, 2, 1, 0);
}
-/// \brief Create a 128-bit vector of [4 x float] with undefined values.
+/// Create a 128-bit vector of [4 x float] with undefined values.
///
/// \headerfile <x86intrin.h>
///
return (__m128)__builtin_ia32_undef128();
}
-/// \brief Constructs a 128-bit floating-point vector of [4 x float]. The lower
+/// Constructs a 128-bit floating-point vector of [4 x float]. The lower
/// 32 bits of the vector are initialized with the specified single-precision
/// floating-point value. The upper 96 bits are set to zero.
///
static __inline__ __m128 __DEFAULT_FN_ATTRS
_mm_set_ss(float __w)
{
- return (__m128){ __w, 0, 0, 0 };
+ return __extension__ (__m128){ __w, 0, 0, 0 };
}
-/// \brief Constructs a 128-bit floating-point vector of [4 x float], with each
+/// Constructs a 128-bit floating-point vector of [4 x float], with each
/// of the four single-precision floating-point vector elements set to the
/// specified single-precision floating-point value.
///
static __inline__ __m128 __DEFAULT_FN_ATTRS
_mm_set1_ps(float __w)
{
- return (__m128){ __w, __w, __w, __w };
+ return __extension__ (__m128){ __w, __w, __w, __w };
}
/* Microsoft specific. */
-/// \brief Constructs a 128-bit floating-point vector of [4 x float], with each
+/// Constructs a 128-bit floating-point vector of [4 x float], with each
/// of the four single-precision floating-point vector elements set to the
/// specified single-precision floating-point value.
///
return _mm_set1_ps(__w);
}
-/// \brief Constructs a 128-bit floating-point vector of [4 x float]
+/// Constructs a 128-bit floating-point vector of [4 x float]
/// initialized with the specified single-precision floating-point values.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128 __DEFAULT_FN_ATTRS
_mm_set_ps(float __z, float __y, float __x, float __w)
{
- return (__m128){ __w, __x, __y, __z };
+ return __extension__ (__m128){ __w, __x, __y, __z };
}
-/// \brief Constructs a 128-bit floating-point vector of [4 x float],
+/// Constructs a 128-bit floating-point vector of [4 x float],
/// initialized in reverse order with the specified 32-bit single-precision
/// float-point values.
///
static __inline__ __m128 __DEFAULT_FN_ATTRS
_mm_setr_ps(float __z, float __y, float __x, float __w)
{
- return (__m128){ __z, __y, __x, __w };
+ return __extension__ (__m128){ __z, __y, __x, __w };
}
-/// \brief Constructs a 128-bit floating-point vector of [4 x float] initialized
+/// Constructs a 128-bit floating-point vector of [4 x float] initialized
/// to zero.
///
/// \headerfile <x86intrin.h>
static __inline__ __m128 __DEFAULT_FN_ATTRS
_mm_setzero_ps(void)
{
- return (__m128){ 0, 0, 0, 0 };
+ return __extension__ (__m128){ 0, 0, 0, 0 };
}
-/// \brief Stores the upper 64 bits of a 128-bit vector of [4 x float] to a
+/// Stores the upper 64 bits of a 128-bit vector of [4 x float] to a
/// memory location.
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c> VPEXTRQ / MOVQ </c> instruction.
+/// This intrinsic corresponds to the <c> VPEXTRQ / PEXTRQ </c> instruction.
///
/// \param __p
/// A pointer to a 64-bit memory location.
static __inline__ void __DEFAULT_FN_ATTRS
_mm_storeh_pi(__m64 *__p, __m128 __a)
{
- __builtin_ia32_storehps((__v2si *)__p, (__v4sf)__a);
+ typedef float __mm_storeh_pi_v2f32 __attribute__((__vector_size__(8)));
+ struct __mm_storeh_pi_struct {
+ __mm_storeh_pi_v2f32 __u;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __mm_storeh_pi_struct*)__p)->__u = __builtin_shufflevector(__a, __a, 2, 3);
}
-/// \brief Stores the lower 64 bits of a 128-bit vector of [4 x float] to a
+/// Stores the lower 64 bits of a 128-bit vector of [4 x float] to a
/// memory location.
///
/// \headerfile <x86intrin.h>
static __inline__ void __DEFAULT_FN_ATTRS
_mm_storel_pi(__m64 *__p, __m128 __a)
{
- __builtin_ia32_storelps((__v2si *)__p, (__v4sf)__a);
+ typedef float __mm_storeh_pi_v2f32 __attribute__((__vector_size__(8)));
+ struct __mm_storeh_pi_struct {
+ __mm_storeh_pi_v2f32 __u;
+ } __attribute__((__packed__, __may_alias__));
+ ((struct __mm_storeh_pi_struct*)__p)->__u = __builtin_shufflevector(__a, __a, 0, 1);
}
-/// \brief Stores the lower 32 bits of a 128-bit vector of [4 x float] to a
+/// Stores the lower 32 bits of a 128-bit vector of [4 x float] to a
/// memory location.
///
/// \headerfile <x86intrin.h>
((struct __mm_store_ss_struct*)__p)->__u = __a[0];
}
-/// \brief Stores a 128-bit vector of [4 x float] to an unaligned memory
+/// Stores a 128-bit vector of [4 x float] to an unaligned memory
/// location.
///
/// \headerfile <x86intrin.h>
_mm_storeu_ps(float *__p, __m128 __a)
{
struct __storeu_ps {
- __m128 __v;
+ __m128_u __v;
} __attribute__((__packed__, __may_alias__));
((struct __storeu_ps*)__p)->__v = __a;
}
-/// \brief Stores a 128-bit vector of [4 x float] into an aligned memory
+/// Stores a 128-bit vector of [4 x float] into an aligned memory
/// location.
///
/// \headerfile <x86intrin.h>
*(__m128*)__p = __a;
}
-/// \brief Stores the lower 32 bits of a 128-bit vector of [4 x float] into
+/// Stores the lower 32 bits of a 128-bit vector of [4 x float] into
/// four contiguous elements in an aligned memory location.
///
/// \headerfile <x86intrin.h>
_mm_store_ps(__p, __a);
}
-/// \brief Stores the lower 32 bits of a 128-bit vector of [4 x float] into
+/// Stores the lower 32 bits of a 128-bit vector of [4 x float] into
/// four contiguous elements in an aligned memory location.
///
/// \headerfile <x86intrin.h>
static __inline__ void __DEFAULT_FN_ATTRS
_mm_store_ps1(float *__p, __m128 __a)
{
- return _mm_store1_ps(__p, __a);
+ _mm_store1_ps(__p, __a);
}
-/// \brief Stores float values from a 128-bit vector of [4 x float] to an
+/// Stores float values from a 128-bit vector of [4 x float] to an
/// aligned memory location in reverse order.
///
/// \headerfile <x86intrin.h>
_mm_store_ps(__p, __a);
}
-#define _MM_HINT_T0 3
-#define _MM_HINT_T1 2
-#define _MM_HINT_T2 1
+#define _MM_HINT_ET0 7
+#define _MM_HINT_ET1 6
+#define _MM_HINT_T0 3
+#define _MM_HINT_T1 2
+#define _MM_HINT_T2 1
#define _MM_HINT_NTA 0
#ifndef _MSC_VER
/* FIXME: We have to #define this because "sel" must be a constant integer, and
Sema doesn't do any form of constant propagation yet. */
-/// \brief Loads one cache line of data from the specified address to a location
+/// Loads one cache line of data from the specified address to a location
/// closer to the processor.
///
/// \headerfile <x86intrin.h>
/// be generated. \n
/// _MM_HINT_T2: Move data using the T2 hint. The PREFETCHT2 instruction will
/// be generated.
-#define _mm_prefetch(a, sel) (__builtin_prefetch((void *)(a), 0, (sel)))
+#define _mm_prefetch(a, sel) (__builtin_prefetch((const void *)(a), \
+ ((sel) >> 2) & 1, (sel) & 0x3))
#endif
-/// \brief Stores a 64-bit integer in the specified aligned memory location. To
+/// Stores a 64-bit integer in the specified aligned memory location. To
/// minimize caching, the data is flagged as non-temporal (unlikely to be
/// used again soon).
///
/// A pointer to an aligned memory location used to store the register value.
/// \param __a
/// A 64-bit integer containing the value to be stored.
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS_MMX
_mm_stream_pi(__m64 *__p, __m64 __a)
{
__builtin_ia32_movntq(__p, __a);
}
-/// \brief Moves packed float values from a 128-bit vector of [4 x float] to a
+/// Moves packed float values from a 128-bit vector of [4 x float] to a
/// 128-bit aligned memory location. To minimize caching, the data is flagged
/// as non-temporal (unlikely to be used again soon).
///
extern "C" {
#endif
-/// \brief Forces strong memory ordering (serialization) between store
+/// Forces strong memory ordering (serialization) between store
/// instructions preceding this instruction and store instructions following
/// this instruction, ensuring the system completes all previous stores
/// before executing subsequent stores.
} // extern "C"
#endif
-/// \brief Extracts 16-bit element from a 64-bit vector of [4 x i16] and
+/// Extracts 16-bit element from a 64-bit vector of [4 x i16] and
/// returns it, as specified by the immediate integer operand.
///
/// \headerfile <x86intrin.h>
/// 2: Bits [47:32] are copied to the destination. \n
/// 3: Bits [63:48] are copied to the destination.
/// \returns A 16-bit integer containing the extracted 16 bits of packed data.
-#define _mm_extract_pi16(a, n) __extension__ ({ \
- (int)__builtin_ia32_vec_ext_v4hi((__m64)a, (int)n); })
+#define _mm_extract_pi16(a, n) \
+ (int)__builtin_ia32_vec_ext_v4hi((__v4hi)a, (int)n)
-/// \brief Copies data from the 64-bit vector of [4 x i16] to the destination,
+/// Copies data from the 64-bit vector of [4 x i16] to the destination,
/// and inserts the lower 16-bits of an integer operand at the 16-bit offset
/// specified by the immediate operand \a n.
///
/// __m64 _mm_insert_pi16(__m64 a, int d, int n);
/// \endcode
///
-/// This intrinsic corresponds to the <c> VPINSRW / PINSRW </c> instruction.
+/// This intrinsic corresponds to the <c> PINSRW </c> instruction.
///
/// \param a
/// A 64-bit vector of [4 x i16].
/// bits in operand \a a.
/// \returns A 64-bit integer vector containing the copied packed data from the
/// operands.
-#define _mm_insert_pi16(a, d, n) __extension__ ({ \
- (__m64)__builtin_ia32_vec_set_v4hi((__m64)a, (int)d, (int)n); })
+#define _mm_insert_pi16(a, d, n) \
+ (__m64)__builtin_ia32_vec_set_v4hi((__v4hi)a, (int)d, (int)n)
-/// \brief Compares each of the corresponding packed 16-bit integer values of
+/// Compares each of the corresponding packed 16-bit integer values of
/// the 64-bit integer vectors, and writes the greater value to the
/// corresponding bits in the destination.
///
/// \param __b
/// A 64-bit integer vector containing one of the source operands.
/// \returns A 64-bit integer vector containing the comparison results.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_max_pi16(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_pmaxsw((__v4hi)__a, (__v4hi)__b);
}
-/// \brief Compares each of the corresponding packed 8-bit unsigned integer
+/// Compares each of the corresponding packed 8-bit unsigned integer
/// values of the 64-bit integer vectors, and writes the greater value to the
/// corresponding bits in the destination.
///
/// \param __b
/// A 64-bit integer vector containing one of the source operands.
/// \returns A 64-bit integer vector containing the comparison results.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_max_pu8(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_pmaxub((__v8qi)__a, (__v8qi)__b);
}
-/// \brief Compares each of the corresponding packed 16-bit integer values of
+/// Compares each of the corresponding packed 16-bit integer values of
/// the 64-bit integer vectors, and writes the lesser value to the
/// corresponding bits in the destination.
///
/// \param __b
/// A 64-bit integer vector containing one of the source operands.
/// \returns A 64-bit integer vector containing the comparison results.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_min_pi16(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_pminsw((__v4hi)__a, (__v4hi)__b);
}
-/// \brief Compares each of the corresponding packed 8-bit unsigned integer
+/// Compares each of the corresponding packed 8-bit unsigned integer
/// values of the 64-bit integer vectors, and writes the lesser value to the
/// corresponding bits in the destination.
///
/// \param __b
/// A 64-bit integer vector containing one of the source operands.
/// \returns A 64-bit integer vector containing the comparison results.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_min_pu8(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_pminub((__v8qi)__a, (__v8qi)__b);
}
-/// \brief Takes the most significant bit from each 8-bit element in a 64-bit
-/// integer vector to create a 16-bit mask value. Zero-extends the value to
+/// Takes the most significant bit from each 8-bit element in a 64-bit
+/// integer vector to create an 8-bit mask value. Zero-extends the value to
/// 32-bit integer and writes it to the destination.
///
/// \headerfile <x86intrin.h>
///
/// \param __a
/// A 64-bit integer vector containing the values with bits to be extracted.
-/// \returns The most significant bit from each 8-bit element in the operand,
-/// written to bits [15:0].
-static __inline__ int __DEFAULT_FN_ATTRS
+/// \returns The most significant bit from each 8-bit element in \a __a,
+/// written to bits [7:0].
+static __inline__ int __DEFAULT_FN_ATTRS_MMX
_mm_movemask_pi8(__m64 __a)
{
return __builtin_ia32_pmovmskb((__v8qi)__a);
}
-/// \brief Multiplies packed 16-bit unsigned integer values and writes the
+/// Multiplies packed 16-bit unsigned integer values and writes the
/// high-order 16 bits of each 32-bit product to the corresponding bits in
/// the destination.
///
/// \param __b
/// A 64-bit integer vector containing one of the source operands.
/// \returns A 64-bit integer vector containing the products of both operands.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_mulhi_pu16(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_pmulhuw((__v4hi)__a, (__v4hi)__b);
}
-/// \brief Shuffles the 4 16-bit integers from a 64-bit integer vector to the
+/// Shuffles the 4 16-bit integers from a 64-bit integer vector to the
/// destination, as specified by the immediate value operand.
///
/// \headerfile <x86intrin.h>
/// 10: assigned from bits [47:32] of \a a. \n
/// 11: assigned from bits [63:48] of \a a.
/// \returns A 64-bit integer vector containing the shuffled values.
-#define _mm_shuffle_pi16(a, n) __extension__ ({ \
- (__m64)__builtin_ia32_pshufw((__v4hi)(__m64)(a), (n)); })
+#define _mm_shuffle_pi16(a, n) \
+ (__m64)__builtin_ia32_pshufw((__v4hi)(__m64)(a), (n))
-/// \brief Conditionally copies the values from each 8-bit element in the first
+/// Conditionally copies the values from each 8-bit element in the first
/// 64-bit integer vector operand to the specified memory location, as
/// specified by the most significant bit in the corresponding element in the
/// second 64-bit integer vector operand.
/// A pointer to a 64-bit memory location that will receive the conditionally
/// copied integer values. The address of the memory location does not have
/// to be aligned.
-static __inline__ void __DEFAULT_FN_ATTRS
+static __inline__ void __DEFAULT_FN_ATTRS_MMX
_mm_maskmove_si64(__m64 __d, __m64 __n, char *__p)
{
__builtin_ia32_maskmovq((__v8qi)__d, (__v8qi)__n, __p);
}
-/// \brief Computes the rounded averages of the packed unsigned 8-bit integer
+/// Computes the rounded averages of the packed unsigned 8-bit integer
/// values and writes the averages to the corresponding bits in the
/// destination.
///
/// \param __b
/// A 64-bit integer vector containing one of the source operands.
/// \returns A 64-bit integer vector containing the averages of both operands.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_avg_pu8(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_pavgb((__v8qi)__a, (__v8qi)__b);
}
-/// \brief Computes the rounded averages of the packed unsigned 16-bit integer
+/// Computes the rounded averages of the packed unsigned 16-bit integer
/// values and writes the averages to the corresponding bits in the
/// destination.
///
/// \param __b
/// A 64-bit integer vector containing one of the source operands.
/// \returns A 64-bit integer vector containing the averages of both operands.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_avg_pu16(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_pavgw((__v4hi)__a, (__v4hi)__b);
}
-/// \brief Subtracts the corresponding 8-bit unsigned integer values of the two
+/// Subtracts the corresponding 8-bit unsigned integer values of the two
/// 64-bit vector operands and computes the absolute value for each of the
/// difference. Then sum of the 8 absolute differences is written to the
/// bits [15:0] of the destination; the remaining bits [63:16] are cleared.
/// \returns A 64-bit integer vector whose lower 16 bits contain the sums of the
/// sets of absolute differences between both operands. The upper bits are
/// cleared.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_sad_pu8(__m64 __a, __m64 __b)
{
return (__m64)__builtin_ia32_psadbw((__v8qi)__a, (__v8qi)__b);
extern "C" {
#endif
-/// \brief Returns the contents of the MXCSR register as a 32-bit unsigned
+/// Returns the contents of the MXCSR register as a 32-bit unsigned
/// integer value.
///
/// There are several groups of macros associated with this
/// <li>
/// For checking rounding modes: _MM_ROUND_NEAREST, _MM_ROUND_DOWN,
/// _MM_ROUND_UP, _MM_ROUND_TOWARD_ZERO. There is a convenience wrapper
-/// _MM_GET_ROUNDING_MODE(x) where x is one of these macros.
+/// _MM_GET_ROUNDING_MODE().
/// </li>
/// <li>
/// For checking flush-to-zero mode: _MM_FLUSH_ZERO_ON, _MM_FLUSH_ZERO_OFF.
/// </li>
/// </ul>
///
-/// For example, the expression below checks if an overflow exception has
+/// For example, the following expression checks if an overflow exception has
/// occurred:
+/// \code
/// ( _mm_getcsr() & _MM_EXCEPT_OVERFLOW )
+/// \endcode
///
-/// The following example gets the current rounding mode:
+/// The following expression gets the current rounding mode:
+/// \code
/// _MM_GET_ROUNDING_MODE()
+/// \endcode
///
/// \headerfile <x86intrin.h>
///
/// register.
unsigned int _mm_getcsr(void);
-/// \brief Sets the MXCSR register with the 32-bit unsigned integer value.
+/// Sets the MXCSR register with the 32-bit unsigned integer value.
///
/// There are several groups of macros associated with this intrinsic,
/// including:
/// _mm_setcsr(_mm_getcsr() | _MM_ROUND_UP)
///
/// The following example sets the DAZ and FTZ flags:
-/// void setFlags() {
-/// _MM_SET_FLUSH_ZERO_MODE(_MM_FLUSH_ZERO_ON)
-/// _MM_SET_DENORMALS_ZERO_MODE(_MM_DENORMALS_ZERO_ON)
-/// }
+/// \code
+/// void setFlags() {
+/// _MM_SET_FLUSH_ZERO_MODE(_MM_FLUSH_ZERO_ON);
+/// _MM_SET_DENORMALS_ZERO_MODE(_MM_DENORMALS_ZERO_ON);
+/// }
+/// \endcode
///
/// \headerfile <x86intrin.h>
///
} // extern "C"
#endif
-/// \brief Selects 4 float values from the 128-bit operands of [4 x float], as
+/// Selects 4 float values from the 128-bit operands of [4 x float], as
/// specified by the immediate value operand.
///
/// \headerfile <x86intrin.h>
/// 10: Bits [95:64] copied from the specified operand. \n
/// 11: Bits [127:96] copied from the specified operand.
/// \returns A 128-bit vector of [4 x float] containing the shuffled values.
-#define _mm_shuffle_ps(a, b, mask) __extension__ ({ \
- (__m128)__builtin_shufflevector((__v4sf)(__m128)(a), (__v4sf)(__m128)(b), \
- 0 + (((mask) >> 0) & 0x3), \
- 0 + (((mask) >> 2) & 0x3), \
- 4 + (((mask) >> 4) & 0x3), \
- 4 + (((mask) >> 6) & 0x3)); })
-
-/// \brief Unpacks the high-order (index 2,3) values from two 128-bit vectors of
+#define _mm_shuffle_ps(a, b, mask) \
+ (__m128)__builtin_ia32_shufps((__v4sf)(__m128)(a), (__v4sf)(__m128)(b), \
+ (int)(mask))
+
+/// Unpacks the high-order (index 2,3) values from two 128-bit vectors of
/// [4 x float] and interleaves them into a 128-bit vector of [4 x float].
///
/// \headerfile <x86intrin.h>
return __builtin_shufflevector((__v4sf)__a, (__v4sf)__b, 2, 6, 3, 7);
}
-/// \brief Unpacks the low-order (index 0,1) values from two 128-bit vectors of
+/// Unpacks the low-order (index 0,1) values from two 128-bit vectors of
/// [4 x float] and interleaves them into a 128-bit vector of [4 x float].
///
/// \headerfile <x86intrin.h>
return __builtin_shufflevector((__v4sf)__a, (__v4sf)__b, 0, 4, 1, 5);
}
-/// \brief Constructs a 128-bit floating-point vector of [4 x float]. The lower
+/// Constructs a 128-bit floating-point vector of [4 x float]. The lower
/// 32 bits are set to the lower 32 bits of the second parameter. The upper
/// 96 bits are set to the upper 96 bits of the first parameter.
///
/// \headerfile <x86intrin.h>
///
-/// This intrinsic corresponds to the <c> VMOVSS / MOVSS </c> instruction.
+/// This intrinsic corresponds to the <c> VBLENDPS / BLENDPS / MOVSS </c>
+/// instruction.
///
/// \param __a
/// A 128-bit floating-point vector of [4 x float]. The upper 96 bits are
static __inline__ __m128 __DEFAULT_FN_ATTRS
_mm_move_ss(__m128 __a, __m128 __b)
{
- return __builtin_shufflevector((__v4sf)__a, (__v4sf)__b, 4, 1, 2, 3);
+ __a[0] = __b[0];
+ return __a;
}
-/// \brief Constructs a 128-bit floating-point vector of [4 x float]. The lower
+/// Constructs a 128-bit floating-point vector of [4 x float]. The lower
/// 64 bits are set to the upper 64 bits of the second parameter. The upper
/// 64 bits are set to the upper 64 bits of the first parameter.
///
return __builtin_shufflevector((__v4sf)__a, (__v4sf)__b, 6, 7, 2, 3);
}
-/// \brief Constructs a 128-bit floating-point vector of [4 x float]. The lower
+/// Constructs a 128-bit floating-point vector of [4 x float]. The lower
/// 64 bits are set to the lower 64 bits of the first parameter. The upper
/// 64 bits are set to the lower 64 bits of the second parameter.
///
return __builtin_shufflevector((__v4sf)__a, (__v4sf)__b, 0, 1, 4, 5);
}
-/// \brief Converts a 64-bit vector of [4 x i16] into a 128-bit vector of [4 x
+/// Converts a 64-bit vector of [4 x i16] into a 128-bit vector of [4 x
/// float].
///
/// \headerfile <x86intrin.h>
/// from the corresponding elements in this operand.
/// \returns A 128-bit vector of [4 x float] containing the copied and converted
/// values from the operand.
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS_MMX
_mm_cvtpi16_ps(__m64 __a)
{
__m64 __b, __c;
return __r;
}
-/// \brief Converts a 64-bit vector of 16-bit unsigned integer values into a
+/// Converts a 64-bit vector of 16-bit unsigned integer values into a
/// 128-bit vector of [4 x float].
///
/// \headerfile <x86intrin.h>
/// destination are copied from the corresponding elements in this operand.
/// \returns A 128-bit vector of [4 x float] containing the copied and converted
/// values from the operand.
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS_MMX
_mm_cvtpu16_ps(__m64 __a)
{
__m64 __b, __c;
return __r;
}
-/// \brief Converts the lower four 8-bit values from a 64-bit vector of [8 x i8]
+/// Converts the lower four 8-bit values from a 64-bit vector of [8 x i8]
/// into a 128-bit vector of [4 x float].
///
/// \headerfile <x86intrin.h>
/// from the corresponding lower 4 elements in this operand.
/// \returns A 128-bit vector of [4 x float] containing the copied and converted
/// values from the operand.
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS_MMX
_mm_cvtpi8_ps(__m64 __a)
{
__m64 __b;
return _mm_cvtpi16_ps(__b);
}
-/// \brief Converts the lower four unsigned 8-bit integer values from a 64-bit
+/// Converts the lower four unsigned 8-bit integer values from a 64-bit
/// vector of [8 x u8] into a 128-bit vector of [4 x float].
///
/// \headerfile <x86intrin.h>
/// operand.
/// \returns A 128-bit vector of [4 x float] containing the copied and converted
/// values from the source operand.
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS_MMX
_mm_cvtpu8_ps(__m64 __a)
{
__m64 __b;
return _mm_cvtpi16_ps(__b);
}
-/// \brief Converts the two 32-bit signed integer values from each 64-bit vector
+/// Converts the two 32-bit signed integer values from each 64-bit vector
/// operand of [2 x i32] into a 128-bit vector of [4 x float].
///
/// \headerfile <x86intrin.h>
/// \returns A 128-bit vector of [4 x float] whose lower 64 bits contain the
/// copied and converted values from the first operand. The upper 64 bits
/// contain the copied and converted values from the second operand.
-static __inline__ __m128 __DEFAULT_FN_ATTRS
+static __inline__ __m128 __DEFAULT_FN_ATTRS_MMX
_mm_cvtpi32x2_ps(__m64 __a, __m64 __b)
{
__m128 __c;
return _mm_cvtpi32_ps(__c, __a);
}
-/// \brief Converts each single-precision floating-point element of a 128-bit
+/// Converts each single-precision floating-point element of a 128-bit
/// floating-point vector of [4 x float] into a 16-bit signed integer, and
/// packs the results into a 64-bit integer vector of [4 x i16].
///
/// A 128-bit floating-point vector of [4 x float].
/// \returns A 64-bit integer vector of [4 x i16] containing the converted
/// values.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_cvtps_pi16(__m128 __a)
{
__m64 __b, __c;
return _mm_packs_pi32(__b, __c);
}
-/// \brief Converts each single-precision floating-point element of a 128-bit
+/// Converts each single-precision floating-point element of a 128-bit
/// floating-point vector of [4 x float] into an 8-bit signed integer, and
/// packs the results into the lower 32 bits of a 64-bit integer vector of
/// [8 x i8]. The upper 32 bits of the vector are set to 0.
/// 128-bit floating-point vector of [4 x float].
/// \returns A 64-bit integer vector of [8 x i8]. The lower 32 bits contain the
/// converted values and the uppper 32 bits are set to zero.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
_mm_cvtps_pi8(__m128 __a)
{
__m64 __b, __c;
return _mm_packs_pi16(__b, __c);
}
-/// \brief Extracts the sign bits from each single-precision floating-point
+/// Extracts the sign bits from each single-precision floating-point
/// element of a 128-bit floating-point vector of [4 x float] and returns the
/// sign bits in bits [0:3] of the result. Bits [31:4] of the result are set
/// to zero.
#define _MM_SHUFFLE(z, y, x, w) (((z) << 6) | ((y) << 4) | ((x) << 2) | (w))
-#define _MM_EXCEPT_INVALID (0x0001)
-#define _MM_EXCEPT_DENORM (0x0002)
-#define _MM_EXCEPT_DIV_ZERO (0x0004)
-#define _MM_EXCEPT_OVERFLOW (0x0008)
-#define _MM_EXCEPT_UNDERFLOW (0x0010)
-#define _MM_EXCEPT_INEXACT (0x0020)
-#define _MM_EXCEPT_MASK (0x003f)
-
-#define _MM_MASK_INVALID (0x0080)
-#define _MM_MASK_DENORM (0x0100)
-#define _MM_MASK_DIV_ZERO (0x0200)
-#define _MM_MASK_OVERFLOW (0x0400)
-#define _MM_MASK_UNDERFLOW (0x0800)
-#define _MM_MASK_INEXACT (0x1000)
-#define _MM_MASK_MASK (0x1f80)
-
-#define _MM_ROUND_NEAREST (0x0000)
-#define _MM_ROUND_DOWN (0x2000)
-#define _MM_ROUND_UP (0x4000)
-#define _MM_ROUND_TOWARD_ZERO (0x6000)
-#define _MM_ROUND_MASK (0x6000)
-
-#define _MM_FLUSH_ZERO_MASK (0x8000)
-#define _MM_FLUSH_ZERO_ON (0x8000)
-#define _MM_FLUSH_ZERO_OFF (0x0000)
+#define _MM_EXCEPT_INVALID (0x0001U)
+#define _MM_EXCEPT_DENORM (0x0002U)
+#define _MM_EXCEPT_DIV_ZERO (0x0004U)
+#define _MM_EXCEPT_OVERFLOW (0x0008U)
+#define _MM_EXCEPT_UNDERFLOW (0x0010U)
+#define _MM_EXCEPT_INEXACT (0x0020U)
+#define _MM_EXCEPT_MASK (0x003fU)
+
+#define _MM_MASK_INVALID (0x0080U)
+#define _MM_MASK_DENORM (0x0100U)
+#define _MM_MASK_DIV_ZERO (0x0200U)
+#define _MM_MASK_OVERFLOW (0x0400U)
+#define _MM_MASK_UNDERFLOW (0x0800U)
+#define _MM_MASK_INEXACT (0x1000U)
+#define _MM_MASK_MASK (0x1f80U)
+
+#define _MM_ROUND_NEAREST (0x0000U)
+#define _MM_ROUND_DOWN (0x2000U)
+#define _MM_ROUND_UP (0x4000U)
+#define _MM_ROUND_TOWARD_ZERO (0x6000U)
+#define _MM_ROUND_MASK (0x6000U)
+
+#define _MM_FLUSH_ZERO_MASK (0x8000U)
+#define _MM_FLUSH_ZERO_ON (0x8000U)
+#define _MM_FLUSH_ZERO_OFF (0x0000U)
#define _MM_GET_EXCEPTION_MASK() (_mm_getcsr() & _MM_MASK_MASK)
#define _MM_GET_EXCEPTION_STATE() (_mm_getcsr() & _MM_EXCEPT_MASK)
#define _m_ _mm_
#undef __DEFAULT_FN_ATTRS
+#undef __DEFAULT_FN_ATTRS_MMX
/* Ugly hack for backwards-compatibility (compatible with gcc) */
#if defined(__SSE2__) && !__building_module(_Builtin_intrinsics)
/*===---- xopintrin.h - XOP intrinsics -------------------------------------===
*
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#include <fma4intrin.h>
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("xop")))
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("xop"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("xop"), __min_vector_width__(256)))
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_maccs_epi16(__m128i __A, __m128i __B, __m128i __C)
return (__m128i)(((__v2du)__A & (__v2du)__C) | ((__v2du)__B & ~(__v2du)__C));
}
-static __inline__ __m256i __DEFAULT_FN_ATTRS
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
_mm256_cmov_si256(__m256i __A, __m256i __B, __m256i __C)
{
return (__m256i)(((__v4du)__A & (__v4du)__C) | ((__v4du)__B & ~(__v4du)__C));
return (__m128i)__builtin_ia32_vprotq((__v2di)__A, (__v2di)__B);
}
-#define _mm_roti_epi8(A, N) __extension__ ({ \
- (__m128i)__builtin_ia32_vprotbi((__v16qi)(__m128i)(A), (N)); })
+#define _mm_roti_epi8(A, N) \
+ (__m128i)__builtin_ia32_vprotbi((__v16qi)(__m128i)(A), (N))
-#define _mm_roti_epi16(A, N) __extension__ ({ \
- (__m128i)__builtin_ia32_vprotwi((__v8hi)(__m128i)(A), (N)); })
+#define _mm_roti_epi16(A, N) \
+ (__m128i)__builtin_ia32_vprotwi((__v8hi)(__m128i)(A), (N))
-#define _mm_roti_epi32(A, N) __extension__ ({ \
- (__m128i)__builtin_ia32_vprotdi((__v4si)(__m128i)(A), (N)); })
+#define _mm_roti_epi32(A, N) \
+ (__m128i)__builtin_ia32_vprotdi((__v4si)(__m128i)(A), (N))
-#define _mm_roti_epi64(A, N) __extension__ ({ \
- (__m128i)__builtin_ia32_vprotqi((__v2di)(__m128i)(A), (N)); })
+#define _mm_roti_epi64(A, N) \
+ (__m128i)__builtin_ia32_vprotqi((__v2di)(__m128i)(A), (N))
static __inline__ __m128i __DEFAULT_FN_ATTRS
_mm_shl_epi8(__m128i __A, __m128i __B)
return (__m128i)__builtin_ia32_vpshaq((__v2di)__A, (__v2di)__B);
}
-#define _mm_com_epu8(A, B, N) __extension__ ({ \
+#define _mm_com_epu8(A, B, N) \
(__m128i)__builtin_ia32_vpcomub((__v16qi)(__m128i)(A), \
- (__v16qi)(__m128i)(B), (N)); })
+ (__v16qi)(__m128i)(B), (N))
-#define _mm_com_epu16(A, B, N) __extension__ ({ \
+#define _mm_com_epu16(A, B, N) \
(__m128i)__builtin_ia32_vpcomuw((__v8hi)(__m128i)(A), \
- (__v8hi)(__m128i)(B), (N)); })
+ (__v8hi)(__m128i)(B), (N))
-#define _mm_com_epu32(A, B, N) __extension__ ({ \
+#define _mm_com_epu32(A, B, N) \
(__m128i)__builtin_ia32_vpcomud((__v4si)(__m128i)(A), \
- (__v4si)(__m128i)(B), (N)); })
+ (__v4si)(__m128i)(B), (N))
-#define _mm_com_epu64(A, B, N) __extension__ ({ \
+#define _mm_com_epu64(A, B, N) \
(__m128i)__builtin_ia32_vpcomuq((__v2di)(__m128i)(A), \
- (__v2di)(__m128i)(B), (N)); })
+ (__v2di)(__m128i)(B), (N))
-#define _mm_com_epi8(A, B, N) __extension__ ({ \
+#define _mm_com_epi8(A, B, N) \
(__m128i)__builtin_ia32_vpcomb((__v16qi)(__m128i)(A), \
- (__v16qi)(__m128i)(B), (N)); })
+ (__v16qi)(__m128i)(B), (N))
-#define _mm_com_epi16(A, B, N) __extension__ ({ \
+#define _mm_com_epi16(A, B, N) \
(__m128i)__builtin_ia32_vpcomw((__v8hi)(__m128i)(A), \
- (__v8hi)(__m128i)(B), (N)); })
+ (__v8hi)(__m128i)(B), (N))
-#define _mm_com_epi32(A, B, N) __extension__ ({ \
+#define _mm_com_epi32(A, B, N) \
(__m128i)__builtin_ia32_vpcomd((__v4si)(__m128i)(A), \
- (__v4si)(__m128i)(B), (N)); })
+ (__v4si)(__m128i)(B), (N))
-#define _mm_com_epi64(A, B, N) __extension__ ({ \
+#define _mm_com_epi64(A, B, N) \
(__m128i)__builtin_ia32_vpcomq((__v2di)(__m128i)(A), \
- (__v2di)(__m128i)(B), (N)); })
+ (__v2di)(__m128i)(B), (N))
#define _MM_PCOMCTRL_LT 0
#define _MM_PCOMCTRL_LE 1
return _mm_com_epi64(__A, __B, _MM_PCOMCTRL_TRUE);
}
-#define _mm_permute2_pd(X, Y, C, I) __extension__ ({ \
+#define _mm_permute2_pd(X, Y, C, I) \
(__m128d)__builtin_ia32_vpermil2pd((__v2df)(__m128d)(X), \
(__v2df)(__m128d)(Y), \
- (__v2di)(__m128i)(C), (I)); })
+ (__v2di)(__m128i)(C), (I))
-#define _mm256_permute2_pd(X, Y, C, I) __extension__ ({ \
+#define _mm256_permute2_pd(X, Y, C, I) \
(__m256d)__builtin_ia32_vpermil2pd256((__v4df)(__m256d)(X), \
(__v4df)(__m256d)(Y), \
- (__v4di)(__m256i)(C), (I)); })
+ (__v4di)(__m256i)(C), (I))
-#define _mm_permute2_ps(X, Y, C, I) __extension__ ({ \
+#define _mm_permute2_ps(X, Y, C, I) \
(__m128)__builtin_ia32_vpermil2ps((__v4sf)(__m128)(X), (__v4sf)(__m128)(Y), \
- (__v4si)(__m128i)(C), (I)); })
+ (__v4si)(__m128i)(C), (I))
-#define _mm256_permute2_ps(X, Y, C, I) __extension__ ({ \
+#define _mm256_permute2_ps(X, Y, C, I) \
(__m256)__builtin_ia32_vpermil2ps256((__v8sf)(__m256)(X), \
(__v8sf)(__m256)(Y), \
- (__v8si)(__m256i)(C), (I)); })
+ (__v8si)(__m256i)(C), (I))
static __inline__ __m128 __DEFAULT_FN_ATTRS
_mm_frcz_ss(__m128 __A)
return (__m128d)__builtin_ia32_vfrczpd((__v2df)__A);
}
-static __inline__ __m256 __DEFAULT_FN_ATTRS
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
_mm256_frcz_ps(__m256 __A)
{
return (__m256)__builtin_ia32_vfrczps256((__v8sf)__A);
}
-static __inline__ __m256d __DEFAULT_FN_ATTRS
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
_mm256_frcz_pd(__m256d __A)
{
return (__m256d)__builtin_ia32_vfrczpd256((__v4df)__A);
}
#undef __DEFAULT_FN_ATTRS
+#undef __DEFAULT_FN_ATTRS256
#endif /* __XOPINTRIN_H */
--- /dev/null
+//===- xray_interface.h -----------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file is a part of XRay, a dynamic runtime instrumentation system.
+//
+// APIs for controlling XRay functionality explicitly.
+//===----------------------------------------------------------------------===//
+
+#ifndef XRAY_XRAY_INTERFACE_H
+#define XRAY_XRAY_INTERFACE_H
+
+#include <cstddef>
+#include <cstdint>
+
+extern "C" {
+
+/// Synchronize this with AsmPrinter::SledKind in LLVM.
+enum XRayEntryType {
+ ENTRY = 0,
+ EXIT = 1,
+ TAIL = 2,
+ LOG_ARGS_ENTRY = 3,
+ CUSTOM_EVENT = 4,
+ TYPED_EVENT = 5,
+};
+
+/// Provide a function to invoke for when instrumentation points are hit. This
+/// is a user-visible control surface that overrides the default implementation.
+/// The function provided should take the following arguments:
+///
+/// - function id: an identifier that indicates the id of a function; this id
+/// is generated by xray; the mapping between the function id
+/// and the actual function pointer is available through
+/// __xray_table.
+/// - entry type: identifies what kind of instrumentation point was
+/// encountered (function entry, function exit, etc.). See the
+/// enum XRayEntryType for more details.
+///
+/// The user handler must handle correctly spurious calls after this handler is
+/// removed or replaced with another handler, because it would be too costly for
+/// XRay runtime to avoid spurious calls.
+/// To prevent circular calling, the handler function itself and all its
+/// direct&indirect callees must not be instrumented with XRay, which can be
+/// achieved by marking them all with: __attribute__((xray_never_instrument))
+///
+/// Returns 1 on success, 0 on error.
+extern int __xray_set_handler(void (*entry)(int32_t, XRayEntryType));
+
+/// This removes whatever the currently provided handler is. Returns 1 on
+/// success, 0 on error.
+extern int __xray_remove_handler();
+
+/// Use XRay to log the first argument of each (instrumented) function call.
+/// When this function exits, all threads will have observed the effect and
+/// start logging their subsequent affected function calls (if patched).
+///
+/// Returns 1 on success, 0 on error.
+extern int __xray_set_handler_arg1(void (*entry)(int32_t, XRayEntryType,
+ uint64_t));
+
+/// Disables the XRay handler used to log first arguments of function calls.
+/// Returns 1 on success, 0 on error.
+extern int __xray_remove_handler_arg1();
+
+/// Provide a function to invoke when XRay encounters a custom event.
+extern int __xray_set_customevent_handler(void (*entry)(void *, std::size_t));
+
+/// This removes whatever the currently provided custom event handler is.
+/// Returns 1 on success, 0 on error.
+extern int __xray_remove_customevent_handler();
+
+/// Set a handler for xray typed event logging. The first parameter is a type
+/// identifier, the second is a payload, and the third is the payload size.
+extern int __xray_set_typedevent_handler(void (*entry)(uint16_t, const void *,
+ std::size_t));
+
+/// Removes the currently set typed event handler.
+/// Returns 1 on success, 0 on error.
+extern int __xray_remove_typedevent_handler();
+
+extern uint16_t __xray_register_event_type(const char *event_type);
+
+enum XRayPatchingStatus {
+ NOT_INITIALIZED = 0,
+ SUCCESS = 1,
+ ONGOING = 2,
+ FAILED = 3,
+};
+
+/// This tells XRay to patch the instrumentation points. See XRayPatchingStatus
+/// for possible result values.
+extern XRayPatchingStatus __xray_patch();
+
+/// Reverses the effect of __xray_patch(). See XRayPatchingStatus for possible
+/// result values.
+extern XRayPatchingStatus __xray_unpatch();
+
+/// This patches a specific function id. See XRayPatchingStatus for possible
+/// result values.
+extern XRayPatchingStatus __xray_patch_function(int32_t FuncId);
+
+/// This unpatches a specific function id. See XRayPatchingStatus for possible
+/// result values.
+extern XRayPatchingStatus __xray_unpatch_function(int32_t FuncId);
+
+/// This function returns the address of the function provided a valid function
+/// id. We return 0 if we encounter any error, even if 0 may be a valid function
+/// address.
+extern uintptr_t __xray_function_address(int32_t FuncId);
+
+/// This function returns the maximum valid function id. Returns 0 if we
+/// encounter errors (when there are no instrumented functions, etc.).
+extern size_t __xray_max_function_id();
+
+/// Initialize the required XRay data structures. This is useful in cases where
+/// users want to control precisely when the XRay instrumentation data
+/// structures are initialized, for example when the XRay library is built with
+/// the XRAY_NO_PREINIT preprocessor definition.
+///
+/// Calling __xray_init() more than once is safe across multiple threads.
+extern void __xray_init();
+
+} // end extern "C"
+
+#endif // XRAY_XRAY_INTERFACE_H
--- /dev/null
+//===-- xray_log_interface.h ----------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file is a part of XRay, a function call tracing system.
+//
+// APIs for installing a new logging implementation.
+//
+//===----------------------------------------------------------------------===//
+///
+/// XRay allows users to implement their own logging handlers and install them
+/// to replace the default runtime-controllable implementation that comes with
+/// compiler-rt/xray. The "flight data recorder" (FDR) mode implementation uses
+/// this API to install itself in an XRay-enabled binary. See
+/// compiler-rt/lib/xray_fdr_logging.{h,cc} for details of that implementation.
+///
+/// The high-level usage pattern for these APIs look like the following:
+///
+/// // We choose the mode which we'd like to install, and check whether this
+/// // has succeeded. Each mode will have their own set of flags they will
+/// // support, outside of the global XRay configuration options that are
+/// // defined in the XRAY_OPTIONS environment variable.
+/// auto select_status = __xray_log_select_mode("xray-fdr");
+/// if (select_status != XRayLogRegisterStatus::XRAY_REGISTRATION_OK) {
+/// // This failed, we should not proceed with attempting to initialise
+/// // the currently selected mode.
+/// return;
+/// }
+///
+/// // Once that's done, we can now attempt to configure the implementation.
+/// // To do this, we provide the string flags configuration for the mode.
+/// auto config_status = __xray_log_init_mode(
+/// "xray-fdr", "verbosity=1 some_flag=1 another_flag=2");
+/// if (config_status != XRayLogInitStatus::XRAY_LOG_INITIALIZED) {
+/// // deal with the error here, if there is one.
+/// }
+///
+/// // When the log implementation has had the chance to initialize, we can
+/// // now patch the instrumentation points. Note that we could have patched
+/// // the instrumentation points first, but there's no strict ordering to
+/// // these operations.
+/// auto patch_status = __xray_patch();
+/// if (patch_status != XRayPatchingStatus::SUCCESS) {
+/// // deal with the error here, if it is an error.
+/// }
+///
+/// // If we want to stop the implementation, we can then finalize it (before
+/// // optionally flushing the log).
+/// auto fin_status = __xray_log_finalize();
+/// if (fin_status != XRayLogInitStatus::XRAY_LOG_FINALIZED) {
+/// // deal with the error here, if it is an error.
+/// }
+///
+/// // We can optionally wait before flushing the log to give other threads a
+/// // chance to see that the implementation is already finalized. Also, at
+/// // this point we can optionally unpatch the instrumentation points to
+/// // reduce overheads at runtime.
+/// auto unpatch_status = __xray_unpatch();
+/// if (unpatch_status != XRayPatchingStatus::SUCCESS) {
+/// // deal with the error here, if it is an error.
+/// }
+///
+/// // If there are logs or data to be flushed somewhere, we can do so only
+/// // after we've finalized the log. Some implementations may not actually
+/// // have anything to log (it might keep the data in memory, or periodically
+/// // be logging the data anyway).
+/// auto flush_status = __xray_log_flushLog();
+/// if (flush_status != XRayLogFlushStatus::XRAY_LOG_FLUSHED) {
+/// // deal with the error here, if it is an error.
+/// }
+///
+/// // Alternatively, we can go through the buffers ourselves without
+/// // relying on the implementations' flushing semantics (if the
+/// // implementation supports exporting this data directly).
+/// auto MyBufferProcessor = +[](const char* mode, XRayBuffer buffer) {
+/// // Check the "mode" to see if it's something we know how to handle...
+/// // and/or do something with an XRayBuffer instance.
+/// };
+/// auto process_status = __xray_log_process_buffers(MyBufferProcessor);
+/// if (process_status != XRayLogFlushStatus::XRAY_LOG_FLUSHED) {
+/// // deal with the error here, if it is an error.
+/// }
+///
+/// NOTE: Before calling __xray_patch() again, consider re-initializing the
+/// implementation first. Some implementations might stay in an "off" state when
+/// they are finalized, while some might be in an invalid/unknown state.
+///
+#ifndef XRAY_XRAY_LOG_INTERFACE_H
+#define XRAY_XRAY_LOG_INTERFACE_H
+
+#include "xray/xray_interface.h"
+#include <stddef.h>
+
+extern "C" {
+
+/// This enum defines the valid states in which the logging implementation can
+/// be at.
+enum XRayLogInitStatus {
+ /// The default state is uninitialized, and in case there were errors in the
+ /// initialization, the implementation MUST return XRAY_LOG_UNINITIALIZED.
+ XRAY_LOG_UNINITIALIZED = 0,
+
+ /// Some implementations support multi-stage init (or asynchronous init), and
+ /// may return XRAY_LOG_INITIALIZING to signal callers of the API that
+ /// there's an ongoing initialization routine running. This allows
+ /// implementations to support concurrent threads attempting to initialize,
+ /// while only signalling success in one.
+ XRAY_LOG_INITIALIZING = 1,
+
+ /// When an implementation is done initializing, it MUST return
+ /// XRAY_LOG_INITIALIZED. When users call `__xray_patch()`, they are
+ /// guaranteed that the implementation installed with
+ /// `__xray_set_log_impl(...)` has been initialized.
+ XRAY_LOG_INITIALIZED = 2,
+
+ /// Some implementations might support multi-stage finalization (or
+ /// asynchronous finalization), and may return XRAY_LOG_FINALIZING to signal
+ /// callers of the API that there's an ongoing finalization routine running.
+ /// This allows implementations to support concurrent threads attempting to
+ /// finalize, while only signalling success/completion in one.
+ XRAY_LOG_FINALIZING = 3,
+
+ /// When an implementation is done finalizing, it MUST return
+ /// XRAY_LOG_FINALIZED. It is up to the implementation to determine what the
+ /// semantics of a finalized implementation is. Some implementations might
+ /// allow re-initialization once the log is finalized, while some might always
+ /// be on (and that finalization is a no-op).
+ XRAY_LOG_FINALIZED = 4,
+};
+
+/// This enum allows an implementation to signal log flushing operations via
+/// `__xray_log_flushLog()`, and the state of flushing the log.
+enum XRayLogFlushStatus {
+ XRAY_LOG_NOT_FLUSHING = 0,
+ XRAY_LOG_FLUSHING = 1,
+ XRAY_LOG_FLUSHED = 2,
+};
+
+/// This enum indicates the installation state of a logging implementation, when
+/// associating a mode to a particular logging implementation through
+/// `__xray_log_register_impl(...)` or through `__xray_log_select_mode(...`.
+enum XRayLogRegisterStatus {
+ XRAY_REGISTRATION_OK = 0,
+ XRAY_DUPLICATE_MODE = 1,
+ XRAY_MODE_NOT_FOUND = 2,
+ XRAY_INCOMPLETE_IMPL = 3,
+};
+
+/// A valid XRay logging implementation MUST provide all of the function
+/// pointers in XRayLogImpl when being installed through `__xray_set_log_impl`.
+/// To be precise, ALL the functions pointers MUST NOT be nullptr.
+struct XRayLogImpl {
+ /// The log initialization routine provided by the implementation, always
+ /// provided with the following parameters:
+ ///
+ /// - buffer size (unused)
+ /// - maximum number of buffers (unused)
+ /// - a pointer to an argument struct that the implementation MUST handle
+ /// - the size of the argument struct
+ ///
+ /// See XRayLogInitStatus for details on what the implementation MUST return
+ /// when called.
+ ///
+ /// If the implementation needs to install handlers aside from the 0-argument
+ /// function call handler, it MUST do so in this initialization handler.
+ ///
+ /// See xray_interface.h for available handler installation routines.
+ XRayLogInitStatus (*log_init)(size_t, size_t, void *, size_t);
+
+ /// The log finalization routine provided by the implementation.
+ ///
+ /// See XRayLogInitStatus for details on what the implementation MUST return
+ /// when called.
+ XRayLogInitStatus (*log_finalize)();
+
+ /// The 0-argument function call handler. XRay logging implementations MUST
+ /// always have a handler for function entry and exit events. In case the
+ /// implementation wants to support arg1 (or other future extensions to XRay
+ /// logging) those MUST be installed by the installed 'log_init' handler.
+ ///
+ /// Because we didn't want to change the ABI of this struct, the arg1 handler
+ /// may be silently overwritten during initialization as well.
+ void (*handle_arg0)(int32_t, XRayEntryType);
+
+ /// The log implementation provided routine for when __xray_log_flushLog() is
+ /// called.
+ ///
+ /// See XRayLogFlushStatus for details on what the implementation MUST return
+ /// when called.
+ XRayLogFlushStatus (*flush_log)();
+};
+
+/// DEPRECATED: Use the mode registration workflow instead with
+/// __xray_log_register_mode(...) and __xray_log_select_mode(...). See the
+/// documentation for those function.
+///
+/// This function installs a new logging implementation that XRay will use. In
+/// case there are any nullptr members in Impl, XRay will *uninstall any
+/// existing implementations*. It does NOT patch the instrumentation points.
+///
+/// NOTE: This function does NOT attempt to finalize the currently installed
+/// implementation. Use with caution.
+///
+/// It is guaranteed safe to call this function in the following states:
+///
+/// - When the implementation is UNINITIALIZED.
+/// - When the implementation is FINALIZED.
+/// - When there is no current implementation installed.
+///
+/// It is logging implementation defined what happens when this function is
+/// called while in any other states.
+void __xray_set_log_impl(XRayLogImpl Impl);
+
+/// This function registers a logging implementation against a "mode"
+/// identifier. This allows multiple modes to be registered, and chosen at
+/// runtime using the same mode identifier through
+/// `__xray_log_select_mode(...)`.
+///
+/// We treat the Mode identifier as a null-terminated byte string, as the
+/// identifier used when retrieving the log impl.
+///
+/// Returns:
+/// - XRAY_REGISTRATION_OK on success.
+/// - XRAY_DUPLICATE_MODE when an implementation is already associated with
+/// the provided Mode; does not update the already-registered
+/// implementation.
+XRayLogRegisterStatus __xray_log_register_mode(const char *Mode,
+ XRayLogImpl Impl);
+
+/// This function selects the implementation associated with Mode that has been
+/// registered through __xray_log_register_mode(...) and installs that
+/// implementation (as if through calling __xray_set_log_impl(...)). The same
+/// caveats apply to __xray_log_select_mode(...) as with
+/// __xray_log_set_log_impl(...).
+///
+/// Returns:
+/// - XRAY_REGISTRATION_OK on success.
+/// - XRAY_MODE_NOT_FOUND if there is no implementation associated with Mode;
+/// does not update the currently installed implementation.
+XRayLogRegisterStatus __xray_log_select_mode(const char *Mode);
+
+/// Returns an identifier for the currently selected XRay mode chosen through
+/// the __xray_log_select_mode(...) function call. Returns nullptr if there is
+/// no currently installed mode.
+const char *__xray_log_get_current_mode();
+
+/// This function removes the currently installed implementation. It will also
+/// uninstall any handlers that have been previously installed. It does NOT
+/// unpatch the instrumentation points.
+///
+/// NOTE: This function does NOT attempt to finalize the currently installed
+/// implementation. Use with caution.
+///
+/// It is guaranteed safe to call this function in the following states:
+///
+/// - When the implementation is UNINITIALIZED.
+/// - When the implementation is FINALIZED.
+/// - When there is no current implementation installed.
+///
+/// It is logging implementation defined what happens when this function is
+/// called while in any other states.
+void __xray_remove_log_impl();
+
+/// DEPRECATED: Use __xray_log_init_mode() instead, and provide all the options
+/// in string form.
+/// Invokes the installed implementation initialization routine. See
+/// XRayLogInitStatus for what the return values mean.
+XRayLogInitStatus __xray_log_init(size_t BufferSize, size_t MaxBuffers,
+ void *Args, size_t ArgsSize);
+
+/// Invokes the installed initialization routine, which *must* support the
+/// string based form.
+///
+/// NOTE: When this API is used, we still invoke the installed initialization
+/// routine, but we will call it with the following convention to signal that we
+/// are using the string form:
+///
+/// - BufferSize = 0
+/// - MaxBuffers = 0
+/// - ArgsSize = 0
+/// - Args will be the pointer to the character buffer representing the
+/// configuration.
+///
+/// FIXME: Updating the XRayLogImpl struct is an ABI breaking change. When we
+/// are ready to make a breaking change, we should clean this up appropriately.
+XRayLogInitStatus __xray_log_init_mode(const char *Mode, const char *Config);
+
+/// Like __xray_log_init_mode(...) this version allows for providing
+/// configurations that might have non-null-terminated strings. This will
+/// operate similarly to __xray_log_init_mode, with the exception that
+/// |ArgsSize| will be what |ConfigSize| is.
+XRayLogInitStatus __xray_log_init_mode_bin(const char *Mode, const char *Config,
+ size_t ConfigSize);
+
+/// Invokes the installed implementation finalization routine. See
+/// XRayLogInitStatus for what the return values mean.
+XRayLogInitStatus __xray_log_finalize();
+
+/// Invokes the install implementation log flushing routine. See
+/// XRayLogFlushStatus for what the return values mean.
+XRayLogFlushStatus __xray_log_flushLog();
+
+/// An XRayBuffer represents a section of memory which can be treated by log
+/// processing functions as bytes stored in the logging implementation's
+/// buffers.
+struct XRayBuffer {
+ const void *Data;
+ size_t Size;
+};
+
+/// Registers an iterator function which takes an XRayBuffer argument, then
+/// returns another XRayBuffer function representing the next buffer. When the
+/// Iterator function returns an empty XRayBuffer (Data = nullptr, Size = 0),
+/// this signifies the end of the buffers.
+///
+/// The first invocation of this Iterator function will always take an empty
+/// XRayBuffer (Data = nullptr, Size = 0).
+void __xray_log_set_buffer_iterator(XRayBuffer (*Iterator)(XRayBuffer));
+
+/// Removes the currently registered buffer iterator function.
+void __xray_log_remove_buffer_iterator();
+
+/// Invokes the provided handler to process data maintained by the logging
+/// handler. This API will be provided raw access to the data available in
+/// memory from the logging implementation. The callback function must:
+///
+/// 1) Not modify the data, to avoid running into undefined behaviour.
+///
+/// 2) Either know the data layout, or treat the data as raw bytes for later
+/// interpretation.
+///
+/// This API is best used in place of the `__xray_log_flushLog()` implementation
+/// above to enable the caller to provide an alternative means of extracting the
+/// data from the XRay implementation.
+///
+/// Implementations MUST then provide:
+///
+/// 1) A function that will return an XRayBuffer. Functions that return an
+/// "empty" XRayBuffer signifies that there are no more buffers to be
+/// processed. This function should be registered through the
+/// `__xray_log_set_buffer_iterator(...)` function.
+///
+/// 2) Its own means of converting data it holds in memory into an XRayBuffer
+/// structure.
+///
+/// See XRayLogFlushStatus for what the return values mean.
+///
+XRayLogFlushStatus __xray_log_process_buffers(void (*Processor)(const char *,
+ XRayBuffer));
+
+} // extern "C"
+
+#endif // XRAY_XRAY_LOG_INTERFACE_H
--- /dev/null
+//===-- xray_records.h ------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file is a part of XRay, a dynamic runtime instrumentation system.
+//
+// This header exposes some record types useful for the XRay in-memory logging
+// implementation.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef XRAY_XRAY_RECORDS_H
+#define XRAY_XRAY_RECORDS_H
+
+#include <cstdint>
+
+namespace __xray {
+
+enum FileTypes {
+ NAIVE_LOG = 0,
+ FDR_LOG = 1,
+};
+
+// FDR mode use of the union field in the XRayFileHeader.
+struct alignas(16) FdrAdditionalHeaderData {
+ uint64_t ThreadBufferSize;
+};
+
+static_assert(sizeof(FdrAdditionalHeaderData) == 16,
+ "FdrAdditionalHeaderData != 16 bytes");
+
+// This data structure is used to describe the contents of the file. We use this
+// for versioning the supported XRay file formats.
+struct alignas(32) XRayFileHeader {
+ uint16_t Version = 0;
+
+ // The type of file we're writing out. See the FileTypes enum for more
+ // information. This allows different implementations of the XRay logging to
+ // have different files for different information being stored.
+ uint16_t Type = 0;
+
+ // What follows are a set of flags that indicate useful things for when
+ // reading the data in the file.
+ bool ConstantTSC : 1;
+ bool NonstopTSC : 1;
+
+ // The frequency by which TSC increases per-second.
+ alignas(8) uint64_t CycleFrequency = 0;
+
+ union {
+ char FreeForm[16];
+ // The current civiltime timestamp, as retrieved from 'clock_gettime'. This
+ // allows readers of the file to determine when the file was created or
+ // written down.
+ struct timespec TS;
+
+ struct FdrAdditionalHeaderData FdrData;
+ };
+} __attribute__((packed));
+
+static_assert(sizeof(XRayFileHeader) == 32, "XRayFileHeader != 32 bytes");
+
+enum RecordTypes {
+ NORMAL = 0,
+ ARG_PAYLOAD = 1,
+};
+
+struct alignas(32) XRayRecord {
+ // This is the type of the record being written. We use 16 bits to allow us to
+ // treat this as a discriminant, and so that the first 4 bytes get packed
+ // properly. See RecordTypes for more supported types.
+ uint16_t RecordType = RecordTypes::NORMAL;
+
+ // The CPU where the thread is running. We assume number of CPUs <= 256.
+ uint8_t CPU = 0;
+
+ // The type of the event. One of the following:
+ // ENTER = 0
+ // EXIT = 1
+ // TAIL_EXIT = 2
+ // ENTER_ARG = 3
+ uint8_t Type = 0;
+
+ // The function ID for the record.
+ int32_t FuncId = 0;
+
+ // Get the full 8 bytes of the TSC when we get the log record.
+ uint64_t TSC = 0;
+
+ // The thread ID for the currently running thread.
+ uint32_t TId = 0;
+
+ // The ID of process that is currently running
+ uint32_t PId = 0;
+
+ // Use some bytes in the end of the record for buffers.
+ char Buffer[8] = {};
+} __attribute__((packed));
+
+static_assert(sizeof(XRayRecord) == 32, "XRayRecord != 32 bytes");
+
+struct alignas(32) XRayArgPayload {
+ // We use the same 16 bits as a discriminant for the records in the log here
+ // too, and so that the first 4 bytes are packed properly.
+ uint16_t RecordType = RecordTypes::ARG_PAYLOAD;
+
+ // Add a few bytes to pad.
+ uint8_t Padding[2] = {};
+
+ // The function ID for the record.
+ int32_t FuncId = 0;
+
+ // The thread ID for the currently running thread.
+ uint32_t TId = 0;
+
+ // The ID of process that is currently running
+ uint32_t PId = 0;
+
+ // The argument payload.
+ uint64_t Arg = 0;
+
+ // The rest of this record ought to be left as padding.
+ uint8_t TailPadding[8] = {};
+} __attribute__((packed));
+
+static_assert(sizeof(XRayArgPayload) == 32, "XRayArgPayload != 32 bytes");
+
+} // namespace __xray
+
+#endif // XRAY_XRAY_RECORDS_H
--- /dev/null
+/*===---- xsavecintrin.h - XSAVEC intrinsic --------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <xsavecintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __XSAVECINTRIN_H
+#define __XSAVECINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("xsavec")))
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_xsavec(void *__p, unsigned long long __m) {
+ __builtin_ia32_xsavec(__p, __m);
+}
+
+#ifdef __x86_64__
+static __inline__ void __DEFAULT_FN_ATTRS
+_xsavec64(void *__p, unsigned long long __m) {
+ __builtin_ia32_xsavec64(__p, __m);
+}
+#endif
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif
--- /dev/null
+/*===---- xsaveintrin.h - XSAVE intrinsic ----------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <xsaveintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __XSAVEINTRIN_H
+#define __XSAVEINTRIN_H
+
+#ifdef _MSC_VER
+#define _XCR_XFEATURE_ENABLED_MASK 0
+#endif
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("xsave")))
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_xsave(void *__p, unsigned long long __m) {
+ __builtin_ia32_xsave(__p, __m);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_xrstor(void *__p, unsigned long long __m) {
+ __builtin_ia32_xrstor(__p, __m);
+}
+
+#ifndef _MSC_VER
+#define _xgetbv(A) __builtin_ia32_xgetbv((long long)(A))
+#define _xsetbv(A, B) __builtin_ia32_xsetbv((unsigned int)(A), (unsigned long long)(B))
+#else
+#ifdef __cplusplus
+extern "C" {
+#endif
+unsigned __int64 __cdecl _xgetbv(unsigned int);
+void __cdecl _xsetbv(unsigned int, unsigned __int64);
+#ifdef __cplusplus
+}
+#endif
+#endif /* _MSC_VER */
+
+#ifdef __x86_64__
+static __inline__ void __DEFAULT_FN_ATTRS
+_xsave64(void *__p, unsigned long long __m) {
+ __builtin_ia32_xsave64(__p, __m);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_xrstor64(void *__p, unsigned long long __m) {
+ __builtin_ia32_xrstor64(__p, __m);
+}
+
+#endif
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif
--- /dev/null
+/*===---- xsaveoptintrin.h - XSAVEOPT intrinsic ----------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <xsaveoptintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __XSAVEOPTINTRIN_H
+#define __XSAVEOPTINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("xsaveopt")))
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_xsaveopt(void *__p, unsigned long long __m) {
+ __builtin_ia32_xsaveopt(__p, __m);
+}
+
+#ifdef __x86_64__
+static __inline__ void __DEFAULT_FN_ATTRS
+_xsaveopt64(void *__p, unsigned long long __m) {
+ __builtin_ia32_xsaveopt64(__p, __m);
+}
+#endif
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif
--- /dev/null
+/*===---- xsavesintrin.h - XSAVES intrinsic --------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <xsavesintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __XSAVESINTRIN_H
+#define __XSAVESINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("xsaves")))
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_xsaves(void *__p, unsigned long long __m) {
+ __builtin_ia32_xsaves(__p, __m);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_xrstors(void *__p, unsigned long long __m) {
+ __builtin_ia32_xrstors(__p, __m);
+}
+
+#ifdef __x86_64__
+static __inline__ void __DEFAULT_FN_ATTRS
+_xrstors64(void *__p, unsigned long long __m) {
+ __builtin_ia32_xrstors64(__p, __m);
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS
+_xsaves64(void *__p, unsigned long long __m) {
+ __builtin_ia32_xsaves64(__p, __m);
+}
+#endif
+
+#undef __DEFAULT_FN_ATTRS
+
+#endif
--- /dev/null
+/*===---- xtestintrin.h - XTEST intrinsic ----------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <xtestintrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef __XTESTINTRIN_H
+#define __XTESTINTRIN_H
+
+/* xtest returns non-zero if the instruction is executed within an RTM or active
+ * HLE region. */
+/* FIXME: This can be an either or for RTM/HLE. Deal with this when HLE is
+ * supported. */
+static __inline__ int
+ __attribute__((__always_inline__, __nodebug__, __target__("rtm")))
+ _xtest(void) {
+ return __builtin_ia32_xtest();
+}
+
+#endif
+++ /dev/null
-/*===-- __clang_cuda_complex_builtins - CUDA impls of runtime complex fns ---===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __CLANG_CUDA_COMPLEX_BUILTINS
-#define __CLANG_CUDA_COMPLEX_BUILTINS
-
-// This header defines __muldc3, __mulsc3, __divdc3, and __divsc3. These are
-// libgcc functions that clang assumes are available when compiling c99 complex
-// operations. (These implementations come from libc++, and have been modified
-// to work with CUDA.)
-
-extern "C" inline __device__ double _Complex __muldc3(double __a, double __b,
- double __c, double __d) {
- double __ac = __a * __c;
- double __bd = __b * __d;
- double __ad = __a * __d;
- double __bc = __b * __c;
- double _Complex z;
- __real__(z) = __ac - __bd;
- __imag__(z) = __ad + __bc;
- if (std::isnan(__real__(z)) && std::isnan(__imag__(z))) {
- int __recalc = 0;
- if (std::isinf(__a) || std::isinf(__b)) {
- __a = std::copysign(std::isinf(__a) ? 1 : 0, __a);
- __b = std::copysign(std::isinf(__b) ? 1 : 0, __b);
- if (std::isnan(__c))
- __c = std::copysign(0, __c);
- if (std::isnan(__d))
- __d = std::copysign(0, __d);
- __recalc = 1;
- }
- if (std::isinf(__c) || std::isinf(__d)) {
- __c = std::copysign(std::isinf(__c) ? 1 : 0, __c);
- __d = std::copysign(std::isinf(__d) ? 1 : 0, __d);
- if (std::isnan(__a))
- __a = std::copysign(0, __a);
- if (std::isnan(__b))
- __b = std::copysign(0, __b);
- __recalc = 1;
- }
- if (!__recalc && (std::isinf(__ac) || std::isinf(__bd) ||
- std::isinf(__ad) || std::isinf(__bc))) {
- if (std::isnan(__a))
- __a = std::copysign(0, __a);
- if (std::isnan(__b))
- __b = std::copysign(0, __b);
- if (std::isnan(__c))
- __c = std::copysign(0, __c);
- if (std::isnan(__d))
- __d = std::copysign(0, __d);
- __recalc = 1;
- }
- if (__recalc) {
- // Can't use std::numeric_limits<double>::infinity() -- that doesn't have
- // a device overload (and isn't constexpr before C++11, naturally).
- __real__(z) = __builtin_huge_valf() * (__a * __c - __b * __d);
- __imag__(z) = __builtin_huge_valf() * (__a * __d + __b * __c);
- }
- }
- return z;
-}
-
-extern "C" inline __device__ float _Complex __mulsc3(float __a, float __b,
- float __c, float __d) {
- float __ac = __a * __c;
- float __bd = __b * __d;
- float __ad = __a * __d;
- float __bc = __b * __c;
- float _Complex z;
- __real__(z) = __ac - __bd;
- __imag__(z) = __ad + __bc;
- if (std::isnan(__real__(z)) && std::isnan(__imag__(z))) {
- int __recalc = 0;
- if (std::isinf(__a) || std::isinf(__b)) {
- __a = std::copysign(std::isinf(__a) ? 1 : 0, __a);
- __b = std::copysign(std::isinf(__b) ? 1 : 0, __b);
- if (std::isnan(__c))
- __c = std::copysign(0, __c);
- if (std::isnan(__d))
- __d = std::copysign(0, __d);
- __recalc = 1;
- }
- if (std::isinf(__c) || std::isinf(__d)) {
- __c = std::copysign(std::isinf(__c) ? 1 : 0, __c);
- __d = std::copysign(std::isinf(__d) ? 1 : 0, __d);
- if (std::isnan(__a))
- __a = std::copysign(0, __a);
- if (std::isnan(__b))
- __b = std::copysign(0, __b);
- __recalc = 1;
- }
- if (!__recalc && (std::isinf(__ac) || std::isinf(__bd) ||
- std::isinf(__ad) || std::isinf(__bc))) {
- if (std::isnan(__a))
- __a = std::copysign(0, __a);
- if (std::isnan(__b))
- __b = std::copysign(0, __b);
- if (std::isnan(__c))
- __c = std::copysign(0, __c);
- if (std::isnan(__d))
- __d = std::copysign(0, __d);
- __recalc = 1;
- }
- if (__recalc) {
- __real__(z) = __builtin_huge_valf() * (__a * __c - __b * __d);
- __imag__(z) = __builtin_huge_valf() * (__a * __d + __b * __c);
- }
- }
- return z;
-}
-
-extern "C" inline __device__ double _Complex __divdc3(double __a, double __b,
- double __c, double __d) {
- int __ilogbw = 0;
- // Can't use std::max, because that's defined in <algorithm>, and we don't
- // want to pull that in for every compile. The CUDA headers define
- // ::max(float, float) and ::max(double, double), which is sufficient for us.
- double __logbw = std::logb(max(std::abs(__c), std::abs(__d)));
- if (std::isfinite(__logbw)) {
- __ilogbw = (int)__logbw;
- __c = std::scalbn(__c, -__ilogbw);
- __d = std::scalbn(__d, -__ilogbw);
- }
- double __denom = __c * __c + __d * __d;
- double _Complex z;
- __real__(z) = std::scalbn((__a * __c + __b * __d) / __denom, -__ilogbw);
- __imag__(z) = std::scalbn((__b * __c - __a * __d) / __denom, -__ilogbw);
- if (std::isnan(__real__(z)) && std::isnan(__imag__(z))) {
- if ((__denom == 0.0) && (!std::isnan(__a) || !std::isnan(__b))) {
- __real__(z) = std::copysign(__builtin_huge_valf(), __c) * __a;
- __imag__(z) = std::copysign(__builtin_huge_valf(), __c) * __b;
- } else if ((std::isinf(__a) || std::isinf(__b)) && std::isfinite(__c) &&
- std::isfinite(__d)) {
- __a = std::copysign(std::isinf(__a) ? 1.0 : 0.0, __a);
- __b = std::copysign(std::isinf(__b) ? 1.0 : 0.0, __b);
- __real__(z) = __builtin_huge_valf() * (__a * __c + __b * __d);
- __imag__(z) = __builtin_huge_valf() * (__b * __c - __a * __d);
- } else if (std::isinf(__logbw) && __logbw > 0.0 && std::isfinite(__a) &&
- std::isfinite(__b)) {
- __c = std::copysign(std::isinf(__c) ? 1.0 : 0.0, __c);
- __d = std::copysign(std::isinf(__d) ? 1.0 : 0.0, __d);
- __real__(z) = 0.0 * (__a * __c + __b * __d);
- __imag__(z) = 0.0 * (__b * __c - __a * __d);
- }
- }
- return z;
-}
-
-extern "C" inline __device__ float _Complex __divsc3(float __a, float __b,
- float __c, float __d) {
- int __ilogbw = 0;
- float __logbw = std::logb(max(std::abs(__c), std::abs(__d)));
- if (std::isfinite(__logbw)) {
- __ilogbw = (int)__logbw;
- __c = std::scalbn(__c, -__ilogbw);
- __d = std::scalbn(__d, -__ilogbw);
- }
- float __denom = __c * __c + __d * __d;
- float _Complex z;
- __real__(z) = std::scalbn((__a * __c + __b * __d) / __denom, -__ilogbw);
- __imag__(z) = std::scalbn((__b * __c - __a * __d) / __denom, -__ilogbw);
- if (std::isnan(__real__(z)) && std::isnan(__imag__(z))) {
- if ((__denom == 0) && (!std::isnan(__a) || !std::isnan(__b))) {
- __real__(z) = std::copysign(__builtin_huge_valf(), __c) * __a;
- __imag__(z) = std::copysign(__builtin_huge_valf(), __c) * __b;
- } else if ((std::isinf(__a) || std::isinf(__b)) && std::isfinite(__c) &&
- std::isfinite(__d)) {
- __a = std::copysign(std::isinf(__a) ? 1 : 0, __a);
- __b = std::copysign(std::isinf(__b) ? 1 : 0, __b);
- __real__(z) = __builtin_huge_valf() * (__a * __c + __b * __d);
- __imag__(z) = __builtin_huge_valf() * (__b * __c - __a * __d);
- } else if (std::isinf(__logbw) && __logbw > 0 && std::isfinite(__a) &&
- std::isfinite(__b)) {
- __c = std::copysign(std::isinf(__c) ? 1 : 0, __c);
- __d = std::copysign(std::isinf(__d) ? 1 : 0, __d);
- __real__(z) = 0 * (__a * __c + __b * __d);
- __imag__(z) = 0 * (__b * __c - __a * __d);
- }
- }
- return z;
-}
-
-#endif // __CLANG_CUDA_COMPLEX_BUILTINS
+++ /dev/null
-/*===---- __stddef_max_align_t.h - Definition of max_align_t for modules ---===
- *
- * Copyright (c) 2014 Chandler Carruth
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __CLANG_MAX_ALIGN_T_DEFINED
-#define __CLANG_MAX_ALIGN_T_DEFINED
-
-#if defined(_MSC_VER)
-typedef double max_align_t;
-#elif defined(__APPLE__)
-typedef long double max_align_t;
-#else
-// Define 'max_align_t' to match the GCC definition.
-typedef struct {
- long long __clang_max_align_nonce1
- __attribute__((__aligned__(__alignof__(long long))));
- long double __clang_max_align_nonce2
- __attribute__((__aligned__(__alignof__(long double))));
-} max_align_t;
-#endif
-
-#endif
+++ /dev/null
-/*===---- __wmmintrin_pclmul.h - PCMUL intrinsics ---------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-#ifndef _WMMINTRIN_PCLMUL_H
-#define _WMMINTRIN_PCLMUL_H
-
-/// \brief Multiplies two 64-bit integer values, which are selected from source
-/// operands using the immediate-value operand. The multiplication is a
-/// carry-less multiplication, and the 128-bit integer product is stored in
-/// the destination.
-///
-/// \headerfile <x86intrin.h>
-///
-/// \code
-/// __m128i _mm_clmulepi64_si128(__m128i __X, __m128i __Y, const int __I);
-/// \endcode
-///
-/// This intrinsic corresponds to the <c> VPCLMULQDQ </c> instruction.
-///
-/// \param __X
-/// A 128-bit vector of [2 x i64] containing one of the source operands.
-/// \param __Y
-/// A 128-bit vector of [2 x i64] containing one of the source operands.
-/// \param __I
-/// An immediate value specifying which 64-bit values to select from the
-/// operands. Bit 0 is used to select a value from operand \a __X, and bit
-/// 4 is used to select a value from operand \a __Y: \n
-/// Bit[0]=0 indicates that bits[63:0] of operand \a __X are used. \n
-/// Bit[0]=1 indicates that bits[127:64] of operand \a __X are used. \n
-/// Bit[4]=0 indicates that bits[63:0] of operand \a __Y are used. \n
-/// Bit[4]=1 indicates that bits[127:64] of operand \a __Y are used.
-/// \returns The 128-bit integer vector containing the result of the carry-less
-/// multiplication of the selected 64-bit values.
-#define _mm_clmulepi64_si128(__X, __Y, __I) \
- ((__m128i)__builtin_ia32_pclmulqdq128((__v2di)(__m128i)(__X), \
- (__v2di)(__m128i)(__Y), (char)(__I)))
-
-#endif /* _WMMINTRIN_PCLMUL_H */
+++ /dev/null
-/*===---- armintr.h - ARM Windows intrinsics -------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-/* Only include this if we're compiling for the windows platform. */
-#ifndef _MSC_VER
-#include_next <armintr.h>
-#else
-
-#ifndef __ARMINTR_H
-#define __ARMINTR_H
-
-typedef enum
-{
- _ARM_BARRIER_SY = 0xF,
- _ARM_BARRIER_ST = 0xE,
- _ARM_BARRIER_ISH = 0xB,
- _ARM_BARRIER_ISHST = 0xA,
- _ARM_BARRIER_NSH = 0x7,
- _ARM_BARRIER_NSHST = 0x6,
- _ARM_BARRIER_OSH = 0x3,
- _ARM_BARRIER_OSHST = 0x2
-} _ARMINTR_BARRIER_TYPE;
-
-#endif /* __ARMINTR_H */
-#endif /* _MSC_VER */
+++ /dev/null
-/*===------------- avx512bwintrin.h - AVX512BW intrinsics ------------------===
- *
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-#ifndef __IMMINTRIN_H
-#error "Never use <avx512bwintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __AVX512BWINTRIN_H
-#define __AVX512BWINTRIN_H
-
-typedef unsigned int __mmask32;
-typedef unsigned long long __mmask64;
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512bw")))
-
-static __inline __m512i __DEFAULT_FN_ATTRS
-_mm512_setzero_qi(void) {
- return (__m512i)(__v64qi){ 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0 };
-}
-
-static __inline __m512i __DEFAULT_FN_ATTRS
-_mm512_setzero_hi(void) {
- return (__m512i)(__v32hi){ 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0 };
-}
-
-/* Integer compare */
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_cmpeq_epi8_mask(__m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_pcmpeqb512_mask((__v64qi)__a, (__v64qi)__b,
- (__mmask64)-1);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpeq_epi8_mask(__mmask64 __u, __m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_pcmpeqb512_mask((__v64qi)__a, (__v64qi)__b,
- __u);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_cmpeq_epu8_mask(__m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)__a, (__v64qi)__b, 0,
- (__mmask64)-1);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpeq_epu8_mask(__mmask64 __u, __m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)__a, (__v64qi)__b, 0,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_cmpeq_epi16_mask(__m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_pcmpeqw512_mask((__v32hi)__a, (__v32hi)__b,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpeq_epi16_mask(__mmask32 __u, __m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_pcmpeqw512_mask((__v32hi)__a, (__v32hi)__b,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_cmpeq_epu16_mask(__m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)__a, (__v32hi)__b, 0,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpeq_epu16_mask(__mmask32 __u, __m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)__a, (__v32hi)__b, 0,
- __u);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_cmpge_epi8_mask(__m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_cmpb512_mask((__v64qi)__a, (__v64qi)__b, 5,
- (__mmask64)-1);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpge_epi8_mask(__mmask64 __u, __m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_cmpb512_mask((__v64qi)__a, (__v64qi)__b, 5,
- __u);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_cmpge_epu8_mask(__m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)__a, (__v64qi)__b, 5,
- (__mmask64)-1);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpge_epu8_mask(__mmask64 __u, __m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)__a, (__v64qi)__b, 5,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_cmpge_epi16_mask(__m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_cmpw512_mask((__v32hi)__a, (__v32hi)__b, 5,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpge_epi16_mask(__mmask32 __u, __m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_cmpw512_mask((__v32hi)__a, (__v32hi)__b, 5,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_cmpge_epu16_mask(__m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)__a, (__v32hi)__b, 5,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpge_epu16_mask(__mmask32 __u, __m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)__a, (__v32hi)__b, 5,
- __u);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_cmpgt_epi8_mask(__m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_pcmpgtb512_mask((__v64qi)__a, (__v64qi)__b,
- (__mmask64)-1);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpgt_epi8_mask(__mmask64 __u, __m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_pcmpgtb512_mask((__v64qi)__a, (__v64qi)__b,
- __u);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_cmpgt_epu8_mask(__m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)__a, (__v64qi)__b, 6,
- (__mmask64)-1);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpgt_epu8_mask(__mmask64 __u, __m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)__a, (__v64qi)__b, 6,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_cmpgt_epi16_mask(__m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_pcmpgtw512_mask((__v32hi)__a, (__v32hi)__b,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpgt_epi16_mask(__mmask32 __u, __m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_pcmpgtw512_mask((__v32hi)__a, (__v32hi)__b,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_cmpgt_epu16_mask(__m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)__a, (__v32hi)__b, 6,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpgt_epu16_mask(__mmask32 __u, __m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)__a, (__v32hi)__b, 6,
- __u);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_cmple_epi8_mask(__m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_cmpb512_mask((__v64qi)__a, (__v64qi)__b, 2,
- (__mmask64)-1);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_mask_cmple_epi8_mask(__mmask64 __u, __m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_cmpb512_mask((__v64qi)__a, (__v64qi)__b, 2,
- __u);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_cmple_epu8_mask(__m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)__a, (__v64qi)__b, 2,
- (__mmask64)-1);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_mask_cmple_epu8_mask(__mmask64 __u, __m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)__a, (__v64qi)__b, 2,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_cmple_epi16_mask(__m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_cmpw512_mask((__v32hi)__a, (__v32hi)__b, 2,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_mask_cmple_epi16_mask(__mmask32 __u, __m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_cmpw512_mask((__v32hi)__a, (__v32hi)__b, 2,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_cmple_epu16_mask(__m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)__a, (__v32hi)__b, 2,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_mask_cmple_epu16_mask(__mmask32 __u, __m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)__a, (__v32hi)__b, 2,
- __u);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_cmplt_epi8_mask(__m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_cmpb512_mask((__v64qi)__a, (__v64qi)__b, 1,
- (__mmask64)-1);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_mask_cmplt_epi8_mask(__mmask64 __u, __m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_cmpb512_mask((__v64qi)__a, (__v64qi)__b, 1,
- __u);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_cmplt_epu8_mask(__m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)__a, (__v64qi)__b, 1,
- (__mmask64)-1);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_mask_cmplt_epu8_mask(__mmask64 __u, __m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)__a, (__v64qi)__b, 1,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_cmplt_epi16_mask(__m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_cmpw512_mask((__v32hi)__a, (__v32hi)__b, 1,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_mask_cmplt_epi16_mask(__mmask32 __u, __m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_cmpw512_mask((__v32hi)__a, (__v32hi)__b, 1,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_cmplt_epu16_mask(__m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)__a, (__v32hi)__b, 1,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_mask_cmplt_epu16_mask(__mmask32 __u, __m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)__a, (__v32hi)__b, 1,
- __u);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_cmpneq_epi8_mask(__m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_cmpb512_mask((__v64qi)__a, (__v64qi)__b, 4,
- (__mmask64)-1);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpneq_epi8_mask(__mmask64 __u, __m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_cmpb512_mask((__v64qi)__a, (__v64qi)__b, 4,
- __u);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_cmpneq_epu8_mask(__m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)__a, (__v64qi)__b, 4,
- (__mmask64)-1);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpneq_epu8_mask(__mmask64 __u, __m512i __a, __m512i __b) {
- return (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)__a, (__v64qi)__b, 4,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_cmpneq_epi16_mask(__m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_cmpw512_mask((__v32hi)__a, (__v32hi)__b, 4,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpneq_epi16_mask(__mmask32 __u, __m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_cmpw512_mask((__v32hi)__a, (__v32hi)__b, 4,
- __u);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_cmpneq_epu16_mask(__m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)__a, (__v32hi)__b, 4,
- (__mmask32)-1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_mask_cmpneq_epu16_mask(__mmask32 __u, __m512i __a, __m512i __b) {
- return (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)__a, (__v32hi)__b, 4,
- __u);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_add_epi8 (__m512i __A, __m512i __B) {
- return (__m512i) ((__v64qu) __A + (__v64qu) __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_add_epi8(__m512i __W, __mmask64 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
- (__v64qi)_mm512_add_epi8(__A, __B),
- (__v64qi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_add_epi8(__mmask64 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
- (__v64qi)_mm512_add_epi8(__A, __B),
- (__v64qi)_mm512_setzero_qi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_sub_epi8 (__m512i __A, __m512i __B) {
- return (__m512i) ((__v64qu) __A - (__v64qu) __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_sub_epi8(__m512i __W, __mmask64 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
- (__v64qi)_mm512_sub_epi8(__A, __B),
- (__v64qi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_sub_epi8(__mmask64 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
- (__v64qi)_mm512_sub_epi8(__A, __B),
- (__v64qi)_mm512_setzero_qi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_add_epi16 (__m512i __A, __m512i __B) {
- return (__m512i) ((__v32hu) __A + (__v32hu) __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_add_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_add_epi16(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_add_epi16(__mmask32 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_add_epi16(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_sub_epi16 (__m512i __A, __m512i __B) {
- return (__m512i) ((__v32hu) __A - (__v32hu) __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_sub_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_sub_epi16(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_sub_epi16(__mmask32 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_sub_epi16(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mullo_epi16 (__m512i __A, __m512i __B) {
- return (__m512i) ((__v32hu) __A * (__v32hu) __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_mullo_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_mullo_epi16(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_mullo_epi16(__mmask32 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_mullo_epi16(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_blend_epi8 (__mmask64 __U, __m512i __A, __m512i __W)
-{
- return (__m512i) __builtin_ia32_selectb_512 ((__mmask64) __U,
- (__v64qi) __W,
- (__v64qi) __A);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_blend_epi16 (__mmask32 __U, __m512i __A, __m512i __W)
-{
- return (__m512i) __builtin_ia32_selectw_512 ((__mmask32) __U,
- (__v32hi) __W,
- (__v32hi) __A);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_abs_epi8 (__m512i __A)
-{
- return (__m512i) __builtin_ia32_pabsb512_mask ((__v64qi) __A,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_abs_epi8 (__m512i __W, __mmask64 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_pabsb512_mask ((__v64qi) __A,
- (__v64qi) __W,
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_abs_epi8 (__mmask64 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_pabsb512_mask ((__v64qi) __A,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_abs_epi16 (__m512i __A)
-{
- return (__m512i) __builtin_ia32_pabsw512_mask ((__v32hi) __A,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_abs_epi16 (__m512i __W, __mmask32 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_pabsw512_mask ((__v32hi) __A,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_abs_epi16 (__mmask32 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_pabsw512_mask ((__v32hi) __A,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_packs_epi32(__m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_packssdw512((__v16si)__A, (__v16si)__B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_packs_epi32(__mmask32 __M, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
- (__v32hi)_mm512_packs_epi32(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_packs_epi32(__m512i __W, __mmask32 __M, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
- (__v32hi)_mm512_packs_epi32(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_packs_epi16(__m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_packsswb512((__v32hi)__A, (__v32hi) __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_packs_epi16(__m512i __W, __mmask64 __M, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
- (__v64qi)_mm512_packs_epi16(__A, __B),
- (__v64qi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_packs_epi16(__mmask64 __M, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
- (__v64qi)_mm512_packs_epi16(__A, __B),
- (__v64qi)_mm512_setzero_qi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_packus_epi32(__m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_packusdw512((__v16si) __A, (__v16si) __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_packus_epi32(__mmask32 __M, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
- (__v32hi)_mm512_packus_epi32(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_packus_epi32(__m512i __W, __mmask32 __M, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__M,
- (__v32hi)_mm512_packus_epi32(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_packus_epi16(__m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_packuswb512((__v32hi) __A, (__v32hi) __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_packus_epi16(__m512i __W, __mmask64 __M, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
- (__v64qi)_mm512_packus_epi16(__A, __B),
- (__v64qi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_packus_epi16(__mmask64 __M, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
- (__v64qi)_mm512_packus_epi16(__A, __B),
- (__v64qi)_mm512_setzero_qi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_adds_epi8 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_paddsb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_adds_epi8 (__m512i __W, __mmask64 __U, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_paddsb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) __W,
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_adds_epi8 (__mmask64 __U, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_paddsb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_adds_epi16 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_paddsw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_adds_epi16 (__m512i __W, __mmask32 __U, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_paddsw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_adds_epi16 (__mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_paddsw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_adds_epu8 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_paddusb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_adds_epu8 (__m512i __W, __mmask64 __U, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_paddusb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) __W,
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_adds_epu8 (__mmask64 __U, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_paddusb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_adds_epu16 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_paddusw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_adds_epu16 (__m512i __W, __mmask32 __U, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_paddusw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_adds_epu16 (__mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_paddusw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_avg_epu8 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pavgb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_avg_epu8 (__m512i __W, __mmask64 __U, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_pavgb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) __W,
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_avg_epu8 (__mmask64 __U, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pavgb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_avg_epu16 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pavgw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_avg_epu16 (__m512i __W, __mmask32 __U, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_pavgw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_avg_epu16 (__mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pavgw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_max_epi8 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmaxsb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_max_epi8 (__mmask64 __M, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmaxsb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_max_epi8 (__m512i __W, __mmask64 __M, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmaxsb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) __W,
- (__mmask64) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_max_epi16 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmaxsw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_max_epi16 (__mmask32 __M, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmaxsw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_max_epi16 (__m512i __W, __mmask32 __M, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmaxsw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) __W,
- (__mmask32) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_max_epu8 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmaxub512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_max_epu8 (__mmask64 __M, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmaxub512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_max_epu8 (__m512i __W, __mmask64 __M, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmaxub512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) __W,
- (__mmask64) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_max_epu16 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmaxuw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_max_epu16 (__mmask32 __M, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmaxuw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_max_epu16 (__m512i __W, __mmask32 __M, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmaxuw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) __W,
- (__mmask32) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_min_epi8 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pminsb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_min_epi8 (__mmask64 __M, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pminsb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_min_epi8 (__m512i __W, __mmask64 __M, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_pminsb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) __W,
- (__mmask64) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_min_epi16 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pminsw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_min_epi16 (__mmask32 __M, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pminsw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_min_epi16 (__m512i __W, __mmask32 __M, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_pminsw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) __W,
- (__mmask32) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_min_epu8 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pminub512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_min_epu8 (__mmask64 __M, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pminub512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_min_epu8 (__m512i __W, __mmask64 __M, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_pminub512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) __W,
- (__mmask64) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_min_epu16 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pminuw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_min_epu16 (__mmask32 __M, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pminuw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_min_epu16 (__m512i __W, __mmask32 __M, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_pminuw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) __W,
- (__mmask32) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_shuffle_epi8(__m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_pshufb512((__v64qi)__A,(__v64qi)__B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_shuffle_epi8(__m512i __W, __mmask64 __U, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
- (__v64qi)_mm512_shuffle_epi8(__A, __B),
- (__v64qi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_shuffle_epi8(__mmask64 __U, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
- (__v64qi)_mm512_shuffle_epi8(__A, __B),
- (__v64qi)_mm512_setzero_qi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_subs_epi8 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_psubsb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_subs_epi8 (__m512i __W, __mmask64 __U, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_psubsb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) __W,
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_subs_epi8 (__mmask64 __U, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_psubsb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_subs_epi16 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_psubsw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_subs_epi16 (__m512i __W, __mmask32 __U, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_psubsw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_subs_epi16 (__mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_psubsw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_subs_epu8 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_psubusb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_subs_epu8 (__m512i __W, __mmask64 __U, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_psubusb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) __W,
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_subs_epu8 (__mmask64 __U, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_psubusb512_mask ((__v64qi) __A,
- (__v64qi) __B,
- (__v64qi) _mm512_setzero_qi(),
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_subs_epu16 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_psubusw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_subs_epu16 (__m512i __W, __mmask32 __U, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_psubusw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_subs_epu16 (__mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_psubusw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask2_permutex2var_epi16 (__m512i __A, __m512i __I,
- __mmask32 __U, __m512i __B)
-{
- return (__m512i) __builtin_ia32_vpermi2varhi512_mask ((__v32hi) __A,
- (__v32hi) __I /* idx */ ,
- (__v32hi) __B,
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_permutex2var_epi16 (__m512i __A, __m512i __I, __m512i __B)
-{
- return (__m512i) __builtin_ia32_vpermt2varhi512_mask ((__v32hi) __I /* idx */,
- (__v32hi) __A,
- (__v32hi) __B,
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_permutex2var_epi16 (__m512i __A, __mmask32 __U,
- __m512i __I, __m512i __B)
-{
- return (__m512i) __builtin_ia32_vpermt2varhi512_mask ((__v32hi) __I /* idx */,
- (__v32hi) __A,
- (__v32hi) __B,
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_permutex2var_epi16 (__mmask32 __U, __m512i __A,
- __m512i __I, __m512i __B)
-{
- return (__m512i) __builtin_ia32_vpermt2varhi512_maskz ((__v32hi) __I
- /* idx */ ,
- (__v32hi) __A,
- (__v32hi) __B,
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mulhrs_epi16 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmulhrsw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_mulhrs_epi16 (__m512i __W, __mmask32 __U, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmulhrsw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_mulhrs_epi16 (__mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmulhrsw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mulhi_epi16 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmulhw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_mulhi_epi16 (__m512i __W, __mmask32 __U, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmulhw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_mulhi_epi16 (__mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmulhw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mulhi_epu16 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmulhuw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_mulhi_epu16 (__m512i __W, __mmask32 __U, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmulhuw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_mulhi_epu16 (__mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_pmulhuw512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maddubs_epi16 (__m512i __X, __m512i __Y) {
- return (__m512i) __builtin_ia32_pmaddubsw512_mask ((__v64qi) __X,
- (__v64qi) __Y,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_maddubs_epi16 (__m512i __W, __mmask32 __U, __m512i __X,
- __m512i __Y) {
- return (__m512i) __builtin_ia32_pmaddubsw512_mask ((__v64qi) __X,
- (__v64qi) __Y,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_maddubs_epi16 (__mmask32 __U, __m512i __X, __m512i __Y) {
- return (__m512i) __builtin_ia32_pmaddubsw512_mask ((__v64qi) __X,
- (__v64qi) __Y,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_madd_epi16 (__m512i __A, __m512i __B) {
- return (__m512i) __builtin_ia32_pmaddwd512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v16si) _mm512_setzero_si512(),
- (__mmask16) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_madd_epi16 (__m512i __W, __mmask16 __U, __m512i __A,
- __m512i __B) {
- return (__m512i) __builtin_ia32_pmaddwd512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v16si) __W,
- (__mmask16) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_madd_epi16 (__mmask16 __U, __m512i __A, __m512i __B) {
- return (__m512i) __builtin_ia32_pmaddwd512_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v16si) _mm512_setzero_si512(),
- (__mmask16) __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm512_cvtsepi16_epi8 (__m512i __A) {
- return (__m256i) __builtin_ia32_pmovswb512_mask ((__v32hi) __A,
- (__v32qi)_mm256_setzero_si256(),
- (__mmask32) -1);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm512_mask_cvtsepi16_epi8 (__m256i __O, __mmask32 __M, __m512i __A) {
- return (__m256i) __builtin_ia32_pmovswb512_mask ((__v32hi) __A,
- (__v32qi)__O,
- __M);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm512_maskz_cvtsepi16_epi8 (__mmask32 __M, __m512i __A) {
- return (__m256i) __builtin_ia32_pmovswb512_mask ((__v32hi) __A,
- (__v32qi) _mm256_setzero_si256(),
- __M);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm512_cvtusepi16_epi8 (__m512i __A) {
- return (__m256i) __builtin_ia32_pmovuswb512_mask ((__v32hi) __A,
- (__v32qi) _mm256_setzero_si256(),
- (__mmask32) -1);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm512_mask_cvtusepi16_epi8 (__m256i __O, __mmask32 __M, __m512i __A) {
- return (__m256i) __builtin_ia32_pmovuswb512_mask ((__v32hi) __A,
- (__v32qi) __O,
- __M);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm512_maskz_cvtusepi16_epi8 (__mmask32 __M, __m512i __A) {
- return (__m256i) __builtin_ia32_pmovuswb512_mask ((__v32hi) __A,
- (__v32qi) _mm256_setzero_si256(),
- __M);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm512_cvtepi16_epi8 (__m512i __A) {
- return (__m256i) __builtin_ia32_pmovwb512_mask ((__v32hi) __A,
- (__v32qi) _mm256_setzero_si256(),
- (__mmask32) -1);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm512_mask_cvtepi16_epi8 (__m256i __O, __mmask32 __M, __m512i __A) {
- return (__m256i) __builtin_ia32_pmovwb512_mask ((__v32hi) __A,
- (__v32qi) __O,
- __M);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm512_maskz_cvtepi16_epi8 (__mmask32 __M, __m512i __A) {
- return (__m256i) __builtin_ia32_pmovwb512_mask ((__v32hi) __A,
- (__v32qi) _mm256_setzero_si256(),
- __M);
-}
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm512_mask_cvtepi16_storeu_epi8 (void * __P, __mmask32 __M, __m512i __A)
-{
- __builtin_ia32_pmovwb512mem_mask ((__v32qi *) __P, (__v32hi) __A, __M);
-}
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm512_mask_cvtsepi16_storeu_epi8 (void * __P, __mmask32 __M, __m512i __A)
-{
- __builtin_ia32_pmovswb512mem_mask ((__v32qi *) __P, (__v32hi) __A, __M);
-}
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm512_mask_cvtusepi16_storeu_epi8 (void * __P, __mmask32 __M, __m512i __A)
-{
- __builtin_ia32_pmovuswb512mem_mask ((__v32qi *) __P, (__v32hi) __A, __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_unpackhi_epi8(__m512i __A, __m512i __B) {
- return (__m512i)__builtin_shufflevector((__v64qi)__A, (__v64qi)__B,
- 8, 64+8, 9, 64+9,
- 10, 64+10, 11, 64+11,
- 12, 64+12, 13, 64+13,
- 14, 64+14, 15, 64+15,
- 24, 64+24, 25, 64+25,
- 26, 64+26, 27, 64+27,
- 28, 64+28, 29, 64+29,
- 30, 64+30, 31, 64+31,
- 40, 64+40, 41, 64+41,
- 42, 64+42, 43, 64+43,
- 44, 64+44, 45, 64+45,
- 46, 64+46, 47, 64+47,
- 56, 64+56, 57, 64+57,
- 58, 64+58, 59, 64+59,
- 60, 64+60, 61, 64+61,
- 62, 64+62, 63, 64+63);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_unpackhi_epi8(__m512i __W, __mmask64 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
- (__v64qi)_mm512_unpackhi_epi8(__A, __B),
- (__v64qi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_unpackhi_epi8(__mmask64 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
- (__v64qi)_mm512_unpackhi_epi8(__A, __B),
- (__v64qi)_mm512_setzero_qi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_unpackhi_epi16(__m512i __A, __m512i __B) {
- return (__m512i)__builtin_shufflevector((__v32hi)__A, (__v32hi)__B,
- 4, 32+4, 5, 32+5,
- 6, 32+6, 7, 32+7,
- 12, 32+12, 13, 32+13,
- 14, 32+14, 15, 32+15,
- 20, 32+20, 21, 32+21,
- 22, 32+22, 23, 32+23,
- 28, 32+28, 29, 32+29,
- 30, 32+30, 31, 32+31);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_unpackhi_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_unpackhi_epi16(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_unpackhi_epi16(__mmask32 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_unpackhi_epi16(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_unpacklo_epi8(__m512i __A, __m512i __B) {
- return (__m512i)__builtin_shufflevector((__v64qi)__A, (__v64qi)__B,
- 0, 64+0, 1, 64+1,
- 2, 64+2, 3, 64+3,
- 4, 64+4, 5, 64+5,
- 6, 64+6, 7, 64+7,
- 16, 64+16, 17, 64+17,
- 18, 64+18, 19, 64+19,
- 20, 64+20, 21, 64+21,
- 22, 64+22, 23, 64+23,
- 32, 64+32, 33, 64+33,
- 34, 64+34, 35, 64+35,
- 36, 64+36, 37, 64+37,
- 38, 64+38, 39, 64+39,
- 48, 64+48, 49, 64+49,
- 50, 64+50, 51, 64+51,
- 52, 64+52, 53, 64+53,
- 54, 64+54, 55, 64+55);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_unpacklo_epi8(__m512i __W, __mmask64 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
- (__v64qi)_mm512_unpacklo_epi8(__A, __B),
- (__v64qi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_unpacklo_epi8(__mmask64 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectb_512((__mmask64)__U,
- (__v64qi)_mm512_unpacklo_epi8(__A, __B),
- (__v64qi)_mm512_setzero_qi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_unpacklo_epi16(__m512i __A, __m512i __B) {
- return (__m512i)__builtin_shufflevector((__v32hi)__A, (__v32hi)__B,
- 0, 32+0, 1, 32+1,
- 2, 32+2, 3, 32+3,
- 8, 32+8, 9, 32+9,
- 10, 32+10, 11, 32+11,
- 16, 32+16, 17, 32+17,
- 18, 32+18, 19, 32+19,
- 24, 32+24, 25, 32+25,
- 26, 32+26, 27, 32+27);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_unpacklo_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_unpacklo_epi16(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_unpacklo_epi16(__mmask32 __U, __m512i __A, __m512i __B) {
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_unpacklo_epi16(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_cvtepi8_epi16(__m256i __A)
-{
- /* This function always performs a signed extension, but __v32qi is a char
- which may be signed or unsigned, so use __v32qs. */
- return (__m512i)__builtin_convertvector((__v32qs)__A, __v32hi);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_cvtepi8_epi16(__m512i __W, __mmask32 __U, __m256i __A)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_cvtepi8_epi16(__A),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_cvtepi8_epi16(__mmask32 __U, __m256i __A)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_cvtepi8_epi16(__A),
- (__v32hi)_mm512_setzero_hi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_cvtepu8_epi16(__m256i __A)
-{
- return (__m512i)__builtin_convertvector((__v32qu)__A, __v32hi);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_cvtepu8_epi16(__m512i __W, __mmask32 __U, __m256i __A)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_cvtepu8_epi16(__A),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_cvtepu8_epi16(__mmask32 __U, __m256i __A)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_cvtepu8_epi16(__A),
- (__v32hi)_mm512_setzero_hi());
-}
-
-
-#define _mm512_cmp_epi8_mask(a, b, p) __extension__ ({ \
- (__mmask64)__builtin_ia32_cmpb512_mask((__v64qi)(__m512i)(a), \
- (__v64qi)(__m512i)(b), (int)(p), \
- (__mmask64)-1); })
-
-#define _mm512_mask_cmp_epi8_mask(m, a, b, p) __extension__ ({ \
- (__mmask64)__builtin_ia32_cmpb512_mask((__v64qi)(__m512i)(a), \
- (__v64qi)(__m512i)(b), (int)(p), \
- (__mmask64)(m)); })
-
-#define _mm512_cmp_epu8_mask(a, b, p) __extension__ ({ \
- (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)(__m512i)(a), \
- (__v64qi)(__m512i)(b), (int)(p), \
- (__mmask64)-1); })
-
-#define _mm512_mask_cmp_epu8_mask(m, a, b, p) __extension__ ({ \
- (__mmask64)__builtin_ia32_ucmpb512_mask((__v64qi)(__m512i)(a), \
- (__v64qi)(__m512i)(b), (int)(p), \
- (__mmask64)(m)); })
-
-#define _mm512_cmp_epi16_mask(a, b, p) __extension__ ({ \
- (__mmask32)__builtin_ia32_cmpw512_mask((__v32hi)(__m512i)(a), \
- (__v32hi)(__m512i)(b), (int)(p), \
- (__mmask32)-1); })
-
-#define _mm512_mask_cmp_epi16_mask(m, a, b, p) __extension__ ({ \
- (__mmask32)__builtin_ia32_cmpw512_mask((__v32hi)(__m512i)(a), \
- (__v32hi)(__m512i)(b), (int)(p), \
- (__mmask32)(m)); })
-
-#define _mm512_cmp_epu16_mask(a, b, p) __extension__ ({ \
- (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)(__m512i)(a), \
- (__v32hi)(__m512i)(b), (int)(p), \
- (__mmask32)-1); })
-
-#define _mm512_mask_cmp_epu16_mask(m, a, b, p) __extension__ ({ \
- (__mmask32)__builtin_ia32_ucmpw512_mask((__v32hi)(__m512i)(a), \
- (__v32hi)(__m512i)(b), (int)(p), \
- (__mmask32)(m)); })
-
-#define _mm512_shufflehi_epi16(A, imm) __extension__ ({ \
- (__m512i)__builtin_shufflevector((__v32hi)(__m512i)(A), \
- (__v32hi)_mm512_undefined_epi32(), \
- 0, 1, 2, 3, \
- 4 + (((imm) >> 0) & 0x3), \
- 4 + (((imm) >> 2) & 0x3), \
- 4 + (((imm) >> 4) & 0x3), \
- 4 + (((imm) >> 6) & 0x3), \
- 8, 9, 10, 11, \
- 12 + (((imm) >> 0) & 0x3), \
- 12 + (((imm) >> 2) & 0x3), \
- 12 + (((imm) >> 4) & 0x3), \
- 12 + (((imm) >> 6) & 0x3), \
- 16, 17, 18, 19, \
- 20 + (((imm) >> 0) & 0x3), \
- 20 + (((imm) >> 2) & 0x3), \
- 20 + (((imm) >> 4) & 0x3), \
- 20 + (((imm) >> 6) & 0x3), \
- 24, 25, 26, 27, \
- 28 + (((imm) >> 0) & 0x3), \
- 28 + (((imm) >> 2) & 0x3), \
- 28 + (((imm) >> 4) & 0x3), \
- 28 + (((imm) >> 6) & 0x3)); })
-
-#define _mm512_mask_shufflehi_epi16(W, U, A, imm) __extension__ ({ \
- (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
- (__v32hi)_mm512_shufflehi_epi16((A), \
- (imm)), \
- (__v32hi)(__m512i)(W)); })
-
-#define _mm512_maskz_shufflehi_epi16(U, A, imm) __extension__ ({ \
- (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
- (__v32hi)_mm512_shufflehi_epi16((A), \
- (imm)), \
- (__v32hi)_mm512_setzero_hi()); })
-
-#define _mm512_shufflelo_epi16(A, imm) __extension__ ({ \
- (__m512i)__builtin_shufflevector((__v32hi)(__m512i)(A), \
- (__v32hi)_mm512_undefined_epi32(), \
- 0 + (((imm) >> 0) & 0x3), \
- 0 + (((imm) >> 2) & 0x3), \
- 0 + (((imm) >> 4) & 0x3), \
- 0 + (((imm) >> 6) & 0x3), \
- 4, 5, 6, 7, \
- 8 + (((imm) >> 0) & 0x3), \
- 8 + (((imm) >> 2) & 0x3), \
- 8 + (((imm) >> 4) & 0x3), \
- 8 + (((imm) >> 6) & 0x3), \
- 12, 13, 14, 15, \
- 16 + (((imm) >> 0) & 0x3), \
- 16 + (((imm) >> 2) & 0x3), \
- 16 + (((imm) >> 4) & 0x3), \
- 16 + (((imm) >> 6) & 0x3), \
- 20, 21, 22, 23, \
- 24 + (((imm) >> 0) & 0x3), \
- 24 + (((imm) >> 2) & 0x3), \
- 24 + (((imm) >> 4) & 0x3), \
- 24 + (((imm) >> 6) & 0x3), \
- 28, 29, 30, 31); })
-
-
-#define _mm512_mask_shufflelo_epi16(W, U, A, imm) __extension__ ({ \
- (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
- (__v32hi)_mm512_shufflelo_epi16((A), \
- (imm)), \
- (__v32hi)(__m512i)(W)); })
-
-
-#define _mm512_maskz_shufflelo_epi16(U, A, imm) __extension__ ({ \
- (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
- (__v32hi)_mm512_shufflelo_epi16((A), \
- (imm)), \
- (__v32hi)_mm512_setzero_hi()); })
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_sllv_epi16(__m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_psllv32hi((__v32hi) __A, (__v32hi) __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_sllv_epi16 (__m512i __W, __mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_sllv_epi16(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_sllv_epi16(__mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_sllv_epi16(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_sll_epi16(__m512i __A, __m128i __B)
-{
- return (__m512i)__builtin_ia32_psllw512((__v32hi) __A, (__v8hi) __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_sll_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m128i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_sll_epi16(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_sll_epi16(__mmask32 __U, __m512i __A, __m128i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_sll_epi16(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_slli_epi16(__m512i __A, int __B)
-{
- return (__m512i)__builtin_ia32_psllwi512((__v32hi)__A, __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_slli_epi16(__m512i __W, __mmask32 __U, __m512i __A, int __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_slli_epi16(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_slli_epi16(__mmask32 __U, __m512i __A, int __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_slli_epi16(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-#define _mm512_bslli_epi128(a, imm) __extension__ ({ \
- (__m512i)__builtin_shufflevector( \
- (__v64qi)_mm512_setzero_si512(), \
- (__v64qi)(__m512i)(a), \
- ((char)(imm)&0xF0) ? 0 : ((char)(imm)>0x0 ? 16 : 64) - (char)(imm), \
- ((char)(imm)&0xF0) ? 1 : ((char)(imm)>0x1 ? 17 : 65) - (char)(imm), \
- ((char)(imm)&0xF0) ? 2 : ((char)(imm)>0x2 ? 18 : 66) - (char)(imm), \
- ((char)(imm)&0xF0) ? 3 : ((char)(imm)>0x3 ? 19 : 67) - (char)(imm), \
- ((char)(imm)&0xF0) ? 4 : ((char)(imm)>0x4 ? 20 : 68) - (char)(imm), \
- ((char)(imm)&0xF0) ? 5 : ((char)(imm)>0x5 ? 21 : 69) - (char)(imm), \
- ((char)(imm)&0xF0) ? 6 : ((char)(imm)>0x6 ? 22 : 70) - (char)(imm), \
- ((char)(imm)&0xF0) ? 7 : ((char)(imm)>0x7 ? 23 : 71) - (char)(imm), \
- ((char)(imm)&0xF0) ? 8 : ((char)(imm)>0x8 ? 24 : 72) - (char)(imm), \
- ((char)(imm)&0xF0) ? 9 : ((char)(imm)>0x9 ? 25 : 73) - (char)(imm), \
- ((char)(imm)&0xF0) ? 10 : ((char)(imm)>0xA ? 26 : 74) - (char)(imm), \
- ((char)(imm)&0xF0) ? 11 : ((char)(imm)>0xB ? 27 : 75) - (char)(imm), \
- ((char)(imm)&0xF0) ? 12 : ((char)(imm)>0xC ? 28 : 76) - (char)(imm), \
- ((char)(imm)&0xF0) ? 13 : ((char)(imm)>0xD ? 29 : 77) - (char)(imm), \
- ((char)(imm)&0xF0) ? 14 : ((char)(imm)>0xE ? 30 : 78) - (char)(imm), \
- ((char)(imm)&0xF0) ? 15 : ((char)(imm)>0xF ? 31 : 79) - (char)(imm), \
- ((char)(imm)&0xF0) ? 16 : ((char)(imm)>0x0 ? 32 : 80) - (char)(imm), \
- ((char)(imm)&0xF0) ? 17 : ((char)(imm)>0x1 ? 33 : 81) - (char)(imm), \
- ((char)(imm)&0xF0) ? 18 : ((char)(imm)>0x2 ? 34 : 82) - (char)(imm), \
- ((char)(imm)&0xF0) ? 19 : ((char)(imm)>0x3 ? 35 : 83) - (char)(imm), \
- ((char)(imm)&0xF0) ? 20 : ((char)(imm)>0x4 ? 36 : 84) - (char)(imm), \
- ((char)(imm)&0xF0) ? 21 : ((char)(imm)>0x5 ? 37 : 85) - (char)(imm), \
- ((char)(imm)&0xF0) ? 22 : ((char)(imm)>0x6 ? 38 : 86) - (char)(imm), \
- ((char)(imm)&0xF0) ? 23 : ((char)(imm)>0x7 ? 39 : 87) - (char)(imm), \
- ((char)(imm)&0xF0) ? 24 : ((char)(imm)>0x8 ? 40 : 88) - (char)(imm), \
- ((char)(imm)&0xF0) ? 25 : ((char)(imm)>0x9 ? 41 : 89) - (char)(imm), \
- ((char)(imm)&0xF0) ? 26 : ((char)(imm)>0xA ? 42 : 90) - (char)(imm), \
- ((char)(imm)&0xF0) ? 27 : ((char)(imm)>0xB ? 43 : 91) - (char)(imm), \
- ((char)(imm)&0xF0) ? 28 : ((char)(imm)>0xC ? 44 : 92) - (char)(imm), \
- ((char)(imm)&0xF0) ? 29 : ((char)(imm)>0xD ? 45 : 93) - (char)(imm), \
- ((char)(imm)&0xF0) ? 30 : ((char)(imm)>0xE ? 46 : 94) - (char)(imm), \
- ((char)(imm)&0xF0) ? 31 : ((char)(imm)>0xF ? 47 : 95) - (char)(imm), \
- ((char)(imm)&0xF0) ? 32 : ((char)(imm)>0x0 ? 48 : 96) - (char)(imm), \
- ((char)(imm)&0xF0) ? 33 : ((char)(imm)>0x1 ? 49 : 97) - (char)(imm), \
- ((char)(imm)&0xF0) ? 34 : ((char)(imm)>0x2 ? 50 : 98) - (char)(imm), \
- ((char)(imm)&0xF0) ? 35 : ((char)(imm)>0x3 ? 51 : 99) - (char)(imm), \
- ((char)(imm)&0xF0) ? 36 : ((char)(imm)>0x4 ? 52 : 100) - (char)(imm), \
- ((char)(imm)&0xF0) ? 37 : ((char)(imm)>0x5 ? 53 : 101) - (char)(imm), \
- ((char)(imm)&0xF0) ? 38 : ((char)(imm)>0x6 ? 54 : 102) - (char)(imm), \
- ((char)(imm)&0xF0) ? 39 : ((char)(imm)>0x7 ? 55 : 103) - (char)(imm), \
- ((char)(imm)&0xF0) ? 40 : ((char)(imm)>0x8 ? 56 : 104) - (char)(imm), \
- ((char)(imm)&0xF0) ? 41 : ((char)(imm)>0x9 ? 57 : 105) - (char)(imm), \
- ((char)(imm)&0xF0) ? 42 : ((char)(imm)>0xA ? 58 : 106) - (char)(imm), \
- ((char)(imm)&0xF0) ? 43 : ((char)(imm)>0xB ? 59 : 107) - (char)(imm), \
- ((char)(imm)&0xF0) ? 44 : ((char)(imm)>0xC ? 60 : 108) - (char)(imm), \
- ((char)(imm)&0xF0) ? 45 : ((char)(imm)>0xD ? 61 : 109) - (char)(imm), \
- ((char)(imm)&0xF0) ? 46 : ((char)(imm)>0xE ? 62 : 110) - (char)(imm), \
- ((char)(imm)&0xF0) ? 47 : ((char)(imm)>0xF ? 63 : 111) - (char)(imm), \
- ((char)(imm)&0xF0) ? 48 : ((char)(imm)>0x0 ? 64 : 112) - (char)(imm), \
- ((char)(imm)&0xF0) ? 49 : ((char)(imm)>0x1 ? 65 : 113) - (char)(imm), \
- ((char)(imm)&0xF0) ? 50 : ((char)(imm)>0x2 ? 66 : 114) - (char)(imm), \
- ((char)(imm)&0xF0) ? 51 : ((char)(imm)>0x3 ? 67 : 115) - (char)(imm), \
- ((char)(imm)&0xF0) ? 52 : ((char)(imm)>0x4 ? 68 : 116) - (char)(imm), \
- ((char)(imm)&0xF0) ? 53 : ((char)(imm)>0x5 ? 69 : 117) - (char)(imm), \
- ((char)(imm)&0xF0) ? 54 : ((char)(imm)>0x6 ? 70 : 118) - (char)(imm), \
- ((char)(imm)&0xF0) ? 55 : ((char)(imm)>0x7 ? 71 : 119) - (char)(imm), \
- ((char)(imm)&0xF0) ? 56 : ((char)(imm)>0x8 ? 72 : 120) - (char)(imm), \
- ((char)(imm)&0xF0) ? 57 : ((char)(imm)>0x9 ? 73 : 121) - (char)(imm), \
- ((char)(imm)&0xF0) ? 58 : ((char)(imm)>0xA ? 74 : 122) - (char)(imm), \
- ((char)(imm)&0xF0) ? 59 : ((char)(imm)>0xB ? 75 : 123) - (char)(imm), \
- ((char)(imm)&0xF0) ? 60 : ((char)(imm)>0xC ? 76 : 124) - (char)(imm), \
- ((char)(imm)&0xF0) ? 61 : ((char)(imm)>0xD ? 77 : 125) - (char)(imm), \
- ((char)(imm)&0xF0) ? 62 : ((char)(imm)>0xE ? 78 : 126) - (char)(imm), \
- ((char)(imm)&0xF0) ? 63 : ((char)(imm)>0xF ? 79 : 127) - (char)(imm)); })
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_srlv_epi16(__m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_psrlv32hi((__v32hi)__A, (__v32hi)__B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_srlv_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_srlv_epi16(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_srlv_epi16(__mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_srlv_epi16(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_srav_epi16(__m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_psrav32hi((__v32hi)__A, (__v32hi)__B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_srav_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_srav_epi16(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_srav_epi16(__mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_srav_epi16(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_sra_epi16(__m512i __A, __m128i __B)
-{
- return (__m512i)__builtin_ia32_psraw512((__v32hi) __A, (__v8hi) __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_sra_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m128i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_sra_epi16(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_sra_epi16(__mmask32 __U, __m512i __A, __m128i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_sra_epi16(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_srai_epi16(__m512i __A, int __B)
-{
- return (__m512i)__builtin_ia32_psrawi512((__v32hi)__A, __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_srai_epi16(__m512i __W, __mmask32 __U, __m512i __A, int __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_srai_epi16(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_srai_epi16(__mmask32 __U, __m512i __A, int __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_srai_epi16(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_srl_epi16(__m512i __A, __m128i __B)
-{
- return (__m512i)__builtin_ia32_psrlw512((__v32hi) __A, (__v8hi) __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_srl_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m128i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_srl_epi16(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_srl_epi16(__mmask32 __U, __m512i __A, __m128i __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_srl_epi16(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_srli_epi16(__m512i __A, int __B)
-{
- return (__m512i)__builtin_ia32_psrlwi512((__v32hi)__A, __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_srli_epi16(__m512i __W, __mmask32 __U, __m512i __A, int __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_srli_epi16(__A, __B),
- (__v32hi)__W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_srli_epi16(__mmask32 __U, __m512i __A, int __B)
-{
- return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
- (__v32hi)_mm512_srli_epi16(__A, __B),
- (__v32hi)_mm512_setzero_hi());
-}
-
-#define _mm512_bsrli_epi128(a, imm) __extension__ ({ \
- (__m512i)__builtin_shufflevector( \
- (__v64qi)(__m512i)(a), \
- (__v64qi)_mm512_setzero_si512(), \
- ((char)(imm)&0xF0) ? 64 : (char)(imm) + ((char)(imm)>0xF ? 48 : 0), \
- ((char)(imm)&0xF0) ? 65 : (char)(imm) + ((char)(imm)>0xE ? 49 : 1), \
- ((char)(imm)&0xF0) ? 66 : (char)(imm) + ((char)(imm)>0xD ? 50 : 2), \
- ((char)(imm)&0xF0) ? 67 : (char)(imm) + ((char)(imm)>0xC ? 51 : 3), \
- ((char)(imm)&0xF0) ? 68 : (char)(imm) + ((char)(imm)>0xB ? 52 : 4), \
- ((char)(imm)&0xF0) ? 69 : (char)(imm) + ((char)(imm)>0xA ? 53 : 5), \
- ((char)(imm)&0xF0) ? 70 : (char)(imm) + ((char)(imm)>0x9 ? 54 : 6), \
- ((char)(imm)&0xF0) ? 71 : (char)(imm) + ((char)(imm)>0x8 ? 55 : 7), \
- ((char)(imm)&0xF0) ? 72 : (char)(imm) + ((char)(imm)>0x7 ? 56 : 8), \
- ((char)(imm)&0xF0) ? 73 : (char)(imm) + ((char)(imm)>0x6 ? 57 : 9), \
- ((char)(imm)&0xF0) ? 74 : (char)(imm) + ((char)(imm)>0x5 ? 58 : 10), \
- ((char)(imm)&0xF0) ? 75 : (char)(imm) + ((char)(imm)>0x4 ? 59 : 11), \
- ((char)(imm)&0xF0) ? 76 : (char)(imm) + ((char)(imm)>0x3 ? 60 : 12), \
- ((char)(imm)&0xF0) ? 77 : (char)(imm) + ((char)(imm)>0x2 ? 61 : 13), \
- ((char)(imm)&0xF0) ? 78 : (char)(imm) + ((char)(imm)>0x1 ? 62 : 14), \
- ((char)(imm)&0xF0) ? 79 : (char)(imm) + ((char)(imm)>0x0 ? 63 : 15), \
- ((char)(imm)&0xF0) ? 80 : (char)(imm) + ((char)(imm)>0xF ? 64 : 16), \
- ((char)(imm)&0xF0) ? 81 : (char)(imm) + ((char)(imm)>0xE ? 65 : 17), \
- ((char)(imm)&0xF0) ? 82 : (char)(imm) + ((char)(imm)>0xD ? 66 : 18), \
- ((char)(imm)&0xF0) ? 83 : (char)(imm) + ((char)(imm)>0xC ? 67 : 19), \
- ((char)(imm)&0xF0) ? 84 : (char)(imm) + ((char)(imm)>0xB ? 68 : 20), \
- ((char)(imm)&0xF0) ? 85 : (char)(imm) + ((char)(imm)>0xA ? 69 : 21), \
- ((char)(imm)&0xF0) ? 86 : (char)(imm) + ((char)(imm)>0x9 ? 70 : 22), \
- ((char)(imm)&0xF0) ? 87 : (char)(imm) + ((char)(imm)>0x8 ? 71 : 23), \
- ((char)(imm)&0xF0) ? 88 : (char)(imm) + ((char)(imm)>0x7 ? 72 : 24), \
- ((char)(imm)&0xF0) ? 89 : (char)(imm) + ((char)(imm)>0x6 ? 73 : 25), \
- ((char)(imm)&0xF0) ? 90 : (char)(imm) + ((char)(imm)>0x5 ? 74 : 26), \
- ((char)(imm)&0xF0) ? 91 : (char)(imm) + ((char)(imm)>0x4 ? 75 : 27), \
- ((char)(imm)&0xF0) ? 92 : (char)(imm) + ((char)(imm)>0x3 ? 76 : 28), \
- ((char)(imm)&0xF0) ? 93 : (char)(imm) + ((char)(imm)>0x2 ? 77 : 29), \
- ((char)(imm)&0xF0) ? 94 : (char)(imm) + ((char)(imm)>0x1 ? 78 : 30), \
- ((char)(imm)&0xF0) ? 95 : (char)(imm) + ((char)(imm)>0x0 ? 79 : 31), \
- ((char)(imm)&0xF0) ? 96 : (char)(imm) + ((char)(imm)>0xF ? 80 : 32), \
- ((char)(imm)&0xF0) ? 97 : (char)(imm) + ((char)(imm)>0xE ? 81 : 33), \
- ((char)(imm)&0xF0) ? 98 : (char)(imm) + ((char)(imm)>0xD ? 82 : 34), \
- ((char)(imm)&0xF0) ? 99 : (char)(imm) + ((char)(imm)>0xC ? 83 : 35), \
- ((char)(imm)&0xF0) ? 100 : (char)(imm) + ((char)(imm)>0xB ? 84 : 36), \
- ((char)(imm)&0xF0) ? 101 : (char)(imm) + ((char)(imm)>0xA ? 85 : 37), \
- ((char)(imm)&0xF0) ? 102 : (char)(imm) + ((char)(imm)>0x9 ? 86 : 38), \
- ((char)(imm)&0xF0) ? 103 : (char)(imm) + ((char)(imm)>0x8 ? 87 : 39), \
- ((char)(imm)&0xF0) ? 104 : (char)(imm) + ((char)(imm)>0x7 ? 88 : 40), \
- ((char)(imm)&0xF0) ? 105 : (char)(imm) + ((char)(imm)>0x6 ? 89 : 41), \
- ((char)(imm)&0xF0) ? 106 : (char)(imm) + ((char)(imm)>0x5 ? 90 : 42), \
- ((char)(imm)&0xF0) ? 107 : (char)(imm) + ((char)(imm)>0x4 ? 91 : 43), \
- ((char)(imm)&0xF0) ? 108 : (char)(imm) + ((char)(imm)>0x3 ? 92 : 44), \
- ((char)(imm)&0xF0) ? 109 : (char)(imm) + ((char)(imm)>0x2 ? 93 : 45), \
- ((char)(imm)&0xF0) ? 110 : (char)(imm) + ((char)(imm)>0x1 ? 94 : 46), \
- ((char)(imm)&0xF0) ? 111 : (char)(imm) + ((char)(imm)>0x0 ? 95 : 47), \
- ((char)(imm)&0xF0) ? 112 : (char)(imm) + ((char)(imm)>0xF ? 96 : 48), \
- ((char)(imm)&0xF0) ? 113 : (char)(imm) + ((char)(imm)>0xE ? 97 : 49), \
- ((char)(imm)&0xF0) ? 114 : (char)(imm) + ((char)(imm)>0xD ? 98 : 50), \
- ((char)(imm)&0xF0) ? 115 : (char)(imm) + ((char)(imm)>0xC ? 99 : 51), \
- ((char)(imm)&0xF0) ? 116 : (char)(imm) + ((char)(imm)>0xB ? 100 : 52), \
- ((char)(imm)&0xF0) ? 117 : (char)(imm) + ((char)(imm)>0xA ? 101 : 53), \
- ((char)(imm)&0xF0) ? 118 : (char)(imm) + ((char)(imm)>0x9 ? 102 : 54), \
- ((char)(imm)&0xF0) ? 119 : (char)(imm) + ((char)(imm)>0x8 ? 103 : 55), \
- ((char)(imm)&0xF0) ? 120 : (char)(imm) + ((char)(imm)>0x7 ? 104 : 56), \
- ((char)(imm)&0xF0) ? 121 : (char)(imm) + ((char)(imm)>0x6 ? 105 : 57), \
- ((char)(imm)&0xF0) ? 122 : (char)(imm) + ((char)(imm)>0x5 ? 106 : 58), \
- ((char)(imm)&0xF0) ? 123 : (char)(imm) + ((char)(imm)>0x4 ? 107 : 59), \
- ((char)(imm)&0xF0) ? 124 : (char)(imm) + ((char)(imm)>0x3 ? 108 : 60), \
- ((char)(imm)&0xF0) ? 125 : (char)(imm) + ((char)(imm)>0x2 ? 109 : 61), \
- ((char)(imm)&0xF0) ? 126 : (char)(imm) + ((char)(imm)>0x1 ? 110 : 62), \
- ((char)(imm)&0xF0) ? 127 : (char)(imm) + ((char)(imm)>0x0 ? 111 : 63)); })
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_mov_epi16 (__m512i __W, __mmask32 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_selectw_512 ((__mmask32) __U,
- (__v32hi) __A,
- (__v32hi) __W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_mov_epi16 (__mmask32 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_selectw_512 ((__mmask32) __U,
- (__v32hi) __A,
- (__v32hi) _mm512_setzero_hi ());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_mov_epi8 (__m512i __W, __mmask64 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_selectb_512 ((__mmask64) __U,
- (__v64qi) __A,
- (__v64qi) __W);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_mov_epi8 (__mmask64 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_selectb_512 ((__mmask64) __U,
- (__v64qi) __A,
- (__v64qi) _mm512_setzero_hi ());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_set1_epi8 (__m512i __O, __mmask64 __M, char __A)
-{
- return (__m512i) __builtin_ia32_pbroadcastb512_gpr_mask (__A,
- (__v64qi) __O,
- __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_set1_epi8 (__mmask64 __M, char __A)
-{
- return (__m512i) __builtin_ia32_pbroadcastb512_gpr_mask (__A,
- (__v64qi)
- _mm512_setzero_qi(),
- __M);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_kunpackd (__mmask64 __A, __mmask64 __B)
-{
- return (__mmask64) __builtin_ia32_kunpckdi ((__mmask64) __A,
- (__mmask64) __B);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_kunpackw (__mmask32 __A, __mmask32 __B)
-{
- return (__mmask32) __builtin_ia32_kunpcksi ((__mmask32) __A,
- (__mmask32) __B);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_loadu_epi16 (__m512i __W, __mmask32 __U, void const *__P)
-{
- return (__m512i) __builtin_ia32_loaddquhi512_mask ((__v32hi *) __P,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_loadu_epi16 (__mmask32 __U, void const *__P)
-{
- return (__m512i) __builtin_ia32_loaddquhi512_mask ((__v32hi *) __P,
- (__v32hi)
- _mm512_setzero_hi (),
- (__mmask32) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_loadu_epi8 (__m512i __W, __mmask64 __U, void const *__P)
-{
- return (__m512i) __builtin_ia32_loaddquqi512_mask ((__v64qi *) __P,
- (__v64qi) __W,
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_loadu_epi8 (__mmask64 __U, void const *__P)
-{
- return (__m512i) __builtin_ia32_loaddquqi512_mask ((__v64qi *) __P,
- (__v64qi)
- _mm512_setzero_hi (),
- (__mmask64) __U);
-}
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm512_mask_storeu_epi16 (void *__P, __mmask32 __U, __m512i __A)
-{
- __builtin_ia32_storedquhi512_mask ((__v32hi *) __P,
- (__v32hi) __A,
- (__mmask32) __U);
-}
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm512_mask_storeu_epi8 (void *__P, __mmask64 __U, __m512i __A)
-{
- __builtin_ia32_storedquqi512_mask ((__v64qi *) __P,
- (__v64qi) __A,
- (__mmask64) __U);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_test_epi8_mask (__m512i __A, __m512i __B)
-{
- return (__mmask64) __builtin_ia32_ptestmb512 ((__v64qi) __A,
- (__v64qi) __B,
- (__mmask64) -1);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_mask_test_epi8_mask (__mmask64 __U, __m512i __A, __m512i __B)
-{
- return (__mmask64) __builtin_ia32_ptestmb512 ((__v64qi) __A,
- (__v64qi) __B, __U);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_test_epi16_mask (__m512i __A, __m512i __B)
-{
- return (__mmask32) __builtin_ia32_ptestmw512 ((__v32hi) __A,
- (__v32hi) __B,
- (__mmask32) -1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_mask_test_epi16_mask (__mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__mmask32) __builtin_ia32_ptestmw512 ((__v32hi) __A,
- (__v32hi) __B, __U);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_testn_epi8_mask (__m512i __A, __m512i __B)
-{
- return (__mmask64) __builtin_ia32_ptestnmb512 ((__v64qi) __A,
- (__v64qi) __B,
- (__mmask64) -1);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_mask_testn_epi8_mask (__mmask64 __U, __m512i __A, __m512i __B)
-{
- return (__mmask64) __builtin_ia32_ptestnmb512 ((__v64qi) __A,
- (__v64qi) __B, __U);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_testn_epi16_mask (__m512i __A, __m512i __B)
-{
- return (__mmask32) __builtin_ia32_ptestnmw512 ((__v32hi) __A,
- (__v32hi) __B,
- (__mmask32) -1);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_mask_testn_epi16_mask (__mmask32 __U, __m512i __A, __m512i __B)
-{
- return (__mmask32) __builtin_ia32_ptestnmw512 ((__v32hi) __A,
- (__v32hi) __B, __U);
-}
-
-static __inline__ __mmask64 __DEFAULT_FN_ATTRS
-_mm512_movepi8_mask (__m512i __A)
-{
- return (__mmask64) __builtin_ia32_cvtb2mask512 ((__v64qi) __A);
-}
-
-static __inline__ __mmask32 __DEFAULT_FN_ATTRS
-_mm512_movepi16_mask (__m512i __A)
-{
- return (__mmask32) __builtin_ia32_cvtw2mask512 ((__v32hi) __A);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_movm_epi8 (__mmask64 __A)
-{
- return (__m512i) __builtin_ia32_cvtmask2b512 (__A);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_movm_epi16 (__mmask32 __A)
-{
- return (__m512i) __builtin_ia32_cvtmask2w512 (__A);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_broadcastb_epi8 (__m128i __A)
-{
- return (__m512i)__builtin_shufflevector((__v16qi) __A,
- (__v16qi)_mm_undefined_si128(),
- 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_broadcastb_epi8 (__m512i __O, __mmask64 __M, __m128i __A)
-{
- return (__m512i)__builtin_ia32_selectb_512(__M,
- (__v64qi) _mm512_broadcastb_epi8(__A),
- (__v64qi) __O);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_broadcastb_epi8 (__mmask64 __M, __m128i __A)
-{
- return (__m512i)__builtin_ia32_selectb_512(__M,
- (__v64qi) _mm512_broadcastb_epi8(__A),
- (__v64qi) _mm512_setzero_si512());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_set1_epi16 (__m512i __O, __mmask32 __M, short __A)
-{
- return (__m512i) __builtin_ia32_pbroadcastw512_gpr_mask (__A,
- (__v32hi) __O,
- __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_set1_epi16 (__mmask32 __M, short __A)
-{
- return (__m512i) __builtin_ia32_pbroadcastw512_gpr_mask (__A,
- (__v32hi) _mm512_setzero_hi(),
- __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_broadcastw_epi16 (__m128i __A)
-{
- return (__m512i)__builtin_shufflevector((__v8hi) __A,
- (__v8hi)_mm_undefined_si128(),
- 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_broadcastw_epi16 (__m512i __O, __mmask32 __M, __m128i __A)
-{
- return (__m512i)__builtin_ia32_selectw_512(__M,
- (__v32hi) _mm512_broadcastw_epi16(__A),
- (__v32hi) __O);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_broadcastw_epi16 (__mmask32 __M, __m128i __A)
-{
- return (__m512i)__builtin_ia32_selectw_512(__M,
- (__v32hi) _mm512_broadcastw_epi16(__A),
- (__v32hi) _mm512_setzero_si512());
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_permutexvar_epi16 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_permvarhi512_mask ((__v32hi) __B,
- (__v32hi) __A,
- (__v32hi) _mm512_undefined_epi32 (),
- (__mmask32) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_permutexvar_epi16 (__mmask32 __M, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_permvarhi512_mask ((__v32hi) __B,
- (__v32hi) __A,
- (__v32hi) _mm512_setzero_hi(),
- (__mmask32) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_permutexvar_epi16 (__m512i __W, __mmask32 __M, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_permvarhi512_mask ((__v32hi) __B,
- (__v32hi) __A,
- (__v32hi) __W,
- (__mmask32) __M);
-}
-
-#define _mm512_alignr_epi8(A, B, N) __extension__ ({\
- (__m512i)__builtin_ia32_palignr512_mask((__v64qi)(__m512i)(A), \
- (__v64qi)(__m512i)(B), (int)(N), \
- (__v64qi)_mm512_undefined_pd(), \
- (__mmask64)-1); })
-
-#define _mm512_mask_alignr_epi8(W, U, A, B, N) __extension__({\
- (__m512i)__builtin_ia32_palignr512_mask((__v64qi)(__m512i)(A), \
- (__v64qi)(__m512i)(B), (int)(N), \
- (__v64qi)(__m512i)(W), \
- (__mmask64)(U)); })
-
-#define _mm512_maskz_alignr_epi8(U, A, B, N) __extension__({\
- (__m512i)__builtin_ia32_palignr512_mask((__v64qi)(__m512i)(A), \
- (__v64qi)(__m512i)(B), (int)(N), \
- (__v64qi)_mm512_setzero_si512(), \
- (__mmask64)(U)); })
-
-#define _mm512_dbsad_epu8(A, B, imm) __extension__ ({\
- (__m512i)__builtin_ia32_dbpsadbw512_mask((__v64qi)(__m512i)(A), \
- (__v64qi)(__m512i)(B), (int)(imm), \
- (__v32hi)_mm512_undefined_epi32(), \
- (__mmask32)-1); })
-
-#define _mm512_mask_dbsad_epu8(W, U, A, B, imm) ({\
- (__m512i)__builtin_ia32_dbpsadbw512_mask((__v64qi)(__m512i)(A), \
- (__v64qi)(__m512i)(B), (int)(imm), \
- (__v32hi)(__m512i)(W), \
- (__mmask32)(U)); })
-
-#define _mm512_maskz_dbsad_epu8(U, A, B, imm) ({\
- (__m512i)__builtin_ia32_dbpsadbw512_mask((__v64qi)(__m512i)(A), \
- (__v64qi)(__m512i)(B), (int)(imm), \
- (__v32hi)_mm512_setzero_hi(), \
- (__mmask32)(U)); })
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_sad_epu8 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_psadbw512 ((__v64qi) __A,
- (__v64qi) __B);
-}
-
-
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif
+++ /dev/null
-/*===------------- avx512cdintrin.h - AVX512CD intrinsics ------------------===
- *
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-#ifndef __IMMINTRIN_H
-#error "Never use <avx512cdintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __AVX512CDINTRIN_H
-#define __AVX512CDINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512cd")))
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_conflict_epi64 (__m512i __A)
-{
- return (__m512i) __builtin_ia32_vpconflictdi_512_mask ((__v8di) __A,
- (__v8di) _mm512_setzero_si512 (),
- (__mmask8) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_conflict_epi64 (__m512i __W, __mmask8 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_vpconflictdi_512_mask ((__v8di) __A,
- (__v8di) __W,
- (__mmask8) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_conflict_epi64 (__mmask8 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_vpconflictdi_512_mask ((__v8di) __A,
- (__v8di) _mm512_setzero_si512 (),
- (__mmask8) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_conflict_epi32 (__m512i __A)
-{
- return (__m512i) __builtin_ia32_vpconflictsi_512_mask ((__v16si) __A,
- (__v16si) _mm512_setzero_si512 (),
- (__mmask16) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_conflict_epi32 (__m512i __W, __mmask16 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_vpconflictsi_512_mask ((__v16si) __A,
- (__v16si) __W,
- (__mmask16) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_conflict_epi32 (__mmask16 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_vpconflictsi_512_mask ((__v16si) __A,
- (__v16si) _mm512_setzero_si512 (),
- (__mmask16) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_lzcnt_epi32 (__m512i __A)
-{
- return (__m512i) __builtin_ia32_vplzcntd_512_mask ((__v16si) __A,
- (__v16si) _mm512_setzero_si512 (),
- (__mmask16) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_lzcnt_epi32 (__m512i __W, __mmask16 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_vplzcntd_512_mask ((__v16si) __A,
- (__v16si) __W,
- (__mmask16) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_lzcnt_epi32 (__mmask16 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_vplzcntd_512_mask ((__v16si) __A,
- (__v16si) _mm512_setzero_si512 (),
- (__mmask16) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_lzcnt_epi64 (__m512i __A)
-{
- return (__m512i) __builtin_ia32_vplzcntq_512_mask ((__v8di) __A,
- (__v8di) _mm512_setzero_si512 (),
- (__mmask8) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_lzcnt_epi64 (__m512i __W, __mmask8 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_vplzcntq_512_mask ((__v8di) __A,
- (__v8di) __W,
- (__mmask8) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_lzcnt_epi64 (__mmask8 __U, __m512i __A)
-{
- return (__m512i) __builtin_ia32_vplzcntq_512_mask ((__v8di) __A,
- (__v8di) _mm512_setzero_si512 (),
- (__mmask8) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_broadcastmb_epi64 (__mmask8 __A)
-{
- return (__m512i) __builtin_ia32_broadcastmb512 (__A);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_broadcastmw_epi32 (__mmask16 __A)
-{
- return (__m512i) __builtin_ia32_broadcastmw512 (__A);
-}
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif
+++ /dev/null
-/*===------------- avx512ifmaintrin.h - IFMA intrinsics ------------------===
- *
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-#ifndef __IMMINTRIN_H
-#error "Never use <avx512ifmaintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __IFMAINTRIN_H
-#define __IFMAINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512ifma")))
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_madd52hi_epu64 (__m512i __X, __m512i __Y, __m512i __Z)
-{
- return (__m512i) __builtin_ia32_vpmadd52huq512_mask ((__v8di) __X,
- (__v8di) __Y,
- (__v8di) __Z,
- (__mmask8) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_madd52hi_epu64 (__m512i __W, __mmask8 __M, __m512i __X,
- __m512i __Y)
-{
- return (__m512i) __builtin_ia32_vpmadd52huq512_mask ((__v8di) __W,
- (__v8di) __X,
- (__v8di) __Y,
- (__mmask8) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_madd52hi_epu64 (__mmask8 __M, __m512i __X, __m512i __Y, __m512i __Z)
-{
- return (__m512i) __builtin_ia32_vpmadd52huq512_maskz ((__v8di) __X,
- (__v8di) __Y,
- (__v8di) __Z,
- (__mmask8) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_madd52lo_epu64 (__m512i __X, __m512i __Y, __m512i __Z)
-{
- return (__m512i) __builtin_ia32_vpmadd52luq512_mask ((__v8di) __X,
- (__v8di) __Y,
- (__v8di) __Z,
- (__mmask8) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_madd52lo_epu64 (__m512i __W, __mmask8 __M, __m512i __X,
- __m512i __Y)
-{
- return (__m512i) __builtin_ia32_vpmadd52luq512_mask ((__v8di) __W,
- (__v8di) __X,
- (__v8di) __Y,
- (__mmask8) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_madd52lo_epu64 (__mmask8 __M, __m512i __X, __m512i __Y, __m512i __Z)
-{
- return (__m512i) __builtin_ia32_vpmadd52luq512_maskz ((__v8di) __X,
- (__v8di) __Y,
- (__v8di) __Z,
- (__mmask8) __M);
-}
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif
+++ /dev/null
-/*===------------- avx512ifmavlintrin.h - IFMA intrinsics ------------------===
- *
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-#ifndef __IMMINTRIN_H
-#error "Never use <avx512ifmavlintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __IFMAVLINTRIN_H
-#define __IFMAVLINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512ifma,avx512vl")))
-
-
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_madd52hi_epu64 (__m128i __X, __m128i __Y, __m128i __Z)
-{
- return (__m128i) __builtin_ia32_vpmadd52huq128_mask ((__v2di) __X,
- (__v2di) __Y,
- (__v2di) __Z,
- (__mmask8) -1);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_madd52hi_epu64 (__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y)
-{
- return (__m128i) __builtin_ia32_vpmadd52huq128_mask ((__v2di) __W,
- (__v2di) __X,
- (__v2di) __Y,
- (__mmask8) __M);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_madd52hi_epu64 (__mmask8 __M, __m128i __X, __m128i __Y, __m128i __Z)
-{
- return (__m128i) __builtin_ia32_vpmadd52huq128_maskz ((__v2di) __X,
- (__v2di) __Y,
- (__v2di) __Z,
- (__mmask8) __M);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_madd52hi_epu64 (__m256i __X, __m256i __Y, __m256i __Z)
-{
- return (__m256i) __builtin_ia32_vpmadd52huq256_mask ((__v4di) __X,
- (__v4di) __Y,
- (__v4di) __Z,
- (__mmask8) -1);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_madd52hi_epu64 (__m256i __W, __mmask8 __M, __m256i __X,
- __m256i __Y)
-{
- return (__m256i) __builtin_ia32_vpmadd52huq256_mask ((__v4di) __W,
- (__v4di) __X,
- (__v4di) __Y,
- (__mmask8) __M);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_madd52hi_epu64 (__mmask8 __M, __m256i __X, __m256i __Y, __m256i __Z)
-{
- return (__m256i) __builtin_ia32_vpmadd52huq256_maskz ((__v4di) __X,
- (__v4di) __Y,
- (__v4di) __Z,
- (__mmask8) __M);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_madd52lo_epu64 (__m128i __X, __m128i __Y, __m128i __Z)
-{
- return (__m128i) __builtin_ia32_vpmadd52luq128_mask ((__v2di) __X,
- (__v2di) __Y,
- (__v2di) __Z,
- (__mmask8) -1);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_madd52lo_epu64 (__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y)
-{
- return (__m128i) __builtin_ia32_vpmadd52luq128_mask ((__v2di) __W,
- (__v2di) __X,
- (__v2di) __Y,
- (__mmask8) __M);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_madd52lo_epu64 (__mmask8 __M, __m128i __X, __m128i __Y, __m128i __Z)
-{
- return (__m128i) __builtin_ia32_vpmadd52luq128_maskz ((__v2di) __X,
- (__v2di) __Y,
- (__v2di) __Z,
- (__mmask8) __M);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_madd52lo_epu64 (__m256i __X, __m256i __Y, __m256i __Z)
-{
- return (__m256i) __builtin_ia32_vpmadd52luq256_mask ((__v4di) __X,
- (__v4di) __Y,
- (__v4di) __Z,
- (__mmask8) -1);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_madd52lo_epu64 (__m256i __W, __mmask8 __M, __m256i __X,
- __m256i __Y)
-{
- return (__m256i) __builtin_ia32_vpmadd52luq256_mask ((__v4di) __W,
- (__v4di) __X,
- (__v4di) __Y,
- (__mmask8) __M);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_madd52lo_epu64 (__mmask8 __M, __m256i __X, __m256i __Y, __m256i __Z)
-{
- return (__m256i) __builtin_ia32_vpmadd52luq256_maskz ((__v4di) __X,
- (__v4di) __Y,
- (__v4di) __Z,
- (__mmask8) __M);
-}
-
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif
+++ /dev/null
-/*===------------- avx512pfintrin.h - PF intrinsics ------------------===
- *
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-#ifndef __IMMINTRIN_H
-#error "Never use <avx512pfintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __AVX512PFINTRIN_H
-#define __AVX512PFINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512pf")))
-
-#define _mm512_mask_prefetch_i32gather_pd(index, mask, addr, scale, hint) __extension__ ({\
- __builtin_ia32_gatherpfdpd((__mmask8)(mask), (__v8si)(__m256i)(index), \
- (long long const *)(addr), (int)(scale), \
- (int)(hint)); })
-
-#define _mm512_prefetch_i32gather_pd(index, addr, scale, hint) __extension__ ({\
- __builtin_ia32_gatherpfdpd((__mmask8) -1, (__v8si)(__m256i)(index), \
- (long long const *)(addr), (int)(scale), \
- (int)(hint)); })
-
-#define _mm512_mask_prefetch_i32gather_ps(index, mask, addr, scale, hint) ({\
- __builtin_ia32_gatherpfdps((__mmask16)(mask), \
- (__v16si)(__m512i)(index), (int const *)(addr), \
- (int)(scale), (int)(hint)); })
-
-#define _mm512_prefetch_i32gather_ps(index, addr, scale, hint) ({\
- __builtin_ia32_gatherpfdps((__mmask16) -1, \
- (__v16si)(__m512i)(index), (int const *)(addr), \
- (int)(scale), (int)(hint)); })
-
-#define _mm512_mask_prefetch_i64gather_pd(index, mask, addr, scale, hint) __extension__ ({\
- __builtin_ia32_gatherpfqpd((__mmask8)(mask), (__v8di)(__m512i)(index), \
- (long long const *)(addr), (int)(scale), \
- (int)(hint)); })
-
-#define _mm512_prefetch_i64gather_pd(index, addr, scale, hint) __extension__ ({\
- __builtin_ia32_gatherpfqpd((__mmask8) -1, (__v8di)(__m512i)(index), \
- (long long const *)(addr), (int)(scale), \
- (int)(hint)); })
-
-#define _mm512_mask_prefetch_i64gather_ps(index, mask, addr, scale, hint) ({\
- __builtin_ia32_gatherpfqps((__mmask8)(mask), (__v8di)(__m512i)(index), \
- (int const *)(addr), (int)(scale), (int)(hint)); })
-
-#define _mm512_prefetch_i64gather_ps(index, addr, scale, hint) ({\
- __builtin_ia32_gatherpfqps((__mmask8) -1, (__v8di)(__m512i)(index), \
- (int const *)(addr), (int)(scale), (int)(hint)); })
-
-#define _mm512_prefetch_i32scatter_pd(addr, index, scale, hint) __extension__ ({\
- __builtin_ia32_scatterpfdpd((__mmask8)-1, (__v8si)(__m256i)(index), \
- (long long *)(addr), (int)(scale), \
- (int)(hint)); })
-
-#define _mm512_mask_prefetch_i32scatter_pd(addr, mask, index, scale, hint) __extension__ ({\
- __builtin_ia32_scatterpfdpd((__mmask8)(mask), (__v8si)(__m256i)(index), \
- (long long *)(addr), (int)(scale), \
- (int)(hint)); })
-
-#define _mm512_prefetch_i32scatter_ps(addr, index, scale, hint) __extension__ ({\
- __builtin_ia32_scatterpfdps((__mmask16)-1, (__v16si)(__m512i)(index), \
- (int *)(addr), (int)(scale), (int)(hint)); })
-
-#define _mm512_mask_prefetch_i32scatter_ps(addr, mask, index, scale, hint) __extension__ ({\
- __builtin_ia32_scatterpfdps((__mmask16)(mask), \
- (__v16si)(__m512i)(index), (int *)(addr), \
- (int)(scale), (int)(hint)); })
-
-#define _mm512_prefetch_i64scatter_pd(addr, index, scale, hint) __extension__ ({\
- __builtin_ia32_scatterpfqpd((__mmask8)-1, (__v8di)(__m512i)(index), \
- (long long *)(addr), (int)(scale), \
- (int)(hint)); })
-
-#define _mm512_mask_prefetch_i64scatter_pd(addr, mask, index, scale, hint) __extension__ ({\
- __builtin_ia32_scatterpfqpd((__mmask8)(mask), (__v8di)(__m512i)(index), \
- (long long *)(addr), (int)(scale), \
- (int)(hint)); })
-
-#define _mm512_prefetch_i64scatter_ps(addr, index, scale, hint) __extension__ ({\
- __builtin_ia32_scatterpfqps((__mmask8)-1, (__v8di)(__m512i)(index), \
- (int *)(addr), (int)(scale), (int)(hint)); })
-
-#define _mm512_mask_prefetch_i64scatter_ps(addr, mask, index, scale, hint) __extension__ ({\
- __builtin_ia32_scatterpfqps((__mmask8)(mask), (__v8di)(__m512i)(index), \
- (int *)(addr), (int)(scale), (int)(hint)); })
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif
+++ /dev/null
-/*===------------- avx512vbmiintrin.h - VBMI intrinsics ------------------===
- *
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-#ifndef __IMMINTRIN_H
-#error "Never use <avx512vbmiintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __VBMIINTRIN_H
-#define __VBMIINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512vbmi")))
-
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask2_permutex2var_epi8 (__m512i __A, __m512i __I,
- __mmask64 __U, __m512i __B)
-{
- return (__m512i) __builtin_ia32_vpermi2varqi512_mask ((__v64qi) __A,
- (__v64qi) __I
- /* idx */ ,
- (__v64qi) __B,
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_permutex2var_epi8 (__m512i __A, __m512i __I, __m512i __B)
-{
- return (__m512i) __builtin_ia32_vpermt2varqi512_mask ((__v64qi) __I
- /* idx */ ,
- (__v64qi) __A,
- (__v64qi) __B,
- (__mmask64) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_permutex2var_epi8 (__m512i __A, __mmask64 __U,
- __m512i __I, __m512i __B)
-{
- return (__m512i) __builtin_ia32_vpermt2varqi512_mask ((__v64qi) __I
- /* idx */ ,
- (__v64qi) __A,
- (__v64qi) __B,
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_permutex2var_epi8 (__mmask64 __U, __m512i __A,
- __m512i __I, __m512i __B)
-{
- return (__m512i) __builtin_ia32_vpermt2varqi512_maskz ((__v64qi) __I
- /* idx */ ,
- (__v64qi) __A,
- (__v64qi) __B,
- (__mmask64) __U);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_permutexvar_epi8 (__m512i __A, __m512i __B)
-{
- return (__m512i) __builtin_ia32_permvarqi512_mask ((__v64qi) __B,
- (__v64qi) __A,
- (__v64qi) _mm512_undefined_epi32 (),
- (__mmask64) -1);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_permutexvar_epi8 (__mmask64 __M, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_permvarqi512_mask ((__v64qi) __B,
- (__v64qi) __A,
- (__v64qi) _mm512_setzero_si512(),
- (__mmask64) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_permutexvar_epi8 (__m512i __W, __mmask64 __M, __m512i __A,
- __m512i __B)
-{
- return (__m512i) __builtin_ia32_permvarqi512_mask ((__v64qi) __B,
- (__v64qi) __A,
- (__v64qi) __W,
- (__mmask64) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_multishift_epi64_epi8 (__m512i __W, __mmask64 __M, __m512i __X, __m512i __Y)
-{
- return (__m512i) __builtin_ia32_vpmultishiftqb512_mask ((__v64qi) __X,
- (__v64qi) __Y,
- (__v64qi) __W,
- (__mmask64) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_multishift_epi64_epi8 (__mmask64 __M, __m512i __X, __m512i __Y)
-{
- return (__m512i) __builtin_ia32_vpmultishiftqb512_mask ((__v64qi) __X,
- (__v64qi) __Y,
- (__v64qi) _mm512_setzero_si512 (),
- (__mmask64) __M);
-}
-
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_multishift_epi64_epi8 (__m512i __X, __m512i __Y)
-{
- return (__m512i) __builtin_ia32_vpmultishiftqb512_mask ((__v64qi) __X,
- (__v64qi) __Y,
- (__v64qi) _mm512_undefined_epi32 (),
- (__mmask64) -1);
-}
-
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif
+++ /dev/null
-/*===------------- avx512vbmivlintrin.h - VBMI intrinsics ------------------===
- *
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-#ifndef __IMMINTRIN_H
-#error "Never use <avx512vbmivlintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __VBMIVLINTRIN_H
-#define __VBMIVLINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512vbmi,avx512vl")))
-
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask2_permutex2var_epi8 (__m128i __A, __m128i __I, __mmask16 __U,
- __m128i __B)
-{
- return (__m128i) __builtin_ia32_vpermi2varqi128_mask ((__v16qi) __A,
- (__v16qi) __I
- /* idx */ ,
- (__v16qi) __B,
- (__mmask16)
- __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask2_permutex2var_epi8 (__m256i __A, __m256i __I,
- __mmask32 __U, __m256i __B)
-{
- return (__m256i) __builtin_ia32_vpermi2varqi256_mask ((__v32qi) __A,
- (__v32qi) __I
- /* idx */ ,
- (__v32qi) __B,
- (__mmask32)
- __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_permutex2var_epi8 (__m128i __A, __m128i __I, __m128i __B)
-{
- return (__m128i) __builtin_ia32_vpermt2varqi128_mask ((__v16qi) __I
- /* idx */ ,
- (__v16qi) __A,
- (__v16qi) __B,
- (__mmask16) -
- 1);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_permutex2var_epi8 (__m128i __A, __mmask16 __U, __m128i __I,
- __m128i __B)
-{
- return (__m128i) __builtin_ia32_vpermt2varqi128_mask ((__v16qi) __I
- /* idx */ ,
- (__v16qi) __A,
- (__v16qi) __B,
- (__mmask16)
- __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_permutex2var_epi8 (__mmask16 __U, __m128i __A, __m128i __I,
- __m128i __B)
-{
- return (__m128i) __builtin_ia32_vpermt2varqi128_maskz ((__v16qi) __I
- /* idx */ ,
- (__v16qi) __A,
- (__v16qi) __B,
- (__mmask16)
- __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_permutex2var_epi8 (__m256i __A, __m256i __I, __m256i __B)
-{
- return (__m256i) __builtin_ia32_vpermt2varqi256_mask ((__v32qi) __I
- /* idx */ ,
- (__v32qi) __A,
- (__v32qi) __B,
- (__mmask32) -
- 1);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_permutex2var_epi8 (__m256i __A, __mmask32 __U,
- __m256i __I, __m256i __B)
-{
- return (__m256i) __builtin_ia32_vpermt2varqi256_mask ((__v32qi) __I
- /* idx */ ,
- (__v32qi) __A,
- (__v32qi) __B,
- (__mmask32)
- __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_permutex2var_epi8 (__mmask32 __U, __m256i __A,
- __m256i __I, __m256i __B)
-{
- return (__m256i) __builtin_ia32_vpermt2varqi256_maskz ((__v32qi) __I
- /* idx */ ,
- (__v32qi) __A,
- (__v32qi) __B,
- (__mmask32)
- __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_permutexvar_epi8 (__m128i __A, __m128i __B)
-{
- return (__m128i) __builtin_ia32_permvarqi128_mask ((__v16qi) __B,
- (__v16qi) __A,
- (__v16qi) _mm_undefined_si128 (),
- (__mmask16) -1);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_permutexvar_epi8 (__mmask16 __M, __m128i __A, __m128i __B)
-{
- return (__m128i) __builtin_ia32_permvarqi128_mask ((__v16qi) __B,
- (__v16qi) __A,
- (__v16qi) _mm_setzero_si128 (),
- (__mmask16) __M);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_permutexvar_epi8 (__m128i __W, __mmask16 __M, __m128i __A,
- __m128i __B)
-{
- return (__m128i) __builtin_ia32_permvarqi128_mask ((__v16qi) __B,
- (__v16qi) __A,
- (__v16qi) __W,
- (__mmask16) __M);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_permutexvar_epi8 (__m256i __A, __m256i __B)
-{
- return (__m256i) __builtin_ia32_permvarqi256_mask ((__v32qi) __B,
- (__v32qi) __A,
- (__v32qi) _mm256_undefined_si256 (),
- (__mmask32) -1);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_permutexvar_epi8 (__mmask32 __M, __m256i __A,
- __m256i __B)
-{
- return (__m256i) __builtin_ia32_permvarqi256_mask ((__v32qi) __B,
- (__v32qi) __A,
- (__v32qi) _mm256_setzero_si256 (),
- (__mmask32) __M);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_permutexvar_epi8 (__m256i __W, __mmask32 __M, __m256i __A,
- __m256i __B)
-{
- return (__m256i) __builtin_ia32_permvarqi256_mask ((__v32qi) __B,
- (__v32qi) __A,
- (__v32qi) __W,
- (__mmask32) __M);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_multishift_epi64_epi8 (__m128i __W, __mmask16 __M, __m128i __X, __m128i __Y)
-{
- return (__m128i) __builtin_ia32_vpmultishiftqb128_mask ((__v16qi) __X,
- (__v16qi) __Y,
- (__v16qi) __W,
- (__mmask16) __M);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_multishift_epi64_epi8 (__mmask16 __M, __m128i __X, __m128i __Y)
-{
- return (__m128i) __builtin_ia32_vpmultishiftqb128_mask ((__v16qi) __X,
- (__v16qi) __Y,
- (__v16qi)
- _mm_setzero_si128 (),
- (__mmask16) __M);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_multishift_epi64_epi8 (__m128i __X, __m128i __Y)
-{
- return (__m128i) __builtin_ia32_vpmultishiftqb128_mask ((__v16qi) __X,
- (__v16qi) __Y,
- (__v16qi)
- _mm_undefined_si128 (),
- (__mmask16) -1);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_multishift_epi64_epi8 (__m256i __W, __mmask32 __M, __m256i __X, __m256i __Y)
-{
- return (__m256i) __builtin_ia32_vpmultishiftqb256_mask ((__v32qi) __X,
- (__v32qi) __Y,
- (__v32qi) __W,
- (__mmask32) __M);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_multishift_epi64_epi8 (__mmask32 __M, __m256i __X, __m256i __Y)
-{
- return (__m256i) __builtin_ia32_vpmultishiftqb256_mask ((__v32qi) __X,
- (__v32qi) __Y,
- (__v32qi)
- _mm256_setzero_si256 (),
- (__mmask32) __M);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_multishift_epi64_epi8 (__m256i __X, __m256i __Y)
-{
- return (__m256i) __builtin_ia32_vpmultishiftqb256_mask ((__v32qi) __X,
- (__v32qi) __Y,
- (__v32qi)
- _mm256_undefined_si256 (),
- (__mmask32) -1);
-}
-
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif
+++ /dev/null
-/*===---- avx512vlcdintrin.h - AVX512VL and AVX512CD intrinsics ---------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-#ifndef __IMMINTRIN_H
-#error "Never use <avx512vlcdintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __AVX512VLCDINTRIN_H
-#define __AVX512VLCDINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx512vl,avx512cd")))
-
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_broadcastmb_epi64 (__mmask8 __A)
-{
- return (__m128i) __builtin_ia32_broadcastmb128 (__A);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_broadcastmb_epi64 (__mmask8 __A)
-{
- return (__m256i) __builtin_ia32_broadcastmb256 (__A);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_broadcastmw_epi32 (__mmask16 __A)
-{
- return (__m128i) __builtin_ia32_broadcastmw128 (__A);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_broadcastmw_epi32 (__mmask16 __A)
-{
- return (__m256i) __builtin_ia32_broadcastmw256 (__A);
-}
-
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_conflict_epi64 (__m128i __A)
-{
- return (__m128i) __builtin_ia32_vpconflictdi_128_mask ((__v2di) __A,
- (__v2di) _mm_undefined_si128 (),
- (__mmask8) -1);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_conflict_epi64 (__m128i __W, __mmask8 __U, __m128i __A)
-{
- return (__m128i) __builtin_ia32_vpconflictdi_128_mask ((__v2di) __A,
- (__v2di) __W,
- (__mmask8) __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_conflict_epi64 (__mmask8 __U, __m128i __A)
-{
- return (__m128i) __builtin_ia32_vpconflictdi_128_mask ((__v2di) __A,
- (__v2di)
- _mm_setzero_di (),
- (__mmask8) __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_conflict_epi64 (__m256i __A)
-{
- return (__m256i) __builtin_ia32_vpconflictdi_256_mask ((__v4di) __A,
- (__v4di) _mm256_undefined_si256 (),
- (__mmask8) -1);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_conflict_epi64 (__m256i __W, __mmask8 __U, __m256i __A)
-{
- return (__m256i) __builtin_ia32_vpconflictdi_256_mask ((__v4di) __A,
- (__v4di) __W,
- (__mmask8) __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_conflict_epi64 (__mmask8 __U, __m256i __A)
-{
- return (__m256i) __builtin_ia32_vpconflictdi_256_mask ((__v4di) __A,
- (__v4di) _mm256_setzero_si256 (),
- (__mmask8) __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_conflict_epi32 (__m128i __A)
-{
- return (__m128i) __builtin_ia32_vpconflictsi_128_mask ((__v4si) __A,
- (__v4si) _mm_undefined_si128 (),
- (__mmask8) -1);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_conflict_epi32 (__m128i __W, __mmask8 __U, __m128i __A)
-{
- return (__m128i) __builtin_ia32_vpconflictsi_128_mask ((__v4si) __A,
- (__v4si) __W,
- (__mmask8) __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_conflict_epi32 (__mmask8 __U, __m128i __A)
-{
- return (__m128i) __builtin_ia32_vpconflictsi_128_mask ((__v4si) __A,
- (__v4si) _mm_setzero_si128 (),
- (__mmask8) __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_conflict_epi32 (__m256i __A)
-{
- return (__m256i) __builtin_ia32_vpconflictsi_256_mask ((__v8si) __A,
- (__v8si) _mm256_undefined_si256 (),
- (__mmask8) -1);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_conflict_epi32 (__m256i __W, __mmask8 __U, __m256i __A)
-{
- return (__m256i) __builtin_ia32_vpconflictsi_256_mask ((__v8si) __A,
- (__v8si) __W,
- (__mmask8) __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_conflict_epi32 (__mmask8 __U, __m256i __A)
-{
- return (__m256i) __builtin_ia32_vpconflictsi_256_mask ((__v8si) __A,
- (__v8si)
- _mm256_setzero_si256 (),
- (__mmask8) __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_lzcnt_epi32 (__m128i __A)
-{
- return (__m128i) __builtin_ia32_vplzcntd_128_mask ((__v4si) __A,
- (__v4si)
- _mm_setzero_si128 (),
- (__mmask8) -1);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_lzcnt_epi32 (__m128i __W, __mmask8 __U, __m128i __A)
-{
- return (__m128i) __builtin_ia32_vplzcntd_128_mask ((__v4si) __A,
- (__v4si) __W,
- (__mmask8) __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_lzcnt_epi32 (__mmask8 __U, __m128i __A)
-{
- return (__m128i) __builtin_ia32_vplzcntd_128_mask ((__v4si) __A,
- (__v4si)
- _mm_setzero_si128 (),
- (__mmask8) __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_lzcnt_epi32 (__m256i __A)
-{
- return (__m256i) __builtin_ia32_vplzcntd_256_mask ((__v8si) __A,
- (__v8si)
- _mm256_setzero_si256 (),
- (__mmask8) -1);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_lzcnt_epi32 (__m256i __W, __mmask8 __U, __m256i __A)
-{
- return (__m256i) __builtin_ia32_vplzcntd_256_mask ((__v8si) __A,
- (__v8si) __W,
- (__mmask8) __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_lzcnt_epi32 (__mmask8 __U, __m256i __A)
-{
- return (__m256i) __builtin_ia32_vplzcntd_256_mask ((__v8si) __A,
- (__v8si)
- _mm256_setzero_si256 (),
- (__mmask8) __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_lzcnt_epi64 (__m128i __A)
-{
- return (__m128i) __builtin_ia32_vplzcntq_128_mask ((__v2di) __A,
- (__v2di)
- _mm_setzero_di (),
- (__mmask8) -1);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_lzcnt_epi64 (__m128i __W, __mmask8 __U, __m128i __A)
-{
- return (__m128i) __builtin_ia32_vplzcntq_128_mask ((__v2di) __A,
- (__v2di) __W,
- (__mmask8) __U);
-}
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_lzcnt_epi64 (__mmask8 __U, __m128i __A)
-{
- return (__m128i) __builtin_ia32_vplzcntq_128_mask ((__v2di) __A,
- (__v2di)
- _mm_setzero_di (),
- (__mmask8) __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_lzcnt_epi64 (__m256i __A)
-{
- return (__m256i) __builtin_ia32_vplzcntq_256_mask ((__v4di) __A,
- (__v4di)
- _mm256_setzero_si256 (),
- (__mmask8) -1);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_lzcnt_epi64 (__m256i __W, __mmask8 __U, __m256i __A)
-{
- return (__m256i) __builtin_ia32_vplzcntq_256_mask ((__v4di) __A,
- (__v4di) __W,
- (__mmask8) __U);
-}
-
-static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_lzcnt_epi64 (__mmask8 __U, __m256i __A)
-{
- return (__m256i) __builtin_ia32_vplzcntq_256_mask ((__v4di) __A,
- (__v4di)
- _mm256_setzero_si256 (),
- (__mmask8) __U);
-}
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif /* __AVX512VLCDINTRIN_H */
+++ /dev/null
-/*===---- clflushoptintrin.h - CLFLUSHOPT intrinsic ------------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __IMMINTRIN_H
-#error "Never use <clflushoptintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __CLFLUSHOPTINTRIN_H
-#define __CLFLUSHOPTINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("clflushopt")))
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_clflushopt(char * __m) {
- __builtin_ia32_clflushopt(__m);
-}
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif
+++ /dev/null
-/*===----------------------- clzerointrin.h - CLZERO ----------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-#ifndef __X86INTRIN_H
-#error "Never use <clzerointrin.h> directly; include <x86intrin.h> instead."
-#endif
-
-#ifndef _CLZEROINTRIN_H
-#define _CLZEROINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS \
- __attribute__((__always_inline__, __nodebug__, __target__("clzero")))
-
-/// \brief Loads the cache line address and zero's out the cacheline
-///
-/// \headerfile <clzerointrin.h>
-///
-/// This intrinsic corresponds to the <c> CLZERO </c> instruction.
-///
-/// \param __line
-/// A pointer to a cacheline which needs to be zeroed out.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_clzero (void * __line)
-{
- __builtin_ia32_clzero ((void *)__line);
-}
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif /* _CLZEROINTRIN_H */
+++ /dev/null
-/*===---- f16cintrin.h - F16C intrinsics -----------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#if !defined __X86INTRIN_H && !defined __EMMINTRIN_H && !defined __IMMINTRIN_H
-#error "Never use <f16cintrin.h> directly; include <emmintrin.h> instead."
-#endif
-
-#ifndef __F16CINTRIN_H
-#define __F16CINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS \
- __attribute__((__always_inline__, __nodebug__, __target__("f16c")))
-
-/// \brief Converts a 16-bit half-precision float value into a 32-bit float
-/// value.
-///
-/// \headerfile <x86intrin.h>
-///
-/// This intrinsic corresponds to the <c> VCVTPH2PS </c> instruction.
-///
-/// \param __a
-/// A 16-bit half-precision float value.
-/// \returns The converted 32-bit float value.
-static __inline float __DEFAULT_FN_ATTRS
-_cvtsh_ss(unsigned short __a)
-{
- __v8hi v = {(short)__a, 0, 0, 0, 0, 0, 0, 0};
- __v4sf r = __builtin_ia32_vcvtph2ps(v);
- return r[0];
-}
-
-/// \brief Converts a 32-bit single-precision float value to a 16-bit
-/// half-precision float value.
-///
-/// \headerfile <x86intrin.h>
-///
-/// \code
-/// unsigned short _cvtss_sh(float a, const int imm);
-/// \endcode
-///
-/// This intrinsic corresponds to the <c> VCVTPS2PH </c> instruction.
-///
-/// \param a
-/// A 32-bit single-precision float value to be converted to a 16-bit
-/// half-precision float value.
-/// \param imm
-/// An immediate value controlling rounding using bits [2:0]: \n
-/// 000: Nearest \n
-/// 001: Down \n
-/// 010: Up \n
-/// 011: Truncate \n
-/// 1XX: Use MXCSR.RC for rounding
-/// \returns The converted 16-bit half-precision float value.
-#define _cvtss_sh(a, imm) __extension__ ({ \
- (unsigned short)(((__v8hi)__builtin_ia32_vcvtps2ph((__v4sf){a, 0, 0, 0}, \
- (imm)))[0]); })
-
-/// \brief Converts a 128-bit vector containing 32-bit float values into a
-/// 128-bit vector containing 16-bit half-precision float values.
-///
-/// \headerfile <x86intrin.h>
-///
-/// \code
-/// __m128i _mm_cvtps_ph(__m128 a, const int imm);
-/// \endcode
-///
-/// This intrinsic corresponds to the <c> VCVTPS2PH </c> instruction.
-///
-/// \param a
-/// A 128-bit vector containing 32-bit float values.
-/// \param imm
-/// An immediate value controlling rounding using bits [2:0]: \n
-/// 000: Nearest \n
-/// 001: Down \n
-/// 010: Up \n
-/// 011: Truncate \n
-/// 1XX: Use MXCSR.RC for rounding
-/// \returns A 128-bit vector containing converted 16-bit half-precision float
-/// values. The lower 64 bits are used to store the converted 16-bit
-/// half-precision floating-point values.
-#define _mm_cvtps_ph(a, imm) __extension__ ({ \
- (__m128i)__builtin_ia32_vcvtps2ph((__v4sf)(__m128)(a), (imm)); })
-
-/// \brief Converts a 128-bit vector containing 16-bit half-precision float
-/// values into a 128-bit vector containing 32-bit float values.
-///
-/// \headerfile <x86intrin.h>
-///
-/// This intrinsic corresponds to the <c> VCVTPH2PS </c> instruction.
-///
-/// \param __a
-/// A 128-bit vector containing 16-bit half-precision float values. The lower
-/// 64 bits are used in the conversion.
-/// \returns A 128-bit vector of [4 x float] containing converted float values.
-static __inline __m128 __DEFAULT_FN_ATTRS
-_mm_cvtph_ps(__m128i __a)
-{
- return (__m128)__builtin_ia32_vcvtph2ps((__v8hi)__a);
-}
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif /* __F16CINTRIN_H */
+++ /dev/null
-/*===---- fma4intrin.h - FMA4 intrinsics -----------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __X86INTRIN_H
-#error "Never use <fma4intrin.h> directly; include <x86intrin.h> instead."
-#endif
-
-#ifndef __FMA4INTRIN_H
-#define __FMA4INTRIN_H
-
-#include <pmmintrin.h>
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("fma4")))
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_macc_ps(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfmaddps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_macc_pd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfmaddpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_macc_ss(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfmaddss((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_macc_sd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfmaddsd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_msub_ps(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfmsubps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_msub_pd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfmsubpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_msub_ss(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfmsubss((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_msub_sd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfmsubsd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_nmacc_ps(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfnmaddps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_nmacc_pd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfnmaddpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_nmacc_ss(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfnmaddss((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_nmacc_sd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfnmaddsd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_nmsub_ps(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfnmsubps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_nmsub_pd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfnmsubpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_nmsub_ss(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfnmsubss((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_nmsub_sd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfnmsubsd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_maddsub_ps(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfmaddsubps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_maddsub_pd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfmaddsubpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_msubadd_ps(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfmsubaddps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_msubadd_pd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfmsubaddpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_macc_ps(__m256 __A, __m256 __B, __m256 __C)
-{
- return (__m256)__builtin_ia32_vfmaddps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_macc_pd(__m256d __A, __m256d __B, __m256d __C)
-{
- return (__m256d)__builtin_ia32_vfmaddpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_msub_ps(__m256 __A, __m256 __B, __m256 __C)
-{
- return (__m256)__builtin_ia32_vfmsubps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_msub_pd(__m256d __A, __m256d __B, __m256d __C)
-{
- return (__m256d)__builtin_ia32_vfmsubpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_nmacc_ps(__m256 __A, __m256 __B, __m256 __C)
-{
- return (__m256)__builtin_ia32_vfnmaddps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_nmacc_pd(__m256d __A, __m256d __B, __m256d __C)
-{
- return (__m256d)__builtin_ia32_vfnmaddpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_nmsub_ps(__m256 __A, __m256 __B, __m256 __C)
-{
- return (__m256)__builtin_ia32_vfnmsubps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_nmsub_pd(__m256d __A, __m256d __B, __m256d __C)
-{
- return (__m256d)__builtin_ia32_vfnmsubpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_maddsub_ps(__m256 __A, __m256 __B, __m256 __C)
-{
- return (__m256)__builtin_ia32_vfmaddsubps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_maddsub_pd(__m256d __A, __m256d __B, __m256d __C)
-{
- return (__m256d)__builtin_ia32_vfmaddsubpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_msubadd_ps(__m256 __A, __m256 __B, __m256 __C)
-{
- return (__m256)__builtin_ia32_vfmsubaddps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_msubadd_pd(__m256d __A, __m256d __B, __m256d __C)
-{
- return (__m256d)__builtin_ia32_vfmsubaddpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
-}
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif /* __FMA4INTRIN_H */
+++ /dev/null
-/*===---- fma4intrin.h - FMA4 intrinsics -----------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __IMMINTRIN_H
-#error "Never use <fmaintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __FMAINTRIN_H
-#define __FMAINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("fma")))
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_fmadd_ps(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfmaddps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_fmadd_pd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfmaddpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_fmadd_ss(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfmaddss((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_fmadd_sd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfmaddsd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_fmsub_ps(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfmsubps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_fmsub_pd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfmsubpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_fmsub_ss(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfmsubss((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_fmsub_sd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfmsubsd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_fnmadd_ps(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfnmaddps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_fnmadd_pd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfnmaddpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_fnmadd_ss(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfnmaddss((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_fnmadd_sd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfnmaddsd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_fnmsub_ps(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfnmsubps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_fnmsub_pd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfnmsubpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_fnmsub_ss(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfnmsubss((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_fnmsub_sd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfnmsubsd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_fmaddsub_ps(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfmaddsubps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_fmaddsub_pd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfmaddsubpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_fmsubadd_ps(__m128 __A, __m128 __B, __m128 __C)
-{
- return (__m128)__builtin_ia32_vfmsubaddps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
-}
-
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_fmsubadd_pd(__m128d __A, __m128d __B, __m128d __C)
-{
- return (__m128d)__builtin_ia32_vfmsubaddpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_fmadd_ps(__m256 __A, __m256 __B, __m256 __C)
-{
- return (__m256)__builtin_ia32_vfmaddps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_fmadd_pd(__m256d __A, __m256d __B, __m256d __C)
-{
- return (__m256d)__builtin_ia32_vfmaddpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_fmsub_ps(__m256 __A, __m256 __B, __m256 __C)
-{
- return (__m256)__builtin_ia32_vfmsubps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_fmsub_pd(__m256d __A, __m256d __B, __m256d __C)
-{
- return (__m256d)__builtin_ia32_vfmsubpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_fnmadd_ps(__m256 __A, __m256 __B, __m256 __C)
-{
- return (__m256)__builtin_ia32_vfnmaddps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_fnmadd_pd(__m256d __A, __m256d __B, __m256d __C)
-{
- return (__m256d)__builtin_ia32_vfnmaddpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_fnmsub_ps(__m256 __A, __m256 __B, __m256 __C)
-{
- return (__m256)__builtin_ia32_vfnmsubps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_fnmsub_pd(__m256d __A, __m256d __B, __m256d __C)
-{
- return (__m256d)__builtin_ia32_vfnmsubpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_fmaddsub_ps(__m256 __A, __m256 __B, __m256 __C)
-{
- return (__m256)__builtin_ia32_vfmaddsubps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_fmaddsub_pd(__m256d __A, __m256d __B, __m256d __C)
-{
- return (__m256d)__builtin_ia32_vfmaddsubpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
-}
-
-static __inline__ __m256 __DEFAULT_FN_ATTRS
-_mm256_fmsubadd_ps(__m256 __A, __m256 __B, __m256 __C)
-{
- return (__m256)__builtin_ia32_vfmsubaddps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
-}
-
-static __inline__ __m256d __DEFAULT_FN_ATTRS
-_mm256_fmsubadd_pd(__m256d __A, __m256d __B, __m256d __C)
-{
- return (__m256d)__builtin_ia32_vfmsubaddpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
-}
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif /* __FMAINTRIN_H */
+++ /dev/null
-/* ===-------- ia32intrin.h ---------------------------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __X86INTRIN_H
-#error "Never use <ia32intrin.h> directly; include <x86intrin.h> instead."
-#endif
-
-#ifndef __IA32INTRIN_H
-#define __IA32INTRIN_H
-
-#ifdef __x86_64__
-static __inline__ unsigned long long __attribute__((__always_inline__, __nodebug__))
-__readeflags(void)
-{
- return __builtin_ia32_readeflags_u64();
-}
-
-static __inline__ void __attribute__((__always_inline__, __nodebug__))
-__writeeflags(unsigned long long __f)
-{
- __builtin_ia32_writeeflags_u64(__f);
-}
-
-#else /* !__x86_64__ */
-static __inline__ unsigned int __attribute__((__always_inline__, __nodebug__))
-__readeflags(void)
-{
- return __builtin_ia32_readeflags_u32();
-}
-
-static __inline__ void __attribute__((__always_inline__, __nodebug__))
-__writeeflags(unsigned int __f)
-{
- __builtin_ia32_writeeflags_u32(__f);
-}
-#endif /* !__x86_64__ */
-
-static __inline__ unsigned long long __attribute__((__always_inline__, __nodebug__))
-__rdpmc(int __A) {
- return __builtin_ia32_rdpmc(__A);
-}
-
-/* __rdtscp */
-static __inline__ unsigned long long __attribute__((__always_inline__, __nodebug__))
-__rdtscp(unsigned int *__A) {
- return __builtin_ia32_rdtscp(__A);
-}
-
-#define _rdtsc() __rdtsc()
-
-#define _rdpmc(A) __rdpmc(A)
-
-#endif /* __IA32INTRIN_H */
+++ /dev/null
-/*===---- immintrin.h - Intel intrinsics -----------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __IMMINTRIN_H
-#define __IMMINTRIN_H
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__MMX__)
-#include <mmintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__SSE__)
-#include <xmmintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__SSE2__)
-#include <emmintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__SSE3__)
-#include <pmmintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__SSSE3__)
-#include <tmmintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || \
- (defined(__SSE4_2__) || defined(__SSE4_1__))
-#include <smmintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || \
- (defined(__AES__) || defined(__PCLMUL__))
-#include <wmmintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__CLFLUSHOPT__)
-#include <clflushoptintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__AVX__)
-#include <avxintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__AVX2__)
-#include <avx2intrin.h>
-
-/* The 256-bit versions of functions in f16cintrin.h.
- Intel documents these as being in immintrin.h, and
- they depend on typedefs from avxintrin.h. */
-
-/// \brief Converts a 256-bit vector of [8 x float] into a 128-bit vector
-/// containing 16-bit half-precision float values.
-///
-/// \headerfile <x86intrin.h>
-///
-/// \code
-/// __m128i _mm256_cvtps_ph(__m256 a, const int imm);
-/// \endcode
-///
-/// This intrinsic corresponds to the <c> VCVTPS2PH </c> instruction.
-///
-/// \param a
-/// A 256-bit vector containing 32-bit single-precision float values to be
-/// converted to 16-bit half-precision float values.
-/// \param imm
-/// An immediate value controlling rounding using bits [2:0]: \n
-/// 000: Nearest \n
-/// 001: Down \n
-/// 010: Up \n
-/// 011: Truncate \n
-/// 1XX: Use MXCSR.RC for rounding
-/// \returns A 128-bit vector containing the converted 16-bit half-precision
-/// float values.
-#define _mm256_cvtps_ph(a, imm) __extension__ ({ \
- (__m128i)__builtin_ia32_vcvtps2ph256((__v8sf)(__m256)(a), (imm)); })
-
-/// \brief Converts a 128-bit vector containing 16-bit half-precision float
-/// values into a 256-bit vector of [8 x float].
-///
-/// \headerfile <x86intrin.h>
-///
-/// This intrinsic corresponds to the <c> VCVTPH2PS </c> instruction.
-///
-/// \param __a
-/// A 128-bit vector containing 16-bit half-precision float values to be
-/// converted to 32-bit single-precision float values.
-/// \returns A vector of [8 x float] containing the converted 32-bit
-/// single-precision float values.
-static __inline __m256 __attribute__((__always_inline__, __nodebug__, __target__("f16c")))
-_mm256_cvtph_ps(__m128i __a)
-{
- return (__m256)__builtin_ia32_vcvtph2ps256((__v8hi)__a);
-}
-#endif /* __AVX2__ */
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__BMI__)
-#include <bmiintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__BMI2__)
-#include <bmi2intrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__LZCNT__)
-#include <lzcntintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__FMA__)
-#include <fmaintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__AVX512F__)
-#include <avx512fintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__AVX512VL__)
-#include <avx512vlintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__AVX512BW__)
-#include <avx512bwintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__AVX512CD__)
-#include <avx512cdintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__AVX512VPOPCNTDQ__)
-#include <avx512vpopcntdqintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__AVX512DQ__)
-#include <avx512dqintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || \
- (defined(__AVX512VL__) && defined(__AVX512BW__))
-#include <avx512vlbwintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || \
- (defined(__AVX512VL__) && defined(__AVX512CD__))
-#include <avx512vlcdintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || \
- (defined(__AVX512VL__) && defined(__AVX512DQ__))
-#include <avx512vldqintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__AVX512ER__)
-#include <avx512erintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__AVX512IFMA__)
-#include <avx512ifmaintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || \
- (defined(__AVX512IFMA__) && defined(__AVX512VL__))
-#include <avx512ifmavlintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__AVX512VBMI__)
-#include <avx512vbmiintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || \
- (defined(__AVX512VBMI__) && defined(__AVX512VL__))
-#include <avx512vbmivlintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__AVX512PF__)
-#include <avx512pfintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__PKU__)
-#include <pkuintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__RDRND__)
-static __inline__ int __attribute__((__always_inline__, __nodebug__, __target__("rdrnd")))
-_rdrand16_step(unsigned short *__p)
-{
- return __builtin_ia32_rdrand16_step(__p);
-}
-
-static __inline__ int __attribute__((__always_inline__, __nodebug__, __target__("rdrnd")))
-_rdrand32_step(unsigned int *__p)
-{
- return __builtin_ia32_rdrand32_step(__p);
-}
-
-#ifdef __x86_64__
-static __inline__ int __attribute__((__always_inline__, __nodebug__, __target__("rdrnd")))
-_rdrand64_step(unsigned long long *__p)
-{
- return __builtin_ia32_rdrand64_step(__p);
-}
-#endif
-#endif /* __RDRND__ */
-
-/* __bit_scan_forward */
-static __inline__ int __attribute__((__always_inline__, __nodebug__))
-_bit_scan_forward(int __A) {
- return __builtin_ctz(__A);
-}
-
-/* __bit_scan_reverse */
-static __inline__ int __attribute__((__always_inline__, __nodebug__))
-_bit_scan_reverse(int __A) {
- return 31 - __builtin_clz(__A);
-}
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__FSGSBASE__)
-#ifdef __x86_64__
-static __inline__ unsigned int __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
-_readfsbase_u32(void)
-{
- return __builtin_ia32_rdfsbase32();
-}
-
-static __inline__ unsigned long long __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
-_readfsbase_u64(void)
-{
- return __builtin_ia32_rdfsbase64();
-}
-
-static __inline__ unsigned int __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
-_readgsbase_u32(void)
-{
- return __builtin_ia32_rdgsbase32();
-}
-
-static __inline__ unsigned long long __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
-_readgsbase_u64(void)
-{
- return __builtin_ia32_rdgsbase64();
-}
-
-static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
-_writefsbase_u32(unsigned int __V)
-{
- return __builtin_ia32_wrfsbase32(__V);
-}
-
-static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
-_writefsbase_u64(unsigned long long __V)
-{
- return __builtin_ia32_wrfsbase64(__V);
-}
-
-static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
-_writegsbase_u32(unsigned int __V)
-{
- return __builtin_ia32_wrgsbase32(__V);
-}
-
-static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("fsgsbase")))
-_writegsbase_u64(unsigned long long __V)
-{
- return __builtin_ia32_wrgsbase64(__V);
-}
-
-#endif
-#endif /* __FSGSBASE__ */
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__RTM__)
-#include <rtmintrin.h>
-#include <xtestintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__SHA__)
-#include <shaintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__FXSR__)
-#include <fxsrintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__XSAVE__)
-#include <xsaveintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__XSAVEOPT__)
-#include <xsaveoptintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__XSAVEC__)
-#include <xsavecintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__XSAVES__)
-#include <xsavesintrin.h>
-#endif
-
-/* Some intrinsics inside adxintrin.h are available only on processors with ADX,
- * whereas others are also available at all times. */
-#include <adxintrin.h>
-
-#endif /* __IMMINTRIN_H */
+++ /dev/null
-/* ===-------- intrin.h ---------------------------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-/* Only include this if we're compiling for the windows platform. */
-#ifndef _MSC_VER
-#include_next <intrin.h>
-#else
-
-#ifndef __INTRIN_H
-#define __INTRIN_H
-
-/* First include the standard intrinsics. */
-#if defined(__i386__) || defined(__x86_64__)
-#include <x86intrin.h>
-#endif
-
-#if defined(__arm__)
-#include <armintr.h>
-#endif
-
-/* For the definition of jmp_buf. */
-#if __STDC_HOSTED__
-#include <setjmp.h>
-#endif
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__))
-
-#ifdef __cplusplus
-extern "C" {
-#endif
-
-#if defined(__MMX__)
-/* And the random ones that aren't in those files. */
-__m64 _m_from_float(float);
-float _m_to_float(__m64);
-#endif
-
-/* Other assorted instruction intrinsics. */
-void __addfsbyte(unsigned long, unsigned char);
-void __addfsdword(unsigned long, unsigned long);
-void __addfsword(unsigned long, unsigned short);
-void __code_seg(const char *);
-static __inline__
-void __cpuid(int[4], int);
-static __inline__
-void __cpuidex(int[4], int, int);
-static __inline__
-__int64 __emul(int, int);
-static __inline__
-unsigned __int64 __emulu(unsigned int, unsigned int);
-unsigned int __getcallerseflags(void);
-static __inline__
-void __halt(void);
-unsigned char __inbyte(unsigned short);
-void __inbytestring(unsigned short, unsigned char *, unsigned long);
-void __incfsbyte(unsigned long);
-void __incfsdword(unsigned long);
-void __incfsword(unsigned long);
-unsigned long __indword(unsigned short);
-void __indwordstring(unsigned short, unsigned long *, unsigned long);
-void __invlpg(void *);
-unsigned short __inword(unsigned short);
-void __inwordstring(unsigned short, unsigned short *, unsigned long);
-void __lidt(void *);
-unsigned __int64 __ll_lshift(unsigned __int64, int);
-__int64 __ll_rshift(__int64, int);
-unsigned int __lzcnt(unsigned int);
-unsigned short __lzcnt16(unsigned short);
-static __inline__
-void __movsb(unsigned char *, unsigned char const *, size_t);
-static __inline__
-void __movsd(unsigned long *, unsigned long const *, size_t);
-static __inline__
-void __movsw(unsigned short *, unsigned short const *, size_t);
-static __inline__
-void __nop(void);
-void __nvreg_restore_fence(void);
-void __nvreg_save_fence(void);
-void __outbyte(unsigned short, unsigned char);
-void __outbytestring(unsigned short, unsigned char *, unsigned long);
-void __outdword(unsigned short, unsigned long);
-void __outdwordstring(unsigned short, unsigned long *, unsigned long);
-void __outword(unsigned short, unsigned short);
-void __outwordstring(unsigned short, unsigned short *, unsigned long);
-unsigned long __readcr0(void);
-unsigned long __readcr2(void);
-static __inline__
-unsigned long __readcr3(void);
-unsigned long __readcr4(void);
-unsigned long __readcr8(void);
-unsigned int __readdr(unsigned int);
-#ifdef __i386__
-static __inline__
-unsigned char __readfsbyte(unsigned long);
-static __inline__
-unsigned __int64 __readfsqword(unsigned long);
-static __inline__
-unsigned short __readfsword(unsigned long);
-#endif
-static __inline__
-unsigned __int64 __readmsr(unsigned long);
-unsigned __int64 __readpmc(unsigned long);
-unsigned long __segmentlimit(unsigned long);
-void __sidt(void *);
-static __inline__
-void __stosb(unsigned char *, unsigned char, size_t);
-static __inline__
-void __stosd(unsigned long *, unsigned long, size_t);
-static __inline__
-void __stosw(unsigned short *, unsigned short, size_t);
-void __svm_clgi(void);
-void __svm_invlpga(void *, int);
-void __svm_skinit(int);
-void __svm_stgi(void);
-void __svm_vmload(size_t);
-void __svm_vmrun(size_t);
-void __svm_vmsave(size_t);
-unsigned __int64 __ull_rshift(unsigned __int64, int);
-void __vmx_off(void);
-void __vmx_vmptrst(unsigned __int64 *);
-void __wbinvd(void);
-void __writecr0(unsigned int);
-static __inline__
-void __writecr3(unsigned int);
-void __writecr4(unsigned int);
-void __writecr8(unsigned int);
-void __writedr(unsigned int, unsigned int);
-void __writefsbyte(unsigned long, unsigned char);
-void __writefsdword(unsigned long, unsigned long);
-void __writefsqword(unsigned long, unsigned __int64);
-void __writefsword(unsigned long, unsigned short);
-void __writemsr(unsigned long, unsigned __int64);
-static __inline__
-void *_AddressOfReturnAddress(void);
-static __inline__
-unsigned char _BitScanForward(unsigned long *_Index, unsigned long _Mask);
-static __inline__
-unsigned char _BitScanReverse(unsigned long *_Index, unsigned long _Mask);
-static __inline__
-unsigned char _bittest(long const *, long);
-static __inline__
-unsigned char _bittestandcomplement(long *, long);
-static __inline__
-unsigned char _bittestandreset(long *, long);
-static __inline__
-unsigned char _bittestandset(long *, long);
-void __cdecl _disable(void);
-void __cdecl _enable(void);
-long _InterlockedAddLargeStatistic(__int64 volatile *_Addend, long _Value);
-unsigned char _interlockedbittestandreset(long volatile *, long);
-unsigned char _interlockedbittestandset(long volatile *, long);
-long _InterlockedCompareExchange_HLEAcquire(long volatile *, long, long);
-long _InterlockedCompareExchange_HLERelease(long volatile *, long, long);
-__int64 _InterlockedcompareExchange64_HLEAcquire(__int64 volatile *, __int64,
- __int64);
-__int64 _InterlockedCompareExchange64_HLERelease(__int64 volatile *, __int64,
- __int64);
-void *_InterlockedCompareExchangePointer_HLEAcquire(void *volatile *, void *,
- void *);
-void *_InterlockedCompareExchangePointer_HLERelease(void *volatile *, void *,
- void *);
-long _InterlockedExchangeAdd_HLEAcquire(long volatile *, long);
-long _InterlockedExchangeAdd_HLERelease(long volatile *, long);
-__int64 _InterlockedExchangeAdd64_HLEAcquire(__int64 volatile *, __int64);
-__int64 _InterlockedExchangeAdd64_HLERelease(__int64 volatile *, __int64);
-void __cdecl _invpcid(unsigned int, void *);
-static __inline__ void
-__attribute__((__deprecated__("use other intrinsics or C++11 atomics instead")))
-_ReadBarrier(void);
-static __inline__ void
-__attribute__((__deprecated__("use other intrinsics or C++11 atomics instead")))
-_ReadWriteBarrier(void);
-unsigned int _rorx_u32(unsigned int, const unsigned int);
-int _sarx_i32(int, unsigned int);
-#if __STDC_HOSTED__
-int __cdecl _setjmp(jmp_buf);
-#endif
-unsigned int _shlx_u32(unsigned int, unsigned int);
-unsigned int _shrx_u32(unsigned int, unsigned int);
-void _Store_HLERelease(long volatile *, long);
-void _Store64_HLERelease(__int64 volatile *, __int64);
-void _StorePointer_HLERelease(void *volatile *, void *);
-static __inline__ void
-__attribute__((__deprecated__("use other intrinsics or C++11 atomics instead")))
-_WriteBarrier(void);
-unsigned __int32 xbegin(void);
-void _xend(void);
-static __inline__
-#define _XCR_XFEATURE_ENABLED_MASK 0
-unsigned __int64 __cdecl _xgetbv(unsigned int);
-void __cdecl _xsetbv(unsigned int, unsigned __int64);
-
-/* These additional intrinsics are turned on in x64/amd64/x86_64 mode. */
-#ifdef __x86_64__
-void __addgsbyte(unsigned long, unsigned char);
-void __addgsdword(unsigned long, unsigned long);
-void __addgsqword(unsigned long, unsigned __int64);
-void __addgsword(unsigned long, unsigned short);
-static __inline__
-void __faststorefence(void);
-void __incgsbyte(unsigned long);
-void __incgsdword(unsigned long);
-void __incgsqword(unsigned long);
-void __incgsword(unsigned long);
-unsigned __int64 __lzcnt64(unsigned __int64);
-static __inline__
-void __movsq(unsigned long long *, unsigned long long const *, size_t);
-static __inline__
-unsigned char __readgsbyte(unsigned long);
-static __inline__
-unsigned long __readgsdword(unsigned long);
-static __inline__
-unsigned __int64 __readgsqword(unsigned long);
-unsigned short __readgsword(unsigned long);
-unsigned __int64 __shiftleft128(unsigned __int64 _LowPart,
- unsigned __int64 _HighPart,
- unsigned char _Shift);
-unsigned __int64 __shiftright128(unsigned __int64 _LowPart,
- unsigned __int64 _HighPart,
- unsigned char _Shift);
-static __inline__
-void __stosq(unsigned __int64 *, unsigned __int64, size_t);
-unsigned char __vmx_on(unsigned __int64 *);
-unsigned char __vmx_vmclear(unsigned __int64 *);
-unsigned char __vmx_vmlaunch(void);
-unsigned char __vmx_vmptrld(unsigned __int64 *);
-unsigned char __vmx_vmread(size_t, size_t *);
-unsigned char __vmx_vmresume(void);
-unsigned char __vmx_vmwrite(size_t, size_t);
-void __writegsbyte(unsigned long, unsigned char);
-void __writegsdword(unsigned long, unsigned long);
-void __writegsqword(unsigned long, unsigned __int64);
-void __writegsword(unsigned long, unsigned short);
-static __inline__
-unsigned char _BitScanForward64(unsigned long *_Index, unsigned __int64 _Mask);
-static __inline__
-unsigned char _BitScanReverse64(unsigned long *_Index, unsigned __int64 _Mask);
-static __inline__
-unsigned char _bittest64(__int64 const *, __int64);
-static __inline__
-unsigned char _bittestandcomplement64(__int64 *, __int64);
-static __inline__
-unsigned char _bittestandreset64(__int64 *, __int64);
-static __inline__
-unsigned char _bittestandset64(__int64 *, __int64);
-long _InterlockedAnd_np(long volatile *_Value, long _Mask);
-short _InterlockedAnd16_np(short volatile *_Value, short _Mask);
-__int64 _InterlockedAnd64_np(__int64 volatile *_Value, __int64 _Mask);
-char _InterlockedAnd8_np(char volatile *_Value, char _Mask);
-unsigned char _interlockedbittestandreset64(__int64 volatile *, __int64);
-static __inline__
-unsigned char _interlockedbittestandset64(__int64 volatile *, __int64);
-long _InterlockedCompareExchange_np(long volatile *_Destination, long _Exchange,
- long _Comparand);
-unsigned char _InterlockedCompareExchange128(__int64 volatile *_Destination,
- __int64 _ExchangeHigh,
- __int64 _ExchangeLow,
- __int64 *_CompareandResult);
-unsigned char _InterlockedCompareExchange128_np(__int64 volatile *_Destination,
- __int64 _ExchangeHigh,
- __int64 _ExchangeLow,
- __int64 *_ComparandResult);
-short _InterlockedCompareExchange16_np(short volatile *_Destination,
- short _Exchange, short _Comparand);
-__int64 _InterlockedCompareExchange64_HLEAcquire(__int64 volatile *, __int64,
- __int64);
-__int64 _InterlockedCompareExchange64_HLERelease(__int64 volatile *, __int64,
- __int64);
-__int64 _InterlockedCompareExchange64_np(__int64 volatile *_Destination,
- __int64 _Exchange, __int64 _Comparand);
-void *_InterlockedCompareExchangePointer_np(void *volatile *_Destination,
- void *_Exchange, void *_Comparand);
-long _InterlockedOr_np(long volatile *_Value, long _Mask);
-short _InterlockedOr16_np(short volatile *_Value, short _Mask);
-__int64 _InterlockedOr64_np(__int64 volatile *_Value, __int64 _Mask);
-char _InterlockedOr8_np(char volatile *_Value, char _Mask);
-long _InterlockedXor_np(long volatile *_Value, long _Mask);
-short _InterlockedXor16_np(short volatile *_Value, short _Mask);
-__int64 _InterlockedXor64_np(__int64 volatile *_Value, __int64 _Mask);
-char _InterlockedXor8_np(char volatile *_Value, char _Mask);
-unsigned __int64 _rorx_u64(unsigned __int64, const unsigned int);
-__int64 _sarx_i64(__int64, unsigned int);
-unsigned __int64 _shlx_u64(unsigned __int64, unsigned int);
-unsigned __int64 _shrx_u64(unsigned __int64, unsigned int);
-static __inline__
-__int64 __mulh(__int64, __int64);
-static __inline__
-unsigned __int64 __umulh(unsigned __int64, unsigned __int64);
-static __inline__
-__int64 _mul128(__int64, __int64, __int64*);
-static __inline__
-unsigned __int64 _umul128(unsigned __int64,
- unsigned __int64,
- unsigned __int64*);
-
-#endif /* __x86_64__ */
-
-#if defined(__x86_64__) || defined(__arm__)
-
-static __inline__
-__int64 _InterlockedDecrement64(__int64 volatile *_Addend);
-static __inline__
-__int64 _InterlockedExchange64(__int64 volatile *_Target, __int64 _Value);
-static __inline__
-__int64 _InterlockedExchangeAdd64(__int64 volatile *_Addend, __int64 _Value);
-static __inline__
-__int64 _InterlockedExchangeSub64(__int64 volatile *_Subend, __int64 _Value);
-static __inline__
-__int64 _InterlockedIncrement64(__int64 volatile *_Addend);
-static __inline__
-__int64 _InterlockedOr64(__int64 volatile *_Value, __int64 _Mask);
-static __inline__
-__int64 _InterlockedXor64(__int64 volatile *_Value, __int64 _Mask);
-static __inline__
-__int64 _InterlockedAnd64(__int64 volatile *_Value, __int64 _Mask);
-
-#endif
-
-/*----------------------------------------------------------------------------*\
-|* Bit Counting and Testing
-\*----------------------------------------------------------------------------*/
-static __inline__ unsigned char __DEFAULT_FN_ATTRS
-_bittest(long const *_BitBase, long _BitPos) {
- return (*_BitBase >> _BitPos) & 1;
-}
-static __inline__ unsigned char __DEFAULT_FN_ATTRS
-_bittestandcomplement(long *_BitBase, long _BitPos) {
- unsigned char _Res = (*_BitBase >> _BitPos) & 1;
- *_BitBase = *_BitBase ^ (1 << _BitPos);
- return _Res;
-}
-static __inline__ unsigned char __DEFAULT_FN_ATTRS
-_bittestandreset(long *_BitBase, long _BitPos) {
- unsigned char _Res = (*_BitBase >> _BitPos) & 1;
- *_BitBase = *_BitBase & ~(1 << _BitPos);
- return _Res;
-}
-static __inline__ unsigned char __DEFAULT_FN_ATTRS
-_bittestandset(long *_BitBase, long _BitPos) {
- unsigned char _Res = (*_BitBase >> _BitPos) & 1;
- *_BitBase = *_BitBase | (1 << _BitPos);
- return _Res;
-}
-#if defined(__arm__) || defined(__aarch64__)
-static __inline__ unsigned char __DEFAULT_FN_ATTRS
-_interlockedbittestandset_acq(long volatile *_BitBase, long _BitPos) {
- long _PrevVal = __atomic_fetch_or(_BitBase, 1l << _BitPos, __ATOMIC_ACQUIRE);
- return (_PrevVal >> _BitPos) & 1;
-}
-static __inline__ unsigned char __DEFAULT_FN_ATTRS
-_interlockedbittestandset_nf(long volatile *_BitBase, long _BitPos) {
- long _PrevVal = __atomic_fetch_or(_BitBase, 1l << _BitPos, __ATOMIC_RELAXED);
- return (_PrevVal >> _BitPos) & 1;
-}
-static __inline__ unsigned char __DEFAULT_FN_ATTRS
-_interlockedbittestandset_rel(long volatile *_BitBase, long _BitPos) {
- long _PrevVal = __atomic_fetch_or(_BitBase, 1l << _BitPos, __ATOMIC_RELEASE);
- return (_PrevVal >> _BitPos) & 1;
-}
-#endif
-#ifdef __x86_64__
-static __inline__ unsigned char __DEFAULT_FN_ATTRS
-_bittest64(__int64 const *_BitBase, __int64 _BitPos) {
- return (*_BitBase >> _BitPos) & 1;
-}
-static __inline__ unsigned char __DEFAULT_FN_ATTRS
-_bittestandcomplement64(__int64 *_BitBase, __int64 _BitPos) {
- unsigned char _Res = (*_BitBase >> _BitPos) & 1;
- *_BitBase = *_BitBase ^ (1ll << _BitPos);
- return _Res;
-}
-static __inline__ unsigned char __DEFAULT_FN_ATTRS
-_bittestandreset64(__int64 *_BitBase, __int64 _BitPos) {
- unsigned char _Res = (*_BitBase >> _BitPos) & 1;
- *_BitBase = *_BitBase & ~(1ll << _BitPos);
- return _Res;
-}
-static __inline__ unsigned char __DEFAULT_FN_ATTRS
-_bittestandset64(__int64 *_BitBase, __int64 _BitPos) {
- unsigned char _Res = (*_BitBase >> _BitPos) & 1;
- *_BitBase = *_BitBase | (1ll << _BitPos);
- return _Res;
-}
-static __inline__ unsigned char __DEFAULT_FN_ATTRS
-_interlockedbittestandset64(__int64 volatile *_BitBase, __int64 _BitPos) {
- long long _PrevVal =
- __atomic_fetch_or(_BitBase, 1ll << _BitPos, __ATOMIC_SEQ_CST);
- return (_PrevVal >> _BitPos) & 1;
-}
-#endif
-/*----------------------------------------------------------------------------*\
-|* Interlocked Exchange Add
-\*----------------------------------------------------------------------------*/
-#if defined(__arm__) || defined(__aarch64__)
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd8_acq(char volatile *_Addend, char _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_ACQUIRE);
-}
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd8_nf(char volatile *_Addend, char _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELAXED);
-}
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd8_rel(char volatile *_Addend, char _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELAXED);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd16_acq(short volatile *_Addend, short _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_ACQUIRE);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd16_nf(short volatile *_Addend, short _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELAXED);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd16_rel(short volatile *_Addend, short _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELEASE);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd_acq(long volatile *_Addend, long _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_ACQUIRE);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd_nf(long volatile *_Addend, long _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELAXED);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd_rel(long volatile *_Addend, long _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELEASE);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd64_acq(__int64 volatile *_Addend, __int64 _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_ACQUIRE);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd64_nf(__int64 volatile *_Addend, __int64 _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELAXED);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd64_rel(__int64 volatile *_Addend, __int64 _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELEASE);
-}
-#endif
-/*----------------------------------------------------------------------------*\
-|* Interlocked Increment
-\*----------------------------------------------------------------------------*/
-#if defined(__arm__) || defined(__aarch64__)
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedIncrement16_acq(short volatile *_Value) {
- return __atomic_add_fetch(_Value, 1, __ATOMIC_ACQUIRE);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedIncrement16_nf(short volatile *_Value) {
- return __atomic_add_fetch(_Value, 1, __ATOMIC_RELAXED);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedIncrement16_rel(short volatile *_Value) {
- return __atomic_add_fetch(_Value, 1, __ATOMIC_RELEASE);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedIncrement_acq(long volatile *_Value) {
- return __atomic_add_fetch(_Value, 1, __ATOMIC_ACQUIRE);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedIncrement_nf(long volatile *_Value) {
- return __atomic_add_fetch(_Value, 1, __ATOMIC_RELAXED);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedIncrement_rel(long volatile *_Value) {
- return __atomic_add_fetch(_Value, 1, __ATOMIC_RELEASE);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedIncrement64_acq(__int64 volatile *_Value) {
- return __atomic_add_fetch(_Value, 1, __ATOMIC_ACQUIRE);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedIncrement64_nf(__int64 volatile *_Value) {
- return __atomic_add_fetch(_Value, 1, __ATOMIC_RELAXED);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedIncrement64_rel(__int64 volatile *_Value) {
- return __atomic_add_fetch(_Value, 1, __ATOMIC_RELEASE);
-}
-#endif
-/*----------------------------------------------------------------------------*\
-|* Interlocked Decrement
-\*----------------------------------------------------------------------------*/
-#if defined(__arm__) || defined(__aarch64__)
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedDecrement16_acq(short volatile *_Value) {
- return __atomic_sub_fetch(_Value, 1, __ATOMIC_ACQUIRE);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedDecrement16_nf(short volatile *_Value) {
- return __atomic_sub_fetch(_Value, 1, __ATOMIC_RELAXED);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedDecrement16_rel(short volatile *_Value) {
- return __atomic_sub_fetch(_Value, 1, __ATOMIC_RELEASE);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedDecrement_acq(long volatile *_Value) {
- return __atomic_sub_fetch(_Value, 1, __ATOMIC_ACQUIRE);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedDecrement_nf(long volatile *_Value) {
- return __atomic_sub_fetch(_Value, 1, __ATOMIC_RELAXED);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedDecrement_rel(long volatile *_Value) {
- return __atomic_sub_fetch(_Value, 1, __ATOMIC_RELEASE);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedDecrement64_acq(__int64 volatile *_Value) {
- return __atomic_sub_fetch(_Value, 1, __ATOMIC_ACQUIRE);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedDecrement64_nf(__int64 volatile *_Value) {
- return __atomic_sub_fetch(_Value, 1, __ATOMIC_RELAXED);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedDecrement64_rel(__int64 volatile *_Value) {
- return __atomic_sub_fetch(_Value, 1, __ATOMIC_RELEASE);
-}
-#endif
-/*----------------------------------------------------------------------------*\
-|* Interlocked And
-\*----------------------------------------------------------------------------*/
-#if defined(__arm__) || defined(__aarch64__)
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedAnd8_acq(char volatile *_Value, char _Mask) {
- return __atomic_fetch_and(_Value, _Mask, __ATOMIC_ACQUIRE);
-}
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedAnd8_nf(char volatile *_Value, char _Mask) {
- return __atomic_fetch_and(_Value, _Mask, __ATOMIC_RELAXED);
-}
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedAnd8_rel(char volatile *_Value, char _Mask) {
- return __atomic_fetch_and(_Value, _Mask, __ATOMIC_RELEASE);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedAnd16_acq(short volatile *_Value, short _Mask) {
- return __atomic_fetch_and(_Value, _Mask, __ATOMIC_ACQUIRE);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedAnd16_nf(short volatile *_Value, short _Mask) {
- return __atomic_fetch_and(_Value, _Mask, __ATOMIC_RELAXED);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedAnd16_rel(short volatile *_Value, short _Mask) {
- return __atomic_fetch_and(_Value, _Mask, __ATOMIC_RELEASE);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedAnd_acq(long volatile *_Value, long _Mask) {
- return __atomic_fetch_and(_Value, _Mask, __ATOMIC_ACQUIRE);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedAnd_nf(long volatile *_Value, long _Mask) {
- return __atomic_fetch_and(_Value, _Mask, __ATOMIC_RELAXED);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedAnd_rel(long volatile *_Value, long _Mask) {
- return __atomic_fetch_and(_Value, _Mask, __ATOMIC_RELEASE);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedAnd64_acq(__int64 volatile *_Value, __int64 _Mask) {
- return __atomic_fetch_and(_Value, _Mask, __ATOMIC_ACQUIRE);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedAnd64_nf(__int64 volatile *_Value, __int64 _Mask) {
- return __atomic_fetch_and(_Value, _Mask, __ATOMIC_RELAXED);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedAnd64_rel(__int64 volatile *_Value, __int64 _Mask) {
- return __atomic_fetch_and(_Value, _Mask, __ATOMIC_RELEASE);
-}
-#endif
-/*----------------------------------------------------------------------------*\
-|* Interlocked Or
-\*----------------------------------------------------------------------------*/
-#if defined(__arm__) || defined(__aarch64__)
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedOr8_acq(char volatile *_Value, char _Mask) {
- return __atomic_fetch_or(_Value, _Mask, __ATOMIC_ACQUIRE);
-}
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedOr8_nf(char volatile *_Value, char _Mask) {
- return __atomic_fetch_or(_Value, _Mask, __ATOMIC_RELAXED);
-}
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedOr8_rel(char volatile *_Value, char _Mask) {
- return __atomic_fetch_or(_Value, _Mask, __ATOMIC_RELEASE);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedOr16_acq(short volatile *_Value, short _Mask) {
- return __atomic_fetch_or(_Value, _Mask, __ATOMIC_ACQUIRE);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedOr16_nf(short volatile *_Value, short _Mask) {
- return __atomic_fetch_or(_Value, _Mask, __ATOMIC_RELAXED);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedOr16_rel(short volatile *_Value, short _Mask) {
- return __atomic_fetch_or(_Value, _Mask, __ATOMIC_RELEASE);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedOr_acq(long volatile *_Value, long _Mask) {
- return __atomic_fetch_or(_Value, _Mask, __ATOMIC_ACQUIRE);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedOr_nf(long volatile *_Value, long _Mask) {
- return __atomic_fetch_or(_Value, _Mask, __ATOMIC_RELAXED);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedOr_rel(long volatile *_Value, long _Mask) {
- return __atomic_fetch_or(_Value, _Mask, __ATOMIC_RELEASE);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedOr64_acq(__int64 volatile *_Value, __int64 _Mask) {
- return __atomic_fetch_or(_Value, _Mask, __ATOMIC_ACQUIRE);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedOr64_nf(__int64 volatile *_Value, __int64 _Mask) {
- return __atomic_fetch_or(_Value, _Mask, __ATOMIC_RELAXED);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedOr64_rel(__int64 volatile *_Value, __int64 _Mask) {
- return __atomic_fetch_or(_Value, _Mask, __ATOMIC_RELEASE);
-}
-#endif
-/*----------------------------------------------------------------------------*\
-|* Interlocked Xor
-\*----------------------------------------------------------------------------*/
-#if defined(__arm__) || defined(__aarch64__)
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedXor8_acq(char volatile *_Value, char _Mask) {
- return __atomic_fetch_xor(_Value, _Mask, __ATOMIC_ACQUIRE);
-}
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedXor8_nf(char volatile *_Value, char _Mask) {
- return __atomic_fetch_xor(_Value, _Mask, __ATOMIC_RELAXED);
-}
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedXor8_rel(char volatile *_Value, char _Mask) {
- return __atomic_fetch_xor(_Value, _Mask, __ATOMIC_RELEASE);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedXor16_acq(short volatile *_Value, short _Mask) {
- return __atomic_fetch_xor(_Value, _Mask, __ATOMIC_ACQUIRE);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedXor16_nf(short volatile *_Value, short _Mask) {
- return __atomic_fetch_xor(_Value, _Mask, __ATOMIC_RELAXED);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedXor16_rel(short volatile *_Value, short _Mask) {
- return __atomic_fetch_xor(_Value, _Mask, __ATOMIC_RELEASE);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedXor_acq(long volatile *_Value, long _Mask) {
- return __atomic_fetch_xor(_Value, _Mask, __ATOMIC_ACQUIRE);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedXor_nf(long volatile *_Value, long _Mask) {
- return __atomic_fetch_xor(_Value, _Mask, __ATOMIC_RELAXED);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedXor_rel(long volatile *_Value, long _Mask) {
- return __atomic_fetch_xor(_Value, _Mask, __ATOMIC_RELEASE);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedXor64_acq(__int64 volatile *_Value, __int64 _Mask) {
- return __atomic_fetch_xor(_Value, _Mask, __ATOMIC_ACQUIRE);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedXor64_nf(__int64 volatile *_Value, __int64 _Mask) {
- return __atomic_fetch_xor(_Value, _Mask, __ATOMIC_RELAXED);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedXor64_rel(__int64 volatile *_Value, __int64 _Mask) {
- return __atomic_fetch_xor(_Value, _Mask, __ATOMIC_RELEASE);
-}
-#endif
-/*----------------------------------------------------------------------------*\
-|* Interlocked Exchange
-\*----------------------------------------------------------------------------*/
-#if defined(__arm__) || defined(__aarch64__)
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedExchange8_acq(char volatile *_Target, char _Value) {
- __atomic_exchange(_Target, &_Value, &_Value, __ATOMIC_ACQUIRE);
- return _Value;
-}
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedExchange8_nf(char volatile *_Target, char _Value) {
- __atomic_exchange(_Target, &_Value, &_Value, __ATOMIC_RELAXED);
- return _Value;
-}
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedExchange8_rel(char volatile *_Target, char _Value) {
- __atomic_exchange(_Target, &_Value, &_Value, __ATOMIC_RELEASE);
- return _Value;
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedExchange16_acq(short volatile *_Target, short _Value) {
- __atomic_exchange(_Target, &_Value, &_Value, __ATOMIC_ACQUIRE);
- return _Value;
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedExchange16_nf(short volatile *_Target, short _Value) {
- __atomic_exchange(_Target, &_Value, &_Value, __ATOMIC_RELAXED);
- return _Value;
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedExchange16_rel(short volatile *_Target, short _Value) {
- __atomic_exchange(_Target, &_Value, &_Value, __ATOMIC_RELEASE);
- return _Value;
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedExchange_acq(long volatile *_Target, long _Value) {
- __atomic_exchange(_Target, &_Value, &_Value, __ATOMIC_ACQUIRE);
- return _Value;
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedExchange_nf(long volatile *_Target, long _Value) {
- __atomic_exchange(_Target, &_Value, &_Value, __ATOMIC_RELAXED);
- return _Value;
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedExchange_rel(long volatile *_Target, long _Value) {
- __atomic_exchange(_Target, &_Value, &_Value, __ATOMIC_RELEASE);
- return _Value;
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedExchange64_acq(__int64 volatile *_Target, __int64 _Value) {
- __atomic_exchange(_Target, &_Value, &_Value, __ATOMIC_ACQUIRE);
- return _Value;
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedExchange64_nf(__int64 volatile *_Target, __int64 _Value) {
- __atomic_exchange(_Target, &_Value, &_Value, __ATOMIC_RELAXED);
- return _Value;
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedExchange64_rel(__int64 volatile *_Target, __int64 _Value) {
- __atomic_exchange(_Target, &_Value, &_Value, __ATOMIC_RELEASE);
- return _Value;
-}
-#endif
-/*----------------------------------------------------------------------------*\
-|* Interlocked Compare Exchange
-\*----------------------------------------------------------------------------*/
-#if defined(__arm__) || defined(__aarch64__)
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedCompareExchange8_acq(char volatile *_Destination,
- char _Exchange, char _Comparand) {
- __atomic_compare_exchange(_Destination, &_Comparand, &_Exchange, 0,
- __ATOMIC_SEQ_CST, __ATOMIC_ACQUIRE);
- return _Comparand;
-}
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedCompareExchange8_nf(char volatile *_Destination,
- char _Exchange, char _Comparand) {
- __atomic_compare_exchange(_Destination, &_Comparand, &_Exchange, 0,
- __ATOMIC_SEQ_CST, __ATOMIC_RELAXED);
- return _Comparand;
-}
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedCompareExchange8_rel(char volatile *_Destination,
- char _Exchange, char _Comparand) {
- __atomic_compare_exchange(_Destination, &_Comparand, &_Exchange, 0,
- __ATOMIC_SEQ_CST, __ATOMIC_RELEASE);
- return _Comparand;
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedCompareExchange16_acq(short volatile *_Destination,
- short _Exchange, short _Comparand) {
- __atomic_compare_exchange(_Destination, &_Comparand, &_Exchange, 0,
- __ATOMIC_SEQ_CST, __ATOMIC_ACQUIRE);
- return _Comparand;
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedCompareExchange16_nf(short volatile *_Destination,
- short _Exchange, short _Comparand) {
- __atomic_compare_exchange(_Destination, &_Comparand, &_Exchange, 0,
- __ATOMIC_SEQ_CST, __ATOMIC_RELAXED);
- return _Comparand;
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedCompareExchange16_rel(short volatile *_Destination,
- short _Exchange, short _Comparand) {
- __atomic_compare_exchange(_Destination, &_Comparand, &_Exchange, 0,
- __ATOMIC_SEQ_CST, __ATOMIC_RELEASE);
- return _Comparand;
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedCompareExchange_acq(long volatile *_Destination,
- long _Exchange, long _Comparand) {
- __atomic_compare_exchange(_Destination, &_Comparand, &_Exchange, 0,
- __ATOMIC_SEQ_CST, __ATOMIC_ACQUIRE);
- return _Comparand;
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedCompareExchange_nf(long volatile *_Destination,
- long _Exchange, long _Comparand) {
- __atomic_compare_exchange(_Destination, &_Comparand, &_Exchange, 0,
- __ATOMIC_SEQ_CST, __ATOMIC_RELAXED);
- return _Comparand;
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedCompareExchange_rel(long volatile *_Destination,
- long _Exchange, long _Comparand) {
- __atomic_compare_exchange(_Destination, &_Comparand, &_Exchange, 0,
- __ATOMIC_SEQ_CST, __ATOMIC_RELEASE);
- return _Comparand;
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedCompareExchange64_acq(__int64 volatile *_Destination,
- __int64 _Exchange, __int64 _Comparand) {
- __atomic_compare_exchange(_Destination, &_Comparand, &_Exchange, 0,
- __ATOMIC_SEQ_CST, __ATOMIC_ACQUIRE);
- return _Comparand;
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedCompareExchange64_nf(__int64 volatile *_Destination,
- __int64 _Exchange, __int64 _Comparand) {
- __atomic_compare_exchange(_Destination, &_Comparand, &_Exchange, 0,
- __ATOMIC_SEQ_CST, __ATOMIC_RELAXED);
- return _Comparand;
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedCompareExchange64_rel(__int64 volatile *_Destination,
- __int64 _Exchange, __int64 _Comparand) {
- __atomic_compare_exchange(_Destination, &_Comparand, &_Exchange, 0,
- __ATOMIC_SEQ_CST, __ATOMIC_RELEASE);
- return _Comparand;
-}
-#endif
-
-/*----------------------------------------------------------------------------*\
-|* movs, stos
-\*----------------------------------------------------------------------------*/
-#if defined(__i386__) || defined(__x86_64__)
-static __inline__ void __DEFAULT_FN_ATTRS
-__movsb(unsigned char *__dst, unsigned char const *__src, size_t __n) {
- __asm__("rep movsb" : : "D"(__dst), "S"(__src), "c"(__n));
-}
-static __inline__ void __DEFAULT_FN_ATTRS
-__movsd(unsigned long *__dst, unsigned long const *__src, size_t __n) {
- __asm__("rep movsl" : : "D"(__dst), "S"(__src), "c"(__n));
-}
-static __inline__ void __DEFAULT_FN_ATTRS
-__movsw(unsigned short *__dst, unsigned short const *__src, size_t __n) {
- __asm__("rep movsw" : : "D"(__dst), "S"(__src), "c"(__n));
-}
-static __inline__ void __DEFAULT_FN_ATTRS
-__stosd(unsigned long *__dst, unsigned long __x, size_t __n) {
- __asm__("rep stosl" : : "D"(__dst), "a"(__x), "c"(__n));
-}
-static __inline__ void __DEFAULT_FN_ATTRS
-__stosw(unsigned short *__dst, unsigned short __x, size_t __n) {
- __asm__("rep stosw" : : "D"(__dst), "a"(__x), "c"(__n));
-}
-#endif
-#ifdef __x86_64__
-static __inline__ void __DEFAULT_FN_ATTRS
-__movsq(unsigned long long *__dst, unsigned long long const *__src, size_t __n) {
- __asm__("rep movsq" : : "D"(__dst), "S"(__src), "c"(__n));
-}
-static __inline__ void __DEFAULT_FN_ATTRS
-__stosq(unsigned __int64 *__dst, unsigned __int64 __x, size_t __n) {
- __asm__("rep stosq" : : "D"(__dst), "a"(__x), "c"(__n));
-}
-#endif
-
-/*----------------------------------------------------------------------------*\
-|* Misc
-\*----------------------------------------------------------------------------*/
-#if defined(__i386__) || defined(__x86_64__)
-static __inline__ void __DEFAULT_FN_ATTRS
-__cpuid(int __info[4], int __level) {
- __asm__ ("cpuid" : "=a"(__info[0]), "=b" (__info[1]), "=c"(__info[2]), "=d"(__info[3])
- : "a"(__level));
-}
-static __inline__ void __DEFAULT_FN_ATTRS
-__cpuidex(int __info[4], int __level, int __ecx) {
- __asm__ ("cpuid" : "=a"(__info[0]), "=b" (__info[1]), "=c"(__info[2]), "=d"(__info[3])
- : "a"(__level), "c"(__ecx));
-}
-static __inline__ unsigned __int64 __cdecl __DEFAULT_FN_ATTRS
-_xgetbv(unsigned int __xcr_no) {
- unsigned int __eax, __edx;
- __asm__ ("xgetbv" : "=a" (__eax), "=d" (__edx) : "c" (__xcr_no));
- return ((unsigned __int64)__edx << 32) | __eax;
-}
-static __inline__ void __DEFAULT_FN_ATTRS
-__halt(void) {
- __asm__ volatile ("hlt");
-}
-static __inline__ void __DEFAULT_FN_ATTRS
-__nop(void) {
- __asm__ volatile ("nop");
-}
-#endif
-
-/*----------------------------------------------------------------------------*\
-|* Privileged intrinsics
-\*----------------------------------------------------------------------------*/
-#if defined(__i386__) || defined(__x86_64__)
-static __inline__ unsigned __int64 __DEFAULT_FN_ATTRS
-__readmsr(unsigned long __register) {
- // Loads the contents of a 64-bit model specific register (MSR) specified in
- // the ECX register into registers EDX:EAX. The EDX register is loaded with
- // the high-order 32 bits of the MSR and the EAX register is loaded with the
- // low-order 32 bits. If less than 64 bits are implemented in the MSR being
- // read, the values returned to EDX:EAX in unimplemented bit locations are
- // undefined.
- unsigned long __edx;
- unsigned long __eax;
- __asm__ ("rdmsr" : "=d"(__edx), "=a"(__eax) : "c"(__register));
- return (((unsigned __int64)__edx) << 32) | (unsigned __int64)__eax;
-}
-
-static __inline__ unsigned long __DEFAULT_FN_ATTRS
-__readcr3(void) {
- unsigned long __cr3_val;
- __asm__ __volatile__ ("mov %%cr3, %0" : "=q"(__cr3_val) : : "memory");
- return __cr3_val;
-}
-
-static __inline__ void __DEFAULT_FN_ATTRS
-__writecr3(unsigned int __cr3_val) {
- __asm__ ("mov %0, %%cr3" : : "q"(__cr3_val) : "memory");
-}
-#endif
-
-#ifdef __cplusplus
-}
-#endif
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif /* __INTRIN_H */
-#endif /* _MSC_VER */
+++ /dev/null
-/*===---- iso646.h - Standard header for alternate spellings of operators---===
- *
- * Copyright (c) 2008 Eli Friedman
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __ISO646_H
-#define __ISO646_H
-
-#ifndef __cplusplus
-#define and &&
-#define and_eq &=
-#define bitand &
-#define bitor |
-#define compl ~
-#define not !
-#define not_eq !=
-#define or ||
-#define or_eq |=
-#define xor ^
-#define xor_eq ^=
-#endif
-
-#endif /* __ISO646_H */
+++ /dev/null
-/*===---- mwaitxintrin.h - MONITORX/MWAITX intrinsics ----------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __X86INTRIN_H
-#error "Never use <mwaitxintrin.h> directly; include <x86intrin.h> instead."
-#endif
-
-#ifndef _MWAITXINTRIN_H
-#define _MWAITXINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("mwaitx")))
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_monitorx(void const * __p, unsigned __extensions, unsigned __hints)
-{
- __builtin_ia32_monitorx((void *)__p, __extensions, __hints);
-}
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_mwaitx(unsigned __extensions, unsigned __hints, unsigned __clock)
-{
- __builtin_ia32_mwaitx(__extensions, __hints, __clock);
-}
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif /* _MWAITXINTRIN_H */
+++ /dev/null
-/*===---- nmmintrin.h - SSE4 intrinsics ------------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef _NMMINTRIN_H
-#define _NMMINTRIN_H
-
-/* To match expectations of gcc we put the sse4.2 definitions into smmintrin.h,
- just include it now then. */
-#include <smmintrin.h>
-#endif /* _NMMINTRIN_H */
+++ /dev/null
-/*===------------- pkuintrin.h - PKU intrinsics ------------------===
- *
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-#ifndef __IMMINTRIN_H
-#error "Never use <pkuintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __PKUINTRIN_H
-#define __PKUINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("pku")))
-
-static __inline__ unsigned int __DEFAULT_FN_ATTRS
-_rdpkru_u32(void)
-{
- return __builtin_ia32_rdpkru();
-}
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_wrpkru(unsigned int __val)
-{
- return __builtin_ia32_wrpkru(__val);
-}
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif
+++ /dev/null
-/*===---- popcntintrin.h - POPCNT intrinsics -------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef _POPCNTINTRIN_H
-#define _POPCNTINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("popcnt")))
-
-/// \brief Counts the number of bits in the source operand having a value of 1.
-///
-/// \headerfile <x86intrin.h>
-///
-/// This intrinsic corresponds to the <c> POPCNT </c> instruction.
-///
-/// \param __A
-/// An unsigned 32-bit integer operand.
-/// \returns A 32-bit integer containing the number of bits with value 1 in the
-/// source operand.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_popcnt_u32(unsigned int __A)
-{
- return __builtin_popcount(__A);
-}
-
-/// \brief Counts the number of bits in the source operand having a value of 1.
-///
-/// \headerfile <x86intrin.h>
-///
-/// This intrinsic corresponds to the <c> POPCNT </c> instruction.
-///
-/// \param __A
-/// A signed 32-bit integer operand.
-/// \returns A 32-bit integer containing the number of bits with value 1 in the
-/// source operand.
-static __inline__ int __DEFAULT_FN_ATTRS
-_popcnt32(int __A)
-{
- return __builtin_popcount(__A);
-}
-
-#ifdef __x86_64__
-/// \brief Counts the number of bits in the source operand having a value of 1.
-///
-/// \headerfile <x86intrin.h>
-///
-/// This intrinsic corresponds to the <c> POPCNT </c> instruction.
-///
-/// \param __A
-/// An unsigned 64-bit integer operand.
-/// \returns A 64-bit integer containing the number of bits with value 1 in the
-/// source operand.
-static __inline__ long long __DEFAULT_FN_ATTRS
-_mm_popcnt_u64(unsigned long long __A)
-{
- return __builtin_popcountll(__A);
-}
-
-/// \brief Counts the number of bits in the source operand having a value of 1.
-///
-/// \headerfile <x86intrin.h>
-///
-/// This intrinsic corresponds to the <c> POPCNT </c> instruction.
-///
-/// \param __A
-/// A signed 64-bit integer operand.
-/// \returns A 64-bit integer containing the number of bits with value 1 in the
-/// source operand.
-static __inline__ long long __DEFAULT_FN_ATTRS
-_popcnt64(long long __A)
-{
- return __builtin_popcountll(__A);
-}
-#endif /* __x86_64__ */
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif /* _POPCNTINTRIN_H */
+++ /dev/null
-/*===---- rdseedintrin.h - RDSEED intrinsics -------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __X86INTRIN_H
-#error "Never use <rdseedintrin.h> directly; include <x86intrin.h> instead."
-#endif
-
-#ifndef __RDSEEDINTRIN_H
-#define __RDSEEDINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("rdseed")))
-
-static __inline__ int __DEFAULT_FN_ATTRS
-_rdseed16_step(unsigned short *__p)
-{
- return __builtin_ia32_rdseed16_step(__p);
-}
-
-static __inline__ int __DEFAULT_FN_ATTRS
-_rdseed32_step(unsigned int *__p)
-{
- return __builtin_ia32_rdseed32_step(__p);
-}
-
-#ifdef __x86_64__
-static __inline__ int __DEFAULT_FN_ATTRS
-_rdseed64_step(unsigned long long *__p)
-{
- return __builtin_ia32_rdseed64_step(__p);
-}
-#endif
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif /* __RDSEEDINTRIN_H */
+++ /dev/null
-/*===---- rtmintrin.h - RTM intrinsics -------------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __IMMINTRIN_H
-#error "Never use <rtmintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __RTMINTRIN_H
-#define __RTMINTRIN_H
-
-#define _XBEGIN_STARTED (~0u)
-#define _XABORT_EXPLICIT (1 << 0)
-#define _XABORT_RETRY (1 << 1)
-#define _XABORT_CONFLICT (1 << 2)
-#define _XABORT_CAPACITY (1 << 3)
-#define _XABORT_DEBUG (1 << 4)
-#define _XABORT_NESTED (1 << 5)
-#define _XABORT_CODE(x) (((x) >> 24) & 0xFF)
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("rtm")))
-
-static __inline__ unsigned int __DEFAULT_FN_ATTRS
-_xbegin(void)
-{
- return __builtin_ia32_xbegin();
-}
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_xend(void)
-{
- __builtin_ia32_xend();
-}
-
-#define _xabort(imm) __builtin_ia32_xabort((imm))
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif /* __RTMINTRIN_H */
+++ /dev/null
-/*===---- s390intrin.h - SystemZ intrinsics --------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __S390INTRIN_H
-#define __S390INTRIN_H
-
-#ifndef __s390__
-#error "<s390intrin.h> is for s390 only"
-#endif
-
-#ifdef __HTM__
-#include <htmintrin.h>
-#endif
-
-#ifdef __VEC__
-#include <vecintrin.h>
-#endif
-
-#endif /* __S390INTRIN_H*/
+++ /dev/null
-/*===---- stdalign.h - Standard header for alignment ------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __STDALIGN_H
-#define __STDALIGN_H
-
-#ifndef __cplusplus
-#define alignas _Alignas
-#define alignof _Alignof
-#endif
-
-#define __alignas_is_defined 1
-#define __alignof_is_defined 1
-
-#endif /* __STDALIGN_H */
+++ /dev/null
-/*===---- stdarg.h - Variable argument handling ----------------------------===
- *
- * Copyright (c) 2008 Eli Friedman
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __STDARG_H
-#define __STDARG_H
-
-#ifndef _VA_LIST
-typedef __builtin_va_list va_list;
-#define _VA_LIST
-#endif
-#define va_start(ap, param) __builtin_va_start(ap, param)
-#define va_end(ap) __builtin_va_end(ap)
-#define va_arg(ap, type) __builtin_va_arg(ap, type)
-
-/* GCC always defines __va_copy, but does not define va_copy unless in c99 mode
- * or -ansi is not specified, since it was not part of C90.
- */
-#define __va_copy(d,s) __builtin_va_copy(d,s)
-
-#if __STDC_VERSION__ >= 199901L || __cplusplus >= 201103L || !defined(__STRICT_ANSI__)
-#define va_copy(dest, src) __builtin_va_copy(dest, src)
-#endif
-
-#ifndef __GNUC_VA_LIST
-#define __GNUC_VA_LIST 1
-typedef __builtin_va_list __gnuc_va_list;
-#endif
-
-#endif /* __STDARG_H */
+++ /dev/null
-/*===---- stdbool.h - Standard header for booleans -------------------------===
- *
- * Copyright (c) 2008 Eli Friedman
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __STDBOOL_H
-#define __STDBOOL_H
-
-/* Don't define bool, true, and false in C++, except as a GNU extension. */
-#ifndef __cplusplus
-#define bool _Bool
-#define true 1
-#define false 0
-#elif defined(__GNUC__) && !defined(__STRICT_ANSI__)
-/* Define _Bool, bool, false, true as a GNU extension. */
-#define _Bool bool
-#define bool bool
-#define false false
-#define true true
-#endif
-
-#define __bool_true_false_are_defined 1
-
-#endif /* __STDBOOL_H */
+++ /dev/null
-/*===---- stdnoreturn.h - Standard header for noreturn macro ---------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __STDNORETURN_H
-#define __STDNORETURN_H
-
-#define noreturn _Noreturn
-#define __noreturn_is_defined 1
-
-#endif /* __STDNORETURN_H */
+++ /dev/null
-/*===---- varargs.h - Variable argument handling -------------------------------------===
-*
-* Permission is hereby granted, free of charge, to any person obtaining a copy
-* of this software and associated documentation files (the "Software"), to deal
-* in the Software without restriction, including without limitation the rights
-* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-* copies of the Software, and to permit persons to whom the Software is
-* furnished to do so, subject to the following conditions:
-*
-* The above copyright notice and this permission notice shall be included in
-* all copies or substantial portions of the Software.
-*
-* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
-* THE SOFTWARE.
-*
-*===-----------------------------------------------------------------------===
-*/
-#ifndef __VARARGS_H
-#define __VARARGS_H
- #error "Please use <stdarg.h> instead of <varargs.h>"
-#endif
+++ /dev/null
-/*===---- vecintrin.h - Vector intrinsics ----------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#if defined(__s390x__) && defined(__VEC__)
-
-#define __ATTRS_ai __attribute__((__always_inline__))
-#define __ATTRS_o __attribute__((__overloadable__))
-#define __ATTRS_o_ai __attribute__((__overloadable__, __always_inline__))
-
-#define __constant(PARM) \
- __attribute__((__enable_if__ ((PARM) == (PARM), \
- "argument must be a constant integer")))
-#define __constant_range(PARM, LOW, HIGH) \
- __attribute__((__enable_if__ ((PARM) >= (LOW) && (PARM) <= (HIGH), \
- "argument must be a constant integer from " #LOW " to " #HIGH)))
-#define __constant_pow2_range(PARM, LOW, HIGH) \
- __attribute__((__enable_if__ ((PARM) >= (LOW) && (PARM) <= (HIGH) && \
- ((PARM) & ((PARM) - 1)) == 0, \
- "argument must be a constant power of 2 from " #LOW " to " #HIGH)))
-
-/*-- __lcbb -----------------------------------------------------------------*/
-
-extern __ATTRS_o unsigned int
-__lcbb(const void *__ptr, unsigned short __len)
- __constant_pow2_range(__len, 64, 4096);
-
-#define __lcbb(X, Y) ((__typeof__((__lcbb)((X), (Y)))) \
- __builtin_s390_lcbb((X), __builtin_constant_p((Y))? \
- ((Y) == 64 ? 0 : \
- (Y) == 128 ? 1 : \
- (Y) == 256 ? 2 : \
- (Y) == 512 ? 3 : \
- (Y) == 1024 ? 4 : \
- (Y) == 2048 ? 5 : \
- (Y) == 4096 ? 6 : 0) : 0))
-
-/*-- vec_extract ------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai signed char
-vec_extract(vector signed char __vec, int __index) {
- return __vec[__index & 15];
-}
-
-static inline __ATTRS_o_ai unsigned char
-vec_extract(vector bool char __vec, int __index) {
- return __vec[__index & 15];
-}
-
-static inline __ATTRS_o_ai unsigned char
-vec_extract(vector unsigned char __vec, int __index) {
- return __vec[__index & 15];
-}
-
-static inline __ATTRS_o_ai signed short
-vec_extract(vector signed short __vec, int __index) {
- return __vec[__index & 7];
-}
-
-static inline __ATTRS_o_ai unsigned short
-vec_extract(vector bool short __vec, int __index) {
- return __vec[__index & 7];
-}
-
-static inline __ATTRS_o_ai unsigned short
-vec_extract(vector unsigned short __vec, int __index) {
- return __vec[__index & 7];
-}
-
-static inline __ATTRS_o_ai signed int
-vec_extract(vector signed int __vec, int __index) {
- return __vec[__index & 3];
-}
-
-static inline __ATTRS_o_ai unsigned int
-vec_extract(vector bool int __vec, int __index) {
- return __vec[__index & 3];
-}
-
-static inline __ATTRS_o_ai unsigned int
-vec_extract(vector unsigned int __vec, int __index) {
- return __vec[__index & 3];
-}
-
-static inline __ATTRS_o_ai signed long long
-vec_extract(vector signed long long __vec, int __index) {
- return __vec[__index & 1];
-}
-
-static inline __ATTRS_o_ai unsigned long long
-vec_extract(vector bool long long __vec, int __index) {
- return __vec[__index & 1];
-}
-
-static inline __ATTRS_o_ai unsigned long long
-vec_extract(vector unsigned long long __vec, int __index) {
- return __vec[__index & 1];
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai float
-vec_extract(vector float __vec, int __index) {
- return __vec[__index & 3];
-}
-#endif
-
-static inline __ATTRS_o_ai double
-vec_extract(vector double __vec, int __index) {
- return __vec[__index & 1];
-}
-
-/*-- vec_insert -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_insert(signed char __scalar, vector signed char __vec, int __index) {
- __vec[__index & 15] = __scalar;
- return __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_insert(unsigned char __scalar, vector bool char __vec, int __index) {
- vector unsigned char __newvec = (vector unsigned char)__vec;
- __newvec[__index & 15] = (unsigned char)__scalar;
- return __newvec;
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_insert(unsigned char __scalar, vector unsigned char __vec, int __index) {
- __vec[__index & 15] = __scalar;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_insert(signed short __scalar, vector signed short __vec, int __index) {
- __vec[__index & 7] = __scalar;
- return __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_insert(unsigned short __scalar, vector bool short __vec, int __index) {
- vector unsigned short __newvec = (vector unsigned short)__vec;
- __newvec[__index & 7] = (unsigned short)__scalar;
- return __newvec;
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_insert(unsigned short __scalar, vector unsigned short __vec, int __index) {
- __vec[__index & 7] = __scalar;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_insert(signed int __scalar, vector signed int __vec, int __index) {
- __vec[__index & 3] = __scalar;
- return __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_insert(unsigned int __scalar, vector bool int __vec, int __index) {
- vector unsigned int __newvec = (vector unsigned int)__vec;
- __newvec[__index & 3] = __scalar;
- return __newvec;
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_insert(unsigned int __scalar, vector unsigned int __vec, int __index) {
- __vec[__index & 3] = __scalar;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_insert(signed long long __scalar, vector signed long long __vec,
- int __index) {
- __vec[__index & 1] = __scalar;
- return __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_insert(unsigned long long __scalar, vector bool long long __vec,
- int __index) {
- vector unsigned long long __newvec = (vector unsigned long long)__vec;
- __newvec[__index & 1] = __scalar;
- return __newvec;
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_insert(unsigned long long __scalar, vector unsigned long long __vec,
- int __index) {
- __vec[__index & 1] = __scalar;
- return __vec;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_insert(float __scalar, vector float __vec, int __index) {
- __vec[__index & 1] = __scalar;
- return __vec;
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_insert(double __scalar, vector double __vec, int __index) {
- __vec[__index & 1] = __scalar;
- return __vec;
-}
-
-/*-- vec_promote ------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_promote(signed char __scalar, int __index) {
- const vector signed char __zero = (vector signed char)0;
- vector signed char __vec = __builtin_shufflevector(__zero, __zero,
- -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1);
- __vec[__index & 15] = __scalar;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_promote(unsigned char __scalar, int __index) {
- const vector unsigned char __zero = (vector unsigned char)0;
- vector unsigned char __vec = __builtin_shufflevector(__zero, __zero,
- -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1);
- __vec[__index & 15] = __scalar;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_promote(signed short __scalar, int __index) {
- const vector signed short __zero = (vector signed short)0;
- vector signed short __vec = __builtin_shufflevector(__zero, __zero,
- -1, -1, -1, -1, -1, -1, -1, -1);
- __vec[__index & 7] = __scalar;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_promote(unsigned short __scalar, int __index) {
- const vector unsigned short __zero = (vector unsigned short)0;
- vector unsigned short __vec = __builtin_shufflevector(__zero, __zero,
- -1, -1, -1, -1, -1, -1, -1, -1);
- __vec[__index & 7] = __scalar;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_promote(signed int __scalar, int __index) {
- const vector signed int __zero = (vector signed int)0;
- vector signed int __vec = __builtin_shufflevector(__zero, __zero,
- -1, -1, -1, -1);
- __vec[__index & 3] = __scalar;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_promote(unsigned int __scalar, int __index) {
- const vector unsigned int __zero = (vector unsigned int)0;
- vector unsigned int __vec = __builtin_shufflevector(__zero, __zero,
- -1, -1, -1, -1);
- __vec[__index & 3] = __scalar;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_promote(signed long long __scalar, int __index) {
- const vector signed long long __zero = (vector signed long long)0;
- vector signed long long __vec = __builtin_shufflevector(__zero, __zero,
- -1, -1);
- __vec[__index & 1] = __scalar;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_promote(unsigned long long __scalar, int __index) {
- const vector unsigned long long __zero = (vector unsigned long long)0;
- vector unsigned long long __vec = __builtin_shufflevector(__zero, __zero,
- -1, -1);
- __vec[__index & 1] = __scalar;
- return __vec;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_promote(float __scalar, int __index) {
- const vector float __zero = (vector float)0;
- vector float __vec = __builtin_shufflevector(__zero, __zero, -1, -1, -1, -1);
- __vec[__index & 3] = __scalar;
- return __vec;
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_promote(double __scalar, int __index) {
- const vector double __zero = (vector double)0;
- vector double __vec = __builtin_shufflevector(__zero, __zero, -1, -1);
- __vec[__index & 1] = __scalar;
- return __vec;
-}
-
-/*-- vec_insert_and_zero ----------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_insert_and_zero(const signed char *__ptr) {
- vector signed char __vec = (vector signed char)0;
- __vec[7] = *__ptr;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_insert_and_zero(const unsigned char *__ptr) {
- vector unsigned char __vec = (vector unsigned char)0;
- __vec[7] = *__ptr;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_insert_and_zero(const signed short *__ptr) {
- vector signed short __vec = (vector signed short)0;
- __vec[3] = *__ptr;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_insert_and_zero(const unsigned short *__ptr) {
- vector unsigned short __vec = (vector unsigned short)0;
- __vec[3] = *__ptr;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_insert_and_zero(const signed int *__ptr) {
- vector signed int __vec = (vector signed int)0;
- __vec[1] = *__ptr;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_insert_and_zero(const unsigned int *__ptr) {
- vector unsigned int __vec = (vector unsigned int)0;
- __vec[1] = *__ptr;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_insert_and_zero(const signed long long *__ptr) {
- vector signed long long __vec = (vector signed long long)0;
- __vec[0] = *__ptr;
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_insert_and_zero(const unsigned long long *__ptr) {
- vector unsigned long long __vec = (vector unsigned long long)0;
- __vec[0] = *__ptr;
- return __vec;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_insert_and_zero(const float *__ptr) {
- vector float __vec = (vector float)0;
- __vec[0] = *__ptr;
- return __vec;
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_insert_and_zero(const double *__ptr) {
- vector double __vec = (vector double)0;
- __vec[0] = *__ptr;
- return __vec;
-}
-
-/*-- vec_perm ---------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_perm(vector signed char __a, vector signed char __b,
- vector unsigned char __c) {
- return (vector signed char)__builtin_s390_vperm(
- (vector unsigned char)__a, (vector unsigned char)__b, __c);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_perm(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c) {
- return (vector unsigned char)__builtin_s390_vperm(
- (vector unsigned char)__a, (vector unsigned char)__b, __c);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_perm(vector bool char __a, vector bool char __b,
- vector unsigned char __c) {
- return (vector bool char)__builtin_s390_vperm(
- (vector unsigned char)__a, (vector unsigned char)__b, __c);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_perm(vector signed short __a, vector signed short __b,
- vector unsigned char __c) {
- return (vector signed short)__builtin_s390_vperm(
- (vector unsigned char)__a, (vector unsigned char)__b, __c);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_perm(vector unsigned short __a, vector unsigned short __b,
- vector unsigned char __c) {
- return (vector unsigned short)__builtin_s390_vperm(
- (vector unsigned char)__a, (vector unsigned char)__b, __c);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_perm(vector bool short __a, vector bool short __b,
- vector unsigned char __c) {
- return (vector bool short)__builtin_s390_vperm(
- (vector unsigned char)__a, (vector unsigned char)__b, __c);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_perm(vector signed int __a, vector signed int __b,
- vector unsigned char __c) {
- return (vector signed int)__builtin_s390_vperm(
- (vector unsigned char)__a, (vector unsigned char)__b, __c);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_perm(vector unsigned int __a, vector unsigned int __b,
- vector unsigned char __c) {
- return (vector unsigned int)__builtin_s390_vperm(
- (vector unsigned char)__a, (vector unsigned char)__b, __c);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_perm(vector bool int __a, vector bool int __b,
- vector unsigned char __c) {
- return (vector bool int)__builtin_s390_vperm(
- (vector unsigned char)__a, (vector unsigned char)__b, __c);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_perm(vector signed long long __a, vector signed long long __b,
- vector unsigned char __c) {
- return (vector signed long long)__builtin_s390_vperm(
- (vector unsigned char)__a, (vector unsigned char)__b, __c);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_perm(vector unsigned long long __a, vector unsigned long long __b,
- vector unsigned char __c) {
- return (vector unsigned long long)__builtin_s390_vperm(
- (vector unsigned char)__a, (vector unsigned char)__b, __c);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_perm(vector bool long long __a, vector bool long long __b,
- vector unsigned char __c) {
- return (vector bool long long)__builtin_s390_vperm(
- (vector unsigned char)__a, (vector unsigned char)__b, __c);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_perm(vector float __a, vector float __b,
- vector unsigned char __c) {
- return (vector float)__builtin_s390_vperm(
- (vector unsigned char)__a, (vector unsigned char)__b, __c);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_perm(vector double __a, vector double __b,
- vector unsigned char __c) {
- return (vector double)__builtin_s390_vperm(
- (vector unsigned char)__a, (vector unsigned char)__b, __c);
-}
-
-/*-- vec_permi --------------------------------------------------------------*/
-
-// This prototype is deprecated.
-extern __ATTRS_o vector signed long long
-vec_permi(vector signed long long __a, vector signed long long __b, int __c)
- __constant_range(__c, 0, 3);
-
-// This prototype is deprecated.
-extern __ATTRS_o vector unsigned long long
-vec_permi(vector unsigned long long __a, vector unsigned long long __b, int __c)
- __constant_range(__c, 0, 3);
-
-// This prototype is deprecated.
-extern __ATTRS_o vector bool long long
-vec_permi(vector bool long long __a, vector bool long long __b, int __c)
- __constant_range(__c, 0, 3);
-
-// This prototype is deprecated.
-extern __ATTRS_o vector double
-vec_permi(vector double __a, vector double __b, int __c)
- __constant_range(__c, 0, 3);
-
-#define vec_permi(X, Y, Z) ((__typeof__((vec_permi)((X), (Y), (Z)))) \
- __builtin_s390_vpdi((vector unsigned long long)(X), \
- (vector unsigned long long)(Y), \
- (((Z) & 2) << 1) | ((Z) & 1)))
-
-/*-- vec_bperm_u128 ---------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_ai vector unsigned long long
-vec_bperm_u128(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vbperm(__a, __b);
-}
-#endif
-
-/*-- vec_sel ----------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_sel(vector signed char __a, vector signed char __b,
- vector unsigned char __c) {
- return ((vector signed char)__c & __b) | (~(vector signed char)__c & __a);
-}
-
-static inline __ATTRS_o_ai vector signed char
-vec_sel(vector signed char __a, vector signed char __b, vector bool char __c) {
- return ((vector signed char)__c & __b) | (~(vector signed char)__c & __a);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_sel(vector bool char __a, vector bool char __b, vector unsigned char __c) {
- return ((vector bool char)__c & __b) | (~(vector bool char)__c & __a);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_sel(vector bool char __a, vector bool char __b, vector bool char __c) {
- return (__c & __b) | (~__c & __a);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_sel(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c) {
- return (__c & __b) | (~__c & __a);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_sel(vector unsigned char __a, vector unsigned char __b,
- vector bool char __c) {
- return ((vector unsigned char)__c & __b) | (~(vector unsigned char)__c & __a);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_sel(vector signed short __a, vector signed short __b,
- vector unsigned short __c) {
- return ((vector signed short)__c & __b) | (~(vector signed short)__c & __a);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_sel(vector signed short __a, vector signed short __b,
- vector bool short __c) {
- return ((vector signed short)__c & __b) | (~(vector signed short)__c & __a);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_sel(vector bool short __a, vector bool short __b,
- vector unsigned short __c) {
- return ((vector bool short)__c & __b) | (~(vector bool short)__c & __a);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_sel(vector bool short __a, vector bool short __b, vector bool short __c) {
- return (__c & __b) | (~__c & __a);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_sel(vector unsigned short __a, vector unsigned short __b,
- vector unsigned short __c) {
- return (__c & __b) | (~__c & __a);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_sel(vector unsigned short __a, vector unsigned short __b,
- vector bool short __c) {
- return (((vector unsigned short)__c & __b) |
- (~(vector unsigned short)__c & __a));
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_sel(vector signed int __a, vector signed int __b,
- vector unsigned int __c) {
- return ((vector signed int)__c & __b) | (~(vector signed int)__c & __a);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_sel(vector signed int __a, vector signed int __b, vector bool int __c) {
- return ((vector signed int)__c & __b) | (~(vector signed int)__c & __a);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_sel(vector bool int __a, vector bool int __b, vector unsigned int __c) {
- return ((vector bool int)__c & __b) | (~(vector bool int)__c & __a);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_sel(vector bool int __a, vector bool int __b, vector bool int __c) {
- return (__c & __b) | (~__c & __a);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_sel(vector unsigned int __a, vector unsigned int __b,
- vector unsigned int __c) {
- return (__c & __b) | (~__c & __a);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_sel(vector unsigned int __a, vector unsigned int __b, vector bool int __c) {
- return ((vector unsigned int)__c & __b) | (~(vector unsigned int)__c & __a);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_sel(vector signed long long __a, vector signed long long __b,
- vector unsigned long long __c) {
- return (((vector signed long long)__c & __b) |
- (~(vector signed long long)__c & __a));
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_sel(vector signed long long __a, vector signed long long __b,
- vector bool long long __c) {
- return (((vector signed long long)__c & __b) |
- (~(vector signed long long)__c & __a));
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_sel(vector bool long long __a, vector bool long long __b,
- vector unsigned long long __c) {
- return (((vector bool long long)__c & __b) |
- (~(vector bool long long)__c & __a));
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_sel(vector bool long long __a, vector bool long long __b,
- vector bool long long __c) {
- return (__c & __b) | (~__c & __a);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_sel(vector unsigned long long __a, vector unsigned long long __b,
- vector unsigned long long __c) {
- return (__c & __b) | (~__c & __a);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_sel(vector unsigned long long __a, vector unsigned long long __b,
- vector bool long long __c) {
- return (((vector unsigned long long)__c & __b) |
- (~(vector unsigned long long)__c & __a));
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_sel(vector float __a, vector float __b, vector unsigned int __c) {
- return (vector float)((__c & (vector unsigned int)__b) |
- (~__c & (vector unsigned int)__a));
-}
-
-static inline __ATTRS_o_ai vector float
-vec_sel(vector float __a, vector float __b, vector bool int __c) {
- vector unsigned int __ac = (vector unsigned int)__a;
- vector unsigned int __bc = (vector unsigned int)__b;
- vector unsigned int __cc = (vector unsigned int)__c;
- return (vector float)((__cc & __bc) | (~__cc & __ac));
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_sel(vector double __a, vector double __b, vector unsigned long long __c) {
- return (vector double)((__c & (vector unsigned long long)__b) |
- (~__c & (vector unsigned long long)__a));
-}
-
-static inline __ATTRS_o_ai vector double
-vec_sel(vector double __a, vector double __b, vector bool long long __c) {
- vector unsigned long long __ac = (vector unsigned long long)__a;
- vector unsigned long long __bc = (vector unsigned long long)__b;
- vector unsigned long long __cc = (vector unsigned long long)__c;
- return (vector double)((__cc & __bc) | (~__cc & __ac));
-}
-
-/*-- vec_gather_element -----------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed int
-vec_gather_element(vector signed int __vec, vector unsigned int __offset,
- const signed int *__ptr, int __index)
- __constant_range(__index, 0, 3) {
- __vec[__index] = *(const signed int *)(
- (__INTPTR_TYPE__)__ptr + (__INTPTR_TYPE__)__offset[__index]);
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_gather_element(vector bool int __vec, vector unsigned int __offset,
- const unsigned int *__ptr, int __index)
- __constant_range(__index, 0, 3) {
- __vec[__index] = *(const unsigned int *)(
- (__INTPTR_TYPE__)__ptr + (__INTPTR_TYPE__)__offset[__index]);
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_gather_element(vector unsigned int __vec, vector unsigned int __offset,
- const unsigned int *__ptr, int __index)
- __constant_range(__index, 0, 3) {
- __vec[__index] = *(const unsigned int *)(
- (__INTPTR_TYPE__)__ptr + (__INTPTR_TYPE__)__offset[__index]);
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_gather_element(vector signed long long __vec,
- vector unsigned long long __offset,
- const signed long long *__ptr, int __index)
- __constant_range(__index, 0, 1) {
- __vec[__index] = *(const signed long long *)(
- (__INTPTR_TYPE__)__ptr + (__INTPTR_TYPE__)__offset[__index]);
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_gather_element(vector bool long long __vec,
- vector unsigned long long __offset,
- const unsigned long long *__ptr, int __index)
- __constant_range(__index, 0, 1) {
- __vec[__index] = *(const unsigned long long *)(
- (__INTPTR_TYPE__)__ptr + (__INTPTR_TYPE__)__offset[__index]);
- return __vec;
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_gather_element(vector unsigned long long __vec,
- vector unsigned long long __offset,
- const unsigned long long *__ptr, int __index)
- __constant_range(__index, 0, 1) {
- __vec[__index] = *(const unsigned long long *)(
- (__INTPTR_TYPE__)__ptr + (__INTPTR_TYPE__)__offset[__index]);
- return __vec;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_gather_element(vector float __vec, vector unsigned int __offset,
- const float *__ptr, int __index)
- __constant_range(__index, 0, 3) {
- __vec[__index] = *(const float *)(
- (__INTPTR_TYPE__)__ptr + (__INTPTR_TYPE__)__offset[__index]);
- return __vec;
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_gather_element(vector double __vec, vector unsigned long long __offset,
- const double *__ptr, int __index)
- __constant_range(__index, 0, 1) {
- __vec[__index] = *(const double *)(
- (__INTPTR_TYPE__)__ptr + (__INTPTR_TYPE__)__offset[__index]);
- return __vec;
-}
-
-/*-- vec_scatter_element ----------------------------------------------------*/
-
-static inline __ATTRS_o_ai void
-vec_scatter_element(vector signed int __vec, vector unsigned int __offset,
- signed int *__ptr, int __index)
- __constant_range(__index, 0, 3) {
- *(signed int *)((__INTPTR_TYPE__)__ptr + __offset[__index]) =
- __vec[__index];
-}
-
-static inline __ATTRS_o_ai void
-vec_scatter_element(vector bool int __vec, vector unsigned int __offset,
- unsigned int *__ptr, int __index)
- __constant_range(__index, 0, 3) {
- *(unsigned int *)((__INTPTR_TYPE__)__ptr + __offset[__index]) =
- __vec[__index];
-}
-
-static inline __ATTRS_o_ai void
-vec_scatter_element(vector unsigned int __vec, vector unsigned int __offset,
- unsigned int *__ptr, int __index)
- __constant_range(__index, 0, 3) {
- *(unsigned int *)((__INTPTR_TYPE__)__ptr + __offset[__index]) =
- __vec[__index];
-}
-
-static inline __ATTRS_o_ai void
-vec_scatter_element(vector signed long long __vec,
- vector unsigned long long __offset,
- signed long long *__ptr, int __index)
- __constant_range(__index, 0, 1) {
- *(signed long long *)((__INTPTR_TYPE__)__ptr + __offset[__index]) =
- __vec[__index];
-}
-
-static inline __ATTRS_o_ai void
-vec_scatter_element(vector bool long long __vec,
- vector unsigned long long __offset,
- unsigned long long *__ptr, int __index)
- __constant_range(__index, 0, 1) {
- *(unsigned long long *)((__INTPTR_TYPE__)__ptr + __offset[__index]) =
- __vec[__index];
-}
-
-static inline __ATTRS_o_ai void
-vec_scatter_element(vector unsigned long long __vec,
- vector unsigned long long __offset,
- unsigned long long *__ptr, int __index)
- __constant_range(__index, 0, 1) {
- *(unsigned long long *)((__INTPTR_TYPE__)__ptr + __offset[__index]) =
- __vec[__index];
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai void
-vec_scatter_element(vector float __vec, vector unsigned int __offset,
- float *__ptr, int __index)
- __constant_range(__index, 0, 3) {
- *(float *)((__INTPTR_TYPE__)__ptr + __offset[__index]) =
- __vec[__index];
-}
-#endif
-
-static inline __ATTRS_o_ai void
-vec_scatter_element(vector double __vec, vector unsigned long long __offset,
- double *__ptr, int __index)
- __constant_range(__index, 0, 1) {
- *(double *)((__INTPTR_TYPE__)__ptr + __offset[__index]) =
- __vec[__index];
-}
-
-/*-- vec_xl -----------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_xl(long __offset, const signed char *__ptr) {
- return *(const vector signed char *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_xl(long __offset, const unsigned char *__ptr) {
- return *(const vector unsigned char *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_xl(long __offset, const signed short *__ptr) {
- return *(const vector signed short *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_xl(long __offset, const unsigned short *__ptr) {
- return *(const vector unsigned short *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_xl(long __offset, const signed int *__ptr) {
- return *(const vector signed int *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_xl(long __offset, const unsigned int *__ptr) {
- return *(const vector unsigned int *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_xl(long __offset, const signed long long *__ptr) {
- return *(const vector signed long long *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_xl(long __offset, const unsigned long long *__ptr) {
- return *(const vector unsigned long long *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_xl(long __offset, const float *__ptr) {
- return *(const vector float *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_xl(long __offset, const double *__ptr) {
- return *(const vector double *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-/*-- vec_xld2 ---------------------------------------------------------------*/
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_xld2(long __offset, const signed char *__ptr) {
- return *(const vector signed char *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_xld2(long __offset, const unsigned char *__ptr) {
- return *(const vector unsigned char *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_xld2(long __offset, const signed short *__ptr) {
- return *(const vector signed short *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_xld2(long __offset, const unsigned short *__ptr) {
- return *(const vector unsigned short *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_xld2(long __offset, const signed int *__ptr) {
- return *(const vector signed int *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_xld2(long __offset, const unsigned int *__ptr) {
- return *(const vector unsigned int *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_xld2(long __offset, const signed long long *__ptr) {
- return *(const vector signed long long *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_xld2(long __offset, const unsigned long long *__ptr) {
- return *(const vector unsigned long long *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector double
-vec_xld2(long __offset, const double *__ptr) {
- return *(const vector double *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-/*-- vec_xlw4 ---------------------------------------------------------------*/
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_xlw4(long __offset, const signed char *__ptr) {
- return *(const vector signed char *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_xlw4(long __offset, const unsigned char *__ptr) {
- return *(const vector unsigned char *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_xlw4(long __offset, const signed short *__ptr) {
- return *(const vector signed short *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_xlw4(long __offset, const unsigned short *__ptr) {
- return *(const vector unsigned short *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_xlw4(long __offset, const signed int *__ptr) {
- return *(const vector signed int *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_xlw4(long __offset, const unsigned int *__ptr) {
- return *(const vector unsigned int *)((__INTPTR_TYPE__)__ptr + __offset);
-}
-
-/*-- vec_xst ----------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai void
-vec_xst(vector signed char __vec, long __offset, signed char *__ptr) {
- *(vector signed char *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-static inline __ATTRS_o_ai void
-vec_xst(vector unsigned char __vec, long __offset, unsigned char *__ptr) {
- *(vector unsigned char *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-static inline __ATTRS_o_ai void
-vec_xst(vector signed short __vec, long __offset, signed short *__ptr) {
- *(vector signed short *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-static inline __ATTRS_o_ai void
-vec_xst(vector unsigned short __vec, long __offset, unsigned short *__ptr) {
- *(vector unsigned short *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-static inline __ATTRS_o_ai void
-vec_xst(vector signed int __vec, long __offset, signed int *__ptr) {
- *(vector signed int *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-static inline __ATTRS_o_ai void
-vec_xst(vector unsigned int __vec, long __offset, unsigned int *__ptr) {
- *(vector unsigned int *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-static inline __ATTRS_o_ai void
-vec_xst(vector signed long long __vec, long __offset,
- signed long long *__ptr) {
- *(vector signed long long *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-static inline __ATTRS_o_ai void
-vec_xst(vector unsigned long long __vec, long __offset,
- unsigned long long *__ptr) {
- *(vector unsigned long long *)((__INTPTR_TYPE__)__ptr + __offset) =
- __vec;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai void
-vec_xst(vector float __vec, long __offset, float *__ptr) {
- *(vector float *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-#endif
-
-static inline __ATTRS_o_ai void
-vec_xst(vector double __vec, long __offset, double *__ptr) {
- *(vector double *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-/*-- vec_xstd2 --------------------------------------------------------------*/
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai void
-vec_xstd2(vector signed char __vec, long __offset, signed char *__ptr) {
- *(vector signed char *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai void
-vec_xstd2(vector unsigned char __vec, long __offset, unsigned char *__ptr) {
- *(vector unsigned char *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai void
-vec_xstd2(vector signed short __vec, long __offset, signed short *__ptr) {
- *(vector signed short *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai void
-vec_xstd2(vector unsigned short __vec, long __offset, unsigned short *__ptr) {
- *(vector unsigned short *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai void
-vec_xstd2(vector signed int __vec, long __offset, signed int *__ptr) {
- *(vector signed int *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai void
-vec_xstd2(vector unsigned int __vec, long __offset, unsigned int *__ptr) {
- *(vector unsigned int *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai void
-vec_xstd2(vector signed long long __vec, long __offset,
- signed long long *__ptr) {
- *(vector signed long long *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai void
-vec_xstd2(vector unsigned long long __vec, long __offset,
- unsigned long long *__ptr) {
- *(vector unsigned long long *)((__INTPTR_TYPE__)__ptr + __offset) =
- __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai void
-vec_xstd2(vector double __vec, long __offset, double *__ptr) {
- *(vector double *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-/*-- vec_xstw4 --------------------------------------------------------------*/
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai void
-vec_xstw4(vector signed char __vec, long __offset, signed char *__ptr) {
- *(vector signed char *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai void
-vec_xstw4(vector unsigned char __vec, long __offset, unsigned char *__ptr) {
- *(vector unsigned char *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai void
-vec_xstw4(vector signed short __vec, long __offset, signed short *__ptr) {
- *(vector signed short *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai void
-vec_xstw4(vector unsigned short __vec, long __offset, unsigned short *__ptr) {
- *(vector unsigned short *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai void
-vec_xstw4(vector signed int __vec, long __offset, signed int *__ptr) {
- *(vector signed int *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai void
-vec_xstw4(vector unsigned int __vec, long __offset, unsigned int *__ptr) {
- *(vector unsigned int *)((__INTPTR_TYPE__)__ptr + __offset) = __vec;
-}
-
-/*-- vec_load_bndry ---------------------------------------------------------*/
-
-extern __ATTRS_o vector signed char
-vec_load_bndry(const signed char *__ptr, unsigned short __len)
- __constant_pow2_range(__len, 64, 4096);
-
-extern __ATTRS_o vector unsigned char
-vec_load_bndry(const unsigned char *__ptr, unsigned short __len)
- __constant_pow2_range(__len, 64, 4096);
-
-extern __ATTRS_o vector signed short
-vec_load_bndry(const signed short *__ptr, unsigned short __len)
- __constant_pow2_range(__len, 64, 4096);
-
-extern __ATTRS_o vector unsigned short
-vec_load_bndry(const unsigned short *__ptr, unsigned short __len)
- __constant_pow2_range(__len, 64, 4096);
-
-extern __ATTRS_o vector signed int
-vec_load_bndry(const signed int *__ptr, unsigned short __len)
- __constant_pow2_range(__len, 64, 4096);
-
-extern __ATTRS_o vector unsigned int
-vec_load_bndry(const unsigned int *__ptr, unsigned short __len)
- __constant_pow2_range(__len, 64, 4096);
-
-extern __ATTRS_o vector signed long long
-vec_load_bndry(const signed long long *__ptr, unsigned short __len)
- __constant_pow2_range(__len, 64, 4096);
-
-extern __ATTRS_o vector unsigned long long
-vec_load_bndry(const unsigned long long *__ptr, unsigned short __len)
- __constant_pow2_range(__len, 64, 4096);
-
-#if __ARCH__ >= 12
-extern __ATTRS_o vector float
-vec_load_bndry(const float *__ptr, unsigned short __len)
- __constant_pow2_range(__len, 64, 4096);
-#endif
-
-extern __ATTRS_o vector double
-vec_load_bndry(const double *__ptr, unsigned short __len)
- __constant_pow2_range(__len, 64, 4096);
-
-#define vec_load_bndry(X, Y) ((__typeof__((vec_load_bndry)((X), (Y)))) \
- __builtin_s390_vlbb((X), ((Y) == 64 ? 0 : \
- (Y) == 128 ? 1 : \
- (Y) == 256 ? 2 : \
- (Y) == 512 ? 3 : \
- (Y) == 1024 ? 4 : \
- (Y) == 2048 ? 5 : \
- (Y) == 4096 ? 6 : -1)))
-
-/*-- vec_load_len -----------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_load_len(const signed char *__ptr, unsigned int __len) {
- return (vector signed char)__builtin_s390_vll(__len, __ptr);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_load_len(const unsigned char *__ptr, unsigned int __len) {
- return (vector unsigned char)__builtin_s390_vll(__len, __ptr);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_load_len(const signed short *__ptr, unsigned int __len) {
- return (vector signed short)__builtin_s390_vll(__len, __ptr);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_load_len(const unsigned short *__ptr, unsigned int __len) {
- return (vector unsigned short)__builtin_s390_vll(__len, __ptr);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_load_len(const signed int *__ptr, unsigned int __len) {
- return (vector signed int)__builtin_s390_vll(__len, __ptr);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_load_len(const unsigned int *__ptr, unsigned int __len) {
- return (vector unsigned int)__builtin_s390_vll(__len, __ptr);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_load_len(const signed long long *__ptr, unsigned int __len) {
- return (vector signed long long)__builtin_s390_vll(__len, __ptr);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_load_len(const unsigned long long *__ptr, unsigned int __len) {
- return (vector unsigned long long)__builtin_s390_vll(__len, __ptr);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_load_len(const float *__ptr, unsigned int __len) {
- return (vector float)__builtin_s390_vll(__len, __ptr);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_load_len(const double *__ptr, unsigned int __len) {
- return (vector double)__builtin_s390_vll(__len, __ptr);
-}
-
-/*-- vec_load_len_r ---------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_ai vector unsigned char
-vec_load_len_r(const unsigned char *__ptr, unsigned int __len) {
- return (vector unsigned char)__builtin_s390_vlrl(__len, __ptr);
-}
-#endif
-
-/*-- vec_store_len ----------------------------------------------------------*/
-
-static inline __ATTRS_o_ai void
-vec_store_len(vector signed char __vec, signed char *__ptr,
- unsigned int __len) {
- __builtin_s390_vstl((vector signed char)__vec, __len, __ptr);
-}
-
-static inline __ATTRS_o_ai void
-vec_store_len(vector unsigned char __vec, unsigned char *__ptr,
- unsigned int __len) {
- __builtin_s390_vstl((vector signed char)__vec, __len, __ptr);
-}
-
-static inline __ATTRS_o_ai void
-vec_store_len(vector signed short __vec, signed short *__ptr,
- unsigned int __len) {
- __builtin_s390_vstl((vector signed char)__vec, __len, __ptr);
-}
-
-static inline __ATTRS_o_ai void
-vec_store_len(vector unsigned short __vec, unsigned short *__ptr,
- unsigned int __len) {
- __builtin_s390_vstl((vector signed char)__vec, __len, __ptr);
-}
-
-static inline __ATTRS_o_ai void
-vec_store_len(vector signed int __vec, signed int *__ptr,
- unsigned int __len) {
- __builtin_s390_vstl((vector signed char)__vec, __len, __ptr);
-}
-
-static inline __ATTRS_o_ai void
-vec_store_len(vector unsigned int __vec, unsigned int *__ptr,
- unsigned int __len) {
- __builtin_s390_vstl((vector signed char)__vec, __len, __ptr);
-}
-
-static inline __ATTRS_o_ai void
-vec_store_len(vector signed long long __vec, signed long long *__ptr,
- unsigned int __len) {
- __builtin_s390_vstl((vector signed char)__vec, __len, __ptr);
-}
-
-static inline __ATTRS_o_ai void
-vec_store_len(vector unsigned long long __vec, unsigned long long *__ptr,
- unsigned int __len) {
- __builtin_s390_vstl((vector signed char)__vec, __len, __ptr);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai void
-vec_store_len(vector float __vec, float *__ptr,
- unsigned int __len) {
- __builtin_s390_vstl((vector signed char)__vec, __len, __ptr);
-}
-#endif
-
-static inline __ATTRS_o_ai void
-vec_store_len(vector double __vec, double *__ptr,
- unsigned int __len) {
- __builtin_s390_vstl((vector signed char)__vec, __len, __ptr);
-}
-
-/*-- vec_store_len_r --------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_ai void
-vec_store_len_r(vector unsigned char __vec, unsigned char *__ptr,
- unsigned int __len) {
- __builtin_s390_vstrl((vector signed char)__vec, __len, __ptr);
-}
-#endif
-
-/*-- vec_load_pair ----------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed long long
-vec_load_pair(signed long long __a, signed long long __b) {
- return (vector signed long long)(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_load_pair(unsigned long long __a, unsigned long long __b) {
- return (vector unsigned long long)(__a, __b);
-}
-
-/*-- vec_genmask ------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_genmask(unsigned short __mask)
- __constant(__mask) {
- return (vector unsigned char)(
- __mask & 0x8000 ? 0xff : 0,
- __mask & 0x4000 ? 0xff : 0,
- __mask & 0x2000 ? 0xff : 0,
- __mask & 0x1000 ? 0xff : 0,
- __mask & 0x0800 ? 0xff : 0,
- __mask & 0x0400 ? 0xff : 0,
- __mask & 0x0200 ? 0xff : 0,
- __mask & 0x0100 ? 0xff : 0,
- __mask & 0x0080 ? 0xff : 0,
- __mask & 0x0040 ? 0xff : 0,
- __mask & 0x0020 ? 0xff : 0,
- __mask & 0x0010 ? 0xff : 0,
- __mask & 0x0008 ? 0xff : 0,
- __mask & 0x0004 ? 0xff : 0,
- __mask & 0x0002 ? 0xff : 0,
- __mask & 0x0001 ? 0xff : 0);
-}
-
-/*-- vec_genmasks_* ---------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_genmasks_8(unsigned char __first, unsigned char __last)
- __constant(__first) __constant(__last) {
- unsigned char __bit1 = __first & 7;
- unsigned char __bit2 = __last & 7;
- unsigned char __mask1 = (unsigned char)(1U << (7 - __bit1) << 1) - 1;
- unsigned char __mask2 = (unsigned char)(1U << (7 - __bit2)) - 1;
- unsigned char __value = (__bit1 <= __bit2 ?
- __mask1 & ~__mask2 :
- __mask1 | ~__mask2);
- return (vector unsigned char)__value;
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_genmasks_16(unsigned char __first, unsigned char __last)
- __constant(__first) __constant(__last) {
- unsigned char __bit1 = __first & 15;
- unsigned char __bit2 = __last & 15;
- unsigned short __mask1 = (unsigned short)(1U << (15 - __bit1) << 1) - 1;
- unsigned short __mask2 = (unsigned short)(1U << (15 - __bit2)) - 1;
- unsigned short __value = (__bit1 <= __bit2 ?
- __mask1 & ~__mask2 :
- __mask1 | ~__mask2);
- return (vector unsigned short)__value;
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_genmasks_32(unsigned char __first, unsigned char __last)
- __constant(__first) __constant(__last) {
- unsigned char __bit1 = __first & 31;
- unsigned char __bit2 = __last & 31;
- unsigned int __mask1 = (1U << (31 - __bit1) << 1) - 1;
- unsigned int __mask2 = (1U << (31 - __bit2)) - 1;
- unsigned int __value = (__bit1 <= __bit2 ?
- __mask1 & ~__mask2 :
- __mask1 | ~__mask2);
- return (vector unsigned int)__value;
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_genmasks_64(unsigned char __first, unsigned char __last)
- __constant(__first) __constant(__last) {
- unsigned char __bit1 = __first & 63;
- unsigned char __bit2 = __last & 63;
- unsigned long long __mask1 = (1ULL << (63 - __bit1) << 1) - 1;
- unsigned long long __mask2 = (1ULL << (63 - __bit2)) - 1;
- unsigned long long __value = (__bit1 <= __bit2 ?
- __mask1 & ~__mask2 :
- __mask1 | ~__mask2);
- return (vector unsigned long long)__value;
-}
-
-/*-- vec_splat --------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_splat(vector signed char __vec, int __index)
- __constant_range(__index, 0, 15) {
- return (vector signed char)__vec[__index];
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_splat(vector bool char __vec, int __index)
- __constant_range(__index, 0, 15) {
- return (vector bool char)(vector unsigned char)__vec[__index];
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_splat(vector unsigned char __vec, int __index)
- __constant_range(__index, 0, 15) {
- return (vector unsigned char)__vec[__index];
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_splat(vector signed short __vec, int __index)
- __constant_range(__index, 0, 7) {
- return (vector signed short)__vec[__index];
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_splat(vector bool short __vec, int __index)
- __constant_range(__index, 0, 7) {
- return (vector bool short)(vector unsigned short)__vec[__index];
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_splat(vector unsigned short __vec, int __index)
- __constant_range(__index, 0, 7) {
- return (vector unsigned short)__vec[__index];
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_splat(vector signed int __vec, int __index)
- __constant_range(__index, 0, 3) {
- return (vector signed int)__vec[__index];
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_splat(vector bool int __vec, int __index)
- __constant_range(__index, 0, 3) {
- return (vector bool int)(vector unsigned int)__vec[__index];
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_splat(vector unsigned int __vec, int __index)
- __constant_range(__index, 0, 3) {
- return (vector unsigned int)__vec[__index];
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_splat(vector signed long long __vec, int __index)
- __constant_range(__index, 0, 1) {
- return (vector signed long long)__vec[__index];
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_splat(vector bool long long __vec, int __index)
- __constant_range(__index, 0, 1) {
- return (vector bool long long)(vector unsigned long long)__vec[__index];
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_splat(vector unsigned long long __vec, int __index)
- __constant_range(__index, 0, 1) {
- return (vector unsigned long long)__vec[__index];
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_splat(vector float __vec, int __index)
- __constant_range(__index, 0, 3) {
- return (vector float)__vec[__index];
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_splat(vector double __vec, int __index)
- __constant_range(__index, 0, 1) {
- return (vector double)__vec[__index];
-}
-
-/*-- vec_splat_s* -----------------------------------------------------------*/
-
-static inline __ATTRS_ai vector signed char
-vec_splat_s8(signed char __scalar)
- __constant(__scalar) {
- return (vector signed char)__scalar;
-}
-
-static inline __ATTRS_ai vector signed short
-vec_splat_s16(signed short __scalar)
- __constant(__scalar) {
- return (vector signed short)__scalar;
-}
-
-static inline __ATTRS_ai vector signed int
-vec_splat_s32(signed short __scalar)
- __constant(__scalar) {
- return (vector signed int)(signed int)__scalar;
-}
-
-static inline __ATTRS_ai vector signed long long
-vec_splat_s64(signed short __scalar)
- __constant(__scalar) {
- return (vector signed long long)(signed long)__scalar;
-}
-
-/*-- vec_splat_u* -----------------------------------------------------------*/
-
-static inline __ATTRS_ai vector unsigned char
-vec_splat_u8(unsigned char __scalar)
- __constant(__scalar) {
- return (vector unsigned char)__scalar;
-}
-
-static inline __ATTRS_ai vector unsigned short
-vec_splat_u16(unsigned short __scalar)
- __constant(__scalar) {
- return (vector unsigned short)__scalar;
-}
-
-static inline __ATTRS_ai vector unsigned int
-vec_splat_u32(signed short __scalar)
- __constant(__scalar) {
- return (vector unsigned int)(signed int)__scalar;
-}
-
-static inline __ATTRS_ai vector unsigned long long
-vec_splat_u64(signed short __scalar)
- __constant(__scalar) {
- return (vector unsigned long long)(signed long long)__scalar;
-}
-
-/*-- vec_splats -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_splats(signed char __scalar) {
- return (vector signed char)__scalar;
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_splats(unsigned char __scalar) {
- return (vector unsigned char)__scalar;
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_splats(signed short __scalar) {
- return (vector signed short)__scalar;
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_splats(unsigned short __scalar) {
- return (vector unsigned short)__scalar;
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_splats(signed int __scalar) {
- return (vector signed int)__scalar;
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_splats(unsigned int __scalar) {
- return (vector unsigned int)__scalar;
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_splats(signed long long __scalar) {
- return (vector signed long long)__scalar;
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_splats(unsigned long long __scalar) {
- return (vector unsigned long long)__scalar;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_splats(float __scalar) {
- return (vector float)__scalar;
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_splats(double __scalar) {
- return (vector double)__scalar;
-}
-
-/*-- vec_extend_s64 ---------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed long long
-vec_extend_s64(vector signed char __a) {
- return (vector signed long long)(__a[7], __a[15]);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_extend_s64(vector signed short __a) {
- return (vector signed long long)(__a[3], __a[7]);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_extend_s64(vector signed int __a) {
- return (vector signed long long)(__a[1], __a[3]);
-}
-
-/*-- vec_mergeh -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_mergeh(vector signed char __a, vector signed char __b) {
- return (vector signed char)(
- __a[0], __b[0], __a[1], __b[1], __a[2], __b[2], __a[3], __b[3],
- __a[4], __b[4], __a[5], __b[5], __a[6], __b[6], __a[7], __b[7]);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_mergeh(vector bool char __a, vector bool char __b) {
- return (vector bool char)(
- __a[0], __b[0], __a[1], __b[1], __a[2], __b[2], __a[3], __b[3],
- __a[4], __b[4], __a[5], __b[5], __a[6], __b[6], __a[7], __b[7]);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_mergeh(vector unsigned char __a, vector unsigned char __b) {
- return (vector unsigned char)(
- __a[0], __b[0], __a[1], __b[1], __a[2], __b[2], __a[3], __b[3],
- __a[4], __b[4], __a[5], __b[5], __a[6], __b[6], __a[7], __b[7]);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_mergeh(vector signed short __a, vector signed short __b) {
- return (vector signed short)(
- __a[0], __b[0], __a[1], __b[1], __a[2], __b[2], __a[3], __b[3]);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_mergeh(vector bool short __a, vector bool short __b) {
- return (vector bool short)(
- __a[0], __b[0], __a[1], __b[1], __a[2], __b[2], __a[3], __b[3]);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_mergeh(vector unsigned short __a, vector unsigned short __b) {
- return (vector unsigned short)(
- __a[0], __b[0], __a[1], __b[1], __a[2], __b[2], __a[3], __b[3]);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_mergeh(vector signed int __a, vector signed int __b) {
- return (vector signed int)(__a[0], __b[0], __a[1], __b[1]);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_mergeh(vector bool int __a, vector bool int __b) {
- return (vector bool int)(__a[0], __b[0], __a[1], __b[1]);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_mergeh(vector unsigned int __a, vector unsigned int __b) {
- return (vector unsigned int)(__a[0], __b[0], __a[1], __b[1]);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_mergeh(vector signed long long __a, vector signed long long __b) {
- return (vector signed long long)(__a[0], __b[0]);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_mergeh(vector bool long long __a, vector bool long long __b) {
- return (vector bool long long)(__a[0], __b[0]);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_mergeh(vector unsigned long long __a, vector unsigned long long __b) {
- return (vector unsigned long long)(__a[0], __b[0]);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_mergeh(vector float __a, vector float __b) {
- return (vector float)(__a[0], __b[0], __a[1], __b[1]);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_mergeh(vector double __a, vector double __b) {
- return (vector double)(__a[0], __b[0]);
-}
-
-/*-- vec_mergel -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_mergel(vector signed char __a, vector signed char __b) {
- return (vector signed char)(
- __a[8], __b[8], __a[9], __b[9], __a[10], __b[10], __a[11], __b[11],
- __a[12], __b[12], __a[13], __b[13], __a[14], __b[14], __a[15], __b[15]);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_mergel(vector bool char __a, vector bool char __b) {
- return (vector bool char)(
- __a[8], __b[8], __a[9], __b[9], __a[10], __b[10], __a[11], __b[11],
- __a[12], __b[12], __a[13], __b[13], __a[14], __b[14], __a[15], __b[15]);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_mergel(vector unsigned char __a, vector unsigned char __b) {
- return (vector unsigned char)(
- __a[8], __b[8], __a[9], __b[9], __a[10], __b[10], __a[11], __b[11],
- __a[12], __b[12], __a[13], __b[13], __a[14], __b[14], __a[15], __b[15]);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_mergel(vector signed short __a, vector signed short __b) {
- return (vector signed short)(
- __a[4], __b[4], __a[5], __b[5], __a[6], __b[6], __a[7], __b[7]);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_mergel(vector bool short __a, vector bool short __b) {
- return (vector bool short)(
- __a[4], __b[4], __a[5], __b[5], __a[6], __b[6], __a[7], __b[7]);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_mergel(vector unsigned short __a, vector unsigned short __b) {
- return (vector unsigned short)(
- __a[4], __b[4], __a[5], __b[5], __a[6], __b[6], __a[7], __b[7]);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_mergel(vector signed int __a, vector signed int __b) {
- return (vector signed int)(__a[2], __b[2], __a[3], __b[3]);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_mergel(vector bool int __a, vector bool int __b) {
- return (vector bool int)(__a[2], __b[2], __a[3], __b[3]);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_mergel(vector unsigned int __a, vector unsigned int __b) {
- return (vector unsigned int)(__a[2], __b[2], __a[3], __b[3]);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_mergel(vector signed long long __a, vector signed long long __b) {
- return (vector signed long long)(__a[1], __b[1]);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_mergel(vector bool long long __a, vector bool long long __b) {
- return (vector bool long long)(__a[1], __b[1]);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_mergel(vector unsigned long long __a, vector unsigned long long __b) {
- return (vector unsigned long long)(__a[1], __b[1]);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_mergel(vector float __a, vector float __b) {
- return (vector float)(__a[2], __b[2], __a[3], __b[3]);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_mergel(vector double __a, vector double __b) {
- return (vector double)(__a[1], __b[1]);
-}
-
-/*-- vec_pack ---------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_pack(vector signed short __a, vector signed short __b) {
- vector signed char __ac = (vector signed char)__a;
- vector signed char __bc = (vector signed char)__b;
- return (vector signed char)(
- __ac[1], __ac[3], __ac[5], __ac[7], __ac[9], __ac[11], __ac[13], __ac[15],
- __bc[1], __bc[3], __bc[5], __bc[7], __bc[9], __bc[11], __bc[13], __bc[15]);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_pack(vector bool short __a, vector bool short __b) {
- vector bool char __ac = (vector bool char)__a;
- vector bool char __bc = (vector bool char)__b;
- return (vector bool char)(
- __ac[1], __ac[3], __ac[5], __ac[7], __ac[9], __ac[11], __ac[13], __ac[15],
- __bc[1], __bc[3], __bc[5], __bc[7], __bc[9], __bc[11], __bc[13], __bc[15]);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_pack(vector unsigned short __a, vector unsigned short __b) {
- vector unsigned char __ac = (vector unsigned char)__a;
- vector unsigned char __bc = (vector unsigned char)__b;
- return (vector unsigned char)(
- __ac[1], __ac[3], __ac[5], __ac[7], __ac[9], __ac[11], __ac[13], __ac[15],
- __bc[1], __bc[3], __bc[5], __bc[7], __bc[9], __bc[11], __bc[13], __bc[15]);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_pack(vector signed int __a, vector signed int __b) {
- vector signed short __ac = (vector signed short)__a;
- vector signed short __bc = (vector signed short)__b;
- return (vector signed short)(
- __ac[1], __ac[3], __ac[5], __ac[7],
- __bc[1], __bc[3], __bc[5], __bc[7]);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_pack(vector bool int __a, vector bool int __b) {
- vector bool short __ac = (vector bool short)__a;
- vector bool short __bc = (vector bool short)__b;
- return (vector bool short)(
- __ac[1], __ac[3], __ac[5], __ac[7],
- __bc[1], __bc[3], __bc[5], __bc[7]);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_pack(vector unsigned int __a, vector unsigned int __b) {
- vector unsigned short __ac = (vector unsigned short)__a;
- vector unsigned short __bc = (vector unsigned short)__b;
- return (vector unsigned short)(
- __ac[1], __ac[3], __ac[5], __ac[7],
- __bc[1], __bc[3], __bc[5], __bc[7]);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_pack(vector signed long long __a, vector signed long long __b) {
- vector signed int __ac = (vector signed int)__a;
- vector signed int __bc = (vector signed int)__b;
- return (vector signed int)(__ac[1], __ac[3], __bc[1], __bc[3]);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_pack(vector bool long long __a, vector bool long long __b) {
- vector bool int __ac = (vector bool int)__a;
- vector bool int __bc = (vector bool int)__b;
- return (vector bool int)(__ac[1], __ac[3], __bc[1], __bc[3]);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_pack(vector unsigned long long __a, vector unsigned long long __b) {
- vector unsigned int __ac = (vector unsigned int)__a;
- vector unsigned int __bc = (vector unsigned int)__b;
- return (vector unsigned int)(__ac[1], __ac[3], __bc[1], __bc[3]);
-}
-
-/*-- vec_packs --------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_packs(vector signed short __a, vector signed short __b) {
- return __builtin_s390_vpksh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_packs(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vpklsh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_packs(vector signed int __a, vector signed int __b) {
- return __builtin_s390_vpksf(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_packs(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vpklsf(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_packs(vector signed long long __a, vector signed long long __b) {
- return __builtin_s390_vpksg(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_packs(vector unsigned long long __a, vector unsigned long long __b) {
- return __builtin_s390_vpklsg(__a, __b);
-}
-
-/*-- vec_packs_cc -----------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_packs_cc(vector signed short __a, vector signed short __b, int *__cc) {
- return __builtin_s390_vpkshs(__a, __b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_packs_cc(vector unsigned short __a, vector unsigned short __b, int *__cc) {
- return __builtin_s390_vpklshs(__a, __b, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_packs_cc(vector signed int __a, vector signed int __b, int *__cc) {
- return __builtin_s390_vpksfs(__a, __b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_packs_cc(vector unsigned int __a, vector unsigned int __b, int *__cc) {
- return __builtin_s390_vpklsfs(__a, __b, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_packs_cc(vector signed long long __a, vector signed long long __b,
- int *__cc) {
- return __builtin_s390_vpksgs(__a, __b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_packs_cc(vector unsigned long long __a, vector unsigned long long __b,
- int *__cc) {
- return __builtin_s390_vpklsgs(__a, __b, __cc);
-}
-
-/*-- vec_packsu -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_packsu(vector signed short __a, vector signed short __b) {
- const vector signed short __zero = (vector signed short)0;
- return __builtin_s390_vpklsh(
- (vector unsigned short)(__a >= __zero) & (vector unsigned short)__a,
- (vector unsigned short)(__b >= __zero) & (vector unsigned short)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_packsu(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vpklsh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_packsu(vector signed int __a, vector signed int __b) {
- const vector signed int __zero = (vector signed int)0;
- return __builtin_s390_vpklsf(
- (vector unsigned int)(__a >= __zero) & (vector unsigned int)__a,
- (vector unsigned int)(__b >= __zero) & (vector unsigned int)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_packsu(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vpklsf(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_packsu(vector signed long long __a, vector signed long long __b) {
- const vector signed long long __zero = (vector signed long long)0;
- return __builtin_s390_vpklsg(
- (vector unsigned long long)(__a >= __zero) &
- (vector unsigned long long)__a,
- (vector unsigned long long)(__b >= __zero) &
- (vector unsigned long long)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_packsu(vector unsigned long long __a, vector unsigned long long __b) {
- return __builtin_s390_vpklsg(__a, __b);
-}
-
-/*-- vec_packsu_cc ----------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_packsu_cc(vector unsigned short __a, vector unsigned short __b, int *__cc) {
- return __builtin_s390_vpklshs(__a, __b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_packsu_cc(vector unsigned int __a, vector unsigned int __b, int *__cc) {
- return __builtin_s390_vpklsfs(__a, __b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_packsu_cc(vector unsigned long long __a, vector unsigned long long __b,
- int *__cc) {
- return __builtin_s390_vpklsgs(__a, __b, __cc);
-}
-
-/*-- vec_unpackh ------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed short
-vec_unpackh(vector signed char __a) {
- return __builtin_s390_vuphb(__a);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_unpackh(vector bool char __a) {
- return (vector bool short)__builtin_s390_vuphb((vector signed char)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_unpackh(vector unsigned char __a) {
- return __builtin_s390_vuplhb(__a);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_unpackh(vector signed short __a) {
- return __builtin_s390_vuphh(__a);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_unpackh(vector bool short __a) {
- return (vector bool int)__builtin_s390_vuphh((vector signed short)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_unpackh(vector unsigned short __a) {
- return __builtin_s390_vuplhh(__a);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_unpackh(vector signed int __a) {
- return __builtin_s390_vuphf(__a);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_unpackh(vector bool int __a) {
- return (vector bool long long)__builtin_s390_vuphf((vector signed int)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_unpackh(vector unsigned int __a) {
- return __builtin_s390_vuplhf(__a);
-}
-
-/*-- vec_unpackl ------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed short
-vec_unpackl(vector signed char __a) {
- return __builtin_s390_vuplb(__a);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_unpackl(vector bool char __a) {
- return (vector bool short)__builtin_s390_vuplb((vector signed char)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_unpackl(vector unsigned char __a) {
- return __builtin_s390_vupllb(__a);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_unpackl(vector signed short __a) {
- return __builtin_s390_vuplhw(__a);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_unpackl(vector bool short __a) {
- return (vector bool int)__builtin_s390_vuplhw((vector signed short)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_unpackl(vector unsigned short __a) {
- return __builtin_s390_vupllh(__a);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_unpackl(vector signed int __a) {
- return __builtin_s390_vuplf(__a);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_unpackl(vector bool int __a) {
- return (vector bool long long)__builtin_s390_vuplf((vector signed int)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_unpackl(vector unsigned int __a) {
- return __builtin_s390_vupllf(__a);
-}
-
-/*-- vec_cmpeq --------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector bool char
-vec_cmpeq(vector bool char __a, vector bool char __b) {
- return (vector bool char)(__a == __b);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_cmpeq(vector signed char __a, vector signed char __b) {
- return (vector bool char)(__a == __b);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_cmpeq(vector unsigned char __a, vector unsigned char __b) {
- return (vector bool char)(__a == __b);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cmpeq(vector bool short __a, vector bool short __b) {
- return (vector bool short)(__a == __b);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cmpeq(vector signed short __a, vector signed short __b) {
- return (vector bool short)(__a == __b);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cmpeq(vector unsigned short __a, vector unsigned short __b) {
- return (vector bool short)(__a == __b);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cmpeq(vector bool int __a, vector bool int __b) {
- return (vector bool int)(__a == __b);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cmpeq(vector signed int __a, vector signed int __b) {
- return (vector bool int)(__a == __b);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cmpeq(vector unsigned int __a, vector unsigned int __b) {
- return (vector bool int)(__a == __b);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmpeq(vector bool long long __a, vector bool long long __b) {
- return (vector bool long long)(__a == __b);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmpeq(vector signed long long __a, vector signed long long __b) {
- return (vector bool long long)(__a == __b);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmpeq(vector unsigned long long __a, vector unsigned long long __b) {
- return (vector bool long long)(__a == __b);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector bool int
-vec_cmpeq(vector float __a, vector float __b) {
- return (vector bool int)(__a == __b);
-}
-#endif
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmpeq(vector double __a, vector double __b) {
- return (vector bool long long)(__a == __b);
-}
-
-/*-- vec_cmpge --------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector bool char
-vec_cmpge(vector signed char __a, vector signed char __b) {
- return (vector bool char)(__a >= __b);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_cmpge(vector unsigned char __a, vector unsigned char __b) {
- return (vector bool char)(__a >= __b);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cmpge(vector signed short __a, vector signed short __b) {
- return (vector bool short)(__a >= __b);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cmpge(vector unsigned short __a, vector unsigned short __b) {
- return (vector bool short)(__a >= __b);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cmpge(vector signed int __a, vector signed int __b) {
- return (vector bool int)(__a >= __b);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cmpge(vector unsigned int __a, vector unsigned int __b) {
- return (vector bool int)(__a >= __b);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmpge(vector signed long long __a, vector signed long long __b) {
- return (vector bool long long)(__a >= __b);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmpge(vector unsigned long long __a, vector unsigned long long __b) {
- return (vector bool long long)(__a >= __b);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector bool int
-vec_cmpge(vector float __a, vector float __b) {
- return (vector bool int)(__a >= __b);
-}
-#endif
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmpge(vector double __a, vector double __b) {
- return (vector bool long long)(__a >= __b);
-}
-
-/*-- vec_cmpgt --------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector bool char
-vec_cmpgt(vector signed char __a, vector signed char __b) {
- return (vector bool char)(__a > __b);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_cmpgt(vector unsigned char __a, vector unsigned char __b) {
- return (vector bool char)(__a > __b);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cmpgt(vector signed short __a, vector signed short __b) {
- return (vector bool short)(__a > __b);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cmpgt(vector unsigned short __a, vector unsigned short __b) {
- return (vector bool short)(__a > __b);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cmpgt(vector signed int __a, vector signed int __b) {
- return (vector bool int)(__a > __b);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cmpgt(vector unsigned int __a, vector unsigned int __b) {
- return (vector bool int)(__a > __b);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmpgt(vector signed long long __a, vector signed long long __b) {
- return (vector bool long long)(__a > __b);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmpgt(vector unsigned long long __a, vector unsigned long long __b) {
- return (vector bool long long)(__a > __b);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector bool int
-vec_cmpgt(vector float __a, vector float __b) {
- return (vector bool int)(__a > __b);
-}
-#endif
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmpgt(vector double __a, vector double __b) {
- return (vector bool long long)(__a > __b);
-}
-
-/*-- vec_cmple --------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector bool char
-vec_cmple(vector signed char __a, vector signed char __b) {
- return (vector bool char)(__a <= __b);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_cmple(vector unsigned char __a, vector unsigned char __b) {
- return (vector bool char)(__a <= __b);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cmple(vector signed short __a, vector signed short __b) {
- return (vector bool short)(__a <= __b);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cmple(vector unsigned short __a, vector unsigned short __b) {
- return (vector bool short)(__a <= __b);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cmple(vector signed int __a, vector signed int __b) {
- return (vector bool int)(__a <= __b);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cmple(vector unsigned int __a, vector unsigned int __b) {
- return (vector bool int)(__a <= __b);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmple(vector signed long long __a, vector signed long long __b) {
- return (vector bool long long)(__a <= __b);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmple(vector unsigned long long __a, vector unsigned long long __b) {
- return (vector bool long long)(__a <= __b);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector bool int
-vec_cmple(vector float __a, vector float __b) {
- return (vector bool int)(__a <= __b);
-}
-#endif
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmple(vector double __a, vector double __b) {
- return (vector bool long long)(__a <= __b);
-}
-
-/*-- vec_cmplt --------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector bool char
-vec_cmplt(vector signed char __a, vector signed char __b) {
- return (vector bool char)(__a < __b);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_cmplt(vector unsigned char __a, vector unsigned char __b) {
- return (vector bool char)(__a < __b);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cmplt(vector signed short __a, vector signed short __b) {
- return (vector bool short)(__a < __b);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cmplt(vector unsigned short __a, vector unsigned short __b) {
- return (vector bool short)(__a < __b);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cmplt(vector signed int __a, vector signed int __b) {
- return (vector bool int)(__a < __b);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cmplt(vector unsigned int __a, vector unsigned int __b) {
- return (vector bool int)(__a < __b);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmplt(vector signed long long __a, vector signed long long __b) {
- return (vector bool long long)(__a < __b);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmplt(vector unsigned long long __a, vector unsigned long long __b) {
- return (vector bool long long)(__a < __b);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector bool int
-vec_cmplt(vector float __a, vector float __b) {
- return (vector bool int)(__a < __b);
-}
-#endif
-
-static inline __ATTRS_o_ai vector bool long long
-vec_cmplt(vector double __a, vector double __b) {
- return (vector bool long long)(__a < __b);
-}
-
-/*-- vec_all_eq -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai int
-vec_all_eq(vector signed char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vceqbs(__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector signed char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vceqbs(__a, (vector signed char)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector bool char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a, __b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_eq(vector unsigned char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector unsigned char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector bool char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_eq(vector bool char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_eq(vector signed short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vceqhs(__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector signed short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vceqhs(__a, (vector signed short)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector bool short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a, __b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_eq(vector unsigned short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector unsigned short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector bool short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_eq(vector bool short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_eq(vector signed int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vceqfs(__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector signed int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vceqfs(__a, (vector signed int)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector bool int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a, __b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_eq(vector unsigned int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector unsigned int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector bool int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_eq(vector bool int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_eq(vector signed long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vceqgs(__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector signed long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vceqgs(__a, (vector signed long long)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector bool long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a, __b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_eq(vector unsigned long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector unsigned long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_eq(vector bool long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_eq(vector bool long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc == 0;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_all_eq(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfcesbs(__a, __b, &__cc);
- return __cc == 0;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_all_eq(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfcedbs(__a, __b, &__cc);
- return __cc == 0;
-}
-
-/*-- vec_all_ne -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai int
-vec_all_ne(vector signed char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vceqbs(__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector signed char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vceqbs(__a, (vector signed char)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector bool char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a, __b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ne(vector unsigned char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector unsigned char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector bool char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ne(vector bool char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ne(vector signed short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vceqhs(__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector signed short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vceqhs(__a, (vector signed short)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector bool short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a, __b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ne(vector unsigned short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector unsigned short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector bool short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ne(vector bool short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ne(vector signed int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vceqfs(__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector signed int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vceqfs(__a, (vector signed int)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector bool int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a, __b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ne(vector unsigned int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector unsigned int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector bool int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ne(vector bool int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ne(vector signed long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vceqgs(__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector signed long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vceqgs(__a, (vector signed long long)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector bool long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a, __b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ne(vector unsigned long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector unsigned long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ne(vector bool long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ne(vector bool long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc == 3;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_all_ne(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfcesbs(__a, __b, &__cc);
- return __cc == 3;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_all_ne(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfcedbs(__a, __b, &__cc);
- return __cc == 3;
-}
-
-/*-- vec_all_ge -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai int
-vec_all_ge(vector signed char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs(__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector signed char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchbs((vector signed char)__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector bool char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs(__b, (vector signed char)__a, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ge(vector unsigned char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs(__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector unsigned char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector bool char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs(__b, (vector unsigned char)__a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector bool char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__b,
- (vector unsigned char)__a, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ge(vector signed short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs(__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector signed short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchhs((vector signed short)__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector bool short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs(__b, (vector signed short)__a, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ge(vector unsigned short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs(__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector unsigned short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector bool short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs(__b, (vector unsigned short)__a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector bool short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__b,
- (vector unsigned short)__a, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ge(vector signed int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs(__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector signed int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchfs((vector signed int)__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector bool int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs(__b, (vector signed int)__a, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ge(vector unsigned int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs(__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector unsigned int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector bool int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs(__b, (vector unsigned int)__a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector bool int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__b,
- (vector unsigned int)__a, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ge(vector signed long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs(__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector signed long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchgs((vector signed long long)__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector bool long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs(__b, (vector signed long long)__a, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_ge(vector unsigned long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector unsigned long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__b, __a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector bool long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__b, (vector unsigned long long)__a, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_ge(vector bool long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__b,
- (vector unsigned long long)__a, &__cc);
- return __cc == 3;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_all_ge(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchesbs(__a, __b, &__cc);
- return __cc == 0;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_all_ge(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchedbs(__a, __b, &__cc);
- return __cc == 0;
-}
-
-/*-- vec_all_gt -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai int
-vec_all_gt(vector signed char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs(__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector signed char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchbs(__a, (vector signed char)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector bool char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs((vector signed char)__a, __b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_gt(vector unsigned char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs(__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector unsigned char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs(__a, (vector unsigned char)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector bool char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector bool char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__a,
- (vector unsigned char)__b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_gt(vector signed short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs(__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector signed short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchhs(__a, (vector signed short)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector bool short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs((vector signed short)__a, __b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_gt(vector unsigned short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs(__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector unsigned short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs(__a, (vector unsigned short)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector bool short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector bool short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__a,
- (vector unsigned short)__b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_gt(vector signed int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs(__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector signed int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchfs(__a, (vector signed int)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector bool int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs((vector signed int)__a, __b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_gt(vector unsigned int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs(__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector unsigned int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs(__a, (vector unsigned int)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector bool int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector bool int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__a,
- (vector unsigned int)__b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_gt(vector signed long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs(__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector signed long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchgs(__a, (vector signed long long)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector bool long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs((vector signed long long)__a, __b, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_gt(vector unsigned long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector unsigned long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__a, (vector unsigned long long)__b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector bool long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__a, __b, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_gt(vector bool long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__a,
- (vector unsigned long long)__b, &__cc);
- return __cc == 0;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_all_gt(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchsbs(__a, __b, &__cc);
- return __cc == 0;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_all_gt(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchdbs(__a, __b, &__cc);
- return __cc == 0;
-}
-
-/*-- vec_all_le -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai int
-vec_all_le(vector signed char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs(__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector signed char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchbs(__a, (vector signed char)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector bool char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs((vector signed char)__a, __b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_le(vector unsigned char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs(__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector unsigned char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs(__a, (vector unsigned char)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector bool char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector bool char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__a,
- (vector unsigned char)__b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_le(vector signed short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs(__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector signed short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchhs(__a, (vector signed short)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector bool short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs((vector signed short)__a, __b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_le(vector unsigned short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs(__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector unsigned short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs(__a, (vector unsigned short)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector bool short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector bool short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__a,
- (vector unsigned short)__b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_le(vector signed int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs(__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector signed int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchfs(__a, (vector signed int)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector bool int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs((vector signed int)__a, __b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_le(vector unsigned int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs(__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector unsigned int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs(__a, (vector unsigned int)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector bool int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector bool int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__a,
- (vector unsigned int)__b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_le(vector signed long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs(__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector signed long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchgs(__a, (vector signed long long)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector bool long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs((vector signed long long)__a, __b, &__cc);
- return __cc == 3;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_le(vector unsigned long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector unsigned long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__a, (vector unsigned long long)__b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector bool long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__a, __b, &__cc);
- return __cc == 3;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_le(vector bool long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__a,
- (vector unsigned long long)__b, &__cc);
- return __cc == 3;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_all_le(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchesbs(__b, __a, &__cc);
- return __cc == 0;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_all_le(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchedbs(__b, __a, &__cc);
- return __cc == 0;
-}
-
-/*-- vec_all_lt -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai int
-vec_all_lt(vector signed char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs(__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector signed char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchbs((vector signed char)__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector bool char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs(__b, (vector signed char)__a, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_lt(vector unsigned char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs(__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector unsigned char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector bool char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs(__b, (vector unsigned char)__a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector bool char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__b,
- (vector unsigned char)__a, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_lt(vector signed short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs(__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector signed short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchhs((vector signed short)__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector bool short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs(__b, (vector signed short)__a, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_lt(vector unsigned short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs(__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector unsigned short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector bool short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs(__b, (vector unsigned short)__a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector bool short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__b,
- (vector unsigned short)__a, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_lt(vector signed int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs(__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector signed int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchfs((vector signed int)__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector bool int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs(__b, (vector signed int)__a, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_lt(vector unsigned int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs(__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector unsigned int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector bool int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs(__b, (vector unsigned int)__a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector bool int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__b,
- (vector unsigned int)__a, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_lt(vector signed long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs(__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector signed long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchgs((vector signed long long)__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector bool long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs(__b, (vector signed long long)__a, &__cc);
- return __cc == 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_all_lt(vector unsigned long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector unsigned long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__b, __a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector bool long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__b, (vector unsigned long long)__a, &__cc);
- return __cc == 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_all_lt(vector bool long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__b,
- (vector unsigned long long)__a, &__cc);
- return __cc == 0;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_all_lt(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchsbs(__b, __a, &__cc);
- return __cc == 0;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_all_lt(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchdbs(__b, __a, &__cc);
- return __cc == 0;
-}
-
-/*-- vec_all_nge ------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_all_nge(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchesbs(__a, __b, &__cc);
- return __cc == 3;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_all_nge(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchedbs(__a, __b, &__cc);
- return __cc == 3;
-}
-
-/*-- vec_all_ngt ------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_all_ngt(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchsbs(__a, __b, &__cc);
- return __cc == 3;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_all_ngt(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchdbs(__a, __b, &__cc);
- return __cc == 3;
-}
-
-/*-- vec_all_nle ------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_all_nle(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchesbs(__b, __a, &__cc);
- return __cc == 3;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_all_nle(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchedbs(__b, __a, &__cc);
- return __cc == 3;
-}
-
-/*-- vec_all_nlt ------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_all_nlt(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchsbs(__b, __a, &__cc);
- return __cc == 3;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_all_nlt(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchdbs(__b, __a, &__cc);
- return __cc == 3;
-}
-
-/*-- vec_all_nan ------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_all_nan(vector float __a) {
- int __cc;
- __builtin_s390_vftcisb(__a, 15, &__cc);
- return __cc == 0;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_all_nan(vector double __a) {
- int __cc;
- __builtin_s390_vftcidb(__a, 15, &__cc);
- return __cc == 0;
-}
-
-/*-- vec_all_numeric --------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_all_numeric(vector float __a) {
- int __cc;
- __builtin_s390_vftcisb(__a, 15, &__cc);
- return __cc == 3;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_all_numeric(vector double __a) {
- int __cc;
- __builtin_s390_vftcidb(__a, 15, &__cc);
- return __cc == 3;
-}
-
-/*-- vec_any_eq -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai int
-vec_any_eq(vector signed char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vceqbs(__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector signed char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vceqbs(__a, (vector signed char)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector bool char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a, __b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_eq(vector unsigned char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector unsigned char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector bool char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_eq(vector bool char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_eq(vector signed short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vceqhs(__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector signed short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vceqhs(__a, (vector signed short)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector bool short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a, __b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_eq(vector unsigned short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector unsigned short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector bool short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_eq(vector bool short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_eq(vector signed int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vceqfs(__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector signed int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vceqfs(__a, (vector signed int)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector bool int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a, __b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_eq(vector unsigned int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector unsigned int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector bool int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_eq(vector bool int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_eq(vector signed long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vceqgs(__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector signed long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vceqgs(__a, (vector signed long long)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector bool long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a, __b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_eq(vector unsigned long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector unsigned long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_eq(vector bool long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_eq(vector bool long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc <= 1;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_any_eq(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfcesbs(__a, __b, &__cc);
- return __cc <= 1;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_any_eq(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfcedbs(__a, __b, &__cc);
- return __cc <= 1;
-}
-
-/*-- vec_any_ne -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai int
-vec_any_ne(vector signed char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vceqbs(__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector signed char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vceqbs(__a, (vector signed char)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector bool char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a, __b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ne(vector unsigned char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector unsigned char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector bool char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ne(vector bool char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vceqbs((vector signed char)__a,
- (vector signed char)__b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ne(vector signed short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vceqhs(__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector signed short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vceqhs(__a, (vector signed short)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector bool short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a, __b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ne(vector unsigned short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector unsigned short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector bool short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ne(vector bool short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vceqhs((vector signed short)__a,
- (vector signed short)__b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ne(vector signed int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vceqfs(__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector signed int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vceqfs(__a, (vector signed int)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector bool int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a, __b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ne(vector unsigned int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector unsigned int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector bool int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ne(vector bool int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vceqfs((vector signed int)__a,
- (vector signed int)__b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ne(vector signed long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vceqgs(__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector signed long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vceqgs(__a, (vector signed long long)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector bool long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a, __b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ne(vector unsigned long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector unsigned long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ne(vector bool long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ne(vector bool long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vceqgs((vector signed long long)__a,
- (vector signed long long)__b, &__cc);
- return __cc != 0;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_any_ne(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfcesbs(__a, __b, &__cc);
- return __cc != 0;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_any_ne(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfcedbs(__a, __b, &__cc);
- return __cc != 0;
-}
-
-/*-- vec_any_ge -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai int
-vec_any_ge(vector signed char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs(__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector signed char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchbs((vector signed char)__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector bool char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs(__b, (vector signed char)__a, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ge(vector unsigned char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs(__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector unsigned char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector bool char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs(__b, (vector unsigned char)__a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector bool char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__b,
- (vector unsigned char)__a, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ge(vector signed short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs(__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector signed short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchhs((vector signed short)__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector bool short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs(__b, (vector signed short)__a, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ge(vector unsigned short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs(__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector unsigned short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector bool short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs(__b, (vector unsigned short)__a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector bool short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__b,
- (vector unsigned short)__a, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ge(vector signed int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs(__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector signed int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchfs((vector signed int)__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector bool int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs(__b, (vector signed int)__a, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ge(vector unsigned int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs(__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector unsigned int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector bool int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs(__b, (vector unsigned int)__a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector bool int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__b,
- (vector unsigned int)__a, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ge(vector signed long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs(__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector signed long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchgs((vector signed long long)__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector bool long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs(__b, (vector signed long long)__a, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_ge(vector unsigned long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector unsigned long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__b, __a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector bool long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__b, (vector unsigned long long)__a, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_ge(vector bool long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__b,
- (vector unsigned long long)__a, &__cc);
- return __cc != 0;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_any_ge(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchesbs(__a, __b, &__cc);
- return __cc <= 1;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_any_ge(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchedbs(__a, __b, &__cc);
- return __cc <= 1;
-}
-
-/*-- vec_any_gt -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai int
-vec_any_gt(vector signed char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs(__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector signed char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchbs(__a, (vector signed char)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector bool char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs((vector signed char)__a, __b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_gt(vector unsigned char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs(__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector unsigned char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs(__a, (vector unsigned char)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector bool char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector bool char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__a,
- (vector unsigned char)__b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_gt(vector signed short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs(__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector signed short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchhs(__a, (vector signed short)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector bool short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs((vector signed short)__a, __b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_gt(vector unsigned short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs(__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector unsigned short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs(__a, (vector unsigned short)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector bool short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector bool short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__a,
- (vector unsigned short)__b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_gt(vector signed int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs(__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector signed int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchfs(__a, (vector signed int)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector bool int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs((vector signed int)__a, __b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_gt(vector unsigned int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs(__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector unsigned int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs(__a, (vector unsigned int)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector bool int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector bool int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__a,
- (vector unsigned int)__b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_gt(vector signed long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs(__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector signed long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchgs(__a, (vector signed long long)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector bool long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs((vector signed long long)__a, __b, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_gt(vector unsigned long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector unsigned long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__a, (vector unsigned long long)__b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector bool long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__a, __b, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_gt(vector bool long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__a,
- (vector unsigned long long)__b, &__cc);
- return __cc <= 1;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_any_gt(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchsbs(__a, __b, &__cc);
- return __cc <= 1;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_any_gt(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchdbs(__a, __b, &__cc);
- return __cc <= 1;
-}
-
-/*-- vec_any_le -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai int
-vec_any_le(vector signed char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs(__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector signed char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchbs(__a, (vector signed char)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector bool char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs((vector signed char)__a, __b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_le(vector unsigned char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs(__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector unsigned char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs(__a, (vector unsigned char)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector bool char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector bool char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__a,
- (vector unsigned char)__b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_le(vector signed short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs(__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector signed short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchhs(__a, (vector signed short)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector bool short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs((vector signed short)__a, __b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_le(vector unsigned short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs(__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector unsigned short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs(__a, (vector unsigned short)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector bool short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector bool short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__a,
- (vector unsigned short)__b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_le(vector signed int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs(__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector signed int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchfs(__a, (vector signed int)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector bool int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs((vector signed int)__a, __b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_le(vector unsigned int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs(__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector unsigned int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs(__a, (vector unsigned int)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector bool int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector bool int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__a,
- (vector unsigned int)__b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_le(vector signed long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs(__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector signed long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchgs(__a, (vector signed long long)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector bool long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs((vector signed long long)__a, __b, &__cc);
- return __cc != 0;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_le(vector unsigned long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector unsigned long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__a, (vector unsigned long long)__b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector bool long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__a, __b, &__cc);
- return __cc != 0;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_le(vector bool long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__a,
- (vector unsigned long long)__b, &__cc);
- return __cc != 0;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_any_le(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchesbs(__b, __a, &__cc);
- return __cc <= 1;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_any_le(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchedbs(__b, __a, &__cc);
- return __cc <= 1;
-}
-
-/*-- vec_any_lt -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai int
-vec_any_lt(vector signed char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs(__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector signed char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchbs((vector signed char)__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector bool char __a, vector signed char __b) {
- int __cc;
- __builtin_s390_vchbs(__b, (vector signed char)__a, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_lt(vector unsigned char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs(__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector unsigned char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector bool char __a, vector unsigned char __b) {
- int __cc;
- __builtin_s390_vchlbs(__b, (vector unsigned char)__a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector bool char __a, vector bool char __b) {
- int __cc;
- __builtin_s390_vchlbs((vector unsigned char)__b,
- (vector unsigned char)__a, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_lt(vector signed short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs(__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector signed short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchhs((vector signed short)__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector bool short __a, vector signed short __b) {
- int __cc;
- __builtin_s390_vchhs(__b, (vector signed short)__a, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_lt(vector unsigned short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs(__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector unsigned short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector bool short __a, vector unsigned short __b) {
- int __cc;
- __builtin_s390_vchlhs(__b, (vector unsigned short)__a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector bool short __a, vector bool short __b) {
- int __cc;
- __builtin_s390_vchlhs((vector unsigned short)__b,
- (vector unsigned short)__a, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_lt(vector signed int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs(__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector signed int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchfs((vector signed int)__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector bool int __a, vector signed int __b) {
- int __cc;
- __builtin_s390_vchfs(__b, (vector signed int)__a, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_lt(vector unsigned int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs(__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector unsigned int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector bool int __a, vector unsigned int __b) {
- int __cc;
- __builtin_s390_vchlfs(__b, (vector unsigned int)__a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector bool int __a, vector bool int __b) {
- int __cc;
- __builtin_s390_vchlfs((vector unsigned int)__b,
- (vector unsigned int)__a, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_lt(vector signed long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs(__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector signed long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchgs((vector signed long long)__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector bool long long __a, vector signed long long __b) {
- int __cc;
- __builtin_s390_vchgs(__b, (vector signed long long)__a, &__cc);
- return __cc <= 1;
-}
-
-static inline __ATTRS_o_ai int
-vec_any_lt(vector unsigned long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector unsigned long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__b, __a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector bool long long __a, vector unsigned long long __b) {
- int __cc;
- __builtin_s390_vchlgs(__b, (vector unsigned long long)__a, &__cc);
- return __cc <= 1;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai int
-vec_any_lt(vector bool long long __a, vector bool long long __b) {
- int __cc;
- __builtin_s390_vchlgs((vector unsigned long long)__b,
- (vector unsigned long long)__a, &__cc);
- return __cc <= 1;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_any_lt(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchsbs(__b, __a, &__cc);
- return __cc <= 1;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_any_lt(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchdbs(__b, __a, &__cc);
- return __cc <= 1;
-}
-
-/*-- vec_any_nge ------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_any_nge(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchesbs(__a, __b, &__cc);
- return __cc != 0;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_any_nge(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchedbs(__a, __b, &__cc);
- return __cc != 0;
-}
-
-/*-- vec_any_ngt ------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_any_ngt(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchsbs(__a, __b, &__cc);
- return __cc != 0;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_any_ngt(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchdbs(__a, __b, &__cc);
- return __cc != 0;
-}
-
-/*-- vec_any_nle ------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_any_nle(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchesbs(__b, __a, &__cc);
- return __cc != 0;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_any_nle(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchedbs(__b, __a, &__cc);
- return __cc != 0;
-}
-
-/*-- vec_any_nlt ------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_any_nlt(vector float __a, vector float __b) {
- int __cc;
- __builtin_s390_vfchsbs(__b, __a, &__cc);
- return __cc != 0;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_any_nlt(vector double __a, vector double __b) {
- int __cc;
- __builtin_s390_vfchdbs(__b, __a, &__cc);
- return __cc != 0;
-}
-
-/*-- vec_any_nan ------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_any_nan(vector float __a) {
- int __cc;
- __builtin_s390_vftcisb(__a, 15, &__cc);
- return __cc != 3;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_any_nan(vector double __a) {
- int __cc;
- __builtin_s390_vftcidb(__a, 15, &__cc);
- return __cc != 3;
-}
-
-/*-- vec_any_numeric --------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_any_numeric(vector float __a) {
- int __cc;
- __builtin_s390_vftcisb(__a, 15, &__cc);
- return __cc != 0;
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_any_numeric(vector double __a) {
- int __cc;
- __builtin_s390_vftcidb(__a, 15, &__cc);
- return __cc != 0;
-}
-
-/*-- vec_andc ---------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector bool char
-vec_andc(vector bool char __a, vector bool char __b) {
- return __a & ~__b;
-}
-
-static inline __ATTRS_o_ai vector signed char
-vec_andc(vector signed char __a, vector signed char __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_andc(vector bool char __a, vector signed char __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_andc(vector signed char __a, vector bool char __b) {
- return __a & ~__b;
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_andc(vector unsigned char __a, vector unsigned char __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_andc(vector bool char __a, vector unsigned char __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_andc(vector unsigned char __a, vector bool char __b) {
- return __a & ~__b;
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_andc(vector bool short __a, vector bool short __b) {
- return __a & ~__b;
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_andc(vector signed short __a, vector signed short __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_andc(vector bool short __a, vector signed short __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_andc(vector signed short __a, vector bool short __b) {
- return __a & ~__b;
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_andc(vector unsigned short __a, vector unsigned short __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_andc(vector bool short __a, vector unsigned short __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_andc(vector unsigned short __a, vector bool short __b) {
- return __a & ~__b;
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_andc(vector bool int __a, vector bool int __b) {
- return __a & ~__b;
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_andc(vector signed int __a, vector signed int __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_andc(vector bool int __a, vector signed int __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_andc(vector signed int __a, vector bool int __b) {
- return __a & ~__b;
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_andc(vector unsigned int __a, vector unsigned int __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_andc(vector bool int __a, vector unsigned int __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_andc(vector unsigned int __a, vector bool int __b) {
- return __a & ~__b;
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_andc(vector bool long long __a, vector bool long long __b) {
- return __a & ~__b;
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_andc(vector signed long long __a, vector signed long long __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_andc(vector bool long long __a, vector signed long long __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_andc(vector signed long long __a, vector bool long long __b) {
- return __a & ~__b;
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_andc(vector unsigned long long __a, vector unsigned long long __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_andc(vector bool long long __a, vector unsigned long long __b) {
- return __a & ~__b;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_andc(vector unsigned long long __a, vector bool long long __b) {
- return __a & ~__b;
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_andc(vector float __a, vector float __b) {
- return (vector float)((vector unsigned int)__a &
- ~(vector unsigned int)__b);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_andc(vector double __a, vector double __b) {
- return (vector double)((vector unsigned long long)__a &
- ~(vector unsigned long long)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector double
-vec_andc(vector bool long long __a, vector double __b) {
- return (vector double)((vector unsigned long long)__a &
- ~(vector unsigned long long)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector double
-vec_andc(vector double __a, vector bool long long __b) {
- return (vector double)((vector unsigned long long)__a &
- ~(vector unsigned long long)__b);
-}
-
-/*-- vec_nor ----------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector bool char
-vec_nor(vector bool char __a, vector bool char __b) {
- return ~(__a | __b);
-}
-
-static inline __ATTRS_o_ai vector signed char
-vec_nor(vector signed char __a, vector signed char __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_nor(vector bool char __a, vector signed char __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_nor(vector signed char __a, vector bool char __b) {
- return ~(__a | __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_nor(vector unsigned char __a, vector unsigned char __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_nor(vector bool char __a, vector unsigned char __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_nor(vector unsigned char __a, vector bool char __b) {
- return ~(__a | __b);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_nor(vector bool short __a, vector bool short __b) {
- return ~(__a | __b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_nor(vector signed short __a, vector signed short __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_nor(vector bool short __a, vector signed short __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_nor(vector signed short __a, vector bool short __b) {
- return ~(__a | __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_nor(vector unsigned short __a, vector unsigned short __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_nor(vector bool short __a, vector unsigned short __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_nor(vector unsigned short __a, vector bool short __b) {
- return ~(__a | __b);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_nor(vector bool int __a, vector bool int __b) {
- return ~(__a | __b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_nor(vector signed int __a, vector signed int __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_nor(vector bool int __a, vector signed int __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_nor(vector signed int __a, vector bool int __b) {
- return ~(__a | __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_nor(vector unsigned int __a, vector unsigned int __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_nor(vector bool int __a, vector unsigned int __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_nor(vector unsigned int __a, vector bool int __b) {
- return ~(__a | __b);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_nor(vector bool long long __a, vector bool long long __b) {
- return ~(__a | __b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_nor(vector signed long long __a, vector signed long long __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_nor(vector bool long long __a, vector signed long long __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_nor(vector signed long long __a, vector bool long long __b) {
- return ~(__a | __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_nor(vector unsigned long long __a, vector unsigned long long __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_nor(vector bool long long __a, vector unsigned long long __b) {
- return ~(__a | __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_nor(vector unsigned long long __a, vector bool long long __b) {
- return ~(__a | __b);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_nor(vector float __a, vector float __b) {
- return (vector float)~((vector unsigned int)__a |
- (vector unsigned int)__b);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_nor(vector double __a, vector double __b) {
- return (vector double)~((vector unsigned long long)__a |
- (vector unsigned long long)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector double
-vec_nor(vector bool long long __a, vector double __b) {
- return (vector double)~((vector unsigned long long)__a |
- (vector unsigned long long)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector double
-vec_nor(vector double __a, vector bool long long __b) {
- return (vector double)~((vector unsigned long long)__a |
- (vector unsigned long long)__b);
-}
-
-/*-- vec_orc ----------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector bool char
-vec_orc(vector bool char __a, vector bool char __b) {
- return __a | ~__b;
-}
-
-static inline __ATTRS_o_ai vector signed char
-vec_orc(vector signed char __a, vector signed char __b) {
- return __a | ~__b;
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_orc(vector unsigned char __a, vector unsigned char __b) {
- return __a | ~__b;
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_orc(vector bool short __a, vector bool short __b) {
- return __a | ~__b;
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_orc(vector signed short __a, vector signed short __b) {
- return __a | ~__b;
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_orc(vector unsigned short __a, vector unsigned short __b) {
- return __a | ~__b;
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_orc(vector bool int __a, vector bool int __b) {
- return __a | ~__b;
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_orc(vector signed int __a, vector signed int __b) {
- return __a | ~__b;
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_orc(vector unsigned int __a, vector unsigned int __b) {
- return __a | ~__b;
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_orc(vector bool long long __a, vector bool long long __b) {
- return __a | ~__b;
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_orc(vector signed long long __a, vector signed long long __b) {
- return __a | ~__b;
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_orc(vector unsigned long long __a, vector unsigned long long __b) {
- return __a | ~__b;
-}
-
-static inline __ATTRS_o_ai vector float
-vec_orc(vector float __a, vector float __b) {
- return (vector float)((vector unsigned int)__a &
- ~(vector unsigned int)__b);
-}
-
-static inline __ATTRS_o_ai vector double
-vec_orc(vector double __a, vector double __b) {
- return (vector double)((vector unsigned long long)__a &
- ~(vector unsigned long long)__b);
-}
-#endif
-
-/*-- vec_nand ---------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector bool char
-vec_nand(vector bool char __a, vector bool char __b) {
- return ~(__a & __b);
-}
-
-static inline __ATTRS_o_ai vector signed char
-vec_nand(vector signed char __a, vector signed char __b) {
- return ~(__a & __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_nand(vector unsigned char __a, vector unsigned char __b) {
- return ~(__a & __b);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_nand(vector bool short __a, vector bool short __b) {
- return ~(__a & __b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_nand(vector signed short __a, vector signed short __b) {
- return ~(__a & __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_nand(vector unsigned short __a, vector unsigned short __b) {
- return ~(__a & __b);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_nand(vector bool int __a, vector bool int __b) {
- return ~(__a & __b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_nand(vector signed int __a, vector signed int __b) {
- return ~(__a & __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_nand(vector unsigned int __a, vector unsigned int __b) {
- return ~(__a & __b);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_nand(vector bool long long __a, vector bool long long __b) {
- return ~(__a & __b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_nand(vector signed long long __a, vector signed long long __b) {
- return ~(__a & __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_nand(vector unsigned long long __a, vector unsigned long long __b) {
- return ~(__a & __b);
-}
-
-static inline __ATTRS_o_ai vector float
-vec_nand(vector float __a, vector float __b) {
- return (vector float)~((vector unsigned int)__a &
- (vector unsigned int)__b);
-}
-
-static inline __ATTRS_o_ai vector double
-vec_nand(vector double __a, vector double __b) {
- return (vector double)~((vector unsigned long long)__a &
- (vector unsigned long long)__b);
-}
-#endif
-
-/*-- vec_eqv ----------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector bool char
-vec_eqv(vector bool char __a, vector bool char __b) {
- return ~(__a ^ __b);
-}
-
-static inline __ATTRS_o_ai vector signed char
-vec_eqv(vector signed char __a, vector signed char __b) {
- return ~(__a ^ __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_eqv(vector unsigned char __a, vector unsigned char __b) {
- return ~(__a ^ __b);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_eqv(vector bool short __a, vector bool short __b) {
- return ~(__a ^ __b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_eqv(vector signed short __a, vector signed short __b) {
- return ~(__a ^ __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_eqv(vector unsigned short __a, vector unsigned short __b) {
- return ~(__a ^ __b);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_eqv(vector bool int __a, vector bool int __b) {
- return ~(__a ^ __b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_eqv(vector signed int __a, vector signed int __b) {
- return ~(__a ^ __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_eqv(vector unsigned int __a, vector unsigned int __b) {
- return ~(__a ^ __b);
-}
-
-static inline __ATTRS_o_ai vector bool long long
-vec_eqv(vector bool long long __a, vector bool long long __b) {
- return ~(__a ^ __b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_eqv(vector signed long long __a, vector signed long long __b) {
- return ~(__a ^ __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_eqv(vector unsigned long long __a, vector unsigned long long __b) {
- return ~(__a ^ __b);
-}
-
-static inline __ATTRS_o_ai vector float
-vec_eqv(vector float __a, vector float __b) {
- return (vector float)~((vector unsigned int)__a ^
- (vector unsigned int)__b);
-}
-
-static inline __ATTRS_o_ai vector double
-vec_eqv(vector double __a, vector double __b) {
- return (vector double)~((vector unsigned long long)__a ^
- (vector unsigned long long)__b);
-}
-#endif
-
-/*-- vec_cntlz --------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cntlz(vector signed char __a) {
- return __builtin_s390_vclzb((vector unsigned char)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cntlz(vector unsigned char __a) {
- return __builtin_s390_vclzb(__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cntlz(vector signed short __a) {
- return __builtin_s390_vclzh((vector unsigned short)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cntlz(vector unsigned short __a) {
- return __builtin_s390_vclzh(__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cntlz(vector signed int __a) {
- return __builtin_s390_vclzf((vector unsigned int)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cntlz(vector unsigned int __a) {
- return __builtin_s390_vclzf(__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_cntlz(vector signed long long __a) {
- return __builtin_s390_vclzg((vector unsigned long long)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_cntlz(vector unsigned long long __a) {
- return __builtin_s390_vclzg(__a);
-}
-
-/*-- vec_cnttz --------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cnttz(vector signed char __a) {
- return __builtin_s390_vctzb((vector unsigned char)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cnttz(vector unsigned char __a) {
- return __builtin_s390_vctzb(__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cnttz(vector signed short __a) {
- return __builtin_s390_vctzh((vector unsigned short)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cnttz(vector unsigned short __a) {
- return __builtin_s390_vctzh(__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cnttz(vector signed int __a) {
- return __builtin_s390_vctzf((vector unsigned int)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cnttz(vector unsigned int __a) {
- return __builtin_s390_vctzf(__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_cnttz(vector signed long long __a) {
- return __builtin_s390_vctzg((vector unsigned long long)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_cnttz(vector unsigned long long __a) {
- return __builtin_s390_vctzg(__a);
-}
-
-/*-- vec_popcnt -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_popcnt(vector signed char __a) {
- return __builtin_s390_vpopctb((vector unsigned char)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_popcnt(vector unsigned char __a) {
- return __builtin_s390_vpopctb(__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_popcnt(vector signed short __a) {
- return __builtin_s390_vpopcth((vector unsigned short)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_popcnt(vector unsigned short __a) {
- return __builtin_s390_vpopcth(__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_popcnt(vector signed int __a) {
- return __builtin_s390_vpopctf((vector unsigned int)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_popcnt(vector unsigned int __a) {
- return __builtin_s390_vpopctf(__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_popcnt(vector signed long long __a) {
- return __builtin_s390_vpopctg((vector unsigned long long)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_popcnt(vector unsigned long long __a) {
- return __builtin_s390_vpopctg(__a);
-}
-
-/*-- vec_rl -----------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_rl(vector signed char __a, vector unsigned char __b) {
- return (vector signed char)__builtin_s390_verllvb(
- (vector unsigned char)__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_rl(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_verllvb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_rl(vector signed short __a, vector unsigned short __b) {
- return (vector signed short)__builtin_s390_verllvh(
- (vector unsigned short)__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_rl(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_verllvh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_rl(vector signed int __a, vector unsigned int __b) {
- return (vector signed int)__builtin_s390_verllvf(
- (vector unsigned int)__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_rl(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_verllvf(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_rl(vector signed long long __a, vector unsigned long long __b) {
- return (vector signed long long)__builtin_s390_verllvg(
- (vector unsigned long long)__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_rl(vector unsigned long long __a, vector unsigned long long __b) {
- return __builtin_s390_verllvg(__a, __b);
-}
-
-/*-- vec_rli ----------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_rli(vector signed char __a, unsigned long __b) {
- return (vector signed char)__builtin_s390_verllb(
- (vector unsigned char)__a, (int)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_rli(vector unsigned char __a, unsigned long __b) {
- return __builtin_s390_verllb(__a, (int)__b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_rli(vector signed short __a, unsigned long __b) {
- return (vector signed short)__builtin_s390_verllh(
- (vector unsigned short)__a, (int)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_rli(vector unsigned short __a, unsigned long __b) {
- return __builtin_s390_verllh(__a, (int)__b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_rli(vector signed int __a, unsigned long __b) {
- return (vector signed int)__builtin_s390_verllf(
- (vector unsigned int)__a, (int)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_rli(vector unsigned int __a, unsigned long __b) {
- return __builtin_s390_verllf(__a, (int)__b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_rli(vector signed long long __a, unsigned long __b) {
- return (vector signed long long)__builtin_s390_verllg(
- (vector unsigned long long)__a, (int)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_rli(vector unsigned long long __a, unsigned long __b) {
- return __builtin_s390_verllg(__a, (int)__b);
-}
-
-/*-- vec_rl_mask ------------------------------------------------------------*/
-
-extern __ATTRS_o vector signed char
-vec_rl_mask(vector signed char __a, vector unsigned char __b,
- unsigned char __c) __constant(__c);
-
-extern __ATTRS_o vector unsigned char
-vec_rl_mask(vector unsigned char __a, vector unsigned char __b,
- unsigned char __c) __constant(__c);
-
-extern __ATTRS_o vector signed short
-vec_rl_mask(vector signed short __a, vector unsigned short __b,
- unsigned char __c) __constant(__c);
-
-extern __ATTRS_o vector unsigned short
-vec_rl_mask(vector unsigned short __a, vector unsigned short __b,
- unsigned char __c) __constant(__c);
-
-extern __ATTRS_o vector signed int
-vec_rl_mask(vector signed int __a, vector unsigned int __b,
- unsigned char __c) __constant(__c);
-
-extern __ATTRS_o vector unsigned int
-vec_rl_mask(vector unsigned int __a, vector unsigned int __b,
- unsigned char __c) __constant(__c);
-
-extern __ATTRS_o vector signed long long
-vec_rl_mask(vector signed long long __a, vector unsigned long long __b,
- unsigned char __c) __constant(__c);
-
-extern __ATTRS_o vector unsigned long long
-vec_rl_mask(vector unsigned long long __a, vector unsigned long long __b,
- unsigned char __c) __constant(__c);
-
-#define vec_rl_mask(X, Y, Z) ((__typeof__((vec_rl_mask)((X), (Y), (Z)))) \
- __extension__ ({ \
- vector unsigned char __res; \
- vector unsigned char __x = (vector unsigned char)(X); \
- vector unsigned char __y = (vector unsigned char)(Y); \
- switch (sizeof ((X)[0])) { \
- case 1: __res = (vector unsigned char) __builtin_s390_verimb( \
- (vector unsigned char)__x, (vector unsigned char)__x, \
- (vector unsigned char)__y, (Z)); break; \
- case 2: __res = (vector unsigned char) __builtin_s390_verimh( \
- (vector unsigned short)__x, (vector unsigned short)__x, \
- (vector unsigned short)__y, (Z)); break; \
- case 4: __res = (vector unsigned char) __builtin_s390_verimf( \
- (vector unsigned int)__x, (vector unsigned int)__x, \
- (vector unsigned int)__y, (Z)); break; \
- default: __res = (vector unsigned char) __builtin_s390_verimg( \
- (vector unsigned long long)__x, (vector unsigned long long)__x, \
- (vector unsigned long long)__y, (Z)); break; \
- } __res; }))
-
-/*-- vec_sll ----------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_sll(vector signed char __a, vector unsigned char __b) {
- return (vector signed char)__builtin_s390_vsl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_sll(vector signed char __a, vector unsigned short __b) {
- return (vector signed char)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_sll(vector signed char __a, vector unsigned int __b) {
- return (vector signed char)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool char
-vec_sll(vector bool char __a, vector unsigned char __b) {
- return (vector bool char)__builtin_s390_vsl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool char
-vec_sll(vector bool char __a, vector unsigned short __b) {
- return (vector bool char)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool char
-vec_sll(vector bool char __a, vector unsigned int __b) {
- return (vector bool char)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_sll(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vsl(__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_sll(vector unsigned char __a, vector unsigned short __b) {
- return __builtin_s390_vsl(__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_sll(vector unsigned char __a, vector unsigned int __b) {
- return __builtin_s390_vsl(__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_sll(vector signed short __a, vector unsigned char __b) {
- return (vector signed short)__builtin_s390_vsl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_sll(vector signed short __a, vector unsigned short __b) {
- return (vector signed short)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_sll(vector signed short __a, vector unsigned int __b) {
- return (vector signed short)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool short
-vec_sll(vector bool short __a, vector unsigned char __b) {
- return (vector bool short)__builtin_s390_vsl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool short
-vec_sll(vector bool short __a, vector unsigned short __b) {
- return (vector bool short)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool short
-vec_sll(vector bool short __a, vector unsigned int __b) {
- return (vector bool short)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_sll(vector unsigned short __a, vector unsigned char __b) {
- return (vector unsigned short)__builtin_s390_vsl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_sll(vector unsigned short __a, vector unsigned short __b) {
- return (vector unsigned short)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_sll(vector unsigned short __a, vector unsigned int __b) {
- return (vector unsigned short)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_sll(vector signed int __a, vector unsigned char __b) {
- return (vector signed int)__builtin_s390_vsl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_sll(vector signed int __a, vector unsigned short __b) {
- return (vector signed int)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_sll(vector signed int __a, vector unsigned int __b) {
- return (vector signed int)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool int
-vec_sll(vector bool int __a, vector unsigned char __b) {
- return (vector bool int)__builtin_s390_vsl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool int
-vec_sll(vector bool int __a, vector unsigned short __b) {
- return (vector bool int)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool int
-vec_sll(vector bool int __a, vector unsigned int __b) {
- return (vector bool int)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_sll(vector unsigned int __a, vector unsigned char __b) {
- return (vector unsigned int)__builtin_s390_vsl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_sll(vector unsigned int __a, vector unsigned short __b) {
- return (vector unsigned int)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_sll(vector unsigned int __a, vector unsigned int __b) {
- return (vector unsigned int)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_sll(vector signed long long __a, vector unsigned char __b) {
- return (vector signed long long)__builtin_s390_vsl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_sll(vector signed long long __a, vector unsigned short __b) {
- return (vector signed long long)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_sll(vector signed long long __a, vector unsigned int __b) {
- return (vector signed long long)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool long long
-vec_sll(vector bool long long __a, vector unsigned char __b) {
- return (vector bool long long)__builtin_s390_vsl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool long long
-vec_sll(vector bool long long __a, vector unsigned short __b) {
- return (vector bool long long)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool long long
-vec_sll(vector bool long long __a, vector unsigned int __b) {
- return (vector bool long long)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_sll(vector unsigned long long __a, vector unsigned char __b) {
- return (vector unsigned long long)__builtin_s390_vsl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_sll(vector unsigned long long __a, vector unsigned short __b) {
- return (vector unsigned long long)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_sll(vector unsigned long long __a, vector unsigned int __b) {
- return (vector unsigned long long)__builtin_s390_vsl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-/*-- vec_slb ----------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_slb(vector signed char __a, vector signed char __b) {
- return (vector signed char)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed char
-vec_slb(vector signed char __a, vector unsigned char __b) {
- return (vector signed char)__builtin_s390_vslb(
- (vector unsigned char)__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_slb(vector unsigned char __a, vector signed char __b) {
- return __builtin_s390_vslb(__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_slb(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vslb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_slb(vector signed short __a, vector signed short __b) {
- return (vector signed short)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_slb(vector signed short __a, vector unsigned short __b) {
- return (vector signed short)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_slb(vector unsigned short __a, vector signed short __b) {
- return (vector unsigned short)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_slb(vector unsigned short __a, vector unsigned short __b) {
- return (vector unsigned short)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_slb(vector signed int __a, vector signed int __b) {
- return (vector signed int)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_slb(vector signed int __a, vector unsigned int __b) {
- return (vector signed int)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_slb(vector unsigned int __a, vector signed int __b) {
- return (vector unsigned int)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_slb(vector unsigned int __a, vector unsigned int __b) {
- return (vector unsigned int)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_slb(vector signed long long __a, vector signed long long __b) {
- return (vector signed long long)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_slb(vector signed long long __a, vector unsigned long long __b) {
- return (vector signed long long)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_slb(vector unsigned long long __a, vector signed long long __b) {
- return (vector unsigned long long)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_slb(vector unsigned long long __a, vector unsigned long long __b) {
- return (vector unsigned long long)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_slb(vector float __a, vector signed int __b) {
- return (vector float)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector float
-vec_slb(vector float __a, vector unsigned int __b) {
- return (vector float)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_slb(vector double __a, vector signed long long __b) {
- return (vector double)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector double
-vec_slb(vector double __a, vector unsigned long long __b) {
- return (vector double)__builtin_s390_vslb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-/*-- vec_sld ----------------------------------------------------------------*/
-
-extern __ATTRS_o vector signed char
-vec_sld(vector signed char __a, vector signed char __b, int __c)
- __constant_range(__c, 0, 15);
-
-extern __ATTRS_o vector bool char
-vec_sld(vector bool char __a, vector bool char __b, int __c)
- __constant_range(__c, 0, 15);
-
-extern __ATTRS_o vector unsigned char
-vec_sld(vector unsigned char __a, vector unsigned char __b, int __c)
- __constant_range(__c, 0, 15);
-
-extern __ATTRS_o vector signed short
-vec_sld(vector signed short __a, vector signed short __b, int __c)
- __constant_range(__c, 0, 15);
-
-extern __ATTRS_o vector bool short
-vec_sld(vector bool short __a, vector bool short __b, int __c)
- __constant_range(__c, 0, 15);
-
-extern __ATTRS_o vector unsigned short
-vec_sld(vector unsigned short __a, vector unsigned short __b, int __c)
- __constant_range(__c, 0, 15);
-
-extern __ATTRS_o vector signed int
-vec_sld(vector signed int __a, vector signed int __b, int __c)
- __constant_range(__c, 0, 15);
-
-extern __ATTRS_o vector bool int
-vec_sld(vector bool int __a, vector bool int __b, int __c)
- __constant_range(__c, 0, 15);
-
-extern __ATTRS_o vector unsigned int
-vec_sld(vector unsigned int __a, vector unsigned int __b, int __c)
- __constant_range(__c, 0, 15);
-
-extern __ATTRS_o vector signed long long
-vec_sld(vector signed long long __a, vector signed long long __b, int __c)
- __constant_range(__c, 0, 15);
-
-extern __ATTRS_o vector bool long long
-vec_sld(vector bool long long __a, vector bool long long __b, int __c)
- __constant_range(__c, 0, 15);
-
-extern __ATTRS_o vector unsigned long long
-vec_sld(vector unsigned long long __a, vector unsigned long long __b, int __c)
- __constant_range(__c, 0, 15);
-
-#if __ARCH__ >= 12
-extern __ATTRS_o vector float
-vec_sld(vector float __a, vector float __b, int __c)
- __constant_range(__c, 0, 15);
-#endif
-
-extern __ATTRS_o vector double
-vec_sld(vector double __a, vector double __b, int __c)
- __constant_range(__c, 0, 15);
-
-#define vec_sld(X, Y, Z) ((__typeof__((vec_sld)((X), (Y), (Z)))) \
- __builtin_s390_vsldb((vector unsigned char)(X), \
- (vector unsigned char)(Y), (Z)))
-
-/*-- vec_sldw ---------------------------------------------------------------*/
-
-extern __ATTRS_o vector signed char
-vec_sldw(vector signed char __a, vector signed char __b, int __c)
- __constant_range(__c, 0, 3);
-
-extern __ATTRS_o vector unsigned char
-vec_sldw(vector unsigned char __a, vector unsigned char __b, int __c)
- __constant_range(__c, 0, 3);
-
-extern __ATTRS_o vector signed short
-vec_sldw(vector signed short __a, vector signed short __b, int __c)
- __constant_range(__c, 0, 3);
-
-extern __ATTRS_o vector unsigned short
-vec_sldw(vector unsigned short __a, vector unsigned short __b, int __c)
- __constant_range(__c, 0, 3);
-
-extern __ATTRS_o vector signed int
-vec_sldw(vector signed int __a, vector signed int __b, int __c)
- __constant_range(__c, 0, 3);
-
-extern __ATTRS_o vector unsigned int
-vec_sldw(vector unsigned int __a, vector unsigned int __b, int __c)
- __constant_range(__c, 0, 3);
-
-extern __ATTRS_o vector signed long long
-vec_sldw(vector signed long long __a, vector signed long long __b, int __c)
- __constant_range(__c, 0, 3);
-
-extern __ATTRS_o vector unsigned long long
-vec_sldw(vector unsigned long long __a, vector unsigned long long __b, int __c)
- __constant_range(__c, 0, 3);
-
-// This prototype is deprecated.
-extern __ATTRS_o vector double
-vec_sldw(vector double __a, vector double __b, int __c)
- __constant_range(__c, 0, 3);
-
-#define vec_sldw(X, Y, Z) ((__typeof__((vec_sldw)((X), (Y), (Z)))) \
- __builtin_s390_vsldb((vector unsigned char)(X), \
- (vector unsigned char)(Y), (Z) * 4))
-
-/*-- vec_sral ---------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_sral(vector signed char __a, vector unsigned char __b) {
- return (vector signed char)__builtin_s390_vsra(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_sral(vector signed char __a, vector unsigned short __b) {
- return (vector signed char)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_sral(vector signed char __a, vector unsigned int __b) {
- return (vector signed char)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool char
-vec_sral(vector bool char __a, vector unsigned char __b) {
- return (vector bool char)__builtin_s390_vsra(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool char
-vec_sral(vector bool char __a, vector unsigned short __b) {
- return (vector bool char)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool char
-vec_sral(vector bool char __a, vector unsigned int __b) {
- return (vector bool char)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_sral(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vsra(__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_sral(vector unsigned char __a, vector unsigned short __b) {
- return __builtin_s390_vsra(__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_sral(vector unsigned char __a, vector unsigned int __b) {
- return __builtin_s390_vsra(__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_sral(vector signed short __a, vector unsigned char __b) {
- return (vector signed short)__builtin_s390_vsra(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_sral(vector signed short __a, vector unsigned short __b) {
- return (vector signed short)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_sral(vector signed short __a, vector unsigned int __b) {
- return (vector signed short)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool short
-vec_sral(vector bool short __a, vector unsigned char __b) {
- return (vector bool short)__builtin_s390_vsra(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool short
-vec_sral(vector bool short __a, vector unsigned short __b) {
- return (vector bool short)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool short
-vec_sral(vector bool short __a, vector unsigned int __b) {
- return (vector bool short)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_sral(vector unsigned short __a, vector unsigned char __b) {
- return (vector unsigned short)__builtin_s390_vsra(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_sral(vector unsigned short __a, vector unsigned short __b) {
- return (vector unsigned short)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_sral(vector unsigned short __a, vector unsigned int __b) {
- return (vector unsigned short)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_sral(vector signed int __a, vector unsigned char __b) {
- return (vector signed int)__builtin_s390_vsra(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_sral(vector signed int __a, vector unsigned short __b) {
- return (vector signed int)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_sral(vector signed int __a, vector unsigned int __b) {
- return (vector signed int)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool int
-vec_sral(vector bool int __a, vector unsigned char __b) {
- return (vector bool int)__builtin_s390_vsra(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool int
-vec_sral(vector bool int __a, vector unsigned short __b) {
- return (vector bool int)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool int
-vec_sral(vector bool int __a, vector unsigned int __b) {
- return (vector bool int)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_sral(vector unsigned int __a, vector unsigned char __b) {
- return (vector unsigned int)__builtin_s390_vsra(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_sral(vector unsigned int __a, vector unsigned short __b) {
- return (vector unsigned int)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_sral(vector unsigned int __a, vector unsigned int __b) {
- return (vector unsigned int)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_sral(vector signed long long __a, vector unsigned char __b) {
- return (vector signed long long)__builtin_s390_vsra(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_sral(vector signed long long __a, vector unsigned short __b) {
- return (vector signed long long)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_sral(vector signed long long __a, vector unsigned int __b) {
- return (vector signed long long)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool long long
-vec_sral(vector bool long long __a, vector unsigned char __b) {
- return (vector bool long long)__builtin_s390_vsra(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool long long
-vec_sral(vector bool long long __a, vector unsigned short __b) {
- return (vector bool long long)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool long long
-vec_sral(vector bool long long __a, vector unsigned int __b) {
- return (vector bool long long)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_sral(vector unsigned long long __a, vector unsigned char __b) {
- return (vector unsigned long long)__builtin_s390_vsra(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_sral(vector unsigned long long __a, vector unsigned short __b) {
- return (vector unsigned long long)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_sral(vector unsigned long long __a, vector unsigned int __b) {
- return (vector unsigned long long)__builtin_s390_vsra(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-/*-- vec_srab ---------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_srab(vector signed char __a, vector signed char __b) {
- return (vector signed char)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed char
-vec_srab(vector signed char __a, vector unsigned char __b) {
- return (vector signed char)__builtin_s390_vsrab(
- (vector unsigned char)__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_srab(vector unsigned char __a, vector signed char __b) {
- return __builtin_s390_vsrab(__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_srab(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vsrab(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_srab(vector signed short __a, vector signed short __b) {
- return (vector signed short)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_srab(vector signed short __a, vector unsigned short __b) {
- return (vector signed short)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_srab(vector unsigned short __a, vector signed short __b) {
- return (vector unsigned short)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_srab(vector unsigned short __a, vector unsigned short __b) {
- return (vector unsigned short)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_srab(vector signed int __a, vector signed int __b) {
- return (vector signed int)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_srab(vector signed int __a, vector unsigned int __b) {
- return (vector signed int)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_srab(vector unsigned int __a, vector signed int __b) {
- return (vector unsigned int)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_srab(vector unsigned int __a, vector unsigned int __b) {
- return (vector unsigned int)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_srab(vector signed long long __a, vector signed long long __b) {
- return (vector signed long long)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_srab(vector signed long long __a, vector unsigned long long __b) {
- return (vector signed long long)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_srab(vector unsigned long long __a, vector signed long long __b) {
- return (vector unsigned long long)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_srab(vector unsigned long long __a, vector unsigned long long __b) {
- return (vector unsigned long long)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_srab(vector float __a, vector signed int __b) {
- return (vector float)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector float
-vec_srab(vector float __a, vector unsigned int __b) {
- return (vector float)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_srab(vector double __a, vector signed long long __b) {
- return (vector double)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector double
-vec_srab(vector double __a, vector unsigned long long __b) {
- return (vector double)__builtin_s390_vsrab(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-/*-- vec_srl ----------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_srl(vector signed char __a, vector unsigned char __b) {
- return (vector signed char)__builtin_s390_vsrl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_srl(vector signed char __a, vector unsigned short __b) {
- return (vector signed char)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_srl(vector signed char __a, vector unsigned int __b) {
- return (vector signed char)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool char
-vec_srl(vector bool char __a, vector unsigned char __b) {
- return (vector bool char)__builtin_s390_vsrl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool char
-vec_srl(vector bool char __a, vector unsigned short __b) {
- return (vector bool char)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool char
-vec_srl(vector bool char __a, vector unsigned int __b) {
- return (vector bool char)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_srl(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vsrl(__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_srl(vector unsigned char __a, vector unsigned short __b) {
- return __builtin_s390_vsrl(__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_srl(vector unsigned char __a, vector unsigned int __b) {
- return __builtin_s390_vsrl(__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_srl(vector signed short __a, vector unsigned char __b) {
- return (vector signed short)__builtin_s390_vsrl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_srl(vector signed short __a, vector unsigned short __b) {
- return (vector signed short)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_srl(vector signed short __a, vector unsigned int __b) {
- return (vector signed short)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool short
-vec_srl(vector bool short __a, vector unsigned char __b) {
- return (vector bool short)__builtin_s390_vsrl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool short
-vec_srl(vector bool short __a, vector unsigned short __b) {
- return (vector bool short)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool short
-vec_srl(vector bool short __a, vector unsigned int __b) {
- return (vector bool short)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_srl(vector unsigned short __a, vector unsigned char __b) {
- return (vector unsigned short)__builtin_s390_vsrl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_srl(vector unsigned short __a, vector unsigned short __b) {
- return (vector unsigned short)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_srl(vector unsigned short __a, vector unsigned int __b) {
- return (vector unsigned short)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_srl(vector signed int __a, vector unsigned char __b) {
- return (vector signed int)__builtin_s390_vsrl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_srl(vector signed int __a, vector unsigned short __b) {
- return (vector signed int)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_srl(vector signed int __a, vector unsigned int __b) {
- return (vector signed int)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool int
-vec_srl(vector bool int __a, vector unsigned char __b) {
- return (vector bool int)__builtin_s390_vsrl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool int
-vec_srl(vector bool int __a, vector unsigned short __b) {
- return (vector bool int)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool int
-vec_srl(vector bool int __a, vector unsigned int __b) {
- return (vector bool int)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_srl(vector unsigned int __a, vector unsigned char __b) {
- return (vector unsigned int)__builtin_s390_vsrl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_srl(vector unsigned int __a, vector unsigned short __b) {
- return (vector unsigned int)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_srl(vector unsigned int __a, vector unsigned int __b) {
- return (vector unsigned int)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_srl(vector signed long long __a, vector unsigned char __b) {
- return (vector signed long long)__builtin_s390_vsrl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_srl(vector signed long long __a, vector unsigned short __b) {
- return (vector signed long long)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_srl(vector signed long long __a, vector unsigned int __b) {
- return (vector signed long long)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool long long
-vec_srl(vector bool long long __a, vector unsigned char __b) {
- return (vector bool long long)__builtin_s390_vsrl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool long long
-vec_srl(vector bool long long __a, vector unsigned short __b) {
- return (vector bool long long)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector bool long long
-vec_srl(vector bool long long __a, vector unsigned int __b) {
- return (vector bool long long)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_srl(vector unsigned long long __a, vector unsigned char __b) {
- return (vector unsigned long long)__builtin_s390_vsrl(
- (vector unsigned char)__a, __b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_srl(vector unsigned long long __a, vector unsigned short __b) {
- return (vector unsigned long long)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_srl(vector unsigned long long __a, vector unsigned int __b) {
- return (vector unsigned long long)__builtin_s390_vsrl(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-/*-- vec_srb ----------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_srb(vector signed char __a, vector signed char __b) {
- return (vector signed char)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed char
-vec_srb(vector signed char __a, vector unsigned char __b) {
- return (vector signed char)__builtin_s390_vsrlb(
- (vector unsigned char)__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_srb(vector unsigned char __a, vector signed char __b) {
- return __builtin_s390_vsrlb(__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_srb(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vsrlb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_srb(vector signed short __a, vector signed short __b) {
- return (vector signed short)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_srb(vector signed short __a, vector unsigned short __b) {
- return (vector signed short)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_srb(vector unsigned short __a, vector signed short __b) {
- return (vector unsigned short)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_srb(vector unsigned short __a, vector unsigned short __b) {
- return (vector unsigned short)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_srb(vector signed int __a, vector signed int __b) {
- return (vector signed int)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_srb(vector signed int __a, vector unsigned int __b) {
- return (vector signed int)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_srb(vector unsigned int __a, vector signed int __b) {
- return (vector unsigned int)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_srb(vector unsigned int __a, vector unsigned int __b) {
- return (vector unsigned int)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_srb(vector signed long long __a, vector signed long long __b) {
- return (vector signed long long)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_srb(vector signed long long __a, vector unsigned long long __b) {
- return (vector signed long long)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_srb(vector unsigned long long __a, vector signed long long __b) {
- return (vector unsigned long long)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_srb(vector unsigned long long __a, vector unsigned long long __b) {
- return (vector unsigned long long)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_srb(vector float __a, vector signed int __b) {
- return (vector float)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector float
-vec_srb(vector float __a, vector unsigned int __b) {
- return (vector float)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_srb(vector double __a, vector signed long long __b) {
- return (vector double)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector double
-vec_srb(vector double __a, vector unsigned long long __b) {
- return (vector double)__builtin_s390_vsrlb(
- (vector unsigned char)__a, (vector unsigned char)__b);
-}
-
-/*-- vec_abs ----------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_abs(vector signed char __a) {
- return vec_sel(__a, -__a, vec_cmplt(__a, (vector signed char)0));
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_abs(vector signed short __a) {
- return vec_sel(__a, -__a, vec_cmplt(__a, (vector signed short)0));
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_abs(vector signed int __a) {
- return vec_sel(__a, -__a, vec_cmplt(__a, (vector signed int)0));
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_abs(vector signed long long __a) {
- return vec_sel(__a, -__a, vec_cmplt(__a, (vector signed long long)0));
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_abs(vector float __a) {
- return __builtin_s390_vflpsb(__a);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_abs(vector double __a) {
- return __builtin_s390_vflpdb(__a);
-}
-
-/*-- vec_nabs ---------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_nabs(vector float __a) {
- return __builtin_s390_vflnsb(__a);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_nabs(vector double __a) {
- return __builtin_s390_vflndb(__a);
-}
-
-/*-- vec_max ----------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_max(vector signed char __a, vector signed char __b) {
- return vec_sel(__b, __a, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_max(vector signed char __a, vector bool char __b) {
- vector signed char __bc = (vector signed char)__b;
- return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_max(vector bool char __a, vector signed char __b) {
- vector signed char __ac = (vector signed char)__a;
- return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_max(vector unsigned char __a, vector unsigned char __b) {
- return vec_sel(__b, __a, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_max(vector unsigned char __a, vector bool char __b) {
- vector unsigned char __bc = (vector unsigned char)__b;
- return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_max(vector bool char __a, vector unsigned char __b) {
- vector unsigned char __ac = (vector unsigned char)__a;
- return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_max(vector signed short __a, vector signed short __b) {
- return vec_sel(__b, __a, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_max(vector signed short __a, vector bool short __b) {
- vector signed short __bc = (vector signed short)__b;
- return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_max(vector bool short __a, vector signed short __b) {
- vector signed short __ac = (vector signed short)__a;
- return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_max(vector unsigned short __a, vector unsigned short __b) {
- return vec_sel(__b, __a, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_max(vector unsigned short __a, vector bool short __b) {
- vector unsigned short __bc = (vector unsigned short)__b;
- return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_max(vector bool short __a, vector unsigned short __b) {
- vector unsigned short __ac = (vector unsigned short)__a;
- return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_max(vector signed int __a, vector signed int __b) {
- return vec_sel(__b, __a, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_max(vector signed int __a, vector bool int __b) {
- vector signed int __bc = (vector signed int)__b;
- return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_max(vector bool int __a, vector signed int __b) {
- vector signed int __ac = (vector signed int)__a;
- return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_max(vector unsigned int __a, vector unsigned int __b) {
- return vec_sel(__b, __a, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_max(vector unsigned int __a, vector bool int __b) {
- vector unsigned int __bc = (vector unsigned int)__b;
- return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_max(vector bool int __a, vector unsigned int __b) {
- vector unsigned int __ac = (vector unsigned int)__a;
- return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_max(vector signed long long __a, vector signed long long __b) {
- return vec_sel(__b, __a, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_max(vector signed long long __a, vector bool long long __b) {
- vector signed long long __bc = (vector signed long long)__b;
- return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_max(vector bool long long __a, vector signed long long __b) {
- vector signed long long __ac = (vector signed long long)__a;
- return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_max(vector unsigned long long __a, vector unsigned long long __b) {
- return vec_sel(__b, __a, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_max(vector unsigned long long __a, vector bool long long __b) {
- vector unsigned long long __bc = (vector unsigned long long)__b;
- return vec_sel(__bc, __a, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_max(vector bool long long __a, vector unsigned long long __b) {
- vector unsigned long long __ac = (vector unsigned long long)__a;
- return vec_sel(__b, __ac, vec_cmpgt(__ac, __b));
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_max(vector float __a, vector float __b) {
- return __builtin_s390_vfmaxsb(__a, __b, 0);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_max(vector double __a, vector double __b) {
-#if __ARCH__ >= 12
- return __builtin_s390_vfmaxdb(__a, __b, 0);
-#else
- return vec_sel(__b, __a, vec_cmpgt(__a, __b));
-#endif
-}
-
-/*-- vec_min ----------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_min(vector signed char __a, vector signed char __b) {
- return vec_sel(__a, __b, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_min(vector signed char __a, vector bool char __b) {
- vector signed char __bc = (vector signed char)__b;
- return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed char
-vec_min(vector bool char __a, vector signed char __b) {
- vector signed char __ac = (vector signed char)__a;
- return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_min(vector unsigned char __a, vector unsigned char __b) {
- return vec_sel(__a, __b, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_min(vector unsigned char __a, vector bool char __b) {
- vector unsigned char __bc = (vector unsigned char)__b;
- return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned char
-vec_min(vector bool char __a, vector unsigned char __b) {
- vector unsigned char __ac = (vector unsigned char)__a;
- return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_min(vector signed short __a, vector signed short __b) {
- return vec_sel(__a, __b, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_min(vector signed short __a, vector bool short __b) {
- vector signed short __bc = (vector signed short)__b;
- return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed short
-vec_min(vector bool short __a, vector signed short __b) {
- vector signed short __ac = (vector signed short)__a;
- return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_min(vector unsigned short __a, vector unsigned short __b) {
- return vec_sel(__a, __b, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_min(vector unsigned short __a, vector bool short __b) {
- vector unsigned short __bc = (vector unsigned short)__b;
- return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned short
-vec_min(vector bool short __a, vector unsigned short __b) {
- vector unsigned short __ac = (vector unsigned short)__a;
- return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_min(vector signed int __a, vector signed int __b) {
- return vec_sel(__a, __b, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_min(vector signed int __a, vector bool int __b) {
- vector signed int __bc = (vector signed int)__b;
- return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed int
-vec_min(vector bool int __a, vector signed int __b) {
- vector signed int __ac = (vector signed int)__a;
- return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_min(vector unsigned int __a, vector unsigned int __b) {
- return vec_sel(__a, __b, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_min(vector unsigned int __a, vector bool int __b) {
- vector unsigned int __bc = (vector unsigned int)__b;
- return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned int
-vec_min(vector bool int __a, vector unsigned int __b) {
- vector unsigned int __ac = (vector unsigned int)__a;
- return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_min(vector signed long long __a, vector signed long long __b) {
- return vec_sel(__a, __b, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_min(vector signed long long __a, vector bool long long __b) {
- vector signed long long __bc = (vector signed long long)__b;
- return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_min(vector bool long long __a, vector signed long long __b) {
- vector signed long long __ac = (vector signed long long)__a;
- return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_min(vector unsigned long long __a, vector unsigned long long __b) {
- return vec_sel(__a, __b, vec_cmpgt(__a, __b));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_min(vector unsigned long long __a, vector bool long long __b) {
- vector unsigned long long __bc = (vector unsigned long long)__b;
- return vec_sel(__a, __bc, vec_cmpgt(__a, __bc));
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_min(vector bool long long __a, vector unsigned long long __b) {
- vector unsigned long long __ac = (vector unsigned long long)__a;
- return vec_sel(__ac, __b, vec_cmpgt(__ac, __b));
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_min(vector float __a, vector float __b) {
- return __builtin_s390_vfminsb(__a, __b, 0);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_min(vector double __a, vector double __b) {
-#if __ARCH__ >= 12
- return __builtin_s390_vfmindb(__a, __b, 0);
-#else
- return vec_sel(__a, __b, vec_cmpgt(__a, __b));
-#endif
-}
-
-/*-- vec_add_u128 -----------------------------------------------------------*/
-
-static inline __ATTRS_ai vector unsigned char
-vec_add_u128(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vaq(__a, __b);
-}
-
-/*-- vec_addc ---------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_addc(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vaccb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_addc(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vacch(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_addc(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vaccf(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_addc(vector unsigned long long __a, vector unsigned long long __b) {
- return __builtin_s390_vaccg(__a, __b);
-}
-
-/*-- vec_addc_u128 ----------------------------------------------------------*/
-
-static inline __ATTRS_ai vector unsigned char
-vec_addc_u128(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vaccq(__a, __b);
-}
-
-/*-- vec_adde_u128 ----------------------------------------------------------*/
-
-static inline __ATTRS_ai vector unsigned char
-vec_adde_u128(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c) {
- return __builtin_s390_vacq(__a, __b, __c);
-}
-
-/*-- vec_addec_u128 ---------------------------------------------------------*/
-
-static inline __ATTRS_ai vector unsigned char
-vec_addec_u128(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c) {
- return __builtin_s390_vacccq(__a, __b, __c);
-}
-
-/*-- vec_avg ----------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_avg(vector signed char __a, vector signed char __b) {
- return __builtin_s390_vavgb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_avg(vector signed short __a, vector signed short __b) {
- return __builtin_s390_vavgh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_avg(vector signed int __a, vector signed int __b) {
- return __builtin_s390_vavgf(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_avg(vector signed long long __a, vector signed long long __b) {
- return __builtin_s390_vavgg(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_avg(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vavglb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_avg(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vavglh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_avg(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vavglf(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_avg(vector unsigned long long __a, vector unsigned long long __b) {
- return __builtin_s390_vavglg(__a, __b);
-}
-
-/*-- vec_checksum -----------------------------------------------------------*/
-
-static inline __ATTRS_ai vector unsigned int
-vec_checksum(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vcksm(__a, __b);
-}
-
-/*-- vec_gfmsum -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_gfmsum(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vgfmb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_gfmsum(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vgfmh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_gfmsum(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vgfmf(__a, __b);
-}
-
-/*-- vec_gfmsum_128 ---------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_gfmsum_128(vector unsigned long long __a, vector unsigned long long __b) {
- return __builtin_s390_vgfmg(__a, __b);
-}
-
-/*-- vec_gfmsum_accum -------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_gfmsum_accum(vector unsigned char __a, vector unsigned char __b,
- vector unsigned short __c) {
- return __builtin_s390_vgfmab(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_gfmsum_accum(vector unsigned short __a, vector unsigned short __b,
- vector unsigned int __c) {
- return __builtin_s390_vgfmah(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_gfmsum_accum(vector unsigned int __a, vector unsigned int __b,
- vector unsigned long long __c) {
- return __builtin_s390_vgfmaf(__a, __b, __c);
-}
-
-/*-- vec_gfmsum_accum_128 ---------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_gfmsum_accum_128(vector unsigned long long __a,
- vector unsigned long long __b,
- vector unsigned char __c) {
- return __builtin_s390_vgfmag(__a, __b, __c);
-}
-
-/*-- vec_mladd --------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_mladd(vector signed char __a, vector signed char __b,
- vector signed char __c) {
- return __a * __b + __c;
-}
-
-static inline __ATTRS_o_ai vector signed char
-vec_mladd(vector unsigned char __a, vector signed char __b,
- vector signed char __c) {
- return (vector signed char)__a * __b + __c;
-}
-
-static inline __ATTRS_o_ai vector signed char
-vec_mladd(vector signed char __a, vector unsigned char __b,
- vector unsigned char __c) {
- return __a * (vector signed char)__b + (vector signed char)__c;
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_mladd(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c) {
- return __a * __b + __c;
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_mladd(vector signed short __a, vector signed short __b,
- vector signed short __c) {
- return __a * __b + __c;
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_mladd(vector unsigned short __a, vector signed short __b,
- vector signed short __c) {
- return (vector signed short)__a * __b + __c;
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_mladd(vector signed short __a, vector unsigned short __b,
- vector unsigned short __c) {
- return __a * (vector signed short)__b + (vector signed short)__c;
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_mladd(vector unsigned short __a, vector unsigned short __b,
- vector unsigned short __c) {
- return __a * __b + __c;
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_mladd(vector signed int __a, vector signed int __b,
- vector signed int __c) {
- return __a * __b + __c;
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_mladd(vector unsigned int __a, vector signed int __b,
- vector signed int __c) {
- return (vector signed int)__a * __b + __c;
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_mladd(vector signed int __a, vector unsigned int __b,
- vector unsigned int __c) {
- return __a * (vector signed int)__b + (vector signed int)__c;
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_mladd(vector unsigned int __a, vector unsigned int __b,
- vector unsigned int __c) {
- return __a * __b + __c;
-}
-
-/*-- vec_mhadd --------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_mhadd(vector signed char __a, vector signed char __b,
- vector signed char __c) {
- return __builtin_s390_vmahb(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_mhadd(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c) {
- return __builtin_s390_vmalhb(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_mhadd(vector signed short __a, vector signed short __b,
- vector signed short __c) {
- return __builtin_s390_vmahh(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_mhadd(vector unsigned short __a, vector unsigned short __b,
- vector unsigned short __c) {
- return __builtin_s390_vmalhh(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_mhadd(vector signed int __a, vector signed int __b,
- vector signed int __c) {
- return __builtin_s390_vmahf(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_mhadd(vector unsigned int __a, vector unsigned int __b,
- vector unsigned int __c) {
- return __builtin_s390_vmalhf(__a, __b, __c);
-}
-
-/*-- vec_meadd --------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed short
-vec_meadd(vector signed char __a, vector signed char __b,
- vector signed short __c) {
- return __builtin_s390_vmaeb(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_meadd(vector unsigned char __a, vector unsigned char __b,
- vector unsigned short __c) {
- return __builtin_s390_vmaleb(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_meadd(vector signed short __a, vector signed short __b,
- vector signed int __c) {
- return __builtin_s390_vmaeh(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_meadd(vector unsigned short __a, vector unsigned short __b,
- vector unsigned int __c) {
- return __builtin_s390_vmaleh(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_meadd(vector signed int __a, vector signed int __b,
- vector signed long long __c) {
- return __builtin_s390_vmaef(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_meadd(vector unsigned int __a, vector unsigned int __b,
- vector unsigned long long __c) {
- return __builtin_s390_vmalef(__a, __b, __c);
-}
-
-/*-- vec_moadd --------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed short
-vec_moadd(vector signed char __a, vector signed char __b,
- vector signed short __c) {
- return __builtin_s390_vmaob(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_moadd(vector unsigned char __a, vector unsigned char __b,
- vector unsigned short __c) {
- return __builtin_s390_vmalob(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_moadd(vector signed short __a, vector signed short __b,
- vector signed int __c) {
- return __builtin_s390_vmaoh(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_moadd(vector unsigned short __a, vector unsigned short __b,
- vector unsigned int __c) {
- return __builtin_s390_vmaloh(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_moadd(vector signed int __a, vector signed int __b,
- vector signed long long __c) {
- return __builtin_s390_vmaof(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_moadd(vector unsigned int __a, vector unsigned int __b,
- vector unsigned long long __c) {
- return __builtin_s390_vmalof(__a, __b, __c);
-}
-
-/*-- vec_mulh ---------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_mulh(vector signed char __a, vector signed char __b) {
- return __builtin_s390_vmhb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_mulh(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vmlhb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_mulh(vector signed short __a, vector signed short __b) {
- return __builtin_s390_vmhh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_mulh(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vmlhh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_mulh(vector signed int __a, vector signed int __b) {
- return __builtin_s390_vmhf(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_mulh(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vmlhf(__a, __b);
-}
-
-/*-- vec_mule ---------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed short
-vec_mule(vector signed char __a, vector signed char __b) {
- return __builtin_s390_vmeb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_mule(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vmleb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_mule(vector signed short __a, vector signed short __b) {
- return __builtin_s390_vmeh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_mule(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vmleh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_mule(vector signed int __a, vector signed int __b) {
- return __builtin_s390_vmef(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_mule(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vmlef(__a, __b);
-}
-
-/*-- vec_mulo ---------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed short
-vec_mulo(vector signed char __a, vector signed char __b) {
- return __builtin_s390_vmob(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_mulo(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vmlob(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_mulo(vector signed short __a, vector signed short __b) {
- return __builtin_s390_vmoh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_mulo(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vmloh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed long long
-vec_mulo(vector signed int __a, vector signed int __b) {
- return __builtin_s390_vmof(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_mulo(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vmlof(__a, __b);
-}
-
-/*-- vec_msum_u128 ----------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-#define vec_msum_u128(X, Y, Z, W) \
- ((vector unsigned char)__builtin_s390_vmslg((X), (Y), (Z), (W)));
-#endif
-
-/*-- vec_sub_u128 -----------------------------------------------------------*/
-
-static inline __ATTRS_ai vector unsigned char
-vec_sub_u128(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vsq(__a, __b);
-}
-
-/*-- vec_subc ---------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_subc(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vscbib(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_subc(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vscbih(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_subc(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vscbif(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_subc(vector unsigned long long __a, vector unsigned long long __b) {
- return __builtin_s390_vscbig(__a, __b);
-}
-
-/*-- vec_subc_u128 ----------------------------------------------------------*/
-
-static inline __ATTRS_ai vector unsigned char
-vec_subc_u128(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vscbiq(__a, __b);
-}
-
-/*-- vec_sube_u128 ----------------------------------------------------------*/
-
-static inline __ATTRS_ai vector unsigned char
-vec_sube_u128(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c) {
- return __builtin_s390_vsbiq(__a, __b, __c);
-}
-
-/*-- vec_subec_u128 ---------------------------------------------------------*/
-
-static inline __ATTRS_ai vector unsigned char
-vec_subec_u128(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c) {
- return __builtin_s390_vsbcbiq(__a, __b, __c);
-}
-
-/*-- vec_sum2 ---------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_sum2(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vsumgh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_sum2(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vsumgf(__a, __b);
-}
-
-/*-- vec_sum_u128 -----------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_sum_u128(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vsumqf(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_sum_u128(vector unsigned long long __a, vector unsigned long long __b) {
- return __builtin_s390_vsumqg(__a, __b);
-}
-
-/*-- vec_sum4 ---------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_sum4(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vsumb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_sum4(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vsumh(__a, __b);
-}
-
-/*-- vec_test_mask ----------------------------------------------------------*/
-
-static inline __ATTRS_o_ai int
-vec_test_mask(vector signed char __a, vector unsigned char __b) {
- return __builtin_s390_vtm((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai int
-vec_test_mask(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vtm(__a, __b);
-}
-
-static inline __ATTRS_o_ai int
-vec_test_mask(vector signed short __a, vector unsigned short __b) {
- return __builtin_s390_vtm((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai int
-vec_test_mask(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vtm((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai int
-vec_test_mask(vector signed int __a, vector unsigned int __b) {
- return __builtin_s390_vtm((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai int
-vec_test_mask(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vtm((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai int
-vec_test_mask(vector signed long long __a, vector unsigned long long __b) {
- return __builtin_s390_vtm((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai int
-vec_test_mask(vector unsigned long long __a, vector unsigned long long __b) {
- return __builtin_s390_vtm((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai int
-vec_test_mask(vector float __a, vector unsigned int __b) {
- return __builtin_s390_vtm((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-#endif
-
-static inline __ATTRS_o_ai int
-vec_test_mask(vector double __a, vector unsigned long long __b) {
- return __builtin_s390_vtm((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-/*-- vec_madd ---------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_madd(vector float __a, vector float __b, vector float __c) {
- return __builtin_s390_vfmasb(__a, __b, __c);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_madd(vector double __a, vector double __b, vector double __c) {
- return __builtin_s390_vfmadb(__a, __b, __c);
-}
-
-/*-- vec_msub ---------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_msub(vector float __a, vector float __b, vector float __c) {
- return __builtin_s390_vfmssb(__a, __b, __c);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_msub(vector double __a, vector double __b, vector double __c) {
- return __builtin_s390_vfmsdb(__a, __b, __c);
-}
-
-/*-- vec_nmadd ---------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_nmadd(vector float __a, vector float __b, vector float __c) {
- return __builtin_s390_vfnmasb(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector double
-vec_nmadd(vector double __a, vector double __b, vector double __c) {
- return __builtin_s390_vfnmadb(__a, __b, __c);
-}
-#endif
-
-/*-- vec_nmsub ---------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_nmsub(vector float __a, vector float __b, vector float __c) {
- return __builtin_s390_vfnmssb(__a, __b, __c);
-}
-
-static inline __ATTRS_o_ai vector double
-vec_nmsub(vector double __a, vector double __b, vector double __c) {
- return __builtin_s390_vfnmsdb(__a, __b, __c);
-}
-#endif
-
-/*-- vec_sqrt ---------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_sqrt(vector float __a) {
- return __builtin_s390_vfsqsb(__a);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_sqrt(vector double __a) {
- return __builtin_s390_vfsqdb(__a);
-}
-
-/*-- vec_ld2f ---------------------------------------------------------------*/
-
-// This prototype is deprecated.
-static inline __ATTRS_ai vector double
-vec_ld2f(const float *__ptr) {
- typedef float __v2f32 __attribute__((__vector_size__(8)));
- return __builtin_convertvector(*(const __v2f32 *)__ptr, vector double);
-}
-
-/*-- vec_st2f ---------------------------------------------------------------*/
-
-// This prototype is deprecated.
-static inline __ATTRS_ai void
-vec_st2f(vector double __a, float *__ptr) {
- typedef float __v2f32 __attribute__((__vector_size__(8)));
- *(__v2f32 *)__ptr = __builtin_convertvector(__a, __v2f32);
-}
-
-/*-- vec_ctd ----------------------------------------------------------------*/
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector double
-vec_ctd(vector signed long long __a, int __b)
- __constant_range(__b, 0, 31) {
- vector double __conv = __builtin_convertvector(__a, vector double);
- __conv *= (vector double)(vector unsigned long long)((0x3ffULL - __b) << 52);
- return __conv;
-}
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector double
-vec_ctd(vector unsigned long long __a, int __b)
- __constant_range(__b, 0, 31) {
- vector double __conv = __builtin_convertvector(__a, vector double);
- __conv *= (vector double)(vector unsigned long long)((0x3ffULL - __b) << 52);
- return __conv;
-}
-
-/*-- vec_ctsl ---------------------------------------------------------------*/
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector signed long long
-vec_ctsl(vector double __a, int __b)
- __constant_range(__b, 0, 31) {
- __a *= (vector double)(vector unsigned long long)((0x3ffULL + __b) << 52);
- return __builtin_convertvector(__a, vector signed long long);
-}
-
-/*-- vec_ctul ---------------------------------------------------------------*/
-
-// This prototype is deprecated.
-static inline __ATTRS_o_ai vector unsigned long long
-vec_ctul(vector double __a, int __b)
- __constant_range(__b, 0, 31) {
- __a *= (vector double)(vector unsigned long long)((0x3ffULL + __b) << 52);
- return __builtin_convertvector(__a, vector unsigned long long);
-}
-
-/*-- vec_doublee ------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_ai vector double
-vec_doublee(vector float __a) {
- typedef float __v2f32 __attribute__((__vector_size__(8)));
- __v2f32 __pack = __builtin_shufflevector(__a, __a, 0, 2);
- return __builtin_convertvector(__pack, vector double);
-}
-#endif
-
-/*-- vec_floate -------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_ai vector float
-vec_floate(vector double __a) {
- typedef float __v2f32 __attribute__((__vector_size__(8)));
- __v2f32 __pack = __builtin_convertvector(__a, __v2f32);
- return __builtin_shufflevector(__pack, __pack, 0, -1, 1, -1);
-}
-#endif
-
-/*-- vec_double -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector double
-vec_double(vector signed long long __a) {
- return __builtin_convertvector(__a, vector double);
-}
-
-static inline __ATTRS_o_ai vector double
-vec_double(vector unsigned long long __a) {
- return __builtin_convertvector(__a, vector double);
-}
-
-/*-- vec_signed -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed long long
-vec_signed(vector double __a) {
- return __builtin_convertvector(__a, vector signed long long);
-}
-
-/*-- vec_unsigned -----------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned long long
-vec_unsigned(vector double __a) {
- return __builtin_convertvector(__a, vector unsigned long long);
-}
-
-/*-- vec_roundp -------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_roundp(vector float __a) {
- return __builtin_s390_vfisb(__a, 4, 6);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_roundp(vector double __a) {
- return __builtin_s390_vfidb(__a, 4, 6);
-}
-
-/*-- vec_ceil ---------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_ceil(vector float __a) {
- // On this platform, vec_ceil never triggers the IEEE-inexact exception.
- return __builtin_s390_vfisb(__a, 4, 6);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_ceil(vector double __a) {
- // On this platform, vec_ceil never triggers the IEEE-inexact exception.
- return __builtin_s390_vfidb(__a, 4, 6);
-}
-
-/*-- vec_roundm -------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_roundm(vector float __a) {
- return __builtin_s390_vfisb(__a, 4, 7);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_roundm(vector double __a) {
- return __builtin_s390_vfidb(__a, 4, 7);
-}
-
-/*-- vec_floor --------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_floor(vector float __a) {
- // On this platform, vec_floor never triggers the IEEE-inexact exception.
- return __builtin_s390_vfisb(__a, 4, 7);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_floor(vector double __a) {
- // On this platform, vec_floor never triggers the IEEE-inexact exception.
- return __builtin_s390_vfidb(__a, 4, 7);
-}
-
-/*-- vec_roundz -------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_roundz(vector float __a) {
- return __builtin_s390_vfisb(__a, 4, 5);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_roundz(vector double __a) {
- return __builtin_s390_vfidb(__a, 4, 5);
-}
-
-/*-- vec_trunc --------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_trunc(vector float __a) {
- // On this platform, vec_trunc never triggers the IEEE-inexact exception.
- return __builtin_s390_vfisb(__a, 4, 5);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_trunc(vector double __a) {
- // On this platform, vec_trunc never triggers the IEEE-inexact exception.
- return __builtin_s390_vfidb(__a, 4, 5);
-}
-
-/*-- vec_roundc -------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_roundc(vector float __a) {
- return __builtin_s390_vfisb(__a, 4, 0);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_roundc(vector double __a) {
- return __builtin_s390_vfidb(__a, 4, 0);
-}
-
-/*-- vec_rint ---------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_rint(vector float __a) {
- // vec_rint may trigger the IEEE-inexact exception.
- return __builtin_s390_vfisb(__a, 0, 0);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_rint(vector double __a) {
- // vec_rint may trigger the IEEE-inexact exception.
- return __builtin_s390_vfidb(__a, 0, 0);
-}
-
-/*-- vec_round --------------------------------------------------------------*/
-
-#if __ARCH__ >= 12
-static inline __ATTRS_o_ai vector float
-vec_round(vector float __a) {
- return __builtin_s390_vfisb(__a, 4, 4);
-}
-#endif
-
-static inline __ATTRS_o_ai vector double
-vec_round(vector double __a) {
- return __builtin_s390_vfidb(__a, 4, 4);
-}
-
-/*-- vec_fp_test_data_class -------------------------------------------------*/
-
-#if __ARCH__ >= 12
-extern __ATTRS_o vector bool int
-vec_fp_test_data_class(vector float __a, int __b, int *__c)
- __constant_range(__b, 0, 4095);
-
-extern __ATTRS_o vector bool long long
-vec_fp_test_data_class(vector double __a, int __b, int *__c)
- __constant_range(__b, 0, 4095);
-
-#define vec_fp_test_data_class(X, Y, Z) \
- ((__typeof__((vec_fp_test_data_class)((X), (Y), (Z)))) \
- __extension__ ({ \
- vector unsigned char __res; \
- vector unsigned char __x = (vector unsigned char)(X); \
- int *__z = (Z); \
- switch (sizeof ((X)[0])) { \
- case 4: __res = (vector unsigned char) \
- __builtin_s390_vftcisb((vector float)__x, (Y), __z); \
- break; \
- default: __res = (vector unsigned char) \
- __builtin_s390_vftcidb((vector double)__x, (Y), __z); \
- break; \
- } __res; }))
-#else
-#define vec_fp_test_data_class(X, Y, Z) \
- ((vector bool long long)__builtin_s390_vftcidb((X), (Y), (Z)))
-#endif
-
-#define __VEC_CLASS_FP_ZERO_P (1 << 11)
-#define __VEC_CLASS_FP_ZERO_N (1 << 10)
-#define __VEC_CLASS_FP_ZERO (__VEC_CLASS_FP_ZERO_P | __VEC_CLASS_FP_ZERO_N)
-#define __VEC_CLASS_FP_NORMAL_P (1 << 9)
-#define __VEC_CLASS_FP_NORMAL_N (1 << 8)
-#define __VEC_CLASS_FP_NORMAL (__VEC_CLASS_FP_NORMAL_P | \
- __VEC_CLASS_FP_NORMAL_N)
-#define __VEC_CLASS_FP_SUBNORMAL_P (1 << 7)
-#define __VEC_CLASS_FP_SUBNORMAL_N (1 << 6)
-#define __VEC_CLASS_FP_SUBNORMAL (__VEC_CLASS_FP_SUBNORMAL_P | \
- __VEC_CLASS_FP_SUBNORMAL_N)
-#define __VEC_CLASS_FP_INFINITY_P (1 << 5)
-#define __VEC_CLASS_FP_INFINITY_N (1 << 4)
-#define __VEC_CLASS_FP_INFINITY (__VEC_CLASS_FP_INFINITY_P | \
- __VEC_CLASS_FP_INFINITY_N)
-#define __VEC_CLASS_FP_QNAN_P (1 << 3)
-#define __VEC_CLASS_FP_QNAN_N (1 << 2)
-#define __VEC_CLASS_FP_QNAN (__VEC_CLASS_FP_QNAN_P | __VEC_CLASS_FP_QNAN_N)
-#define __VEC_CLASS_FP_SNAN_P (1 << 1)
-#define __VEC_CLASS_FP_SNAN_N (1 << 0)
-#define __VEC_CLASS_FP_SNAN (__VEC_CLASS_FP_SNAN_P | __VEC_CLASS_FP_SNAN_N)
-#define __VEC_CLASS_FP_NAN (__VEC_CLASS_FP_QNAN | __VEC_CLASS_FP_SNAN)
-#define __VEC_CLASS_FP_NOT_NORMAL (__VEC_CLASS_FP_NAN | \
- __VEC_CLASS_FP_SUBNORMAL | \
- __VEC_CLASS_FP_ZERO | \
- __VEC_CLASS_FP_INFINITY)
-
-/*-- vec_cp_until_zero ------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_cp_until_zero(vector signed char __a) {
- return (vector signed char)__builtin_s390_vistrb((vector unsigned char)__a);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_cp_until_zero(vector bool char __a) {
- return (vector bool char)__builtin_s390_vistrb((vector unsigned char)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cp_until_zero(vector unsigned char __a) {
- return __builtin_s390_vistrb(__a);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_cp_until_zero(vector signed short __a) {
- return (vector signed short)__builtin_s390_vistrh((vector unsigned short)__a);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cp_until_zero(vector bool short __a) {
- return (vector bool short)__builtin_s390_vistrh((vector unsigned short)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cp_until_zero(vector unsigned short __a) {
- return __builtin_s390_vistrh(__a);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_cp_until_zero(vector signed int __a) {
- return (vector signed int)__builtin_s390_vistrf((vector unsigned int)__a);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cp_until_zero(vector bool int __a) {
- return (vector bool int)__builtin_s390_vistrf((vector unsigned int)__a);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cp_until_zero(vector unsigned int __a) {
- return __builtin_s390_vistrf(__a);
-}
-
-/*-- vec_cp_until_zero_cc ---------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_cp_until_zero_cc(vector signed char __a, int *__cc) {
- return (vector signed char)
- __builtin_s390_vistrbs((vector unsigned char)__a, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_cp_until_zero_cc(vector bool char __a, int *__cc) {
- return (vector bool char)
- __builtin_s390_vistrbs((vector unsigned char)__a, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cp_until_zero_cc(vector unsigned char __a, int *__cc) {
- return __builtin_s390_vistrbs(__a, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_cp_until_zero_cc(vector signed short __a, int *__cc) {
- return (vector signed short)
- __builtin_s390_vistrhs((vector unsigned short)__a, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cp_until_zero_cc(vector bool short __a, int *__cc) {
- return (vector bool short)
- __builtin_s390_vistrhs((vector unsigned short)__a, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cp_until_zero_cc(vector unsigned short __a, int *__cc) {
- return __builtin_s390_vistrhs(__a, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_cp_until_zero_cc(vector signed int __a, int *__cc) {
- return (vector signed int)
- __builtin_s390_vistrfs((vector unsigned int)__a, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cp_until_zero_cc(vector bool int __a, int *__cc) {
- return (vector bool int)__builtin_s390_vistrfs((vector unsigned int)__a,
- __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cp_until_zero_cc(vector unsigned int __a, int *__cc) {
- return __builtin_s390_vistrfs(__a, __cc);
-}
-
-/*-- vec_cmpeq_idx ----------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_cmpeq_idx(vector signed char __a, vector signed char __b) {
- return (vector signed char)
- __builtin_s390_vfeeb((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpeq_idx(vector bool char __a, vector bool char __b) {
- return __builtin_s390_vfeeb((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpeq_idx(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vfeeb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_cmpeq_idx(vector signed short __a, vector signed short __b) {
- return (vector signed short)
- __builtin_s390_vfeeh((vector unsigned short)__a,
- (vector unsigned short)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpeq_idx(vector bool short __a, vector bool short __b) {
- return __builtin_s390_vfeeh((vector unsigned short)__a,
- (vector unsigned short)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpeq_idx(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vfeeh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_cmpeq_idx(vector signed int __a, vector signed int __b) {
- return (vector signed int)
- __builtin_s390_vfeef((vector unsigned int)__a,
- (vector unsigned int)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpeq_idx(vector bool int __a, vector bool int __b) {
- return __builtin_s390_vfeef((vector unsigned int)__a,
- (vector unsigned int)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpeq_idx(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vfeef(__a, __b);
-}
-
-/*-- vec_cmpeq_idx_cc -------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_cmpeq_idx_cc(vector signed char __a, vector signed char __b, int *__cc) {
- return (vector signed char)
- __builtin_s390_vfeebs((vector unsigned char)__a,
- (vector unsigned char)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpeq_idx_cc(vector bool char __a, vector bool char __b, int *__cc) {
- return __builtin_s390_vfeebs((vector unsigned char)__a,
- (vector unsigned char)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpeq_idx_cc(vector unsigned char __a, vector unsigned char __b,
- int *__cc) {
- return __builtin_s390_vfeebs(__a, __b, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_cmpeq_idx_cc(vector signed short __a, vector signed short __b, int *__cc) {
- return (vector signed short)
- __builtin_s390_vfeehs((vector unsigned short)__a,
- (vector unsigned short)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpeq_idx_cc(vector bool short __a, vector bool short __b, int *__cc) {
- return __builtin_s390_vfeehs((vector unsigned short)__a,
- (vector unsigned short)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpeq_idx_cc(vector unsigned short __a, vector unsigned short __b,
- int *__cc) {
- return __builtin_s390_vfeehs(__a, __b, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_cmpeq_idx_cc(vector signed int __a, vector signed int __b, int *__cc) {
- return (vector signed int)
- __builtin_s390_vfeefs((vector unsigned int)__a,
- (vector unsigned int)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpeq_idx_cc(vector bool int __a, vector bool int __b, int *__cc) {
- return __builtin_s390_vfeefs((vector unsigned int)__a,
- (vector unsigned int)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpeq_idx_cc(vector unsigned int __a, vector unsigned int __b, int *__cc) {
- return __builtin_s390_vfeefs(__a, __b, __cc);
-}
-
-/*-- vec_cmpeq_or_0_idx -----------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_cmpeq_or_0_idx(vector signed char __a, vector signed char __b) {
- return (vector signed char)
- __builtin_s390_vfeezb((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpeq_or_0_idx(vector bool char __a, vector bool char __b) {
- return __builtin_s390_vfeezb((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpeq_or_0_idx(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vfeezb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_cmpeq_or_0_idx(vector signed short __a, vector signed short __b) {
- return (vector signed short)
- __builtin_s390_vfeezh((vector unsigned short)__a,
- (vector unsigned short)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpeq_or_0_idx(vector bool short __a, vector bool short __b) {
- return __builtin_s390_vfeezh((vector unsigned short)__a,
- (vector unsigned short)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpeq_or_0_idx(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vfeezh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_cmpeq_or_0_idx(vector signed int __a, vector signed int __b) {
- return (vector signed int)
- __builtin_s390_vfeezf((vector unsigned int)__a,
- (vector unsigned int)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpeq_or_0_idx(vector bool int __a, vector bool int __b) {
- return __builtin_s390_vfeezf((vector unsigned int)__a,
- (vector unsigned int)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpeq_or_0_idx(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vfeezf(__a, __b);
-}
-
-/*-- vec_cmpeq_or_0_idx_cc --------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_cmpeq_or_0_idx_cc(vector signed char __a, vector signed char __b,
- int *__cc) {
- return (vector signed char)
- __builtin_s390_vfeezbs((vector unsigned char)__a,
- (vector unsigned char)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpeq_or_0_idx_cc(vector bool char __a, vector bool char __b, int *__cc) {
- return __builtin_s390_vfeezbs((vector unsigned char)__a,
- (vector unsigned char)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpeq_or_0_idx_cc(vector unsigned char __a, vector unsigned char __b,
- int *__cc) {
- return __builtin_s390_vfeezbs(__a, __b, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_cmpeq_or_0_idx_cc(vector signed short __a, vector signed short __b,
- int *__cc) {
- return (vector signed short)
- __builtin_s390_vfeezhs((vector unsigned short)__a,
- (vector unsigned short)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpeq_or_0_idx_cc(vector bool short __a, vector bool short __b, int *__cc) {
- return __builtin_s390_vfeezhs((vector unsigned short)__a,
- (vector unsigned short)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpeq_or_0_idx_cc(vector unsigned short __a, vector unsigned short __b,
- int *__cc) {
- return __builtin_s390_vfeezhs(__a, __b, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_cmpeq_or_0_idx_cc(vector signed int __a, vector signed int __b, int *__cc) {
- return (vector signed int)
- __builtin_s390_vfeezfs((vector unsigned int)__a,
- (vector unsigned int)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpeq_or_0_idx_cc(vector bool int __a, vector bool int __b, int *__cc) {
- return __builtin_s390_vfeezfs((vector unsigned int)__a,
- (vector unsigned int)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpeq_or_0_idx_cc(vector unsigned int __a, vector unsigned int __b,
- int *__cc) {
- return __builtin_s390_vfeezfs(__a, __b, __cc);
-}
-
-/*-- vec_cmpne_idx ----------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_cmpne_idx(vector signed char __a, vector signed char __b) {
- return (vector signed char)
- __builtin_s390_vfeneb((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpne_idx(vector bool char __a, vector bool char __b) {
- return __builtin_s390_vfeneb((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpne_idx(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vfeneb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_cmpne_idx(vector signed short __a, vector signed short __b) {
- return (vector signed short)
- __builtin_s390_vfeneh((vector unsigned short)__a,
- (vector unsigned short)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpne_idx(vector bool short __a, vector bool short __b) {
- return __builtin_s390_vfeneh((vector unsigned short)__a,
- (vector unsigned short)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpne_idx(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vfeneh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_cmpne_idx(vector signed int __a, vector signed int __b) {
- return (vector signed int)
- __builtin_s390_vfenef((vector unsigned int)__a,
- (vector unsigned int)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpne_idx(vector bool int __a, vector bool int __b) {
- return __builtin_s390_vfenef((vector unsigned int)__a,
- (vector unsigned int)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpne_idx(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vfenef(__a, __b);
-}
-
-/*-- vec_cmpne_idx_cc -------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_cmpne_idx_cc(vector signed char __a, vector signed char __b, int *__cc) {
- return (vector signed char)
- __builtin_s390_vfenebs((vector unsigned char)__a,
- (vector unsigned char)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpne_idx_cc(vector bool char __a, vector bool char __b, int *__cc) {
- return __builtin_s390_vfenebs((vector unsigned char)__a,
- (vector unsigned char)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpne_idx_cc(vector unsigned char __a, vector unsigned char __b,
- int *__cc) {
- return __builtin_s390_vfenebs(__a, __b, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_cmpne_idx_cc(vector signed short __a, vector signed short __b, int *__cc) {
- return (vector signed short)
- __builtin_s390_vfenehs((vector unsigned short)__a,
- (vector unsigned short)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpne_idx_cc(vector bool short __a, vector bool short __b, int *__cc) {
- return __builtin_s390_vfenehs((vector unsigned short)__a,
- (vector unsigned short)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpne_idx_cc(vector unsigned short __a, vector unsigned short __b,
- int *__cc) {
- return __builtin_s390_vfenehs(__a, __b, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_cmpne_idx_cc(vector signed int __a, vector signed int __b, int *__cc) {
- return (vector signed int)
- __builtin_s390_vfenefs((vector unsigned int)__a,
- (vector unsigned int)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpne_idx_cc(vector bool int __a, vector bool int __b, int *__cc) {
- return __builtin_s390_vfenefs((vector unsigned int)__a,
- (vector unsigned int)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpne_idx_cc(vector unsigned int __a, vector unsigned int __b, int *__cc) {
- return __builtin_s390_vfenefs(__a, __b, __cc);
-}
-
-/*-- vec_cmpne_or_0_idx -----------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_cmpne_or_0_idx(vector signed char __a, vector signed char __b) {
- return (vector signed char)
- __builtin_s390_vfenezb((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpne_or_0_idx(vector bool char __a, vector bool char __b) {
- return __builtin_s390_vfenezb((vector unsigned char)__a,
- (vector unsigned char)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpne_or_0_idx(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vfenezb(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_cmpne_or_0_idx(vector signed short __a, vector signed short __b) {
- return (vector signed short)
- __builtin_s390_vfenezh((vector unsigned short)__a,
- (vector unsigned short)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpne_or_0_idx(vector bool short __a, vector bool short __b) {
- return __builtin_s390_vfenezh((vector unsigned short)__a,
- (vector unsigned short)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpne_or_0_idx(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vfenezh(__a, __b);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_cmpne_or_0_idx(vector signed int __a, vector signed int __b) {
- return (vector signed int)
- __builtin_s390_vfenezf((vector unsigned int)__a,
- (vector unsigned int)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpne_or_0_idx(vector bool int __a, vector bool int __b) {
- return __builtin_s390_vfenezf((vector unsigned int)__a,
- (vector unsigned int)__b);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpne_or_0_idx(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vfenezf(__a, __b);
-}
-
-/*-- vec_cmpne_or_0_idx_cc --------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_cmpne_or_0_idx_cc(vector signed char __a, vector signed char __b,
- int *__cc) {
- return (vector signed char)
- __builtin_s390_vfenezbs((vector unsigned char)__a,
- (vector unsigned char)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpne_or_0_idx_cc(vector bool char __a, vector bool char __b, int *__cc) {
- return __builtin_s390_vfenezbs((vector unsigned char)__a,
- (vector unsigned char)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpne_or_0_idx_cc(vector unsigned char __a, vector unsigned char __b,
- int *__cc) {
- return __builtin_s390_vfenezbs(__a, __b, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_cmpne_or_0_idx_cc(vector signed short __a, vector signed short __b,
- int *__cc) {
- return (vector signed short)
- __builtin_s390_vfenezhs((vector unsigned short)__a,
- (vector unsigned short)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpne_or_0_idx_cc(vector bool short __a, vector bool short __b, int *__cc) {
- return __builtin_s390_vfenezhs((vector unsigned short)__a,
- (vector unsigned short)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpne_or_0_idx_cc(vector unsigned short __a, vector unsigned short __b,
- int *__cc) {
- return __builtin_s390_vfenezhs(__a, __b, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_cmpne_or_0_idx_cc(vector signed int __a, vector signed int __b, int *__cc) {
- return (vector signed int)
- __builtin_s390_vfenezfs((vector unsigned int)__a,
- (vector unsigned int)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpne_or_0_idx_cc(vector bool int __a, vector bool int __b, int *__cc) {
- return __builtin_s390_vfenezfs((vector unsigned int)__a,
- (vector unsigned int)__b, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpne_or_0_idx_cc(vector unsigned int __a, vector unsigned int __b,
- int *__cc) {
- return __builtin_s390_vfenezfs(__a, __b, __cc);
-}
-
-/*-- vec_cmprg --------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector bool char
-vec_cmprg(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c) {
- return (vector bool char)__builtin_s390_vstrcb(__a, __b, __c, 4);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cmprg(vector unsigned short __a, vector unsigned short __b,
- vector unsigned short __c) {
- return (vector bool short)__builtin_s390_vstrch(__a, __b, __c, 4);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cmprg(vector unsigned int __a, vector unsigned int __b,
- vector unsigned int __c) {
- return (vector bool int)__builtin_s390_vstrcf(__a, __b, __c, 4);
-}
-
-/*-- vec_cmprg_cc -----------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector bool char
-vec_cmprg_cc(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c, int *__cc) {
- return (vector bool char)__builtin_s390_vstrcbs(__a, __b, __c, 4, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cmprg_cc(vector unsigned short __a, vector unsigned short __b,
- vector unsigned short __c, int *__cc) {
- return (vector bool short)__builtin_s390_vstrchs(__a, __b, __c, 4, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cmprg_cc(vector unsigned int __a, vector unsigned int __b,
- vector unsigned int __c, int *__cc) {
- return (vector bool int)__builtin_s390_vstrcfs(__a, __b, __c, 4, __cc);
-}
-
-/*-- vec_cmprg_idx ----------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmprg_idx(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c) {
- return __builtin_s390_vstrcb(__a, __b, __c, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmprg_idx(vector unsigned short __a, vector unsigned short __b,
- vector unsigned short __c) {
- return __builtin_s390_vstrch(__a, __b, __c, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmprg_idx(vector unsigned int __a, vector unsigned int __b,
- vector unsigned int __c) {
- return __builtin_s390_vstrcf(__a, __b, __c, 0);
-}
-
-/*-- vec_cmprg_idx_cc -------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmprg_idx_cc(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c, int *__cc) {
- return __builtin_s390_vstrcbs(__a, __b, __c, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmprg_idx_cc(vector unsigned short __a, vector unsigned short __b,
- vector unsigned short __c, int *__cc) {
- return __builtin_s390_vstrchs(__a, __b, __c, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmprg_idx_cc(vector unsigned int __a, vector unsigned int __b,
- vector unsigned int __c, int *__cc) {
- return __builtin_s390_vstrcfs(__a, __b, __c, 0, __cc);
-}
-
-/*-- vec_cmprg_or_0_idx -----------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmprg_or_0_idx(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c) {
- return __builtin_s390_vstrczb(__a, __b, __c, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmprg_or_0_idx(vector unsigned short __a, vector unsigned short __b,
- vector unsigned short __c) {
- return __builtin_s390_vstrczh(__a, __b, __c, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmprg_or_0_idx(vector unsigned int __a, vector unsigned int __b,
- vector unsigned int __c) {
- return __builtin_s390_vstrczf(__a, __b, __c, 0);
-}
-
-/*-- vec_cmprg_or_0_idx_cc --------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmprg_or_0_idx_cc(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c, int *__cc) {
- return __builtin_s390_vstrczbs(__a, __b, __c, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmprg_or_0_idx_cc(vector unsigned short __a, vector unsigned short __b,
- vector unsigned short __c, int *__cc) {
- return __builtin_s390_vstrczhs(__a, __b, __c, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmprg_or_0_idx_cc(vector unsigned int __a, vector unsigned int __b,
- vector unsigned int __c, int *__cc) {
- return __builtin_s390_vstrczfs(__a, __b, __c, 0, __cc);
-}
-
-/*-- vec_cmpnrg -------------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector bool char
-vec_cmpnrg(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c) {
- return (vector bool char)__builtin_s390_vstrcb(__a, __b, __c, 12);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cmpnrg(vector unsigned short __a, vector unsigned short __b,
- vector unsigned short __c) {
- return (vector bool short)__builtin_s390_vstrch(__a, __b, __c, 12);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cmpnrg(vector unsigned int __a, vector unsigned int __b,
- vector unsigned int __c) {
- return (vector bool int)__builtin_s390_vstrcf(__a, __b, __c, 12);
-}
-
-/*-- vec_cmpnrg_cc ----------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector bool char
-vec_cmpnrg_cc(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c, int *__cc) {
- return (vector bool char)__builtin_s390_vstrcbs(__a, __b, __c, 12, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_cmpnrg_cc(vector unsigned short __a, vector unsigned short __b,
- vector unsigned short __c, int *__cc) {
- return (vector bool short)__builtin_s390_vstrchs(__a, __b, __c, 12, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_cmpnrg_cc(vector unsigned int __a, vector unsigned int __b,
- vector unsigned int __c, int *__cc) {
- return (vector bool int)__builtin_s390_vstrcfs(__a, __b, __c, 12, __cc);
-}
-
-/*-- vec_cmpnrg_idx ---------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpnrg_idx(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c) {
- return __builtin_s390_vstrcb(__a, __b, __c, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpnrg_idx(vector unsigned short __a, vector unsigned short __b,
- vector unsigned short __c) {
- return __builtin_s390_vstrch(__a, __b, __c, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpnrg_idx(vector unsigned int __a, vector unsigned int __b,
- vector unsigned int __c) {
- return __builtin_s390_vstrcf(__a, __b, __c, 8);
-}
-
-/*-- vec_cmpnrg_idx_cc ------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpnrg_idx_cc(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c, int *__cc) {
- return __builtin_s390_vstrcbs(__a, __b, __c, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpnrg_idx_cc(vector unsigned short __a, vector unsigned short __b,
- vector unsigned short __c, int *__cc) {
- return __builtin_s390_vstrchs(__a, __b, __c, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpnrg_idx_cc(vector unsigned int __a, vector unsigned int __b,
- vector unsigned int __c, int *__cc) {
- return __builtin_s390_vstrcfs(__a, __b, __c, 8, __cc);
-}
-
-/*-- vec_cmpnrg_or_0_idx ----------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpnrg_or_0_idx(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c) {
- return __builtin_s390_vstrczb(__a, __b, __c, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpnrg_or_0_idx(vector unsigned short __a, vector unsigned short __b,
- vector unsigned short __c) {
- return __builtin_s390_vstrczh(__a, __b, __c, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpnrg_or_0_idx(vector unsigned int __a, vector unsigned int __b,
- vector unsigned int __c) {
- return __builtin_s390_vstrczf(__a, __b, __c, 8);
-}
-
-/*-- vec_cmpnrg_or_0_idx_cc -------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_cmpnrg_or_0_idx_cc(vector unsigned char __a, vector unsigned char __b,
- vector unsigned char __c, int *__cc) {
- return __builtin_s390_vstrczbs(__a, __b, __c, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_cmpnrg_or_0_idx_cc(vector unsigned short __a, vector unsigned short __b,
- vector unsigned short __c, int *__cc) {
- return __builtin_s390_vstrczhs(__a, __b, __c, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_cmpnrg_or_0_idx_cc(vector unsigned int __a, vector unsigned int __b,
- vector unsigned int __c, int *__cc) {
- return __builtin_s390_vstrczfs(__a, __b, __c, 8, __cc);
-}
-
-/*-- vec_find_any_eq --------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector bool char
-vec_find_any_eq(vector signed char __a, vector signed char __b) {
- return (vector bool char)
- __builtin_s390_vfaeb((vector unsigned char)__a,
- (vector unsigned char)__b, 4);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_find_any_eq(vector bool char __a, vector bool char __b) {
- return (vector bool char)
- __builtin_s390_vfaeb((vector unsigned char)__a,
- (vector unsigned char)__b, 4);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_find_any_eq(vector unsigned char __a, vector unsigned char __b) {
- return (vector bool char)__builtin_s390_vfaeb(__a, __b, 4);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_find_any_eq(vector signed short __a, vector signed short __b) {
- return (vector bool short)
- __builtin_s390_vfaeh((vector unsigned short)__a,
- (vector unsigned short)__b, 4);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_find_any_eq(vector bool short __a, vector bool short __b) {
- return (vector bool short)
- __builtin_s390_vfaeh((vector unsigned short)__a,
- (vector unsigned short)__b, 4);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_find_any_eq(vector unsigned short __a, vector unsigned short __b) {
- return (vector bool short)__builtin_s390_vfaeh(__a, __b, 4);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_find_any_eq(vector signed int __a, vector signed int __b) {
- return (vector bool int)
- __builtin_s390_vfaef((vector unsigned int)__a,
- (vector unsigned int)__b, 4);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_find_any_eq(vector bool int __a, vector bool int __b) {
- return (vector bool int)
- __builtin_s390_vfaef((vector unsigned int)__a,
- (vector unsigned int)__b, 4);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_find_any_eq(vector unsigned int __a, vector unsigned int __b) {
- return (vector bool int)__builtin_s390_vfaef(__a, __b, 4);
-}
-
-/*-- vec_find_any_eq_cc -----------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector bool char
-vec_find_any_eq_cc(vector signed char __a, vector signed char __b, int *__cc) {
- return (vector bool char)
- __builtin_s390_vfaebs((vector unsigned char)__a,
- (vector unsigned char)__b, 4, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_find_any_eq_cc(vector bool char __a, vector bool char __b, int *__cc) {
- return (vector bool char)
- __builtin_s390_vfaebs((vector unsigned char)__a,
- (vector unsigned char)__b, 4, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_find_any_eq_cc(vector unsigned char __a, vector unsigned char __b,
- int *__cc) {
- return (vector bool char)__builtin_s390_vfaebs(__a, __b, 4, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_find_any_eq_cc(vector signed short __a, vector signed short __b,
- int *__cc) {
- return (vector bool short)
- __builtin_s390_vfaehs((vector unsigned short)__a,
- (vector unsigned short)__b, 4, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_find_any_eq_cc(vector bool short __a, vector bool short __b, int *__cc) {
- return (vector bool short)
- __builtin_s390_vfaehs((vector unsigned short)__a,
- (vector unsigned short)__b, 4, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_find_any_eq_cc(vector unsigned short __a, vector unsigned short __b,
- int *__cc) {
- return (vector bool short)__builtin_s390_vfaehs(__a, __b, 4, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_find_any_eq_cc(vector signed int __a, vector signed int __b, int *__cc) {
- return (vector bool int)
- __builtin_s390_vfaefs((vector unsigned int)__a,
- (vector unsigned int)__b, 4, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_find_any_eq_cc(vector bool int __a, vector bool int __b, int *__cc) {
- return (vector bool int)
- __builtin_s390_vfaefs((vector unsigned int)__a,
- (vector unsigned int)__b, 4, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_find_any_eq_cc(vector unsigned int __a, vector unsigned int __b,
- int *__cc) {
- return (vector bool int)__builtin_s390_vfaefs(__a, __b, 4, __cc);
-}
-
-/*-- vec_find_any_eq_idx ----------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_find_any_eq_idx(vector signed char __a, vector signed char __b) {
- return (vector signed char)
- __builtin_s390_vfaeb((vector unsigned char)__a,
- (vector unsigned char)__b, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_eq_idx(vector bool char __a, vector bool char __b) {
- return __builtin_s390_vfaeb((vector unsigned char)__a,
- (vector unsigned char)__b, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_eq_idx(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vfaeb(__a, __b, 0);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_find_any_eq_idx(vector signed short __a, vector signed short __b) {
- return (vector signed short)
- __builtin_s390_vfaeh((vector unsigned short)__a,
- (vector unsigned short)__b, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_eq_idx(vector bool short __a, vector bool short __b) {
- return __builtin_s390_vfaeh((vector unsigned short)__a,
- (vector unsigned short)__b, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_eq_idx(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vfaeh(__a, __b, 0);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_find_any_eq_idx(vector signed int __a, vector signed int __b) {
- return (vector signed int)
- __builtin_s390_vfaef((vector unsigned int)__a,
- (vector unsigned int)__b, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_eq_idx(vector bool int __a, vector bool int __b) {
- return __builtin_s390_vfaef((vector unsigned int)__a,
- (vector unsigned int)__b, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_eq_idx(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vfaef(__a, __b, 0);
-}
-
-/*-- vec_find_any_eq_idx_cc -------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_find_any_eq_idx_cc(vector signed char __a, vector signed char __b,
- int *__cc) {
- return (vector signed char)
- __builtin_s390_vfaebs((vector unsigned char)__a,
- (vector unsigned char)__b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_eq_idx_cc(vector bool char __a, vector bool char __b, int *__cc) {
- return __builtin_s390_vfaebs((vector unsigned char)__a,
- (vector unsigned char)__b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_eq_idx_cc(vector unsigned char __a, vector unsigned char __b,
- int *__cc) {
- return __builtin_s390_vfaebs(__a, __b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_find_any_eq_idx_cc(vector signed short __a, vector signed short __b,
- int *__cc) {
- return (vector signed short)
- __builtin_s390_vfaehs((vector unsigned short)__a,
- (vector unsigned short)__b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_eq_idx_cc(vector bool short __a, vector bool short __b,
- int *__cc) {
- return __builtin_s390_vfaehs((vector unsigned short)__a,
- (vector unsigned short)__b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_eq_idx_cc(vector unsigned short __a, vector unsigned short __b,
- int *__cc) {
- return __builtin_s390_vfaehs(__a, __b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_find_any_eq_idx_cc(vector signed int __a, vector signed int __b,
- int *__cc) {
- return (vector signed int)
- __builtin_s390_vfaefs((vector unsigned int)__a,
- (vector unsigned int)__b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_eq_idx_cc(vector bool int __a, vector bool int __b, int *__cc) {
- return __builtin_s390_vfaefs((vector unsigned int)__a,
- (vector unsigned int)__b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_eq_idx_cc(vector unsigned int __a, vector unsigned int __b,
- int *__cc) {
- return __builtin_s390_vfaefs(__a, __b, 0, __cc);
-}
-
-/*-- vec_find_any_eq_or_0_idx -----------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_find_any_eq_or_0_idx(vector signed char __a, vector signed char __b) {
- return (vector signed char)
- __builtin_s390_vfaezb((vector unsigned char)__a,
- (vector unsigned char)__b, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_eq_or_0_idx(vector bool char __a, vector bool char __b) {
- return __builtin_s390_vfaezb((vector unsigned char)__a,
- (vector unsigned char)__b, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_eq_or_0_idx(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vfaezb(__a, __b, 0);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_find_any_eq_or_0_idx(vector signed short __a, vector signed short __b) {
- return (vector signed short)
- __builtin_s390_vfaezh((vector unsigned short)__a,
- (vector unsigned short)__b, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_eq_or_0_idx(vector bool short __a, vector bool short __b) {
- return __builtin_s390_vfaezh((vector unsigned short)__a,
- (vector unsigned short)__b, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_eq_or_0_idx(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vfaezh(__a, __b, 0);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_find_any_eq_or_0_idx(vector signed int __a, vector signed int __b) {
- return (vector signed int)
- __builtin_s390_vfaezf((vector unsigned int)__a,
- (vector unsigned int)__b, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_eq_or_0_idx(vector bool int __a, vector bool int __b) {
- return __builtin_s390_vfaezf((vector unsigned int)__a,
- (vector unsigned int)__b, 0);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_eq_or_0_idx(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vfaezf(__a, __b, 0);
-}
-
-/*-- vec_find_any_eq_or_0_idx_cc --------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_find_any_eq_or_0_idx_cc(vector signed char __a, vector signed char __b,
- int *__cc) {
- return (vector signed char)
- __builtin_s390_vfaezbs((vector unsigned char)__a,
- (vector unsigned char)__b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_eq_or_0_idx_cc(vector bool char __a, vector bool char __b,
- int *__cc) {
- return __builtin_s390_vfaezbs((vector unsigned char)__a,
- (vector unsigned char)__b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_eq_or_0_idx_cc(vector unsigned char __a, vector unsigned char __b,
- int *__cc) {
- return __builtin_s390_vfaezbs(__a, __b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_find_any_eq_or_0_idx_cc(vector signed short __a, vector signed short __b,
- int *__cc) {
- return (vector signed short)
- __builtin_s390_vfaezhs((vector unsigned short)__a,
- (vector unsigned short)__b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_eq_or_0_idx_cc(vector bool short __a, vector bool short __b,
- int *__cc) {
- return __builtin_s390_vfaezhs((vector unsigned short)__a,
- (vector unsigned short)__b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_eq_or_0_idx_cc(vector unsigned short __a,
- vector unsigned short __b, int *__cc) {
- return __builtin_s390_vfaezhs(__a, __b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_find_any_eq_or_0_idx_cc(vector signed int __a, vector signed int __b,
- int *__cc) {
- return (vector signed int)
- __builtin_s390_vfaezfs((vector unsigned int)__a,
- (vector unsigned int)__b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_eq_or_0_idx_cc(vector bool int __a, vector bool int __b,
- int *__cc) {
- return __builtin_s390_vfaezfs((vector unsigned int)__a,
- (vector unsigned int)__b, 0, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_eq_or_0_idx_cc(vector unsigned int __a, vector unsigned int __b,
- int *__cc) {
- return __builtin_s390_vfaezfs(__a, __b, 0, __cc);
-}
-
-/*-- vec_find_any_ne --------------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector bool char
-vec_find_any_ne(vector signed char __a, vector signed char __b) {
- return (vector bool char)
- __builtin_s390_vfaeb((vector unsigned char)__a,
- (vector unsigned char)__b, 12);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_find_any_ne(vector bool char __a, vector bool char __b) {
- return (vector bool char)
- __builtin_s390_vfaeb((vector unsigned char)__a,
- (vector unsigned char)__b, 12);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_find_any_ne(vector unsigned char __a, vector unsigned char __b) {
- return (vector bool char)__builtin_s390_vfaeb(__a, __b, 12);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_find_any_ne(vector signed short __a, vector signed short __b) {
- return (vector bool short)
- __builtin_s390_vfaeh((vector unsigned short)__a,
- (vector unsigned short)__b, 12);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_find_any_ne(vector bool short __a, vector bool short __b) {
- return (vector bool short)
- __builtin_s390_vfaeh((vector unsigned short)__a,
- (vector unsigned short)__b, 12);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_find_any_ne(vector unsigned short __a, vector unsigned short __b) {
- return (vector bool short)__builtin_s390_vfaeh(__a, __b, 12);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_find_any_ne(vector signed int __a, vector signed int __b) {
- return (vector bool int)
- __builtin_s390_vfaef((vector unsigned int)__a,
- (vector unsigned int)__b, 12);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_find_any_ne(vector bool int __a, vector bool int __b) {
- return (vector bool int)
- __builtin_s390_vfaef((vector unsigned int)__a,
- (vector unsigned int)__b, 12);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_find_any_ne(vector unsigned int __a, vector unsigned int __b) {
- return (vector bool int)__builtin_s390_vfaef(__a, __b, 12);
-}
-
-/*-- vec_find_any_ne_cc -----------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector bool char
-vec_find_any_ne_cc(vector signed char __a, vector signed char __b, int *__cc) {
- return (vector bool char)
- __builtin_s390_vfaebs((vector unsigned char)__a,
- (vector unsigned char)__b, 12, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_find_any_ne_cc(vector bool char __a, vector bool char __b, int *__cc) {
- return (vector bool char)
- __builtin_s390_vfaebs((vector unsigned char)__a,
- (vector unsigned char)__b, 12, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool char
-vec_find_any_ne_cc(vector unsigned char __a, vector unsigned char __b,
- int *__cc) {
- return (vector bool char)__builtin_s390_vfaebs(__a, __b, 12, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_find_any_ne_cc(vector signed short __a, vector signed short __b,
- int *__cc) {
- return (vector bool short)
- __builtin_s390_vfaehs((vector unsigned short)__a,
- (vector unsigned short)__b, 12, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_find_any_ne_cc(vector bool short __a, vector bool short __b, int *__cc) {
- return (vector bool short)
- __builtin_s390_vfaehs((vector unsigned short)__a,
- (vector unsigned short)__b, 12, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool short
-vec_find_any_ne_cc(vector unsigned short __a, vector unsigned short __b,
- int *__cc) {
- return (vector bool short)__builtin_s390_vfaehs(__a, __b, 12, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_find_any_ne_cc(vector signed int __a, vector signed int __b, int *__cc) {
- return (vector bool int)
- __builtin_s390_vfaefs((vector unsigned int)__a,
- (vector unsigned int)__b, 12, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_find_any_ne_cc(vector bool int __a, vector bool int __b, int *__cc) {
- return (vector bool int)
- __builtin_s390_vfaefs((vector unsigned int)__a,
- (vector unsigned int)__b, 12, __cc);
-}
-
-static inline __ATTRS_o_ai vector bool int
-vec_find_any_ne_cc(vector unsigned int __a, vector unsigned int __b,
- int *__cc) {
- return (vector bool int)__builtin_s390_vfaefs(__a, __b, 12, __cc);
-}
-
-/*-- vec_find_any_ne_idx ----------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_find_any_ne_idx(vector signed char __a, vector signed char __b) {
- return (vector signed char)
- __builtin_s390_vfaeb((vector unsigned char)__a,
- (vector unsigned char)__b, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_ne_idx(vector bool char __a, vector bool char __b) {
- return __builtin_s390_vfaeb((vector unsigned char)__a,
- (vector unsigned char)__b, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_ne_idx(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vfaeb(__a, __b, 8);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_find_any_ne_idx(vector signed short __a, vector signed short __b) {
- return (vector signed short)
- __builtin_s390_vfaeh((vector unsigned short)__a,
- (vector unsigned short)__b, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_ne_idx(vector bool short __a, vector bool short __b) {
- return __builtin_s390_vfaeh((vector unsigned short)__a,
- (vector unsigned short)__b, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_ne_idx(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vfaeh(__a, __b, 8);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_find_any_ne_idx(vector signed int __a, vector signed int __b) {
- return (vector signed int)
- __builtin_s390_vfaef((vector unsigned int)__a,
- (vector unsigned int)__b, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_ne_idx(vector bool int __a, vector bool int __b) {
- return __builtin_s390_vfaef((vector unsigned int)__a,
- (vector unsigned int)__b, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_ne_idx(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vfaef(__a, __b, 8);
-}
-
-/*-- vec_find_any_ne_idx_cc -------------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_find_any_ne_idx_cc(vector signed char __a, vector signed char __b,
- int *__cc) {
- return (vector signed char)
- __builtin_s390_vfaebs((vector unsigned char)__a,
- (vector unsigned char)__b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_ne_idx_cc(vector bool char __a, vector bool char __b, int *__cc) {
- return __builtin_s390_vfaebs((vector unsigned char)__a,
- (vector unsigned char)__b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_ne_idx_cc(vector unsigned char __a, vector unsigned char __b,
- int *__cc) {
- return __builtin_s390_vfaebs(__a, __b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_find_any_ne_idx_cc(vector signed short __a, vector signed short __b,
- int *__cc) {
- return (vector signed short)
- __builtin_s390_vfaehs((vector unsigned short)__a,
- (vector unsigned short)__b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_ne_idx_cc(vector bool short __a, vector bool short __b,
- int *__cc) {
- return __builtin_s390_vfaehs((vector unsigned short)__a,
- (vector unsigned short)__b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_ne_idx_cc(vector unsigned short __a, vector unsigned short __b,
- int *__cc) {
- return __builtin_s390_vfaehs(__a, __b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_find_any_ne_idx_cc(vector signed int __a, vector signed int __b,
- int *__cc) {
- return (vector signed int)
- __builtin_s390_vfaefs((vector unsigned int)__a,
- (vector unsigned int)__b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_ne_idx_cc(vector bool int __a, vector bool int __b, int *__cc) {
- return __builtin_s390_vfaefs((vector unsigned int)__a,
- (vector unsigned int)__b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_ne_idx_cc(vector unsigned int __a, vector unsigned int __b,
- int *__cc) {
- return __builtin_s390_vfaefs(__a, __b, 8, __cc);
-}
-
-/*-- vec_find_any_ne_or_0_idx -----------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_find_any_ne_or_0_idx(vector signed char __a, vector signed char __b) {
- return (vector signed char)
- __builtin_s390_vfaezb((vector unsigned char)__a,
- (vector unsigned char)__b, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_ne_or_0_idx(vector bool char __a, vector bool char __b) {
- return __builtin_s390_vfaezb((vector unsigned char)__a,
- (vector unsigned char)__b, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_ne_or_0_idx(vector unsigned char __a, vector unsigned char __b) {
- return __builtin_s390_vfaezb(__a, __b, 8);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_find_any_ne_or_0_idx(vector signed short __a, vector signed short __b) {
- return (vector signed short)
- __builtin_s390_vfaezh((vector unsigned short)__a,
- (vector unsigned short)__b, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_ne_or_0_idx(vector bool short __a, vector bool short __b) {
- return __builtin_s390_vfaezh((vector unsigned short)__a,
- (vector unsigned short)__b, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_ne_or_0_idx(vector unsigned short __a, vector unsigned short __b) {
- return __builtin_s390_vfaezh(__a, __b, 8);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_find_any_ne_or_0_idx(vector signed int __a, vector signed int __b) {
- return (vector signed int)
- __builtin_s390_vfaezf((vector unsigned int)__a,
- (vector unsigned int)__b, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_ne_or_0_idx(vector bool int __a, vector bool int __b) {
- return __builtin_s390_vfaezf((vector unsigned int)__a,
- (vector unsigned int)__b, 8);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_ne_or_0_idx(vector unsigned int __a, vector unsigned int __b) {
- return __builtin_s390_vfaezf(__a, __b, 8);
-}
-
-/*-- vec_find_any_ne_or_0_idx_cc --------------------------------------------*/
-
-static inline __ATTRS_o_ai vector signed char
-vec_find_any_ne_or_0_idx_cc(vector signed char __a, vector signed char __b,
- int *__cc) {
- return (vector signed char)
- __builtin_s390_vfaezbs((vector unsigned char)__a,
- (vector unsigned char)__b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_ne_or_0_idx_cc(vector bool char __a, vector bool char __b,
- int *__cc) {
- return __builtin_s390_vfaezbs((vector unsigned char)__a,
- (vector unsigned char)__b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned char
-vec_find_any_ne_or_0_idx_cc(vector unsigned char __a, vector unsigned char __b,
- int *__cc) {
- return __builtin_s390_vfaezbs(__a, __b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed short
-vec_find_any_ne_or_0_idx_cc(vector signed short __a, vector signed short __b,
- int *__cc) {
- return (vector signed short)
- __builtin_s390_vfaezhs((vector unsigned short)__a,
- (vector unsigned short)__b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_ne_or_0_idx_cc(vector bool short __a, vector bool short __b,
- int *__cc) {
- return __builtin_s390_vfaezhs((vector unsigned short)__a,
- (vector unsigned short)__b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned short
-vec_find_any_ne_or_0_idx_cc(vector unsigned short __a,
- vector unsigned short __b, int *__cc) {
- return __builtin_s390_vfaezhs(__a, __b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector signed int
-vec_find_any_ne_or_0_idx_cc(vector signed int __a, vector signed int __b,
- int *__cc) {
- return (vector signed int)
- __builtin_s390_vfaezfs((vector unsigned int)__a,
- (vector unsigned int)__b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_ne_or_0_idx_cc(vector bool int __a, vector bool int __b,
- int *__cc) {
- return __builtin_s390_vfaezfs((vector unsigned int)__a,
- (vector unsigned int)__b, 8, __cc);
-}
-
-static inline __ATTRS_o_ai vector unsigned int
-vec_find_any_ne_or_0_idx_cc(vector unsigned int __a, vector unsigned int __b,
- int *__cc) {
- return __builtin_s390_vfaezfs(__a, __b, 8, __cc);
-}
-
-#undef __constant_pow2_range
-#undef __constant_range
-#undef __constant
-#undef __ATTRS_o
-#undef __ATTRS_o_ai
-#undef __ATTRS_ai
-
-#else
-
-#error "Use -fzvector to enable vector extensions"
-
-#endif
+++ /dev/null
-/*===---- wmmintrin.h - AES intrinsics ------------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef _WMMINTRIN_H
-#define _WMMINTRIN_H
-
-#include <emmintrin.h>
-
-#include <__wmmintrin_aes.h>
-
-#include <__wmmintrin_pclmul.h>
-
-#endif /* _WMMINTRIN_H */
+++ /dev/null
-/*===---- x86intrin.h - X86 intrinsics -------------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __X86INTRIN_H
-#define __X86INTRIN_H
-
-#include <ia32intrin.h>
-
-#include <immintrin.h>
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__3dNOW__)
-#include <mm3dnow.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__BMI__)
-#include <bmiintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__BMI2__)
-#include <bmi2intrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__LZCNT__)
-#include <lzcntintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__POPCNT__)
-#include <popcntintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__RDSEED__)
-#include <rdseedintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__PRFCHW__)
-#include <prfchwintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__SSE4A__)
-#include <ammintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__FMA4__)
-#include <fma4intrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__XOP__)
-#include <xopintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__TBM__)
-#include <tbmintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__LWP__)
-#include <lwpintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__F16C__)
-#include <f16cintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__MWAITX__)
-#include <mwaitxintrin.h>
-#endif
-
-#if !defined(_MSC_VER) || __has_feature(modules) || defined(__CLZERO__)
-#include <clzerointrin.h>
-#endif
-
-#endif /* __X86INTRIN_H */
+++ /dev/null
-/*===---- xsavecintrin.h - XSAVEC intrinsic ------------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __IMMINTRIN_H
-#error "Never use <xsavecintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __XSAVECINTRIN_H
-#define __XSAVECINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("xsavec")))
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_xsavec(void *__p, unsigned long long __m) {
- __builtin_ia32_xsavec(__p, __m);
-}
-
-#ifdef __x86_64__
-static __inline__ void __DEFAULT_FN_ATTRS
-_xsavec64(void *__p, unsigned long long __m) {
- __builtin_ia32_xsavec64(__p, __m);
-}
-#endif
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif
+++ /dev/null
-/*===---- xsaveintrin.h - XSAVE intrinsic ------------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __IMMINTRIN_H
-#error "Never use <xsaveintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __XSAVEINTRIN_H
-#define __XSAVEINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("xsave")))
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_xsave(void *__p, unsigned long long __m) {
- return __builtin_ia32_xsave(__p, __m);
-}
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_xrstor(void *__p, unsigned long long __m) {
- return __builtin_ia32_xrstor(__p, __m);
-}
-
-#ifdef __x86_64__
-static __inline__ void __DEFAULT_FN_ATTRS
-_xsave64(void *__p, unsigned long long __m) {
- return __builtin_ia32_xsave64(__p, __m);
-}
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_xrstor64(void *__p, unsigned long long __m) {
- return __builtin_ia32_xrstor64(__p, __m);
-}
-#endif
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif
+++ /dev/null
-/*===---- xsaveoptintrin.h - XSAVEOPT intrinsic ------------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __IMMINTRIN_H
-#error "Never use <xsaveoptintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __XSAVEOPTINTRIN_H
-#define __XSAVEOPTINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("xsaveopt")))
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_xsaveopt(void *__p, unsigned long long __m) {
- return __builtin_ia32_xsaveopt(__p, __m);
-}
-
-#ifdef __x86_64__
-static __inline__ void __DEFAULT_FN_ATTRS
-_xsaveopt64(void *__p, unsigned long long __m) {
- return __builtin_ia32_xsaveopt64(__p, __m);
-}
-#endif
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif
+++ /dev/null
-/*===---- xsavesintrin.h - XSAVES intrinsic ------------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __IMMINTRIN_H
-#error "Never use <xsavesintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __XSAVESINTRIN_H
-#define __XSAVESINTRIN_H
-
-/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("xsaves")))
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_xsaves(void *__p, unsigned long long __m) {
- __builtin_ia32_xsaves(__p, __m);
-}
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_xrstors(void *__p, unsigned long long __m) {
- __builtin_ia32_xrstors(__p, __m);
-}
-
-#ifdef __x86_64__
-static __inline__ void __DEFAULT_FN_ATTRS
-_xrstors64(void *__p, unsigned long long __m) {
- __builtin_ia32_xrstors64(__p, __m);
-}
-
-static __inline__ void __DEFAULT_FN_ATTRS
-_xsaves64(void *__p, unsigned long long __m) {
- __builtin_ia32_xsaves64(__p, __m);
-}
-#endif
-
-#undef __DEFAULT_FN_ATTRS
-
-#endif
+++ /dev/null
-/*===---- xtestintrin.h - XTEST intrinsic ---------------------------------===
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- *
- *===-----------------------------------------------------------------------===
- */
-
-#ifndef __IMMINTRIN_H
-#error "Never use <xtestintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef __XTESTINTRIN_H
-#define __XTESTINTRIN_H
-
-/* xtest returns non-zero if the instruction is executed within an RTM or active
- * HLE region. */
-/* FIXME: This can be an either or for RTM/HLE. Deal with this when HLE is
- * supported. */
-static __inline__ int
- __attribute__((__always_inline__, __nodebug__, __target__("rtm")))
- _xtest(void) {
- return __builtin_ia32_xtest();
-}
-
-#endif
projects / platform / framework / web / chromium-efl.git / commitdiff